IUB Student Halima Develops Powerful AI Model to Improve Power Supply

Artificial Intelligence (AI) technology is truly an incredible invention. It is making various aspects of our daily lives easier and more efficient. For example, AI is revolutionizing fields such as healthcare, education, communication, and business.

Halima Haque is an Electrical and Electronic Engineering (EEE) student at Independent University, Bangladesh (IUB), created an AI powered forecasting model to improve Dhaka’s electrical supply and eliminate load shedding. This model can forecast electricity consumption, especially during high summer months, therefore ensuring efficient power distribution. As a consequence, load shedding will be reduced, and renewable energy will be utilized more efficiently.

Halima’s research was supervised by Dr. Md. Abdur Razzak, an EEE professor at IUB. The study examined approximately 6.5 million data points gathered from Dhaka Electric Supply Company Limited (DESCO) between 2020 and 2023. It studied power consumption trends in the residential, commercial, and industrial sectors, taking into account weather conditions, usage patterns, and pricing structures.

“The demand for electricity is constantly increasing due to urbanization and industrialization. We can improve power distribution efficiency, reduce load shedding and better manage renewable energy sources by accurately forecasting demand. This AI driven technology will help power distribution companies and policy makers make informed decisions, ensuring a more stable and sustainable power supply,” Halima said.

K-Nearest Neighbors, Random Forest, Light Gradient Boosting Model, Extreme Gradient Boosting, and Multiple Linear Regression were the five machine learning models that were examined in the study. The results have already been published in the esteemed IEEE Transactions on Industry Applications journal and presented at IEEE conferences in Thailand and the Maldives.

Professor Dr. Md. Abdur Razzak declared that “This study is important for modernizing and guaranteeing the sustainability of Bangladesh’s power sector.” He also said “We can create more effective policies by using machine learning to better understand trends of electricity consumption. This can lead to a reduction in load shedding and the realization of sustainable urban design.

Source: Dhaka Post