No abstract

We herein report the development of a distributed heat pump system that can utilize a variety of renewable energy sources to meet different building heating and cooling demands (i.e., a multiple source and multiple use heat pump system, MMHP). In this system, a water circulating loop is used to connect ground heat exchangers, a unique sky-source heat pump, and various heat pumps for heating and cooling purposes to form a thermal network within a building. This distribution increases the flexibility of the system and allows an improved matching of supply and demand. To evaluate the system performance, an experimental house was constructed, and a winter field experiment was conducted. We found that the reported heat pump for floor heating achieved a stable operation with a high coefficient of performance of ~11.5, while the heat collecting operation performance of the sky-source heat pump varied significantly depending on the amount of solar radiation and the outside air temperature. Finally, since the sky-source heat pump contributes to an improvement in the whole system performance, it appears that there is still room for improved regarding the whole system performance by adjusting the operating and control strategy.

This paper is based on the translation of the Japanese version with some slight modifications.

In this study, an artificial neural network (ANN) was used to model the thermal performance of a novel direct-expansion solar-assisted sky-source heat pump (SSHP) during winter. The input parameters of the ANN take into account the weather conditions, water loop characteristics, and the compressor characteristics of the SSHP. The following four output parameters were adopted to evaluate the SSHP performance: the outlet water temperature of the water loop, electricity consumption, heat production, and the coefficient of performance. To increase the accuracy of the ANN and simultaneously investigate the effects of each of the input parameters on the performance of the SSHP, the combination of input parameters for the validation data set was varied in multiple case studies. Additionally, learning curves were introduced to clarify the relationship between the training data size and the generalization performance of the ANN. Finally, the ANNs with the best performance were selected and evaluated based on the test data set by using metrics such as the root mean square error. The reported results demonstrated that the ANN model has comparatively high SSHP winter performance prediction accuracy.