改进粒子群算法优化的SVM模拟电路故障诊断

发布时间：2018-02-13 10:29

本文关键词： 支持向量机改进粒子群算法模拟电路故障诊断　出处：《电子测量与仪器学报》2017年08期 　论文类型：期刊论文

【摘要】：针对粒子群(PSO)算法优化支持向量机(SVM)参数存在容易陷入局部最优的问题,通过引入新的动态惯性权重、全局邻域搜索、收缩因子和遗传算法中的变异操作,提出了一种基于改进粒子群(IPSO)算法优化SVM参数(IPSO-SVM)的改进型分类器。采用UCI机器学习库中的公共数据集Iris、Wine和seeds来测试其分类效果,结果表明IPSO-SVM分类器在分类准确率和分类时间上优于GS-SVM、AFSA-SVM、GA-SVM和PSO-SVM分类器。最后,将IPSO-SVM分类器应用于Sallen-Key带通滤波器、四运放双二次高通滤波器及非线性整流电路的故障诊断中,结果表明IPSO-SVM分类器具有较强的全局收敛能力和较快的收敛速度。
[Abstract]:In order to solve the problem that the parameters of support vector machine (SVM) optimization based on particle swarm optimization (PSO) algorithm are easy to fall into local optimum, new dynamic inertia weight, global neighborhood search, shrinkage factor and mutation operation in genetic algorithm are introduced. An improved classifier based on improved particle swarm optimization (PSO) algorithm is proposed to optimize the SVM parameters (IPSO-SVM). Iris Wine and seeds, a common data set in the UCI machine learning database, are used to test the classification effect. The results show that IPSO-SVM classifier is superior to GS-SVMU AFSA-SVMU GA-SVM and PSO-SVM classifier in classification accuracy and classification time. Finally, the IPSO-SVM classifier is applied to the fault diagnosis of Sallen-Key bandpass filter, four-op amplifier double-quadratic high-pass filter and nonlinear rectifier circuit. The results show that the IPSO-SVM classifier has strong global convergence ability and faster convergence speed.
【作者单位】：合肥工业大学电气与自动化工程学院;
【基金】：国家自然科学基金(51577046,11105037);国家自然科学基金重点项目(51637004) 安徽省科技计划重点项目(1301022036) 国际热核聚变实验堆(ITER)计划专项(2015GB102000) 教育部科学技术研究重大项目(313018)资助
【分类号】：TN710;TP18

【相似文献】