本数据集基于2020年、2021年、2024年历史洪水过程数据，构建三峡-葛洲坝梯级水库群闸门优化调度模型，以最小化最大出库流量和闸门调整次数为目标函数，考虑水量平衡约束、水库水位约束、出库流量约束、泄水设施泄流约束、闸门离散开度约束、闸门启闭顺序约束等约束条件。利用深度强化学习近端策略优化算法求解，结合环境交互与策略评估阶段智能体的决策，选择累积奖励最大的决策过程作为相应入库流量过程下的最优闸门运行方案，通过不断更新策略，逐步优化最优调度过程，最终确定三峡、葛洲坝出库流量过程、水库水位过程、闸门运行方案。所制定的水库调度方案，在提高防洪能力的同时有效减小闸门启闭次数。
\n数据集每个excel包含4个sheet文件，分别为2020年7月、8月，2021年9月及2024年7月三峡和葛洲坝水库的来流（m³/s）、泄流（m³/s）、水位（m）及各闸门（二江左区、二江中区、二江右区、大江冲沙、深孔）开度（m）数据", "ds_source": "
来自长江电力股份有限公司
", "ds_process_way": "以最小化最大出库流量和闸门调整次数为目标函数，考虑水量平衡约束、水库水位约束、出库流量约束、泄水设施泄流约束、闸门离散开度约束、闸门启闭顺序约束等约束条件。利用深度强化学习近端策略优化算法求解，结合环境交互与策略评估阶段智能体的决策，选择累积奖励最大的决策过程作为相应入库流量过程下的最优闸门运行方案，通过不断更新策略，逐步优化最优调度过程，最终确定三峡、葛洲坝出库流量过程、水库水位过程、闸门运行方案。
", "ds_quality": "数据比较准确，达到精度要求。
", "ds_acq_start_time": "2020-07-16 00:00:00", "ds_acq_end_time": "2024-07-17 00:00:00", "ds_acq_place": "葛洲坝水库,三峡水库", "ds_acq_lon_east": 111.0, "ds_acq_lat_south": 29.26, "ds_acq_lon_west": 106.83, "ds_acq_lat_north": 31.42, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "login-access", "ds_total_size": 173870, "ds_files_count": 0, "ds_format": "*.xlsx", "ds_space_res": "", "ds_time_res": "日", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "644159a8-6834-453e-972c-065e6db390fd.png", "ds_thumb_from": 2, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": "", "organization_id": "44547705-9513-4685-a641-661bdf406520", "ds_serv_man": "", "ds_serv_phone": "", "ds_serv_mail": "", "doi_value": "", "subject_codes": [ "170.55" ], "quality_level": 0, "publish_time": "2026-07-14 17:17:02", "last_updated": "2026-07-14 17:17:02", "protected": false, "protected_to": "2027-01-01 00:00:00", "lang": "zh", "cstr": "", "i18n": { "en": { "title": "Hourly Dataset of Outflow Discharge, Reservoir Water Level, and Gate Operation Schemes for the Three Gorges and Gezhouba Reservoirs Optimized Based on Historical Flood Processes （2020、2021、2024）", "ds_format": "*.xlsx", "ds_source": "From Changjiang Electric Power Co., Ltd.
", "ds_quality": "The data is relatively accurate and meets the accuracy requirements.
", "ds_ref_way": "", "ds_abstract": "This dataset is based on historical flood process data in 2020, 2021, and 2024 to build an optimal dispatching model for the gates of the Three Gorges and Gezhouba cascade reservoirs, taking minimizing the maximum outflow flow and the number of gate adjustments as the objective functions, and considering constraints such as water balance constraints, reservoir water level constraints, outflow flow constraints, discharge constraints of water discharge facilities, discrete gate opening constraints, gate opening and closing sequence constraints. Using the deep reinforcement learning near-end policy optimization algorithm to solve the problem, combining the environmental interaction and the decision-making of the agent in the policy evaluation stage, selecting the decision-making process with the largest cumulative reward as the optimal gate operation plan under the corresponding inflow flow process. By continuously updating the strategy, gradually Optimize the optimal scheduling process and finally determine the outflow process of the Three Gorges and Gezhouba Dam, reservoir water level process, and gate operation plan. The formulated reservoir dispatching plan not only improves flood control capacity but also effectively reduces the number of gate openings and closures.
\r\nEach excel data set contains 4 sheet files, which are the inflow (m³/s), discharge (m ³/s), water level (m) and opening (m³/s) of the Three Gorges and Gezhouba Reservoir in July and August 2020, September 2021 and July 2024, and the opening (m) of each gate (the left area of the second river, the middle area of the second river, the right area of the second river, the sand flushing of the large river, and the deep hole) data", "ds_time_res": "", "ds_acq_place": "Gezhouba Reservoir, Three Gorges Reservoir", "ds_space_res": "", "ds_projection": "", "ds_process_way": "
Taking minimizing the maximum outflow flow and the number of gate adjustments as the objective function, considering constraints such as water balance constraints, reservoir water level constraints, outflow flow constraints, discharge facility discharge constraints, gate discrete opening constraints, gate opening and closing sequence constraints, etc. Using the deep reinforcement learning near-end policy optimization algorithm to solve the problem, combining the environmental interaction and the decision-making of the agent in the policy evaluation stage, selecting the decision-making process with the largest cumulative reward as the optimal gate operation plan under the corresponding inflow flow process. By continuously updating the strategy, gradually Optimize the optimal scheduling process and finally determine the outflow process of the Three Gorges and Gezhouba Dam, reservoir water level process, and gate operation plan.
", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "recommendation_value": 0, "license_type": "https://creativecommons.org/licenses/by/4.0/", "doi_reg_from": "reg_local", "cstr_reg_from": "reg_local", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "is_paper_in_submitting": false, "belong_to_nieer": false, "ds_topic_tags": [ "水库运行", "流量", "水位" ], "ds_subject_tags": [ "水文学" ], "ds_class_tags": [], "ds_locus_tags": [ "三峡", "葛洲坝" ], "ds_time_tags": [ 2020, 2021, 2024 ], "ds_contributors": [ { "true_name": "张奥南", "email": "zhangaonan@whu.edu.cn", "work_for": "武汉大学", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "张奥南", "email": "zhangaonan@whu.edu.cn", "work_for": "武汉大学", "country": "中国" } ], "ds_managers": [ { "true_name": "张奥南", "email": "zhangaonan@whu.edu.cn", "work_for": "武汉大学", "country": "中国" } ], "category": "水文" }