Cost Prediction for Grid Technical Renovation Projects based on XGBoost and SMA

Forecasting the costs of power grid technology renovation projects is instrumental in enabling power utilities to plan investments effectively, ensuring timely equipment updates to maintain the health and operation of the power grid. This paper presents an innovative cost forecasting model, SMA-XGBoost, for power grid technological renovation projects. It integrates the XGBoost algorithm with the slime mould algorithm to address the complexities of cost prediction in an evolving energy landscape. The model leverages the predictive power of XGBoost and the adaptive search capabilities of SMA to improve accuracy and reliability. The study demonstrates the model’s superior performance over existing benchmarks, offering a robust tool for strategic planning and project execution within budgetary constraints.