A hybrid framework for optimal site selection and energy resource forecasting for off-grid hybrid energy systems: integrating GIS, hesitant fuzzy linguistic MCDM, and forecasting tools

Abstract

Transitioning to sustainable renewable energy is essential for achieving a carbon-neutral economy. Decentralized hybrid energy systems, which utilize locally available resources, can help bridge the gap between energy demand and supply. However, identifying optimal locations and forecasting renewable resource availability remain major challenges. This study proposes an integrated framework combining Geographic Information Systems (GIS), hesitant fuzzy multi-criteria decision-making, and fuzzy forecasting to address these issues. The primary goal is to identify the most suitable site for decentralized hybrid energy deployment. Sensitivity and obstacle degree analyses are conducted to test the robustness of the site selection and highlight key influencing factors. The methodology, demonstrated using spatial data from a central Indian state, is adaptable and broadly applicable. Among nine alternatives, Sailana emerged as the most favorable location due to its strong resource potential and favorable geographic, economic, and social conditions. Additionally, the fuzzy forecasting method showed superior accuracy over optimized neural network models, reducing mean relative error by 33–80 %. This research contributes both a practical tool for stakeholders and an enhancement to theoretical models for renewable energy site selection.