Optimum Control System Design of a Distributed Parameter Process

摘要

It is seen that the optimum controller setting on the basis of simulated process model for complex processes performs better than more generalized model assumptions like FOPDT approximation. To exemplify the notion, a double pipe heat exchanger is numerically simulated as the process. A control system with a PID controller and ratio type final control element to realize temperture forcing is simulated around the process for a set of process patameter values. From open loop step response of the heat exchanger, the FOPDT model parameters are found and controller settings are worked out following the FOPDT guideline. For the same set of process parameter values, a three parameter gradient search optimiation routine is employed to minimize integral time absolute error (ITAE) calculated from the close loop set point step response of heat exchanger control system by adjusting the three PID controller parameters. Using controller parameter values obtained from FOPDT guideline as initial guess, the search has been nitiated. After establishing the absolite minimum point, the controller parameter values are recorded corresponding to the process parameter value. The ITAE value, thus obtained, is compared to that of the close-loop response with controller settings according to FOPDT guidelnes. The procedure has been repeated for a number of combinations of process parameter values. The imprvement in the close-loop performance is appreciable. A set of graphs have been prepared to obtain optimum controller setting for a given set of process parameters. Such an approach may be used in conjunction with an adaptive control algorithm for online optimum parameter estimation.