Adaptive double-Loop PID control method of DC motor based on the GA-FNC algorithm

The DC torque motor is an important part in the missile guidance system, and its servo control strategy is related to tracking performance closely. This paper not only established but also analyzed a mathematical model of the DC torque motor to improve the missile´s hitting accuracy and flying stability. Then the genetic algorithm and the fuzzy BP neural network technology are applied into the position-speed dual closed-loop PID controlling algorithm of the DC torque motor. Because the DC torque motor controlling has the characteristics of time-varying and nonlinearity, in order to achieve the minimum position deviation of the guidance system. The controlling strategy uses the genetic algorithm to off-line optimize the parameters (c_cj,b_j) of the Gaussian membership function and the network structure of the fuzzy controller which have a global impact on the system, and uses the BP algorithm to on-line adjust the weight parameters of the controlling output which has a localized impact on the system, The simulation results show that the method which the genetic algorithm and the fuzzy neural network technology are applied into the dual closed-loop PID controlling algorithm of the DC torque motor greatly improves the self-learning ability and the robustness of the system, and significantly improved the stability of the controlling process.