In the hydropower sector, accurate estimation of short-term reservoir inflows is an essential element to ensure efficient and safe management of water resources. Short-term forecasting supports the optimization of energy production, prevention of uncontrolled water discharges, planning of equipment maintenance, and adaption of operational strategies. In the absence of data on topography, vegetation, and basin characteristics (required in distributed or semi-distributed models), data-driven approaches can serve as effective alternatives for inflow prediction. This study proposes a novel hybrid approach that reverses the conventional LSTM (Long Short-Term Memory)—ARIMA (Autoregressive Integrated Moving Average) sequence: LSTM is first used to capture nonlinear hydrological patterns, followed by ARIMA to model residual linear trends.The model was calibrated using daily inflow data in the Izvorul Muntelui–Bicaz reservoir in Romania from 2012 to 2020, tested for prediction on the day ahead in a repetitive loop of 365 days corresponding to 2021 and further evaluated through multiple seven-day forecasts randomly selected to cover all 12 months of 2021. For the tested period, the proposed model significantly outperforms the standalone LSTM, increasing the R² from 0.93 to 0.96 and reducing RMSE from 9.74 m³/s to 6.94 m³/s for one-day-ahead forecasting. For multistep forecasting (84 values, randomly selected, 7 per month), the model improves R² from 0.75 to 0.89 and lowers RMSE from 18.56 m³/s to 12.74 m³/s. Thus, the hybrid model offers notable improvements in multi-step forecasting by capturing both seasonal patterns and nonlinear variations in hydrological data. The approach offers a replicable data-driven solution for inflow prediction in reservoirs with limited physical data.

This paper investigates the integration of hydropower and solar energy within the Lower Olt River cascade as a pathway toward sustainable energy generation in Romania. The study focuses on the conceptual design of future hybrid power plants consisting of existing hydropower facilities where floating photovoltaic panels are proposed to be installed on the reservoir’s surfaces. An estimation of electricity production from both sources was performed, followed by the formulation of a trading strategy for the July–September 2025 period. The paper also explores the interaction between tactical and strategic management in hydropower operation and planning, describing how forecasting and decision-making processes are structured within the institutional framework. Finally, results for the selected hydropower plants demonstrate the positive influence of floating photovoltaic deployment on company performance, the national energy mix, and the overall sustainability of energy generation in Romania.

Accurate forecasting of reservoir inflows based on historical data is essential for effective water resources planning and management. Water flow forecasting presents a major challenge due to the nonlinear, non-stationary, stochastic, and highly noisy nature of water flows in rivers and inflows into reservoirs. These complex characteristics hinder the modeling process and limit the effectiveness of conventional prediction methods, making it difficult to achieve high accuracy results. This study explores the use of the Elephant Herding Optimization (EHO) algorithm for hyperparameter tuning in three established forecasting models: Autoregressive Integrated Moving Average (ARIMA), Gated Recurrent Unit (GRU), and Random Forest (RF). The proposed framework enables efficient exploration of the parameter space and adaptive learning of inflow patterns, aiming to reduce overfitting and improve predictive accuracy. All models are trained using historical inflow data and evaluated for one-day-ahead forecasts over a 365-day period in an open-loop configuration. By combining data-driven methods with meta-heuristic optimization, this work contributes to the development of robust forecasting tools for water resources management, enhancing the resilience and sustainability of hydropower system operation.