Forecasting of Wind Speed Using Deep Learning Method: A Case Study in Andhrapradesh

Wind power forecasting is crucial for effective grid planning and load scheduling, leading to efficient grid management, reduced power loss, and lower production costs. However, as the analysis of large datasets becomes more complex, achieving accurate wind power forecasts becomes increasingly challenging. Therefore, implementing effective forecasting methods is essential. This paper introduces a novel wind speed estimating model using the Long short-term memory (LSTM) technique. To evaluate the model's effectiveness, it was compared with four other methods: Linear regression (LR), Random Forest (RF), Support Vector Machine (SVM) and Decision tree (DT). The model was trained and tested using wind speed and meteorological data from the Narasaraopet, Andhra Pradesh region. Results show that all methods have, Mean Square Error (MSE), Mean absolute error (MAE), R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>, Mean Absolute Percentage error (MAPE), and Root Mean Square Error (RMSE) values between 0.017 and 12.1870, 0.102 and 2.596, 0.048 and 0.878, 1.149 and 60.098, 0.131 and 3.491 respectively. Support Vector Regression consistently performed the poorest across the board in terms of all metrics. The current work ends by emphasizing that, while all machine learning methods investigated in this research can be used to predict daily wind speed prediction data with little error, the RF and the LSTM algorithms produced the best results with the least amount of error. Then, that is followed in orderly succession by decision trees and linear regression.