基于改进蝴蝶算法优化SVM的TE过程故障诊断  

TE Process Fault Diagnosis Based on Improved Butterfly Algorithm Optimization SVM

作　　者：赵文虎 ZHAO Wenhu(College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050)

机构地区：[1]兰州理工大学电气工程与信息工程学院,兰州730050

出　　处：《计算机与数字工程》2024年第5期1569-1575,共7页Computer & Digital Engineering

摘　　要：针对支持向量机在田纳西-伊斯曼过程故障诊断中存在不稳定且分类精度低以及蝴蝶优化算法在优化支持向量机时存在的易陷入局部最优和收敛速度慢等问题,提出了采用改进蝴蝶算法优化支持向量机的模型。首先利用反向学习策略、自适应调节策略和混沌局部搜索策略来优化蝴蝶算法,改善蝴蝶算法的寻优性能和开发能力。然后采用改进蝴蝶算法来优化支持向量机的参数,最后将改进蝴蝶算法优化的支持向量机模型应用到田纳西-伊斯曼过程的故障诊断中。通过函数测试和对比实验,结果表明:改进蝴蝶算法具有较好的寻优效果,用改进蝴蝶算法优化的支持向量机具有更好的稳定性和更高的分类准确率。Aiming at the problems of instability and low classification accuracy of support vector machine in Tennessee East-man process fault diagnosis,easy to fall into local optimization and slow convergence of butterfly optimization algorithm in optimiz-ing support vector machine,a model of using improved butterfly algorithm to optimize support vector machine is proposed.Firstly,reverse learning strategy,adaptive adjustment strategy and chaotic local search strategy are used to optimize the butterfly algorithm to improve the optimization performance and development ability of the butterfly algorithm.Then the improved butterfly algorithm is used to optimize the parameters of support vector machine.Finally,the support vector machine model optimized by the improved butterfly algorithm is applied to the fault diagnosis of Tennessee Eastman process.Through function test and comparative experi-ment,the results show that the improved butterfly algorithm has better optimization effect,and the support vector machine opti-mized by the improved butterfly algorithm has better stability and higher classification accuracy.

关 键 词：支持向量机 蝴蝶优化算法 故障诊断 函数测试 分类准确率 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]