研究坝体的溃决过程与溃坝洪水演进对于处置由溃坝引起的洪水灾害、提升水利安全具有重要的意义。鉴于目前大多模型均将坝体溃决过程与溃坝洪水演进分别进行模拟,不能反映土体与水流相互耦合的特点,模拟结果精度有限。基于对土体有限抗冲能力的考虑,选取双曲线型冲蚀速率表达式描述坝体冲蚀、采用简化Bishop法搜索临界滑裂面描述溃口边坡坍塌和具有总变差不增特性的MacCormack有限体积法离散控制方程,建立了坝体溃决过程与溃坝洪水演进耦合的平面二维数值模型。实际算例表明模型合理地模拟了溃口的发展过程与洪水演进过程,在溃口急缓流转换区展现了较强稳定性,守恒性良好,可作为溃坝洪水风险评估与洪灾预报的有力工具。

The study of dam break process and dam break flood routing is of great significance for dealing with flood disasters caused by dam break and improving water security. Most models at present could not reflect the interaction between soil and water flow by only simulating dam break process or flood routing, hence resulting in a limited accuracy. In this article, a 2-D numerical model coupling dam break process and flood routing was established in consideration of the finite erosion resistance of soil. The erosion rate expressions of hyperbolic type was employed to describe the rate of dam erosion; simplified Bishop method was adopted to search the critical slip surface for describing slope collapse at breach; and the Mac-Cormack finite volume method with total variation diminishing was adopted to discrete the control equations. A computation example demonstrates that the model could reasonably simulate the process of breach development and the flood routing process with strong stability and good conservativeness in breach transition zone from rapid flow to slow flow. The model can be used as a powerful tool for flood risk assessment and flood disaster forecast.

对于寒冷地区的混凝土大坝,由于表层保温层的影响,其内部温度往往滞后于气温变化。当内部温度计缺失时,使用水力-季节-时间(HST)模型进行大坝预测时存在较大的误差,且即使利用内部温度计进行多元回归(MR)模型的建模也无法反映温度与变形的非线性关系。因此,针对现阶段对高寒区变形预测精度低的问题,提出利用反向学习后的鲸群(OWOA)算法对RReliefF因子加权支持向量机(RFWSVR)与分布滞后线性模型(DLM)的温度因子的超参数进行寻优,以构建缺乏内部温度计的寒区混凝土大坝变形预测模型。结果表明:通过对所建立的变形预测模型与传统统计模型和其余常用机器学习算法的性能比较,证明所建立模型具有较高的预测精度,能更好地反映保温混凝土大坝的工作特点。

A memory-based network that provides estimates of continuous variables and converges to the underlying (linear or nonlinear) regression surface is described. The general regression neural network (GRNN) is a one-pass learning algorithm with a highly parallel structure. It is shown that, even with sparse data in a multidimensional measurement space, the algorithm provides smooth transitions from one observed value to another. The algorithmic form can be used for any regression problem in which an assumption of linearity is not justified.

为提升堤坝溃决险情处置和溃决灾害防御能力,多年来长江科学院河流研究所采用物理模型试验、水槽试验、理论分析、数值模拟等方法研究了堤坝漫溢溃决的机理、模型与模拟技术。其主要成果包括:揭示了堤坝漫溢溃决机理,解析了“溯源陡坎冲刷”在堤坝溃决过程中的作用,提出了溯源陡坎冲刷模式和堤坝漫溢溃决模式;基于物理机制,研发了溯源陡坎冲刷二维数学模型和堤坝漫溢溃决数学模型;发展了适应溃坝水流急变特征的一、二、三维溃坝洪水运动模拟技术及地形处理方法,并初步探索了溃坝水流的三维流场与动压特性;总结评述了相关领域的研究进展。研究成果成功应用于唐家山、白格等历次堰塞湖溃决险情的应急处置和决策制定,并为今后堤坝(含堰塞坝)溃决险情的科学应对提供了技术参考和经验借鉴。