Hydraulic transient-aware day-ahead dispatch of hydro-wind-solar hybrid energy system

Hydro-wind-solar hybrid energy system (HWSHES) plays a key role in coping with uncertain wind and solar power. However, the dynamic hydraulic transient characteristics of hydropower units (e.g. water hammer effects) are often overlooked in short-term dispatch studies. Ignoring these critical phenomena could lead to operational instability and accelerated equipment wear, particularly during the coordinated operation of hydro, wind, and solar resources. To bridge this gap, this study proposes a hydraulic transient-aware day-ahead dispatch strategy that synergizes hydraulic transient characteristics into the day-ahead dispatch process via a data-driven surrogate model-based approach. This strategy consists of two stages: (1) offline training: A refined hydraulic transient simulation model is developed using an improved transfer function method. A radial basis function (RBF) neural network is then trained as a surrogate model to approximate the nonlinear relationship between power regulation and hydraulic transient characteristics of hydropower units; (2) online optimization: A dynamic constraint-updating heuristic method is proposed to solve the resulting nonlinear closed-loop optimization problem efficiently. The proposed dispatch strategy is validated through a case study of the Ludila hydropower station-dominated HWSHES in China. Results show that (1) the surrogate model achieves high accuracy (RMSE = 0.1882, R2 = 0.9684), effectively capturing hydraulic transient characteristics and serving as a reliable alternative to time-domain simulation-based model; (2) compared to the general short-term dispatch strategy that ignores hydraulic transient characteristics, the proposed strategy enhances the hydraulic transient performance of HWSHES by an average of 23.93 %, with only a 0.72 % increase in water consumption across four seasons.