基于粗糙集与支持向量机的液压挖掘机故障诊断系统研究

发布时间：2018-11-09 20:22

【摘要】：液压挖掘机是一种重要的工程机械,广泛应用于土石方施工作业中。液压挖掘机的工作环境恶劣、工作强度大,发生故障的概率也随之增大,而且液压挖掘机的故障诊断往往伴随着随机性和突发性,因此,如何实现液压挖掘机故障的智能诊断是一直以来备受关注的问题。随着机电液一体化技术的广泛应用以及系统复杂度与集成度的增加,故障的诊断越来越困难,仅依靠传统的经验诊断方法往往不能做出快速准确的判断,这种单纯依靠经验来实现液压挖掘机故障诊断的方法已经难以满足要求。因此,本文以EC230履带式液压挖掘机为研究对象,结合粗糙集、支持向量机和粒子群优化算法,研究液压挖掘机故障诊断方法及其系统。本文首先在研究和分析EC230液压挖掘机结构及故障机理的基础上,对液压挖掘机故障诊断方法进行了设计与可行性分析,针对液压挖掘机故障样本相互关联性异常复杂的问题,运用粗糙集理论对液压挖掘机故障样本进行特征提取,消除故障样本的冗余特征,提高故障样本的有效性；然后,研究基于支持向量机的液压挖掘机故障诊断多分类算法及其模型,并对基于支持向量机故障诊断模型的参数寻优进行了研究,旨在提高故障诊断模型的泛化能力和诊断准确率。最后,基于液压挖掘机故障诊断多分类算法和VC++6.0编程工具实现了一个液压挖掘机故障诊断系统,仿真结果表明,基于粗糙集与支持向量机的液压挖掘机故障诊断,效果良好,能够实现液压挖掘机故障的快速、准确诊断,该系统可以替代故障诊断专家部分或全部工作,有效减少了液压挖掘机因故障而停机的时间,为液压挖掘机带来更高的经济效益。
[Abstract]:Hydraulic excavator is an important construction machinery, which is widely used in earthwork construction. The working environment of hydraulic excavator is bad, the working intensity is high, the probability of failure also increases, and the fault diagnosis of hydraulic excavator is often accompanied by randomness and sudden, so, How to realize intelligent fault diagnosis of hydraulic excavator has been a problem of great concern. With the wide application of electromechanical and hydraulic integration technology and the increase of system complexity and integration, fault diagnosis is becoming more and more difficult. The method of fault diagnosis of hydraulic excavator based on experience has been difficult to meet the requirements. Therefore, taking EC230 crawler hydraulic excavator as the research object, combining rough set, support vector machine and particle swarm optimization algorithm, this paper studies the fault diagnosis method and its system of hydraulic excavator. Based on the research and analysis of the structure and fault mechanism of EC230 hydraulic excavator, the design and feasibility analysis of fault diagnosis method for hydraulic excavator are carried out in this paper. The rough set theory is used to extract the fault sample of hydraulic excavator to eliminate the redundant feature of the fault sample and to improve the effectiveness of the fault sample. Then, the multi-classification algorithm and its model of hydraulic excavator fault diagnosis based on support vector machine are studied, and the parameter optimization based on support vector machine fault diagnosis model is studied. The purpose of this paper is to improve the generalization ability and diagnostic accuracy of fault diagnosis model. Finally, a hydraulic excavator fault diagnosis system is realized based on the multi-classification algorithm of hydraulic excavator fault diagnosis and VC 6.0 programming tool. The simulation results show that the hydraulic excavator fault diagnosis based on rough set and support vector machine. The system can replace some or all of the fault diagnosis experts, and effectively reduce the downtime of hydraulic excavator due to failure. For hydraulic excavator to bring higher economic benefits.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TU621;TU607

【参考文献】