發(fā)布時(shí)間：2018-03-29 20:00

  本文選題：相關(guān)向量機(jī)　切入點(diǎn)：核參數(shù)　出處：《江南大學(xué)》2017年碩士論文

【摘要】：幾乎所有的工業(yè)生產(chǎn)的最終目標(biāo)都是獲得滿足要求的高質(zhì)量產(chǎn)品,因此在保證安全生產(chǎn)的前提下,質(zhì)量控制是生產(chǎn)過(guò)程的核心。軟測(cè)量技術(shù)的出現(xiàn)雖然一定程度上克服和彌補(bǔ)了傳感器以及離線檢測(cè)的不足,但是工業(yè)生產(chǎn)過(guò)程要求測(cè)量數(shù)據(jù)實(shí)時(shí)和精準(zhǔn),逐漸對(duì)軟測(cè)量的發(fā)展提出了更高的要求。因此建模方法的改進(jìn)和優(yōu)化算法的引入,對(duì)簡(jiǎn)化軟測(cè)量模型的結(jié)構(gòu)、提高軟測(cè)量模型的估計(jì)精度和提高建模的速度上有重要意義。本文重點(diǎn)研究了兩種基于核函數(shù)的建模方法—相關(guān)向量機(jī)和快速相關(guān)向量機(jī),以及通過(guò)引入優(yōu)化算法對(duì)其性能的相關(guān)改進(jìn):1、提出了一種組合核函數(shù)的軟測(cè)量建模方法。為了同時(shí)得到較強(qiáng)的回歸能力和較好的稀疏性,在對(duì)相關(guān)向量機(jī)構(gòu)造一個(gè)組合核函數(shù)的同時(shí),又構(gòu)建了一個(gè)綜合回歸性能和稀疏性的適應(yīng)度函數(shù),并利用遺傳算法優(yōu)化相關(guān)向量機(jī)組合核的權(quán)系數(shù)和核參數(shù)。將該方法用于一個(gè)雙酚A生產(chǎn)流程中裂解回收單元的建模仿真。仿真實(shí)例表明,所提方法的估計(jì)精度(0.3721)、稀疏性(41)等指標(biāo)均優(yōu)于一般的支持向量機(jī)組合核模型(0.7327)和GA-RVM單一核相關(guān)向量機(jī)模型(poly核0.9422、gauss核0.7571)。2、現(xiàn)代化流程工業(yè)設(shè)備眾多、工藝復(fù)雜,影響生產(chǎn)過(guò)程中各環(huán)節(jié)技術(shù)指標(biāo)的因素繁多,導(dǎo)致某些質(zhì)量變量在線檢測(cè)困難。因此,如何從可檢測(cè)的生產(chǎn)操作變量中提取有效特征并快速有效的確定模型參數(shù)一直是研究的熱點(diǎn)。針對(duì)此問(wèn)題,本章利用核主成分分析對(duì)軟測(cè)量模型的多特征輸入變量進(jìn)行特征提取,建立相關(guān)向量機(jī)軟測(cè)量回歸模型,鑒于核參數(shù)對(duì)于核主成分分析和相關(guān)向量機(jī)模型性能的影響,又采用HS算法對(duì)KPCA和RVM的核參數(shù)進(jìn)行同時(shí)尋優(yōu),構(gòu)建了一個(gè)基于HS算法優(yōu)化的KPCA-RVM軟測(cè)量模型。仿真結(jié)果表明,通過(guò)HS算法優(yōu)化的KPCA-RVM軟測(cè)量模型估計(jì)精度(0.234888)和運(yùn)算速度(159.57)明顯優(yōu)于HS-SVM算法(0.255969)、GA-RVM算法(0.25423)和HS-RVM算法(0.254186),取得了良好的效果。3、針對(duì)應(yīng)用在現(xiàn)代流程工業(yè)中的軟測(cè)量模型需要滿足數(shù)據(jù)處理量大、估計(jì)精度高、實(shí)時(shí)性強(qiáng)的要求,將快速相關(guān)向量機(jī)(FRVM)代替相關(guān)向量機(jī)用于建立軟測(cè)量回歸模型,降低了計(jì)算復(fù)雜度、減少了訓(xùn)練時(shí)間;同時(shí),為了快速準(zhǔn)確的確定快速相關(guān)向量機(jī)的核函數(shù)參數(shù),提出了一種具有非線性音調(diào)微調(diào)概率的和改進(jìn)優(yōu)化變量的初始選擇方法的改進(jìn)和聲搜索算法用于尋優(yōu)FRVM的核參數(shù)。仿真結(jié)果表明,本章提出的改進(jìn)方法有效的解決了和聲搜索算法容易陷入局部最優(yōu)的不足,并且該方法的估計(jì)精度(0.2324)和運(yùn)行速度(94.76)明顯優(yōu)于基于線性變化PAR的HS算法(0.2815)和固定PAR的HS算法(0.2782)。
[Abstract]:The ultimate goal of almost all industrial production is to obtain high quality products that meet the requirements, thus ensuring the safety of production, Quality control is the core of the production process. Although the emergence of soft sensing technology to some extent overcomes and makes up for the shortcomings of sensor and off-line detection, the industrial production process requires real-time and accurate measurement data. Therefore, the improvement of modeling method and the introduction of optimization algorithm can simplify the structure of soft sensor model. It is very important to improve the estimation accuracy of soft sensor model and the speed of modeling. In this paper, two modeling methods based on kernel function, correlation vector machine and fast correlation vector machine, are studied. And by introducing the correlation improvement of optimization algorithm to its performance, a soft sensor modeling method of combining kernel function is proposed. In order to obtain better regression ability and better sparseness at the same time, a soft sensor modeling method of combining kernel function is proposed. At the same time, a combination kernel function is constructed for the correlation vector mechanism, and a fitness function combining regression performance and sparsity is constructed. The genetic algorithm is used to optimize the weight coefficient and kernel parameters of the combined kernel of correlation vector machine. The method is applied to modeling and simulation of the pyrolysis recovery unit in a bisphenol A production process. The simulation example shows that, The estimated accuracy of the proposed method is 0.3721, and the sparsity of the proposed method is better than that of the general support vector machine combined kernel model (0.7327) and the GA-RVM single kernel correlation vector machine model (Poly kernel 0.9422 gauss 0.7571g 路2). The modern process industrial equipment is numerous and the process is complex. There are many factors that affect the technical index of every link in the process of production, which leads to the difficulty of on-line detection of some quality variables. How to extract effective features from detectable production operation variables and determine model parameters quickly and effectively has been a hot topic. In this chapter, the kernel principal component analysis (KPCA) is used to extract the multi-feature input variables of the soft-sensor model, and the soft-sensing regression model of the correlation vector machine is established. In view of the effect of kernel parameters on the performance of the kernel PCA and the correlation-vector machine model, HS algorithm is used to simultaneously optimize the kernel parameters of KPCA and RVM, and a KPCA-RVM soft sensor model based on HS algorithm is constructed. The simulation results show that, The estimation accuracy of KPCA-RVM soft sensor model optimized by HS algorithm is 0.234888) and the speed of calculation is 159.57). It is obviously superior to HS-SVM algorithm 0.255969U GA-RVM algorithm 0.25423) and HS-RVM algorithm 0.254186.The result is good. 3. According to the requirement of soft sensor model applied in modern process industry, it is better than that of HS-SVM algorithm (0.25423) and HS-RVM algorithm (0.2541866). Meet the large amount of data processing, The fast correlation vector machine (FRVM) is used to build the soft sensor regression model instead of the correlation vector machine, which reduces the computational complexity and the training time, at the same time, the fast correlation vector machine (FRVM) is used instead of the correlation vector machine to build the soft sensor regression model. In order to determine the kernel function parameters of fast correlation vector machine quickly and accurately, An improved harmonic search algorithm with nonlinear tonal fine-tuning probability and an improved initial selection method for optimization variables is proposed to optimize the kernel parameters of FRVM. The simulation results show that, The improved method proposed in this chapter effectively solves the problem that the harmonic search algorithm is easy to fall into local optimum, and its estimation accuracy is 0.2324) and its running speed is 94.76), which is obviously superior to the HS algorithm based on linearly varying PAR (0.2815) and the HS algorithm based on fixed PAR (0.2782).
【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)】：TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 黃凱;陳勇;母志為;何躍;;基于人工神經(jīng)網(wǎng)絡(luò)和遺傳算法的甲烷制氫催化劑設(shè)計(jì)[J];化工學(xué)報(bào);2016年08期

2 黨華;仇異;郭軍強(qiáng);王衛(wèi)江;;基于相空間重構(gòu)和核主分量的水聲信號(hào)增強(qiáng)[J];北京理工大學(xué)學(xué)報(bào);2016年03期

3 李潔;張兆薇;;基于和聲搜索算法和相關(guān)向量機(jī)的網(wǎng)絡(luò)安全態(tài)勢(shì)預(yù)測(cè)方法[J];計(jì)算機(jī)應(yīng)用;2016年01期

4 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

5 白霜;;基于PSO優(yōu)化混合RVM模型的進(jìn)出口貿(mào)易預(yù)測(cè)算法[J];計(jì)算機(jī)與現(xiàn)代化;2014年08期

6 周雅蘭;黃韜;;和聲搜索算法改進(jìn)與應(yīng)用[J];計(jì)算機(jī)科學(xué);2014年S1期

7 歐陽(yáng)海濱;高立群;鄒德旋;孔祥勇;;和聲搜索算法探索能力研究及其修正[J];控制理論與應(yīng)用;2014年01期

8 何志昆;劉光斌;趙曦晶;王明昊;;高斯過(guò)程回歸方法綜述[J];控制與決策;2013年08期

9 王宏;;認(rèn)識(shí)基于數(shù)據(jù)驅(qū)動(dòng)的工業(yè)過(guò)程控制[J];控制工程;2013年02期

10 高立群;依玉峰;鄭平;程偉;;和聲搜索算法在求解最短路徑問(wèn)題中的應(yīng)用[J];東北大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年06期

相關(guān)會(huì)議論文 前1條

1 陳軍;張素貞;徐順喜;;PET終縮聚過(guò)程質(zhì)量控制方案的研究[A];1994中國(guó)控制與決策學(xué)術(shù)年會(huì)論文集[C];1994年

相關(guān)碩士學(xué)位論文 前7條

1 董陶;基于K-means聚類的軟測(cè)量建模研究[D];江南大學(xué);2013年

2 鄧衛(wèi)衛(wèi);多模型軟測(cè)量建模方法研究及其應(yīng)用[D];江南大學(xué);2012年

3 劉曉飛;基于核函數(shù)優(yōu)化的KPCA方法及其在發(fā)酵過(guò)程中應(yīng)用[D];東北大學(xué);2011年

4 陳定三;基于聚類的多模型建模及其在軟測(cè)量中的應(yīng)用[D];江南大學(xué);2011年

5 成亮鋮;雙酚A生產(chǎn)過(guò)程軟測(cè)量混合建模的研究[D];江南大學(xué);2009年

6 高緒偉;核PCA特征提取方法及其應(yīng)用研究[D];南京航空航天大學(xué);2009年

7 瞿偉;基于數(shù)據(jù)驅(qū)動(dòng)的軟測(cè)量建模方法研究及其工業(yè)應(yīng)用[D];浙江大學(xué);2008年

，

本文編號(hào)：1682621