Research On The Effect Of Selection Of Control System Objective Function On Optimization Results Quality

With the rapid development of industrial, the unit capacity is increasing, the quality of control system provides higher needs. That is to say, the unit capacity is not only required fast quasi-steady, but also given consideration to both economy and the utilization rate of equipment, which on the parameters optimization of PID control system has higher requirements. As everyone knows,there is the objective function and optimization strategy two aspects to be solved. In this paper, a mature particle swarm algorithm was applied in single-loop system and the main steam temperature double loop system as the optimization strategy, Learn from the optimized curve and parameters,direct objective function in meeting their own constrains at the same time, the system reached steady state time slowly. However the Indirect target function was more better,though the system had concussion. the advantages of a comprehensive integrated objective function takes into every aspects of the system. Because of the complexity of the system, no self-balancing object, nonlinear control systems, pure delay system and inertial object system had been analysised and given the optimization parameters and curves. We can see it clearly, the optimized system of direct objective function can meet the gi ven value requirements but sacrifice the system’s speedability. Contrary to the error type function, the optimized system can reach steady time faster and the stability is just passible. The non-self-balancing system, nonliner system and pure delay system of optimized by comprehensive objective function had better optimization, but for the large inertia object after optimized various indicators were poor and undesirable in engineering, The achievement in this paper provided a reference for power plant staff. Finally,we designed a simple engineering realization method which on objective function optimization PID parameters. Through the validation of simulation proved its feasibility in practical application and provided convenience production for future applications.