A knowledge-guided LSTM reservoir outflow model and its application to streamflow simulation in reservoir-regulated basins

Accurate reservoir outflow simulation is crucial for modeling streamflow in reservoir-regulated basins. In this study, we introduce a knowledge-guided Long Short-Term Memory model (KG-LSTM) to simulate the outflow of reservoirs-Fengshuba, Xinfengjiang, and Baipenzhu in the Dongjiang River Basin, China. KG-LSTM is built on the standard hyperparameters-optimized-LSTM and the loss function considering reservoir operation knowledge. Model uncertainty is analyzed using the bootstrap method. We then propose a hybrid approach that combines KG-LSTM with the Three-parameter monthly hydrological Model based on the Proportionality Hypothesis (KG-LSTM-TMPH) for streamflow simulation. The propagation of inflow errors to outflow simulations is studied across the three reservoirs. Results show that KG-LSTM enhances accuracy and reduces uncertainty in outflow simulations for three reservoirs compared to LSTM, particularly for the multi-year regulated Xinfengjiang Reservoir: KG-LSTM improves Nash-Sutcliffe efficiency (NSE) from 0.59 to 0.64, reduces root mean squared error (RMSE) from 55.59 m³/s to 54.84 m³/s, and decreases the uncertainty index relative width (RW) from 0.55 to 0.51 during the testing period. For streamflow simulations at four downstream hydrological stations, the hybrid model KG-LSTM-TMPH achieves NSE values above 0.87 and outperforms LSTM-TMPH, particularly in the dry season. Inflow errors impact outflow most significantly for the Xinfengjiang Reservoir in April and May, for the Fengshuba Reservoir throughout the year, and for the Baipenzhu Reservoir in July and August. This study enhances reservoir outflow modeling by integrating reservoir operation knowledge with deep learning. The hybrid KG-LSTM-TMPH approach shows practical potential for streamflow simulation in reservoir-regulated basins, offering valuable applications for water resource management.