Statistical Analysis of Four Artificial Intelligence Algorithms for Multi-Step Short-Term Wind Speed Forecasting in Four Indian Cities

Short-term wind speed prediction is critical for optimizing renewable energy systems and minimizing negative impacts on power systems. Robust algorithms for short-term wind speed forecasting are critical, as wind power is a significant source of renewable energy. We used artificial intelligence techniques to find a suitable algorithm for forecasting short-term wind speeds, multi-step ahead, in four Indian cities: Chennai, Kolkata, New Delhi, and Thiruvananthapuram. We used the LSTM, BiLSTM, ConvLSTM, and Prophet approaches on data from 2017 to 2020 for training and 2021 for testing. The time horizon considered in the study is 1 h. The methods were compared on the basis of Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Coefficient of determination (R2) score. The results revealed that the ConvLSTM technique outperformed all others, with the lowest RMSE (Chennai: 0.801, New Delhi: 0.695, Kolkata: 0.863, Thiruvananthapuram: 0.721), lowest MAE (Chennai: 0.569, New Delhi: 0.479, Kolkata: 0.626, Thiruvananthapuram: 0.540), and highest R2 score (Chennai: 0.79, New Delhi: 0.81, Kolkata: 0.81, Thiruvananthapuram: 0.79) for all stations examined. This sophisticated method can forecast hourly wind speeds even when data is missing, making it a valuable tool.