Microgrid short-term load prediction method

Abstract

The invention relates to a microgrid short-term load prediction method which comprises the following steps: S1, acquiring load historical data, and carrying out data quality improvement on a singularvalue interval by adopting cubic spline difference values; S2, establishing a gray variation coefficient model based on amplitude compression, and introducing an improved human body comfort index as an influence factor in the modeling process; S3, calculating a prediction parameter sequence of the cumulative variation coefficient sequence and the mutational variation coefficient sequence; S4, optimizing the prediction parameter sequence by using a genetic simulated annealing algorithm to obtain an optimal matching parameter; S5, reconstructing an optimal matching parameter to obtain an optimalvariable coefficient sequence; and S6, carrying out reverse solving to obtain a load prediction value, and controlling the working state of the distributed power supply. Compared with the prior art,the method has the advantages that the air quality condition is introduced into the human body comfort index, the gray variation coefficient model is established, solving is carried out based on the genetic simulated annealing algorithm, the prediction speed is high, and the prediction precision is high.

Description

Technical field [0001] The invention relates to the technical field of load forecasting, in particular to a short-term load forecasting method of a microgrid. Background technique [0002] Microgrid is a small power generation and distribution system connected to the user side, which can realize self-control, protection and management, and has the characteristics of low cost, low voltage and low pollution. Short-term load forecasting of microgrids is an important prerequisite for achieving high-efficiency, energy-saving and optimized operation of microgrids. Accurate prediction of short-term load changes is of great significance to the arrangement of power system generation and supply plans and the benefits of the power market. Compared with large power grids, the user-side microgrid load has the characteristics of large volatility and uncertainty due to external conditions, and low similarity of historical load curves. The user-side microgrid load changes have many influencing ...

AI Technical Summary

Problems solved by technology

Most of the above methods do not consider or only consider the influence of certain meteorological factors in the modeling, which makes the prediction results less accurate. Aiming at the influence of meteorological factors on the load, a human comfort index is introduced, and a human comfort index based model is proposed. In the short-term load forecasting method, the human comfort index in the prior art usually includes temperature, humidity and wind speed. However, with the acce

Images