Enhanced maximum Power Point tracking algorithm and DC-DC converters optimal design methodology powered by the earthquake optimization algorithm

Nowadays, owing to the growing interest in cleaner energy systems, energy harvesting from Photovoltaic (PV) sources has gained greater relevance due to their worldwide suitability. PV systems are responsible for supplying more than 500,000 [GW] of the electrical energy consumed worldwide. Therefore, different power converters topologies, control algorithms and techniques have been developed to maximize the energy harvested by PV sources. Among the research topics related to PV applications, Maximum Power Point Tracking (MPPT) methods are usually employed together with DC/DC converters to control the impedance at the output of PV arrays, which allows changing the current and voltage supplied by the PV source to achieve a dynamic optimization of the energy transferred. Classical MPPT algorithms such as, Perturb and Observe (P&O) guarantee correct tracking behavior with low calibration parameter dependence but with a compromised relationship between the settling time and steady-state oscillations. Thus, methods like Particle Swarm Optimization (PSO) based techniques have improved the settling time and the steady-state oscillations, but the performance of the PSO-MPPT is highly susceptible to a correct and precise parameter calibration, which may not always ensure the expected behavior. Therefore, this work presents a novel alternative for MPPT applications, based on the Earthquake Optimization Algorithm (EA), which contributes a solution with an easy parameters calibration and improved dynamic behavior. Hence, results show that the contributed MPPT can be easily suited to different power applications and converter topologies, where the proposed solution reduced between 12% and 36% of the energy wasted compared to the P&O and PSO-based proposals. Yet, aggressive dynamic changes may cause the metaheuristic MPPTs to get stuck in local minimum solutions due to their convergence properties, which is why this work also presents an Artificial Neural Networks (ANN) contribution as a reliable reinitialization signal for metaheuristic MPPT algorithms, whose results show that the solar irradiation changes detection through the ANN achieved over 99\% of accuracy. Still, the experimental validation of the contributed MPPT control structure requires an efficient and reliable testbed for the tests, which is why MPPTs are usually implemented through DC-DC converters. Yet, components selection and precise estimation of circuit parameters are issues that can improve the converter’s performance; which is why, metaheuristic optimization algorithms can be applied using the mathematical model of DC-DC converters in order to optimize their performance through an optimal components selection. Therefore, this work also contributes a novel optimal design methodology for DC-DC converters, where the validation designs are optimized to enable an optimal dynamic behavior regarding the validating application. Henceforth, The experimental results validate the design methodology, showing ripple improvements and operating power range extension, which are key features to have an efficient performance in DC-DC converters. Thereby, the contributions of this work were completely validated through an integration case study, which will be later addressed in this work. The technology developed through the EA contributions in this work, reached a Technology Readiness Level (TRL) 5.