基于改进MF-DFA的液压泵退化特征提取方法

发布时间：2018-01-12 14:11

本文关键词：基于改进MF-DFA的液压泵退化特征提取方法　出处：《振动.测试与诊断》2017年01期 　论文类型：期刊论文

【摘要】：针对液压泵振动信号通常具有非线性强和信噪比低的特点,提出了一种基于改进多重分形去趋势波动分析(multi-fractal detrended fluctuation analysis,简称MF-DFA)的液压泵性能退化特征提取方法。首先,引入滑动窗口技术改进传统MF-DFA方法在时间序列数据分割过程中存在的不足,提高了MF-DFA方法的计算精度;然后,利用改进的MF-DFA方法计算液压泵多重分形谱参数,分析了不同分形谱参数对液压泵退化状态的反映能力,选取奇异指数α0和多重分形谱宽度Δα作为退化特征量;最后,以液压泵不同退化状态下的实测数据为例验证了该算法的有效性。试验结果表明,该方法能够准确提取液压泵退化特征,提高了退化状态识别的准确率。
[Abstract]:The vibration signal of hydraulic pump usually has the characteristics of strong nonlinearity and low signal-to-noise ratio (SNR). A multi-fractal detrended fluctuation analysis based on improved multifractal detrend fluctuation analysis is proposed. Firstly, the sliding window technique is introduced to improve the traditional MF-DFA method in the process of time series data segmentation. The accuracy of MF-DFA method is improved. Then, the multifractal parameters of hydraulic pump are calculated by using the improved MF-DFA method, and the ability of reflecting different fractal spectrum parameters to the degradation state of hydraulic pump is analyzed. Singular exponent 伪 0 and multifractal spectrum width 螖伪 are selected as degenerate eigenvalues. Finally, the validity of the algorithm is verified by taking the measured data of hydraulic pump under different degenerative states as an example. The experimental results show that the proposed method can accurately extract the degradation characteristics of hydraulic pumps and improve the accuracy of degenerative state recognition.
【作者单位】：军械工程学院导弹工程系;
【基金】：国家自然科学基金资助项目(51275524)
【分类号】：TH137.51
【正文快照】： 引言液压泵是整个液压系统的“心脏”,其性能好坏不仅直接影响液压系统的可靠性,甚至对整个系统的安全运行产生决定性的影响[1]。退化特征提取是实现液压泵故障预测的关键。由于液压泵振动信号受到流体的压缩性、泵源与伺服系统的流固耦合作用以及自身具有的大幅固有机械振动

【参考文献】