The high proportion of new energy integration into the basin’s cascaded hydropower system significantly increases the uncertainty of grid scheduling. To address the highly coupled, nonlinear, multi-constraint, and multi-objective optimization challenges faced by the cascaded hydro-wind-photovoltaic complementary system, an optimal day-ahead scheduling method based on a physically-embedded neural network was proposed. In the modeling phase, core physical mechanisms, such as the turbine efficiency curve, reservoir water balance equation, and wind power probability distribution, were embedded as physical constraint loss terms and integrated into a fully connected neural network architecture. This integration was aimed at enhancing the interpretability of scheduling outcomes and overcoming the limitations of traditional purely data-driven methods. During the training phase, by leveraging the DeepXDE framework, the physical constraint loss term and the data-driven loss term were jointly optimized. An adaptive weight adjustment mechanism was introduced to coordinate multi-source information between historical operational data and physical constraint equations, thereby improving the physical plausibility of the scheduling results and their consistency with actual operations. Furthermore, an adaptive weighting optimization strategy for multi-source losses was adopted to mitigate numerical instability and convergence difficulties caused by gradient magnitude discrepancies during training, thereby facilitating the generation of optimal scheduling plans. Comparative validation using an actual basin dataset demonstrated the model’s validity and generalization capability. It is shown to produce generation scheduling strategies that balance system reliability and economy, significantly enhancing the system’s robustness in coping with fluctuations in wind and solar power, as well as the dynamic synergistic optimization performance of hydro-wind-solar multi-energy gradient utilization.
TianJingjing, XingKang. Analysis of wind power photovoltaic grid connection vulnerability based on complex network and electrical characteristics[J]. Science Technology and Engineering, 2025, 25(6): 2379-2388.
ZhengXiaoliang, YangXiaoliang, LaiWenhao. Classification of prediction methods for new energy power generation and research progress[J]. Science Technology and Engineering, 2024, 24(34): 14503-14521.
[7]
王函. 风光发电功率与用电负荷联合预测方法研究[D]. 北京: 华北电力大学, 2021.
[8]
WangHan. Research on joint prediction method of wind and solar power generation power and electricity load[D]. Beijing: North China Electric Power University, 2021.
YuYi, LiuZixuan, ZhaoGuohan, et al. Spatiotemporal correlation scenario generation of wind-solar complementary systems based on SGMM-MCopula[J]. Science Technology and Engineering, 2025, 25(10): 4156-4167.
GaoFan, BaoDaorina, ZhaoMingzhi, et al. Smoothing characteristics of hybrid energy storage on wind-solar coupled output fluctuations under multi-scenario planning[J]. Proceedings of the China Electrotechnical Society, 2025, 40(9): 2827-2839.
[13]
LiuM, HuoH, XuY, et al. Two-stage stochastic robust optimization for the planning and operation of rural integrated energy systems considering source-load uncertainties[J]. Renewable Energy Focus, 2025, 55: 100735.
[14]
ZhengL, ChenY, LiuG, et al. A unified robust optimization approach for problems with costly simulation-based objectives and constraints[J]. Computers and Operations Research, 2025, 183: 107179.
[15]
LuZ, YuJ. An OGSM-based multi-objective optimization model for partner selection in fresh produce supply chain considering carbon emissions[J]. Computers & Industrial Engineering, 2024, 194: 110402.
[16]
ZouG, LiZ, HeC. A new product configuration model for low product cost and carbon-neutral expenditure[J]. Sustainability, 2023, 15(13): 10358.
ChenLuyao, WenXin, TanQiaofeng, et al. Research on real-time load optimization allocation of cascade hydropower stations based on DQN[J]. Hydraulic and Hydropower Technology, 2025, 56(7): 26-40.
CuiYang, JiangTao, ZhongWuzhi, et al. Economic dispatch of regional integrated energy systems considering wind power consumption and source-load coordination[J]. Power System Technology, 2020, 44(7): 2474-2483.
MoJiang, YanWangji. A high-dimensional stochastic system identification framework based on physics-informed embedded neural networks[C]// Proceedings of the 15th National Conference on Vibration Theory and Applications. Nanjing: Chinese Society of Vibration Engineering, 2023: 161.
ChenShili, LüLingling, TongJunkai, et al. Guided wave imaging of pipe wall based on physical information embedded neural network[J]. Journal of Electronic Measurement and Instrumentation, 2023, 37(8): 136-145.
WangLigang, LiQun, JinYuan, et al. Optimal scheduling model of cascade hydropower station considering the uncertainty of incoming water[J]. Acta Solar, 2025, 46(6): 360-366.
[29]
WuC, XuY, SunH, et al. Bayesian distributionally robust optimization based day-ahead optimal dispatch in hybrid AC-DC distribution networks[J]. Energy, 2025, 333: 137094.
DongXiaochong, SunYingyun, PuTianjiao. Method for day-ahead scenario generation of renewable energy based on conditional generative adversarial network[J]. Proceedings of the Chinese Society for Electrical Engineering, 2020, 40(17): 5527-5536.
SangShengang, LiGuipeng, WangXiangwei, et al. Model predictive control of wind turbine yaw system based on reinforcement learning[J]. Thermal Power Generation, 2025, 54(9): 86-94.
ZhangYi, PengPeipei, TangBingjie, et al. Active power control strategy of wind power considering prediction error[J]. Journal of Xi’an University of Technology, 2021, 35(6): 76-82.
[38]
QuD, YaoY, XiaoY, et al. Short term output prediction of distributed photovoltaic grid connected generation in low-voltage distribution networks based on spectral clustering[J]. Journal of Physics: Conference Series, 2025, 2979(1): 012013.
[39]
TanQ, WangY, WenX, et al. Cross-regional peak-shaving scheduling for the hybrid pumped storage-wind-photovoltaic complementary system[J]. Energy, 2025, 326: 136119.
[40]
AlharbiT, IqbalS. Novel hybrid data-driven models for enhanced renewable energy prediction[J]. Frontiers in Energy Research, 2024, 12: 1416201.
LiPeng, HuangWenqi, WangXin, et al. Review on the application of artificial intelligence method driven by data and knowledge in power dispatching[J]. Automation of Electric Power Systems, 2024, 48(1): 160-175.
LiuXiuheng, ZuoXiang, MaJianbo, et al. Study on tide level forecasting model for tidal river sections based on mechanism and data joint-driven approach[J]. China Rural Water and Hydropower, 2025(11): 6-12.
WangXinying, YanDong, ShiZhan, et al. Optimization algorithms empowered by machine learning and their applications and prospects in novel power systems[J]. Proceedings of the China Electrotechnical Society, 2024, 44(16): 6367-6385.