ANN CONTROLLER FOR MIMO SYSTEM

Abstract

The primary objective of controller is to generate the actuating signal to the_ final control element of a given physical process to yield its desired response. The objective of the design of an intelligent controller—is—s-i-gin--filar -to that for an adaptive control system. The object with intelligent control is to design a system with acceptable performance characteristics over a- wide range of uncertainties. The term ANN controller refers to the Artificial neural network controller. The advantages of neural networks are two fold, one is its learning ability and another is its versatile mapping capabilities from input to output.Ver-satile mapping should provide a means of controlling complex systems which cannot be carried out well with conventional feedback controllers. The learning ability can reduce- human effort in designing controllers and it even suggests in discovering better control schemes than presently known. .In this dissertation work The concept of direct feedback controller-., a-sample configuration_ in which the neural- network serves as a direct feedback controller, is -implemented using Modified Genetic Algorithm based self trained ANN controller. The modelling of a simple stirred -mixing tank have been developed and the process objective is -.to control the level (in -effect total flow) and temperature of water in the tank. There are- -two inputs to the process namely the flow of cold water and the Clow of hot water into the tank. So the controlled var-iables are temperature . and total flow and the manipulated variables are cold water and' hot water flow rates. (v) For the developed mathematical model ANN controller is trained for the performance of minimizing the integral square error (ISE). For the training of controller a self learning scheme is used and the neural network is trained by batch learning process. Standard version of Genetic Algorithm is modified to apply it for neural network training. In Genetic Algorithm the improvement is triggered by Crossover, Mutation and Natural Selection. The system behavior for various types of disturbances is studied for this M.I.M.O. system with ANN controller so designed, this response is then compared for the same system controlled using PI controller.