2025-09-22
AI-Powered Cooling: Revolutionising Energy Efficiency in West Kowloon Cultural District
Figure 1 – WKCD DCS Plant
Hong Kong District Cooling Company Limited (HKDC), a subsidiary of Veolia, operates and maintains The West Kowloon District Cooling System (DCS) in the West Kowloon Cultural District (WKCD). This system significantly contributes to energy efficiency and environmental sustainability within WKCD.
To best serve WKCD, the HKDC team utilises an advanced AI optimisation programme for its district cooling system. This programme, adapted from a European tool for local application, aims to reduce electricity consumption, carbon footprints, and operational costs.
It analyses historical plant data and weather information to recommend optimal chiller sequencing and staging strategies for operators. The AI tool employs machine learning models to predict cooling loads and simulate chiller efficiency using the Coefficient of Performance (COP), enabling optimised control of the District Cooling Chiller Plant in WKCD.
In a trial conducted in 2024, HKDC's plant operators implemented the AI's recommendations and achieved preliminary electricity savings of 3.2% compared to a simulated baseline. These savings were independently verified using the International Performance Measurement and Verification Protocol (IPMVP), which confirmed the high reliability and accuracy of the simulation model used for verification.
Taking into account factors like the existing direct seawater cooling system, the trial successfully laid a robust groundwork for future AI integration. Moving forward, the focus will be on further enhancing and expanding the AI tool. This includes refining its prediction models by incorporating occupancy data to significantly boost efficiency and sustainability.
To best serve WKCD, the HKDC team utilises an advanced AI optimisation programme for its district cooling system. This programme, adapted from a European tool for local application, aims to reduce electricity consumption, carbon footprints, and operational costs.
It analyses historical plant data and weather information to recommend optimal chiller sequencing and staging strategies for operators. The AI tool employs machine learning models to predict cooling loads and simulate chiller efficiency using the Coefficient of Performance (COP), enabling optimised control of the District Cooling Chiller Plant in WKCD.
In a trial conducted in 2024, HKDC's plant operators implemented the AI's recommendations and achieved preliminary electricity savings of 3.2% compared to a simulated baseline. These savings were independently verified using the International Performance Measurement and Verification Protocol (IPMVP), which confirmed the high reliability and accuracy of the simulation model used for verification.
Taking into account factors like the existing direct seawater cooling system, the trial successfully laid a robust groundwork for future AI integration. Moving forward, the focus will be on further enhancing and expanding the AI tool. This includes refining its prediction models by incorporating occupancy data to significantly boost efficiency and sustainability.
Figure 2 – Dashboard display of energy analysis