Dynamic modeling and control of nonlinear processes using neural network techniques

Karsai, Gábor; Andersen, Kristinn; Cook, Gerald; Ramaswamy, K.

doi:10.1109/isic.1989.238681

Cited by 4 publications

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method presented here uses neural networks to memorise the parameter updating process of standard RLS algorithms and to relate these parameters t o the operating conditions. The more normal methods use the delayed sequence of the process input and output signals as input patterns to neural net-0-7803-1 206-6/93/$3.00 Q1993 IEEE works which then model the process dynamics by comparing the measured process output with that of the neural networks [5]. This method however combines the simplicity and speed of convergence of RLS algorithms with the ability of neural networks to learn any complex nonlinear functions with any desired accuracy [6].…”

Section: Introductionmentioning

confidence: 99%

A neural network based RLS parameter estimation algorithm and its application in predictive control

Geng

Geary

Proceedings of 8th IEEE International Symposium on Intelligent Control

View full text Add to dashboard Cite

This paper presents a novel method which uses neural networks to improve the performance of RLS algorithms in estimating the parameters of nonlinear processes. Its application in a process gain-adaptive GPC algorithm is also described.Normally in order to model the dynamics of a process, a delayed sequence of its input and output signals is used as input patterns to neural networks. This method however uses neural networks to learn the parameter updating process of standard RLS algorithms and to relate these parameters to the operating conditions. Experimental results when using this method in modelling and control of a heat transfer process of an airhandling plant are reported and show great potential.

show abstract

Section: Introductionmentioning

confidence: 99%

A neural network based RLS parameter estimation algorithm and its application in predictive control

Geng

Geary

Proceedings of 8th IEEE International Symposium on Intelligent Control

View full text Add to dashboard Cite

show abstract

Neural networks and the inverse kinematics problem

et al. 1993

View full text Add to dashboard Cite

Inverse kinematics is a fundamental problem in robotics. Past solutions for this problem have been realized through the use of various algebraic or algorithmic procedures. In this paper the use of feedforward neural networks to solve the inverse kinematics problem is examined for three different cases. A closed kinematic linkage is used for mapping input joint angles to output joint angles. A three-degree-of-freedom manipulator in 3D space is used to test mappings from both cartesian and spherical coordinates to manipulator joint coordinates. A majority of the results have average errors which fall below 1% of the robot workspace. The accuracy indicates that neural networks are an alternate method for performing the inverse kinematics estimation, thus introducing the fault-tolerant and high-speed advantages of neural networks to the inverse kinematics problem.This paper also shows the use of a new technique which reduces neural network mapping errors with the use of error compensation networks. The results of the work are put in perspective with a survey of current applications of neural networks in robotics.

show abstract