The Fractional-Order Proportional-Integral-Derivative Controller Tuned with Hybrid Particle Swarm Optimization-Artificial Bee Colony Algorithm for Boost Converters: An Experimental Study

Abstract This paper presents Fractional-Order Proportional-Integral-Derivative (FO-PID) controllers for a voltage feedback boost converter operated in continuous conduction mode. If parameters of a classical PID controller are considered, it is seen that a FO-PID has two additional parameters to be set. Achieving optimum values for those parameters is a laborious task. Here, a hybrid optimization algorithm is developed as a serial connection of particle swarm optimization (PSO) and artificial bee colony (ABC) algorithm to tune FO-PID parameters. Handling the cost functions of integral of absolute error and integral of squared error (ISE) with PSO, ABC and hybrid algorithm, six different FO-PID controllers are achieved. Then, integer order approximates of those FO-PIDs are found, and discrete transfer functions are derived utilizing Backward Euler method. With the help of simulation results and experimental results, the performances of a PID controller and six different FO-PID controllers are examined. To evaluate controllers’ performance, simulation data and experimental data are processed and settling time, overshoot and mean squared error (MSE) performance indices are calculated. The FO-PID HYB(ISE) controller, which has been developed by using the hybrid algorithm and ISE cost function, achieved 49.4% improvement in overshoot performance indicator with regard to PI controller in the simulations, and provided 23.04% performance increase in the MSE scale compared to PID controller in the experiments. Consequently, the control performance measures validated that the FO-PID HYB(ISE) controller is superior to the others.