The demand for new service connections and electric energy meter installations is significantly driven by electricity consumption growth.To improve forecasting accuracy and resource allocation efficiency

this study investigates a demand forecasting approach for newly installed meters based on electricity consumption characteristics.Key influencing factors are identified

and typical growth patterns are extracted through analysis of user electricity consumption across different industries and regions.A combined forecasting model integrating time series and machine learning algorithms is constructed

incorporating multi-source data and dynamic trends.Simulation results demonstrate that the proposed model outperforms traditional methods in terms of accuracy

stability

and adaptability

providing effective decision support for power enterprise expansion planning.