Multi-Objective Optimization and Feasibility Study of Hybrid Energy System Considering Future Load of 2041 using Machine Learning

Electricity usage in Bangladesh is drastically rising, as a result of industrial expansion and rising living standards, particularly in the major cities. Despite the country’s ever-increasing demand, she is already facing a power crisis around the nation. Furthermore, the majority of the electricity generation units use traditional sources, which have finite limits. Chittagong, one of the country’s important cities, also faces these problems. This research work addresses an issue regarding the ever-increasing energy consumption in Chittagong City of Bangladesh and expresses its concerns for the continuation of the recent energy crisis in the city. To address these issues, the study proposes an AI-based Load Forecasting Model to predict the future load of 2041 and a Multi-Objective Hybrid Energy System (HES) that combines renewable and conventional sources to meet the predicted demand in an optimized way. The goal of the study is to propose a long-term energy plan for a densely populated region in Chittagong (KHULSHI) using HES to reduce the pressure on the national energy grid and eradicate the deficiency of electricity by the year 2041. Deep Learning Models like Long-Short-Term-Memory (LSTM) and Gated-Recurrent-Unit (GRU) are built, trained, and tested using the collected load data where the GRU model outperformed LSTM, with an accuracy of GRU (97.9%) and LSTM (96.9%). Then, the forecasted 2041 load data is applied to the Hybrid Energy System (HES) model, with HOMER optimizing for Net Present Cost (NPC), Levelized Cost of Energy (LCOE), and CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions. RETScreen software is used to conduct a feasibility analysis by analyzing technical aspects, and economic aspects of the research project and reducing Greenhouse-gas (GHG) production.