2019 Volume 41 Issue 3 Pages 75-89
A long-discussed problem of ground source heat pump systems is to determine required sizes (length and number) of closed-type ground heat exchangers under various conditions in terms of building, heat pump, facility and underground. The purpose of this study is to propose a simulation-based methodology for the size determination over the land through a combination of geo-property estimation and heat pump simulation. In this process, ground thermal conductivity was estimated as a weighted-probability average of individual thermal conductivities of eight soil/rock types. The probabilities were also estimated by indicator kriging with surrounding borehole data. This study collected about 46 thousand borehole data to apply the estimation method in any location, especially of urbanized areas. This study also showed two nation-scale maps of ground thermal conductivity on a 10 km regular grid at a depth of 50 m and 150 m, indicating its depth-dependence was seen in the mountain areas, but was unobvious in the plain areas where the unconsolidated sediments were over one hundred meters thick. The simulation-based method was performed under two conditions in terms of required temperature of the circulating fluids and target performance of the system. The simulation results were carried out among different sizes of ground heat exchangers, and the required sizes were interpolated as a minimum size to satisfy both conditions in each location. As a case study, this study applied the methodology for a household system consisting of a 10 kW ground source heat pump and a single borehole heat exchanger on a regular 10 km grid over the land, Japan. The simulation results were compared in three cases: Case 1; the necessary temperature condition only, Case 2; with target performance condition (the seasonal performance factor is over 4.0), and Case 3; for a highly insulated residence with an overall heat transfer coefficient of 1.0 W/(m2･K). In Case 1, the determined lengths were found to vary between 30 m and 180 m (104 m in average), with differences among adjacent grids arising owing to the varying climatic and geologic conditions. In Case 2, the lengths increased to satisfy the target performance (115 m in average), especially in the northern areas of low ground temperatures. In Case 3, the lengths for the highly insulated house were decreased (76 m in average) in the northern and mid-areas, but not in the southern areas where the cooling loads by the solar radiation were dominant. In conclusion, the proposed methodology could be applied in any location over the land with various conditions, contributing to more public implementation of GSHP systems.