论文部分内容阅读
针对液压系统性能参数的退化特点,提出了一种小波包变换和隐马尔科夫模型(HMM)相结合的液压系统故障预测方法.该方法对采集的振动信号进行小波包变换,提取能量特征,分别使用正常状态,经过轻度退化状态、中度退化态和高度退化态,最终达到故障状态的全过程数据训练HMM,建立性能评估模型,然后进行模式识别,实现液压系统的故障预测.最后,通过试验研究,验证了所提出的方法的可行性和有效性.
A hydraulic system fault prediction method based on wavelet packet transform and hidden Markov model (HMM) is proposed for degradation characteristics of hydraulic system performance parameters. The proposed method performs wavelet packet transform on the acquired vibration signals to extract energy features, The HMM was trained using the normal state, the light degraded state, the moderately degenerated state and the highly degenerated state, and finally the fault state was reached, and the performance evaluation model was established, and then the pattern recognition was carried out to realize the fault prediction of the hydraulic system.Finally, Through the experimental study, the feasibility and effectiveness of the proposed method are verified.