针对人为选择支持向量机参数的随机性和盲目性, 将蚁群算法的全局收敛和并行计算的特点引入到支持向量机参数的优化中, 建立了基于蚁群算法优化支持向量机参数的模型, 使两种算法的优点有机结合, 通过对支持向量机的惩罚因子和核函数参数进行优化, 使支持向量机分类效果达到最好, 并与遗传支持向量机模型比较, 结果表明：蚁群算法优化支持向量机参数的方法不仅能够提高支持向量机的分类正确率, 而且算法循环时间比较少；最后对Elliptical Filter电路进行仿真, 应用小波分析提取响应信号的能量作为故障特征并建立故障样本集, 利用蚁群支持向量机模型实现了Elliptical Filter电路的故障诊断, 分类正确率达到100%。

Considering that the parameters of Support Vector Machine (SVM) were chosen randomly and blindly,the Ant Colony Optimization (ACO) algorithm,which has the features of global convergence and parallel computation,were introduced in the parameter optimization of SVM.A parameter optimization model based on ACO algorithm was established in order to combine the advantages of the algorithms.The classifying performance of SVM could reach the best state by optimizing the penalty factor and the kernel function parameter of it.Compared with the model based on GA-SVM,the ACO based SVM parameter optimization could improve the classification accuracy of SVM,and the time cost was less.At last,simulation was carried out for Elliptical Filter circuit,in which wavelet analysis was used to pick up the energy distribution of response signal and fault sample set was established,then fault diagnosis based on ACO-SVM was implemented which showed that the classification accuracy is 100%.