Wind power forecasting method based on genetic algorithm optimization BP neural network

Abstract

The invention discloses a wind power forecasting method based on a genetic algorithm optimization BP neural network, comprising the steps: acquiring forecasting reference data from a data processing module of a wind power forecasting system; establishing a forecasting model of the BP neural network to the reference data, adopting a plurality of population codes corresponding to different structures of the BP neural network, encoding the weight number and threshold of the neural network by every population to generate individuals with different lengths, evolving and optimizing every populationby using selection, intersection and variation operations of the genetic algorithm, and finally judging convergence conditions and selecting optimal individual; then initiating the neural network, further training the network by using momentum BP algorithm with variable learning rate till up to convergence, forecasting wind power by using the network; and finally, repeatedly using a forecasted valve to carry out a plurality of times of forecasting in a circle of forecast for realizing multi-step forecasting with spacing time interval. In the invention, the forecasting precision is improved, the calculation time is decreased, and the stability is enhanced.

Description

technical field [0001] The invention relates to a method for predicting wind power in a wind farm, in particular to a method for predicting wind power based on a genetic algorithm optimized BP neural network. Background technique [0002] With the expansion of wind power scale, a series of problems caused by wind power grid connection need to be solved urgently, including the scheduling problem of power system including wind farms. Accurate prediction of wind power power is the key to solving this problem. Accurate wind power forecasting helps the power sector to formulate effective power generation plans, thereby ensuring the safe, stable and economical operation of the power system. Therefore, wind power forecasting is a necessary prerequisite and important support for wind power dispatching. The current prediction methods mainly include: v-P curve method, that is, to predict the wind speed first, and then calculate the wind speed and power value according to the curve of ...

AI Technical Summary

Problems solved by technology

[0016] (1) During the prediction process, the convergence speed of the BP neural network is slow and the training process may be trapped in a local minimum, resulting in an increase in the prediction time and a decrease in the prediction accuracy. This requires multiple trainings by changing the initial values ​​of the weights and thresholds of the BP neural network. improve;

[0017] (2) Since there is no effective method to determine the structure of the BP neural network, the number of neurons in the hidden layer is mostly determined by a trial algorithm in the actual prediction process. When the data changes, the trained BP neural network may not continue to be applicable, further affecting the prediction result;

[0018] (3) Existing prediction methods are highly dependent on data. When encountering bad data, BP neural network training will become difficult to converge. At this time, the calculation time will increase significantly, and the prediction system is not stable enough.

[0019] The above problems existing in the existing forecasting method based on BP neural network limit its application in actual wind power forecasting to a certain extent.

Images