Research on the Localization Method of Meter Box Leakage Current Combining Data Fusion Technology

With the increasing emphasis on the operational safety of electrical equipment, the detection and localization of leakage current in meter boxes has become an important issue in power system maintenance. This paper proposes a leakage current localization method for meter boxes (DF -CNN) that combines data fusion technology. The core of this method lies in the use of data collected from multiple sensors, which are processed through data fusion algorithms to achieve precise localization of the leakage current. First, we introduced a multi-source data collection mechanism, including current sensors, temperature sensors, and humidity sensors, among others, which can collect multidimensional information about the meter box and its surrounding environment in real-time. By comprehensively processing this information, we can gain a more comprehensive understanding of the operating status of the meter box and potential leakage risks. Second, in data preprocessing and feature extraction, we employed a data fusion algorithm that can integrate data from different sensors, extracting information related to the location of the leakage current. By effectively integrating these different data types, we improve the positioning accuracy. Finally, in terms of algorithm design, we first constructed a leakage current model and used a convolutional neural network to train and classify the features after fusion. Through optimizing algorithm parameters and feature selection, we ultimately achieved rapid localization of the leakage current source. The experimental results show that the positioning method combining data fusion technology has a high positioning accuracy in the detection of leakage current, significantly superior to the existing methods. This research provides new ideas and methods for the fault diagnosis and maintenance of electrical equipment, which is of great significance for improving the safety of the power system.