Control Strategy for the DC-DC Converter in Fuel Cell Vehicles.

Abstract

A BP neural network PID control strategy optimized based on a genetic algorithm is proposed to meet the requirements of a wide buck-boost range, low input current ripple, and a wide input voltage range for fuel cell vehicle buck-boost converters. By leveraging the global optimization capability of the genetic algorithm, the optimal initial weights of the BP neural network are calculated, avoiding the issue of local optima caused by dependence on initial weights during the BP neural network iteration process. Simulation experiments were conducted in MATLAB/Simulink, and the results show that compared with the traditional PID control strategy and the BP neural network PID control strategy, the system's response time was reduced by 62.5% and 59.46%, respectively, in the absence of load disturbances. When load disturbances were introduced, the system's average error was reduced by 81.36% and 25.57%, respectively, demonstrating the stronger robustness and the faster recovery to the desired voltage value. These results indicate that the proposed control strategy exhibits strong robustness and disturbance rejection, fast response speed, and high accuracy.