基于经验模态分解与深度森林的液压泵健康评估

提出了一种基于经验模态分解和深度森林的方法,用于分析液压泵出口压力,从而进行健康状态评估。通过试验系统采集了不同工作时间下液压泵的出口压力信号,利用经验模态分解的方法对其进行分解,分别得到一组本征模态函数,提取其特征;结合原始信号典型时域特征,最终构成信号的特征向量。采用深度森林的方法进行不同健康状态的分类。实验结果表明,所提方法的分类结果准确率可达97%,采用经验模态分解和深度森林结合的方法可以有效提高液压泵健康状态评估的准确率。

Health A ssessment of Hydraulic Pump Based on Empirical Mode Decomposition and Deep Forest

A method based on empirical mode decomposition (EMD) and deep forest was proposed to analyze the pressure at the pump outlet to evaluate the health status. The pressure signals of hydraulic pump outlets under different working hours are collected by the test system, and they are decomposed by the EMD method to obtain a set of intrinsic mode function (IMF). The features of the IMFs are extracted and combined with the typical time-domain features of the original signal to form the feature vectors of the signal. Deep forest method is used to classify different health states. The experimental results show that the accuracy of the classification results of the proposed method can reach 97%, and the combination of EMD and deep forest can effectively improve the accuracy of the health assessment of the hydraulic pump.