LAUSR

Abstract Computational intelligence algorithm (CIA) has been widely applied in HVAC fields and several papers have reviewed about those researches and applications. However, their application study in GSHP system are still required to be further enriched. Since the structure of GSHP system is more complex than that of conventional HVAC system, whose operation has obvious nonlinear and dynamic characteristics. CIAs such as artificial neural network (ANN), adaptive network-based fuzzy inference system (ANFIS) and support vector machine (SVM) are suitable for the analysis of such nonlinear problems. This paper conducted a review about the researches relating to CIA applications in GSHP system up to now. These studies were divided into five sections according to the procedure from the system design to system operation of GSHP system, i.e., air conditioning load forecasting, basic design data acquisition (soil thermal properties and ground temperature), heat transfer property and design of ground heat ex-changer, system performance and optimization, together with system control and diagnosis. It has been found from the review that, for the application of CIA in GSHP field, there are some aspects’ works have been excessively done while some other areas’ works have been rarely involved or deviate from GSHP system. For example, although many CIA applications on GSHP system performance prediction and system control strategies have been done, most of them were accomplished basing on small experimental platform or software simulation, while few of CIA applications were conducted towards to actual GSHP system. Ground temperature predictions with CIA technologies were also carried out frequently, but the CIA prediction depths mainly concentrate within 100 cm, which is far from the usual buried depth of GSHP system. In summary, there are many excellent CIA models have been promoted and applied to GSHP field, but the application scope and degree should be further strengthened in the future, especially for the CIA model predictive control that can be directly applied to the actual operational GSHP system, which is the critical to the system energy conservation.