Abstract: Taking office building in Shenyang region as an example, a solar coupled ground source heat pump system（SGSHPS） model with different operation modes was established based on TRNSYS platform to optimize the configuration of SGSHPS. The soil temperature imbalance ratios and total energy consumption of the system after operation for 30 years in different modes were compared and analyzed. Using the parallel transition season thermal storage mode as the research object, different sets of solar collector systems and buried tube heat exchanger combination methods were simulated. Under the premise of stable system operation, the configuration of the solar collector area and the number of buried tube heat exchanger boreholes were optimized for minimum dynamic annual cost value. PSO-MSVR algorithm was introduced. PSO algorithm was used to generate the MSVR model parameters, as well as to find the optimal values. Results show that compared with the configuration method determined by the traditional heat balance method, the system energy consumption after operation for 30 years reduces by 379 522 kW for the optimized configuration method, which means energy saving of 11.2%. The annual cost reduces by 61 863 yuan, while the soil temperature only increases by 0.73 K. For different solar collector and buried tube heat exchanger costs, the annual costs after optimization reduce by different degrees compared with before optimization, and the soil temperature has not been greatly shifted.