Comparison of Decomposition-based Machine Learning Models for Multi-step Time Series Forecasting of Wind Power Generation

Time Series Forecasting With Volatility Activation Function

“…Malhan [10] has compared three decomposition based machine learning algorithms in order to evaluate the performance of wind energy in multi-step univariate time series forecasting and the models he selected were STL-ARIMA, CEEMD-BiLSTM and CEEMDAN-BiLSTM. The accuracy of these models has been tested for different time frames ranging between short term namely one day before and long term namely three years ago.…”