帶時(shí)滯估計(jì)的軟測(cè)量建模方法研究
發(fā)布時(shí)間：2018-09-03 07:13
【摘要】：化工過(guò)程對(duì)象呈現(xiàn)顯著的非線性和時(shí)變性,為了對(duì)過(guò)程實(shí)施高效的監(jiān)控策略,廣泛以軟測(cè)量技術(shù)的手段對(duì)反映產(chǎn)品質(zhì)量指標(biāo)的難測(cè)變量(即主導(dǎo)變量)進(jìn)行推斷估計(jì)。如今,隨著過(guò)程工況復(fù)雜度的日益增加,工業(yè)界對(duì)于軟測(cè)量技術(shù)的精度和可靠性方面的要求也相應(yīng)提高。在實(shí)際過(guò)程中,主導(dǎo)變量的獲取通常受到裝置成本、儀表可靠性或技術(shù)瓶頸等方面的限制,存在很大的測(cè)量滯后性。盡管軟測(cè)量建模領(lǐng)域的研究不斷邁向自適應(yīng)時(shí)代,時(shí)滯信息卻往往不被考慮在建模過(guò)程中。為了進(jìn)一步改善傳統(tǒng)軟測(cè)量建模方法的預(yù)測(cè)精度,本論文不僅針對(duì)工業(yè)過(guò)程的時(shí)變和非線性特征,同時(shí)還考慮了過(guò)程數(shù)據(jù)集中隱含的時(shí)滯信息,在現(xiàn)有的軟測(cè)量技術(shù)研究成果的基礎(chǔ)上,以時(shí)間差高斯過(guò)程回歸算法為基礎(chǔ),對(duì)帶時(shí)滯估計(jì)的自適應(yīng)軟測(cè)量建模方法進(jìn)行了研究。全文的主要研究?jī)?nèi)容如下:1.針對(duì)建模數(shù)據(jù)時(shí)序匹配不一致和變量漂移的問(wèn)題,提出一種基于模糊曲線分析(Fuzzy Curve Analysis,FCA)的時(shí)間差高斯過(guò)程回歸(Time Difference Gaussian Process Regression,TDGPR)建模方法。該方法利用離線估計(jì)的時(shí)滯參數(shù)重新匹配建模樣本時(shí)序,對(duì)于查詢樣本,采用TDGPR模型對(duì)主導(dǎo)變量進(jìn)行在線預(yù)測(cè)。2.針對(duì)傳統(tǒng)全局時(shí)間差(Time Difference,TD)模型的“老化”問(wèn)題,基于選擇性集成思想,提出一種局部時(shí)間差高斯過(guò)程回歸(Local Time Difference Gaussian Process Regression,LTDGPR)的自適應(yīng)建模方法。首先,對(duì)數(shù)據(jù)庫(kù)中的時(shí)滯動(dòng)態(tài)信息進(jìn)行挖掘,并利用該信息對(duì)建模數(shù)據(jù)進(jìn)行重構(gòu);然后,采取局部化策略對(duì)差分后的重構(gòu)樣本進(jìn)行統(tǒng)計(jì)劃分,得到LTDGPR模型集。對(duì)于查詢樣本,在線選擇部分泛化能力強(qiáng)的LTDGPR子模型進(jìn)行集成,估計(jì)出含一定時(shí)間差的主導(dǎo)變量動(dòng)態(tài)偏移值;最后,基于TD模型思想對(duì)主導(dǎo)變量值進(jìn)行實(shí)時(shí)預(yù)測(cè)。3.考慮到過(guò)程非線性和時(shí)滯呈現(xiàn)出的階段性特征,提出一種基于局部時(shí)滯重構(gòu)(Local Time-delay Reconstruction,LTR)的滑動(dòng)窗時(shí)間差高斯過(guò)程回歸(Moving Window Time Difference Gaussian Process Regression,MWTDGPR)建模方法。該方法以滑動(dòng)窗和TD組合策略的方式逐步跟蹤過(guò)程局部非線性突變和緩變特征,同步實(shí)現(xiàn)了局部樣本集的時(shí)序校正和重構(gòu),為其它的時(shí)滯非線性時(shí)變化工過(guò)程建模問(wèn)題提供了一種可行的框架。論文通過(guò)實(shí)際過(guò)程的數(shù)據(jù)仿真研究驗(yàn)證了上述方法的可行性和精度,仿真結(jié)果充分顯示考慮時(shí)滯估計(jì)的自適應(yīng)軟測(cè)量建模對(duì)于化工過(guò)程的經(jīng)濟(jì)效益和安全平穩(wěn)運(yùn)行具有重大意義。
[Abstract]:Chemical process objects show significant nonlinear and time-varying characteristics. In order to implement an efficient monitoring strategy, the difficult variables (i.e. dominant variables) which reflect the product quality index are inferred and estimated by soft sensing technology. Nowadays, with the increasing complexity of the process, the requirements of the industry for the accuracy and reliability of soft sensing technology are also increased. In the actual process, the acquisition of dominant variables is usually limited by the cost of the device, the reliability of the instrument or the technical bottleneck, etc., so there is a great lag in measurement. Although the research in the field of soft sensor modeling is moving towards the era of adaptation, time-delay information is often not considered in the modeling process. In order to further improve the prediction accuracy of traditional soft sensor modeling methods, this paper not only aims at the time-varying and nonlinear characteristics of industrial processes, but also takes into account the implicit time-delay information in the process data set. Based on the existing research results of soft sensing technology and based on the time difference Gao Si process regression algorithm, the adaptive soft sensor modeling method with time delay estimation is studied. The main contents of this paper are as follows: 1. Aiming at the inconsistency of time series matching and variable drift of modeling data, a time difference Gao Si process regression (Time Difference Gaussian Process Regression,TDGPR (Time Difference Gaussian Process Regression,TDGPR) modeling method based on fuzzy curve analysis (Fuzzy Curve Analysis,FCA) is proposed. In this method, the time series of modeling samples is rematched by off-line estimation of time-delay parameters. For query samples, the TDGPR model is used to predict the dominant variables online. Aiming at the problem of "aging" of traditional global time difference (Time Difference,TD) model, an adaptive modeling method of local time-difference Gao Si process regression (Local Time Difference Gaussian Process Regression,LTDGPR) is proposed based on the idea of selective integration. Firstly, the time-delay dynamic information in the database is mined, and the modeling data is reconstructed using this information. Then, the LTDGPR model set is obtained by statistical partitioning of the reconstructed samples after the difference by using the localization strategy. For the query samples, the LTDGPR submodel with strong generalization ability is selected online to integrate to estimate the dynamic offset value of the dominant variable with a certain time difference. Finally, based on the idea of TD model, the dominant variable value is predicted in real time. 3. Considering the stage characteristics of process nonlinearity and time-delay, a sliding window time-difference Gao Si process regression (Moving Window Time Difference Gaussian Process Regression,MWTDGPR modeling method based on local time-delay reconstruction (Local Time-delay Reconstruction,LTR) is proposed. The method uses sliding window and TD combination strategy to track the local nonlinear mutation and slowly varying feature step by step, and synchronously realizes the timing correction and reconstruction of the local sample set. It provides a feasible framework for other nonlinear time-varying chemical process modeling problems with time delay. The feasibility and accuracy of the above method are verified by the data simulation of the actual process. The simulation results show that the adaptive soft-sensor modeling with time-delay estimation is of great significance for the economic benefit and safe and stable operation of chemical process.
【學(xué)位授予單位】：江南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TQ018
本文編號(hào)：2219287
[Abstract]:Chemical process objects show significant nonlinear and time-varying characteristics. In order to implement an efficient monitoring strategy, the difficult variables (i.e. dominant variables) which reflect the product quality index are inferred and estimated by soft sensing technology. Nowadays, with the increasing complexity of the process, the requirements of the industry for the accuracy and reliability of soft sensing technology are also increased. In the actual process, the acquisition of dominant variables is usually limited by the cost of the device, the reliability of the instrument or the technical bottleneck, etc., so there is a great lag in measurement. Although the research in the field of soft sensor modeling is moving towards the era of adaptation, time-delay information is often not considered in the modeling process. In order to further improve the prediction accuracy of traditional soft sensor modeling methods, this paper not only aims at the time-varying and nonlinear characteristics of industrial processes, but also takes into account the implicit time-delay information in the process data set. Based on the existing research results of soft sensing technology and based on the time difference Gao Si process regression algorithm, the adaptive soft sensor modeling method with time delay estimation is studied. The main contents of this paper are as follows: 1. Aiming at the inconsistency of time series matching and variable drift of modeling data, a time difference Gao Si process regression (Time Difference Gaussian Process Regression,TDGPR (Time Difference Gaussian Process Regression,TDGPR) modeling method based on fuzzy curve analysis (Fuzzy Curve Analysis,FCA) is proposed. In this method, the time series of modeling samples is rematched by off-line estimation of time-delay parameters. For query samples, the TDGPR model is used to predict the dominant variables online. Aiming at the problem of "aging" of traditional global time difference (Time Difference,TD) model, an adaptive modeling method of local time-difference Gao Si process regression (Local Time Difference Gaussian Process Regression,LTDGPR) is proposed based on the idea of selective integration. Firstly, the time-delay dynamic information in the database is mined, and the modeling data is reconstructed using this information. Then, the LTDGPR model set is obtained by statistical partitioning of the reconstructed samples after the difference by using the localization strategy. For the query samples, the LTDGPR submodel with strong generalization ability is selected online to integrate to estimate the dynamic offset value of the dominant variable with a certain time difference. Finally, based on the idea of TD model, the dominant variable value is predicted in real time. 3. Considering the stage characteristics of process nonlinearity and time-delay, a sliding window time-difference Gao Si process regression (Moving Window Time Difference Gaussian Process Regression,MWTDGPR modeling method based on local time-delay reconstruction (Local Time-delay Reconstruction,LTR) is proposed. The method uses sliding window and TD combination strategy to track the local nonlinear mutation and slowly varying feature step by step, and synchronously realizes the timing correction and reconstruction of the local sample set. It provides a feasible framework for other nonlinear time-varying chemical process modeling problems with time delay. The feasibility and accuracy of the above method are verified by the data simulation of the actual process. The simulation results show that the adaptive soft-sensor modeling with time-delay estimation is of great significance for the economic benefit and safe and stable operation of chemical process.
【學(xué)位授予單位】：江南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TQ018
【參考文獻(xiàn)】
相關(guān)期刊論文 前10條
1 熊偉麗;張偉;徐保國(guó);;一種基于EGMM的高斯過(guò)程回歸軟測(cè)量建模[J];信息與控制;2016年01期
2 孫茂偉;楊慧中;;基于改進(jìn)Bagging算法的高斯過(guò)程集成軟測(cè)量建模[J];化工學(xué)報(bào);2016年04期
3 雙翼帆;顧幸生;;基于改進(jìn)的快速搜索聚類算法和高斯過(guò)程回歸的催化重整脫氯前氫氣純度多模型建模方法[J];化工學(xué)報(bào);2016年03期
4 張偉;熊偉麗;徐保國(guó);;基于實(shí)時(shí)學(xué)習(xí)的高斯過(guò)程回歸多模型融合建模[J];信息與控制;2015年04期
5 李翔宇;高憲文;侯延彬;;基于在線動(dòng)態(tài)高斯過(guò)程回歸抽油井動(dòng)液面軟測(cè)量建模[J];化工學(xué)報(bào);2015年06期
6 阮宏鎂;田學(xué)民;王平;;基于聯(lián)合互信息的動(dòng)態(tài)軟測(cè)量方法[J];化工學(xué)報(bào);2014年11期
7 邵偉明;田學(xué)民;王平;;基于局部PLS的多輸出過(guò)程自適應(yīng)軟測(cè)量建模方法(英文)[J];Chinese Journal of Chemical Engineering;2014年07期
8 王振雷;唐苦;王昕;;一種基于D-S和ARIMA的多模型軟測(cè)量方法[J];控制與決策;2014年07期
9 阮宏鎂;田學(xué)民;王平;;帶時(shí)延估計(jì)的時(shí)間差分PLS軟測(cè)量建模方法[J];石油化工自動(dòng)化;2013年06期
10 許少鵬;韓紅桂;喬俊飛;;基于模糊遞歸神經(jīng)網(wǎng)絡(luò)的污泥容積指數(shù)預(yù)測(cè)模型[J];化工學(xué)報(bào);2013年12期
，本文編號(hào)：2219287
本文鏈接：http://sikaile.net/kejilunwen/huagong/2219287.html
最近更新
教材專著
