| 熊伟丽,李妍君,姚乐,徐保国.一种动态校正的AGMM-GPR多模型软测量建模方法[J].,2016,56(1):77-85 |
| 一种动态校正的AGMM-GPR多模型软测量建模方法 |
| A dynamically corrected AGMM-GPR multi-model soft sensor modeling method |
|
| DOI：10.7511/dllgxb201601012 |
| 中文关键词:自适应多模型动态校正高斯过程回归ARIMA模型 |
| 英文关键词:adaptivemulti-modeldynamic correctionGaussian process regressionARIMA model |
| 基金项目:国家自然科学基金资助项目(2120605321276111)；江苏省“六大人才高峰”计划资助项目(2013-DZXX-043)；江苏省产学研资助项目(BY2014023-27)；江苏高校优势学科建设工程资助项目(PAPD). |
|
| 摘要点击次数:1628 |
| 全文下载次数:2099 |
| 中文摘要: |
| 工业过程常常是强非线性的，并有多个工况，传统的软测量方法存在预测能力差，不能有效利用误差信息等缺点．为了有效解决这些问题，提出一种基于自适应高斯混合模型-高斯过程回归(AGMM-GPR)的多模型动态校正软测量建模方法．首先，通过贝叶斯信息准则构建自适应高斯混合模型(AGMM)，得到优化的子模型个数；然后，利用GPR方法建立各局部模型，当新的数据到来时，将其隶属于各局部模型的后验概率和预测值融合得到多模型输出；最后，为了进一步提高模型的精度，构建自回归积分滑动平均(ARIMA)模型对多模型输出进行动态反馈校正．通过数值仿真和硫回收装置(SRU)中H 2S浓度的估计，验证了所提方法具有良好的预测精度和泛化性能． |
| 英文摘要: |
| Industrial processes often encounter strong nonlinearity and multiple operating modes. Traditional soft sensor methods cannot effectively take advantage of error information, which accounts for unsatisfactory predictive results. To effectively address these problems, a dynamically corrected multi-model soft sensor modeling method based on adaptive Gaussian mixture model-Gaussian process regression (AGMM-GPR) is proposed. Firstly, an adaptive Gaussian mixture model is constructed using Bayesian information criterion and optimized sub-model number is obtained. Then, each local model is built through GPR method. For the new data, its posterior probability and prediction value belonging to each local model can be combined to get multi-model output. Finally, to further improve model accuracy, an autoregressive integrated moving average (ARIMA) model is employed to conduct a dynamic feedback correction to multi-model output. Numerical simulation and H 2S concentration estimation in sulfur recovery unit (SRU) indicate that the proposed method has good prediction accuracy and generalization performance. |
| 查看全文查看/发表评论下载PDF阅读器 |
| --> | 关闭 |
优惠券领取后72小时内有效，10万种最新考研考试考证类电子打印资料任你选。涵盖全国500余所院校考研专业课、200多种职业资格考试、1100多种经典教材，产品类型包含电子书、题库、全套资料以及视频，无论您是考研复习、考证刷题，还是考前冲刺等，不同类型的产品可满足您学习上的不同需求。 ...考试优惠券 本站小编 Free壹佰分学习网 2022-09-19