| 基于核偏最小二乘回归方法的发电机关键运行参数预测分析 |
| |
| 引用本文： | 马贺贺,李辉,张小虎. 基于核偏最小二乘回归方法的发电机关键运行参数预测分析[J]. 大电机技术, 2017, 0(4). DOI: 10.3969/j.issn.1000-3983.2017.04.002 |
| |
| 作者姓名： | 马贺贺 李辉 张小虎 |
| |
| 作者单位： | 1. 上海电气集团股份有限公司中央研究院,上海,200070;2. 上海电气电站设备有限公司上海发电机厂,上海,200240 |
| |
| 基金项目： | 上海市科委企业合作专项项目 |
| |
| 摘 要： | 通过对发电机关键运行参数的实时预测能够辅助实现发电机运行状态的有效监控。传统偏最小二乘斱法(Partial Least Squares,PLS)在应用过程中没有考虑变量间的非线性关系,为提高发电机关键运行参数的预测精度,引入核函数迚行变量空间的非线性映射变化,实现非线性回归分析,幵将核偏最小二乘回归斱法(Kernel Partial Least Squares Regression,KPLSR)应用于发电机定子线圈出水温度预测,实际运行数据的对比分析验证了斱法的有效性。
|
| 关 键 词： | 发电机 偏最小二乘方法 核函数 参数预测 |
Prediction Analysis of Key Generator Operation Parameters Based on Kernel Partial Least Squares Regression Method |
| |
| MA Hehe,LI Hui,ZHANG Xiaohu. Prediction Analysis of Key Generator Operation Parameters Based on Kernel Partial Least Squares Regression Method[J]. Large Electric Machine and Hydraulic Turbine, 2017, 0(4). DOI: 10.3969/j.issn.1000-3983.2017.04.002 |
| |
| Authors: | MA Hehe LI Hui ZHANG Xiaohu |
| |
| Abstract: | Real-time prediction of key operation parameters can assist to achieve effective status monitoring of the generator. Traditional partial least squares method is used without considering the nonlinear relationship between variables. In order to improve the prediction accuracy of generator operation parameters, Kernel function for nonlinear mapping of the variable space is applied in this paper. The kernel partial least squares regression method is used to predict generator stator coil outlet temperature. The validity of KPLSR is verified by contrastive analysis of the real operation data. |
| |
| Keywords: | generator partial least squares kernel function parameter prediction |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
