Abstract:In the health status warning of substation equipment, due to the many factors that affect the health status of equipment, the analysis process of equipment health status is complex, resulting in low accuracy of the results. To alleviate this problem, a multi-dimensional warning model for the health status of substation equipment based on multi-source data fusion has been proposed. Collect multi-dimensional substation equipment data and expand the data samples through resampling processing. Perform multi-source data fusion processing on the resampled data samples to evaluate the health status of substation equipment using the fused data. On this basis, a multidimensional warning model is established with XGBoost algorithm as the core, and corresponding warning processing is made according to the set evaluation system. Through experimental testing, it is known that the model has demonstrated a high level of accuracy in practical applications. Compared with other methods, the model has more accurate analysis of equipment health status, better ROC curve, and higher early warning accuracy, which has a good application prospect in the actual operation and maintenance management of substation equipment.