摘要
为了解决因缺少大量故障数据样本而制约机械故障智能诊断发展的问题,提出了一种基于支持向量机的故障诊断模型。该模型建立在VC维理论和结构风险最小原理基础上,根据有限的样本信息在模型的复杂性和学习能力之间寻求最佳折衷。在选取诊断模型输入向量时,对故障信号功率谱进行小波分解,简化了故障特征向量的提取。仿真结果表明该模型可以有效地对旋转机械设备故障进行诊断。
In order to solve the problem in development of machinery fault intelligent diagnosis due to needing many fault data sampies, a fault diagnosis model based on support vector machine (SVM) is proposed. The model is built on VC dimension and structural risk minimization principle. According to limited sample information, the model seeks the optimal approach between complexity and study ability of the model. In selecting input vectors, the power spectrum of fault signals are decomposed by wavelet analysis, which predigests choosing method of fault eigenvectors. The simulation results show the model can effectively diagnose rotating machinery facility faults.
出处
《微计算机信息》
北大核心
2006年第12S期184-185,199,共3页
Control & Automation
基金
中国博士后科学基金资助项目(2005038515)
关键词
小波包分析
故障诊断
支持向量机
核函数
Wavelet packet analysis
Fault diagnosis
Support vector machine
Kernel function