| 基于LSSVR的模拟电路故障诊断 |
| |
| 引用本文： | 侯青剑,王宏力.基于LSSVR的模拟电路故障诊断[J].上海航天,2009,26(5):61-64. |
| |
| 作者姓名： | 侯青剑 王宏力 |
| |
| 作者单位： | 第二炮兵工程学院,陕西,西安,710025 |
| |
| 摘 要： | 对基于最小二乘回归型支持向量机（LSSVR）的模拟电路故障诊断方法进行了研究。根据模型故障诊断原理，设计了LSSVR残差生成器，用遗传算法优化LSSVR参数，以残差的高阶统计量为故障特征识别电路故障。仿真结果表明：该法能有效识别模拟电路的故障，具一定的实用性。
|
| 关 键 词： | 模拟电路 故障诊断 最小二乘支持向量机 遗传算法 残差 |
Application of LSSVR on Analog Circuit Fault Diagnosis |
| |
| HOU Qing-jian,WANG Hong-li.Application of LSSVR on Analog Circuit Fault Diagnosis[J].Aerospace Shanghai,2009,26(5):61-64. |
| |
| Authors: | HOU Qing-jian WANG Hong-li |
| |
| Institution: | (The Second Artillery Engineering Institute, Xi＇an Shaanxi 710025, China) |
| |
| Abstract: | The fault diagnosis of analog circuit based on the least square support vector regression （LSSVR） was developed on this paper. According to the principle of the fault diagnosis method based on model, the residual error generator was designed and the parameters of LSSVR were optimized by genetic algorithm. The circuit faults were identified through the fault character of higher-order statistics of residual error. The results of simulative experiments on analog circuit showed that the method was effective and practical. |
| |
| Keywords: | Analog circuit Fault diagnosis Least square support vector machine Genetic algorithm Residual error |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
