MBZUAI researchers are applying federated learning to optimize smart grids while protecting user data privacy. This approach leverages techniques from smart healthcare systems to enhance energy efficiency and local energy sharing. The research addresses the challenge of balancing grid optimization with the risk of user identity theft associated with traditional data-intensive smart grids. Why it matters: This research demonstrates a practical application of privacy-preserving AI in critical infrastructure, addressing key concerns around data security and fostering trust in smart grid technologies.
Khaled Alrashed, president and CEO of Saudi Electricity Company for Projects Development, discussed the challenges of future smart cities at a KAUST event. He emphasized the importance of smart grids, AI, and large-scale optimization for improving urban living. The Saudi Electricity Company is partnering with KAUST, including using the Shaheen supercomputer, to develop these technologies and predict grid load. Why it matters: This collaboration highlights Saudi Arabia's ambition to become a leader in smart city technology and renewable energy, leveraging local expertise and resources.
KAUST hosted the "Human-Machine Networks and Intelligent Infrastructures" conference, co-organized by Prof. Jeff Shamma and Asst. Prof. Meriem Laleg. The conference explored the blend of engineered devices and human elements in large-scale systems like smart grids. Keynote speaker Dr. Pramod Khargonekar discussed cyber-physical-social systems and emerging trends. Why it matters: The conference highlights the growing importance of understanding the interplay between AI, infrastructure, and human behavior in the development of smart cities and intelligent systems in the region.
To meet Net Zero Emissions goals, governments and corporations need to drive a wholesale energy transition. Current energy grids are outdated and need to be updated to handle renewable energy's specific demands. Research at MBZUAI is helping to create smarter grids by using AI to monitor and measure energy flow in real-time. Why it matters: AI-empowered smart grids can help accelerate the energy transition by enabling more efficient and reliable use of renewable energy sources.
MBZUAI researchers are using federated learning to optimize energy production and use in microgrids, balancing individual and grid-level needs with a focus on sustainability. They presented a multi-agent framework called MAHTM at the ICLR 2023 workshop, aiming to minimize the carbon footprint of electrical grids. The system uses three layers of decision-making agents to minimize cost, decrease carbon impact, and balance production. Why it matters: This research offers a novel approach to integrating renewable energy sources into existing grids, potentially accelerating the transition to more sustainable energy systems in the region and globally.
KAUST researchers are simulating cyberattacks on microgrids to assess their impact and develop detection/suppression methods. They used the Canadian urban distribution model with four inverter-based distributed generations (DGs) to capture system dynamics. The simulations considered attacks altering measurement data, modifying control signals, and causing sudden load changes, all of which had damaging effects. Why it matters: This research is crucial for ensuring the resilience of increasingly complex microgrids against cyber threats, especially as they become more integrated into critical infrastructure.
KAUST researchers have developed a dual-use wireless sensor system that monitors both traffic congestion and flood incidents in cities. The system combines ultrasonic range finders and infrared thermal sensors to provide real-time, accurate data on traffic flow and roadway flooding. Data is sent to central servers and assimilated with satellite data to form real-time maps and forecasts. Why it matters: This technology can provide up-to-the-minute warnings for flash floods and traffic, enabling rapid emergency response and potentially saving lives in urban environments.
This paper presents a reinforcement learning framework for optimizing energy pricing in peer-to-peer (P2P) energy systems. The framework aims to maximize the profit of all components in a microgrid, including consumers, prosumers, the service provider, and a community battery. Experimental results on the Pymgrid dataset demonstrate the approach's effectiveness in price optimization, considering the interests of different components and the impact of community battery capacity.