In the heart of Hong Kong, researchers are revolutionizing the way buildings stay cool, and it’s not just about the technology—it’s about making it work in the real world. Lingyun Xie, an associate professor at The Hong Kong Polytechnic University’s Department of Building Environment and Energy Engineering, is leading the charge with a groundbreaking approach to optimizing building cooling systems using artificial intelligence.
Imagine a world where your office building’s air conditioning system doesn’t just keep you comfortable, but does so while saving energy and money. That’s the promise of Xie’s research, published in the journal ‘Advances in Applied Energy’ (translated from Chinese as ‘Advances in Applied Energy’). The study introduces AI-empowered online control optimization technologies designed to bridge the gap between academic research and practical implementation.
The challenge has always been making AI work efficiently and reliably in real-time applications. “Traditional AI algorithms can be computationally intensive,” Xie explains, “which makes them impractical for online control systems that need to make decisions in real-time.” To overcome this, Xie and her team developed a simplified deep learning-enabled Genetic Algorithm. This algorithm accelerates the optimization process, ensuring it can keep up with the demands of online applications. But that’s not all—it also significantly reduces CPU and memory usage, making it possible to deploy on miniaturized control stations right in the field.
But what about reliability? Buildings can’t afford to have their cooling systems go haywire. That’s why Xie’s team introduced a robust assurance scheme. If something goes wrong, the system switches to expert knowledge-based control, ensuring stability and reliability. “We can’t just rely on AI to do everything,” Xie notes. “We need a safety net, a fallback plan.”
To test their strategy, the team conducted hardware-in-the-loop tests using a physical smart station controlling a simulated real-time dynamic cooling system. The results were impressive: the optimal control strategy achieved 7.66% energy savings and exhibited strong operational robustness.
So, what does this mean for the future of building energy efficiency? For one, it could lead to significant cost savings for building owners and operators. But more importantly, it could pave the way for smarter, more efficient buildings that are better equipped to handle the challenges of a changing climate.
Xie’s research is a step towards making AI a practical tool for building energy management. As buildings become smarter and more connected, the need for efficient, reliable control systems will only grow. This work could shape the future of building automation, making our cities more sustainable one cool building at a time.
