220 / 2024-04-28 15:23:18

Current Characteristics Analysis and Defect Evaluation of Switching Coils Based on Virtual-Real Coordination

Switching Coils,Current analysis,synergy between real and virtual reality,Defect Prediction,deep learning

The switching coil is a crucial component of the circuit breaker triggering mechanism. Extracting its current waveform characteristics is an important way to evaluate the defect state of the circuit breaker. However, the machine learning method that uses a large amount of current waveform data has high data requirements and poor interpretability. To address these issues, this paper proposes a method for analyzing the current characteristics and defects of circuit breaker coils based on virtual-real synergy. Firstly, the circuit parameter model is constructed according to the coil loop entity. The virtual-real synergy is then realized using real-zone and virtual-virtual and parameter iteration. Analog and digital mirroring of the normal state and typical fault state of the coil and waveform data acquisition, combined with deep learning for feature analysis and defect evaluation. Taking VSD model typical circuit breaker as an example, the maximum error is less than 5% and the average error is less than 3% in terms of current virtual and real coordination. In terms of defect identification, the average accuracy rate of multiple faults is higher than 96%. The results show that the method based on virtual-real synergy proposed in this paper can achieve accurate circuit breaker feature analysis and defect evaluation under the condition of a small data volume, which can provide a new idea and method for power system operation and maintenance.