LS-SVM model based nonlinear predictive control for MCFC system

This paper describes a nonlinear model predictive controller for regulating a molten carbonate fuel cell (MCFC). In order to improve MCFC’s generating performance, prolong its life and guarantee safety, it must be controlled efficiently. First, the output voltage of an MCFC stack is identified by a least squares support vector machine (LS-SVM) method with radial basis function (RBF) kernel so as to implement nonlinear predictive control. And then, the optimal control sequences are obtained by applying genetic algorithm (GA). The model and controller have been realized in the MATLAB environment. Simulation results indicated that the proposed controller exhibits satisfying control effect.     

LS-SVM的核参数对概率筛筛分效率预测影响

以探索概率筛振动参数与筛分效率之间的关系,为概率筛结构的进一步改进提供指导意义为研究目的,将LS-SVM分类算法引入自同步概率筛筛分效率预测建模,探讨LS-SVM建模的可行性。基于各个不同的应用领域,可以构造不同的核函数,针对核函数需要优化特征参数的问题,应用网格搜索和交叉验证算法,对核参数的选择进行优化。通过研究得出用多项式(Poly)核函数建模对预测样本的最高预测识别率达到96.7%,采用RBF核函数建模对预测样本达到了零错分率,表明将LS-SVM算法引入概率筛筛分效率预测建模是可行的。     

基于核方法的改进可能聚类模型

A novel model of fuzzy clustering using kernel methods is proposed. This model is called kernel modified possibilisticc-means (KMPCM) model. The proposed model is an extension of the modified possibilistic c-means (MPCM) algorithm byusing kernel methods. Different from MPCM and fuzzy c-means (FCM) model which are based on Euclidean distance, theproposed model is based on kernel-induced distance. Furthermore, with kernel methods the input data can be mappedimplicitly into a high-dimensional feature space where the nonlinear pattern now appears linear. It is unnecessary to docalculation in the high-dimensional feature space because the kernel function can do it. Numerical experiments show thatKMPCM outperforms FCM and MPCM.     

Interior-point algorithm based on general kernel function for monotone linear complementarity problem

A polynomial interior-point algorithm is presented for monotone linear complementarity problem (MLCP) based on:a class of kernel functions with the general barrier term, which are called general kernel functions. Under the mild conditions for the barrier term, the complexity bound of algorithm in terms of such kernel function and its derivatives is obtained. The approach is actually an extension of the existing work which only used the specific kernel functions for the MLCP.     

Kernel Model Applied in Kernel Direct Discriminant Analysis for the Recognition of Face with Nonlinear Variations

A kernel-based discriminant analysis method called kernel direct discriminant analysis is employed, which combines the merit of direct linear discriminant analysis with that of kernel trick. In order to demonstrate its better robustness to the complex and nonlinear variations of real face images , such as illumination, facial expression, scale and pose variations, experiments are carried out on the Olivetti Research Laboratory, Yale and self-built face databases. The results indicate that in contrast to kernel principal component analysis and kernel linear discriminant analysis, the method can achieve lower (7%) error rate using only a very small set of features. Furthermore, a new corrected kernel model is proposed to improve the recognition performance. Experimental results confirm its superiority (1% in terms of recognition rate) to other polynomial kernel models.     

Nonlinear decoupling controller design based on least squares support vector regression

INTRODUCTION Most practical systems are multivariate nonlin- ear systems. In general, the MIMO (multiple inputs and multiple outputs) systems are coupled. This cou- pling affects the effectiveness of a specific loop con- troller on the corresponding output, and in some case, may become serious and cause many difficulties to the control system design. How to decouple the mul- tivariate systems and design practical controllers is one of the major issues in nonlinear control area. In recen…     

基于LS-SVM的福房价格指数预测

提出一种基于最小二乘支持向量机(LS-SVM)的福房指数预测方法.采用感知机核函数、多项式核函数和高斯核函数进行仿真模拟,经过参数选优建立了精度较高的福房指数预测模型.预测结果表明,利用LS-SVM模型进行预测具有误差小、拟合程度高等优点,可适用于房地产价格指数的预测.     

基于递阶遗传算法的神经网络预测控制(英文)

针对具有时延非线性系统提出一种神经网络预测控制．相应的预测模型通过递阶遗传算法离线训练数据并实时多步递推得到;性能函数中引入模型误差校正以克服模型失配及干扰等对系统造成的影响．仿真结果表明该系统具有良好动态响应及鲁棒性．     

Predictive control of a class of bilinear systems based on global off-line models

INTRODUCTION Bilinear systems are a kind of important nonlinear systems with relatively simple structure, and many industrial processes can be described as a bilinear system. Thus research on the control of this kind of systems is very important. On the other hand, model predictive control (MPC) (Clarke et al., 1987) has been widely used in industrial applications and many predictive control methods focusing on bilinear systems are emerging (Bloemen et al., 2001; Fontes et al., 2004; He…     

基于二元B样条基的核函数及支持向量机研究

在支持向量机方法中,核函数起着关键的作用.利用2-型三角剖分上的样条函数空间(S21(Δmn))的一组二元B样条基构造了一种新的核函数,并应用到二元函数的逼近中,取得了良好的效果.     

Robust model predictive control for discrete uncertain nonlinear systems with time-delay via fuzzy model

INTRODUCTION In recent years, research on nonlinear model predictive control has attracted significant attention because most practical processes are nonlinear. In industry process control, model predictive controllers based on linear models (LMPC) are often used to control nonlinear systems. When the operating con- ditions undergo significant changes, the performance of LMPC can deteriorate drastically. Under such conditions, predictive control based on nonlinear models (NMPC) shoul…     

一类非线性系统的广义预测控制

对于时延的非线性系统,将其等价为含有纯滞后的时变线性系统,利用三次样条函数逼近时变参数,利用渐消记忆递推最小二乘法在线辨识多项式的参数以实现时变参数估计,将基于Hammerstein模型的非线性系统线性化.最后运用广义预测控制策略实现对该系统的仿真,结果表明被控对象在广义预测控制算法的控制下系统闭环稳定性好、稳态误差小.     

Continuous robust tracking controllers for a class of uncertain nonlinear dynamical systems

Robust tracking controller for a class of uncertain nonlinear dynamical systems, which are linearizable by input-output feedback with matching uncertainties, was investigated. In this study, uniform ultimate bound or uniformly asymptotic stability of tracking errors were obtained by different choice of the control gain. A simulation to determine the effectiveness of the proposed approach showed that the control performance was better than that of VSC (Variable Structure Control). Project(69934030) supported by Natural Science Foundation of China.     

Web客户的核聚类

提出了一种基于核的聚类方法,增加对样本特征的优化;通过核函数,把数据样本空间映射到一个高维的特征空间;在特征空间对数据样本进行k-中心点聚类,并通过计算类内距离作为适应度准则,取其最优的结果。通过Web日志挖掘中的Web客户聚类应用比较,表明核聚类方法在性能上比经典的聚类算法有较大的改进,从而实现更为准确的聚类。     

基于快速并行优化算法的非线性预测控制

从建立稳定、精确的非线性预测模型出发,以Broyden,Fletcher,Goldfarb和Sharmo(BFGS)的拟牛顿优化算法为基础,在神经网络的训练过程中引入两个含有不同参数的拟牛顿校正公式并行的确定搜索方向,通过不精确搜索法求取最优步长,最后根据性能指标取最优的一个搜索方向和相应的步长对网络各层之间的权值进行修正。控制器采用神经网络递推多步预测、自补偿式在线闭环反馈校正以及迭代学习求解进行优化的方法。Matlab仿真结果验证了该方法的有效性。     

基于小波核的LS—SVM算法在城市交通流预测中与其他核函数对比研究

针对城市交通流短时流量预测在智能交通系统中的重要性,在以往研究基础上采用Morlet函数作为小波核函数,进一步提高了模型的预测精度及泛化能力．将其与其他几种常用核函数模型进行比较,其效果明显优于其他核函数模型,能够满足智能交通控制和诱导的要求．     

CONTINUOUS ROBUST TRACKING CONTROLLERS FOR A CLASS OF UNCERTAIN NONLINEAR DYNAMICAL SYSTEMS

ＩＮＴＲＯＤＵＣＴＩＯＮＩｎｒｅｃｅｎｔｙｅａｒｓ,ｒｏｂｕｓｔｃｏｎｔｒｏｌｏｆｄｙｎａｍｉｃａｌｓｙｓｔｅｍｓｗｉｔｈｓｉｇｎｉｆｉｃａｎｔｕｎｃｅｒｔａｉｎｔｉｅｓｗａｓｗｉｄｅｌｙｒｅｓｅａｒｃｈｅｄ .Ｍａｎｙｎｅｗａｐｐｒｏａｃｈｅｓｈａｖｅｂｅｅｎｐｒｏｐｏｓｅｄｂａｓｅｄｏｎｓｏｍｅｄｅｓｉｒｅｄｐｅｒｆｏｒｍａｎｃｅｓ.ＴｈｅＬｙａｐｕｎｏｖｆｕｎｃｔｉｏｎｃｏｍｍｏｍｌｙｕｓｅｄｔｏｓｙｎ ｔｈｅｓｉｚｅｔｈｅｃｏｎｔｒｏｌｌｅｒ(Ｌｉ,1 996;Ｙａｌｌａｐｒａｇａｄａｅｔａｌ.,1 996…     

基于黄金分割法的RBF神经网络预测控制

针对非线性系统的控制问题,提出了一种基于黄金分割法的RBF神经网络预测控制算法。该算法以神经网络作为预测模型,用黄金分割法优化控制器,其中以控制变量的约束条件作为优化的初始区间。针对化工过程蒸馏塔控制系统,通过仿真计算验证了该方法的有效性。     

常系数线性分数阶微分方程组的解

文章研究了常系数线性分数阶微分方程的求解问题,利用Mittag—Leffler函数及其Laplace变换,提出了某些类别的常系数线性分数阶微分方程的求解问题,且得到了一些解线性分数阶微分方程的方法．     

动态投入产出模型的新状态方程

本文采用各期各部门生产量的增量为控制变量,构造动态投入产出模型的控制系统,使所建立的控制系统实际意义具体、对原模型无特别的要求、能减少各种有关运算。