Analysis of a Ground-Coupled Heat Pump During a Heating Season

A ground-coupled heat pump, installed in an 170 square meter residential building in Knox County Tennessee, has been examined both experimentally and analytically. The ground coupling was realized with a 210 meter coil buried 1.2 meters beneath the soil, through which circulated a secondary heat transfer fluid. The house, the ground and the power consuming devices were instrumented, and hourly readings of temperatures, heat flows, power consumption, and weather data were recorded.

The experiment was carried out during the 1982-1983 heating season beginning in November and lasting 26 weeks. The overall heating seasonal performance factor was 2.57. No backup heat was required throughout the duration of the experiment.

An analysis with the TRNSYS and GROCS computer programs was also undertaken. These programs were specifically modified to model the heating system, and the ground heat exchanger. The model prediction of overall seasonal performance factor was within 5% of the measured values. The effect of varying several parameters such as ground coil length, soil thermal properties, and the heat pump performance curve, was examined. The model was found to be relatively insensitive to small variations in thermal properties of the soil. The maximum seasonal performance factor for the Knoxville area was predicted to be 2.85 using the maximum estimated thermal conductivity, twice the original coil length, and the observed heat pump performance curve for the system.