Comparative Analysis of Performance of 1D-CNN and DNN Model in Line Loss Rate Prediction

Area line loss suggestion is an important economic indicator for power supply enterprise evaluation, but theoretical calculation line loss suggestion in complex area has been a difficult problem. In recent years, with the depth of the rise of the network learning algorithm, many scholars use data mining, deep learning and other technologies to comprehensively analyze the influencing factors of line loss rate of distribution network, and thus establish a predictive relationship model between line loss related data and line loss rate. On this basis, this paper proposes a line loss rate prediction model based on 1D-CNN algorithm and DNN algorithm respectively, and compares and analyzes the predictive performance of the two algorithms by using the data of a regional power grid in Suzhou, which verifies that the 1D-CNN algorithm is better than the DNN algorithm in the prediction of line loss rate.