Medium and Long Term Wind Power Generation Forecasting Method Based on Multi-model Fusion

Abstract

Abstract Accurate, long-term wind power forecasting is crucial for effective power grid operation planning. It can enhance the stability and security of the power system. We propose a combined medium and long-term wind power generation prediction method based on multi-model fusion to study forecasting. By analyzing the connection between meteorological data and wind power generation, we identify the pivotal factors that impact wind power output and determine the best input data scheme. Based on the predictive outcomes, we select the suitable core predictive sub-models and use the particle swarm optimization algorithm to achieve the dynamic optimization of the weight for each sub-model. Afterwards, we allocate the optimized weights to each sub-model and establish a fused multi-model advantage-based combined predictive platform for wind power generation for the medium and long term. The wind power forecast is enhanced through the dynamic weighted combination forecasting technique, which significantly enhances forecast precision.