Automatic Hyper-Heuristic to Generate Heuristic-based Adaptive Sliding Mode Controller Tuners for Buck-Boost Converters

Metaheuristics are commonly used to solve complex and challenging problems, particularly in electrical system applications. Nevertheless, there is a colorful palette of metaheuristics to select from, which can be overwhelming for a practitioner with solid experience in controlling electrical systems. Still, it is well-known that empirical or analytical tuning of controllers is no trivial labor. This work implements a methodology to address the Metaheuristic Composition Optimization Problem to coin a heuristic-based technique that guides an initially random population of individuals to find the optimal set of parameters of an Adaptive Sliding Mode Controller in the search space. As a case study, we implement this methodology for controlling the dynamic response of a DC-DC Buck-Boost converter under different conditions, including nonlinear perturbations. The objective is clear: find the optimal heuristic-based solver to determine the control parameters satisfying a predefined performance. Our experimental results reveal the reliability and potential of the proposed methodology when finding suitable solutions for control system design applications. We also support our findings with statistical analyses performed on the results obtained by the tailored metaheuristic against other classical heuristic-based methods.