基于核函数及参数优化的KPLS质量预测研究
2021年电子技术应用第12期
陈 路,郑 丹,童楚东
宁波大学 信息科学与工程学院,浙江 宁波315211
摘要: 核偏最小二乘(KPLS)在工业过程监测和质量预测中得到了广泛的应用,核函数和核参数的选取对KPLS质量预测结果有重要影响。然而,如何选择核函数类型和核参数一直是该方法应用的瓶颈。针对以上问题,提出一种改进遗传算法的核函数优化方法。该方法将核的种类及核参数作为优化的决策变量,以均方根误差为目标,分别从编码方案、遗传策略、适应度函数优化、交叉和变异算法等方面进行设计,以保证核函数种类的多样性,利用2折交叉验证法对训练结果进行验证。以田纳西-伊斯曼过程(TE)与MATLAB结合进行仿真实验,仿真结果表明,该方法能寻找到最优核函数以及其核参数,具有很好的稳定性和一致性。
中圖分類號: TN081;TP277
文獻(xiàn)標(biāo)識碼: A
DOI:10.16157/j.issn.0258-7998.201259
中文引用格式: 陳路,鄭丹,童楚東. 基于核函數(shù)及參數(shù)優(yōu)化的KPLS質(zhì)量預(yù)測研究[J].電子技術(shù)應(yīng)用,2021,47(12):100-104.
英文引用格式: Chen Lu,Zheng Dan,Tong Chudong. The optimization of the kind and parameters of kernel function in KPLS for quality prediction[J]. Application of Electronic Technique,2021,47(12):100-104.
文獻(xiàn)標(biāo)識碼: A
DOI:10.16157/j.issn.0258-7998.201259
中文引用格式: 陳路,鄭丹,童楚東. 基于核函數(shù)及參數(shù)優(yōu)化的KPLS質(zhì)量預(yù)測研究[J].電子技術(shù)應(yīng)用,2021,47(12):100-104.
英文引用格式: Chen Lu,Zheng Dan,Tong Chudong. The optimization of the kind and parameters of kernel function in KPLS for quality prediction[J]. Application of Electronic Technique,2021,47(12):100-104.
The optimization of the kind and parameters of kernel function in KPLS for quality prediction
Chen Lu,Zheng Dan,Tong Chudong
Faculty of Electrical Engineering and Computer Science,Ningbo University,Ningbo 315211,China
Abstract: Kernel partial least squares(KPLS) has been widely used in industrial process monitoring and quality prediction. The choice of kernel function and kernel parameters has an important impact on the KPLS quality prediction results. However, how to choose the kernel function type and kernel parameters has always been the bottleneck of the application of this method. To solve the above problems, a kernel function optimization method based on improved genetic algorithm is proposed. In this method, the kernel type and kernel parameters are used as the optimal decision variables, and the root mean square error is targeted. It is designed in terms of coding scheme, genetic strategy, fitness function optimization, crossover and mutation algorithms to ensure the variety of kernel functions, and uses the 2-fold cross-validation method to verify the training results. The Tennessee-Eastman Process(TE) is combined with MATLAB for simulation experiments. The simulation results show that the method can find the optimal kernel function and its kernel parameters, and has good stability and consistency.
Key words : kernel partial least squares;genetic algorithm;quality prediction;k-fold cross-validation
0 引言
質(zhì)量預(yù)測與分析是實(shí)現(xiàn)工業(yè)過程閉環(huán)控制的基礎(chǔ)和關(guān)鍵[1]?;贙PLS的方法可以提高質(zhì)量預(yù)測精度,許多研究人員以KPLS方法為基石,提出了許多解決非線性問題的方法[1-8]。
核函數(shù)是KPLS方法的關(guān)鍵,而KPLS選擇核函數(shù)并不是任意的,必須要滿足Mercer定理。特定的內(nèi)核函數(shù)選擇隱含地決定了映射和特征空間。在KPLS中,由于提取系統(tǒng)非線性特征的程度是基于核函數(shù)的,因此核函數(shù)的選擇是最重要的。如何給基于KPLS的質(zhì)量預(yù)測選擇理想的核函數(shù)和核參數(shù)是一個開放的問題[9-10]。而且,一旦設(shè)置了核函數(shù),就需要設(shè)置適當(dāng)?shù)暮藚?shù)。但是,沒有一個理論框架能尋找到指定核函數(shù)的參數(shù)最最優(yōu)值,也就是說基于KPLS的質(zhì)量預(yù)測很大程度上取決于選擇的核函數(shù)和核參數(shù)。
本文詳細(xì)內(nèi)容請下載:http://m.ihrv.cn/resource/share/2000003880。
作者信息:
陳 路,鄭 丹,童楚東
(寧波大學(xué) 信息科學(xué)與工程學(xué)院,浙江 寧波315211)

此內(nèi)容為AET網(wǎng)站原創(chuàng),未經(jīng)授權(quán)禁止轉(zhuǎn)載。
