A new wind power forecasting algorithm based on long short‐term memory neural network

Abstract

Wind power is one of the most large-scale new energy sources. But wind power instability will affect power grid safety, which is in a great need for wind power forecasting algorithms. To accurately predict wind power and reduce power grid fluctuations, it proposes a new wind power forecasting (WPF) algorithm based on long short-term memory (LSTM) neural network using wind farm real operation data. First, the wind farm power data are de-averaged and divided into two different sets in order to meet the requirements of the algorithm. Then, structure of the LSTM neural network is designed and hyperparameters are adjusted to improve accuracy of forecasting. Finally, the definite LSTM neural network is used for forecasting the power data in time series to derive the power forecasting value, which is reduced and evaluated according to the original size. The results show that compared with other forecasting methods, in short-term WPF at different time scales, this algorithm has smaller errors in power forecasting results, which is also suitable for longterm WPF. K E Y W O R D Sdeep learning, long short-term memory neural network, time series forecasting, wind power forecasting | INTRODUCTIONWind energy has received widespread attention in recent years. The advantages of wind power lie in its low cost and low pollution. 1 But wind power has large fluctuations and will greatly impact the grid when integrated into the grid. Countermeasures can be taken based on WPF. In general, a comprehensive study of the local wind energy should be done before the wind farm is installed. 2 WPF can estimate the future output power of wind turbine. By the forecast period, there are two forecasting: short-term and long-term. Short-term forecasting generally lasts less than a few months, while long-term forecasting generally lasts more than 1 year. Short-term forecasting can be used for downtime maintenance of wind turbine,

Machine learning brings new insights for reducing salinization disaster

“…In ANN, it is assumed that the output only depends on the current input, which is not true in the real world (Gu et al, 2019). LSTM allows us to infer the potential relationships among the content of the context, because it is a recurrent neural network (RNN)-the output depends on the current input and memory (Klimov et al, 2020;Huang et al, 2021;Li et al, 2022;Gorgij et al, 2023). The basic idea of RNN is to build a hidden state for acquiring the information at the previous time point and the global parameters are calculated from the current time and all previous memories.…”

Optimized ensemble model for wind power forecasting using hybrid whale and dipper-throated optimization algorithms

“…Since employing raw features might produce ineffective results, optimal feature selection can be essential in providing precise forecasts. Consequently, several techniques employed various feature selection strategies before using the data to train a model (Han et al, 2019;Huang et al, 2021;El-kenawy et al, 2022;Takieldeen et al, 2022). Relevant features are selected from raw data using the binary version of the proposed optimization algorithm, described by the steps presented in Algorithm 2.…”