Research on fault diagnosis of geothermal heat pump based on EEMD and kmeans++

With the development of technology and sustainable concept, ground source heat pump has gradually emerged in the heating industry, which is a new heating technology with the advantages of clean and high efficiency. However, with the continuous operation of the heat pump system, various faults may occur in the internal equipment of the system, such as evaporator fouling, compressor leakage, refrigerant line blockage, etc. It is of great practical significance to identify the fault types accurately and effectively. In this paper, the signal processing method adopts the ensemble empirical mode decomposition, which decomposes the collected ground source heat pump system signals, retains the relevant modal parameters and removes the residual terms, then evaluates each modal component using the approximate entropy as the feature evaluation index, and clusters the entropy values into feature vectors to determine the fault types. The experimental results show that the algorithm based on the combination of ensemble empirical mode decomposition and clustering can be applied to the fault diagnosis accuracy of ground source heat pump systems.