On parity problems and the functional-link artificial neural network

This paper contains the proof of a theorem on the capability of functional-link artificial neural networks both to represent and to learn the n-dimensional parity problem. The result is obtained by an embedding of the problem into a space of dimension 2ⁿ — 1. It is shown that the Volterra expansion of the data in n-dimensions provides the necessary transformation. By computing the parity function, it is shown that a suitable set of neural network weights can be deduced. Finally, it is demonstrated that 2ⁿ — 1 is the minimum embedding dimension for the problem.The contribution of A Zuderell of the University of Innsbruck is acknowledged.     

Multiple models and neural networks based decoupling control of ball mill coal-pulverizing systems

Using a ball mill coal-pulverizing system as a motivating/application example, a class of complex industrial processes is investigated in this paper, which has strong couplings among loops, high nonlinearities and time-varying dynamics under different operation conditions. Focusing on such processes, an intelligent decoupling control method is developed, where the effects of nonlinearities are dealt with by neural network compensations and coupling effects are handled by specifically designed decoupling compensators, while the effect of time-varying dynamics is treated by a switching mechanism among multiple models. The stability and convergence of the closed-loop system are analyzed. The proposed method has been applied to the ball mill coal-pulverizing systems of 200 MW units in a heat power plant in China. Application results show that the system outputs are maintained in desired scopes, the electric energy consumption per unit coal has been reduced by 10.3%, and the production rate has been increased by 8%.     

纯反馈非线性离散系统的自适应神经网络控制

针对一类未知的纯反馈非线性离散系统,提出了基于反步法设计的自适应神经网络控制方法.为避免反步法设计中可能出现的因果矛盾问题,首先将系统进行等价变换,然后利用隐函数定理证实了理想虚拟控制输入和实际控制输入的存在性.利用高阶神经网络估计这些控制量,并基于反步法设计自适应神经网络控制系统,证明了闭环系统半全局一致最终有界.仿真结果验证了所提出方法的有效性.     

A hierarchical optimization neural network for large-scale dynamic systems

A recurrent neural network for dynamical hierarchical optimization of nonlinear discrete large-scale systems is presented. The proposed neural network consists of hierarchically structured sub-networks: one coordination sub-network at the upper level and several local optimization sub-networks at the lower level. In particular, the coordination sub-network and the local optimization sub-networks work simultaneously. This feature makes the proposed method outperform in computational efficiency the conventional iterative algorithms where there usually exists an alternately waiting time during the coordination and local optimization processes. Moreover, the state equations of the subsystems of the large-scale system are imbedded into their corresponding local optimization sub-networks. This imbedding technique not only overcomes the difficulty in treating the constraints imposed by the state equations, but also leads to significant reduction in the network size. We present stability analysis to prove that the neural network is asymptotically stable and this stable state corresponds to the optimal solution to the original optimal control problem. Finally, we illustrate the performance of the proposed method by an example.     

On realizability of neural networks-based input-output models in the classical state-space form

This paper proves that the typical neural network-based input/output model does not have a state-space realization and suggests the Additive Nonlinear Auto-Regressive with eXogenous input (ANARX) structure as an excellent choice for neural-network-based input-output models. The advantage of the ANARX model is that the time-steps in the argument are pair-wise decomposed, which allows the ANARX model to be realized in state space, and to be linearized via dynamic output feedback. Moreover, accessibility of the state-space realization has been proved.     

Adaptive neural network control for strict-feedback nonlinear systems using backstepping design

This paper focuses on adaptive control of strict-feedback nonlinear systems using multilayer neural networks (MNNs). By introducing a modified Lyapunov function, a smooth and singularity-free adaptive controller is firstly designed for a first-order plant. Then, an extension is made to high-order nonlinear systems using neural network approximation and adaptive backstepping techniques. The developed control scheme guarantees the uniform ultimate boundedness of the closed-loop adaptive systems. In addition, the relationship between the transient performance and the design parameters is explicitly given to guide the tuning of the controller. One important feature of the proposed NN controller is the highly structural property which makes it particularly suitable for parallel processing in actual implementation. Simulation studies are included to illustrate the effectiveness of the proposed approach.     

Improving returns on stock investment through neural network selection

The Artificial Neural Network (ANN) is a technique that is heavily researched and used in applications for engineering and scientific fields for various purposes ranging from control systems to artificial intelligence. Its generalization powers have not only received admiration from the engineering and scientific fields, but in recent years, the finance researchers and practitioners are taking an interest in the application of ANN. Bankruptcy prediction, debt-risk assessment and security market applications are the three areas that are heavily researched in the finance arena. The results, this far, have been encouraging as ANN displays better generalization power as compared to conventional statistical tools or benchmark.

With such intensive research and proven ability of the ANN in the area of security market application and the growing importance of the role of equity securities in Singapore, it has motivated the conceptual development of this project in using the ANN in stock selection. With its proven generalization ability, the ANN is able to infer from historical patterns the characteristics of performing stocks. The performance of stocks is reflective of their profitability and the quality of management of the underlying company. Such information is reflected in financial and technical variables. As such, the ANN is used as a tool to uncover the intricate relationships between the performance of stocks and the related financial and technical variables. Historical data such as financial variables (inputs) and performance of the stock (output) are used in this ANN application. Experimental results obtained this far have been very encouraging.     

Selection of artificial neural network models for survival analysis with Genetic Algorithms

In follow-up clinical studies, the main time end-point is the failure from a specific starting point (e.g. treatment, surgery). A deeper investigation concerns the causes of failure. Statistical analysis typically focuses on the study of the cause specific hazard functions of possibly censored survival data. In the framework of discrete time models and competing risks, a multilayer perceptron was already proposed as an extension of generalized linear models with multinomial errors using a non-linear predictor (PLANNCR). According to standard practice, weight-decay was adopted to modulate model complexity. A Genetic Algorithm is considered for the complexity control of PLANNCR allowing to regularize independently each parameter of the model. The ICOMP information criterion is used as fitness function. To demonstrate the criticality and the benefits of the technique an application to a case series of 1793 women with primary breast cancer without axillary lymph node involvement is presented.     

自组织多模型逼近性能分析与改进

针对基于自组织映射神经网络的非线性函数逼近,研究其方法和原理,指出它与一般前向神经网络在逼近原理上的不同．在此基础上,进一步研究该方法的逼近性能,分析其两个不足之处,进而提出一种提高逼近性能的改进神经网络训练策略．最后通过仿真实例验证了所得结论,表明了改进方法的有效性．     

Adaptive neural network predictive control for nonlinear pure feedback systems with input delay

This paper presents an adaptive neural control design for nonlinear pure-feedback systems with an input time-delay. Novel state variables and the corresponding transform are introduced, such that the state-feedback control of a pure-feedback system can be viewed as the output-feedback control of a canonical system. An adaptive predictor incorporated with a high-order neural network (HONN) observer is proposed to obtain the future system states predictions, which are used in the control design to circumvent the input delay and nonlinearities. The proposed predictor, observer and controller are all online implemented without iterative predictive calculations, and the closed-loop system stability is guaranteed. The conventional backstepping design and analysis for pure-feedback systems are avoided, which renders the developed scheme simpler in its synthesis and application. Practical guidelines on the control implementation and the parameter design are provided. Simulation on a continuous stirred tank reactor (CSTR) and practical experiments on a three-tank liquid level process control system are included to verify the reliability and effectiveness.     

Adaptive neural network control for a class of uncertain nonlinear systems in pure-feedback form

A procedure is developed for the design of adaptive neural network controller for a class of SISO uncertain nonlinear systems in pure-feedback form. The design procedure is a combination of adaptive backstepping and neural network based design techniques. It is shown that, under appropriate assumptions, the solution of the closed-loop system is uniformly ultimately bounded.     

Artificial neural network models for indoor temperature prediction: investigations in two buildings

The problem how to identify prediction models of the indoor climate in buildings is discussed. Identification experiments have been carried out in two buildings and different models, such as linear ARX-, ARMAX- and BJ-models as well as non-linear artificial neural network models (ANN-models) of different orders, have been identified based on these experiments. In the models, many different input signals have been used, such as the outdoor and indoor temperature, heating power, wall temperatures, ventilation flow rate, time of day and sun radiation. For both buildings, it is shown that ANN-models give more accurate temperature predictions than linear models. For the first building, it is shown that a non-linear combination of sun radiation and time of day is important when predicting the indoor temperature. For the second building, it is shown that the indoor temperature is non-linearly dependent on the ventilation flow rate.     

Adaptive driver model using a neural network

An adaptive driver model for longitudinal movements of a vehicle has been developed. It incorporates a conventional feedback brake controller, and both fixed and adaptive neural network controllers to produce the throttle demand. It has been interfaced with a vehicle model in a Simulink environment, and simulation studies indicate a high level of performance. Implementation of the adaptive driver model within a real-time environment has also been realized successfully. This work was presented in part at the 7th International Symposium on Artificial Life and Robotics, Oita, Japan, January 16–18, 2002     

Constructing neural network sediment estimation models using a data-driven algorithm

Artificial neural network (ANN) models are designed for suspended sediment estimation using statistical pre-processing of the data. Statistical properties such as cross-, auto- and partial auto-correlation of the data series are used for identifying a unique input vector to the ANN that best represents the sediment estimation process for a basin. The methodology is evaluated using the flow and sediment data from the stations Quebrada Blanca and Rio Valenciano in USA. The result of the study indicates that the statistical pre-processing of the data could significantly reduce the effort and computational time required in developing an ANN model. Three ANN training algorithms are also compared with each other for the selected input vector.     

A comparative study for forecasting using neural networks vs genetically identified Box&Jenkins models

This paper aims to discuss the results and conclusions of an extensive comparative study on the forecasting performance between two different techniques: a genetic expert system in which a genetic algorithm carries out the identification stage embraced in the three- phase Box&Jenkins univariate methodology; and a connectionist approach. At the heart of the former, an expert system rules the identification-estimation-diagnostic checking cyclical process to end up with the predictions provided by the SARIMA model which best fits the data. We will present the connectionist approach as technically equivalent to the latter process and due to its, alas, lack of any conclusive existent algorithm able to identify both the optimal model and architecture for a given problem, the three most common models presently at use and 20 different architectures for each model will be examined. It seems natural that if a comparison is to be made in order to provide a straight answer as to whether or not a connectionist approach outperforms the univariate Box&Jenkins methodology, the benchmark should clearly be the set of time series analysed in the work Time Series Analysis. Forecasting and Control by G. E. Box and G. M. Jenkins. Series BJA through to BJG give a total of 1200 plus measures to evaluate and compare the predictive power for different models, architectures, prediction horizons and pre-processing transformations.     

Neural network compensation control for mechanical systems with disturbances

Two novel compensation schemes based on accelerometer measurements to attenuate the effect of external vibrations on mechanical systems are proposed in this paper. The first compensation algorithm exploits the neural network as the feedback-feedforward compensator whereas the second is the neural network feedforward compensator. Each compensation strategy includes a feedback controller and a neural network compensator with the help of a sensor to detect external vibrations. The feedback controller is employed to guarantee the stability of the mechanical systems, while the neural network is used to provide the required compensation input for trajectory tracking. Dynamics knowledge of the plant, disturbances and the sensor is not required. The stability of the proposed schemes is analyzed by the Lyapunov criterion. Simulation results show that the proposed controllers perform well for a hard disk drive system and a two-link manipulator.     

基于二维主成分分析与卷积神经网络的手写体汉字识别

随着计算能力的飞速增长、训练数据的不断积累以及非线性激活函数的不断完善,卷积神经网络（CNN）在手写体汉字识别中表现出较好的识别性能。针对CNN识别手写体汉字识别速度慢的问题,将二维主成分分析（2DPCA）与CNN相结合识别手写体汉字。首先,利用2DPCA提取手写体汉字的投影特征向量;然后,将得到的投影特征向量组成特征矩阵;其次,用组成的特征矩阵作为CNN的输入;最后,用Softmax函数进行分类。与基于AlexNet的CNN模型相比,所提方法的运行时间降低了78%,与基于ACNN与DCNN的模型相比,所提方法的运行时间分别降低了80%与73%。实验结果表明,该方法在不降低识别精度的同时,可以减少识别手写体汉字的运行时间。     

Dissipative-based adaptive neural control for nonlinear systems

A dissipative-based adaptive neural control scheme was developed for a class of nonlinear uncertain systems with unknown nonlinearities that might not be linearly parameterized. The major advantage of the present work was to relax the requirement of matching condition, i.e., the unknown nonlinearities appear on the same equation as the control input in a state-space representation, which was required in most of the available neural network controllers. By synthesizing a state-feedback neural controller to make the closed-loop system dissipative with respect to a quadratic supply rate, the developed control scheme guarantees that the L2-gain of controlled system was less than or equal to a prescribed level. And then, it is shown that the output tracking error is uniformly ultimate bounded. The design scheme is illustrated using a numerical simulation.     

RealNet: a neural network architecture for real-time systems scheduling

Real-time embedded systems are spreading to more and more new fields and their scope and complexity have grown dramatically in the last few years. Nowadays, real-time embedded computers or controllers can be found everywhere, both in very simple devices used in everyday life and in professional environments. Real-time embedded systems have to take into account robustness, safety and timeliness. The most-used schedulability analysis is the worst-case response time proposed by Joseph and Pandya (Comput J 29:390–395,1986). This test provides a bivaluated response (yes/no) indicating whether the processes will meet their corresponding deadlines or not. Nevertheless, sometimes the real-time designer might want to know, more exactly, the probability of the processes meeting their deadlines, in order to assess the risk of a failed scheduling depending on critical requirements of the processes. This paper presents RealNet, a neural network architecture that will generate schedules from timing requirements of a real-time system. The RealNet simulator will provide the designer, after iterating and averaging over some trials, an estimation of the probability that the system will not meet the deadlines. Moreover, the knowledge of the critical processes in these schedules will allow the designer to decide whether changes in the implementation are required.This revised version was published online in November 2004 with a correction to the accepted date.