Abstract: Emergency repair is a critical component for ensuring the safe and stable operation of the power grid. Personnel dispatch and coordination costs represent a significant portion of emergency repair expenses, necessitating the establishment of scientifically sound estimation standards to achieve precise cost control. Based on typhoon repair cases in Hainan, this paper constructs a cost estimation model integrating multiple regression analysis and stratified sampling. The model quantifies key variables such as dispatch distance, mobilization scale, and standby duration. Historical data were collected via stratified sampling to analyze cost drivers through multiple regression. Empirical results indicate a high goodness-of-fit (adjustedR² reaching 0.85), while sensitivity analysis confirms the model's robustness. This research provides power grid companies with a transparent cost-estimation tool, reducing settlement disputes and enhancing compliance. Furthermore, the standardized mechanism incentivizes construction units to respond more promptly, aiding power grid enterprises in optimizing emergency repair cost management.