Feedback Correction Of Control Systems Based On Differential Evolution

In industrial production, as the increasement of the performance requirements of controlled object, it’s usually very difficult to meet all the target parameters at the same time according to the automatic control system which is designed according to the controlled object and the given technical requirements. Correction is an effective way to improve the overall characteristics of the system. With different locations in the system, correction device with variable parameters is corresponding to various correction methods. Feedback correction is a commonly used method for correction of the control system. The steps of the traditional method of feedback correction are very complex with large amount of calculation. Therefore, it’s particularly important to find an effective, simple and convenient method for configuration parameters of feedback correction device.This paper presents a differential evolution-genetic algorithm, with the operation of replacing the crossover part of differential evolution algorithm with recombinant of genetic algorithm, to enrich the crossover methods of differential evolution algorithm and improve the ability of local search. This paper briefly introduces the designing ideas of feedback correction control system of three different connection mode, and the correcting device linked with the part to be corrected. Regard frequency domain margin and phase as the control parameters, which control the system feedback correction. In this way, the feedback correction is transformed into the problem of intelligent optimization algorithm seeking the optimal solution of the problem. For each connection, feedback correction device is respectively taken low order and high order structure, through the comparative analysis of the optimization results, determine the structure and parameters of the feedback correction device on a better allocation.In order to further simplify the process of correction system, this paper use MATLAB GUI to complete the design of the graphical user interface. The users only needs to input system and expectations, the real control parameters and the corresponding correction device parameters and before and after correction of Bode diagram and step response diagram can be obtained, making the before and after correction performance more convenient and intuitive.