Method for accurately filtering high-frequency random noise in vibration signal of pumped storage unit

Abstract

The invention relates to the field of hydroelectric generating sets of electric power systems, and discloses a method for accurately filtering high-frequency random noise in a vibration signal of a pumped storage unit. The method comprises the following steps of: 1) constructing a Hankel matrix of a vibration monitoring signal of the unit, and obtaining a diagonal matrix containing a singular value sequence after singular value decomposition; 2) sequentially inverting signal components associated with a single singular value based on a Hankel matrix construction principle; 3) respectively calculating autocorrelation functions of each signal component and the original vibration monitoring signal, and solving correlation coefficients between the signal components and the original vibration monitoring signal one by one; and 4) screening a threshold value according to a preset correlation coefficient, filtering high-frequency random noise sequence components, and obtaining effective vibration monitoring signal components through reconstruction. According to the method, a pumped storage unit can perform accurate noise reduction on the vibration monitoring signal under a high backgroundnoise environment, a relatively pure vibration signal is obtained, and a reliable data basis is provided for implementing unit state analysis and evaluation and improving the safe and stable operationperformance of the unit.

Description

technical field [0001] The invention relates to the field of hydropower generating sets in electric power systems, in particular to a method for precisely filtering out high-frequency random noises in oscillation signals of pumped storage units. Background technique [0002] With the rapid development of economy and society, the power load increases rapidly, and the peak-to-valley difference continues to increase. The grid has higher requirements for stability, and the insufficient peak-shaving capacity will become a key problem restricting the development of the power system. As a world-recognized flexible and reliable peak-shaving power source, the pumped-storage power station, with its unique operating characteristics of peak-shaving and valley-filling, plays the role of regulating load, promoting energy saving of the power system and maintaining the safe and stable operation of the power system, and has gradually become an effective power system. an indispensable means o...

AI Technical Summary

Problems solved by technology

[0003] Existing signal denoising methods are mainly based on algorithm theories such as blind signal separation, wavelet transform, and empirical mode decomposition, which are easily limited by factors such as the number of sources, wavelet types, threshold selection, and modal aliasing in practical applications; at the same time, for The high-frequency random noise contained in the oscillation signal, the above-mentioned noise reduction algorithm has made a proper attempt in noise filtering, but there is generally a problem of insufficient denoising performance

At present, there is no in-depth and systematic research on the precise filtering of high-frequency random noise in the oscillation signal of pumped storage units, and there is a lack of scientific and effective theoretical system for accurate filtering of high-frequency random noise in vibration signals of pumped storage units. Meet the functional requirements for accurate analysis and evaluation of the operating status of the unit

Images