Research On Optimal Design Of Controllers With Improved Mind Evolutionary Algorithm

Mind Evolutionary Algorithms (MEA) were proposed by simulating the processes of human mind. It is a new kind of potential evolutionary algorithms. There are many different applications of MEA methods such as numerical optimization problems, some non-numerical optimization problems, a traveling salesman problem, job-shop scheduling, and modeling for systems of ordinary differential equations, and so on. These problems are all solved successfully with MEA. As a new search algorithm, MEA still has some disadvantages and defects. The main content of this thesis includes the following:1. A survey of the origin and the development status of Evolutionary Algorithms is summarized and the status of the parameter optimization of controllers is introduced. Also the general principle and methods of MEA are introduced, the faults and the improvements of MEA are discussed.2. The method of self-adjustive mind evolutionary algorithm (SMEA) is presented in the thesis, which adds two factors that can self-adjust according to the evolution direction and time to the basic MEA. The simulation results show that the optimal PID controller using the SMEA has satisfactory performance, and which are better than that of the basic genetic algorithm.3. SMEA is used to design optimal parameters of Fuzzy-PID control, which belong to a typical single variable intelligent controller. The simulation tests are made and the results demonstrate the efficiency of the proposed controller.4. The multivariate neuron controller based on SMEA is designed for a unit power plant. Simulation experiments show that the performance of the optimal controller is improved a lot.