On-Line Batch Process Monitoring Using Multiway Kernel Partial Least Squares

An approach for batch processes monitoring and fault detection based on multiway kernel partial least squares(MKPLS) was presented.It is known that conventional batch process monitoring methods,such as multiway partial least squares(MPLS),are not suitable due to their intrinsic linearity when the variations are nonlinear.To address this issue,kernel partial least squares(KPLS) was used to capture the nonlinear relationship between the latent structures and predictive variables.In addition,KPLS requires only linear algebra and does not involve any nonlinear optimization.In this paper,the application of KPLS was extended to on-line monitoring of batch processes.The proposed batch monitoring method was applied to a simulation benchmark of fed-batch penicillin fermentation process.And the results demonstrate the superior monitoring performance of MKPLS in comparison to MPLS monitoring.

A Novel Extension of Kernel Partial Least Squares Regression

Based on continuum power regression(CPR) method, a novel derivation of kernel partial least squares(named CPR-KPLS) regression is proposed for approximating arbitrary nonlinear functions.Kernel function is used to map the input variables(input space) into a Reproducing Kernel Hilbert Space(so called feature space),where a linear CPR-PLS is constructed based on the projection of explanatory variables to latent variables(components). The linear CPR-PLS in the high-dimensional feature space corresponds to a nonlinear CPR-KPLS in the original input space. This method offers a novel extension for kernel partial least squares regression(KPLS),and some numerical simulation results are presented to illustrate the feasibility of the proposed method.

Near-Infrared Spectroscopy Coupled with Kernel Partial Least Squares-Discriminant Analysis for Rapid Screening Water Containing Malathion

Near-infrared spectroscopy coupled with kernel partial least squares-discriminant analysis was used to rapidly screen water containing malathion. In the wavenumber of 4348 cm-1 to 9091 cm-1, the overall correct classification rate of kernel partial least squares-discriminant analysis was 100% for training set, and 100% for test set, with the lowest concentration detected malathion residues in water being 1 μg·ml-1. Kernel partial least squares-discriminant analysis was able to have a good performance in classifying data in nonlinear systems. It was inferred that Near-infrared spectroscopy coupled with the kernel partial least squares-discriminant analysis had a potential in rapid screening other pesticide residues in water.

Improved Kernel PLS-based Fault Detection Approach for Nonlinear Chemical Processes

In this paper, an improved nonlinear process fault detection method is proposed based on modified kernel partial least squares(KPLS). By integrating the statistical local approach(SLA) into the KPLS framework, two new statistics are established to monitor changes in the underlying model. The new modeling strategy can avoid the Gaussian distribution assumption of KPLS. Besides, advantage of the proposed method is that the kernel latent variables can be obtained directly through the eigen value decomposition instead of the iterative calculation, which can improve the computing speed. The new method is applied to fault detection in the simulation benchmark of the Tennessee Eastman process. The simulation results show superiority on detection sensitivity and accuracy in comparison to KPLS monitoring.

QSAR Studies on PCDD/Fs by Kernel PLS

QSPR models of PCDD/Fs were generated by means of kernel partial least squares. The molecular distance-edge vector method was used as descriptors to get model I for predicting PCDD/Fs retention behavior. The chlorinated positions were also used and model II was obtained. In studied cases, the predictive ability of the KPLS model is comparable or superior to those of PLS and ANN. The results indicate that KPLS can be used as an alternative powerful modeling tool for QSPR studies.

Quantitative Analysis of Methanol in Methanol Gasoline by Calibration Transfer Strategy Based on Kernel Domain Adaptive Partial Least Squares(kda-PLS)

The application of near-infrared(NIR)spectroscopy combined with multivariate calibration methods can achieve the rapid analysis of methanol gasoline.However,instrumental or environmental differences found for spectra make it impossible to continuously apply the previously developed calibration model.Therefore,the calibration transfer technique would be required to solve the time-consuming and laborious problem of reestablishing a new model.In this work,a calibration transfer method named kernel domain adaptive partial least squares(kda-PLS)was applied to the calibration transfer from the primary instrument to the secondary ones.Firstly,wavelet transform(WT)and variable importance in projection(VIP)were employed to enhance the predictive performance of the kda-PLS transfer model.Then,the results found for the calibration transfer by piecewise direct standardization(PDS)and domain adaptive partial least squares(da-PLS)were compared to verify the calibration transfer(CT)effect of kda-PLS.The results point that the kda-PLS method can transfer the PLS model developed on the primary instrument to the secondary ones,and achieve results comparable to the those of reestablishing a new PLS model on the secondary instrument,with R_(P)^(2)=0.9979(R_(P)^(2):coefficients of determination of the prediction set),RMSEP=0.0040(RMSEP:root mean square error of the prediction set),and MREP=3.03%(MREP:mean relative error of the prediction set).Therefore,kda-PLS will provide a new method for quantitative analysis of methanol content in methanol gasoline.

Short Term Electric Load Prediction by Incorporation of Kernel into Features Extraction Regression Technique

Accurate load prediction plays an important role in smart power management system, either for planning, facing the increasing of load demand, maintenance issues, or power distribution system. In order to achieve a reasonable prediction, authors have applied and compared two features extraction technique presented by kernel partial least square regression and kernel principal component regression, and both of them are carried out by polynomial and Gaussian kernels to map the original features’ to high dimension features’ space, and then draw new predictor variables known as scores and loadings, while kernel principal component regression draws the predictor features to construct new predictor variables without any consideration to response vector. In contrast, kernel partial least square regression does take the response vector into consideration. Models are simulated by three different cities’ electric load data, which used historical load data in addition to weekends and holidays as common predictor features for all models. On the other hand temperature has been used for only one data as a comparative study to measure its effect. Models’ results evaluated by three statistic measurements, show that Gaussian Kernel Partial Least Square Regression offers the more powerful features and significantly can improve the load prediction performance than other presented models.

基于MW-MKEPLS的多重时变间歇生产过程质量预测

间歇生产过程的多重时变特性和非线性使得质量预测问题变得复杂。为了提高间歇过程质量预测精度,提出了滑动窗多向核熵偏最小二乘(moving window multiway kernel entropy partial least squares,MW-MKEPLS)方法。首先采用滑动窗进行数据的动态更新获取,构建了滑动窗多重时变模型;然后在滑动窗多重时变模型下通过核函数将数据映射到高维特征空间,采用Renyi熵贡献度进行数据特征提取,更好地获取数据的信息熵和非线性;最后在KECA处理后的高维特征空间进行质量预测。通过青霉素生产发酵过程进行了实验验证,并与MKPLS和MKEPLS进行对比分析,结果表明所提方法的质量预测精度更高。

基于iPLS-KPCA的高温燃气红外光谱特征提取方法研究

高温燃气红外光谱特征是判断燃气成分和浓度的有效途径。针对高温燃气红外辐射特性复杂、建模难度高的问题,研究了一种基于间隔偏最小二乘(interval Partial Least Squares,iPLS)和核主成分分析(Kernel Principal Component Analysis,KPCA)的特征提取算法。首先通过iPLS进行预筛选,确定具有最优预测能力的特征光谱波段,避免单个子区间建模过程中有用吸收峰信息的遗失;其次,利用KPCA降低数据维度,保留贡献率高的关键特征,降低成分预测模型的复杂度。仿真结果表明,经过iPLS-KPCA方法特征提取后,预测模型的复杂度大幅下降,且预测能力显著提升。

基于核PLS方法的非线性过程在线监控

针对过程监控数据的非线性特点,提出了一种基于核偏最小二乘(KPLS)的监控方法。KPLS方法是将原始输入数据通过核函数映射到高维特征空间,然后在高维特征空间再进行偏最小二乘(PLS)运算。与线性PLS相比,KPLS方法能充分利用样本空间信息,建立起输入输出变量之间的非线性关系。与其他非线性PLS方法不同,KPLS方法只需要进行线性运算,从而避免非线性优化问题。在对过程进行监控时,首先采用KPLS方法建立模型,得到得分向量,然后计算出T2和SPE统计量及其相应的控制限。Tennessee Eastman(TE)模型上的仿真研究结果表明,所提方法比线性PLS方法具有更好的过程监控性能。

基于统计量模式分析的MKPLS间歇过程监控与质量预报

传统MKPLS是对数据矩阵的协方差矩阵进行分解,没有利用数据的高阶统计量等有用信息,在进行特征提取时会造成数据有用信息的丢失,导致故障识别效果差。为了解决此问题,提出了统计量模式分析(SPA)与多向核偏最小二乘(MKPLS)相结合的多向统计量模式分析的核偏最小二乘方法(MSPAKPLS)。该方法首先引入滑动窗技术构造样本的不同阶次统计量,将数据从原始的数据空间映射到统计量样本空间,然后利用核函数将统计量样本空间映射到高维核空间进行偏最小二乘分析,并对产品质量进行预测。最后将该方法应用到工业青霉素发酵过程中,并与传统方法进行比较,发现该方法具有更好的监控性能和预测性能。

基于核函数变换的PLS回归的非线性结构分析

基于核函数变换的PLS非线性回归模型既吸取了核函数能够拟合适应任意连续变化曲线的优点,又借鉴了偏最小二乘回归方法能够有效解决自变量集合高度相关的技术。在本文中针对多元加法模型,从理论和仿真试验的角度分别验证了,对于多个独立自变量对单因变量为非线性关系的数据系统,基于核函数变换的PLS回归方法不仅能够有效实现自变量对因变量的整体预测,而且能够提取各维自变量对因变量的单独非线性作用特征,从而确定数据系统内部的复杂非线性结构关系,增强了模型的可解释性。

基于KPLS模型的间歇过程产品质量控制

针对间歇过程所具有的非线性特性,提出了一种基于核偏最小二乘(KPLS)模型的最终产品质量控制策略。利用初始条件、批次展开后的过程数据以及最终产品质量建立了间歇过程的KPLS模型;采用基于主成分分析(PCA)映射的预估方法对未知的过程数据进行补充,实现了最终产品质量的在线预测。为了解决最终产品质量的控制,利用T2统计量确定KPLS模型的适用范围,并作为约束引入产品质量控制问题,提高控制策略的可行性;采用粒子群优化(PSO)算法实现了优化问题的高效求解。仿真结果表明,与基于偏最小二乘(PLS)模型的控制策略相比,所提出的方法具有更高的预测精度,且能有效解决产品质量控制中出现的各种问题。

在线KPLS建模方法及在磨机负荷参数集成建模中的应用

针对过程非线性、基于历史数据构建的离线模型泛化性差以及基于滑动窗口和每样本递推更新的在线建模方法难以均衡建模精度和建模速度等问题,提出了一种在线核偏最小二乘(On-line kernel partial least squares,OLKPLS)建模方法.该方法依据新样本与建模样本间的近似线性依靠(Approximate linear dependence,ALD)值和代表工业过程特性漂移幅度的阈值,选择有价值样本更新KPLS模型,并采用合成数据和Benchmark平台数据对该方法进行了仿真验证.针对基于离线历史数据建立的融合多传感器信息的磨机负荷参数集成模型难以适应磨矿过程时变特性的问题,提出了基于OLKPLS和在线自适应加权融合算法的在线集成建模方法,并通过实验球磨机的实际运行数据仿真验证了方法的有效性.

基于核函数的PLS丁苯橡胶聚合转化率软测量

针对企业对丁苯橡胶聚合转化率需在线高精度预测的需求,考虑实际工况复杂性和偏最小二乘(PLS)算法非线性处理能力的不足,分别建立了引入单核和混合核函数的丁苯橡胶聚合转化率PLS预测模型。工业数据仿真结果表明:基于核的PLS模型均可满足企业生产对预测精度的要求,即聚合转化率预测绝对误差大于1.5的比例不大于样本总数的10%,尤其是混合核PLS因兼有局部和全局特性,表现出更优的性能。

基于KPLS特征提取的WNN模拟电路软故障诊断

为提高模拟电路的软故障诊断能力,提出一种基于KPLS特征提取和WNN的集成诊断方法。首先利用KPLS良好的特征提取能力,构建故障样本集的主元特征集;然后,利用WNN解决复杂非线性问题的优势,建立主元特征集的故障识别模型;最后,由所建模型对各种故障模式进行诊断判定。Sallen-Key带通滤波器的仿真测试表明:该集成方法仅通过不到300次迭代计算即完成模型训练,诊断的总正确率达到96.7%,且9种模式中的6种达到100%正确率,从而验证了其可行性和有效性。

KPLS-SVM在缺失飞参数据估计中的应用

飞行参数的缺失给飞行事故调查工作带来了很大困难。将核的偏最小二乘法与支持向量机耦合,建立基于状态匹配的飞行参数估计模型可以较好地解决缺失飞参数据的估计问题。首先将初始输入映射到高维特征空间,进而利用偏最小二乘法在特征空间中提取对缺失飞参数据影响较强的得分向量,最后将提取的得分向量作为输入建立支持向量机模型。既克服了输入变量间的相关性问题,又降低了支持向量机的输入维数。仿真也说明了使用该方法估计缺失飞行参数的可行性和有效性。

基于GA-KPLSR的转炉终点碳含量的预测研究

终点碳含量是决定钢的种类和质量的关键因素,是转炉炼钢过程中最难控制的变量之一。建立了基于遗传算法的核偏最小二乘回归(GA-KPLSR)方法的终点碳含量的预测模型。数据仿真结果表明,基于GA-KPLSR的预测模型,不仅能高效的处理变量之间的非线性关系,而且能快速收敛至最优解,得出预测结果的均方误差比主元回归(PCR)和偏最小二乘回归(PLSR)分别降低了25.77%、23.27%;相对误差降低了29.55%、26.83%;绝对误差降低了27.22%、24.84%。该方法可为实际生产中的终点控制提供参考,提高生产效益。

局部KPLS特征提取的LSSVM软测量建模方法

针对复杂工业过程的非线性、变量间的强相关性以及工况时变的特点,提出了一种基于局部KPLS特征提取的LSSVM建模方法。该方法通过属性加权的欧式距离指标选取局部训练样本子集,利用KPLS算法对该子集进行特征提取,使用LSSVM算法在线建立局部软测量模型。实验结果表明,该方法可以有效实现特征提取,具有更好的推广能力和预测精度,比基于全局KPLS特征提取的LSSVM模型和未经特征提取的全局LSSVM模型具有更好的泛化能力。

基于混合核函数和PLS的胎面尺寸预测模型

为实现对生产过程中胎面尺寸的准确预测,分析胎面生产工艺,结合混合核函数和偏最小二乘法(PLS)的特点构造了一种基于混合核函数和PLS的软测量模型。该预测模型把辅助变量通过输入非线性函数映射到高维特征空间,在高维特征空间中采用线性建模方法建立模型后经过非线性函数映射回原数据空间形成非线性模型。应用效果表明,该模型具有较强的学习和泛化能力,并且具有较高的预测精度,能较好地满足胎面实际生产要求。