| 541 | 3 | 1044 |
| 下载次数 | 被引频次 | 阅读次数 |
【目的】流域梯级水电系统规模持续扩大与运行环境日趋复杂，传统优化调度方法难以适应流域复杂多样的调控要求，且其决策精度与求解效率均有限。【方法】以耗水量最小为主要目标，构建了兼顾电调-水调的梯级水电优化调度模型，并研发了基于深度强化学习(DQN)的高效求解方法。以大渡河中游梯级水电系统为研究实例，分别设置中等负荷，低负荷和高负荷三种工况，输入实际运行数据对模型进行训练，并结合耗水量、水位过程等角度对模型优化效果进行评估。【结果】结果显示：DQN算法可显著减少计算耗时，将计算效率提升约41.37倍；同时，DQN算法可以很好地平衡水位和流量等水调需求之间的冲突，相较于优化前，DQN可在将水位波动指数平均降低约0.058 m/min的同时将平均总耗水量减少1 158万m3;除此之外，提出的模型适用于多种工况，具有良好的稳定性。【结论】结果表明：基于DQN的负荷分配方法可有效增强系统运行稳定性与安全性，实现调度科学性与计算效率的双重突破；智能决策框架通过实时优化电站出力分配，显著降低水位波动与发电耗水，验证了电调-水调协同优化的可行性。该方法为梯级水电系统智能化调度与新时期复杂场景下的优化调控提供了新的技术路径。
Abstract：[Objective]With the continuous expansion of cascade hydropower systems and increasingly complex operational environments, traditional optimized scheduling method struggle to meet the complex and diverse regulation requirements of river basins, while their decision-making accuracy and computational efficiency remain limited.[Methods]To address these limitations, an optimized scheduling model was established for cascade hydropower systems that considered power generation scheduling and water resource regulation, with minimum water consumption as the primary objective. Additionally, an efficient solution method based on deep reinforcement learning(Deep Q-Network, DQN) was developed. Using the cascade hydropower system in the middle reaches of the Dadu River as a case study, three operating conditions(medium, low, and high load) were established. The model was trained using actual operational data and evaluated through water consumption and water level processes.[Results]The result showed that the DQN algorithm reduced computational time by approximately 41.37 times compared to conventional method. Furthermore, DQN effectively balanced conflicting water regulation demands(e.g., water level stability and flow control), achieving an average reduction of 0.058 m/min in the water level fluctuation index and a total water consumption decrease of 11.58 million m3 compared to pre-optimization. Notably, the proposed model exhibited good stability across diverse operating conditions.[Conclusion]The findings indicate that the DQN-based load distribution method enhances system operational stability and safety while achieving breakthroughs in both scientific scheduling and computational efficiency. By dynamically optimizing real-time power output distribution among stations, the intelligent decision-making framework significantly mitigates water level fluctuations and reduces water consumption in power generation, thereby validating the feasibility of coordinated power-water optimization. This method provides a novel technical approach for intelligent scheduling of cascade hydropower systems and their optimized regulation under complex operational scenarios in the new era.
[1] 李国英.进一步全面深化水利改革为推动水利高质量发展、保障我国水安全作出新的贡献：在2025年全国水利工作会议上的讲话[J].水利发展研究，2025,25(1):1-12.LI G Y.Further comprehensively deepen water conservancy reform to promote high-quality development of water conservancy and safeguard national water security:A speech at the 2025 national water conservancy work conference[J].Water Resources Development Research,2025,25(1):1-12.
[2] 过夏明，秦毓毅，魏少岩.节能发电调度下的水电短期优化调度[J].电力系统及其自动化学报，2010,22(1):138-141.GUO X M,QIN Y Y,WEI S Y.Short-term hydro power optimal scheduling in energy saving generation dispatch[J].Proceedings of the Chinese Society of Universities for Electric Power System and Its Automation,2010,22(1):138-141.
[3] 马宇航，黄媛，刘俊勇，等.考虑日内来水不确定和电网断面约束的梯级水电日前调度[J].电力建设，2020,41(9):39-49.MA Y H,HUANG Y,LIU J Y,et al.Day-ahead dispatching of cascade hydropower stations considering daily interval flow uncertainty and the section constraint of power grid[J].Electric Power Construction,2020,41(9):39-49.
[4] 李国英.为以中国式现代化全面推进强国建设、民族复兴伟业提供有力的水安全保障：在2024年全国水利工作会议上的讲话[J].水利发展研究，2024,24(1):1-10.LI G Y.Improved water security for China’s efforts to build itself into a stronger country and rejuvenate the Chinese nation on all fronts by pursuing Chinese modernization:Speech at the 2024 National Water Conservancy Work Conference[J].Water Resources Development Research,2024,24(1):1-10.
[5] 郭富强，郭生练，刘攀，等.清江梯级水电站实时负荷分配模型研究[J].水力发电学报，2011,30(1):5-11.GUO F Q,GUO S L,LIU P,et al.Real-time load distribution models for the Qingjiang cascade hydropower plants[J].Journal of Hydroelectric Engineering,2011,30(1):5-11.
[6] LEI K X,CHANG J X,WANG Y M,et al.Cascade hydropower stations short-term operation for load distribution considering water level synchronous variation[J].Renewable Energy,2022,196:683-693.
[7] 葛晓琳，张粒子，王春丽.多目标短期梯级水电优化调度混合整数模型[J].电力系统保护与控制，2013,41(4):55-60.GE X L,ZHANG L Z,WANG C L.Mixed integer model for multi-objective short-term cascaded hydropower optimal scheduling[J].Power System Protection and Control,2013,41(4):55-60.
[8] 吴丰昌.我国水体污染控制与治理成效、科技支撑与展望[J].水利发展研究，2023,23(12):1-8.WU F C.Effectiveness,scientific and technological support,and prospects for water pollution control and management in China[J].Water Resources Development Research,2023,23(12):1-8.
[9] 张建云，刘九夫，金君良.关于智慧水利的认识与思考[J].水利水运工程学报，2019(6):1-7.ZHANG J Y,LIU J F,JIN J L.Understanding and thinking of smart water conservancy[J].Hydro-Science and Engineering,2019(6):1-7.
[10] 牛文静，冯仲恺，程春田.梯级水电站群优化调度多目标量子粒子群算法[J].水力发电学报，2017,36(5):47-57.NIU W J,FENG Z K,CHENG C T.Multi-objective quantum-behaved particle swarm optimization for operation of cascade hydropower stations[J].Journal of Hydroelectric Engineering,2017,36(5):47-57.
[11] 梅亚东，熊莹，陈立华.梯级水库综合利用调度的动态规划方法研究[J].水力发电学报，2007,26(2):1-4.MEI Y D,XIONG Y,CHEN L H.A dynamic programming method for the multi-purpose operation of cascade reservoirs[J].Journal of Hydroelectric Engineering,2007,26(2):1-4.
[12] 董子敖，闫建生，刘文彬，等.径流时空相关时梯级水库群补偿调节和调度的多目标多层次优化法[J].水力发电学报，1986,5(2):1-15.DONG Z A,YAN J S,LIU W B,et al.Multiobjective and multilevel method of optimal compensation adjustmentand operation of cascade reservoirs taking into account correlations of river flows in time and space[J].Journal of Hydroelectric Engineering,1986,5(2):1-15.
[13] NEEDHAM J T,WATKINS D W,LUND J R,et al.Linear programming for flood control in the Iowa and des Moines Rivers[J].Journal of Water Resources Planning and Management,2000,126(3):118-127.
[14] 王浩，王旭，雷晓辉，等.梯级水库群联合调度关键技术发展历程与展望[J].水利学报，2019,50(1):25-37.WANG H,WANG X,LEI X H,et al.The development and prospect of key techniques in the cascade reservoir operation[J].Journal of Hydraulic Engineering,2019,50(1):25-37.
[15] 郭生练，陈炯宏，刘攀，等.水库群联合优化调度研究进展与展望[J].水科学进展，2010,21(4):496-503.GUO S L,CHEN J H,LIU P,et al.State-of-the-art review of joint operation for multi-reservoir systems[J].Advances in Water Science,2010,21(4):496-503.
[16] 周建中，李英海，肖舸，等.基于混合粒子群算法的梯级水电站多目标优化调度[J].水利学报，2010,41(10):1212-1219.ZHOU J Z,LI Y H,XIAO G,et al.Multi-objective optimal dispatch of cascade hydropower stations based on shuffled particle swarm operation algorithm[J].Journal of Hydraulic Engineering,2010,41(10):1212-1219.
[17] 舒卫民，马光文，黄炜斌，等.基于人工神经网络的梯级水电站群调度规则研究[J].水力发电学报，2011,30(2):11-14.SHU W M,MA G W,HUANG W B,et al.Study on hydropower stations operation rule by an artificial neural network model[J].Journal of Hydroelectric Engineering,2011,30(2):11-14.
[18] 李国英.深入贯彻落实党的二十大精神扎实推动新阶段水利高质量发展：在2023年全国水利工作会议上的讲话[J].水利发展研究，2023,23(1):1-11.LI G Y.Thoroughly implement the spirit of the 20th National Congress of the Communist Party of China and solidly promote the high-quality development of water conservancy in the new stage:Speech at the National Water Conservancy Work Conference in 2023[J].Water Resources Development Research,2023,23(1):1-11.
[19] 谭乔凤，宋嘉伟，闻昕，等.基于DQN算法的水电站站内负荷优化分配研究[J].水利学报，2024,55(11):1345-1355.TAN Q F,SONG J W,WEN X,et al.Research on optimal load allocation in hydropower station based on DQN algorithm[J].Journal of Hydraulic Engineering,2024,55(11):1345-1355.
[20] 程春田.碳中和下的水电角色重塑及其关键问题[J].电力系统自动化，2021,45(16):29-36.CHENG C T.Function remolding of hydropower systems for carbon neutral and its key problems[J].Automation of Electric Power Systems,2021,45(16):29-36.
[21] WANG H N,LIU N,ZHANG Y Y,et al.Deep reinforcement learning:A survey[J].Frontiers of Information Technology & Electronic Engineering,2020,21(12):1726-1744.
[22] LEE J,KAO H G,YANG S H.Service innovation and smart analytics for industry 4.0 and big data environment[J].Procedia CIRP,2014,16:3-8.
[23] 李文武，郑凯新，刘江鹏，等.基于SARSA算法的风电：抽蓄联合系统日随机优化研究[J].水电能源科学，2020,38(11):72-76.LI W W,ZHENG K X,LIU J P,et al.Research on daily stochastic optimization of combined wind PowerPumping system based on SARSA algorithm[J].Water Resources and Power,2020,38(11):72-76.
[24] 黄显峰，冉超越，周文，等.基于深度强化学习算法的水光互补优化调度研究[J].水利水电技术(中英文),2025,56(4):235-247.HUANG X F,RAN C Y,ZHOU W,et al.Research on water-light complementary optimal scheduling based on deep reinforcement learning algorithm[J].Water Resources and Hydropower Engineering,2025,56(4):235-247.
[25] 李成健，宋姝谊，粟宇，等.深度强化学习求解多目标.旅行商问题的研究综述[J/OL].计算机工程与应用，2025:1-18.[2025-03-16].https://kns.cnki.net/kcms/detail/11.2127.TP.20250116.1216.004.html.LI C J,SONG S Y,SU Y,et al.Review of multi-objective traveling salesman problem based on deep reinforcement learning[J/OL].Computer Engineering and Applications,2025:1-18.[2025-03-16].https://kns.cnki.net/kcms/detail/11.2127.TP.20250116.1216.004.html.
[26] 李文武，周佳妮，裴本林，等.梯级水库深度强化学习长期随机优化调度研究[J].水力发电学报，2023,42(11):21-32.LI W W,ZHOU J N,PEI B L,et al.Study on long-term stochastic optimal operation of cascade reservoirs by deep reinforcement learning[J].Journal of Hydroelectric Engineering,2023,42(11):21-32.
[27] 杨思明，单征，丁煜，等.深度强化学习研究综述[J].计算机工程，2021,47(12):19-29.YANG S M,SHAN Z,DING Y,et al.Survey of research on deep reinforcement learning[J].Computer Engineering,2021,47(12):19-29.
[28] 刘全，翟建伟，章宗长，等.深度强化学习综述[J].计算机学报，2018,41(1):1-27.LIU Q,ZHAI J W,ZHANG Zongchang,et al.A survey on deep reinforcement learning[J].Chinese Journal of Computers,2018,41(1):1-27.
[29] ARULKUMARAN K,DEISENROTH M P,BRUNDAGE M,et al.Deep reinforcement learning:A brief survey[J].IEEE Signal Processing Magazine,2017,34(6):26-38.
[30] 李佳，曲田，朱艳军，等.大型梯级水电站枯水期多模式优化调度模型研究[J].水资源与水工程学报，2024,35(1):124-132.LI J,QU T,ZHU Y J,et al.Multi-mode scheduling of large-scale cascade hydropower stations in dry season[J].Journal of Water Resources and Water Engineering,2024,35(1):124-132.
[31] 程海花，寇宇，周琳，等.面向清洁能源消纳的流域型风光水多能互补基地协同优化调度模式与机制[J].电力自动化设备，2019,39(10):61-70.CHENG H H,KOU Y,ZHOU L,et al.Collaborative optimal dispatching mode and mechanism of watershed-type wind-solar-water multi-energy complementary bases for clean energy absorption[J].Electric Power Automation Equipment,2019,39(10):61-70.
[32] 尤渺.大渡河流域下游梯级电站发电优化调度研究[D].重庆：重庆大学，2008.YOU M.Study on the United Optimal Dispatching of Cascaded Hydropower Plant in Lower Reaches of Dadu River Basin[D].Chongqing:Chongqing University,2008.
[33] 闻昕，刘凡骞，谭乔凤，等.复杂水利-电力约束下梯级水电站短期多层级多目标优化调度研究[J].水利学报，2023,54(4):392-404.WEN X,LIU F Q,TAN Q F,et al.Research on short-term multi-level and multi-objective optimal scheduling of cascade hydropower stations under complex water resources and power constraints[J].Journal of Hydraulic Engineering,2023,54(4):392-404.
[34] BARROS M T L,TSAI F T,YANG S-L,et al.Optimization of large-scale hydropower system operations[J].Journal of Water Resources Planning and Management,2003,129(3):178-188.
[35] 郑明芬，谭乔凤，闻昕，等.兼顾水调-电调需求的梯级水光互补调度及影响分析研究[J/OL].工程科学与技术，2024:1-14.[2025-03-05].https://kns.cnki.net/kcms/detail/51.1773.TB.20241105.1408.004.html.ZHENG M F,TAN Q F,WEN X,et al.Cascaded hydro-PV complementary dispatching and its impact analysis considerating both water and power demands[J/OL].Advanced Engineering Sciences,2024:1-14.[2025-03-05].https://kns.cnki.net/kcms/detail/51.1773.TB.20241105.1408.004.html.Production,2023,385:135704.
[36] 胡瑞.基于调度知识经验的梯级水库群优化调度方法研究[D].武汉：华中科技大学，2017.HU R.Research on Optimal Operation for Cascaded Hydropower Stations Based on Optimizing Knowledge Experience[D].Wuhan:Huazhong University of Science and Technology,2017.
[37] 舒生茂.基于深度学习的梯级水电站发电调度及系统设计研究[D].武汉：华中科技大学，2019.SHU S M.Research on Power Generation Dispatching and System Design of Cascade Hydropower Station Based on Deep Learning[D].Wuhan:Huazhong University of Science and Technology,2019.
[38] JIANG W Y,LIU Y Q,FANG G H,et al.Research on short-term optimal scheduling of hydro-wind-solar multi-energy power system based on deep reinforcement learning[J].Journal of Cleaner Production,2023,385:135704.
[39] FENG Z K,LUO T,NIU W J,et al.A LSTM-based approximate dynamic programming method for hydropower reservoir operation optimization[J].Journal of Hydrology,2023,625:130018.
[40] SHANG Y Z,LU S B,GONG J G,et al.Improved genetic algorithm for economic load dispatch in hydropower plants and comprehensive performance comparison with dynamic programming method[J].Journal of Hydrology,2017,554:306-316.
基本信息:
DOI：10.13928/j.cnki.wrahe.2025.07.003
中图分类号:TV737
引用信息:
[1]陈鹿尧,闻昕,谭乔凤,等.基于DQN的梯级水电站实时负荷优化分配研究[J].水利水电技术(中英文),2025,56(07):26-40.DOI:10.13928/j.cnki.wrahe.2025.07.003.
基金信息:
国家自然科学基金项目(52479013)
2025-05-21
2025-05-21
2025-05-21