Optimizing biomass energy production in the southern region of Iran: A deterministic MCDM and machine learning approach in GIS

Abstract

This study employs a deterministic approach, distinguishing itself from other renewable energy evaluations, to assess the potential of electrical energy derived from biomass sources in the southern region of Iran. The primary objectives include pinpointing optimal locations for maximal biomass production and subsequent energy generation within distinct climates and topographies, using fuzzy- Analytic Hierarchy Process (AHP). Additionally, Principal Component Analysis (PCA) identify key factors influencing biomass and energy production. The study quantifies electrical and thermal energy derived from biomass sources across various climates. The findings indicate that regions with lower altitudes and humid climates (1530 km²) demonstrate superior biomass performance, leading to increased electrical and thermal energy production. The feature selection process highlights the significant impact of climate and soil characteristics on biomass production and energy output. Analysis of biomass energy production reveals maximum electrical energy production ranging from 674.88 kWh/ha to 711.36 kWh/ha. The results of the Long Short-Term Memory (LSTM) method confirm its high accuracy in estimating electrical energy, with a significant correlation coefficient of 0.98. We conclude that by identifying locations with the best biomass sources based on climate, it is possible to increase the derived electrical energy. These insights are critical for informing energy policies aimed at optimizing biomass energy production and its integration into sustainable power grids.