Parameter identification of Bouc-Wen model for Magnetorheological (MR) fluid Damper by a Novel Genetic Algorithm

In this research, novel genetic algorithm (nGA) is proposed for Bouc-Wen modle parameters esstimation for magnetorheological (MR) fluid dampers. The optimization efficiency is improved by modifying the crossover and mutation steps of a GA. In the crossover stage, the probability of reproducing offspring from the same parent (same mother and father chromosome) is done to be zero, which may happen in the standard GA, and the probability of a chromosome to be selected for mating is based on error rank weighting of the chromosomes. Additional fitness evaluation of chromosomes will take place in between the crossover and mutation steps to save the best chromosome found so far, which is not implemented in the standard genetic algorithm (GA). The model is validated by comparing its simulation output force (F-sim) with experimentally generated MR damper force (F-exp). The mean absolute error, standard deviation and number of generations for convergence are taken as a criterias for performance evaluation. With these ctriterias, the proposed novel GA outperform better than the other researches. The accuracy is improved by 46.67% compared to standard GA. The proposed novel GA for Bouc-Wen model parameter identification can be used for any MR damper control system with better accuracy.