Long-Term Wind Speed and Power Forecasting Based on LSTM: A Comprehensive Study

One of the aspects of sustainable development is attaining grid operation safety management at the right quality level. Since wind energy is also based on renewable energy, its development and improvement are in line with sustainable development goals. However, there are challenges to managing this energy that needs to be addressed. Considering the behavior of wind speed and power of turbines are two effective issues in controlling wind power since lack of this awareness leads to a decline in the quality of customer service or energy waste. Predicting these two factors is one of the fundamental issues in the design and operation of power generation and distribution systems. In this paper, the wind speed and wind power of wind turbines are predicted based on the combined model AM CNN Bi-LSTM. To show the performance of this model, the results of their accuracy and prediction error are compared with the results of LSTM and CNN-LSTM. In all datasets, the results of the proposed model were more successful than other current models, and in the best case, an accuracy of 98.17% was obtained for power forecasting.