【作者】 刘晨；

【导师】 符杰； 唐金波；

【作者基本信息】 西华大学 ， 土木水利硕士（专业学位）， 2024， 硕士

【摘要】 泥石流作为一种具备极高破坏性的地质灾害,其触发机制主要与暴雨、冰雪融化及地震等自然因素紧密相关。并且泥石流流动过程不仅以其快速、具有破坏力和强侵蚀性而著称,而且能在其流动路径中侵蚀携带大量物质,使得泥石流规模增大几倍甚至几十倍,对沟道下游地区的人民财产安全和基础设施构成极大的威胁。泥石流的侵蚀作用是泥石流流体和沟道底床物质相互作用的结果,其同时受到泥石流自身流体动力学特性与沟道物质物理属性等多方面因素的共同影响。由于泥石流流体内部复杂的固-液相互作用及其非牛顿流体特性,其流态相较于传统水流呈现更加复杂。本研究基于蒋家沟野外考察数据,利用相似理论构建泥石流动力学实验装置。开展了60组水槽实验,研究泥石流侵蚀率与沟床物性参数(包括底床含水率、底床细颗粒含量)和流体动力学参数之间的互馈关系。实验采用高速摄像机采集泥石流运动和侵蚀过程中的高帧率视频(采集频率为240 Hz),并通过逐帧图片分析获取泥石流侵蚀率数据。实验采用开源工具Pivlab对采集到的高帧率视频进行分析和计算,获取了泥石流在运动过程中的峰值流速和流速时程曲线。采用“db5”小波分析对实验过程中采集到的流深数据进行降噪处理,并根据流量公式计算得到泥石流峰值流量和流量时程曲线。研究发现,泥石流的侵蚀现象主要发生在泥石流龙头和龙身前半段流经侵蚀段的过程中。泥石流的剪切能力与沟道底床的抗剪能力之间的复杂相互作用关系导致泥石流侵蚀率随时间的发展呈现先增加后降低再增加的波动趋势,但泥石流侵蚀率随底床含水率的增加趋于降低。泥石流的动力学参数(流速、流深、流量)总体趋势呈现出随时间先快速增加后缓慢降低的趋势。本研究采用底床含水率和底床细颗粒含量表征沟道底床物性,并进一步采用一系列无量纲参数(Froude数、Savage数、Bagnold数)表征泥石流流态,全面分析了泥石流侵蚀过程中无量纲侵蚀率与沟道底床物性、泥石流流态之间的互馈关系。并基于无量纲参数分析和神经网络拟合方法,建立了泥石流无量纲侵蚀率与底床物性参数和泥石流动力学参数之间的数学模型,拟合度为0.81。模型预测的无量纲侵蚀率与实际观测值的误差在可接受范围内,表明模型能够较为准确地揭示无量纲侵蚀率与底床物性参数和泥石流流体动力学参数之间的关联性。通过深入理解泥石流动力学与侵蚀过程的相互作用机理,本研究旨在为防治泥石流灾害,减少其对人类社会和自然环境的影响,并提供全面的科学依据和技术解决方案。更多还原

【Abstract】 Debris flows,as a highly destructive geological disaster,are mainly triggered by natural factors such as torrential rains,melting of ice and snow,and earthquakes.The flow process of debris flows is known for its speed,destructiveness,and strong erosive power,and it can erode and carry a large amount of material along its flow path,increasing the scale of the debris flow several times or even dozens of times,posing a great threat to the safety of people’s property and infrastructure in the downstream areas of the gully.The erosion of debris flows is the result of the interaction between the debris flow fluid and the materials on the gully bed,and it is influenced by various factors including the hydrodynamic characteristics of the debris flow itself and the physical properties of the gully materials.Due to the complex solid-liquid interactions within the debris flow fluid and its non-Newtonian fluid characteristics,its flow behavior is more complex than that of traditional water flows.This study,based on field data from Jiangjia Gully,uses similarity theory to construct a debris flow dynamics experimental setup.A total of 60 flume experiments were conducted to study the feedback relationship between the erosion rate of debris flows and the physical parameters of the gully bed(including bed moisture content and fine particle content)and the hydrodynamic parameters of the flow.High-speed cameras were used to capture high-frame-rate videos(at a capture rate of 240 Hz)of the motion and erosion process of debris flows,and frame-by-frame image analysis was used to obtain data on the erosion rate of debris flows.The open-source tool Pivlab was used to analyze and calculate the high-frame-rate videos,obtaining the peak flow velocity and flow velocity-time curves of debris flows during motion."db5" wavelet analysis was used to denoise the flow depth data collected during the experiments,and the peak flow rate and flow rate-time curves of debris flows were calculated based on the flow rate formula.The study found that the erosion phenomenon of debris flows mainly occurs during the process of the debris flow head and the front half of the body passing through the erosion section.The complex interaction between the shear capacity of the debris flow and the shear resistance of the gully bed bottom leads to a fluctuating trend in the erosion rate of the debris flow over time,increasing first,then decreasing,and then increasing again,but the erosion rate of the debris flow tends to decrease with the increase in bed moisture content.The hydrodynamic parameters of the debris flow(flow velocity,flow depth,flow rate)generally show a trend of rapid increase followed by slow decrease over time.This study characterizes the physical properties of the gully bed using bed moisture content and fine particle content and further characterizes the flow state of the debris flow using a series of dimensionless parameters(Froude number,Savage number,Bagnold number),comprehensively analyzing the feedback relationship between the dimensionless erosion rate of the debris flow and the physical properties of the gully bed and the flow state of the debris flow.Based on dimensionless parameter analysis and neural network fitting methods,a mathematical model was established between the dimensionless erosion rate of debris flows and the physical parameters of the bed and the hydrodynamic parameters of debris flows,with a fitting degree of 0.81.The error between the model-predicted dimensionless erosion rate and the actual observed values is within an acceptable range,indicating that the model can accurately reveal the correlation between the dimensionless erosion rate,the physical parameters of the bed,and the hydrodynamic parameters of the debris flow.By deepening the understanding of the interaction mechanism between the dynamics and erosion process of debris flows,this study aims to provide comprehensive scientific basis and technical solutions for the prevention and control of debris flow disasters and to reduce their impact on human society and the natural environment.更多还原