Method and system for detecting harmonic current signal strength of traction converter

Abstract

The invention relates to a method for detecting the harmonic current signal strength of a traction converter. The method comprises the steps of acquiring an output current signal of a traction converter; amplifying and filtering the output current signal; performing analog-digital conversion on the output signal to obtain a digital signal; setting the center frequency, passband and cutoff frequency of a digital band-pass filter, and processing the digital signal using digital band-pass filter; and obtaining the RMS of the output signal of the digital band-pass filter by using a sliding windowcalculator, and then obtaining a signal strength value of the traction converter. The method for detecting the harmonic current signal intensity of the traction converter in the invention performs pure time domain signal processing on the signal, and the processing process is simpler. The characteristics of the digital band-pass filter in the filtering process are basically not affected by environmental factors. In addition, the sliding window calculator is used to process the data, the change of the input signal can be tracked in real time, and the defect of failing to respond well to an abrupt signal can be solved.

Description

technical field [0001] The invention relates to the technical field of track circuits, in particular to a method for detecting the strength of the harmonic current signal of a traction converter, and also to a detection system for the strength of the harmonic current signal of the traction converter. Background technique [0002] Traction converters are widely used in rail vehicles. The function of traction converters is to convert the electric energy between DC and AC systems, and to control and regulate various traction motors, so as to control the operation of vehicles. The detection of the harmonic current signal strength of the converter directly affects the normal operation of the vehicle. In the prior art, time-frequency domain signal processing technology is often used to detect the signal strength of the traction converter harmonic current. The main detection methods include detection methods based on analog band-pass or band-stop filters, detection methods based on...

AI Technical Summary

Problems solved by technology

However, the neural network is based on large sample learning, which is greatly affected by samples, which sometimes leads to unsuccessful training.

And affected by the initial value, sometimes it will fall into a local minimum

Moreover, the neural network can only be trained offline and detected online, and its generalization ability is poor.

Images