�㽶��ý

Working on a paper that was recently published in Nature helped Md. Samiur Rahman (M.S. CS 2nd Year) prove that an integrated Internet of Things (IoT) architecture can optimize a renewable energy production and storage system to power a single-family home.

“” describes an IoT-integrated hybrid renewable energy system for smart homes that combines solar panels and wind turbines for generation, battery storage systems, IoT sensors for real-time monitoring, and smart controllers for adaptive energy management. The system demonstrated up to 72.3 percent average efficiency improvement, 61 percent reduction in energy costs, and a more than 61 percent reduction in carbon dioxide emissions compared to conventional systems.

“I was motivated to participate in this project for several reasons,” Samiur says. “This project is addressing the critical global challenges of energy demands and greenhouse gas emissions. It was also an opportunity to work at the intersection of cutting-edge technologies and develop a comprehensive solution that combines solar, wind, storage, and intelligent controls.”

IoT sensors aggregate and securely transmit data to a cloud platform. In the cloud, advanced machine-learning algorithms incorporate past and current data to conduct real-time analytics, which forecast energy demand patterns, optimize energy production, and identify potential shortfalls or system inefficiencies. Outcomes are displayed on an accessible user dashboard, allowing homeowners to monitor system performance, analyze energy statistics, and receive proactive operational optimization alerts.

“My role in this research project was to develop the IoT architecture and hybrid energy integration framework,” Samiur says. “I also worked on optimizing the algorithms, creating predictive analytics, and control strategies.”

He says that the project presented several challenges, which included creating a fully integrated system combining solar, wind, and storage under a single IoT control. Samiur also found himself addressing computational limitations of optimization algorithms and managing real-world IoT communication issues including device interoperability.

“I implemented an adaptive parameter control to address convergence issues and used parallelization to improve runtime efficiency,” he says. “I also conducted rigorous testing, including Monte Carlo scenarios, sensitivity analysis, and runtime benchmarking, and created comprehensive validation through a 30-day simulated smart home case study.”

Samiur worked with researchers from China, Malaysia, and Bangladesh on the project. He says the lead author of the paper, Amam Hossain Bagdadee, was a colleague at Presidency University in Dhaka, Bangladesh, where he worked as a lecturer. Bagdadee’s supervisor at Hohai University in China, Li Zhang, also worked on the project.

“This was an international collaborative research project, likely facilitated through academic networks, shared research interests in renewable energy and IoT systems, or institutional partnerships between universities,” he says.

Samiur says he would one day like to see a pilot deployment in actual residential settings, address scalability for wider adoption, to improve communication reliability in real IoT networks, and to enhance machine-learning models for better prediction. Eventually, he would like to integrate the system with smart grid infrastructure, expand to multi-home or community-level systems, incorporate additional renewable sources, develop more sophisticated demand response strategies, and advance toward fully autonomous smart energy ecosystems.

“The research positions itself as providing ‘a pathway toward sustainable IoT-enabled smart home energy ecosystems’ with significant room for expansion and real-world implementation,” he says.