Improvement In The Mind Evolutionary Algorithm And Its Application

Parameter optimization of controller directly influences control results and has in touch with safety and economy of units. Mind evolutionary algorithm(MEA) as a new evolutionary algorithm performs efficient and heuristic search, not blindfold and completely stochastic searching. So MEA is used to optimize controller's parameter in this thesis. According to the defects lying in mind evolutionary algorithm (MEA),two ways have been developed to improve MEA in the paper. First, simple method is an effective local optimization method. So it is introduced into similar taxis to enhance the efficiency of local optimization; Second, niche is introduced to MEA improved the ability of global search. Niche can effectively maintain population diversity and generate dynamic search space. The improved mind evolutionary algorithm is used to optimize the PID controller parameters. The controlling of main steam temperature in power plant is studied in this paper. The simulation results show that the improved MEA is very effective and robust.