Support vector machine-based multi-model predictive control

In this paper, a support vector machine-based multi-model predictive control is proposed, in which SVM classification combines well with SVM regression. At first, each working environment is modeled by SVM regression and the support vector machine network-based model predictive control （SVMN-MPC） algorithm corresponding to each environment is developed, and then a multi-class SVM model is established to recognize multiple operating conditions. As for control, the current environment is identified by the multi-class SVM model and then the corresponding SVMN-MPC controller is activated at each sampling instant. The proposed modeling, switching and controller design is demonstrated in simulation results.     

Max-plus-linear model-based predictive control for constrained hybrid systems：linear programming solution

In this paper, a linear programming method is proposed to solve model predictive control for a class of hybrid systems. Firstly, using the （max, ＋） algebra, a typical subclass of hybrid systems called max-plus-linear （MPL） systems is obtained. And then, model predictive control （MPC） framework is extended to MPL systems. In general, the nonlinear optimization approach or extended linear complementarity problem （ELCP） were applied to solve the MPL-MPC optimization problem. A new optimization method based on canonical forms for max-min-plus-scaling （MMPS） functions （using the operations maximization, minimization, addition and scalar multiplication） with linear constraints on the inputs is presented. The proposed approach consists in solving several linear programming problems and is more efficient than nonlinear optimization. The validity of the algorithm is illustrated by an example.     

A multiple-kernel LSSVR method for separable nonlinear system identification

In some nonlinear dynamic systems, the state variables function usually can be separated from the control variables function, which brings much trouble to the identification of such systems. To well solve this problem, an improved least squares support vector regression （LSSVR） model with multiple-kernel is proposed and the model is applied to the nonlinear separable system identification. This method utilizes the excellent nonlinear mapping ability of Morlet wavelet kernel function and combines the state and control variables information into a kernel matrix. Using the composite wavelet kernel, the LSSVR includes two nonlinear functions, whose variables are the state variables and the control ones respectively, in this way, the regression function can gain better nonlinear mapping ability, and it can simulate almost any curve in quadratic continuous integral space. Then, they are used to identify the two functions in the separable nonlinear dynamic system. Simulation results show that the multiple-kernel LSSVR method can greatly improve the identification accuracy than the single kernel method, and the Morlet wavelet kernel is more efficient than the other kernels.     

基于模糊神经模型的电厂协调预测控制

In unit steam-boiler generation, a coordinated control strategy is required to ensure a higher rate of load change without violating thermal constraints. The process is characterized by nonlinearity and uncertainty. While neural networks can model highly complex nonlinear dynamical systems, they produce black box models. This has led to significant interest in neuro-fuzzy networks (NFNs) to represent a nonlinear dynamical process by a set of locally valid and simpler submodels. Two alternative methods of exploiting the NFNs within a generalised predictive control (GPC) framework for nonlinear model predictive control are described. Coordinated control of steam-boiler generation using the two nonlinear GPC methods show excellent tracking and disturbance rejection results and improved performance compared with conventional linear GPC.     

Generalized Predictive Control with Online Least Squares Support Vector Machines

This paper proposes a practical generalized predictive control （GPC） algorithm based on online least squares support vector machines （LS-SVM） which can deal with nonlinear systems effectively. At each sampling period the algorithm recursively modifies the model by adding a new data pair and deleting the least important one out of the consideration on realtime property. The data pair deleted is determined by the absolute value of lagrange multiplier from last sampling period. The paper gives the recursive algorithm of model parameters when adding a new data pair and deleting an existent one, respectively, and thus the inversion of a large matrix is avoided and the memory can be controlled by the algorithm entirely. The nonlinear LS-SVM model is applied in GPC algorithm at each sampling period. The experiments of generalized predictive control on pH neutralizing process show the effectiveness and practicality of the proposed algorithm.     

A new sliding function for discrete predictive sliding mode control of time delay systems

The control of time delay systems is still an open area for research. This paper proposes an enhanced model predictive discrete-time sliding mode control with a new sliding function for a linear system with state delay. Firstly, a new sliding function including a present value and a past value of the state, called dynamic surface, is designed by means of linear matrix inequalities （LMIs）. Then, using this dynamic function and the rolling optimization method in the predictive control strategy, a discrete predictive sliding mode controller is synthesized. This new strategy is proposed to eliminate the undesirable effect of the delay term in the closed loop system. Also, the designed control strategy is more robust, and has a chattering reduction property and a faster convergence of the system s state. Finally, a numerical example is given to illustrate the effectiveness of the proposed control.     

Temperature prediction control based on least squares support vector machines

A prediction control algorithm is presented based on least squares support vector machines (LS-SVM) model for a class of complex systems with strong nonlinearity. The nonlinear off-line model of the controUed plant is built by LS-SVM with radial basis function (RBF) kernel. In the process of system running, the off-line model is linearized at each sampling instant, and the generalized prediction control (GPC) algorithm is employed to implement the prediction control for the controlled plant. The obtained algorithm is applied to a boiler temperature control system with complicated nonlinearity and large time delay. The results of the experiment verify the effectiveness and merit of the algorithm.     

An improved robust model predictive control approach to systems with linear fractional transformation perturbations

In this paper, a robust model predictive control approach is proposed for a class of uncertain systems with time-varying, linear fractional transformation perturbations. By adopting a sequence of feedback control laws instead of a single one, the control performance can be improved and the region of attraction can be enlarged compared with the existing model predictive control (MPC) approaches. Moreover, a synthesis approach of MPC is developed to achieve high performance with lower on-line computational burden. The effectiveness of the proposed approach is verified by simulation examples.     

High-order Volterra Model Predictive Control and its application to a nonlinear polymerisation process

Model Predictive Control (MPC) has recently found wide acceptance in the process industry, but existing design and implementation methods are restricted to linear process models. A chemical process, however, involves severe nonlinearity which cannot be ignored in practice. This paper aims to solve this nonlinear control problem by extending MPC to accommodate nonlinear models. It develops an analytical framework for nonlinear model predictive control (NMPC). It also offers a third-order Volterra series based nonparametric nonlinear modelling technique for NMPC design, which relieves practising engineers from the need for deriving a physical-principles based model first. An on-line realisation technique for implementing NMPC is then developed and applied to a Mitsubishi Chemicals polymerisation reaction process. Results show that this nonlinear MPC technique is feasible and very effective. It considerably outperforms linear and low-order Volterra model based methods. The advantages of the developed approach lie not only in control performance superior to existing NMPC methods, but also in eliminating the need for converting an analytical model and then convert it to a Volterra model obtainable only up to the second order.     

Variable structure control with sliding mode prediction for discrete-time nonlinear systems

A new variable structure control algorithm based on sliding mode prediction for a class of discrete-time nonlinear systems is presented. By employing a special model to predict future sliding mode value, and combining feedback correction and receding horizon optimization methods which are extensively applied on predictive control strategy, a discrete-time variable structure control law is constructed. The closed-loop systems are proved to have robustness to uncertainties with unspecified boundaries. Numerical simulation and pendulum experiment results illustrate that the closed-loop systems possess desired performance, such as strong robustness, fast convergence and chattering elimination.     

基于支持向量机的非线性预测控制技术

探讨了利用支持向量机进行非线性系统辨识的方法,并将支持向量机模型应用到非线性预测控制,提出了基于支持向量机模型的非线性预测控制算法.对一个CSTR反应器的仿真表明,支持向量机在小样本情况下具有良好的非线性建模能力和泛化能力.基于支持向量机的预测控制具有很好的控制性能,为通用非线性控制提供了一种新的控制思路.􀁽     

基于多项式核RVM的非线性模型预测控制

针对非线性、不确定性对象不易建模的特点,提出了基于多项式核关联向量机(RVM)的解析型非线性预测控制方法,该方法采用多项式核RVM进行模型辨识,得到的对象模型作为预测模型。由于RVM具有较好的非线性建模能力,弥补了SVM参数设定难和稀疏性不强等弱点;同时,多项式形式的模型表达式使二次型优化目标函数可以通过函数解析方法求得最优控制输入,即简化了滚动优化模块,增强了控制的实时性。通过对一个标准的非线性Benchmark问题进行仿真实验,结果表明该方法具有良好的控制性能。     

混合核函数稀疏LS-SVM软测量建模与应用

针对最小二乘支持向量机存在的稀疏性欠缺和单核函数局限性问题,本文提出一种基于混合核函数稀疏最小二乘支持向量机的软测量建模方法.该方法使用多项式核函数和RBF核函数线性加权构成混合核函数,兼顾最小二乘支持向量机的全局拟合能力与局部拟合能力,以矢量基学习作为稀疏解算法,改善最小二乘支持向量机的稀疏性,在精简模型结构的同时,避免冗余信息中的噪声过多的拟合到模型参数中,进而采用粒子群算法优化模型部分参数.将此方法分别应用于Mackey- Glasss混沌模型的时间序列预测和乙烯精馏塔塔釜乙烯浓度预测,应用结果表明该方法较最小二乘支持向量机、稀疏最小二乘支持向量机以及混合核最小二乘支持向量机具有更好的泛化效果和预报精度,兆示出其良好的应用潜力.     

基于并行支持向量机的多变量非线性模型预测控制

提出一种基于并行支持向量机的多变量系统非线性模型预测控制算法．首先，通过考虑输入、输出间的耦合，建立基于并行支持向量机的多步预测模型；然后，将该模型用于非线性预测控制,提出新的适用于并行预测模型的反馈校正策略，得到最优控制律．连续搅拌槽式反应器（CSTR）的控制仿真结果表明,该算法的性能优于基于并行神经网络的非线性模型预测控制和基于集成模型的非线性模型预测控制．     

基于SVM和PSO的新型非线性模型预测控制

对于具有强非线性或复杂非线性约束的系统,通过非线性模型的线性化和二次规划优化实现非线性模型预测控制,难以取得满意的结果。提出了一种基于支持向量机模型和粒子群优化的非线性模型预测控制系统的算法。仿真实例表明了支持向量机模型的泛化能力和粒子群优化的寻优速度及能力,证明了将其运用于非线性模型预测控制中的可行性。     

Sparse kernel regression technique for self-cleansing channel design

The application of a robust learning technique is inevitable in the development of a self-cleansing sediment transport model. This study addresses this problem and advocates the use of sparse kernel regression (SKR) technique to design a self-cleaning model. The SKR approach is a regression technique operating in the kernel space which also benefits from the desirable properties of a sparse solution. In order to develop a model applicable to a wide range of channel characteristics, five different experimental data sets from 14 different channels are utilized in this study. In this context, the efficacy of the SKR model is compared against the support vector regression (SVR) approach along with several other methods from the literature. According to the statistical analysis results, the SKR method is found to outperform the SVR and other regression equations. In particular, while empirical regression models fail to generate accurate results for other channel cross-section shapes and sizes, the SKR model provides promising results due to the inclusion of a channel parameter at the core of its structure and also by operating on an extensive range of experimental data. The superior efficacy of the SKR approach is also linked to its formulation in the kernel space while also benefiting from a sparse representation method to select the most useful training samples for model construction. As such, it also circumvents the requirement to evaluate irrelevant or noisy observations during the test phase of the model, and thus improving on the test phase running time.     

Least squares support vector machines with tuning based on chaotic differential evolution approach applied to the identification of a thermal process

In the past decade, support vector machines (SVMs) have gained the attention of many researchers. SVMs are non-parametric supervised learning schemes that rely on statistical learning theory which enables learning machines to generalize well to unseen data. SVMs refer to kernel-based methods that have been introduced as a robust approach to classification and regression problems, lately has handled nonlinear identification problems, the so called support vector regression. In SVMs designs for nonlinear identification, a nonlinear model is represented by an expansion in terms of nonlinear mappings of the model input. The nonlinear mappings define a feature space, which may have infinite dimension. In this context, a relevant identification approach is the least squares support vector machines (LS-SVMs). Compared to the other identification method, LS-SVMs possess prominent advantages: its generalization performance (i.e. error rates on test sets) either matches or is significantly better than that of the competing methods, and more importantly, the performance does not depend on the dimensionality of the input data. Consider a constrained optimization problem of quadratic programing with a regularized cost function, the training process of LS-SVM involves the selection of kernel parameters and the regularization parameter of the objective function. A good choice of these parameters is crucial for the performance of the estimator. In this paper, the LS-SVMs design proposed is the combination of LS-SVM and a new chaotic differential evolution optimization approach based on Ikeda map (CDEK). The CDEK is adopted in tuning of regularization parameter and the radial basis function bandwith. Simulations using LS-SVMs on NARX (Nonlinear AutoRegressive with eXogenous inputs) for the identification of a thermal process show the effectiveness and practicality of the proposed CDEK algorithm when compared with the classical DE approach.     

一种基于Wiener模型的非线性预测控制算法

针对一类Wiener模型描述的非线性系统,提出了一种改进的非线性预测控制算法.该算法利用Laguerre函数描述Wiener模型动态线性部分的控制信号，将预测控制中在预测时域内优化求解未来控制输入序列转化为优化求解一组无记忆的Laguerre系数,以减少优化所需的计算量.利用静态模糊模型来逼近Wiener模型的非线性部分,将非线性预测控制优化问题转化为线性预测控制优化问题,克服了求控制输入时解非线性方程的困难,进而推导出了预测控制输入的解析式.CSTR过程的仿真结果表明了本文算法的有效性和可行性.     

Non-Mercer hybrid kernel for linear programming support vector regression in nonlinear systems identification

As a new sparse kernel modeling method, support vector regression (SVR) has been regarded as the state-of-the-art technique for regression and approximation. In [V.N. Vapnik, The Nature of Statistical Learning Theory, second ed., Springer-Verlag, 2000], Vapnik developed the ?-insensitive loss function for the support vector regression as a trade-off between the robust loss function of Huber and one that enables sparsity within the support vectors. The use of support vector kernel expansion provides us a potential avenue to represent nonlinear dynamical systems and underpin advanced analysis. However, in the standard quadratic programming support vector regression (QP-SVR), its implementation is often computationally expensive and sufficient model sparsity cannot be guaranteed. In an attempt to mitigate these drawbacks, this article focuses on the application of the soft-constrained linear programming support vector regression (LP-SVR) with hybrid kernel in nonlinear black-box systems identification. An innovative non-Mercer hybrid kernel is explored by leveraging the flexibility of LP-SVR in choosing the kernel functions. The simulation results demonstrate the ability to use more general kernel function and the inherent performance advantage of LP-SVR to QP-SVR in terms of model sparsity and computational efficiency.     

稀疏贝叶斯模型与相关向量机学习研究

虽然支持向量机在模式识别的相关领域得到了广泛应用,但它自身固有许多不足之处.相关向量机是在稀疏贝叶斯框架下提出的稀疏模型,模型没有规则化系数,核函数不要求满足Mercer条件.相关向量机不仅具备良好的泛化能力,而且还能够得到具有统计意义的预测结果.首先介绍了稀疏贝叶斯回归和分类模型,通过参数推断过程,将相关向量机学习转化为最大化边缘似然函数估计,并分析了3种估计方法,给出了快速序列稀疏贝叶斯学习算法流程.