| The shock characteristic of debris flow is one of important scientific problems indebris flow dynamics. Paying attention to the multiphase medium such as artificialdebris flow composed of slurry with various viscosity, gradation particles and particleconcentrations (ratio of solid particle to liquid), the study use research methods such asmodeling experiments in laboratory, the wavelet signal processing and the frequencyspectrum analysis to make studies systematically on model test of debris flow, dataprocessing, frequency characteristics of debris flow shock action, energy distribution ofdebris flow shock signal, and debris flow wallop. Depending on the resolution of somekey scientific problems such as the stability control to supply debris flow slurry and putin particles in modeling experiments and the denoising of shock signals, the studiesdiscuss the response relations of debris flow instantaneous shock loads, probabilitydensity types of debris flow shock load to slurry, and gradation particles and particleconcentrations. Further, extract debris flow shock signals with various frequency bandsin shock spectrums of debris flow, analyze the local characteristics and discover theenergy distribution pattern of the shock signals under experiment conditions. Finally,put forward a new formula to estimate debris flow wallop under experiment conditions.Those innovate achievements obtained in this thesis can be summarized as follow.(1) A laboratory experimental model for shock characteristics of debris flow hasbuilt in dynamic model ground in Chongqing Jiaotong University, and the groove isbuilt with the length is13.0m, the width is0.5m, and the height is0.6m, in which twocurved conduit is designed. Water supply and material supply are realized by watersupply box and ballast hopper, and then, valve in water supply box and drawing plate inballast hopper are the key to make water and materials released steadily. Mean while,high-accuracy transducer (HS800type) and Dynamic signal testing and analysis system(DH5922type) make the reception and recording of shock signals of debris flowaccurately.(2) According to vested classification of debris flow,5slurry viscosities of0Pa·s,0.13Pa·s,0.37P Pa·s,0.72Pa·s and0.93Pa·s,5solid phase ratios of0,0.02,0.04,0.08,and0.16,5grain diameters of0,0.075~0.2cm,0.3~0.5cm,0.8~1.0cm and1.3~1.5cmare selected to draft85group working conditions by the orthogonal combination ofthem. The slurry viscosity is prepared in laboratory with amylaceous ether, cooked rubber power and water, and the different diameter’s grain adopt coarse sand, pea gravel,middling gravel and big gravel.(3) Kinematic and impacting process under different working conditions arerecorded by high-speed camera, then, the kinematic characteristics of debris flow underexperimental condition are analyzed by the distill of kineticc time, kinetic modality ofdebris flow head. Under experimental condition, intermittent stream(surge) phenomenaand debris flow head are prominent. And debris flow head is obvious when slurryviscosity is small, and shock of head is also the maxium in whole impacting process,while, with the increase of slurry viscosity, the debris flow head getting more and moreunconspicuous, and disapper when slurry viscosity arrives0.72Pa·s. According to this,debris flow head modality can be divided to high-speed-intensity form, transition formand low-speed-intensity form. Velocity of debris flow has a positive correlation with theincrease of slurry viscosity, solid phase ratio and grain diameter, and its value has avariation during1~3m/s.(4) More than five million shock data of debris flow are obtained in those85groupworking condition, while, in order to get relative true signal, db wavelet is selected to do8layers soft threshold denose based on wavelet analysis theory. After analyzing thestatistic characteristics of denosied signal, the characteristics of impacting signal’sprobability density is gained. It shows that the probability density curve of impactingforce obeys Gauss distribution and the R-square can reach to9%, the RMSE is less than1%. Meanwhile, the modality of the probability density curve can be divided tounimodal distribution form(corresponding to the combination of low slurry viscosityand small grain diameter condition), strange unimodal distribution form(correspondingto the combination of low solid phase ratio and grain diameter between0.3~0.5cm and0.8~1.0cm condition), multimodal distribution form (corresponding to the combinationof high solid phase ratioand big grain diameter condition) and hump-shaped pad blockform (corresponding to the combination of slurry viscosity bigger than0.37Pa·s, solidphase ratio bigger than0.08and grain diameter smaller than0.5cmcondition).(5) Debris flow impacting signals are reconstructed by wavelet decompose andwavelet reconstruction, and signals after be reconstructed are disposaled by Fast FourierTransform(FFT), then, the time-frequency curves of impacting signal are gained. Basedon this, the time-frenquency characteristics and energy features are analyzed.The resultsshows that impacting signals can be broken down to nine frequency range(0~1.958Hz,1.958~3.906Hz,3.906~7.812Hz,7.812~15.625Hz,15.625~31.25Hz,31.25~62.5Hz, 62.5~125Hz,125~250Hz,250~500Hz), and the energy distribution characteristics ofeach frequency range were calculated. The results show that, firstly, more than96percentages of the energy colleted at0~1.958Hz frequency range, and that means thereal signal is exist at low frequency range. Secondly, the impacting energy increaseswith the increase of the slurry viscosity under the same solid phase ration and graindiameter. Thirdly, when the grain diameter is1.3~1.5cm, an evident peak value comesforth in the impacting signal of debris flow under the same slurry viscosity condition.(6) Based on experimental data, mean value and wallop standard deviationofwallop under working condition are got, then, sensitivity analysis between slurryviscosity, solid phase ratio, grain diameter and wallop is done.The results show that,there is a nonlinear positive correlation between debris flow wallop and the increase ofslurry viscosity, phase ratio and grain diameter. While, when the slurry viscosity raisesto0.73Pa·s, increasing amplitude of wallop decreases clearly, and grain diameter raisesfrom middle gravel to large gravel, it has a biggest increasing amplitude.(7) Checking calculating mean velocity and wallop of debris flow gained in themodel test to vested empirical calculating formula, it shows velocity data got in testagrees with empirical velocity calculating formula and wallop data agrees withhydrodynamic method. Based on this, a new calculating method for wallop of debrisflow considering the alteration of slurry viscosity, solid phase ratio and grain diameter isestablished. New formula is proved by actual data in Dongchuan debris flow, and theresult shows it works effectively.The achievements are valuable in promoting the studies on fine Geomorphologicdevelopment of banks of debris flow valley, and the governance level of debris flowdisaster along highways. |
PDF Full Text Request