Research on Multi-Factor Power Load Forecasting Based on Coupling Probability Density Method

Abstract

Power load forecasting plays an important role in power system dispatch, planning and production planning. Accurate load forecasting can ensure the smooth operation of the power system. In real life, electrical loads are usually affected by many external factors. In order to solve the problem of high-dimensional data and reduce the uncertainty in the forecasting process, a multi-element power load forecasting method based on combined probability density is proposed. You can predict the medium-term load probability density of a specific date based on a continuous conditional quantile function. And get more information about mid-term load forecasting. The forecast result can not only obtain a complete future load probability density curve, but also a more accurate future load forecast value and fluctuation range. It is critical to explore an effective way to deal with the load forecasting situation in the big data environment.