Optimization study on the pumping station-hydropower station-reservoir system based on an improved NSGA3 algorithm

ABSTRACT Existing joint operation models often treat pumping and hydropower stations as independent units, neglecting the critical synergies of tailwater reuse and energy self-supply. This study aims to bridge this gap by establishing a ‘full-coupled’ simulation-optimization framework for medium-scale pump-hydropower-reservoir systems. The specific objective is to simultaneously maximize annual water transfer, power generation, ecological satisfaction, and notably, the energy self-sufficiency rate (ESSR). To address the challenge of computationally expensive simulations, we propose an improved NSGA3 algorithm, which enhances search efficiency using pre-selection and inverse modelling. Validated against walking fish group benchmarks, this algorithm was applied to a practical case study in Shanxi, China. The results demonstrate the superiority of the proposed framework: the full-coupled mode achieves a 12.2% increase in ESSR compared with traditional decoupled models by strategically trading off marginal gross water transfer volume. The analysis reveals that the system successfully converts extensive resource quantity into intensive operational autonomy, optimizing the synchronization of energy generation and consumption. Integrating tailwater utilization with internal energy coupling reduces grid dependence, offering an economically viable solution for sustainable water-energy nexus management.