SHORT-TERM WIND POWER PREDICTION BASED ON IWOA-SA-ELMAN NEURAL NETWORK
Due to the randomness and uncertainty of wind power generation,it is very difficult to predict its short-term power,and the neural network model has a wide range of applications in the field of wind power prediction relying on its powerful self-learning ability.However,the prediction accuracy of neural network is greatly affected by the initial weight,and prone to over-fitting problems.In this paper,an Elman neural network short-term wind power prediction model based on improved whale optimization algorithm(IWOA)and simulated annealing(SA)combined optimization is constructed.Firstly,the improved whale optimization algorithm combined with simulated annealing strategy is used to obtain the initial weights of high-quality neural network,and then the regularization loss function is introduced to prevent overfitting.Finally,the short-term wind power of a wind power plant in Valencia,Spain is taken as the research object.The algorithm is compared with back propagation(BP),long short-term memory(LSTM),Elman,WOA-Elman and IWOA-Elman neural network algorithms.The results show that the prediction error of IWOA-SA-Elman neural network model is the smallest,which verifies the rationality and effectiveness of the algorithm.
NETL
NSTL
万方数据
维普
