Mixed H2/H8 State Feedback Controllers of EPS System Based on Genetic Algorithm

Electric power steering system model was set up in this thesis. The mixed H2/H∞ state feedback controllers with robustness and better system performance are designed through making H∞ controllers have H2 optimization by GA. In this method, the coding is float, the selection operator is rank-based fitness assignment and elitist model, the crossover operator is real valued recombination, and the mutation operator is real mutation. Simulation have made to the pinion and rack electric power steering, the results show that the mixed H2/H∞ state-feedback controllers have not only better H∞ performance but also better H2 performance.