过闸流量的精确计算是实现调水系统科学调度以及平稳运行的基础，同时也是调水工程数字孪生建设的重要水利专业模型之一，调度运行中虽已积累大量监测数据，但水流状态受节制闸调度指令影响，当前稳定输水状态被破坏，待调度完成后再次达到新的稳定状态，而监测设备无法识别过渡状态，虽然监测数据符合实际情况，但仍存在较大的波动性不利于参数率定，直接影响过流计算的可靠性。针对上述问题，提出一种稳定输水状态辨识方法，通过分析原始监测数据下流量系数变化值的分布规律得到其变化阈值。基于此，从监测数据中筛选出具有代表性的稳态数据集，结合量纲分析方法利用稳态数据率定参数，进行过闸流量计算。以南水北调中线工程中的金水河节制闸、淇河节制闸和七里河节制闸为例，对比分析利用原始监测数据和稳态数据率定参数后的流量误差效果。结果表明：流量平均相对误差从7.26%、3.35%和3.80%减少至6.55%、3.22%和2.19%。该研究提高了闸门水力计算精度，为进行高精度输水调度模拟预演及决策运行提供有力支撑。

Control gates play a crucial role in managing water levels and flow rates in water transfer canal systems. Arc gates are particularly favored due to their hydraulic efficiency and lightweight construction. Accurate flow rate calculation through these gates is essential for hydraulic simulation models and water management decisions. However, traditional empirical formulas face challenges due to the complex nature of arc gates, leading to the proposal of dimensionless analysis-based approaches. Combined with emerging technologies like artificial intelligence, these approaches improve adaptability and flow calculation accuracy. Yet, challenges persist, such as the need for representative data for parameter calibration and the impact of factors like equipment failures and dispatch instruction operations on monitoring data accuracy. In digital twin basin construction, accurately characterizing gate flow characteristics is crucial for effective water management. Therefore, identifying stable water delivery states and obtaining representative hydrological data are essential steps for analyzing gate flow coefficients and ensuring accurate flow rate calculations, ultimately supporting real-time monitoring and decision-making in water transfer projects. A stable water conveyance state identification method was introduced to accurately characterize stable water delivery states and select representative data for gate parameter calibration in digital twin basin construction. Leveraging dimensionless analysis, it contrasts flow rate calculation accuracy between monitoring and stable state data, validating the method's effectiveness. The aim is to provide scientific basis and technical support for precise gate flow capacity depiction and real-time gate state synchronization in water transfer projects. The methodology involves deriving discharge formulas, stable state identification, and dimensionless analysis. Threshold values for discharge coefficient change and cumulative change are determined by selecting stable state data from historical monitoring data. The dimensionless analysis method establishes a mathematical model for gate flow calculation. Additionally, the dimensionless analysis method establishes a mathematical model for gate flow calculation. Evaluation criteria, including R2,ERMS,EMA,EMAP, and ENS, assess method accuracy and performance. This comprehensive approach ensures reliable gate parameter calibration and enhances the robustness of water management decisions in open channel water transfer systems. The study examines three control gates from different South-to-North Water Transfers Project segments: Jinshui River Control Gate, Qi River Control Gate, and Qili River Control Gate. Using one year data from July 2022 to July 2023, at 2-hour intervals, stable state identification involved normality testing of comprehensive flow coefficient changes, revealing a bell-shaped distribution for three gates. Thresholds, based on a 95% confidence interval and a 4-hour cumulative change duration, identified stable water conveyance states. Specific thresholds were set for change values and cumulative changes at each gate, ensuring reliable data for water transfer management decisions. Stable state data showed greater representatives, utilizing stable state data identified through dimensionless analysis, the determination coefficients of the comprehensive flow coefficients for the Jinshui River Control Gate, Qi River Control Gate, and Qili Control River Control Gate were all improved compared to original monitoring data. Additionally, the root mean square error (ERMS) significantly decreased, with reductions of 43%, 47%, and 29%, respectively. Moreover, the accuracy of flow rate calculations using stable state data surpassed that of original monitoring data, reducing the average relative errors for the Jinshui River Control Gate, Qi River Control Gate, and Qili River Control Gate from 7.26%, 3.35%, and 3.80% to 6.55%, 3.22%, and 2.19%, respectively. Significant insights emerge when comparing results derived from original monitoring data and stable state-identified data. First, parameter calibration utilizing stable state-identified data enhances the determination coefficient of the comprehensive flow coefficient for all three gates, leading to notable reductions in root mean square error (ERMS). Second, the precision of flow calculations improves when using stable state data, resulting in decreased average relative errors in flow for each gate. Third, the proposed stable water conveyance state identification method enables the extraction of representative datasets for different scheduling conditions, and offering robust support for high-precision water transfer scheduling simulations and canal hydraulic capacity analyses. In conclusion, this method demonstrates promising applicability and potential for widespread adoption in practice.