Abstract: Under the background of continuous expansion of coal mine production scale and increasingly intelligent power grid construction,aiming at the problems of low prediction accuracy and large calculation error of capacitance current in high voltage power supply system of coal mine,an intelligent prediction method of capacitance current in high voltage power supply system of coal mine is proposed. According to some existing cable parameters,we use BP neural network to establish the prediction model of capacitance current,and then introduce particle swarm optimization algorithm to optimize the prediction model. The characteristic parameters are selected,the data are normalized,and the prediction process is designed. The average relative error and other indicators are used to analyze the error size and evaluate the accuracy of the method. The capacitance current prediction method is compared and analyzed using the measured data. The results show that the relative error of this method is 2. 52%,realizing the accurate prediction of capacitance current of high voltage power supply system in coal mine,and providing a new idea for its intelligent prediction.