Performance index-based evaluation of Draco Lizard Optimizer for future MPPT applications under partial shading conditions

Hashim, Norazlan and Abdullah, Mohd Huzaifah and Muhammad, Khairul Safuan and Abd Samat, Ahmad Asri and Ab. Rahman, Azhan (2026) Performance index-based evaluation of Draco Lizard Optimizer for future MPPT applications under partial shading conditions. Journal of Electrical and Electronic Systems Research (JEESR), 29 (1): 23. pp. 187-195. ISSN e-ISSN : 3030-640X
Identification Number (DOI): 10.24191/jeesr.v29i1.023
Abstract

New metaheuristic optimization algorithms are introduced each year to address increasingly complex multimodal optimization problems. Consequently, selecting and evaluating a suitable optimization method for photovoltaic maximum power point tracking (MPPT) remains challenging, particularly because no standardized evaluation framework is available for this application. This study presents a performance index-based comparative evaluation of the Draco Lizard Optimizer (DLO) for global maximum power point (GMPP) identification under partial shading conditions (PSCs) as a preliminary assessment for future MPPT implementation. DLO is benchmarked against three established population-based metaheuristic algorithms: particle swarm optimization (PSO), whale optimization algorithm (WOA), and grey wolf optimizer (GWO). The proposed performance index (PI) integrates four weighted criteria relevant to assessing the suitability of metaheuristic algorithms for future MPPT applications: success rate (SR), convergence speed (CS), CPU time, and search agent standard deviation (STD), with respective weights of 0.55, 0.20, 0.15, and 0.10. Each algorithm was evaluated over 1,000 independent trials under three static PSC patterns representing left-, center-, and right-located GMPPs. DLO achieved the highest PI of 0.9350, followed by PSO, WOA, and GWO with PI values of 0.9137, 0.8718, and 0.8650, respectively. Although DLO did not attain the best value for every individual metric, it provided the strongest overall balance among GMPP identification reliability, convergence speed, computational efficiency, and search stability. These findings support the further investigation of DLO through converter-level MPPT simulation and experimental validation under dynamic partial shading conditions.

Item Details
Edit Item
Edit Item
Downloads & Files
[thumbnail of 149173.pdf]
Text
149173.pdf
Download (1MB)
Indexing & Metrics