Study on Health Indicator Construction and Health Status Evaluation of Hydraulic Pumps Based on LSTM–VAE

This paper addresses the difficulty of evaluating operating status in widely used gear pumps. A method for constructing hydraulic pump health indicators and is proposed based on LSTM–VAE. In this study, vibration signal data source pumps was assessed accelerated life test. Firstly, normalized feature vectors whole-life operation were extracted by wavelet packet decomposition amplitude extraction. Combining an LSTM algorithm with a VAE algorithm, LSTM–VAE proposed. By learning varying conditions, one-dimensional HI curve obtained. Then, to predict According volume efficiency pumps, divided into four states: health, sub-health, deterioration, failure. The accurately evaluated indicator. Finally, compared traditional selection PCA dimensionality reduction. indicator constructed superior terms tendency, robustness, monotonicity, which proves effectiveness paper.