Electric power data fusion method and equipment

The invention relates to an electric power data fusion method. The method comprises the following steps: respectively acquiring time sequence data from a plurality of data sources; calculating the similarity between the time sequence data to obtain a similarity matrix; accumulating matrix elements in the same row or the same column in the similarity matrix to obtain a similarity accumulated value of each data source and other data sources; longitudinal parameter fusion is carried out on the data sources with the similarity accumulated values higher than a first threshold value, and the longitudinal parameter fusion comprises the steps that data preprocessing is carried out on the time sequence data; for the time series data after data preprocessing, calculating the grey correlation degree between every two time series data; filtering the time sequence data of which the grey correlation degree is lower than a third threshold value; performing data fusion on the time sequence data by using a BP neural network t