Performance of a Ground-Coupled Heat Pump during a Cooling Season

A ground-coupled heat pump system was installed in TECH House I at the Tennessee Energy Conservation in Housing site located on the University of Tennessee campus. The performance of the heat pump system was evaluated for the cooling season of 1983. Data on system energy flows, power consumption, house temperatures, soil temperatures and weather conditions were gathered on an hourly basis for the entire cooling season. Weekly coefficients of performance, soil thermal conductivitys, and a seasonal performance factor were calculated from experimental data.

The performance was found to be poor when compared to a conventional air-to-air heat pump system. The seasonal performance factor for the ground-coupled heat pump system was 1.11 while typical conventional heat pump system seasonal performance factors for the Knoxville area are around 2.3. A trend of performance deterioration over the season due in part to decreasing soil thermal conductivity was found. The decreasing soil thermal conductivity was caused by a decrease in soil moisture.

The ground-coupled heat pump system was modeled using the TRNSYS and GROCS computer programs. Comparison with experimental data validated the computer model. The difference between the predicted and experimental seasonal performance factors was approximately 3 percent.

Parametric studies were performed to determine system performance under various conditions. Factors varied include soil moisture content, density, cooling load per unit coil length and far-field soil temperatures.

The predictions show that the ground-coupled heat pump should give best performance in coarse grained, well packed, moisture saturated soil. Worst performance was predicted to occur in fine grained, loose packed, dry soil.

The model was also used to predict the maximum performance of a well designed ground-coupled heat pump in the Knoxville area. The maximum predicted seasonal performance factor was 2.04.