Abstract:
In order to solve the problem of low forecasting accuracy of traditional grey model when power load increases fast, the paper presented a method of using crossover genetic particle swarm optimization algorithm instead of least-squares algorithm to optimize parameters a and b in GM(1, 1) model, introduced grey prediction theory and its mathematical model, CGPSO algorithm and optimized grey model, and completed simulation based on actual load data. The results showed the prediction accuracy of the optimized grey model was significantly higher than GM (1, 1) model when power load increases fast, and the optimized grey model could be applied to forecast medium and long term load of power system, so as to expand applying scope of grey model.