Integrating Multifactorial Regression Models, Artificial Neural Networks, and Eco-Dynamic Simulations for Assessing Energy Sustainability in the European Union and Emerging Member States

This study aims to develop an integrated analytical framework for assessing energy sustainability in the EU and emerging member states. It examines the interrelationships between environmental quality—measured by greenhouse gas emissions—economic growth, and energy consumption from both renewable and non-renewable sources. Multifactorial linear regression models are employed and validated through artificial neural networks to enhance robustness and generalization capacity. Based on the identified correlations, the Environmental Kuznets Curve hypothesis is tested to assess patterns of environmental degradation. The analysis then advances to a dynamic framework using the Grey Lotka–Volterra model, which captures nonlinear effects and the mechanisms of competition and cooperation among variables. By exploring simultaneous and evolutionary interactions, the study reveals how these relationships shift over time, identifies inflexion points, and simulates public policy scenarios that may accelerate or hinder decarbonization. This approach provides a comprehensive and policy-relevant perspective on the dynamics of the energy transition.