Showing papers by "Hong Wang published in 1998"
TL;DR: It is shown that the control signals obtained can also make the real system output close to the set point, and the applicability of the proposed method is demonstrated.
Abstract: In this paper a direct adaptive neural-network control strategy for unknown nonlinear systems is presented. The system considered is described by an unknown NARMA model, and a feedforward neural network is used to learn the system. Taking the neural network as a neural model of the system, control signals are directly obtained by minimizing either the instant difference or the cumulative differences between a set point and the output of the neural model. Since the training algorithm guarantees that the output of the neural model approaches that of the actual system, it is shown that the control signals obtained can also make the real system output close to the set point. An application to a flow-rate control system is included to demonstrate the applicability of the proposed method and desired results are obtained.
260 citations
16 Dec 1998
TL;DR: In this paper, two robust solutions to the control of the output probability density function for multi-input and multi-output stochastic systems are presented, where the purpose of control input design is to minimize the difference between the probability density functions of the system output and a given one.
Abstract: This paper presents two robust solutions to the control of the output probability density function for multi-input and multi-output stochastic systems, where the purpose of control input design is to minimise the difference between the probability density function of the system output and a given one. The probability density function of the system output is approximated by a B-spline neural network with all its weights dynamically related to the control input. The measured probability density function of the system output is directly used to construct two robust control algorithms which are insensitive to the unknown input. The stability of the closed loop system are proved under certain conditions. An illustrative example is included to demonstrate the use of the developed control algorithms and desired results have been obtained.
46 citations
TL;DR: An extented solution to the control of the probability density function for the output of a class of nonlinear stochastic systems based on the fact that there exist many control systems where the requirements are set to control the shape of the probabilities density function of the system output.
24 citations
01 Sep 1998
TL;DR: A methodology based on bilinear system modelling and multilayer perceptron (MLP) neural network for modelling of such a complex system and genetic algorithm (GA) search and optimisation technique is proposed to train the neural network weights.
Abstract: The dynamic modelling of the wet end of the paper machines has been recognised as a challenging problem due to its nonlinear, complex, time-varying, time-delayed, and multivariable interactive properties. This paper presents a methodology based on bilinear system modelling and multilayer perceptron (MLP) neural network for modelling of such a complex system. Genetic algorithm (GA) search and optimisation technique is proposed to train the neural network weights. This logical combination has advantages of both physical and genetic neural modellings.
11 citations
Proceedings Article•
01 Jan 1998
8 citations
21 Jun 1998
TL;DR: In this article, a real-time control approach for a class of nonlinear unknown systems is presented, where all the involved nonlinear functions are online estimated by fuzzy logic units, using these online estimations, an adaptive nonlinear control algorithm is established which consists two loops, the inner loop and outer loop.
Abstract: This paper presents a novel real-time control approach for a class of nonlinear unknown systems. The system is assumed to be represented by an affine model where all the involved nonlinear functions are online estimated by fuzzy logic units. Using these online estimations, an adaptive nonlinear control algorithm is established which consists two loops, the inner loop and outer loop. The inner loop is used to compensate the unknown nonlinearities whilst the outer loop is utilised to eliminate the DC component in the tracking error for the closed loop system. A real application to a pH control process in paper making has been made and desired results have been obtained.
7 citations
TL;DR: In this article, an approach for the change detection of the output probability density functions for dynamic stochastic systems using B-splines neural network is presented. But this approach is limited to the detection of unexpected changes of particle size in paper-making.
7 citations
TL;DR: A heuristic approach for the problem of fault tolerant control of unknown nonlinear systems is discussed, which uses a heuristically determined feedback function for the compensation of the system response and a neural network model.
3 citations
01 Sep 1998
TL;DR: In this paper, a stable adaptive control algorithm for the control of the output probability density function for unknown time-invariant stochastic systems is presented, which is based upon the same type of dynamic model proposed by Wang (1997, 1998).
Abstract: This short paper presents a stable adaptive control algorithm for the control of the output probability density function for unknown time-invariant stochastic systems. An online parameter estimation algorithm is constructed using the measured output probability density functions of the system. Based upon the same type of dynamic model proposed by Wang (1997, 1998), a functional weighted type performance function is employed in the formulation of the adaptive control algorithm. It is shown that the stability of the closed loop system can be guaranteed under certain conditions. An applicability study of the proposed algorithm to the solid flocculation control in paper making is included, where a simulated example is employed to illustrate the use of the developed control algorithm.
2 citations