南水北调中线总干渠冰情预测模型参数敏感性分析及率定

南水北调中线工程黄河以北总干渠冬季由于受气温的影响,将有不同程度的冰情产生,总干渠将处于无冰、流冰、冰盖下输水等多种复杂运行状态,运行不当则可能发生冰塞、冰坝危害。为了更准确地预测冰情,文章利用详细热量交换法对不同条件下渠道内一维冰盖生消进行数值模拟,根据2012年冬季的冰清观测结果以均方根差最小的原则率定模型参数。随后通过改变反射系数α、云量C、虚拟冰盖厚度Δh、风速Va、折减系数Cc及Ce、初始水温等分析冰盖厚度对于模型参数的敏感性。仿真结果显示,模型参数中反射系数α、虚拟冰盖厚度Δh这两个参数对冰盖厚度影响较大,气温数据中云量C对冰厚结果影响较大,且在高云量区域敏感性为极敏感,这些规律可以为寒区渠道系统冰情预测模型的开发及率定提供参考。

Parameter sensitivity analysis and calibration of the ice cover perdition model for the main canal of the middle route of South-to-North Water Transfer Project

Because of the low temperature in winter ,different degree of ice cover will grow in the main canal of the middle route of South‐to‐North Water Transfer Project in the north of the Yellow River .The main canal would be operated under different manner such as ice‐free ,drifting ice and ice cover mode ,improper operation will lead to disaster like ice jam and ice dam .In order to anticipate the forming procedure of ice cover in main canals ,a mathematical model in this paper was used to simulate the growth and decay of ice cover under different condition by calculating the detailed heat exchange .According to the field observ‐ing data of ice cover in 2012 ,aset of parameters were calibrated which simulation result by minimizing the quadratic error with measured ice thickness .Thenthe sensitivity of six parameters in the ice prediction model was evaluated :reflection coefficient α, cloudiness C ,virtual ice thickness Δh ,wind speed V a ,the reduction factor Cc and Ce ,initial water temperature .The results showed that the reflection coefficient α,cloudiness Cand virtual ice thickness Δh are more sensitive to the prediction of ice thick‐ness .When the value of cloudiness is high ,the sensitivity of it is extremely high .These results can be used to developed and cal‐ibrate the ice cover prediction model of canal system in clod region .