In our project, a novel gear health monitoring system called a “smart gear sensor system” has been being developed. This system has a smart gear sensor and a monitoring antenna. The smart gear sensor has been created by using a laser sintering technique that allows a conductive ink to be sintered directly on the gear surface to form a sensor chain. This technique is also used to form an antenna pattern of the monitoring antenna. Experimentally, when the smart gear with the integrated sensor chain is brought parallelly close to the monitoring antenna while this antenna is connected to a network analyzer, a magnetic coupling happens. This wireless magnetic coupling is presented via the return loss signal gained by the network analyzer. Moreover, the conductivity of the sensor chain on the smart gear has been proven to affect the magnetically coupled return loss signal. Specifically, the shape of the return loss signal will be changed correspondingly. Since the sensor chain is sintered directly on the surface of the gear, the physical characteristics of the gear such as healthy or cracked will result in the quality of the sensor conductivity. Therefore, the smart gear's healthy or cracked characteristics can be observed via the return loss signal chart obtained by the network analyzer. In other words, the return loss signal can be considered essential data for the gear health monitoring process. Nevertheless, because the temperature will have a certain influence on the physical state of not only the gear but also the sintered sensor, the evaluation of the temperature to the magnetic coupling and its return loss signal becomes unavoidable. As a result, this study focuses majorly on the consideration of the influence of the temperature so that the accuracy of the proposed gear health monitoring method can be improved. The final experimental result and analysis demonstrate that the temperature also significantly impacts the return loss of the smart gear system.
This research evaluates the influence of phase fluctuation of a high-speed smart gear on the return loss of monitoring antenna in our developing “smart gear sensor system”. This smart health monitoring system of gear comprising of an antenna connected to a network analyer and another similar antenna integrating crack detection sensors directly printed on a gear. The principle of this system is based on magnetic coupling phenomenon between samples of antenna and crack detection sensors with antenna printed on polyacetal (POM) plates so that the characteristics of the gear side elements can be obtained wirelessly. In this paper, the effect of the relative phase fluctuation when the smart gear is operating at high speed on this coupling signal perhaps is considered. Practically, a special experimental rig consists of a motor with a speed control button, an antenna on a polyacetal (POM) plate with a radio frequency connector, and a smart gear with integrated sensors antenna was created. Thanks to this test rig, the smart gear is able to rotate with variant rotational velocities in range of 500 rpm to 5000 rpm while keeping pairing concentrically with the fixed monitoring antenna at a constant distance. Then, return losses received from this antenna via the network analyzer then are recorded at each specific speed. In comparison with the return losses gained in case of similar experimental setting but in stationary state, the shape form of the high-speed phase fluctuation return loss is indicated.
Recently, data-driven machine health monitoring has become more popular due to the wide-spread deployment of lowcost sensors and deep learning algorithms’ achievements. The detection of failures of machines can be determined based on failure classification results using deep learning architectures. On this tendency, we constructed a plastic gear failure detection structure using a convolutional neural network. In this study, raw vibration data was converted to frequencydomain data. Amplitudes of frequencies in the monitored frequency band were transferred into images, which then were labeled as crack or non-crack by a high-speed camera. Although deep learning architectures have great potential to automatically learn from complex features of input data, the high-amplitude frequencies reflecting the main vibration causes such as gear meshing frequency and its harmonics or shaft frequency affect the accuracy of learning. Besides, the low-amplitude frequencies in a low-frequency band, which are sensitive to gear failures, show efficiency in early failure signs of the plastic gear. Thus, this paper proposed an image visualization and labeling method by focusing on lowamplitude frequency features in the low-frequency band and lessening high-amplitude frequency features. The results show that the proposed system learning from new visualized images can detect plastic gear’s early failure situation before the initial crack happened.
Nowadays, deep learning (DL) has become a rapidly growing and provides useful tools for processing and analyzing big machinery data. Many research projects achieved success in failure classification from machinery data using convolutional neural networks (CNNs), one of the most extensive study aspects of DL. On this trend, we constructed a crack detection system of POM (Polyoxymethylene) gears using a deep convolutional neural network (DCNN). In our work, vibration data collected from plastic gears was visualized and labelled as crack data or non-crack images. A DCNN based on pre-trained VGG16, which firstly pre-learned from ImageNet’s data and then re-learned from the labelled images, is utilized to classify crack or non-crack situations of plastic gears. In this case of study, the image quality distortions of the dataset such as blur, noise or contrast are stable and do not affect the performance of the DCNN. However, the image size, which keep a vital role to reach high performance of the detection system, has been unknown. Hence, this paper reveals an optimized size of images created from vibration data for high-accuracy of learning.
This paper shows crack detection systems based on deep neural networks, which analyze meshing vibration of plastic gears. A gear operating test rig has an acceleration sensor attached on a bearing housing and a high-speed camera. The meshing vibration of plastic gears during operation was measured and teeth images that enable us to decide whether cracks exists were captured. After transferring the meshing vibration data in the time domain to the frequency domain by FFT, the amplitude and phase information of the meshing vibration was converted to image data. According to the images from the high-speed camera, the imaged vibration data were separated to two classes, with or without crack, as the training data for deep neural networks. Furthermore, two convolutional neural networks, 4 layers and 16 layers were constructed for classification of crack existence or non-existence, and the systems were learned from the labeled data set. In the training, the random weighting functions of the convolution were prepared, and the number of images were 350 and the number of epoch was 125. The learning of the 4 layers convolutional neural network was finished appropriately, however, the learning of the 16 layers convolutional neural network did not progress at all. Then, the transfer learning method was used for the 16 layers convolutional neural network. The transfer learning of the 16 layers convolutional neural network was finished appropriately, and the accuracy at 125 learning steps reached to 97.2%.
Health monitoring of rotating machine elements, such as gears, is challenging because of rotation at high speed in gearboxes, geometric complexity, or space limitation for measurements. The long-term objective of the present research is to develop smart sensor systems for detecting gear failure signs. As the very first step, we proposed a new method to manufacture electrical circuits, such as sensors or antennas, on gears. We had begun to develop a 4-axis laser printing system and showed the laser sintering conditions of the conductive ink splayed on steel plates insulated by polyimide layers. In this paper, a crack detection sensor was designed and printed. The printed sensor can monitor the condition of a plastic gear whose module and number of teeth are 1.0 mm and 48. In addition, an antenna designed for the same size gear was printed on a plastic plate, and the frequency property of the antenna was investigated. As a result, the printed antenna had the 1st natural frequency at 0.3GHz. Finally, monitoring experiments was carried out to check the condition of a smart system consisting of the sensor and antenna from the other antenna having the same dimension. As a result of the experiment, the monitoring of the return loss of the external antenna shows the sensor is healthy or not. The sensor and antenna system will allow for the development of better equipment and detection techniques for health monitoring of gears.
Failures detection of rotating machine elements, such as gears, is an important issue. The purpose of this study was to try to solve this issue by printing conductive ink on gears to manufacture condition-monitoring sensors. In this work, three types of crack detection sensor were designed and the sprayed conductive ink was directly sintered on polyimide (PI) - coated polyamide (PA) 66 gears by laser. The result showed that it was possible to produce narrow circuit lines of the conductive ink including Ag by laser sintering technique and the complex shape sensors on the lateral side of the PA66 gears, module 1.0 mm and tooth number 48. A preliminary operation test was carried out for investigation of the function of the sensors. As a result of the test, the sensors printed in this work should be effective for detecting cracks at tooth root of the gears and will allow for the development of better equipment and detection techniques for health monitoring of gears.
Health monitoring methods for machines have been the subject of considerable efforts to maintain it at an appropriate timing. Failures of rotating machine elements can cause severe accidents, thus, to detect such failures is an important issue. However, health monitoring of rotating machine elements, such as gears, is challenging because of rotation at high speed in gearboxes, geometric complexity, space limitation for measurements, or another operation conditions. The long-term objective of the present research is to develop smart sensor systems for detecting gear failure signs. As the very first step, this paper proposes a new method to manufacture electrical circuits, such as sensors or antennas, on gears. We print these circuits directly on the gear surface using a laser sintering technique of conductive ink. For this purpose, we have begun to develop a 4-axis laser printing system. This paper shows the laser sintering conditions of the conductive ink splayed on steel plates insulated by polyimide layers. The conductivity of the printed lines was evaluated through observation with a miniature scanning electron microscope. Finally, according to the obtained laser sintering conditions, a meander line antenna was printed as a part of smart sensor systems.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.