
Adaptive optimisation model cuts microgrid requirements
An optimization model developed in South Korea improves microgrid operation with lower edge compute requirements to boost energy efficiency and ensure reliable power supply.
Researchers at Incheon National University developed the model to adapt to unexpected changes in power supply and demand, ensuring stable and efficient energy systems. By addressing power outages and varying energy needs, this approach enhances the reliability and sustainability of microgrids, making it suitable for real-world use in areas with unstable power grids.
As the world transitions to renewable energy sources such as solar and wind power, microgrids are becoming more essential. However, managing these systems is challenging due to the uncertainties in energy supply and demand, such as power outages or fluctuations in energy usage, and stochastic islanding — situations where parts of the power grid unexpectedly become isolated, disrupting the power supply.
- A microgrid for the UK’s Royal Mint
- AI model boosts grid optimisation
- Virtual power plant boosts grid resiliance
Traditional methods for optimizing microgrid operations, such as multistage models, are computationally expensive and impractical for real-world use. These models consider different scenarios over time, but the complexity increases exponentially, making their application difficult at a large scale.
The researchers have simplified these models while maintaining their effectiveness, by reducing the number of possible scenarios and introducing a process called replanning, where the optimization model adapts over time as new information emerges. This approach significantly reduced the computational burden, enabling them to be more efficient in real-world settings on edge computers
“Our goal was to create a method that makes microgrid operation more adaptable and cost-effective, especially in regions with unreliable grids or frequent disruptions,” said Assistant Professor Jongheon Lee. “By simplifying the models and using replanning, we can achieve effective operation plan without the heavy computational cost.”
This approach can help in cities as well, where the energy demand is rising, and grids are under strain. Scalable optimization models can improve the overall energy management. Adapting to changes in supply and demand in real time helps boost grid resilience, supporting the transition to sustainable energy. Moreover, these models are flexible, making them suitable for both small and large systems.
“These optimization methods will be vital for improving energy security, particularly in areas with unreliable power. They also support global sustainability goals by promoting renewable energy,” said Lee.
10.1016/j.apenergy.2024.124040
