In the heart of Malaysia, a groundbreaking energy management system is quietly revolutionizing the way we think about peak demand reduction. Researchers at the Universiti Tunku Abdul Rahman (UTAR) have developed an innovative controller for battery-based energy storage systems (BESS) that promises to reshape the energy landscape, making it more efficient and sustainable. At the helm of this project is MD Mahmudul Hasan, an assistant professor in the Electrical and Electronic Engineering Department, who has been leading the charge in this cutting-edge research.
The crux of the innovation lies in an adaptive threshold-based controller that utilizes deep learning to forecast daily load profiles. This isn’t just another simulation study; Hasan and his team have implemented their controller on a real 200 kWh BESS installed on the UTAR campus. The system uses a one-dimensional convolutional neural network (1D-CNN) to predict the next day’s power demand, allowing the BESS to discharge power precisely when needed to shave off peak demands.
“The beauty of this system is its adaptability,” Hasan explains. “The threshold for discharging power is not fixed. It adjusts dynamically based on the actual and forecasted power demand, as well as the preceding peak of the grid power. This adaptability is key to its effectiveness.”
The results speak for themselves. Over six months of simulations, the adaptive threshold-based controller achieved an average daily peak reduction factor of 41.62%, with a monthly peak reduction failure rate of just 16.55%. These figures outperform other controllers, including fixed-threshold and advanced active and fuzzy controllers. But the real test came when the controller was implemented in a live setting. Over 21 days of real-world operation, the system delivered even more impressive results, with an average peak reduction factor of 49.45% and a failure rate of only 4.76%.
The implications for the energy sector are profound. As energy demand continues to grow, particularly during peak hours, the ability to predict and manage these demands more effectively can lead to significant cost savings and reduced strain on the grid. For commercial and industrial consumers, this means lower energy bills and a more reliable power supply. For utility companies, it offers a tool to optimize their operations and potentially defer costly infrastructure upgrades.
Hasan envisions a future where such adaptive controllers are integrated into smart grids, working in tandem with renewable energy sources to create a more resilient and sustainable energy ecosystem. “This technology has the potential to be a game-changer,” he says. “It’s not just about reducing peak demands; it’s about creating a smarter, more efficient energy system.”
The research, published in the IEEE Access journal, titled “An Innovative Adaptive Threshold-Based BESS Controller Utilizing Deep Learning Forecast for Peak Demand Reductions,” is a significant step forward in the field of energy management. As the world continues to grapple with the challenges of climate change and energy security, innovations like this offer a beacon of hope. They remind us that with the right blend of technology and ingenuity, we can build a future where energy is not just abundant but also sustainable and efficient. The journey towards this future has begun, and it’s being led by pioneers like MD Mahmudul Hasan and his team at UTAR.