Multivariable control experiments of non-linear chemical processes using non-linear feedback transformation

The decoupling and linearization control theory based on non-linear feedback transformation for non-linear systems was applied to two non-linear and interactive chemical control processes. The first is a level and temperature control process and the second a level and pH control process. To compensate for the mismatch between the real process and the process model, external PI controllers, which track the outputs from the reference model, were added. The experimental results showed that even if the process is both non-linear and interactive and the process model is not exactly known, a satisfactory control performance can be obtained by applying this theory.     

Neural network model for the efficient calculation of Green's functions in layered media

In this article, neural networks are employed for fast and efficient calculation of Green's functions in a layered medium. Radial basis function networks (RBFNs) are effectively trained to estimate the coefficients and the exponents that represent a Green's function in the discrete complex image method (DCIM). Results show very good agreement with the DCIM, and the trained RBFNs are very fast compared with the corresponding DCIM. © 2003 Wiley Periodicals, Inc. Int J RF and Microwave CAE 13: 128–135, 2003.     

Regional stabilisation of polynomial non-linear systems using rational Lyapunov functions

In this article, we propose a new non-linear stabilisation approach based on the popular linear parameter-varying control techniques. The regional state-feedback control problem of polynomial non-linear systems will be studied using rational Lyapunov functions of states. By bounding the variation rates of each state, the domain of attraction will be embedded in the region specified by the non-linear vector field. As a result, the state-feedback stabilisation conditions will be formulated as a set of polynomial matrix inequalities and can be solved efficiently by sum-of-squares programming. The resulting Lyapunov matrix and state-feedback gains are typically state-dependent rational matrix functions. This approach is also extended to a class of output-dependent non-linear systems where the stabilising output-feedback controller can be synthesised using rational Lyapunov functions of outputs. Finally, several examples will be used to demonstrate the proposed stabilisation approach and clarify the effect of various choices of Lyapunov function forms and state constraints.     

紧支径向基函数插值实现多维数据可视化

通过分析某城市空气质量数值预报数据的时空组织结构,构建出了多维空间数据的整体框架。论述了几种插值方法的优缺点,在比较的基础上,将新的紧支径向基函数局部径向点插值方法引入到多维数据处理中,在空间、时间维度上对数据进行局部插值,从而实现数据的重构。以新的基于封装回调函数的多线程方法实现了大规模空气质量预报数据的三维动态可视化。实验结果表明,以上方法应用于大规模数据可视化时,其质量和运算速度都能满足实际需要。     

Compactly supported radial basis functions for adaptive process control

An adaptive nonlinear control strategy based on networks of compactly supported radial basis functions is proposed. The local influence of the basis functions allows efficient on-line adaptation that is performed using a gradient law, and new basis functions are added to the network only when new regions in state space are encountered and the prediction error exceeds a pre-specified tolerance. The approximate model is used to construct an input-output linearizing control law. The adaptive control strategy is applied to a nonlinear chemical reactor model.     

Combined projection and kernel basis functions for classification in evolutionary neural networks

This paper proposes a hybrid neural network model using a possible combination of different transfer projection functions (sigmoidal unit, SU, product unit, PU) and kernel functions (radial basis function, RBF) in the hidden layer of a feed-forward neural network. An evolutionary algorithm is adapted to this model and applied for learning the architecture, weights and node typology. Three different combined basis function models are proposed with all the different pairs that can be obtained with SU, PU and RBF nodes: product–sigmoidal unit (PSU) neural networks, product–radial basis function (PRBF) neural networks, and sigmoidal–radial basis function (SRBF) neural networks; and these are compared to the corresponding pure models: product unit neural network (PUNN), multilayer perceptron (MLP) and the RBF neural network. The proposals are tested using ten benchmark classification problems from well known machine learning problems. Combined functions using projection and kernel functions are found to be better than pure basis functions for the task of classification in several datasets.     

Modeling of microstrip lines using neural networks: Applications to the design and analysis of distributed microstrip circuits

In this article, two models for solving microstrip lines are presented. The models utilize radial-basis-function neural networks. Using the first model, one estimates the effective dielectric constant and the width of the microstrip line, knowing its characteristic impedance and the frequency. The second model provides the effective dielectric constant and the characteristic impedance of the line based on knowledge of its width and the frequency. Besides their remarkably fast responses, the proposed models are capable of estimating the required quantities with very high accuracy. The potential of the proposed models is demonstrated in the design and analysis of two distributed microstrip circuits. © 2004 Wiley Periodicals, Inc. Int J RF and Microwave CAE 14, 166–173, 2004.     

An improved adaptive online neural control for robot manipulator systems using integral Barrier Lyapunov functions

Conventional Neural Network (NN) control for robots uses radial basis function (RBF) and for n-link robot with online control, the number of nodes and weighting matrix increases exponentially, which requires a number of calculations to be performed within a very short duration of time. This consumes a large amount of computational memory and may subsequently result in system failure. To avoid this problem, this paper proposes an innovative NN robot control using a dimension compressed RBF (DCRBF) for a class of n-degree of freedom (DOF) robot with full-state constraints. The proposed DCRBF NN control scheme can compress the nodes and weighting matrix greatly and provide an output that meets the prescribed tracking performance. Additionally, adaption laws are designed to compensate for the internal and external uncertainties. Finally, the effectiveness of the proposed method has been verified by simulations. The results indicate that the proposed method, integral Barrier Lyapunov Functions (iBLF), avoids the existing defects of Barrier Lyapunov Functions (BLF) and prevents the constraint violations.     

Radial basis functions method for numerical solution of the modified equal width equation

The numerical solution of the modified equal width equation is investigated by using meshless method based on collocation with the well-known radial basis functions. Single solitary wave motion, two solitary waves interaction and three solitary waves interaction are studied. Results of the meshless methods with different radial basis functions are presented.     

自适应RBF-LBF串联神经网络结构与参数优化方法

研究了前向单层径基函数(RBF)网络和前向单层线性基本函数(LBF)网络的分类机理，提出了RBF的中心和宽度应通过学习自动确定，在学习过程中根据错分样本被错分入的类别自动生成新的核函数这一观点．如果两个或两个以上核函数属于同一类，在输入空间相距较近且未被其它类别的样本分隔开来的情况下，则应考虑将之合并，或者使它们的作用区域部分重叠．从理论上阐明了采用Sigmoid活化函数的单层感知器的分类阈值为0．5，进而提出了由单层RBF网络和单层感知器组成的串联RBF—LBF神经网络．文中详细给出了确定该串联RBF—LBF神经网络结构、核函数个数、位置与宽度的优化算法．一般来说，该算法的计算复杂性比前向单隐层感知器采用的误差反传算法要小或至少相当．对几个经典的模式分类难题的处理结果表明，与一般RBF网络和前向单隐层感知器网络相比，该串联RBF—LBF网络及其自适应学习算法具有收敛速度快，分类精度高，易于得到最小结构，在学习过程中不易陷入局部极小点等优点，有利于实现实时分析．实验结果同时也验证了单层LBF网络对提高RBF—LBF网络分类正确率的重要性．     

Approximate implicitization of parametric surfaces by using compactly supported radial basis functions

In this paper, a new approach is proposed to solve the approximate implicitization of parametric surfaces. It is primarily based on multivariate interpolation of scattered data by using compactly supported radial basis functions. Experimental results are provided to illustrate the proposed method is flexible and effective.     

Fast tunable gradient RBF networks for online modeling of nonlinear and nonstationary dynamic processes

Since most real-world processes exhibit both nonlinear and time-varying characteristics, there exists a need for accurate and efficient models that can adapt in nonstationary environments. Also for adaptive control purpose, it is vital that an adaptive model has a fixed small model size. In this paper, we propose an adaptive tunable gradient radial basis function (GRBF) network for online modeling of nonlinear dynamic processes, which meets these practical requirements. Specifically, a compact GRBF model is constructed by the orthogonal least squares algorithm in training, which is capable of modeling variations of local mean and trend in the data well. During online operation, the adaptive GRBF model tacks the time-varying process’s dynamics by replacing a worst performing node with a new node which encodes the current new data. By exploiting the local predictor property of the GRBF node, the new node optimization can be done extremely efficiently. The proposed approach combining the advantages of both the GRBF network structure and fast tunable node mechanism is capable of tracking the time-varying nonlinear dynamics accurately and effectively. Extensive simulation results demonstrate that the proposed fast tunable GRBF network significantly outperforms the existing state-of-the-art methods, in terms of both adaptive modeling accuracy and online computational complexity.     

Modelling and prediction of complex non-linear processes by using Pareto multi-objective genetic programming

In this paper, a new multi-objective genetic programming (GP) with a diversity preserving mechanism and a real number alteration operator is presented and successfully used for Pareto optimal modelling of some complex non-linear systems using some input–output data. In this study, two different input–output data-sets of a non-linear mathematical model and of an explosive cutting process are considered separately in three-objective optimisation processes. The pertinent conflicting objective functions that have been considered for such Pareto optimisations are namely, training error (TE), prediction error (PE), and the length of tree (complexity of the network) (TL) of the GP models. Such three-objective optimisation implementations leads to some non-dominated choices of GP-type models for both cases representing the trade-offs among those objective functions. Therefore, optimal Pareto fronts of such GP models exhibit the trade-off among the corresponding conflicting objectives and, thus, provide different non-dominated optimal choices of GP-type models. Moreover, the results show that no significant optimality in TE and PE may occur when the TL of the corresponding GP model exceeds some values.     

Representation of Finite State Automata in Recurrent Radial Basis Function Networks

In this paper, we propose some techniques for injecting finite state automata into Recurrent Radial Basis Function networks (R2BF). When providing proper hints and constraining the weight space properly, we show that these networks behave as automata. A technique is suggested for forcing the learning process to develop automata representations that is based on adding a proper penalty function to the ordinary cost. Successful experimental results are shown for inductive inference of regular grammars.     

Topology preservation evaluation of compact-support radial basis functions for image registration

Many Radial Basis Function (RBF)-based transformations are used to model the deformations in image registration, and they have different topology preservation properties. This paper compares analytically and experimentally the topology preservation performance of compact-support thin-plate Spline (CSTPS), locally constrained cosine (Cos), Wendland, Gaussian, Buhmann and Wu functions in landmark-based image registration. In addition, the topology preservation characteristics of thin-plate Spline (TPS) and elastic body Spline (EBS)-based transformations are compared for global-support deformations. The comparative results show that, for local deformation CSTPS and Buhmann preserve topology better than others. The Cos and Gaussian functions could easily produce topology violations for relatively dense-landmark matching. For global-support transformations, CSTPS, Wendland ψ_3,1, Buhmann, Wu and TPS outperformed others because they preserve topology better. The Cos, Gaussian and EBS functions perform poorly because folds and tears of the deformation surface occur easily. With very large support, CSTPS produces similar results as TPS, and Wendland ψ_3,1 has similar performance with Wu functions. Also, Cos and Gaussian performed similarly in this case. In the experiments, these theoretical results are evaluated extensively using transformations on random point sets, artificial images, and medical images.     

Radial basis functions method for parabolic inverse problem

The radial basis functions (RBFs) method is employed to handle a class of multi-dimensional parabolic inverse problems. Because they are not modelled by classical parabolic initial-boundary value problems, theoretical behaviour and numerical approximation of these problems have been active areas of research. Based on the idea of RBF approximation, a fast and highly accurate meshless method is developed for solving an inverse problem with a control parameter. Moreover, with the meshless property, it can be used to handle multi-dimensional parabolic inverse problems defined on very complicated geometries.     

Robust adaptive control of uncertain non-linear systems using neural networks

This paper presents a robust adaptive output feedback control design method for uncertain non-affine non-linear systems, which does not rely on state estimation. The approach is applicable to systems with unknown but bounded dimensions and with known relative degree. A neural network is employed to approximate the unknown modelling error. In fact, a neural network is considered to approximate and adaptively make ineffective unknown plant non-linearities. An adaptive law for the weights in the hidden layer and the output layer of the neural network are also established so that the entire closed-loop system is stable in the sense of Lyapunov. Moreover, the robustness of the system against the approximation error of neural network is achieved with the aid of an additional adaptive robustifying control term. In addition, the tracking error is guaranteed to be uniformly and asymptotically stable, rather than uniformly ultimately bounded, by using this additional control term. The proposed control algorithm is relatively straightforward and no restrictive conditions on the design parameters for achieving the systems stability are required. The effectiveness of the proposed scheme is shown through simulations of a non-affine non-linear system with unmodelled dynamics, and is compared with a second-sliding mode controller.     

基于RBF神经网络的非线性时间序列在线预测

针对非线性非高斯时间序列, 提出观测噪声服从隐马尔可夫模型(HMM)的径向基函数(RBF)神经网络(RBF-HMM)预测模型, 其特点在于模型输入包含误差反馈项、RBF网络隐含层节点数的可变性和观测噪声的隐马尔可夫性; 并采用序列蒙特卡罗(SMC)方法实现基于RBF-HMM模型的时间序列在线预测. 最后采用太阳黑子数平滑月均值数据和CRU国际钢材价格指数月数据进行实证研究, 结果表明该模型的有效性.     

径向基函数网络的功能分析与应用的研究

径向基函数网络与BP网络在网络结构上都属于前向网络,但它们对网络权值训练所采用的算法是完全不同的。另外,径向基函数网络的网络结构与模糊系统有很紧密的关联。该文从径向基函数网络的结构入手,分别对其所具有的特点、权值训练、网络设计方法及其应用等方面,通过分析与实例,采用对比的方式,给予实验的验证。