A prediction method and system for hybrid wind power generation

Abstract

The invention discloses a hybrid wind power generation prediction method and system. The hybrid wind power generation prediction method includes: obtaining historical data of wind direction, wind speed and corresponding wind power output power of a wind farm, and sampling the historical data to obtain sample data; Judging and analyzing the statistical characteristics of the sample data, obtaining the wind direction where the wind frequency is concentrated and the wind power output power difference reaches the difference threshold, and according to the acquired wind direction and its corresponding relationship with wind speed and wind power output power, the fuzzy hierarchical clustering method is used. The sample data is divided into three categories; the neural network algorithm is used to train each category of samples to form three specific types of wind power forecasting models, which are then merged to establish a hybrid wind power forecasting model for forecasting wind power capacity. Therefore, by implementing the present invention, targeted model prediction can be realized for different wind directions and wind speeds, and the prediction accuracy of wind power generation power can be improved.

Description

technical field [0001] The invention relates to the technical field of wind power forecasting, in particular to a forecasting method and system for hybrid wind power generation. Background technique [0002] When a wind power plant generates wind power, the output power of wind power is not only affected by the wind speed, but also the wind direction is a factor that cannot be ignored. However, due to the effect of atmospheric pressure, the wind speed and wind direction will change at any time, so the output power of the wind turbine is also volatile. The fluctuation of power output is not conducive to the stability and safety of the overall operation of the regional power grid. Therefore, it is necessary to make a reasonable forecast of wind power output power to facilitate effective scheduling by the power operation department. [0003] At present, in terms of short-term wind power forecasting, two methods are mainly used: physical model forecasting and statistical model ...

AI Technical Summary

Problems solved by technology

[0007] However, for statistical model prediction, the data used is single, and the short-term or ultra-short-term wind power prediction results can meet the accuracy requirements

However, for longer-term predictions, the accuracy of the prediction results is often not enough, and it requires long-term measurement data, a large amount of data processing work and additional training.

In addition, it does not handle mutation information well, rare weather conditions rarely appear during the training phase, and it is difficult to predict accurately

Images