Aiming at the problem that it is difficult to clean and verify the data of distribution terminal automation joint debugging due to the large number of data sources, a method of data verification for distribution terminal automation joint debugging based on self-organizing map is proposed. Collect such data as remote signaling, remote metering, protection remote signaling and protection remote metering of power distribution terminal; A self-organizing map neural network model consisting of an input lay and a competition layer is constructed, that collected joint debug data is used as model input, the feature vectors of different types of joint debugging data are subjected to automatic clustering, whether the joint debugging data of the power distribution terminal automation are abnormal or not is determined by judging the correlation among the joint debugging data, and the aim of checking the joint debugging data of the power distribution terminal automation is fulfilled. At the same time, the competition mechanism of the adaptive optimization model is used to alleviate the advantage of over-learning, and the weights of the model are dynamically optimized through the analysis of grey relations to suppress the negative impact of impurities in the neighborhood neurons and optimize the calibration performance of the self-organizing map neural network model on the joint debugging data. The experimental results show that the method has good clustering performance, can effectively achieve the purpose of checking the joint debugging data, and improve the effect of joint debugging of terminal equipment.