基于粒子群算法优化最小二乘支持向量机的电路故障诊断方法 |
| |
引用本文: | 程思嘉,张昌宏. 基于粒子群算法优化最小二乘支持向量机的电路故障诊断方法[J]. 四川兵工学报, 2016, 0(3): 98-101. DOI: 10.11809/scbgxb2016.03.024 |
| |
作者姓名: | 程思嘉 张昌宏 |
| |
作者单位: | 海军工程大学信息安全系,武汉,430033 |
| |
基金项目: | 全军军事类研究生课题(2013JY430) |
| |
摘 要: | 针对数/模混合电路故障的特点,采用将粒子群算法与最小二乘支持向量机相结合的故障诊断方法,在保证诊断过程准确率的基础上,实现多类故障的快速诊断。在诊断过程中,支持向量机的参数寻优过程存在随意性、盲目性和效率低等问题,采用改进的粒子群算法优化支持向量机的参数,建立基于支持向量机的故障分类模型。实验结果表明,与其他方法相比,该方法提高了故障诊断的精度,具有明显的实用价值。
|
关 键 词: | 故障诊断 粒子群算法 最小二乘支持向量机 |
Fault Diagnosis Method of Circuit Using LS-SVM and Improved PSO |
| |
Abstract: | In allusion to the features of hybrid circuit fault,this text adopted a diagnostic method combined least squares support vector machine with particle swarm optimization.The method completes the diagnos-tic test frequently on the base of accuracy rate.Support vector machine’s parameters overcomes the ran-domness,blindness and inefficiency of the searching process by using modified particle swam optimization. The fault classification model based on support vector machine was established.Experimental results show the method raises the accuracy rate of fault diagnosis compared to other methods and has excellent practical value. |
| |
Keywords: | fault diagnosis particle swarm least square support vector machine |
本文献已被 万方数据 等数据库收录! |