This PDF file contains the front matter associated with SPIE Proceedings Volume 13439, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

First of all, the test requirements of national standard 38383-2019 "Dishwasher energy efficiency water efficiency limit value and grade" are interpreted, and then the test requirements of dishwasher water efficiency test are put forward. Secondly, according to the laboratory test method, the principle design scheme is proposed and the function of dishwasher water efficiency device is analyzed and explained to prove the feasibility and superiority of this structure. Finally, the importance of dishwasher water efficiency testing in the industry is explained.

In the measurement process of the portable braking performance tester, it is necessary to first ensure that the sensors of the portable braking performance tester are fixed on the calibration platform of the static calibration device of the portable braking performance tester, and then ensure that the calibration platform is in a horizontal state. This is a prerequisite for the calibration of parameters such as indication error and repeatability in subsequent static calibration. The article introduces a portable static calibration device for braking performance testing based on a dual-sided laser digital display tilt box, discusses the technical principle, structural design, and component functions of the calibration device, and proposes specific operating methods and steps for this calibration device. According to the analysis of experimental data after implementation, it can be concluded that the measurement of the portable braking performance tester using this calibration device and its operating method can meet the technical requirements of JJF1168-2007 “Portable Braking Performance Tester”.

In order to solve the dynamic synchronization problem of two laser trackers, two laser trackers are used in this paper, equipped with two measurement target balls placed on a large indoor standard device platform. By moving the platform at a constant speed at a certain measurement distance, the dynamic measurement distance between the two target balls and the speed of the platform during the dynamic condition are analyzed. The following conclusions are drawn: In synchronous measurement, when the sampling frequency of the two laser trackers is set to 50Hz, the maximum value of the speed difference between the platforms during the uniform speed stage is 3.1μm/s. When the sampling frequency of the two laser trackers is set to 10Hz, 20Hz, and 50Hz, the maximum relative distance between the two target balls is 1.778mm, 1.765mm, and 1.764mm respectively. The measurement method used in this paper can be further studied in dynamic synchronous measurement through certain transformations, and up to five laser trackers can be connected for synchronous measurement, which has great application prospects.

Direct mercury analyzer can directly test samples without sample pre-treatment, which leads to the advantages of no reagent pollution, less sampling volume, high sensitivity, good precision, and low cost. Therefore, it is widely used to analyze mercury in food, drinking water, cosmetics, soil, etc. As the key equipment for analyzing mercury, the accuracy of a direct mercury analyzer is essential to ensure the accuracy and reliability of the test results of mercury, which has great significance for environmental monitoring and protection. This article developed a calibration method to evaluate the metrological characteristics of a direct mercury analyzer by studying its working principle. Multiple different types of direct mercury analyzers were selected for pre-testing, and national-certified reference materials were chosen as references. Critical parameters such as linear error, detection limit, and repeatability were selected as test items. The experimental results show that the method proposed and reference materials selected can effectively evaluate the metrological performance of the instrument, which is of great significance for the quality control and quantity traceability of direct mercury analyzers.

In response to the insufficient qualitative and quantitative detection of thimble tube wear using conventional BOBBIN probe, it was proposed for the first time to use array eddy current testing technology to study thimble tube wear. This method is more conducive to accurately characterizing the distribution patterns and morphological features of thimble tube wear. Through simulation and experimental analysis, it was found that the experimental test results of the array were basically consistent with the simulation. It was proven that array eddy current can enhance the detection capability of BOBBIN probes, improve the detection process, significantly enhance the qualitative and quantitative level of defect detection, and have significant implications for guiding the maintenance of sleeve tubes in nuclear power plants. it also fills the gap in array eddy current testing of thimble tube in domestic nuclear power plants.

This article presents a pulse-type ultrasonic Doppler fetal heart detection system. An ultrasonic probe based on pulse detection technology was designed, including a single piezoelectric ceramic chip transducer and an ultrasonic transmitting and receiving module. A pulse ultrasonic detection signal transformation model was constructed, and a fetal heart detection system was built, which was composed of a main control module, a timing control module, an ultrasonic transmitting module, an echo receiving module, a signal demodulation module, a signal conditioning module, a human-computer interaction module, a Bluetooth communication module, an output processing module, a power management module, and other key circuits such as four-level low-noise amplifier and filter were designed. The designed system was tested according to JJG 893-2007 Ultrasonic Source of Ultrasonic Doppler Foetal Meters, and its indication error and repeatability of fetal heart rate can meet the requirements of JJG 893-2007. Meanwhile the designed system can complete fetal heart detection efficiently and accurately.

In order to remote monitor and control the natural gas valves, the design and implementation of an intelligent remote control system for natural gas valves based on the Internet of Things is proposed in this paper. The proposed system enables real-time monitoring and remote control of valve states. Additionally, the system introduces a scheme for automatically controlling valve opening values based on the data from the sensors on valves. In order to facilitate user operation and configuration the thresholds for scheme judgement, the corresponding app has been designed for remote control of the valve. The test results demonstrate that the system successfully enables intelligent remote control of natural gas valves. Compared with other natural gas valve systems, this system uses electro-hydraulic actuators with ESD function and reduces the cost to some extent while ensuring the efficiency of data transmission. Therefore, this result is of great practical significance for remote control and automation of natural gas valves.

It is essential that accurate measurement of projectile in-bore kinematic parameters for the design optimization, performance and effectiveness evaluation of projectiles, especially for the optimization of the design of new artillery shell functional components to resist the high overload of launching and the evaluation of the overload resistance. Combined with the engineering practice of new type of artillery shell, based on the analysis of the launching environment of artillery and the motion characteristics of the projectile in the bore, this paper discusses the basic measurement conditions required by the relevant testing systems, and expounds the principles and characteristics of the main testing methods for the motion speed and rotation speed of the projectile in the bore and their corresponding acceleration and jitter posture. The research deficiencies and subsequent development in this field are summarized and prospected, and solutions are put forward for synchronous measurement of multi-parameters of projectile motion characteristics and the accurate and efficient processing of test data.

In this study, a deep multilayer perceptron machine network (DMLP) framework that integrates feature engineering and weighted voting integration methods, called spectral profile feature engineering and weighted voting method (SPFE-WV), is proposed. In order to accurately predict the gas concentration in an experimental environment where various noises and background baseline interferences are present, through the preprocessing of the second-harmonic data time series, a feature engineering and weighted voting integration method to successfully extract key features such as the height and area of the main peak and the height difference between different concentrations.

Microwave technology has been widely used in clinical applications at home and abroad, especially in eliminating a variety of inflammation, relieving patients' pain, promoting blood circulation has shown unique therapeutic effects, and has been widely recognized by the medical community. Inaccurate output power and frequency of the instrument, poor stability, excessive harmonic frequency, etc., which affect the treatment effect and generate interference signals in the surrounding environment. Therefore, the design and development of a testing device, according to the use of the instrument, regular performance evaluation and maintenance of the equipment is of great significance for ensuring the safety of microwave equipment and treatment effect, and preventing adverse events.

This paper addresses the issue of inaccurate detection caused by noise interference in the echo signals of electromagnetic ultrasonic thickness measurement. A method to improve detection accuracy is proposed. An experimental platform was established, where planar spiral coils and permanent magnets were used to excite ultrasonic shear waves for thickness measurement experiments on metal pipes of varying thicknesses. The echo signals were denoised using an adaptive variational mode decomposition algorithm based on the Chimp Optimization Algorithm (ChOA-VMD), and effective peaks were extracted using the Findpeaks function. The final wall thickness data were then calculated. Experimental results show that this method reduces measurement errors to less than 0.1%, effectively filters out noise, extracts useful signals, and significantly improves detection efficiency and accuracy.

The variety of cigarette box printing inks, diverse patterns, and extensive use of reflective laser materials pose significant challenges for quality inspection. Currently, cigarette box quality inspection relies on slow and subjective manual methods, falling short of accurate and objective standards. However, while digital inspection holds promise, it faces challenges from strong reflections during image capture, leading to data distortion. Aiming to address this concern, we propose a network named Dual-stream Pyramid Gated-SwinUnet (DPG-UNet). Firstly, the dual-stream interaction mechanism enhances information exchange between features at different scales. Secondly, the pyramid model more effectively captures multiscale features, improving reflection removal accuracy. Finally, the proposed Gated-SwinUnet architecture, which significantly enhances feature representation capabilities by integrating the gating mechanism with the SwinUnet model. Our design uses the adjustment effect of the gating mechanism to optimize the interaction between the reflection layer and the projection layer, achieving deep fusion and effective separation of features. The experimental outcomes confirm the effectiveness of our approach significantly enhances the ability to remove reflections from cigarette box surfaces, providing higher-quality image inputs for subsequent analysis and automated inspection.

With the acceleration of urbanization and advancements in construction technology, the utilization of underground spaces has become increasingly important. However, the complexity and invisibility of underground environments pose significant challenges for monitoring and management. To address these issues, this paper proposes a design scheme for an autonomous obstacle-avoiding underground environment monitoring device based on the STM32 series microcontroller. This device integrates various high-precision sensors and advanced data processing technologies, enabling real-time monitoring of various underground environmental parameters. The data can be transmitted wirelessly to an online terminal for further processing and analysis. Additionally, the device features autonomous mobility and intelligent obstacle avoidance, enhancing the safety and efficiency of underground construction. This design not only achieves technical innovation but also provides new solutions and technical support for the intelligent construction and sustainable development of underground spaces.

The current water supply and drainage pipe network coverage monitoring node deployment is generally single point coverage, and the monitoring area is small, resulting in an increase in the drop value of the detection pressure. For this reason, the optimization of regulation and control of water supply and drainage pipe network leakage under variable pressure based on the identification algorithm of attenuated underwater acoustic signal is proposed. Based on the current measurement, the multi-level method is adopted to expand the monitoring area, and the multi-level deployment of water supply and drainage pipe network covers the monitoring nodes to capture the leakage abnormal signal and regulate the fuzzy delimitation of coverage. Based on this, an optimization model for the regulation and control of water supply and drainage pipe network leakage under variable pressure is constructed by identifying and calculating the attenuated underwater acoustic signal, and the adaptive balance adjustment method is adopted to realize the optimization of scheduling. The test results show that compared with the sensitive design pipeline pressure regulation optimization detection Fluent water pipeline optimization scheduling method, the attenuation underwater acoustic signal identification algorithm of this design, the pipe network uncertain pressure leakage control optimization method, the final detection pressure drop value is relatively high, which shows that the design of the scheduling optimization method has high coverage and stability, the scheduling results are significantly optimized, and further reliable practical verification.

Develop a series of quality control materials (QCMs) for ruby color grading. Ruby was synthesized using flame melting method, and the ruby samples were cut and polished to prepare samples of the same size and glossiness. According to GB/T 32863-2016, D50 standard light source is used, and 9 experienced jewelry grading technicians collaborate to determine the standard value and uncertainty. The determined values mainly include color tone, brightness, and saturation, with Munsell hue ranging from 2.78RP to 1.25R, Munsell value ranging from 3.4 to 6.7, and Munsell chroma ranging from 7.4 to 11.0. This series of substances provides technical support for quality control of ruby products, confirmation of experimental methods, and personnel identification and evaluation.

As a key component of the tank circuit breaker, the closing resistance mainly inhibits the closing inrush current of the circuit breaker in the process of opening and closing, and accidents occur frequently during operation. Based on this, the thermal and mechanical simulation model of the closing resistance is established to analysis the failure of the impact current on the closing resistance; Firstly, the temperature rise process under different impact accurent with the amplitudes of 1.12kA, 1.68kA and 2.24kA respectively. Then and the stress distribution is also calculated. The results show that Excessive impact current causes faster temperature rise, which can lead to thermal damage of the closing resistor. While the impulse current reaches about 1.76kA, the average stress of the closing resistance reaches the material limited stress of the closing resistance and the maximum stress is on the outer edge, so it turns out the early destruction occurs this place. These conclusions can guide the analysis of the failure mechanism of the closing resistor.

In UHV and UHV transmission lines, tank circuit breakers need to be switched frequently, and the pre-insertion resistors is subject to the mixed impact of AC and DC at the same time, so the pre-insertion resistors is easy to be damaged, which affects the reliability of the power system. The pre-insertion resistors needs to withstand a lot of heat when acting, and the temperature distribution and heat dissipation capacity inside the pre-insertion resistors are particularly important. Based on this, this study first observed the internal morphology of the pre-insertion resistors through electron microscopy, established a Voronoi diagram model according to the internal morphology, built a simulation model of the pre-insertion resistors and calculated the temperature and electric field distribution of the pre-insertion resistors under different porosity. The results show that after injection of impulse current pulse with peak value of 2.03kA, the highest temperature inside the pre-insertion resistors at 366.5°C appears at the pore next to the carbon black conductive chain. Higher rate of porosity will make the temperature inside the pre-insertion resistors more evenly distributed, but will bring about an increase in the strength of the overall electric field inside.

To improve the trajectory tracking accuracy and disturbance rejection capabilities of multi-degree-of-freedom robotic arms, this paper proposes an enhanced Active Disturbance Rejection Control (ADRC) strategy. This strategy involves designing a modified Extended State Observer (ESO) based on nonlinear functions to monitor internal disturbances and the extended state of the control system, which encompasses nonlinear factors, internal model uncertainties, and external disturbances. A Fuzzy proportional-derivative(PD) control controller is employed to mitigate residual errors. Simulation and experimental results demonstrate that the improved ADRC-based trajectory tracking system significantly outperforms traditional control strategies in terms of tracking accuracy and disturbance rejection.

Currently, industrial robots typically monitor and warn of environmental contact forces during press-fitting operations; however, their control modes are relatively simplistic, making it challenging to achieve compliant force processing. To enhance the press-fitting performance of industrial robots in manufacturing environments and to enable smoother force-controlled press-fitting, an innovative approach is proposed in this study. The method involves force tracking through an inner loop of position control, where the position is adjusted according to the magnitude of the force deviation using a fuzzy controller. This leads to higher accuracy in force tracking. The fuzzy controller is specifically designed to address the high stiffness and uncertainties inherent in the end-effector's contact environment, thereby enabling the industrial robot to exhibit strong adaptability. Experimental results demonstrate significant improvements in contact force tracking accuracy, along with excellent stability and robustness. This method facilitates soft and smooth press-fitting by industrial robots, thus improving both precision and efficiency. Furthermore, the adaptability of the robot system to the external working environment is enhanced, exhibiting superior disturbance rejection capabilities.

In order to realize the design and development of the rope-driven soft body spring robot, a kinematic model is established by three-dimensional Cartesian coordinate system and Euler angle transformation method for a drive system with three ropes as active drive and three springs as passive drive, the relationship between the motion position and joint variables of the model and the length of ropes, and the theoretical simulation and virtual prototype simulation and comparison of the model motion are carried out based on the software of Matab and ADAMS. Based on Matab and ADAMS software, the theoretical simulation and virtual prototype simulation of the model were carried out and compared. By analyzing the potential energy and kinetic energy under different working conditions, and based on the Lagrangian dynamics method, the general equations of dynamics are obtained. Adams is utilized to simulate the spring force at a given speed. In the whole process of kinematics and dynamics modeling and simulation, the simulation results and theoretical calculations are in good agreement, and the results provide an important theoretical basis for the development and design of the robot, which is of guiding significance for the design work.

Selective super-wetting surfaces maintain the contrasting super-wetting properties for the oil and the water, and that have received widespread attention since 2000 due to their high surface energy, particularly in oil-water separation applications. Concerning about these surfaces, superoleophobic/superhydrophilic surfaces are effective due to their oil-repellent characteristics.In this paper, Ultraviolet light polymerization was used to polymerize fluorinated substances (TFOA) and hydrophilic substances (MMA), forming a coating on chemically etched copper foam . The resulting porous copper foam exhibited superhydrophilicity and superoleophobicity, facilitating effective oil-water separation tests. The fluorinated substances imparted oleophobic properties to the surface, while the hydrophilic substances provided hydrophilic characteristics to the copper foam. Wear resistance tests using sandpaper and a Taber abrasion tester confirmed that the copper foam exhibits good mechanical durability. The superoleophobic/superhydrophilic copper foam can offer the solution that can overcome continuous oil/water separation process, and help us develop antifouling fabrics, together with selfcleaning surfaces.

The essay explores smart meter status assessment technology within the context of Advanced Metering Infrastructure (AMI), aiming to enhance the reliability and maintenance efficiency of smart meters. By implementing advanced monitoring and fault detection, health status evaluation, remote diagnostics, and data analysis methods, the study achieves real-time evaluation of smart meter status. The findings indicate that these technologies effectively improve meter accuracy, extend operational life, and reduce maintenance resources, thereby optimizing overall smart grid management.

Traditional ant colony algorithms often fall into local optima and deadlocks, leading to unsolvable or suboptimal solutions in the path planning of unmanned underwater vehicles (UUVs). Furthermore, they lack the ability to dynamically avoid collisions. To address these issues, this paper proposes a dynamic path planning method that combines global path planning using an improved ant colony algorithm with a local dynamic collision avoidance strategy. First, the traditional ant colony algorithm is optimized to prevent local optima and deadlocks by employing a roulette wheel selection method and enhancing memory-based backtracking strategies. Subsequently, a Kalman filter-based trajectory prediction method is introduced to handle irregularly moving obstacles, along with specific collision avoidance strategies for different types of collisions. The proposed method ultimately achieves dynamic path planning for UUVs. Simulation experiments demonstrate that the improved ant colony algorithm achieves a convergence speed approximately 16.7% faster than the traditional algorithm and can effectively perform path planning in dynamic environments, resulting in an optimal collision-free path.

The frequency response method testing of power transformers is a basic test project for transformers. In practical operation, this project requires a long test cable to facilitate the introduction of test signals into the testing instrument. At the same time, the response signal of this test can reach a minimum of microvolts. Due to the presence of test cables, the measurement results of small signals are greatly affected by the testing environment, wiring status, and external sudden RF interference. Different results are often obtained from the same measurement, which poses difficulties for correctly analyzing the status of transformers.

We have developed a winding deformation tester based on telemetry technology, which utilizes radio frequency communication technology and integrated high-frequency signal measurement technology. The testing unit can be directly connected to the sleeve, eliminating the influence of the testing line in the experiment. In principle, it achieves one-time wiring and multiple measurements, greatly improving the accuracy and efficiency of the experiment. After trial use on the transformer and comparison with the original test data, good test results have been achieved.

A multifunctional tire lifting device has been designed to directly act on the tires and achieve the functions of lifting and moving cars, as the large number of cars requires high convenience in maintenance and repair. This device can simplify the operation process of tire replacement, wheel hub maintenance, tire balance adjustment, and inflation. By lifting the tires to the appropriate height, it enhances the convenience and safety of car maintenance. This design not only meets the basic functional requirements of car lifting, but also achieves significant improvements in user experience and work efficiency.

With the continuous progress of science and technology, the requirements of measurement technology become more and more accurate and diversified. In the field of electrical measurement, the digital multimeter, as an important tool, is widely used in the measurement of parameters such as voltage, current and resistance and so on. However, the traditional manual measurement method is less efficient and easily affected by human error. Therefore, a software for automated measurement and control of the digital multimeter based on LabVIEW is designed in this paper, which is able to meet the requirements of today's measurement. The software can automatically perform an over-difference analysis to judge whether the measurement data meet the standards through the process in which, by using graphical programming LabVIEW controls the digital multimeter and also inputs signal to it with a DC voltage source in order to automatically collect and compare the measurement results of it. In this case, it can be said that the software can effectively improve the accuracy and efficiency of measurement, and can be widely applied with large potential.

Aiming at the problems of many steps and large amount of data in manual verification of oscilloscope DC gain, the remote verification module of oscilloscope DC gain is designed based on LabVIEW. According to the parameter setting requirements of oscilloscope DC gain in the digital oscilloscope calibration regulation GJB7691-2012, the real-time setting and data reading of the RTM2054 oscilloscope parameters are designed, and the calculation and storage of the DC gain data are completed, which greatly improves the verification efficiency of the oscilloscope DC gain verification.

Electrical resistance tomography (ERT) is a powerful method for two-phase flow measurement in various applications, however, the image reconstruction quality of ERT is often unsatisfactory due to the ill-posed nature of the inverse problem. In order to improve the imaging quality of ERT, a deep learning approach based on Transformer architecture is proposed. To reduce the risk of overfitting on small datasets caused by the weak inductive bias, a convolutional layer is used before the Transformer Encoder module in this paper. Trained on boundary voltages and images of the target location, the Transformer model establishes a nonlinear mapping relationship between them. The simulation results demonstrate that the proposed Transformer model achieves superior image reconstruction performance compared to the traditional image reconstruction algorithm, and have good generalization and noise resistance capabilities.

This article proposes a new method for bearing acoustic detection based on a parabolic acoustic mirror. The parabolic acoustic mirror is used instead of the traditional bearing end cover, and a directional microphone is placed at the focal point of the parabolic acoustic mirror. The fault sound emitted by the bearing is focused on the directional microphone after being reflected by the parabolic acoustic mirror, achieving signal gain acquisition. Due to the inherent properties of paraboloids, non-parallel incident sound waves cannot converge at the focal point, thus achieving directional acquisition of bearing fault signals. Further, the single channel underdetermined blind source separation algorithm is used to extract the sound signal emitted by the bearing fault point, achieving high-quality acquisition of fault signals. Through experimental comparison and analysis with vibration signals and microphone direct sampling signals, it has been proven that the bearing fault signals collected by the parabolic acoustic mirror have certain advantages compared to the other two methods, indicating the feasibility of collecting bearing fault signals by the parabolic acoustic mirror.

In this paper, a new method for bearing acoustic detection is proposed based on the parabolic acoustic mirror, which uses the parabolic acoustic mirror instead of the traditional bearing end cap, places a directional microphone at the focal point of the parabolic acoustic mirror, and focuses the fault information signal emitted by the bearing at the directional microphone after the reflection of the parabolic acoustic mirror to achieve gain acquisition. At the same time, due to the inherent properties of the parabola, the non-parallel incident sound waves cannot converge at the focal point, so as to realize the directional acquisition of the bearing fault signal. Combined with the support vector machine, the acoustic features are extracted, and the high-dimensional features are reduced to three-dimensional features by using the KJADE algorithm, and they are input into the support vector machine as feature vectors for identification and classification of various bearings. Experiments show that the proposed method can accurately classify bearing faults.

The importance of detecting bearing failures lies in ensuring the reliability, safety, and efficiency of mechanical systems. A failed or damaged bearing can lead to severe mechanical breakdowns, injuries, and equipment damage. However, current bearing fault diagnosis algorithms lack accuracy and efficiency. To address these issues, we propose a bearing fault diagnosis method called Bearing Fault Detection based on Area Equalization (BFAE) algorithm, which incorporates outlier detection algorithms from machine learning. The algorithm divides the dataset uniformly, identifies the optimal measurement radius, establishes a neighborhood for each object based on this radius, and evaluates the consistency of object density with the neighborhood’s average density. Subsequently, outliers are labeled, with objects exhibiting higher outliers classified as anomalous. A comparative analysis of the BFAE algorithm against five other algorithms (LOF, COF, Autoencoder) on ten high-dimensional real datasets demonstrates that the BFAE algorithm outperforms others in key metrics such as Area Under the Curve (AUC) and Accuracy (ACC). Bearing fault detection.

Based on the material's inherent functional properties, the work established a method of detecting the depths of double cracks for conductive concrete by use of a simplified electrical model. Kirchhoff's law is used to construct a computation method for the boundary potential response under current excitation, and an inversion method for double cracks length detection is established. In order to verify this method, the potential response data of double-crack damage in conductive concrete specimens at various locations and sizes were obtained using ANSYS Parametric Design Language (APDL) command flow. Good agreement was found in the results of theoretical calculations and numerical simulations.

This article provides a comprehensive introduction to the methods and scoring criteria of the Mobile Progressive Deformable Barrier (MPDB) crash tests under the China New Car Assessment Program (C-NCAP). It also conducts a statistical analysis based on the results from 32 tested vehicle models. Initially, it describes the test methods and dummy injury scoring criteria of the MPDB crash tests. Based on this, it presents a comprehensive statistical analysis of the test results, covering the scoring details for various parts of the test dummies and analyzing injury indicators for key points where points were lost. Finally, it summarizes the findings and offers recommendations for vehicle design optimization and safety performance evaluation.

This paper presents a quantitative evaluation method for assessing the fault isolability of gas pressure regulators without available system models. The method involves generating data-driven residuals using input and output data from gas pressure regulators in a fault-free scenario. The fault isolation performance analysis is then reformulated as a difference measure in residuals for faulty cases. Furthermore, the paper proposes fault isolability evaluation indices that combine distance and direction similarity to comprehensively evaluate the difficulty in fault isolation. The effectiveness of this method is demonstrated through the use of experimental data from gas regulators.

The reliability of Programmable Logic Controller (PLC) systems is critical for the safe and efficient operation of oil and gas pipelines. These systems, which control various components such as valve houses, station and yard technology, oil pump control, and fire gas systems, face significant challenges due to the gradual accumulation of failure programs over time and through system updates. This paper investigates the causes of PLC failure programs, emphasizing their hidden nature and the resulting system overcrowding, resource consumption, decreased operational efficiency, and potential risks to safety and reliability. Drawing on a decade of practical experience in failure prevention, the study analyzed data from 8720 points across different control components. The proposed measures, which include enhancing management systems, standardizing workflows, and improving personnel competencies, resulted in significant reductions in failure rates: 25% for valve house switching values, 30% for station and yard technology switching values, 20% for analog quantities, 28% for oil pump control switching values, 22% for analog quantities, and 18% for fire gas systems. Overall, these measures led to a 26% average reduction in system overcrowding and a 22% increase in operational efficiency. The study highlights the importance of systematic management and continuous improvement in maintaining the reliability of PLC systems, providing a framework for enhancing the stability and safety of national oil and gas pipeline networks.

Crash test dummies are important instruments for measuring the passive safety of automobiles. The changes in their mass property parameters could lead to changes in their kinematics during the test and eventually affect the accuracy of the test results. To solve the problem that mass characteristic parameters of collision dummies can not be measured, this study investigates the principles of center-of-mass measurements, and also designs measurement devices that can measure the mass property parameters of each part of the crash dummy assembly based on this test principle. Moreover, it analyzes the systematic error and random error of the devices, and conducts a measurement analysis of the mass property parameters of the head assembly of the crash test dummy. Such devices can measure the mass property parameters of the crash test dummy assembly with a small error, which establishes the test conditions for the development of Chinese crash test dummies.

In order to supply the demands of modern hydropower stations, a fault diagnosis system based on SOA Architecture was built. This system adopts Depth First Search Strategy with Pruning Processing in the matching process between fault and feature, which can improve the efficiency of inference engine. The inference is achieved by completely matching or partially matching with question. By storing the data of knowledge base with Neo4j, this system can manage expert knowledge base effectively. Based on "Confidence level", this system implement dynamic learning with expert assisting. Through this system, hydropower stations can prevent accidents and improve the maintenance mode, reduce the cost of operation, and improve the efficiency of machine.

With the development of Industrial Internet, the idea of Industrial Internet has penetrated into all walks of life, actively promoting the transformation and upgrading of China's manufacturing industry. In the metallurgical industry, for some special and typical low-speed and heavy-duty equipment, they play an extremely crucial role in the production process. However, currently, the operation and maintenance of such equipment are mainly manual, and usually only when the equipment has already malfunctioned or the malfunction is relatively serious, human workers will discover it and take post maintenance measures. In view of this issue, this article introduces a unique technology - stress wave technology, which can accurately locate and analyze faults in the early stages of faults. It has been well applied in the metallurgical industry, especially in typical low-speed and heavy-duty equipment. The online monitoring and diagnosis of the converter tilting primary and secondary reducers in a steelmaking plant of a certain steel enterprise was completed using this technology. Finally, the monitoring data showed that this monitoring technology can comprehensively reflect the operating status of the converter tilting reducers and effectively monitor the health status of the equipment.

In response to the national "carbon peaking and carbon neutralization strategy", the transportation industry, especially ports, needs to vigorously promote the construction of energy consumption monitoring systems, increase the stability and real-time of monitoring data, and provide a basis for "carbon peaking". In view of the weak status of the port fuel online monitoring system, this paper combs and analyzes the current status of the port fuel equipment energy efficiency statistics, pointing out the deficiencies of the statistical methods at the present stage. Through the comparison and analysis of existing fuel online monitoring schemes, the vehicle emission diagnosis system is selected to collect the fuel consumption data of equipment in real time. Through sorting out and analyzing the data of some vehicles, the real-time fuel consumption data and corresponding proportion under different working conditions are obtained, and suggestions for optimization and improvement of production scheduling are put forward. Through online monitoring of fuel equipment, the online monitoring of all energy elements of the port is realized, providing data support for the port enterprises to achieve green and low-carbon development.

Aiming at the difficulty of identifying the failure state of wellhead structure, a method of monitoring and identifying the failure state of wellhead device is proposed. The minimum envelope entropy is used as the fitness function of GWO to optimize VMD parameters K and a. The retained IMF is selected according to the spectral characteristics, energy and correlation, and the interference of noise and irrelevant components is removed. The time domain, frequency domain and entropy eigenvalues were extracted from the signal after de-noising reconstruction, and the extracted eigenvalues were dimensionally reduced by PCA to obtain the main eigenvalues containing signal information. Finally, the signal was classified and recognized by support vector machine (SVM). The experimental results show that this method can identify the damage of wellhead equipment effectively.

As the core component of the power system, gas-insulated metal-enclosed switchgear (GIS) plays a crucial role in ensuring the safe and stable operation of the power system, so the fault diagnosis research of GIS has significant practical significance. At present, GIS fault diagnosis often adopts the insulation diagnosis method based on acousto-optic signals, but due to the complexity of the GIS operating environment, the ultra-high frequency signals in the environment will cause interference to the traditional partial discharge detection device. In order to solve the above problems, this paper builds a GIS fault diagnosis platform, effectively extracts the vibration signals under the fault condition through the vibration signal analysis method, and combines the short-time Fourier transform and convolutional neural network to diagnose the faults of the GIS by using the end-to-end fault pattern recognition model. The results show that the model has good accuracy and practicability for GIS fault diagnosis, and provides a new idea for the fault diagnosis of GIS equipment.

Aiming to address the low efficiency and accuracy issues of traditional methods for diagnosing faults in belt conveyor rollers under noisy conditions, we propose a fault detection model named DSCNN-BiLSTM. This model integrates a multichannel Deep Separable Convolutional Neural Network (DSCNN) with a Bidirectional Long Short-Term Memory (BiLSTM) network, enhanced by an attention mechanism. The multi-channel Deep Separable Convolutional Network in this model extracts diverse local features by applying different convolutional kernels to each channel, which reduces the number of parameters and computational requirements, thereby significantly improving computational efficiency. The BiLSTM network manages time-series dependencies, allowing the model to leverage both local features and global sequence information effectively. The incorporation of a channel attention mechanism enables the model to adaptively select channels containing fault features, enhancing both accuracy and noise resistance. Experimental results demonstrate that this model performs effectively in fault detection, offering certain reference value and practical significance.

In this paper, a open-cut subway station project in Beijing is taken as Beijing. By analyzing a large amount of monitoring data, the factors affecting the accuracy of machine vision instrument are analyzed. The influence degree of each factor on the machine vision instrument is quantified by calculating the correlation coefficient. The results show that light intensity, distance and vibration are the main factors affecting machine vision instrument, while temperature, humidity, air quality and other factors have little influence on instrument accuracy.

This paper addresses the critical challenge of rapid power restoration in distribution networks by presenting a Multi-Agent System (MAS) designed to minimize network losses. The study begins with a comprehensive background on the increasing complexity of modern power distribution systems and the associated need for advanced fault management technologies. The proposed MAS integrates fault detection, isolation, and load management through a decentralized network of agents that collaborate in real-time. The system leverages smart switchgear for rapid fault isolation and utilizes distributed control mechanisms for efficient rerouting of power. Additionally, continuous network topology optimization is employed to reduce losses. Simulation results demonstrate the MAS's effectiveness in significantly improving restoration speed and enhancing overall network efficiency, making it a promising solution for smart grid applications.

This paper presents an innovative fault isolation technology for distribution networks using wide-area information processing. The technology addresses limitations of existing methods by integrating adaptive protection mechanisms and real-time analytics to enhance fault detection accuracy, reduce isolation time, and improve network stability. The approach is particularly effective in complex network environments, ensuring safer fault current levels and higher overall reliability.

In order to solve the problem of real-time synchronous display of wafer warping monitoring, a wafer warping monitoring method based on DTing is proposed to realize the interactive symbiosis of wafer physical space and virtual space. The wafer warping monitoring system based on the five-dimensional DT model is constructed, including digital space construction, MCC-DT data drive, twin data transmission and management, and service system. The physical wafer warping is simulated by displacement loading, and the virtual wafer warping is obtained synchronously by the established twin system. The experimental results show that the accuracy of the proposed system can reach 90%, and provide help for wafer digital manufacturing.

With the continuous extension of oil pipelines to polar, ocean and geological unstable areas, the failure behaviors related to local flexion, instability and ductility fracture are easy to occur under the action of operating conditions such as high pressure and large deformation. The quantitative detection of the plastic damage degree of pipeline steel material needs to be further discussed. In this project, the typical bipolar pipeline steel is taken as the research object, and the GTN thin damage model is used to describe the plastic damage of pipeline steel, analyze the influence of thin structure on the deterioration of the mechanical properties of materials; The ultrasonic technology was introduced to detect the plastic damage degree of pipeline steel, and the correlation between ultrasonic wave speed and the volume fraction of the damage variable f was studied, and the correlation model between the plastic damage behavior and ultrasonic speed of pipeline steel material was established. provides a basis for quantitative assessment of plastic steel and safe service status of pipeline.

This paper presents a fault detection method based on locally linear embedding (LLE) for the high-speed train traction systems. The method maps high-dimensional complex data into low-dimensional data, analyzes the spatial characteristics of these data within their local neighborhoods, and achieves overall system state monitoring and fault prediction. The paper provides a detailed introduction to the theoretical foundation of LLE fault detection, including techniques such as local linear embedding, fault feature extraction, and analysis, and discusses the roles of these techniques in practical applications. Through the construction of simulation models and experimental data, the effectiveness and robustness of the proposed method are verified.

At present, due to the improvement of equipment performance, the strain value in load testing is increasing. In this paper, the systematic error in calibration process is analyzed quantitatively in theory, and the compensation method of calibration error is designed. The compensation method is verified to be effective and ensure the accuracy of strain data in load testing.

Autogram enables precise extraction of both demodulation band and fault characteristics. However, in presence of strong noise and complex interference, fault features extracted by Autogram may become indistinct, contributing to inaccurate identification of faults. This work proposes a fault detection approach for rolling bearings, integrating an enhanced Feature Mode Decomposition (FMD) algorithm with Autogram technique. First, The parameters of the FMD algorithm are finetuned through the Sparrow Search Algorithm. Following optimization, the FMD is utilized to break down the original vibration signal into several modal components. Subsequently, permutation entropy is calculated for each component to identify optimal one. Finally, Autogram is used to diagnose faults based on selected optimal component. Data from an experimental platform demonstrates that, compared to Fast Kurtogram and Autogram, proposed method significantly reduces noise and random interference. This improvement facilitates precise extraction of fault characteristic frequencies in rolling bearings.

To address the issues of excessive maintenance and untimely maintenance of bearings, this paper proposes a performance evaluation method for bearing condition monitoring based on the combination of Principal Component Analysis (PCA) and Self-Organizing Map (SOM) networks. This method utilizes PCA to reduce the dimensionality of dual-domain features and conducts clustering analysis using the SOM network, constructing a minimal quantization error as a degradation indicator to assess the bearing degradation state. This is an unsupervised approach that does not require the setting of training labels for the network. Experiments were conducted on the IMS dataset, and the results indicate that this method can effectively identify early bearing faults and reduce reliance on human expertise, thereby enhancing both overall efficiency and the intelligence level of assessments, ultimately achieving intelligent bearing fault recognition.

The cathodic protection was one of the main corrosion protection methods for buried metal pipelines. However, the cathodic protection between multiple pipelines in the pipeline corridor interfered with each other, and combined protection was the key to solve this problem. Under this condition, the target pipeline's cathodic protection potential was affected by the other pipelines, leading to inaccurate test results. Therefore, this paper studied the influence of the different coatings on IR drop in pipelines. The results showed that the IR drop of the target pipeline was mainly related to the quality of the total anticorrosive layer in the pipelines, showing a logarithmic relationship, and the more pipelines in the pipeline network was, the smaller the IR drop of the target pipeline caused by the stray current caused by the difference of anticorrosive coatings between pipelines was.

The existing fault location analysis methods for static reactive power compensators require laboratory testing or simulation results of faults, which are time-consuming and costly. The proposed method for analyzing the SVC fault location does not rely on simulation or laboratory testing results. It can be used to identify the type of fault and locate its source in a quick manner. This method can improve the efficiency of the analysis and fault investigation process.

In the examination of surface defects in newly manufactured aluminum alloys, Penetration Testing and Eddy Current Testing were employed to identify flaws in the welding process. The causes of defects in the welding of these newly manufactured aluminum alloys were highlighted, and the reasons for not using other non-destructive testing methods were also explained. Finally, a practical comparison of the test outcomes from both methods revealed their effectiveness in accurately detecting defects in practice, and ensuring the omissions of overlooked weld imperfections.

As the core data exchange equipment of substation, the stability and reliability of the switch is very important for the efficient operation of substation. However, due to the unique structure of the switch and strict industry test standards, the manual test method is widely used in the whole machine test at present, which has the disadvantages of labor-intensive and low automation level. This paper has carried out in-depth research on these technical problems that perplex the automatic test of switches. Firstly, the image recognition technology is introduced in the automatic recognition of LED panel lights of the switch, and successfully realize the automatic recognition of LED lights. Secondly, when solving the problem of intelligent judgment and handling of different types of switches, we introduce the handling robot to solve the problem of unmanned handling, and develop various types of flexible testing accessorial tool to adapt to different types of switches. Finally, in the aspect of switch flow detection, we propose an S-type flow test technology, and combine the automatic chemical equipment and flow tester to achieve a fast and effective flow pressure test. Based on the above research, a set of flexible intelligent test system for switches is developed. At present, the system has been put into use in an intelligent manufacturing demonstration workshop in Nanjing, Jiangsu Province, which effectively improves the testing efficiency of the switch, and has a good demonstration significance in the industry.

As cardiovascular disease rates continue to climb, atrial fibrillation (AF), a common type of arrhythmia, increasingly threatens public health. Traditional electrocardiogram (ECG) diagnostics depend heavily on the subjective interpretation by doctors, making them prone to human error. This paper presents a method that combines traditional feature engineering with deep learning techniques. It extracts key statistical features of the ECG signal through feature engineering, while also using a deep learning model based on a shifted window attention mechanism to further extract temporal dynamic features. Experimental results demonstrate that this method surpasses traditional techniques in detecting AF, especially improving specificity and sensitivity, thereby significantly enhancing diagnostic accuracy.

Common diseases of rice include blight, brown, dead heart, downy mildew, etc. These diseases can cause damage to rice leaves. If not recognized in time, it will lead to a decrease in rice yield and even the death of part or the entire crop in the land. Using deep learning technology to monitor rice crop diseases can provide early warning of rice crops, and then promptly specify prevention strategies when diseases occur, helping farmers and agricultural management departments to make scientific decisions to improve the efficiency and sustainability of rice production. Based on the core concept of ecological and environmental protection, we proposed an innovation to the You Only Look Once (YOLOv8) technical framework, focusing on optimizing the accurate identification technology of rice diseases and pests. This improvement aims to promote the sustainable development of agricultural production in a more environmentally friendly and efficient way. This article uses the improved You Only Look Once (YOLOv8) model to accurately detect the disease types in the collected data set. This article uses multi-scale convolution kernels and Attention to improve the C2f module of the yolov8 model. The improved model is in the data set An accuracy rate of 96.4% was achieved. This research helps to gain a deeper understanding of the applicability of computer vision in crop disease detection and further explore its application potential in agricultural production.

Recently, neural networks dominate various aspects of the field of ground-based cloud detection, serving as an advanced and robust method for cloud observation. Since loss functions are crucial for optimizing neural networks, the existing networks with advanced loss functions are also a vital area. This paper explores the impact of three distinct loss functions and their mixture effects on the optimization of network performance and gives strategies for selecting loss functions. To ascertain the efficacy of various loss functions, we carried out a comprehensive set of experiments on a dataset dedicated to ground-based cloud detection. Our findings indicate that employing a mixture of loss functions significantly enhances the training process for models focused on ground-based cloud identification.

As manufacturing transitions toward digitization, there have been significant changes in the manufacturing process compared to traditional methods. However, quality control methods have not yet undergone a new transformation to adapt to the increasingly intelligent production methods. During the inspection planning phase, due to the transmission of two-dimensional paper-based information, inspection information becomes cluttered and lacks reusability. This leads to identification issues for inspectors regarding the inspection objects, resulting in false positives and missed defects. To prevent losses caused by false positives and missed defects, this article proposes a lean quality inspection and control model for critical components of high-end equipment during the inspection planning phase. The model utilizes the Model- Based Definition (MBD) as the information representation carrier and XML as the data storage and transmission method, bridging accurate data transfer across different systems. This provides model support for the manufacturing industry in improving inspection efficiency and reducing inspection planning costs.

This paper explores various Maximum Power Point Tracking (MPPT) techniques in photovoltaic systems, focusing on their efficiency under different environmental conditions. Traditional methods like Perturb and Observe (P&O) and Incremental Conductance (IncCond) are compared with advanced fuzzy logic controls. Enhanced by fuzzy logic, the improved MPPT technique demonstrates superior adaptability and reduced oscillations near the Maximum Power Point, leading to increased overall energy output and system reliability. Experimental validations confirm significant enhancements in response times and stability, marking a step forward in optimizing photovoltaic energy systems.

Due to the complexity and uncertainty of substation operation environment, as well as the problem of insufficient sample data or insufficient diversity that may exist in practical applications. Therefore, this paper proposes a multilayer fuzzy detection of substation operation risk based on PSO optimisation BP neural network. The main advantage of this approach is that it combines the powerful nonlinear mapping ability of BP neural network and the global search ability of PSO algorithm, thus improving the performance of the detection system. Firstly, the operational data from substations are collected for normalisation or standardisation to suit the training needs of the neural network. Second, the structure of the BP neural network and the parameters of the PSO algorithm are designed to find the optimal neural network parameters by iteratively updating the speed and position of the particles. Finally, the optimised BP neural network is used for training and the network parameters are adjusted to improve the prediction accuracy. The experimental results show that this technique optimises the parameters of the BP neural network through the PSO algorithm, which helps the BP neural network to better adapt to the complexity and uncertainty of the operating environment of the substation, so as to more accurately detect and assess the operational risks.

Given the complexity of quality defects in aviation product development and production, which results in a lack of effective intelligent detection methods, this paper addresses these challenges by systematically reviewing key technologies for intelligent detection based on digital images. The paper summarizes the concepts and implementation methods of intelligent detection for various application scenarios and specific defect characteristics. For aviation castings with complex defect features and stringent detection requirements, a defect feature dataset for digital radiographic inspection of aerospace castings was constructed using data augmentation techniques such as cropping, flipping, overlap cutting, and Mosaic. Subsequently, an improved Mask-RCNN algorithm, incorporating the global feature pyramid network, was designed and optimized. This algorithm was used to test and verify the detection of three types of defects in aviation castings: looseness, cracks, and high-density inclusions. Experimental results demonstrate that the detection accuracy of this optimized algorithm is 93.25%, with a recall rate of 96.51%. Finally, based on the research findings, intelligent detection and evaluation software for aviation castings, grounded in deep learning, was developed and applied to the detection of a specific type of aviation engine blade casting.

In the process of installation and debugging in the industrial field, it is worth exploring whether the micrometer can be used to replace the high-precision sensors and supporting equipment to measure the amount of jitter of the motion plane. By splitting the micrometer, measuring, simplifying, building a simulation in ADAMS, designing a test scheme according to the scene of LCD panel inspection, and carrying out experimental research and data analysis and processing at different speeds, the results show that the micrometer is capable of measuring the plane jitter of LCD panel production equipment within a certain moving speed. This study provides a feasible basis for the use of micrometers for measurements under mobile conditions, which can significantly reduce the cost of enterprises.

This paper explores the advancements in visual inspection technology for substation insulator sweeping robots. It highlights the integration of high-resolution cameras and robust image processing software, utilizing machine learning and deep learning frameworks for precise defect detection.

Quality inspection of aerospace electrical connectors is one of the crucial safeguards for ensuring the safety of aerospace systems. However, the traditional method of manual sampling inspection struggles to meet the increasing production demands and carries the risk of undetected issues. This paper investigates the debris detection technology for aerospace electrical connectors based on machine vision and develops an automated vision inspection system to replace manual inspection for complex debris detection tasks in aerospace electrical connectors. In the study, based on the recognized pin points and same sequence pin alignment, the pin and the background of the printed line for the detection of debris are shielded, using the mask method and RANSAC algorithm, and the suspected debris region is extracted. Subsequently, the detection accuracy of the residual detection models obtained by different methods is compared, and the best-performing model, ShuffleNet-V2, is optimized. In the final test, the model trained by improved ShuffleNet-V2 network with the attention module ECA has the best detection effect with an accuracy of 99.2%.

Target detection technology is crucial in the domain of machine vision. Target detection technology, when paired with intelligent robots, enables industrial production to accurately determine the position of the target object. This integration enhances the level of automation in production, leading to improved production efficiency. In order to enhance the accuracy of target recognition, target detection technology relies on a diverse range of object samples. However, target detection models used in industrial production often face challenges such as limited real-time applicability due to their excessive model size and numerous parameters. In order to solve this problem, this paper proposes a method to train a detection model by combining the production of detection samples with digital twin technology, and to lighten and improve the model trained with this class of synthetic samples, we propose an improved target detection algorithm based on synthetic samples of yolov8n, called fast_yolov8. In the fast_yolov8 model, Dualconv structure and Adown structure are introduced in the backbone part, and AKConv is introduced in the Head part as a lightweight detection head, and the introduction of these three lightweight structures significantly reduces the computational cost and the number of parameters while guaranteeing the detection accuracy of the model. The experimental results show that compared with the pre-trained yolov8n model, the model parameter quantity of fast_yolov8n based on the synthetic sample set is reduced by 25%, the floating-point operations (FLOPs) are reduced by 23%, and the detection accuracy of mAP on the synthetic sample set is improved by 1.3%, which realizes the lightweighting of the model and maintains the detection accuracy of the model.

Wind turbine generator equipment is mainly distributed in more remote and harsh environment areas. The paddle of wind turbine operates all day for a long time and suffers from wind and sand attacks, resulting in paddle surface damage. It has become mainstream to further study detection algorithms on the surface defects of the object in place of manual inspection. After the fieldwork and the dataset making, it is found that there are less actual data samples collected, which is not enough to make the detection model training make the training of the detection model better. In this paper, in order to solve the problem of insufficient defect data required for the training of the detection model, a defect generation method combining the Transformer idea and the generative adversarial network is proposed to generate a large amount of defect data closer to the real distribution. And then the generated data is combined and divided with the original data, and trained and tested by the YOLOv7 detection algorithm. The experimental results show that after the training of the data samples generated in this paper, the mAP value of the detector reaches 87.8% in the training results, the recall rate increases by 1.6%, and the final test precision rate improves from 80.4% to 84.1%, and the leakage rate is 0%. Compared the results after training with the original data, it improves the accuracy and stability of defect detection.

This paper focused on the two key points of three-dimensional space irregularity and water level recognition in the monitoring of inflow from diversion tunnel. The water volume of diversion tunnel was obtained based on the spatial modeling technology of three-dimensional laser scanning and the detection algorithm of water level edge line based on Canny. Firstly, the mapping relationship between diversion tunnel elevation and water volume was obtained through three-dimensional space modeling. On this basis, a suitable viewing Angle was selected to deploy the camera, and the mapping data of the actual height between the reference points and the image pixel height were collected and fitted through reference point calibration. Finally, the pictures including the vertical wall and water level of the working platform of the diversion tunnel were obtained regularly. Canny algorithm was used to identify the horizontal edge line of the working platform and water level line, and the pixel height of the water level line from the edge of the platform was obtained. Then, the pixel height of the water level from the edge of the platform was converted into the actual water level, and the relationship between the elevation and the volume of available water was substituted to obtain the water volume. By comparing the actual measured water level height with the calculated water level in this scheme, the error was only 0.75%. Based on this method, the selection and start-stop control of auxiliary diversion tunnel drainage pump unit have been verified by practice.

A detection method of ice bottom interface melting based on a transmissive fiber optic sensor has been proposed for the wind turbine blade de-icing system. A glass fiber reinforced composite sheet with an implanted transmissive fiber optic sensor was stacked on top of laminates having a polymer electric heating film, and then hot pressed together to form a heating film heating icing de-icing experimental system. A heating mode with a heating power density of 700 W/m² was used to heat and de-ice a low-temperature icing board at -15°C, while monitoring the optical power changes of the transmissive fiber optic sensor. The experimental results show that when heated for 1 minute, the optical power of the transmissive fiber optic sensor changes significantly, effectively characterizing the melting state of the ice bottom.

To address the complexities and incompleteness of manual electroencephalogram (EEG) feature extraction, this paper proposes a fatigue detection method based on a multi-feature fusion convolutional neural network with an attention mechanism. This approach aims to achieve more accurate and comprehensive fatigue detection and analysis. The EEG data from the SEED-VIG dataset is denoised and segmented, with 4D features extracted and generated. These features are then input into spatial, frequency, and temporal attention modules for fusion, resulting in the output of final fatigue detection results. Testing on SEED-VIG samples demonstrates that the proposed algorithm achieves a correct recognition rate of 82.21%, thus validating its effectiveness.

In the context of the sustained and rapid development of the global textile industry, accurate detection of fiber composition is of great significance, which is not only related to the product quality of enterprises and the rights and interests of consumers, but also a key link to promote industrial innovation and development. Facing the increasing demand for textile inspection, the single chemical dissolution method or near infrared spectroscopy analysis method has been widely used in inspection practice, but it often presents certain limitations in inspection efficiency and accuracy of inspection results. In this regard, based on the actual project requirements, this paper will deeply analyze the feasibility of the application of artificial intelligence technology and deep neural network in this field, and combine the convolution neural network (CNN) and near infrared spectroscopy analysis technology to construct a brand-new detection method of textile fiber compositions, so as to realize the rapid and accurate detection of textile fiber compositions. Practice has proved that the collected near infrared spectroscopy data of textile samples are input into CNN network to complete feature recognition and pattern learning, and the results are predicted and output by a strong classifier, which realizes rapid and accurate identification and quantitative analysis of different fiber compositions in textiles, providing strong technical support for the high-quality development of textile industry.

For major oil fields under PetroChina, the wellhead equipment is usually abundant and diverse, selected based on the depth of the reservoirs and the extraction processes. During service, wellhead equipment is prone to erosion and corrosion, which can lead to safety risks such as leaks. Therefore, corrosion detection of high-risk wellhead equipment in service is crucial for ensuring the safe operation of these devices. Considering the operating conditions and structural characteristics of wellhead equipment, the applicability of various non-destructive testing methods was compared, and a defect detection scheme for wellhead equipment, primarily based on phased array ultrasonic testing(PAUT), was determined. Using PAUT technology, an online inspection was conducted on the wellhead equipment of 10 oil and gas wells in the Keshen block of Tarim Oilfield, which were prone to corrosion. It was found that in two of these wells, the passage and flange end surface of the production wing connection short section were both corroded, with the corrosion groove having a depth of approximately 5.0 mm and a width of approximately 50 mm. The on-site inspection results validated the feasibility of applying PAUT for online inspection of wellhead equipment.

This study investigates the temperature measurement accuracy of thermostatic metal bath, aiming to improve their precision through theoretical analysis and experimental validation. Based on the heat transfer model, the effects of sensor thermal conductivity, contact area and heat transfer path on measurement accuracy were analysed. Experiments were conducted to evaluate the performance of PT100 temperature sensing elements with various materials and structures. Results indicate that measurement accuracy is closely related to the sensor’s thermal properties and structural design. The optimized sensing element, featuring a copper circular head, ultra-high thermal conductivity potting compound, and a low thermal conductivity PEEK support tube, achieved remarkably low error rates of 0.07% and 0.08% at set temperatures of 125°C and 165°C, respectively. This research provides both theoretical basis and practical guidance for enhancing the accuracy of dry thermostatic metal bath calibrators.

Quantifying internal crack detection signals in steel pipelines requires a high level of expertise from researchers. The traditional signal features cannot fully characterize the actual signals. This paper uses steel plate specimens to simulate steel pipelines for crack detection signal quantification research. Firstly, a differential eddy current test platform based on incremental permeability extraction is built, and the self-developed eddy current detection technology is used to detect the crack defects and form a quantitative database of crack defects; then, end-to-end crack detection signal quantification models DRSN1d and DRSN2d are established; finally, noise is added to the database to compare the traditional model with DRSN2d. The results show that the constructed deep learning model achieves an average crack depth inversion accuracy of ±0.018mm and an average crack width inversion accuracy of ±0.015mm, which meets the industrial requirements. The deep learning quantization model outperforms the traditional machine learning model on high-noise data, and the features formed from the end-to-end model training are also better than the traditional features.

Based on the inverse square relationship between nuclear radiation intensity and distance from a radioactive source, this study explores the feasibility of using a single detector to calculate and quickly locate the location of the radioactive source based on its travel path position and counting rate, facilitating the source finding operation of lightweight unmanned aerial vehicles and robots; In order to reduce the influence of background and other radioactive substances, a simple and convenient pulse width spectrum is used to obtained the photoelectric peak count of the target radiation source.

In the practical application of the vector network analyzer (VNA), the thru-reflect-match (TRM) calibration kits exhibit non-ideal performance due to the inevitable hardware damage and imperfect manufacture of precision devices. Consequently, the S-parameters of the tested devices deviate from their ideal values, leading to uncertainty in the measurements. This paper provides a detailed analysis of how non-ideal TRM calibration kits affect the uncertainty of Sparameter measurements. To start, we simplify error correction equations through the utilization of the concept of general nodal equations and matrix operations, culminating in a generalized equation describing the deviation of measured Sparameters relative to error terms. On this basis, expressions for the S-parameter sensitivity coefficient using TRM calibration technology are derived. Finally, the accuracy of these expressions is verified through comparative experiments. The research presented in this paper can determine the influence degree of various errors on S-parameters, thus optimizing the calibration process and providing guidance for quality control.

In order to explore whether the infrared imaging characteristics of zero-value insulator string can be used as the basis for proposing field detection methods, we analyzed the influence of different humidity and contamination conditions and different positions of zero-value insulator on the string on infrared thermal image characteristics through experimental research. The research shows that the temperature rise of different positions on the surface of porcelain insulators is closely related to the distance of the center axis of rotation, and the two are negatively correlated. The maximum temperature rise point is at the point where the iron foot of the insulator meets its lower surface. The position of the zero-value insulator in the string also has an effect on the temperature distribution. The closer the position is to the high-voltage or low-voltage side, the greater the temperature difference between the normal insulator and the zero-value insulator. The ambient humidity and surface contamination of the insulator will aggravate the overall temperature appreciation of the insulator, and the impact on the temperature difference between the zero-value insulator and the adjacent normal insulator is more obvious. According to the above research results, a zero-value insulator detection method based on infrared imaging characteristics can be proposed, and its accuracy is verified by examples under the corresponding detection conditions.

With the development of intelligent technology, the textile industry has introduced advanced sensors, control systems, and data analysis technologies to achieve more precise control operations, improve production efficiency, reduce manual labor, lower production costs, and enhance product quality and stability. Yarn tension control directly affects the production efficiency and product quality in the textile industry. Based on recent research progress in tension control, this article reviews the advancements in yarn tension sensors, tension detection, and control technologies. It summarizes the current development trends in yarn tension control technology and points out that the integrated research on the application of new sensor technologies, artificial intelligence technology, intelligent control, and novel control algorithms in tension control is crucial for achieving stable control and driving the continuous progress of tension control technology.

When handling with detection in driving distraction and road rage, the object detection model using visible light like image often suffers from the difficulty of environmental disturbance while the detection model using EEG signal often faces the difficulty of inconspicuous classification and inconvenient collection. In this paper, a novel and improved model fusing visible light and EEG signals is proposed, which the ECA-YOLOv8 model based on the ECANet is used for deep learning to identify the current state of the driver for the visible light images and next the EEG signal spectrum analysis technology is used to identify it again if the output threshold is lower than 0.6. Additionally, a simple and practical detection system is set up. Compared with single source detection, fusing visible light and EEG signals shows an effective increase in accuracy and stability.

In this paper, the study focuses on the evolution mechanism underlying stress concentration formation in oil and gas transmission pipelines when subjected to various factors such as medium internal pressure and material property deterioration. It seeks to explore the relationship between the law of existing magnetism and stress changes induced by geomagnetism within these pipelines, as well as to develop a stress detection methodology. Theoretical models are compared and analyzed for the magnetic behavior under both unidirectional and complex stress conditions within oil and gas pipelines. Simulations are performed to assess the magnetic signal distribution across the pipeline wall, yielding insights into the behavior concerning the change law of magnetic flux signals under complex stress. In addition, a test platform is constructed to facilitate stress and magnetic flux testing on a model of an X80 pipeline exhibiting varying degrees of damage. These experiments aim to validate the theoretical stress detection model based on observed magnetic field changes. The results demonstrate a positive correlation between the existing magnetic flux signal and the applied stress, with a roughly linear relationship. This research establishes the feasibility of utilizing the existing magnetic flux signal to detect stress distribution within oil and gas pipelines.

Ultrasonic testing is a commonly used non-destructive testing technology that utilizes the propagation characteristics of high-frequency sound waves in materials to detect and evaluate their internal structure, defects, and integrity. Due to its high sensitivity and non-destructive nature, ultrasonic testing is suitable for detecting defects in cylinders. This paper designs a defect detection method for CNG cylinders based on the ultrasonic pulse echo method and HRNet. First, a comprehensive scan of the CNG cylinder is conducted using the ultrasonic pulse echo method. After obtaining the scan results, the Kalman filtering method is used to remove noise interference, resulting in a more accurate waveform. An improved HRNet is then utilized for defect target detection. The main improvement of HRNet lies in using global and local feature fusion to achieve higher precision defect identification and introducing the ASPP module to enhance the network's feature extraction and anti-interference capabilities. In experiments, the method can effectively identify groove-type defects and corrosion defects in CNG cylinders.

The water content in stable crude oil emulsions has been widely researched. However, the emulsion in the flowing conditions has higher emulsified water content, which affecting the flow characteristics of oil-water two-phase in gathering lines. This study focusses on exploring the electrical characteristics associated with water content of the flowing emulsions. To describe the impedance characteristics of the flowing emulsions, a RC equivalent circuit model was established basing on the frequency dispersion phenomenon due to interfacial polarization of emulsions. A stirring system with different rotating rates was developed to simulate flowing conditions of the emulsion. Emulsions with varied amount of emulsified water content (0~0.9) were prepared and their impedance spectroscopy were measured with frequency from 50Hz to 1MHz.The experimental results demonstrated that within the observed range, the frequency of interfacial polarization in impedance exhibited a linear relationship with the water content of the crude oil, in strong agreement with the established theoretical model. The principle of linearity can be used to measure the water content of flowing crude oil.

Based on the GB38383-2019 dishwasher energy efficiency and water efficiency limit value and the method of limiting the power consumption of the working cycle and the energy consumption of the dishwasher in the grade, as well as the method of limiting the water consumption and dishwasher water efficiency limit value of the working cycle, refer to GB/T 20290- 2016 In the performance test method of household electric dishwashers, it is required to measure the power consumption, water consumption, etc. of the dishwasher throughout the whole washing process. In view of the above, the problems existing in the detection of dishwasher products at this stage are introduced, and a design of control detection system based on Siemens 200 smart PLC and configuration king host computer is proposed, including system hardware design, control logic design, data storage and analysis design, The process and control requirements of the detection and control control system are analyzed in detail, and the control of PLC and the monitoring of configuration king are realized through the final online debugging, and the use value of the detection system is large, and it has a good detection application prospect.

Ship intelligence has been the main trend of current ship development, the main research of this paper is the ship in the berthing process, the use of AIS (Automatic Identification System) system and other technologies composed of intelligent berthing safety monitoring technology, the first collection of berthing environment-related data of the berthing database as well as the ship - shore and ship - ship interaction between the connection, in order to improve the safety of ship berthing berth. Smart anchorage safety monitoring technology research based on vessel network is through the AIS system to obtain the vessel anchorage process around the environment and anchorage area information for analysis and assessment, the use of database analysis technology on the collected data for comparison and analysis, timely detection and elimination of potential safety risks, to achieve the optimum anchorage path of the vessel, to achieve the smart anchorage safety technology, and to promote the smart development of vessels.

This paper discusses the current status of the application of big data analytics technology in industrial intelligent inspection system as well as the challenges and opportunities. Through combing the related literature, it summarizes three major application scenarios of big data analytics in intelligent inspection system: equipment fault diagnosis, production process monitoring and efficiency management, and safety risk identification and early warning, which provide strong support for the safety, efficiency and quality of industrial production. In addition, we summarize the key technologies applied in this field, including machine learning and deep learning, big data analysis and processing technology, as well as the Internet of Things and sensor technology. The application of these key technologies effectively enhances the intelligent level and practical value of the intelligent inspection system.

In this paper, the AE signal collected in the acoustic emission experiment of the mural collection is taken as the research object. The collected continuous mural state data are analyzed and processed, and the abnormal state information of mural cracks and falling off is extracted to obtain the sample set of disease characteristics. Wavelet transform was used to convert the collected AE signal of crack damage into a two-dimensional time-frequency image. Finally, deep reinforcement learning DQN network was used to identify the damage type. Compared with convolutional neural network, the test results showed that the damage recognition accuracy rate could reach 98%.

In order to achieve the possibility and probability of discovering the violation of operation tasks through previous data, this paper proposes an intelligent identification algorithm for safety risk of transmission line maintenance operations based on deep learning. Firstly, a large amount of transmission line maintenance operation image data is collected, secondly, the size of image chunks is determined by using a priori information and experiments to reduce the number of layers of convolution and reduce the computing time. Finally, a standard library of image samples of transmission lines is successfully constructed, which achieves the standardised collection of inspection defect samples with key features of multi-source data and the standardised management of transmission equipment information, unified labelling transformation and efficient storage and retrieval of defect information. The experimental results show that real-time monitoring and early warning of potential safety risks in transmission line maintenance operations can be achieved through the deep learning-based intelligent identification algorithm for safety risks in transmission line maintenance operations, thus improving the safety and efficiency of maintenance operations.

Aiming at the problems of traditional manual infusion system, such as time consuming and low safety, and lack of medical staff, a transfusion monitoring system based on LabVIEW and contact switch was designed. The real-time data of infusion volume was collected by using the contact switch, LabVIEW software was used to analyze and process the liquid drip rate. The system can realize the functions of infusion speed monitoring, displaying the residual volume of the infusion bottle, infusion remaining time prediction, low volume alarm and so on. The proposed system has the advantages of simple structure, convenient operation, accurate results, good reproducibility and intuitive display, which is convenient for medical personnel to analyze and deal the infusion process, effectively improve the working efficiency of medical staff and reduce the occurrence of abnormal situations. The method provides an effective way for the monitoring of intravenous infusion and the automation of medical care.

There is still room for improvement in the calibration process of digital acquisition devices. Firstly, the specific indicators still use the method of quantification based on the critical points of intervals, which lacks rigor. Secondly, some test results rely on manual interpretation, which inevitably introduces measurement errors and seriously affects the accuracy of the evaluation results. Therefore, this study introduces more performance indicators that are recognized by the testing industry and can be quantified digitally, proposing a more scientific testing method that is also compatible with current metrological testing standards. This method achieves standardization and comprehensiveness of the performance evaluation criteria for acquisition devices, thereby ensuring the accuracy and credibility of the evaluation results.

With the expansion of head-mounted displays into various professional domains, there is an increasing demand for adaptive complex illumination image enhancement algorithms that are computationally efficient and resource-conservative. This study introduces a method to categorize input images into high-light, normal, and low-light images based on brightness thresholds. Enhanced Retinex-based algorithms are proposed to process these categories. For low-light images, operations such as histogram equalization and sharpening are applied to enhance the details of the illumination component. For highlight images, an illumination component estimation method is utilized to effectively reduce noise and enhance contour information, followed by normalization using a sigmoid function. The effectiveness of the low-light enhancement algorithm is validated using the LOL dataset. The high-light enhancement algorithm is validated using a self-constructed dataset.

In this work, an intelligent monitoring method based on target detection and pose estimation is proposed for personal protective equipment monitoring and deep analysis of safety behavior. Different from previous image processing methods, this work uses object detection algorithm to calculate the types and positions of operators and security tools in image data. The pose estimation algorithm is introduced to calculate the key points of the human skeleton in the image data and identify the human pose, and then the relative position relationship of the key point coordinates is compared to analyze and judge the pose of the operator. The experimental results show that the intelligent monitoring method proposed in this work can improve the safety protection effect of operators to the greatest extent on the premise of ensuring the smooth completion of operations.

In order to improve the performance of image recognition, the traditional YOLOv5s network was improved and applied to wheat pest recognition. Based on the analysis of YOLOv5s, the activation function, attention module, feature fusion network and loss function are improved. By comparing YOLOv5s with other image recognition technologies, it is pointed out that the improved YOLOv5s network has good performance in three aspects: accuracy, recall rate and average accuracy. Compared with other target recognition algorithms, the improved YOLOv5s network can be better applied to target recognition in complex scenes containing small targets. It has certain practical value for improving the efficiency of wheat pest detection and realizing the transformation from traditional agriculture to modern agriculture.

Intelligent robots rely on machine vision technology to accomplish tasks in industrial production environments. As one of the important technologies in robotic the vision of the robot arm, the target detection model of the robot arm can accurately detect the position of the target object and guide the robotic arm to grab the object. However, the detection effect of the model often depends on the quality of the dataset, and it is difficult for industrial robots to collect effective datasets when performing tasks, which makes the robotic arm face the great challenge of inaccurate localization when grasping the target object. This paper presents a method for generating a synthetic dataset specifically designed for training robotic arms in grasping tasks. The method combined digital twins technology to generate a substantial dataset for robotic arm gripping. The target identification model trained by the synthetic dataset can precisely identify the real target object. It then converts the object's coordinates into the robotic arm's coordinate system in order to guide the arm to successfully do the grasping job. The experimental results demonstrate that synthetic datasets can efficiently substitute manually annotated real datasets in the context of the robotic arm's grasping and object localization tasks. This method leads to a reduction in the development cycle of the robotic arm vision system and a decrease in the associated costs of development.

Ground Penetrating Radar is widely used in non-destructive pavement engineering. However, its data interpretation requires human experience and considerable time. To address this issue, a void recognition method was proposed to accurate identify voids area based on GPR time-frequency features and sparse autoencoder (SAE). Datasets was constructed by the field tests. After standardization and resampling of the pre-processed GPR signals, 18 time-domain and 12 frequency-domain features representing voids were extracted. These feature parameters were used as inputs to train the sparse autoencoder, and the performance was compared with the multi-layer perceptron (MLP) and support vector machine (SVM) models on the same dataset. The results showed that SAE achieved the highest recognition performance, with an accuracy of SAE (90.75%) > SVM (90.08%) > MLP (53.65%). The SAE model can achieve automatic recognition of onsite data, which provides a basis for accurate maintenance of pavement voids.

Optical imaging of space targets using ground-based or space-based telescopes is typically affected by complex noise. Due to the sparse features and limited data, the denoising performance of star images is often suboptimal. In this paper, we propose a lightweight zero-shot star image denoising framework featuring an improved 3-layer U-Net backbone, which can efficiently complete the denoising task without a complete dataset. This network extracts feature information through two pair of down-sampling and up-sampling layers, as well as several convolution modules. The spatial attention module is employed to focus on attention regions, enhancing the model efficiency and generalization ability. In the experiments conducted with real star images, the denoising pipeline primarily consists of three steps: image preprocessing, network training and inference. The results demonstrate that our method effectively removes noise from star images and outperforms existing techniques, facilitating the accurate detection and extraction of space targets in subsequent research.

This paper presents the design and implementation of an intelligent car system based on Raspberry Pi CM4, focusing on the camera module in the intelligent car. The camera module incorporates functionalities such as camera calibration, visual line following, color tracking, facial recognition, and label recognition. Utilizing image acquisition and processing techniques, this module can accurately perform various visual tasks. The system utilizes hardware components including the Raspberry Pi CM4 core board, relevant expansion boards, and corresponding visual processing equipment. The software portion is developed based on the Raspberry Pi Linux operating system, establishing a development environment and control interface suitable for the camera module's functions. System demonstrations and tests validate the stability and reliability of the camera module in autonomous driving and visual recognition tasks.

To overcome the challenges associated with debris search during space launch missions, this paper proposes a method for rapid aerial search utilizing UAVs (Unmanned Aerial Vehicles). By thoroughly analyzing the specific requirements and characteristics of space search missions, an optimal search strategy is determined through a pre-assessment of relevant parameters using pattern control. The proposed method integrates both hierarchical and holistic approaches, employing mathematical abstraction to develop a unified model that encompasses flight trajectory and gimbal control. This model is then solved to derive a generalized planning method for UAV operations in space search missions.

Action recognition and scoring have important application value in health monitoring, sports analysis, and physical education. With the development of wearable devices and sensor technology, sensor sequences based action recognition and scoring have become a hot research topic. However, the complexity and individual differences of action sequences require recognition models to have high generalization. In addition, the key features in action sensor sequence data are often sparse. It is challenging to combine the global features and local details for collaborative analysis. To address these challenges, a novel action recognition and scoring method CA-TimeMixer based on contrastive learning and TimeMixer is proposed in this paper. In terms of model architecture, the Past Decomposable Mixing (PDM) of TimeMixer is employed for feature extraction. Cross Attention (CA) is introduced to fuse different scaled features of the sensor sequences, thereby enhancing the model’s ability to extract local-global features. A supervised contrastive loss is adopted to perform contrastive learning on sequence samples of different actions for action recognition. During inference, the feature similarity between the test sample and the benchmark action sequence is computed to determine whether the category of test sample is the same with the benchmark action. For the action scoring, the test sequence and the benchmark sequence are input into CA-TimeMixer. The difference tensor of these two features predicts a score with a linear layer. Comparative experiments and ablation studies were conducted on the collected Taichi action sequence dataset. Compared with the benchmark models, the proposed CA-TimeMixer achieves Accuracy of 94.7% in action recognition and Mean Absolute Error (MAE) of 0.36 in action scoring. Comprehensive experimental analysis demonstrates the superiority of the proposed method.

The primary methods for oil and gas pipeline defect identification currently rely on acoustic and magnetic techniques, with visual solutions remaining scarce. However, recent artificial intelligence (AI) advancements suggest promising avenues for real-time visual defect detection. This research explores the capabilities of AI for real-time detection and aims to refine and optimize this methodology. We achieve this by implementing widely utilized deep learning neural networks, such as YOLO and DETR, for training and processing on-site video images. Our model demonstrates a recognition accuracy exceeding 87% through subsequent application testing, indicating significant potential for real-world implementation in oil and gas pipeline defect detection.

To implement the fuzzy control algorithm in a microcontroller, a fuzzy control system for an automatic washing machine was established in the MATLAB/Simulink environment. Based on this fuzzy control system, a data extraction model was first established, and the data extraction algorithm was written using an M-file to obtain a two-dimensional array. Then, an RTW automatic code generation model was built, and the obtained two-dimensional array was imported into the 2-D Lookup Table block of the model. Using the RTW function in MATLAB, automatic code was generated, and the two-dimensional lookup interpolation algorithm in the code was analyzed. Finally, the algorithm was written into the microcontroller and experimentally verified. The results showed that the operational outcomes of the microcontroller were basically consistent with the offline simulation results of the model, and the designed fuzzy controller met the real-time and accuracy requirements of the control.

The in-motion initial alignment of the strapdown inertial navigation system (SINS) assisted by the Doppler Velocity Log (DVL) plays a significant role in the efficient operation of the autonomous underwater vehicle (AUV). In the case of severe maneuvering or low accuracy of inertial sensors, the output error of the gyroscope gradually increases, which leads to a decrease in the initial alignment accuracy of the inertial navigation system. In this paper, an in-motion alignment method for DVL-assisted SINS is proposed, which eliminates the effect of attitude matrix variations on the observation noise by establishing an accurate filtering model for the constructed inertial vectors. Experimental results verify that the proposed algorithm can accurately estimate the attitude matrix, and the initial alignment accuracy is improved as compared with existing algorithms.

With the progression of computer and automation technologies, alongside advancements in the shipbuilding industry, the automation level in ship manufacturing has been steadily increasing. This paper addresses the issue of unclear textures and uneven lighting in the ship bottom environment, which prevent the ORB-SLAM2 algorithm from extracting an adequate number of feature points. To address this issue, this paper proposes a method that combines point and line features based on the LSD algorithm. When the number of feature points is insufficient, line features are extracted to replace point features, and feature matching computations are performed exclusively in the mapping thread. Finally, a simulated model of the real shipyard environment and the robot was created using Gazebo software, and comparative validation demonstrated that the improved algorithm proposed in this paper achieves higher accuracy than ORB-SLAM2.

To achieve automated sorting and transportation of workpieces in the factory, AMR is used to assist in workpiece handling. This paper proposes a two-stage Workpiece Recognition Algorithm based on Mask-RCNN and template matching for determining the processing stage of workpieces. First, Mask R-CNN is used for instance segmentation and reverse masking to obtain an image of the target workpiece, which serves as the image to be matched. Relying on the three-view angle of the workpiece's external shape during the processing stage, a template image is created, and SIFT-FLANN is used for image matching to identify the processing stage. Finally, experiments show that the method achieves an accuracy rate of 81.27% for recognizing categories of workpieces, with an average matching time is close to 20ms for identifying process stages of disk and sleeve-type workpieces, and a matching accuracy of 98.75%, allows for accurate determination of workpiece type and processing stage. This provides a basis for AMR to transport workpieces and contributes to improving factory production efficiency and reducing production costs.

For the first time, the Luotuoqiao Station of Ningbo Metro Line 3 Phase II Project has achieved an intelligent power distribution system that combines virtual and reality in the lighting distribution room of domestic rail transit. In the field of view of the helmet worn by a person, there is an image of a real-world power distribution box, and virtual objects are overlaid on top of the image. The overlaid objects can interact with the real-world power distribution box scene in an immersive manner. This article mainly elaborates on the implementation of the metaverse intelligent power distribution system.

The energy management system of Ningbo Rail Transit is based on a cloud platform, which constructs a traction side "supply use" collaborative integration technology based on bidirectional variable current traction power supply technology, permanent magnet synchronization technology, and dedicated track return current technology. It explores a new energy-saving mode for stations mainly based on ventilation and air conditioning energy-saving technology and DC lighting technology, studies the energy consumption data collection and analysis system based on the cloud platform, establishes an energy consumption standard index system covering various equipment systems and production and operation links of rail transit, and creates a national energy model line that is resource saving, green, low-carbon, intelligent and efficient.

Automated driving is in the stage of development and industry standardization. In order to ensure the safety of autonomous driving, intervention by the driver is required at any time to address unexpected situations during the autonomous driving process. Due to the fatigue and other factors, drivers are often unable to intervene in time to respond to unexpected situations, and autonomous driving accidents occur from time to time. Some companies have already installed DMS in their vehicles, however, many physiological conditions caused by fatigue vary from person to person, and there is still a need for a more reliable and cost-effective method to monitor the driver's condition. Based on the study of a large number of driver monitoring systems (DMS), we propose the use of image-based photoplethysmography (iPPG) for driver physiological condition monitoring, which can effectively solve most of the pain points of the DMS. In this paper, we present an innovative dual-camera measurement system designed to tackle measurement errors arising from environmental variability and motion artifacts. This dual-camera configuration aims to improve the accuracy of physiological signal monitoring, thereby providing more reliable assessments of the driver's condition during autonomous driving.

In modern industry, construction industry, military industry and other fields, there are large containers or complex crossing pipelines, the weld at the junction will be defective due to improper operation during welding, or the corrosion of the weld in the working environment for a long time. Therefore, the detection of weld defects is crucial. Traditional manual detection is not only time-consuming and laborious, but also easy to bring harm to workers' health. With the rise of machine vision technology and image processing technology , robot instead of manual welding seam inspection has been applied in many occasions. On the one hand, the wall-climbing robot with welding seam recognition ability has high recognition accuracy, and it is not easy to miss detection. On the other hand, compared with manual detection, the welding seam detection robot has low cost and high efficiency. Therefore, this kind of wall-climbing robot with the ability of weld detection and recognition has been widely recognized.

In response to the challenges associated with detecting occluded and small targets in automatic driving scenarios, as well as the issues of low detection accuracy and a high miss rate caused by complex background interference, an enhanced road target detection algorithm based on YOLOv8 is proposed. A plural lightweight convolutional module (PLConv) is devised to replace the C2f module with the PL-C3 module, thereby reducing network parameters and enhancing the network's feature extraction capabilities. Moreover, a P2 small target detection head is integrated at the model's apex, facilitating improved extraction of shallow features and enhancing the model's performance in detecting small targets. Experimental results demonstrate that the enhanced model has seen increases of 5.8%, 5.2%, and 3.5% in mAP50, mAP50-90, and Recall respectively, thus better aligning with the demands of object detection tasks in automatic driving scenarios.

This study explores the application of SpectralNet in soil structural characterization based on the SpectralNet method, aiming at the effective identification and clustering of soil samples by this algorithm. SpectralNet is a novel method that combines deep learning and spectral clustering and achieves the clustering objective by learning a low-dimensional representation of the data. In this paper, we first introduce the preprocessing process of soil sample data, then describe in detail the working principle of the SpectralNet algorithm and its application in soil clustering, and analyze the parameter sensitivity of SpectralNet in detail. The experimental results show that compared with traditional clustering methods, SpectralNet-based soil clustering can effectively process soil data and improve the interpretability of clustering results while maintaining high clustering accuracy. This study provides a new data analysis tool for the field of soil science and provides reference and inspiration for future research and application in related fields.

This paper focuses on the research of low-power consumption control for wireless sensor networks in coal mine roof and floor monitoring. Considering the characteristics of coal mine roof and floor sensors, which require long-term continuous operation and rely on battery power and wireless transmission, low-power design becomes crucial for extending their operational lifespan, reducing battery consumption, and lowering maintenance costs. Therefore, a low-power wireless sensor network for coal mine roof and floor monitoring is designed. This network integrates LoRa wireless transmission technology and WaveMesh network, enabling self-organizing networks, relay transmission, and long-distance data transmission for sensor nodes, effectively meeting the demands of monitoring the roof and floor conditions within coal mine tunnels. By optimizing low-power control strategies, the energy efficiency of the entire sensor network is enhanced, providing reliable technical support for coal mine safety production.

The drilling grouting pile, with its mature construction technology, high bearing capacity, and wide range of applications, has been widely used in the foundation of road, railway, and bridge structures. Nonetheless, improper placement of the concrete injection interface within the drilling hole can easily cause issues like over-injection and other engineering difficulties, leading to wasted materials and higher expenses. The traditional method of judge cement covering using a rope and a weighty mallet is both time-overwhelming and labor-intensive, creation it difficult to achieve accurate consequence. To direction this topic, this newspaper current a novel technology for identify the cement vaccination connection, which leverages a multi-sensor data acquisition cloud platform in conjunction with the PSO-GA-BP neural network approach. By collecting and analyzing data on turbidity, conductivity, and pH levels, the PSO-GA-BP neural network model effectively identifies the injection interface, determines its position and height, and facilitates intelligent monitoring of the concrete injection process. This innovation effectively resolves challenges associated with both over-injection and under-injection.