本数据集面向三峡水库汛期运行水位动态控制研究，汇集了典型洪水过程下，在不同预报信息组织方式（单一预报、多尺度预报耦合）及不同预报不确定性水平条件下的水库调度模拟结果。数据包含6小时尺度的入库流量、下泄流量、库水位及控制域上下限等时序数据，并给出预报误差离散情景概率、不同预见期下的最高/平均水位等统计指标。可用于评估多尺度预报信息在三峡水库汛期动态控制中的应用效果，为洪水风险约束下的水资源利用与联合调度方案比选提供数据支撑。
\n数据文件为Excel工作簿，按研究情景分多个Sheet组织，主要包括：\n
1）预报不确定性参数与情景概率：\n
- Sheet2：预报误差倍比x（-1.0～1.0，步长0.1）的离散取值及不同合格率情景（40%～85%等）的概率。\n
- “不确定性变化”类Sheet：给出合格率α（%）、许可误差（0.2）、由正态分布反算的均方差σ，以及不同预见期下最高/平均水位等统计。\n
2）三峡水库调度模拟时序结果（典型洪水过程）： “多尺度1954-100a出图”“多尺度2020出图”等Sheet：6 h时间步长的调度过程线。\n
关键字段含义：Time（时间/时段序号）；入流-××（三峡入库流量，m³/s）；出流-××（下泄流量，m³/s）；水位-××（库水位，m）；弃水（弃水流量，m³/s）；控制域下限/上限（汛限水位动态控制域边界，m）。\n
- “动态控制”“M_动态控制”“规则联合/单库调度”等为不同调度方案或对照方案的结果列名标识；部分Sheet还包含3 d预报入流、预报无误/有误情景等中间变量列，用于复现实验计算过程。\n
数据精度：流量一般保留至0.001 m³/s，水位保留至0.01 m。", "ds_source": "
（1）典型洪水入库过程：包括1954年典型大洪水过程（100a情景）与2020年实测汛期洪水过程的三峡入库流量序列（6 h尺度）。\n
（2）预报不确定性信息：以“合格率”刻画预报误差水平，假定预报误差服从正态分布，并按误差倍比x（-1.0～1.0，步长0.1）离散化得到情景概率。\n
（3）调度约束与方案：基于三峡水库汛期运行与防洪调度约束，构建不同预报信息组织方式（单一预报/多尺度预报耦合）下的汛限水位动态控制方案，并设置规则调度/单库调度等对照方案。", "ds_process_way": "
1）按给定合格率α及许可误差0.2，利用正态分布分位数反算误差标准差σ，并将误差倍比x离散为-1.0～1.0（步长0.1）的情景；计算各情景概率并汇总于Sheet2。\n
2）在典型洪水入库边界条件下，采用6 h时间步长进行三峡水库调度模拟，输出不同方案（动态控制、多尺度耦合、规则/单库对照等）的出库流量与库水位过程。\n
3）从时序结果中统计最高水位、平均水位及最大下泄流量等指标，形成“××不确定性变化”类统计表。", "ds_quality": "
数据由统一的计算流程批量生成并汇总，已进行完整性与一致性检查，包括：时间步长一致性（6 h）、关键字段单位与量纲核对、缺失值检查、以及入库-出库-库水位变化的水量平衡一致性抽查。对于1954年时序数据中由Excel时间精度导致的毫秒级误差（如5.995 h/6 h），不影响过程统计与应用；2020年部分Sheet的Time列为日期重复记录，对应同日内4个6 h时段，可按记录顺序恢复为00/06/12/18时。
", "ds_acq_start_time": "1954-06-03 00:00:00", "ds_acq_end_time": "2020-12-31 00:00:00", "ds_acq_place": "中国湖北省宜昌市（长江干流三峡水库坝址及库区），并涉及其上游来水控制流域。", "ds_acq_lon_east": 111.05, "ds_acq_lat_south": 29.183333333333334, "ds_acq_lon_west": 106.5, "ds_acq_lat_north": 31.183333333333334, "ds_acq_alt_low": 145.0, "ds_acq_alt_high": 175.0, "ds_share_type": "login-access", "ds_total_size": 1586368, "ds_files_count": 0, "ds_format": "*.xlsx", "ds_space_res": "不适用（属性/时间序列数据，无栅格或矢量空间分辨率）。", "ds_time_res": "6小时（典型洪水调度过程线）；不确定性参数与统计表为情景/预见期尺度。", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "990aae10-7464-453c-99e0-1ed46326386d.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 15:25:07", "last_updated": "2026-07-14 15:25:07", "protected": false, "protected_to": "2027-01-01 00:00:00", "lang": "zh", "cstr": "", "i18n": { "en": { "title": "Dataset on Dynamic Control of Water Levels During the Flood Season at the Three Gorges Reservoir for the Coupled Utilization of Multi-Scale Forecast Information (1954, 2020)", "ds_format": "*.xlsx", "ds_source": "(1) Typical flood inflow process: The Three Gorges inflow flow sequence (6 h scale) including the typical major flood process in 1954 (100a scenario) and the measured flood process in 2020 during the flood season.\r\n
(2) Prediction uncertainty information: The forecast error level is characterized by the \"pass rate\", assuming that the forecast error follows a normal distribution, and discretization according to the error multiple ratio x (-1.0 to 1.0, step size 0.1) to obtain the scenario probability.\r\n
(3) Dispatch constraints and plans: Based on the Three Gorges Reservoir's flood season operation and flood control dispatch constraints, dynamic control plans for flood limit water levels under different forecast information organization methods (single forecast/multi-scale forecast coupling) are constructed, and regular dispatch/single reservoir dispatch are set up. Control plans such as scheduling.", "ds_quality": "
Data is generated and summarized in batches by a unified calculation process, and has been checked for completeness and consistency, including: time step consistency (6 hours), key field unit and dimension verification, missing value inspection, and water balance consistency spot check for changes in indoor-out-reservoir water level. For the millisecond-level error caused by Excel time accuracy in 1954 series data (such as 5.995 h/6h), it will not affect process statistics and applications; in 2020, the Time of some Sheets is listed as duplicate dates, corresponding to 4 6-hour periods within the same day, which can be restored to 00/06/12/18 in the recording order.
", "ds_ref_way": "", "ds_abstract": "This dataset is aimed at the dynamic control of the operating water level of the Three Gorges Reservoir in flood season. It collects reservoir dispatch simulation results under typical flood processes, different forecast information organization methods (single forecast, multi-scale forecast coupling) and different forecast uncertainty levels. The data includes time series data such as inflow flow, discharge flow, reservoir water level and upper and lower limits of the control region on a 6-hour scale, and provides statistical indicators such as discrete scenario probability of forecast error and maximum/average water level under different forecast periods. It can be used to evaluate the application effect of multi-scale forecast information in dynamic control of the Three Gorges Reservoir during flood season, and provide data support for the comparison and selection of water resource utilization and joint dispatch plans under flood risk constraints.
\r\nThe data files are Excel workbooks, organized into multiple Sheets according to research scenarios, mainly including:\r\n
1) Forecast uncertainty parameters and scenario probability:\r\n
- Sheet2: The discrete value of the forecast error multiple ratio x (-1.0 ￣ 1.0, step size 0.1) and the probability of different pass rate scenarios (40% ￣ 85%, etc.).\r\n
- \"Uncertainty Change\" Sheet: Give statistics such as pass rate α ( %), allowable error (0.2), mean square error σ calculated back from normal distribution, and maximum/average water level under different foreseeable periods.\r\n
2) Three Gorges Reservoir dispatch simulation time series results (typical flood process): Sheets such as \"Multi-scale 1954- 100a Mapping\" and \"Multi-scale 2020 Mapping\": Dispatch process line with 6 h time steps.\r\n
Key field meanings: Time (time/time period number); inflow-×× (Three Gorges inflow flow, m³/s); outflow-×× (discharge flow, m³/s); water level-×× (reservoir water level, m); Abandoned water (abandoned water flow, m³/s); Lower limit/upper limit of the control domain (dynamic control domain boundary of flood limit water level, m).\r\n
- \"Dynamic Control\",\"M_Dynamic Control\",\"Rule Union/Single Library Scheduling\", etc., are column names to identify the results of different scheduling schemes or comparison schemes; some Sheets also include intermediate variables such as 3D forecast inflow, error-correct/error-correct scenarios, etc., used to replicate the experimental calculation process.\r\n
Data accuracy: The flow rate is generally kept to 0.001 m³/s, and the water level is kept to 0.01 m.", "ds_time_res": "", "ds_acq_place": "Yichang City, Hubei Province, China (the dam site and reservoir area of the Three Gorges Reservoir on the main stream of the Yangtze River), and involves its upstream water control basin.", "ds_space_res": "", "ds_projection": "", "ds_process_way": "
1) According to the given pass rate α and the allowable error of 0.2, the standard deviation of the error σ is back-calculated using the normal distribution quantiles, and the error multiple ratio x is discretized into scenarios of-1.0 to 1.0 (step size 0.1); the probability of each scenario is calculated and summarized in Sheet2.\r\n
2) Under typical flood inflow boundary conditions, a 6-hour time step was used to simulate the Three Gorges Reservoir dispatch, and the outflow flow and reservoir water level processes of different schemes (dynamic control, multi-scale coupling, rule/single reservoir comparison, etc.) were output.\r\n
3) Count indicators such as the highest water level, average water level and maximum discharge flow from the time series results to form a statistical table of \"×× Uncertainty Change\".", "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": [ "三峡水库", "汛期运行水位", "动态控制", "多尺度预报", "预报不确定性", "联合调度" ], "ds_subject_tags": [ "水文学" ], "ds_class_tags": [], "ds_locus_tags": [ "长江上游", "三峡库区", "湖北宜昌" ], "ds_time_tags": [ 1954, 2020 ], "ds_contributors": [ { "true_name": "王浅宁", "email": "745303874@qq.com", "work_for": "大连理工大学", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "王浅宁", "email": "745303874@qq.com", "work_for": "大连理工大学", "country": "中国" } ], "ds_managers": [ { "true_name": "王浅宁", "email": "745303874@qq.com", "work_for": "大连理工大学", "country": "中国" } ], "category": "水文" }