Fault diagnosis of hydropower units based on UPEMD integrating RCMCSE and ALWOA-BP
The diagnosis of vibration signals in hydropower units is crucial to the safe and stable operation of the u-nits.This article proposes a fault diagnosis method for hydropower units based on uniform phase empirical mode de-composition(UPEMD)combined with refined composite multiscale cosine similarity entropy(RCMCSE)and an im-proved whale optimization algorithm(ALWOA)optimized back propagation neural network(BP).The UPEMD is used to decompose the original signal,and then a WOA-BP fault diagnosis model is established.To solve the problem of WO A algorithm quickly falling into local optimum and premature convergence,an adaptive weight and Levy flight are used to optimize the WO A algorithm.Experimental results show that the accuracy of this method reached 100％.To explore the noise resistance performance of the proposed model,a noise with a signal-to-noise ratio of 2 dB was introduced for re-analysis,and the diagnostic result was 94.44％,which was significantly better than other unoptimized models.This study can provide a valuable complement to existing fault diagnosis methods for hydropower units.
国家科技期刊平台
NSTL
万方数据
