河道洪水流量预测方法研究

Study of Flood Flow Forecasting in Rivers

及时准确地预报某区域内河道指定区段洪水流量及发生时间,对合理实施该区域的防洪预案、落实抗洪抢险措施、组织调度人员及防汛物资具有重要意义。目前河道洪水预报普遍采用马斯京根流量演算法及加里宁—米尔加科夫洪水演进法,两种方法的参数率定存在局限性,对应支流的河道分段处理也存在问题。本文依据最小二乘法,建立含有预测河段上游干流、支流水文站(或水位站)流量或水位预测模型,该模型不受其他水文参数的率定精度影响,直接利用以往洪水及当次洪水上游、下游站的观测资料建立回归预测模型,并通过递推方式完成当次洪水预测,表达形式简单直观、便于实际应用。利用该模型完成了嫩江干流齐齐哈尔水文站2013年洪水流量预测,经与实测成果比较,洪峰流量最大拟合误差小于5.2%,具有较好的计算精度。

The widely used methods for forecasting river flood flow are the Muskingum flood routing algorithm and the Kalinin Milgakov method. However, these traditional methods are limited by the difficulty in parameter calibration and the flow balancing at confluences of multiple tributaries. In this study, the least square method is proposed to predicting flood flows or flood stages at main river and tributaries using stream gage data. The proposed model, utilizing historical flood data and real-time measurements, predict flood flow and stage by establishing recursive scheme using regression algorithm. The accuracy of the proposed model is independent of other parameters, and the method is simple and easy to be applied to practice. Taking Qigihar stream gage station at Nenjiang River as an example, flood flow in 2013 was forecasted. The results show that the model output is less sensitive to other hydrological parameters, and the forecasting precision is higher than traditional methods with a maximum error less than 5%.