FIR filtering differential algorithm applicable to research on transformer vibration signal

Abstract

The present invention discloses a differential algorithm based on an FIR (finite impulse response) filter. The differential algorithm based on the FIR filter is a novel differential computation method designed according to features of an actual transformer vibration signal and an idea of a minimum mean square error; a filter length self-adjusting strategy is adopted; the length of the filter is gradually changed and finally, a differential result is optimized (as shown in figures) by a linear adjusting method; the problem of a large result error span, which is obtained in different frequency ranges, of differential computing results obtained by conventional FIR filtering is solved; and the differential algorithm is applicable to research on the power transformer vibration signal.

Description

technical field [0001] The invention relates to a differential algorithm based on an FIR (finite impulse response) filter, which is suitable for research on vibration signals of power transformers. technical background [0002] Since the advent of digital signal processing methods, the research on digital differentiation has not been interrupted, because differentiation is the basic link of many measurements and applications, and various methods have appeared for different applications. Commonly used are Euler's method, Runge-Kutta method, approximation algorithm based on FIR filter, digital differentiation algorithm based on Fourier transform, Tikhonov regularization method, etc. The anti-interference performance of the above methods is not very good. In the case of white noise, some methods will directly diverge, and some methods are difficult to achieve linear phase, and it is difficult to achieve good results when differentiating complex signals. Later, with the continu...

AI Technical Summary

Problems solved by technology

[0005] The vibration signal of the power transformer collected on site is a speed signal, and now we need to obtain the displacement signal of the vibration through differential calculation. In order to solve the problem that the differential calculation results obtained by traditional FIR filtering have a large error gap in different frequency ranges, the present invention A filter length self-adjustment strategy is adopted, by gradually changing the length of the filter, and finally through a linear adjustment method to optimize the differential result

Images