Application of fuzzy control based on adaptive neural network in high voltage output boost circuit

Abstract

To improve the performance of the boost circuit, a fuzzy inference systems (FIS) design method based on adaptive neural networks (ANN) system identification is proposed for the boost circuit. Using ANN for system identification based on training data to generate initial first-order Takagi-Sugeno (T-S) FIS. Adjust the FIS parameters by comparing them with the testing and checking data, and iterate until the error is within an acceptable range to form the final FIS. The steady-state and dynamic capabilities of the boost circuit under FIS control have been verified through simulation and experiments to be superior to traditional proportion integral differential (PID) control. The experimental results show that when the input voltage jumps from 28 V to 22 V, the boost speed of the boost circuit based on FIS control is improved by 21.5% compared to PID.