Simulation used for the selection and sizing of a centralized cooling plant: Methodology and limitations

During the last ten years, many HVAC simulation models and calculation tools were developed. However, until now, the practical use of simulation tools is still very limited in building and HVAC design as well as in the development of control strategies. This paper focuses on the selection and sizing of a centralized cooling plant. Its purpose is to show how the simulation can help the designer and to point out the difficulties encountered by the simulation user. In the design stage, various alternative cooling systems are available. Since the paper stresses on the methodology, only two design alternatives of centralized chilling plant are considered: with air-cooled reciprocating chillers and with water-cooled twin-screw chillers. The sizing of these two cooling plants is based on the sizing criteria and the building thermal loads in design conditions. The models used to simulate the primary equipment are extracted from the simulation toolkits developed by ASHRAE and are briefly described in the paper. The model parameters need to be adjusted to represent accurately the behavior of the components. At the design stage, the information available about the components is provided by the manufactures catalogs. The paper describes the data usually supplied by these catalogs and analyses the difficulties encountered during the procedure of parameter identification of the models. The simulation of the thermal behavior of both cooling plants is handled by the modular simulation program TRNSYS. It is based on the identified component parameters, but also on the implementation of regulation rules. The simulation allows the designer to check if both cooling plant alternatives can meet the building cooling loads and to estimate the energy consumption and, thus, the operating cost associated with each design alternative. The paper also points out that the auxiliary consumptions (fans, pumps) have to be considered in the determination of the operating cost. In addition to the comparison of design alternatives, the simulation can be used at the design stage to evaluate the effects of variations of the characteristics of the systems considered. The paper shows how the system coefficient of performance is affected by an increase of the set point water temperature at the exhaust evaporator and by a different sizing of the water-cooled twin-screw chillers.