FEEDBACK CONTROL OF INDUSTRIAL PROCESSING USING ON-LINE OPTIMIZATION TECHNIQUE

Abstract

Now-a-days industrial processes are becoming complex and nonlinear in nature. In competitive world today the most. necessary thing for any industry to compete is the quality control of its product which may not be achieved without definite control of processes. But as these processes are becoming more and more complex in nature, one time control of the process can not be said to be perfect. But one has to change the control scheme according to product desired. It will be a very costly affair if one does this by hardware and that too will not be flexible. But by the use of digital computers one can achieve the flexible control of the processes. The word flexible indicates here that when one wants to change its control scheme one does not need to change its complete setup, but what all is needed to do is to change some of the parameters in the program or at the maximum one has to change the software used. By doing merely this one can achieve the desired control of the processes. In our present work we have developed algorithm for adaptive control for controlling the processes and an algorithm for, parameter estimation (using recursive least square error method) which involves lesser number of parameters to be estimated. Our subroutines OPTIMIZE ( to de mine the optimum input to the system ) and NEWPARAMETER ( to estimate model parameter ) takes execution time 0.02 sec. and 0.04 sec. respectively on UNIX V/386 system. For simulation purpose we I have assumed sampling time to be 0.2 sec. for armature controlled DC motor example and 0.2 hr. for concentration control example