Charging pile fault prediction method based on inverse learning algorithm

Aiming at the existing charging pile operation abnormality prediction methods with large prediction error and low confidence, the charging pile fault prediction method based on the inverse learning algorithm is proposed. First, according to the working principle of the charging pile, the fault type of the charging pile is divided, and the characteristics of the abnormal type are set as the standard for judging the abnormal operation of the equipment. Second, the multi-dimensional operation data of the charging pile is collected, the influencing factors of charging pile operation from human operation and weather are determined respectively, and the supplier evaluation model of the charging pile is constructed. Again, under the premise of considering the initial quality of the charging pile and the influencing factors, the inverse learning algorithm is utilized to estimate the power load and other operating parameters. Finally, based on the charging pile operation data, the fault prediction results of the charging pile are derived. Through the comparison experiment with the set threshold, it is concluded that the power load prediction errors of the proposed method are lower than 0.3 kW, and the confidence level is always higher than 85%.