| 即时学习多模型加权GPR软测量方法 |
| |
| 引用本文： | 王莉,孙玉梅,杨凯,刘学杰,陈祥光.即时学习多模型加权GPR软测量方法[J].北京理工大学学报,2018,38(2):196-199,204. |
| |
| 作者姓名： | 王莉 孙玉梅 杨凯 刘学杰 陈祥光 |
| |
| 作者单位： | 北京理工大学化学与化工学院,北京,100081;烟台南山学院电气工程学院,山东,龙口 265713;中国人民公安大学国际警务执法学院,北京,100038 |
| |
| 基金项目： | 中国博士后科学基金资助项目（20100480208）；山东省自然科学基金面上项目（ZR2016FM28） |
| |
| 摘 要： | 针对工业过程中难以实现实时在线测量的重要过程变量，在主成分降解变量分组的基础上，提出了一种基于即时学习与集成学习的多模型高斯过程回归建模方法.该方法首先利用多变量组合实现集成模型的多样性，然后借助即时学习的自适应能力进行即时建模，最终多模型加权获得最终的预测结果.将所提方法应用于实际的工业炼胶过程，实验结果表明，该方法具有很好的预测性能.
|
| 关 键 词： | 软测量 即时学习 集成建模 高斯过程回归 |
| 收稿时间： | 2016/7/7 0:00:00 |
A Soft Sensor Method Based on Just in Time Learning and Multi-Model Ensemble GPR |
| |
| WANG Li,SUN Yu-mei,YANG Kai,LIU Xue-jie and CHEN Xiang-guang.A Soft Sensor Method Based on Just in Time Learning and Multi-Model Ensemble GPR[J].Journal of Beijing Institute of Technology(Natural Science Edition),2018,38(2):196-199,204. |
| |
| Authors: | WANG Li SUN Yu-mei YANG Kai LIU Xue-jie and CHEN Xiang-guang |
| |
| Institution: | 1. Department Chemistry and of Chemical Engineering, Beijing Institute of Technology, Beijing 100081, China;2. College of Electronic Engineering, Yantai Nanshan University, Longkou, Shandong 265713, China;3. People's Public Security University of China, Beijing, 100038, China |
| |
| Abstract: | For some difficulties to realize real-time measurements of vital variables, a novel soft sensor based on the just in time learning (JITL) and ensemble learning Gaussian process regression (GPR) modeling was proposed. Firstly, different sub-blocks were created by principal component decomposition. Then a set of JITL-GPR models were developed based on various sub-blocks. Finally, some better results from JITL-GPR models were combined to adaptively obtain the prediction result. The proposed soft sensor was applied to the industrial rubber producing process. Result verifies the superiority of the proposed method. |
| |
| Keywords: | soft sensor just in time learning integrated modeling Gaussian process regression |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《北京理工大学学报》浏览原始摘要信息 |
| 点击此处可从《北京理工大学学报》下载免费的PDF全文 |
|
