A multi-source data fusion method for distribution network based on improved evidence theory

In response to the problem of unsatisfactory multi-source data fusion in distribution network is not ideal under the measurement uncertainty and no prior information request, this paper proposes a multi-source data fusion method based on improved evidence theory for distribution network. The dimensionality and quantity of multi-source data in distribution network are unified. A multi-source heterogeneous data processing method for distribution network based on Box Cox transformation Z-score is proposed to improve the data offset problem in the Z-score normalization process by introducing Box Cox transformation. The membership function in fuzzy set theory is introduced as the support function to allocate initial evidence to multi-source data, and the initial evidence is corrected according to the degree of data deviation. Divergence is used to measure the degree of conflict and difference between evidence, and proportional weights are assigned to each evidence based on the principle of conflict allocation. And the evidence is synthesized through evidence synthesis rules, and the data is weighted and summed to obtain the data fusion result.