| 进化蒙特卡洛优化的SVM在故障诊断中的应用 |
| |
| 引用本文： | 刘晓平,郑海起,祝天宇.进化蒙特卡洛优化的SVM在故障诊断中的应用[J].振动．测试与诊断,2011,31(1):115-118. |
| |
| 作者姓名： | 刘晓平 郑海起 祝天宇 |
| |
| 作者单位： | 1. 军械工程学院一系,石家庄,050003
2. 武汉军械士官学校,武汉,430075 |
| |
| 基金项目： | 国家自然科学基金资助项目 |
| |
| 摘 要： | 针对支持向量机超参数选择问题,将进化蒙特卡洛算法引入支持向量机的参数优化.以交叉验证误差作为目标函数,并行运行多条马尔可夫链,设计了一种改进的变异操作.使得算法在进化的初期具有较强的全局搜索能力,在进化的后期具有精细的局部搜索能力,从而加速马尔可夫链的混合,提高算法的寻优效率,最后将优化的支持向量机应用于滚动轴承故障诊断.试验结果表明,该方法具有较高的寻优效率和参数优化能力,可提高故障识别的精度.
|
| 关 键 词： | 进化蒙特卡洛 支持向量机 参数优化 故障诊断 |
| 收稿时间： | 2010/6/24 0:00:00 |
| 修稿时间： | 2010/10/11 0:00:00 |
Support Vector Machines Using Evolutionary Monte Carlo Parameter Optimization and Its Application to Fault Diagnosis |
| |
| Liu Xiaoping,Zheng Haiqi,Zhu Tianyu.Support Vector Machines Using Evolutionary Monte Carlo Parameter Optimization and Its Application to Fault Diagnosis[J].Journal of Vibration,Measurement & Diagnosis,2011,31(1):115-118. |
| |
| Authors: | Liu Xiaoping Zheng Haiqi Zhu Tianyu |
| |
| Affiliation: | Liu Xiaoping1,Zheng Haiqi1,Zhu Tianyu2(1The First Department,Ordnance Engineering College Shijiazhuang,050003,China)(2Wuhan Ordnance Non-Commissioned Officer Academy Wuhan,430075,China) |
| |
| Abstract: | The performance of support vector machine(SVM) heavily relies on the setting of the hyper-parameters.An evolutionary Monte Carlo(EMC) method was introduced for SVM hyper-parameters optimization.The target function was cross validation accuracy,a population of samples was simulated in parallel.An improved mutation operation was designed for accelerating the mixing of the Markov chains and improving the searching efficiency.It makes the algorithm possess strong global searching ability in the initial stages a... |
| |
| Keywords: | evolutionary Monte Carlo support vector machine parameters optimization fault diagnosis |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《振动．测试与诊断》浏览原始摘要信息 |
|
点击此处可从《振动．测试与诊断》下载全文 |
|
