典型人工神经网络的结构、功能及其在智能系统中的应用

人工神经网络已在各个领域得到广泛的应用, 尤其是在智能系统中的非线性建模及其控制器的设计、模式分类与模式识别、联想记忆和优 化计算等方面更是得到人们的极大关注．本文从网络在智能系统中建模及控制器设计的具体 训练结构入手,详细介绍了BP网络在系统控制中的典型应用方式,并根据不同网络所具有的 功能,从性能对比的角度对人工神经网络在上述各方面的应用给予综述．     

FPGA‐based adaptive dynamic sliding‐mode neural control for a brushless DC motor

In the adaptive neural control design, since the number of hidden neurons is finite for real‐time applications, the approximation errors introduced by the neural network cannot be inevitable. To ensure the stability of the adaptive neural control system, a switching compensator is designed to dispel the approximation error. However, it will lead to substantial chattering in the control effort. In this paper, an adaptive dynamic sliding‐mode neural control (ADSNC) system composed of a neural controller and a fuzzy compensator is proposed to tackle this problem. The neural controller, using a radial basis function neural network, is the main controller and the fuzzy compensator is designed to eliminate the approximation error introduced by the neural controller. Moreover, a proportional‐integral‐type adaptation learning algorithm is developed based on the Lyapunov function; thus not only the system stability can be guaranteed but also the convergence of the tracking error and controller parameters can speed up. Finally, the proposed ADSNC system is implemented based on a field programmable gate array chip for low‐cost and high‐performance industrial applications and is applied to control a brushless DC (BLDC) motor to show its effectiveness. The experimental results demonstrate the proposed ADSNC scheme can achieve favorable control performance without encountering chattering phenomena. Copyright © 2010 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society     

Image processing with neural networks—a review

We review more than 200 applications of neural networks in image processing and discuss the present and possible future role of neural networks, especially feed-forward neural networks, Kohonen feature maps and Hopfield neural networks. The various applications are categorised into a novel two-dimensional taxonomy for image processing algorithms. One dimension specifies the type of task performed by the algorithm: preprocessing, data reduction/feature extraction, segmentation, object recognition, image understanding and optimisation. The other dimension captures the abstraction level of the input data processed by the algorithm: pixel-level, local feature-level, structure-level, object-level, object-set-level and scene characterisation. Each of the six types of tasks poses specific constraints to a neural-based approach. These specific conditions are discussed in detail. A synthesis is made of unresolved problems related to the application of pattern recognition techniques in image processing and specifically to the application of neural networks. Finally, we present an outlook into the future application of neural networks and relate them to novel developments.     

Application of integrated recurrent neural network with multivariate adaptive regression splines on SPC-EPC process

The integration of statistical process control and engineering process control has been reported as an effective way to monitor and control the autocorrelated process. However, because engineering process control compensates for the effects of underlying disturbances, the disturbance patterns become very hard to recognize, especially when various abnormal control chart patterns are mixed and co-existed in the engineering process. In this study, a new control chart pattern recognition model which integrates multivariate adaptive regression splines and recurrent neural network is proposed to not only address the problem of feature selection (i.e., lagged process measurements) but also improve the pattern recognition accuracy. The performance of the proposed method is evaluated by comparing the recognition results of multivariate adaptive regression splines and recurrent neural network with the results of four competing approaches (multivariate adaptive regression splines-extreme learning machine, multivariate adaptive regression splines-random forest, single recurrent neural network, and single random forest) on the simulated individual process data. The experimental study shows that the proposed multivariate adaptive regression splines and recurrent neural network approach can not only solve the problem of variable selection but also outperform other competing models. Moreover, according to the lagged process measurements selected by the proposed approach, lagged observations that exerted significant impact on the construction of the control chart pattern recognition model can be identified successfully. This study has significant implications for research and practice in production management and provides a valuable reference for manufacturing process managers to better understand and develop strategies for control chart pattern recognition.     

An algorithm for segmenting cytological images

An adaptive algorithm for automatic segmentation of objects in cytological images is described. The algorithm is based on the well-known seeded region growing method (SGR), is robust to noise, and can handle low contrast images. The algorithm allows for automatic adjustment of its parameter values and initialization of cluster growth. Oleg L. Konevsky. Born 1973. Received master’s degree in engineering from Novgorod State University in 1995 and candidate’s degree (Eng.) from St. Petersburg State Technical University in 1998. Since 2000, an associate professor at the Information Technologies and Systems Department, Novgorod State University. Scientific interests: image segmentation, mathematical morphology, and neural networks. Author of nearly 30 papers in the field of pattern recognition and image analysis. Member of IEEE, IEEE Computer Society, and IEEE Signal Processing Society. Yurii V. Stepanets. Born 1980. Received master’s degree in engineering from Novgorod State University in 2002. Currently, post-graduate student at the same university. Scientific interests: image segmentation, artificial intelligence, and neural networks. Author of ten papers in the field of pattern recognition and image analysis.     

Adaptive neural complementary sliding-mode control via functional-linked wavelet neural network

Chaos control can be applied in the vast areas of physics and engineering systems, but the parameters of chaotic system are inevitably perturbed by external inartificial factors and cannot be exactly known. This paper proposes an adaptive neural complementary sliding-mode control (ANCSC) system, which is composed of a neural controller and a robust compensator, for a chaotic system. The neural controller uses a functional-linked wavelet neural network (FWNN) to approximate an ideal complementary sliding-mode controller. Since the output weights of FWNN are equipped with a functional-linked type form, the FWNN offers good learning accuracy. The robust compensator is designed to eliminate the effect of the approximation error introduced by the neural controller upon the system stability in the Lyapunov sense. Without requiring preliminary offline learning, the parameter learning algorithm can online tune the controller parameters of the proposed ANCSC system to ensure system stable. Finally, it shows by the simulation results that favorable control performance can be achieved for a chaotic system by the proposed ANCSC scheme.     

Decoupled control using neural network-based sliding-mode controller for nonlinear systems

In this paper, an adaptive neural network sliding-mode controller design approach with decoupled method is proposed. The decoupled method provides a simple way to achieve asymptotic stability for a class of fourth-order nonlinear system. The adaptive neural sliding-mode control system is comprised of neural network (NN) and a compensation controller. The NN is the main regulator controller, which is used to approximate an ideal computational controller. The compensation controller is designed to compensate for the difference between the ideal computational controller and the neural controller. An adaptive methodology is derived to update weight parts of the NN. Using this approach, the response of system will converge faster than that of previous reports. The simulation results for the cart–pole systems and the ball–beam system are presented to demonstrate the effectiveness and robustness of the method. In addition, the experimental results for seesaw system are given to assure the robustness and stability of system.     

A scalable hybrid decision system (HDS) for Roman word recognition using ANN SVM: study case on Malay word recognition

An off-line handwriting recognition (OFHR) system is a computerized system that is capable of intelligently converting human handwritten data extracted from scanned paper documents into an equivalent text format. This paper studies a proposed OFHR for Malaysian bank cheques written in the Malay language. The proposed system comprised of three components, namely a character recognition system (CRS), a hybrid decision system and lexical word classification system. Two types of feature extraction techniques have been used in the system, namely statistical and geometrical. Experiments show that the statistical feature is reliable, accessible and offers results that are more accurate. The CRS in this system was implemented using two individual classifiers, namely an adaptive multilayer feed-forward back-propagation neural network and support vector machine. The results of this study are very promising and could generalize to the entire Malay lexical dictionary in future work toward scaled-up applications.     

Self-organizing neural control system design for dynamic processes

This paper discusses the achievable nominal performance of a well-parametrized neural feedback control system, and proposes an efficient training method for parametrizing such a controller. A self-organizing neural control (SONC) system is presented in which a layered feedforward neural network is adopted as the controller structure in order to apply directly existing back-propagated learning techniques. A self-organizing methodology is introduced to provide the training set for adjusting parameters of the neural controller. One important feature of the proposed adaptive mechanism is that, though it should lack extensive knowledge of the process dynamics at the outset of controller design, it will still be able to achieve its desired results by employing the subjective experience of control specialists as its training aids. Tuning variables of the SONC system are reviewed through exploring their effects on five typical transfer functions. The applicability of the SONC system is also demonstrated on a continuous stirred tank reactor. Simulation results show that a well-parametrized neural controller can improve nominal performance for a wide variety of different processes, and the proposed self-organizing mechanism can direct a controller to achieve the desired final parametrization.     

具有磁滞输入的可调金属切削系统鲁棒自适应动态面控制

针对既有时滞环节又存在磁滞输入的可调金属切削系统,提出了一种改进的自适应动态面控制方法,其特点为:1)设计了带有跟踪误差性能指标函数的鲁棒自适应动态面控制算法,并结合神经网络,使其能够保证系统的跟踪误差及其过渡过程在预先任意给定的范围内;2)克服了反推控制方案中的"微分爆炸"问题,简化控制器结构;3)估计神经网络权值向量的范数而不是估计权值向量,极大地减少系统的计算负担,便于实时控制.仿真结果验证了该控制方法的有效性.     

Adaptive PI Hermite neural control for MIMO uncertain nonlinear systems

This paper presents an adaptive PI Hermite neural control (APIHNC) system for multi-input multi-output (MIMO) uncertain nonlinear systems. The proposed APIHNC system is composed of a neural controller and a robust compensator. The neural controller uses a three-layer Hermite neural network (HNN) to online mimic an ideal controller and the robust compensator is designed to eliminate the effect of the approximation error introduced by the neural controller upon the system stability in the Lyapunov sense. Moreover, a proportional–integral learning algorithm is derived to speed up the convergence of the tracking error. Finally, the proposed APIHNC system is applied to an inverted double pendulums and a two-link robotic manipulator. Simulation results verify that the proposed APIHNC system can achieve high-precision tracking performance. It should be emphasized that the proposed APIHNC system is clearly and easily used for real-time applications.     

Adaptive neural control for uncertain systems subject to actuator failure

In this paper, the design of controller based on neural network is investigated for a class of uncertain systems subject to actuator failures. An adaptive neural controller is designed by utilizing the approximation technique of neural network. The key feature in this work is to remove the requirement on the boundedness of unknown nonlinear functions that is usually encountered in the existing works. Moreover, sufficient conditions are derived such that the closed-loop system is robustly stable. Finally, numerical simulation results are given.     

一种改进的GRU-InFCN人体行为识别模型

为了克服传统机器学习方法在采用传感器数据进行人体行为识别领域上识别效果对人工特征选取依赖严重、识别准确率不高等问题,提出一种改进的全卷积神经网络和多层循环神经网络并联的深度学习模型(GRU-InFCN),并对传感器数据特征进行自动提取,实现人体动作的识别。该模型通过多尺度卷积神经网络和双层GRU网络(Gated Recurrent Unit,GRU)分别对传感器数据进行特征提取,将特征矩阵在矩阵维度上进行特征拼接再通过Softmax完成特征分类。实验结果表明,在开源人体行为识别(HAR)数据集上采用该方法进行人体行为识别,准确率达到了97.76%。该模型在取得高准确率的同时,避免了复杂的信号预处理和特征工程。     

Development and application of an integrated neural system for anHDCL

This study presents the development and industrial application of an integrated neural system in coating weight control for a modern hot dip coating line (HDCL) in a steel mill. The neural system consists of two multilayered feedforward neural networks and a neural adaptive controller. They perform coating weight real-time prediction, feedforward control (FFC), and adaptive feedback control (FBC), respectively. The production line analysis, neural system architecture, learning, associative memories, generalization and real-time applications are addressed in this paper. This integrated neural system has been successfully implemented and applied to an HDCL at Burns Harbor Division, Bethlehem Steel Co., Chesterton, IN. The industrial application results have shown significant improvements in reduction of coating weight transitional footage, variation of the error between the target and actual coating weight, and the coating material used. Some practical aspects for applying a neural system to industrial control are discussed as concluding remarks.     

An Adaptive RBF Neural Network Control Method for a Class of Nonlinear Systems

This paper focuses on designing an adaptive radial basis function neural network (RBFNN) control method for a class of nonlinear systems with unknown parameters and bounded disturbances. The problems raised by the unknown functions and external disturbances in the nonlinear system are overcome by RBFNN, combined with the single parameter direct adaptive control method. The novel adaptive control method is designed to reduce the amount of computations effectively. The uniform ultimate boundedness of the closed-loop system is guaranteed by the proposed controller. A coupled motor drives (CMD) system, which satisfies the structure of nonlinear system, is taken for simulation to confirm the effectiveness of the method. Simulations show that the developed adaptive controller has favorable performance on tracking desired signal and verify the stability of the closed-loop system.      

Specificity enhancement in classification of breast MRI lesion based on multi-classifier

In the conventional CMAC-based adaptive controller design, a switching compensator is designed to guarantee system stability in the Lyapunov stability sense but the undesirable chattering phenomenon occurs. This paper proposes a CMAC-based smooth adaptive neural control (CSANC) system that is composed of a neural controller and a saturation compensator. The neural controller uses a CMAC neural network to online mimic an ideal controller and the saturation compensator is designed to dispel the approximation error between the ideal controller and neural controller without any chattering phenomena. The parameter adaptive algorithms of the CSANC system are derived in the sense of Lyapunov stability, so the system stability can be guaranteed. Finally, the proposed CSANC system is applied to a Chua’s chaotic circuit and a DC motor driver. Simulation and experimental results show the CSANC system can achieve a favorable tracking performance. It should be emphasized that the development of the proposed CSANC system doesn’t need the knowledge of the system dynamics.     

Neural network algorithms for analyzing multidimensional time series for predicting events and their application to study of Sun-Earth relations

In the paper, an original neural network algorithm for analysis of time series is presented. This algorithm allows predicting the occurrence of a certain event and finding a time interval to which a phenomenon (a precursor or a cause of the event) belongs. The characteristics of the algorithm functioning are investigated applied to the study of the solar-terrestrial relationship. Yu. V. Orlov. Candidate in Physics and Mathematics. Researcher at the Institute of Nuclear Physics, Moscow State University. Scientific interests: neural networks, genetic algorithms, algorithms of pattern recognition and image analysis. Yu. S. Shugai. Researcher at the Institute of Nuclear Physics, Moscow State University. Scientific interests: neural networks, genetic algorithms, algorithms of pattern recognition and image analysis, algorithms of classification and prediction. I. G. Persiantsev. Professor, Doctor in Mathematics and Physics. Head of the Laboratory, Leading Researcher at the Institute of Nuclear Physics, Moscow State University Scientific interests: neural networks, genetic algorithms, algorithms of pattern recognition and image analysis, algorithms of classification and prediction, inverse problems. Laureate of the USSR State Prize. S. A. Dolenko. Candidate in Physics and Mathematics. Senior Researcher at the Institute of Nuclear Physics, Moscow State University. Scientific interests: neural networks, genetic algorithms, algorithms of pattern recognition and image analysis, algorithms of classification and prediction, inverse problems.     

基于自适应神经网络的AUVs路径跟踪控制

为解决由于随时间变化水动力阻尼引起的参数变化和不确定性的问题,提出了基于径向基函数神经网络的未知评估算法,引入自适应算法以保证神经网络权值的最优评估.基于Lyapunov稳定性理论,设计一种自适应神经网络控制器以保证路径跟踪系统中所有误差状态都趋于稳定.为了验证该控制器的可行性,对系统施加如位置误差、方向误差等虚拟干扰,证明该控制器可将误差消减为零.另一方面,机器人在以恒定的速度行驶时,每个航点被指定一个适合半径的圆弧可以保证其有较高的精度.为了评估路径跟踪控制器的性能,提出直线型和直线加圆弧型路径方案.仿真结果表明,该控制器可以有效地消除机器人非线性和模型不确定性造成的干扰.     

基于神经网络Smith预估器的预测控制

针对非线性系统时滞问题,给出了一种新型的单神经元Smith预测控制算法.神经网络的预测控制器由不完全微分的单神经元自适应PID控制器和神经网络的Smith预估器组成.预估器对输出进行多步预测,控制器超前动作以消除时滞对系统的影响.不完全微分的单神经元自适应PID控制器通过改进的Hebb学习规则实现其权值调节,通过权系数的在线调整实现自适应控制.仿真实验证明了该方法具有较快的响应速度和较好的响应性能.     

基于神经网络的水下机器人三维航迹跟踪控制

本文研究了水下机器人三维航迹跟踪控制问题.在充分考虑了模型中不确定水动力系数和外界海流干扰的基础上,提出了基于神经网络的自适应输出反馈控制方法.控制器由3部分组成:基于动态补偿器的输出反馈控制项、神经网络自适应控制项和鲁棒控制项.神经网络所需的自适应学习信号由线性观测器提供.基于Lyapunov稳定性理论证明了控制系统的稳定性.最后针对某AUV进行了空间三维航迹跟踪控制仿真实验,结果表明设计的控制器可以较好地克服时变非线性水动力阻尼对系统的影响,并对外界海流干扰有较好的抑制作用,可以实现三维航迹的精确跟踪.