摘要: 针对支持向量机(SVM)分类模型参数选取困难的问题,提出基于遗传免疫的改进粒子群优化算法,克服传统粒子群算法前期收敛快、后期易陷入局部最优的缺陷。将该算法与优化支持向量机分类模型相结合,建立基于遗传免疫粒子群和支持向量机的诊断模型,并用于轴承故障诊断中。结果表明,基于遗传免疫粒子群算法优化的SVM可实现对SVM分类模型参数的自动优化,并能提高SVM分类模型的故障诊断精度,对分散程度较大、聚类性较差的故障样本分类有较强的适用性。
关键词:
支持向量机,
故障诊断,
粒子群优化,
遗传免疫,
轴承,
交叉验证
Abstract: In order to resolve the difficulty that the choice of parameters influence the accuracy of Support Vector Machine(SVM) fault diagnosis model, a genetic-immune Particle Swarm Optimization(PSO) algorithm based on genetic evolution algorithm and immune selection algorithm is presented and used to optimize model parameters of SVM. The forecasting model based on a genetic-immune PSO algorithm and SVM is proposed and used to diagnose bearing fault. The results show that diagnosis model of SVM optimized by genetic-immune PSO algorithm can achieve automatic optimization of parameters, increase diagnosis accuracy of the conventional cross-validation algorithm, and is more fitting to classify the faulty samples scattered greatly.
Key words:
Support Vector Machine(SVM),
fault diagnosis,
Particle Swarm Optimization(PSO),
genetic-immune,
bearing,
cross-validation
中图分类号:
杨柳松, 何光宇. 基于改进粒子群优化的SVM故障诊断方法[J]. 计算机工程, 2013, 39(3): 187-190,196.
YANG Liu-Song, HE Guang-Yu. Support Vector Machine Fault Diagnosis Method Based on Improved Particle Swarm Optimization[J]. Computer Engineering, 2013, 39(3): 187-190,196.