本文選題：軟測(cè)量　切入點(diǎn)：輔助變量選擇　出處：《浙江大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：近年來(lái)軟測(cè)量建模技術(shù)在化工生產(chǎn)過(guò)程中得到了廣泛應(yīng)用。軟測(cè)量技術(shù)根據(jù)某一最優(yōu)準(zhǔn)則,選擇一組與主導(dǎo)變量相關(guān)的且易測(cè)量的輔助變量,構(gòu)造以輔助變量為輸入,主導(dǎo)變量為輸出的數(shù)學(xué)模型,實(shí)現(xiàn)對(duì)主導(dǎo)變量的在線估計(jì)。雖然軟測(cè)量的核心是建模,但是只有選擇與主導(dǎo)變量密切相關(guān)的輔助變量才能建立一個(gè)有效的軟測(cè)量模型。變量選擇就是在軟測(cè)量模型基礎(chǔ)之上,從一系列預(yù)先給定的自變量集合中,確定一個(gè)在某種準(zhǔn)則下可以對(duì)主導(dǎo)變量進(jìn)行最佳描述的變量子集。假定有p個(gè)候選輔助變量,總共能產(chǎn)生2p-1個(gè)候選模型,即使p不是很大的時(shí)候,也能陷入組合爆炸的困境。因此,研究如何快速高效的變量選擇方法,在保證模型預(yù)測(cè)性能的前提下,盡可能地減少冗余變量,是很有必要的。針對(duì)該問(wèn)題,本論文開(kāi)展了較為系統(tǒng)化的軟測(cè)量變量選擇方法研究。本文的主要研究?jī)?nèi)容和成果如下:1.將蒙特卡洛無(wú)信息變量消除算法(MC-UVE)、遺傳算法和偏最小二乘(GA-PLS)三者相結(jié)合,首先利用MC-UVE算法剔除無(wú)信息變量,在MC-UVE所選出的有信息變量的基礎(chǔ)上,使用GA算法進(jìn)一步精選變量子集,提出了MC-UVE-GA-PLS變量選擇方法。最后,通過(guò)UCI數(shù)據(jù)對(duì)算法的可行性、有效性、模型預(yù)測(cè)性能及模型復(fù)雜度等方面進(jìn)行驗(yàn)證,并與A11-PLS模型和GA-PLS模型進(jìn)行了對(duì)比,結(jié)果表明了算法的可行性與可靠性。2.考慮到變量選擇本質(zhì)上是數(shù)學(xué)優(yōu)化問(wèn)題,以多元線性回歸(MLR)模型為基礎(chǔ),通過(guò)引入0-1決策變量,利用BIC準(zhǔn)則,將變量選擇問(wèn)題描述成一個(gè)嵌套的混合整數(shù)二次規(guī)劃(MIQP)問(wèn)題,并提出了嵌套式MIQP-MLR變量選擇方法,同時(shí)實(shí)現(xiàn)特征變量的選擇與預(yù)測(cè)模型的建立。最后,通過(guò)UCI數(shù)據(jù)對(duì)所提出的方法進(jìn)行驗(yàn)證,并與傳統(tǒng)的逐步回歸(Stepwise)變量選擇方法進(jìn)行對(duì)比,結(jié)果驗(yàn)證了所提出方法的有效性和實(shí)用性。3.在基于嵌套式MIQP-MLR變量選擇方法基礎(chǔ)上,進(jìn)一步將模型結(jié)構(gòu)從MLR拓展至魯棒性更強(qiáng)的支持向量回歸(SVR)模型,并利用改進(jìn)的MSE準(zhǔn)則,將變量選擇描述成一個(gè)混合整數(shù)線性規(guī)劃(MILP)問(wèn)題,提出了 MILP-SVR變量選擇方法。所提出的方法不僅不需要事先指定模型中的變量個(gè)數(shù),避免懲罰因子的調(diào)節(jié),而且求解效率更高。此外,SVR模型可以利用核技巧,實(shí)現(xiàn)非線性函數(shù)的擬合。最后通過(guò)UCI數(shù)據(jù)測(cè)試了算法的可行性與有效性,并與A11-SVR、Pearson-SVR和RFE-SVR對(duì)比,驗(yàn)證了方法的可靠性。4.將上述變量選擇方法應(yīng)用至某一工業(yè)精餾塔間苯二胺純度的軟測(cè)量建模中,為間苯二胺純度軟測(cè)量模型選擇輔助變量,并建立相應(yīng)的軟測(cè)量模型。實(shí)際的工業(yè)數(shù)據(jù)的仿真研究結(jié)果表明所提出方法的可靠性與高性能。
[Abstract]:In recent years, soft sensor modeling technology has been widely used in chemical production process. According to an optimal criterion, soft sensing technology selects a set of auxiliary variables related to dominant variables and easy to measure, and constructs auxiliary variables as input. The dominant variable is an output mathematical model, which realizes the on-line estimation of the dominant variable. Although the core of soft sensing is modeling, But only by selecting auxiliary variables closely related to dominant variables can an effective soft-sensor model be established. Variable selection is based on the soft-sensing model and from a set of predefined independent variables. Determine a subset of variables that can best describe the dominant variable under certain criteria. Assuming that there are p candidate auxiliary variables, a total of 2p-1 candidate models can be generated, even if p is not very large. Therefore, it is necessary to study how to select variables quickly and efficiently and reduce the redundant variables as much as possible while ensuring the prediction performance of the model. The main contents and achievements of this paper are as follows: 1. Combining Monte Carlo algorithm with MC-UVEG, genetic algorithm and partial least squares GA-PLS. First of all, the MC-UVE algorithm is used to eliminate the information variable, and on the basis of the information variable selected by MC-UVE, the MC-UVE-GA-PLS variable selection method is proposed by using GA algorithm to further select the subset of variables. Finally, the feasibility and validity of the algorithm are obtained through the UCI data. The model prediction performance and model complexity are verified and compared with A11-PLS model and GA-PLS model. The results show that the algorithm is feasible and reliable. Based on the multivariate linear regression model, by introducing 0-1 decision variables and using BIC criterion, the variable selection problem is described as a nested mixed integer quadratic programming (MIQP) problem, and a nested MIQP-MLR variable selection method is proposed. At the same time, the selection of feature variables and the establishment of prediction model are realized. Finally, the proposed method is verified by UCI data, and compared with the traditional stepwise regression method. The results verify the validity and practicability of the proposed method. Based on the nested MIQP-MLR variable selection method, the model structure is further extended from MLR to a more robust support vector regression (SVR) model, and the improved MSE criterion is used. The variable selection is described as a mixed integer linear programming (MILP) problem, and the MILP-SVR variable selection method is proposed. The proposed method not only does not need to specify the number of variables in the model in advance, but also avoids the adjustment of the penalty factor. In addition, the kernel technique can be used to fit nonlinear functions. Finally, the feasibility and effectiveness of the algorithm are tested by UCI data, and compared with A11-SVRSn-Pearson-SVR and RFE-SVR. The reliability of the method is verified. 4. The above variable selection method is applied to the soft sensor modeling of the purity of resorcinenediamine in a certain industrial distillation column, and the auxiliary variable is selected for the soft sensing model of the purity of resorcinenediamine. The simulation results of practical industrial data show the reliability and high performance of the proposed method.
【學(xué)位授予單位】：浙江大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TQ021.8;TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前6條

1 簡(jiǎn)葳s，

本文編號(hào)：1591706