Design of a novel robust adaptive cascade controller for DC‐DC buck‐boost converter optimized with neural network and fractional‐order PID strategies

Abstract

AbstractA cascade technique with two control loops is designed for a DC\DC Buck‐Boost converter that is a right half‐plane zero (RHPZ) structure called a non‐minimum phase system. This concept presents several challenging constraints for designing well‐behaved control techniques. Cascade controllers can provide various benefits compared with single loop controllers such as higher safety, higher robustness, and higher stability. This strategy assumes the system as a black‐box structure without the need for a mathematical model of the system. This benefit can decrease the computational burden and provides faster dynamics along with ease of implementation. This technique consisted of an outer Fractional‐order PID voltage controller tuned with the Antlion Optimizer (ALO) algorithm, which provides a reference current for the inner control loop of the Neural Network‐based LQR (NN‐LQR) controller. The basic principle in cascade controllers is a more rapid performance of the inner loop that has been satisfied with the NN‐LQR strategy, which optimizes and tunes the gains of the LQR controller and shows faster dynamics and higher robustness. It should be mentioned that the number of neurons is limited to 2 and 4 in each layer to decrease the computational burden with lower complexity. Also, the ALO algorithm is a modern nature‐inspired algorithm used to tune the PID gains with better results under‐constrained problems with diverse search spaces. Considering the negative impacts of various disturbances on a power converter, a Fractional‐order‐based PID (FO‐PID) control technique is a proper alternative since it shows higher robustness in load uncertainties along with better dynamical responses based on its extra degree of freedom. Moreover, to evaluate the superiority of this controller, two other controllers are designed using the PSO algorithm for PID and FO‐PID controllers. Finally, the presented cascade controller has been tested in various working conditions through simulation and experiment results.