CH-24-C067 - Chiller Energy Optimization using Artificial Intelligence: Experience in Hong Kong

Air-conditioning has contributed to around 30% of Hong Kong’s total electricity consumption by End-use in 2021, according to the Hong Kong Energy End-use Data 2023 compiled by Electrical and Mechanical Services Department (EMSD) of the Hong Kong Special Administrative Region (HKSAR) of the People’s Republic of China [1]. To achieve carbon neutrality, a roadmap was set out by the government in the Hong Kong’s Climate Action Plan (“the Plan”) [2] addressing the goal to reduce the electricity consumption in commercial buildings by 30-40% by 2050, of which trimming down the energy use of air-conditioning system in particular its core component i.e. chiller plant would definitely be one of the key measures to attain the target. This paper aims to present the challenges and hurdles faced as well as the experience gained during the implementation of Artifical Intelligence (AI) chiller plant optimization in several in-service government and public buildings in Hong Kong. It elaborates the most favorable plant configuration for setting up AI and illustrates the optimization strategy by utilizing the artificial neural network (ANN) technique and particle swarm optimization (PSO) algorithm. The outcome outlines an encouraging achievement of 5–10% energy savings for the revitalized chiller plants with intelligence and vibrancy, which helps to step up the schedule towards carbon-neutral.