A Hybrid Model of Ensemble Empirical Mode Decomposition and Sparrow Search Algorithm-Based Long Short-Term Memory Neural Networks for Monthly Runoff Forecasting

Monthly runoff forecasting plays a vital role in reservoir ecological operation, which can reduce the negative impact of dam construction and operation on river ecosystem. Numerous studies have been conducted to improve monthly forecast accuracy, machine learning methods paid much attention due their unique advantages. In this study, conjunction model, EEMD-SSA-LSTM for short, comprises ensemble empirical mode decomposition (EEMD) sparrow search algorithm (SSA)–based long short-term neural networks (LSTM), has proposed forecasting. The model is mainly carried out three steps. First, original time series data decomposed into several sub-sequences. Second, each sub-sequence simulated by LSTM, hyperparameters are optimized SSA. Finally, results summarized as final results. obtained from two reservoirs located China used validate performance. Meanwhile, four commonly statistical evaluation indexes utilized evaluate demonstrate that compared benchmark models, yield satisfactory be conducive improving accuracy.