Power grid line loss prediction method and device, storage medium and processor

The invention discloses a power grid line loss prediction method and device, a storage medium and a processor. The method comprises the following steps: acquiring electric power data of a target area, and processing the electric power data to obtain matrix data and data features; processing the matrix data by using a convolutional neural network, and determining a feature map; processing the data features by adopting a full-connection network to obtain feature coefficients; and a target prediction model is adopted to carry out splicing processing on the feature map and the feature coefficient, and a line loss interval is determined. The technical problems of poor applicability and low accuracy of the line loss prediction method in the prior art are solved. 本发明公开了一种电网线损预测方法及装置、存储介质和处理器。其中，该方法包括：获取目标区域的电力数据，并对电力数据进行处理，得到矩阵数据和数据特征；采用卷积神经网络处理上述矩阵数据，确定特征图；采用全连接网络处理上述数据特征，得到特征系数；采用目标预测模型对上述特征图和上述特征系数进行拼接处理，确定线损区间。本发明解决了先有技术中线损预测方法适用性差且准确度较低的技术问题。