Abstract
Wind energy is a kind of sustainable energy with strong uncertainty. With a large amount of wind power injected into the power grid, it will inevitably affect the security, stability and economic operation of the power grid. High-precision wind power spot prediction and fluctuation interval information can provide more adequate decision-making support for grid scheduling and optimization. Hence, this paper proposes a K-Means-long short-term memory (K-Means-LSTM) network model for wind power spot prediction, and a nonparametric kernel density estimation (KDE) model with bandwidth optimization for wind power probabilistic interval prediction. The long short-term memory (LSTM) network has a strong memory function, which can establish the correlation between the data before and after, so as to improve the prediction accuracy. The K-Means clustering method forms different clusters of wind power impact factors to generate a new LSTM sub-prediction model. The optimization of the bandwidth in the nonparametric KDE is implemented by the mean integrated squared error criterion. In addition, a part of the dataset is deliberately demarcated from the wind power historical dataset to generate reasonable wind power prediction errors. The simulation results show that the proposed K-Means-LSTM network model has higher prediction accuracy than the back propagation (BP) neural networks, Elman neural networks, support vector regression (SVR) and LSTM network models. Compared with the KDE model with random bandwidth and the Gaussian distribution model, the bandwidth optimization model proposed in this paper has more narrow prediction intervals with higher interval coverage rates.
Highlights
In recent years, with the increasing attention of countries around the world to the development of renewable energy, the research and development of wind energy is increasing day by day [1]
The method of the wind power prediction within the wind power historical data in the way of constructing supervised learning has serious delays, and the delay effect differs by a single unit time
In order to avoid the above problems, this paper introduces the historical datasets of wind power with meteorological data as the research object, which comes from two wind farms in the Northeast of China
Summary
With the increasing attention of countries around the world to the development of renewable energy, the research and development of wind energy is increasing day by day [1]. This paper will conduct an in-depth study from the perspective of the NWP data and the LSTM network model to achieve wind power spot prediction. The wind power spot prediction methods applied in many literatures are often inconsistent with the theoretical research methods in practical applications Such error data cannot represent the error dataset generated by the actual prediction method. B) A spot prediction method based on long short-term memory (LSTM) networks for wind power is proposed. Compared with the general division of dataset, a part of the error generation dataset is obtained, which can obtain the related predicted power error distribution based on different prediction algorithms, and prepare for the follow-up data work for the wind power probabilistic interval prediction. Where, u and w are the weight values; b is the bias values; σ is the activation function and the sigmoid function is applied in this paper; is the Hadamard product
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
