Research on Multi-source Heterogeneous Data Preprocessing Based on Improved Crow Search Algorithm

The integrity of multi-source heterogeneous data in vehicle-grid interaction is crucial to the security of the vehicle-grid system. Aiming at the problem that the missing of multi-source heterogeneous data can easily lead to inaccurate analysis results, which in turn affects the charging scheduling of electric vehicles and the load management of the power grid, this paper proposes a multi-source heterogeneous data filling algorithm based on the improved crow search algorithm. First, based on the characteristics of multi-source heterogeneous data in vehicle-grid interaction, a data filling model is constructed to clarify the correspondence between parameters and data filling models. Secondly, in view of the problem of insufficient search accuracy of the crow search algorithm, an optimal solution solidification mechanism and an adaptive step size strategy are proposed to improve the algorithm. Then, the improved crow search algorithm is used to find the optimal solution for the data filling model. Finally, the filling model optimized by the algorithm is used to construct a heterogeneous data set. Taking accuracy as the performance index, the proposed algorithm is compared with the unimproved crow search algorithm and whale optimization algorithm, and the results show that the data accuracy of the proposed algorithm is increased by more than 20%, which effectively improves the data integrity.