摘要
针对柴油机故障诊断样本少的实际问题,结合支持向量机的特性和微分进化算法良好的全局优化性能,提出了一种微分进化算法优化支持向量机方法。利用柴油机气门振动信号实测数据,经小波变换作为诊断模型的故障特征,建立了基于微分进化算法优化支持向量机的柴油机气门间隙故障诊断模型,并与反向传播神经网络算法、基于遗传算法优化支持向量机和基于粒子群优化支持向量机的模型相比较,结果表明:应用微分进化算法优化支持向量机比其他三种算法具有更优性能,能够有效地进行柴油机的故障诊断。
Aiming at the practical problem of few samples in diesel engine fault diagnosis,combining the characteristics of support sector machine and good global optimization performance of differential evolution algorithm,a mixed strategy of using differential evolution algorithm to optimize the parameters of support vector machine is proposed.After wavelet transform,the measured data of diesel engine valve vibration signal are used as the characteristic values of the diagnostic model;the diesel engine valve gap fault diagnosis model is established,which is based on differential evolution algorithm optimized SVM.The proposed algorithm is compared with those of the BPNN,GA-SVM and PSO-SVM models.Comparison results show that the performance of the proposed DE-SVM algorithm is better than that of the other three algorithms.The proposed method can realize diesel engine fault diagnosis practically and effectively.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2011年第2期323-328,共6页
Chinese Journal of Scientific Instrument
基金
科技部国际科技合作项目(2007DFR10420)
重庆理工大学汽车零部件制造及检测技术教育部重点实验室开放基金(2009-10)资助项目
关键词
微分进化算法
支持向量机
故障诊断
小波变换
diffierential evolution
support vector machine
fault diagnosis
wavelet transform
