A techno-economic and ai-based optimization framework for hybrid energy systems supplying rural telecom base stations

Area of Science:

• Renewable Energy Systems
• Artificial Intelligence in Energy Management
• Telecommunications Infrastructure

Background:

• Remote Base Transceiver Stations (BTS) often rely on conventional, fuel-intensive power sources.
• Integrating Hybrid Renewable Energy Systems (HRES) presents technical and economic challenges for decentralized power.
• Optimizing energy management is crucial for the reliability and sustainability of remote telecom infrastructure.

Purpose of the Study:

• To introduce a strict AI-based framework for analyzing HRES in technical and economic aspects for remote BTS.
• To evaluate the performance of various forecasting algorithms for energy management in HRES.
• To demonstrate the potential of intelligent energy management to reduce fossil fuel dependency and improve system efficiency.

Main Methods:

• Development of an AI-based framework for HRES analysis.
• Utilizing a year of hourly simulation data to train and validate forecasting algorithms (Linear Regression, Decision Tree, SVM, GPR, KARMA, NN).
• Energy Management System (EMS) simulations to assess load serving capabilities and fuel consumption.

Main Results:

• The proposed HRES system provides 1.2 kW at -48 V and 23 A, meeting telecom load requirements.
• Hybrid solar-wind energy constituted an average of 78.6% of the total daily load served.
• Fuel-based system usage was reduced by over 76% compared to conventional systems.

Conclusions:

• Intelligent forecasting and optimal dispatch strategies significantly improve HRES efficiency.
• The AI-driven framework effectively reduces fossil fuel dependency in decentralized telecom towers.
• The study confirms enhanced sustainability of HRES for remote BTS applications through advanced energy management.