Design and Application of a Fault Diagnosis and Monitoring System for Electric Vehicle Charging Equipment Based on Improved Deep Belief Network

It is of great significance to accurately obtain the operating state of DC charging equipment for electric vehicles (Abbreviated as “charging equipment”) and to detect and identify the occurring faults in time. This paper introduces the design and application of a novel fault diagnosis and monitoring system for charging equipment. Firstly, the system based on a five-layer structure is constructed and analyzed. Secondly, the useful information in the massive operating data is extracted by using big data technology, and the data features that can characterize the health condition of charging equipment are filtered for the construction of the fault diagnosis model. Then, the improved deep belief network (DBN), using a linear restricted Boltzmann machine (LRBM) to initialize the output layer parameters of the network, is chosen to construct the fault classifier. The improved DBN can reduce the chance of the model falling into a locally optimal solution caused by the random initialization of the output layer parameters of the typical DBN. Finally, the system is deployed to a certain project, and the operating state of charging equipment is determined according to the designed fault diagnosis process. The performance of the proposed method is validated by comparing it with the traditional fault diagnosis methods and typical DBN, and the results show that the improved DBN exhibits a more satisfactory performance in the fault diagnosis of charging equipment.