【作者】 王庆华；

【导师】 孙玉清；

【作者基本信息】 大连海事大学 ， 轮机工程， 2004， 硕士

【摘要】 分形是一门新的学科，它特别适用于分析复杂系统，而把分形理论应用于机械系统的故障诊断领域，是近年来国际学术界的新动向。可以采用分形理论分析机械系统的状态信号，从中提取出分形维数作为特征信息，与故障档案进行比较，从而实现机械设备的故障诊断。 本文首先对液压系统故障诊断技术的研究现状进行了综述性回顾，以故障特征提取为切入点，提出了分形用于液压系统故障诊断的新方法。然后，分别介绍了液压系统故障的复杂性理论和分形理论，并对液压系统中的各种状态信号进行了分析。在这些基础上，研究了基于分形的液压系统故障诊断的方法及流程，提出了其诊断模型。 本文以液压泵源系统为研究对象，运用分形理论对其进行故障诊断的仿真研究和实验研究，其中包括仿真模型的建立、实验系统的设计、数据的采集、特征参数的提取等。结果显示，液压泵壳体振动信号和管道压力信号在一定尺度范围内具有分形特征，不同工作(故障)状态下的液压泵壳体振动信号和管道压力信号的关联维数是不同的，具有明显可分性。关联维数对设备故障比较敏感，能够反映系统的动态特性。因此，关联维数作为液压设备故障诊断的敏感因子是可行的。这种方法简单、直观、易行，大大克服传统方法分析上故障特征的提取、分析的困难。 最后对课题的研究作了总结和展望。更多还原

【Abstract】 Fractal is a new subject which is especially suitable for the analysis of complex system. Recently signal processing methods based on fractal have been applied to mechanical system fault diagnosis. The objective of the proposed research is to develop a method based on fractal theory for analyzing the operating signals of mechanical system. This method will provide further useful information about the underlying dynamics. It is possible to implement mechanical system fault diagnosis by using this method compared with fault files.Firstly, this paper generalizes and reviews state of art and development in the field of fault diagnosis for hydraulic system faults and, starting with fault signal identification, it raises a new method of hydraulic system fault diagnosis based on fractal. Then, it describes separately hydraulic system’s complexity theory and fractal theory and analyzes all kinds of operating signals in hydraulic system. On these bases, it studies the method of fractal-based fault diagnosis of hydraulic system, puts forward its diagnosis model.Hydraulic pump is the research object in this paper. In this research, the simulation and experiment of fault diagnosis of hydraulic pump are established, including model establishing, data sampling, obtaining character parameters and so on. The results indicate that the different correlation dimensions of vibration and pressure signals of hydraulic pump under different operating states and the correlation dimension is sensitive to hydraulic pump faults. It reflects the dynamic characteristic of system. The method is simple, intuitionistic and feasible. It greatly overcomes the difficulty of characteristic identification in traditional methods.At last, the research is summarized and prospected in this paper.更多还原