Aiming at the problem that the inverter of three-level communication power supply in 10 kV power supply system is easily disturbed by noise and the fault diagnosis is difficult,this paper proposes a fault diagnosis method of communication power supply inverter in power supply system based on deep convolution neural network.By collecting and converting the output current signal of the inverter,the time domain form of the current signal is obtained,and the characteristics of the fault signal are extracted.On this basis,based on the theory of deep convolution neural network,the fault diagnosis model of communication power inverter in power supply system is constructed to realize the fault diagnosis of communication power inverter in power supply system.The test shows that the fault diagnosis time of this method is short,and the fault diagnosis result is more accurate.The fault location with zero error can be realized in the 66th iteration,and the application effect is good.
10 kV power supply systemthree-level invertercommunication power supplyfault diagnosistime-domain conversion
NETL
NSTL
万方数据
