Séminaire, SDA, Dr. João Soares,Researcher, GECAD, Polytechnic of Porto, Porto, Portugal , 13 Oct. 2025

	Master and Doctoral School Unit « Sustainable Development Applications »

Evolutionary Computation for Complex Energy Problems

João Soares
Polytechnic of Porto, Porto, Portugal

October 13, 2025, 14pm – 16pm
Agora 1, ESPRIT, campus Cité Scientifique

Abstract

Modern power systems combine large-scale, nonlinear physics, and highly uncertain conditions under which classical optimization often breaks down due to scalability, runtime, and memory limits. This talk shows how Computational Intelligence (CI), with an emphasis on Evolutionary Computation (EC), complements exact methods by delivering high-quality solutions under tight time and data constraints. Prof. Joao Soares discuss
families of EC methods, highlighting Differential Evolution and hybrid-adaptive variants (e.g., HyDE/HyDE-DF)— and the scientific practices (fair comparisons, statistical testing, and automatic configuration) that make them reliable for energy applications. Through case studies, Prof. Soares illustrates EC for day-ahead energy resource management with risk (CVaR), local electricity markets under distribution-network constraints, and EV smart charging/V2G dispatch. To improve trust and adoption, the talk introduces eXplainergy, an explainablemetaheuristics framework that links search behavior, parameter choices, and landscape analysis to transparent performance reports. He also presents GECAD’s open platforms, Meta-ERM and the ERM competitions series, which provide accessible algorithms, benchmarks, and results validated across international events since 2017.
Attendees will leave with practical guidance on when EC is better, how to configure it  responsibly, and where to obtain open tools and datasets for real-world deployment.

About the Speaker

João Soares

Researcher, GECAD – Research Group on Intelligent Engineering and Computing for Advanced
Innovation and Development, Polytechnic of Porto, Porto, Portugal
Email: jan@isep.ipp.pt
url: https://www.researchgate.net/profile/Joao_Soares5
url2: https://scholar.google.pt/citations?user=tkYnl0YAAAAJ

João Soares ottained his PhD degree in Electrical and Computer Engineering from UTAD University in 2017. He currently conducts research at GECAD – Research Group on Intelligent Engineering and Computing for Advanced Innovation and Development, in the School of Engineering of the Polytechnic of Porto. He has been a regularly invited visiting professor at Ecole Centrale de Lille – L2EP since 2019. He has contributed to over 250 publications in the fields of power systems and computational intelligence. His works have been cited over 6500 times (h-index 45 in Google Scholar). He has participated in several national and international research projects funded by various organizations. João Soares is currently chair of the IEEE Computational Intelligence Society (CIS) Task Force 3 and has been contributing to bridging the gap between computational intelligence and power system engineers, with numerous efforts, particularly through dedicated algorithm competitions in international venues since 2017.