|
- 2019
基于水文相似性的预报误差修正Keywords: 水文相似,误差修正因子,误差模拟模型,预报误差修正,怒江,hydrological similarity,errors correction factor,errors simulation model,hydrological forecasting errors correction,Nujiang River Abstract: 如何进一步提高水文预报精度,降低预报误差,更好的服务于防洪抢险、水资源调度是水文学重要研究方向。基于水文预报误差主要影响因素在相似流域具有相似性,致预报误差具有相似性的特性,以相似流域各预报站预见期不同为切入点,优选误差修正因子,基于二次正交回归设计,构建了预报误差模拟模型以修正预报结果,提高精度。经怒江道街坝水文站模拟预报检验,修正后模拟确定性系数由0-933提高至0-998,洪峰流量平均绝对误差由693 m3/s降低至116 m3/s,洪峰流量平均相对误差由16%降低至3%,洪峰出现时间预报误差由平均5-4 h降低至0-9 h,方案精度由乙级提升至甲级。提出的基于水文相似洪水预报修正方法结构简单、计算方便、修正效果明显,是一种高效提高洪水预报精度的方法,适用于相似流域不同预报站洪水过程无明显差异的情形。 How to further improve the accuracy of hydrological forecasting,reduce forecasting errors,better serve flood control and rescue and water resources dispatching is a hot topic in hydrology.A forecasting error correction model is constructed in this study to improve the forecasting accuracy,based on the similarity theory that the same main influencing factors of hydrological forecasting in similar watersheds result in the similarity of forecasting errors.Different forecasting periods were taken as a starting point in similar watersheds.The error correction factors were based on the quadratic orthogonal regression design.The coefficient of determination increased from 0-933 to 0-998 for the Nujiang Daojieba hydrological station application.The average absolute error of the flood peak flow decreased from 693 m3/s to 116 m3/s,and the average relative error of the flood peak flow decreased from 16% to 3%.The flood peak time of forecasting error reduced from an average of 5-4 hours to 0-9 hours,and the accuracy of the scheme was upgraded from Grade B to Grade A.The hydrological similar flood forecasting correction method proposed in this paper has the advantages of simple structure,easy calculation and significant correction effects,to imrpove the accuracy of flood forecasting. It is also applicable to the case where the water area ratios of different forecasting sections are not much different in similar watersheds,and the flooding process is not much different. 国家自然科学基金(91547205)
|