基于支持向量机和振动场的故障诊断方法

Fault Diagnosis Based on Support Vector Machines and Vibration Field

故障样本量是制约智能故障诊断发展的关键因素之一,然而实践中往往难以获取充足的故障样本。支持向量机是一种新型的机器学习和模式识别方法,在解决小样本、非线性及高维模式识别问题中表现出优越的性能。本文将欧氏距离分类引入到支持向量机解决支持向量机多类分类问题,提出了基于支持向量机和振动场的故障诊断方法。实验结果表明,该方法在故障诊断上计算速度和准确度令人满意,为类似的研究提供了借鉴意义和参考。

The Shortage of fault samples is one of the main reasons that restrict the development of intelligent, but in practice it is often difficult to obtain sufficient fault samples. Support vector machine is a new type of machine learning and pattern recognition methods, in addressing the small sample, nonlinear and high dimensionalpattern recognition problems showed superior performance. This article will introduce to the Euclidean distance classifier support vector machine to solve multi-class support vector machineclassification is proposed based on support vector machines andmechanical vibration fault diagnosis field. Experimental resultsshow that the method in machinery fault diagnosis on computing speed and accuracy is satisfactory for a similar study providesreference and reference.