Optimization Control of Building Cooling System

Accurate and reliable chiller plant load measurement is essential for chiller sequencing control, performance monitoring and optimization. Traditional approach is simply by measuring the differential temperature of the chilled water supply and return and the total water flow rate to obtain the cooling demand. However many studies and site observations have indicated that errors in actual cooling demand estimation may be as high as 30 % by such measurement. In practical chiller system, temperature sensors may not be able to represent water temperature accurately due to high water pressure, which results in high potential of temperature measurement noise, outlier, systematic error and uncertainty. The errors in cooling load estimation would be amplified when the cooling load is estimated by the differential temperature. This study aims to develop an efficient way of optimizing the central A/C system operation for office buildings in Hong Kong. A total cooling load estimation algorithm would be developed to better represent building cooling load for sequencing control of central chiller plants for energy saving. It is expected that with the new total cooling demand estimation, the electricity consumption of the central A/C system would be reduced by 5 %. In other words, for a typical office building that consumes 8 million kWh per annum, the saving would be around 0.16 million kWh per annum.