According to Technical Memoranda TM18:1994 of CIBSE, the inability to control energy demands in relation to energy supplies has generated new interest in the use and application of thermal storage in general and ice storage in particular. The main advantage of ice storage is to shift the electrical demand from the summer afternoon peak period to nighttime. In subtropical cities, such as Hong Kong, weather in summer is not stable due to rainy days and even thunderstorms. It will be a great waste if ice is blindly stored without any reference to the actual demand on the following day. It will also result in poor performance of the whole air-conditioning system if ice is blindly discharged during normal operation. Our firm belief is that ice charging and discharging should be tightly coupled with the actual weather in order to achieve an optimal goal where accurate weather prediction and cloud tracking are the major tools. Such information can be freely downloaded from the Internet. A useful linguistic-numerical conversion scheme has been established based on our statistical analyses for the past two years. We have also developed a chiller capacity-based control algorithm for ice discharging. Three cases have been simulated, and we have concluded that substantial energy saving and excellent performance can be guaranteed with this new weather information-based intelligent ice storage control system. The research results may be useful to aid real-time air-conditioning control based on weather prediction and real-time monitoring.

Units: SI