Design of Multi-Source Heterogeneous Data Fusion Algorithm for Distribution Networks Based on Improved Kalman Filter

The complexity of multi-source heterogeneous data in distribution network is high and the deviation is large, which makes it more difficult to fuse multi-source heterogeneous data in distribution network. Therefore, a multi-source heterogeneous data fusion algorithm based on Improved Kalman filter is designed. The deviation data in the multi-source heterogeneous data of the initial distribution network is corrected by the correction process. The least square method and Lagrange interpolation method are used to perform time registration and missing filling processing on the corrected data. The distribution map method and Kalman filter algorithm are fused to improve the Kalman filter algorithm. The processed multi-source heterogeneous data are input into the improved algorithm. After removing the disturbed data from the distribution map, the data fusion is completed by Kalman filter. The experimental results show that the algorithm can achieve accurate missing data filling processing, and the average error between the current data, voltage data and actual data is 0.0082 and 0.0149 respectively, which are within the ideal error range. For all kinds of distribution network data with disturbance data, it can achieve high-quality data fusion processing, and the covariance difference of the fusion result is close to the demand covariance difference of 0.06, which meets the actual application requirements.