Current Monitoring And Fault Diagnosis Technology For Opening And Closing Coils Of Low Voltage Circuit Breakers

The low-voltage circuit breaker is an important device in the medium-voltage energy feed device,and its operating state is of great significance to the stable operation of the medium-voltage energy feed system.The faults of circuit breakers are mostly concentrated in mechanical defects and electrical defects.The electrical defects are mainly the faults of the secondary circuit of the opening and closing coils.The opening and closing coil current can reflect whether the coil is undervoltage,overvoltage,aging,whether the electromagnet is stuck or not.It is an important parameter for detecting the operating state of the circuit breaker.Therefore,this dissertation intends to monitor the opening and closing current signal to monitor the circuit breaker.Troubleshooting.First,the primary system based on the medium voltage energy feed device collects and processes the signals of the opening and closing coils.The system has a signal acquisition module,a hardware processing module,a transmission module and a software processing module.The hardware processing module includes signal conditioning circuit and data acquisition circuit,and the software processing module includes upper computer and MATLAB software.Secondly,the current characteristics of the opening and closing coils are analyzed and determined.First,the equivalent circuit model of the opening and closing coils is established,and the corresponding operating states of the opening and closing coil current waveforms at different times are analyzed.The typical characteristic quantities are given as local features,and then the mean,variance,skewness,and skewness are included.As a global feature,degree is collectively used as a feature matrix for subsequent recognition.Once again,feature extraction is performed on the current signals of the opening and closing coils.Before feature extraction,the signal is filtered.This paper presents three methods of traditional filtering,five-point triple smoothing,and wavelet threshold denoising,and analyzes their advantages and disadvantages respectively.Then,the maximum and minimum values and the corresponding time points are obtained as local features by the successive difference method of the current waveform,and the proposed global feature parameters are obtained through certain mathematical formulas,which are combined into a comprehensive fault feature matrix.Finally,the comprehensive fault feature matrix is identified and diagnosed.First,perform the principal component dimensionality reduction processing on the fault characteristics,and then use the BP neural network and genetic algorithm optimized BP neural network to identify the fault characteristics,and compare and analyze the advantages and disadvantages of the two methods.The genetic algorithm has achieved good results.Forecast effect.This paper can improve the online monitoring capability of low-voltage circuit breakers,reflect the operating conditions of the circuit breakers in real time,avoid major production accidents,and greatly help the stable operation of the medium-voltage energy feed system.Figures: 44,Tables: 20,References: 50.