KEYWORDS: Power consumption, Blockchain, Power supplies, Safety, Computer security, Network architectures, Reliability, Network security, Deep learning, Decision making
In order to meet the power demand of low-voltage users and reduce the power cost, a low-voltage user demand response management platform based on blockchain technology is proposed. Based on the blockchain structure, the overall architecture of the demand response management platform is designed, including the response layer, the main chain layer and the decision-making layer. The platform hardware mainly includes the consensus mechanism module and the orderly power consumption management module of the low-voltage user demand response chain. The platform software part calculates the power supply cost of the power terminal and the power consumption cost of the low-voltage user, and constructs the objective function according to the cost calculation result. The reverse induction algorithm is used to calculate the optimal power consumption of low-voltage users, and the demand impact management problem of low-voltage users is transformed into a single objective solution problem. By solving the objective function, the demand response management of low-voltage users is completed. The experimental results show that the platform can play a good peak shaving effect, and can effectively improve the safety of the platform and the safety of low-voltage users.
Complete and available data is of great significance for improving the theoretical line loss calculation in the low-voltage transformer area. However, with the upgrading of equipment, the electrical data presents the characteristics of small granularity and high complexity, resulting in insufficient line loss calculation accuracy. This paper proposes a missing data reconstruction method based on Kmeans and GBDT combined model. Since it is difficult to unify the reconstruction models of different types of data, the data set is clustered and divided by Kmeans, and GBDT is used for training respectively. During the test, the corresponding GBDT model is used for reconstruction according to the sample category. The simulation results show that the proposed method is suitable for actual data and has higher reconstruction accuracy compared with the traditional mean filling, KNN and decision tree methods. The proposed method can reconstruct the missing of multiple datasets and has good generalization.
In order to promote the digitalization of the power grid, many smart meters are installed on the distribution network side to monitor the operating status of the power grid and equipment information in an all-round way. Frequent data missing phenomenon will lead to misjudgment in the thematic analysis of abnormal electricity consumption in distribution network. In this paper, a missing data reconstruction method based on Improved Bayesian Ridge Regression (IBRR) is proposed. Regularization methods are added to the posterior distribution estimation of parameters to automatically filter redundant information in massive data, thereby avoiding Overfitting phenomenon of maximum likelihood estimation, and improve model training speed and generalization ability. A mutation processing mechanism is proposed for the abnormal fluctuations in the reconstruction results. The results show that, compared with the traditional method, the proposed method has better reconstruction accuracy, and the reconstruction speed is greatly improved.
Situational awareness of electric vehicle charging behavior is an important prerequisite for active distribution network to realize charging demand analysis and controllable resource regulation. However, the subjective difference and randomness of charging behavior greatly affect the accuracy of charging demand analysis. In this regard, this paper proposes a method for predicting the subjective charging behavior of electric vehicles based on deep residual networks. Firstly, the K-means clustering algorithm is used to obtain the non-differentiated typical charging behavior. Secondly, a differentiated subjective behavior characteristic model is constructed with multi-dimensional influencing factors as input and charging behavior characteristics as output of the model. The input and output are correlated, and the non-linear related parts are deep knowledge mined and non-linear curve fit through the deep residual network. Taking a Chinese electric vehicle data set as an example for simulation verification, the results show that the proposed method can effectively distinguish the linear and nonlinear relationships between multi-dimensional factors and charging behavior characteristics, and has high-precision charging behavior prediction capabilities.
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.