A multivariate nonlinear regression prediction model for the performance of cooling tower assisted ground source heat pump system

Abstract

Cooling tower-assisted ground source heat pump (GSHP) systems have been widely used in the regions where both cooling and heating are required in recent years. However, the issue of system design and management is still under discussion. The ratio of heat removed by cooling tower to the absorbed by the ground would influence the operation performance of hybrid system. This study developed a method to predict the system comprehensive coefficient of performance (SCOP) of hybrid system to optimize system structure and operation. Taking a cooling tower-assisted GSHP in a residential district in a hot summer and cold winter region as an example, a multivariate nonlinear regression prediction model for SCOP was derived based on the data recorded from May to September 2021 by the Building Energy Management System (BEMS) and the simulation results using TRNSYS software. Outdoor dry bulb, wet bulb temperatures, soil temperature, and auxiliary cooling ratio (ACR) are involved in the model. Based on model prediction and system simulation, the ACR of cooling tower-chiller unit should take 0.7 of the accumulated cooling load, considering the SCOP in summer and sustainability for long-term. An operation strategy has been proposed, prioritizing the operation of cooling tower-chiller and controlling the temperature difference between supply and return chilled water within 6℃. The average SCOP of the existing hybrid system can reach 5.56, and the soil temperature rise is within 4℃ over 15 years. The model can predict the variation of average SCOP with ACR during the cooling season in different regions. The calculation results serve as reference for designing and operating hybrid ground source heat pump (HGSHP) systems, ensuring system sustainability while achieving optimal SCOP.