具有体积分数梯度的连通装置甲烷-空气爆炸特性数值模拟

为了研究具有体积分数梯度的连通装置内甲烷-空气爆炸特性,以60 L圆柱体容器和20 L圆柱体容器通过3 m长,截面为0.035 m×0.035 m的方形管道而连接形成的容器管道连通装置作为研究对象,利用Fluidyn软件对均一体积分数的连通装置以及具有体积分数梯度的连通装置中甲烷-空气爆炸的特性进行了数值模拟。结果表明:连通装置中甲烷的均一体积分数为6.517%～8.067%时,并由大容器中心点火工况时,最大爆炸压力、最大爆炸压力上升速率、最高温度和最大速度,以及这些爆炸参数达到最大值时的时刻值随体积分数的变化约呈线性关系;连通装置大容器甲烷体积分数6.0%体积分数梯度为2.0%～8.0%且大容器中心点火时,最大爆炸压力、最大爆炸压力上升速率、最高温度和最大速度随体积分数梯度总体上呈现先增大后减小趋势;大容器中心点火时,最大爆炸压力位于小容器,最大压力上升速率位于管道1或管道2,最大速度位于管道3,速度值可达400～600m/s。本研究可为连通装置内可燃气体爆炸事故防控提供理论指导。     

波纹管道阻火器内火焰传播的实验与数值模拟研究

对乙烯-空气预混火焰在波纹管道阻火器中的传播与淬熄过程进行了实验和数值模拟研究,实验结果显示:当乙烯接近当量浓度时,预混气体爆炸压力变化过程可分为4个阶段,等压燃烧阶段、缓慢上升阶段、快速上升阶段和压力振荡阶段;在爆炸过程中,由于反射压力波和火焰相互作用的影响,超压值出现多次振荡,压力振荡阶段一般可以持续数十毫秒;乙烯-空气火焰传播速度随管径增加、阻火单元波纹高度减小呈递增趋势,而且随着阻火单元厚度的增加,阻火器的阻火能力明显提高,可以更有效地使火焰淬熄。数值模拟结果显示:在管道封闭端点火后,火焰面呈半球形并以层流扩散的方式向四周传播;当火焰传播到管道壁面时,在管道壁面的约束作用下,火焰面发生变形,壁面附近的火焰逐渐超过了管道轴线附近的火焰,最后形成了“郁金香”状的火焰结构;当爆燃火焰经过阻火单元时,高温已燃气体被其吸收大量热量,同时在反应区产生的稀疏波作用下,气体温度逐渐降低、化学反应速率迅速减小,最终导致火焰被熄灭。通过模拟计算结果可以看出,在整个爆炸过程中,火焰传播速度与爆炸压力波动均较为明显。并提出了孔隙率和阻火单元厚度对火焰传播的影响机制。基于传热学理论模型,并结合实验数据,得出了爆燃火焰速度与爆炸压力之间的关系,为工业装置阻火器的设计和选型提供更为准确的参考依据。     

Compressible gas-liquid mixture flow at abrupt pipe enlargements

A detailed investigation was made of the flow of compressible gas-liquid mixtures through sudden enlargements in diameter of circular pipes. One-dimensional analysis shows that the dimensionless pressure rise varies with mixture void fraction and mixture momentum, while the establishment of choking conditions at the enlargement is controlled by the length of pipe downstream in which frictional pipe flow occurs. The flows were found to exhibit two characteristic modes, jet flow and submerged flow, with intermediate flows displaying unsteady oscillation between these modes. The distance to the downstream position of maximum pressure increased steadily with mixture void fraction when the upstream pipe outlet was choked, varying from 5 to 50 times the downstream pipe diameter. If the flow was not choked, this distance was much smaller and showed discrete fixed values associated with the mode of flow.

One-dimensional analysis accurately predicted maximum pressure, but when flow was choked at the enlargement the calculation was sensitive to the pressure in the region of separated flow surrounding the central jet in the enlargement. Although analysis of maximum pressure in terms of flow expansion and normal shock gave a general indication of the maximum pressure (which was thus concluded to depend on the general flow processes expected in the enlargement), accurate prediction of maximum pressures will depend on empirical knowledge of the separated flow region pressures. The maximum pressure rise was found to be in the range extending down to 0.3 of the upstream pipe outlet pressure and reduced with void fraction; it was also influenced by the enlargement area ratio. Flows in the approach and outlet pipes were found to be compressible, frictional pipe flows of the Fanno type, with somewhat reduced friction factors occurring in the outlet pipe.     

Experimental distinction of damping mechanisms during hydraulic transients in pipe flow

The aim of the present paper is to contribute to the identification of the principal mechanical–hydraulical relationships during hydraulic transients by means of the analysis of observed transient pressure and strain waves in different pipe rigs. Four different experimental set-ups are analysed: a straight copper pipe with either moving or anchored downstream pipe-end, a coil copper pipe and a coil polyethylene pipe. Discussion highlights differences in the response of each system in terms of wave shape, damping, and dispersion. The straight copper pipe behaviour, for an anchored pipe-end, has shown the closest dynamic response to the expected one from classical waterhammer theory, being unsteady friction the most relevant damping mechanism. Fluid structure interaction dominates when the valve is released in the straight copper pipe and also has an important role in the coil copper pipe. Viscoelasticity dominates in the polyethylene pipe.     

Measurement of dispersed two-phase gas-liquid flow by cross correlation of modulated ultrasonic signals

A technique for metering two-phase gas-liquid mixtures is described. The measuring apparatus consists of a concentric constriction device and an ultrasonic cross-correlation velocity meter (Coulthard 1973). The constriction device breaks up the gaseous phase and generates a near-homogeneous type of flow, with characteristics largely independent of those existing in the approach section. The mean velocity of the resulting flow is obtained by cross-correlating the modulations of two parallel ultrasonic beams, transmitted across the pipe. The peak value of the cross-correlation function is found to occur at a time delay virtually equal to the mean transit time of the mixture between the centre lines of the beams.

The main advantages of this type of non-intrusive meter are its averaging capabilities and relative insensitivity to pressure transients.     

微型管中液滴流动的三维数值模拟

对于微型设备中的低雷诺数流动，毛细力和黏性力起主导作用. 应用相场方法，引 入自由能泛函，研究了二相流体在微型管中流动问题及表面浸润现象，并给出了微型管中二 相流体的无量纲输运方程. 针对方形微管道，利用差分法给出了输运方程的数值求解方法. 最后，模拟了方形直管中的液滴流动和变形的过程，并给出了液滴前后压力差与其它主要物 理参数之间的变化关系. 结果表明，压力差随液滴半径增大而增加，而随毛细管系数的增大 而减小.     

颗粒物质在竖直管中下行蠕动流动的实验测量

颗粒流蠕动行为是颗粒物质在竖直管中流动时常见的一种流动现象,其产生机理较复杂。为此,本文在在内径为150mm、高为5000mm的竖直管实验装置上,以FCC催化剂为固体颗粒物料,采用PV6型颗粒速度测量仪,测量不同颗粒流率下竖直管中的颗粒下行蠕动流动速度以及颗粒固含率,系统地考察了颗粒物质在竖直管中下行流动时的蠕动流动特性及产生机理。实验结果表明,颗粒物质在竖直管中下行流动时的流动行为可划分为两种形式。在颗粒流率较小时,颗粒物质下行速度呈现脉冲式变化,有速度停滞,可称之为蠕动I型流动。随着颗粒流率的增加,颗粒下行速度停顿消失,但仍是起伏变化,为蠕动II型流动。当颗粒流率增大到一定值后,颗粒物质下行蠕动行为消失,转变为流化流动。颗粒物质下行的蠕动行为是出口区颗粒成拱与崩塌、颗粒与器壁滑动摩擦和颗粒力链作用的综合反映。     

Simultaneous measurements of velocity and deformation in flows through compliant diaphragms

Flow through a circular orifice in a deformable diaphragm mounted in a pipe was studied experimentally as a simple yet suitable case for validating numerical fluid/structure interaction (FSI) codes including structures with significant deformation and strain. The flow was characterized using pressure taps, particle image velocimetry (PIV), and hot-film anemometry while deformation of the compliant diaphragm was determined directly from PIV images. The diaphragm material properties were measured independently by a uniaxial tensile testing machine. The diaphragm material modulus, orifice diameter, and pipe Reynolds number were varied over ranges appropriate for simulations of flows through heart valves. Pipe Reynolds numbers ranged from 600 (laminar upstream condition) to 8800 (turbulent upstream condition). The pressure drop across the diaphragm resulted in a concave deformation for all cases studied. For the range of Reynolds number tested, the Euler number decreased with increasing Reynolds number as a result of orifice expansion. The flow immediately downstream of compliant diaphragms was jet-like with strong inward radial velocity components and vena contracta. Laminar low Reynolds number flow (Re=600) through both rigid and compliant diaphragms yielded early and regular roll up of coherent vortex rings at a fixed frequency in contrast to turbulent higher Reynolds number flow (Re=3900), which yielded a broad range of vortex passage frequencies. Expansion of the compliant orifice for Re=3900 resulted in an initially broader slower jet with delayed shear layer development compared with the equivalent rigid case.     

Trapped water displacement from low sections of oil pipelines

The study is motivated by the problem of pipeline corrosion due to water accumulation at low spots. Lab-scale experiments were conducted to identify the critical conditions required for the onset of water displacement by oil flow from a low horizontal section into an upward inclined section of the pipeline. Two test loops with pipe diameters of 27 mm and 41 mm I.D. with diesel flow were used. Water withdrawal from tapping valves distributed along the up-hill section enabled to follow the water displacement for oil flow rates exceeding the critical value.     

Effect of surfactants on liquid loading in vertical wells

Most gas wells produce some amount of liquid. The liquid is either condensate or water. At high rates, the gas is able to entrain liquid to the surface; however, as gas well depletes, the liquid drops back in a gas well (called liquid loading) creating a back pressure on the reservoir formation. Addition of surfactants to the well to remove liquid is one of the common methods used in gas wells. Liquid loading in vertical gas wells with and without surfactant application was investigated in this study. Anionic, two types of amphoteric (amphoteric I and amphoteric II), sulphonate and cationic surfactants were tested in 2-inch and 4-inch 40-feet vertical pipes. Pressure gradient and liquid holdup are measured. Visual observation with a high speed camera was used to gain insight into the direction of foam flow in intermittent flow and foam film flow under annular flow conditions.Liquid loading is initiated when the liquid film attached to the wall in annular flow starts flowing downwards. Introduction of foam causes the gas velocity at which film reversal occurs to decrease; this shift increases with increasing surfactant concentration and it is more pronounced in 2-inch pipe than in 4-inch pipe. That is, the benefit of surfactants is much more pronounced in 2-inch pipe than in 4-inch pipe. The reason for postponement of liquid loading is reduction in the liquid holdup at low gas velocities which reduces the liquid holdup in foam flow compared to air-water flow. However, at higher gas velocities, the pressure drop in 2-inch compared to 4-inch pipe increases rapidly as the surfactant concentration increases. The selection of optimum concentration of the surfactant is a balance between the reductions in the gas velocity at which liquid loading occurs compared to increase in the frictional loss as the concentration increases. We provide guidelines about the selection of the surfactant concentration.Visual observations using high speed camera show differences in the behavior under foam flow conditions. Unlike air-water flow, the liquid film attached to the wall is replaced by thick foam capturing the gas bubbles. The type of roll waves which carry the liquid in 2-inch pipe is different than what was observed in 4-inch pipe. Compared to 4-inch pipe, the roll waves in 2-inch pipe are much thicker. This partly explains the differences in 2-inch versus 4-inch pipe behavior.     

Thermodynamic analytical model of a loop heat pipe

A thermodynamics analytical model is developed to explore different parameters effects on a loop heat pipe (LHP). The LHP is a two-phase device with extremely high effective thermal conductivity that utilizes the thermodynamic pressure difference to circulate a cooling fluid. The effects of pipe length, pipe diameter, condenser temperature, and heat load are reported. As pipe length increases and/or pipe diameter decreases, a higher temperature is expected in the evaporator.     

Interaction between an external compression shock and a blunt body in a hypersonic stream

A complex shock configuration with two triple points can occur during the interaction between an external oblique compression shock and the detached shock ahead of a blunt body (for instance, ahead of a wing or stabilizer edge). This results in the formation of a high-pressure, low-entropy supersonic gas jet [1–6]. Here two flow modes are possible [1], which differ substantially in the intensity of the thermal and dynamic effects of the stream on the blunt body: mode I corresponds to the impact of a supersonic jet [2–6], while the supersonic jet in mode II does not reach the body surface in the domain of shock interaction because of curvature under the effect of a pressure drop. Conditions for the realization of the above-mentioned flow modes are investigated experimentally and theoretically, and an approximate method is proposed to determine the magnitude of the compression shock standoff in the interaction domain. Blunt bodies with plane and cylindrical leading edges are examined. The results of a computation agree satisfactorily with experimental data.Translated from Izvestiya Akademii Nauk SSSR, Mekhanika Zhidkosti i Gaza, No. 1, pp. 97–103, January–February, 1976.The author is grateful to V. V. Lunev for discussing the research and for useful remarks.     

NUMERICAL SIMULATION OF THREE-DIMENSIONAL SHOCK-SHOCK INTERACTIONS ON A BLUNT BODY

A numerical study is conducted to simulate the effects of extraneous shock impingement on a blunt body in viscous hypersonic flow. The interaction of extraneous shock with the leading-edge shock results in a very complex flow field that contains local regions of high pressure and intense heating. The heating and pressure can be orders of magnitude higher than the peak values in the absence of shock impingement. The flow field is calculated by solving thin-layer Navier-Stokes equations with a finite-volume flux splitting technique developed by van Leer. For a zero or small sweep of the body, a type IV interaction occurs, which produces a lambda shock structure with a supersonic jet embedded in the otherwise subsonic flow; for a moderate sweep of about 25^°, a type V interaction occurs in which a subsonic shear layer sandwiched in supersonic flow is produced with a transmitted shock. In the present study, both type IV and type V interactions are investigated. Results of the present numerical investigation are compared with available experimental results. For the present conditions, the peak pressure is 2.2 times the unimpinged stagnation point pressure and the peak heating is 3 times the unimpinged stagnation point heating. The flow for a type IV interaction is found to be unsteady.     

The formation of hydrodynamic slugs by the interaction of waves in gas–liquid two-phase pipe flow

This paper examines the effects of wave interaction on the formation of hydrodynamic slugs in two-phase pipe flow at relatively low gas and liquid superficial velocities. The experiments were conducted using a horizontal 31 m long, D = 10 cm internal diameter transparent pipe at atmospheric pressure. High resolution photography allowed the location of the gas–liquid interface to be measured with a high degree of accuracy at 5 Hz. Image analysis allowed individual waves to be tracked over a 14D section of the pipe. Regular waves having similar properties such as speed, amplitude and length were seen far from the region of slug formation. However, near the transition region, where hydrodynamic slugs were formed, significant differences between wave properties were observed which resulted in wave interaction leading to a type of sub-harmonic resonance and slug formation. The formation of hydrodynamic slugs due to wave interaction differs from predictions for slug formation using long wavelength stability theory. The properties of the waves were quantified which gave detailed information on the resonance mechanism found near the transition to slugging.     

The energy transfer mechanism of a perturbed solid-body rotation flow in a rotating pipe

Three-dimensional direct numerical simulations of a solid-body rotation superposed on a uniform axial flow entering a rotating constant-area pipe of finite length are presented. Steady in time profiles of the radial, axial, and circumferential velocities are imposed at the pipe inlet. Convective boundary conditions are imposed at the pipe outlet.The Wang and Rusak(Phys. Fluids 8:1007–1016, 1996.doi:10.1063/1.86882) axisymmetric instability mechanism is retrieved at certain operational conditions in terms of incoming flow swirl levels and the Reynolds number. However, at other operational conditions there exists a dominant,three-dimensional spiral type of instability mode that is consistent with the linear stability theory of Wang et al.(J. Fluid Mech. 797: 284–321, 2016). The growth of this mode leads to a spiral type of flow roll-up that subsequently nonlinearly saturates on a large amplitude rotating spiral wave. The energy transfer mechanism between the bulk of the flow and the perturbations is studied by the Reynolds-Orr equation. The production or loss of the perturbation kinetic energy is combined of three components: the viscous loss, the convective loss at the pipe outlet, and the gain of energy at the outlet through the work done by the pressure perturbation. Theenergy transfer in the nonlinear stage is shown to be a natural extension of the linear stage with a nonlinear saturated process.     

无限均质弹性体中圆片裂纹受均布载荷时的超奇异积分方程封闭解

本文运用一种特殊技巧将一个受均布压力作用的圆片裂纹超奇异积分方程化为Ａｂｅｌ积分形式，从而可获得超奇异积分方程中未知位移间断的封闭解。再利用这个封闭解和应力强度因子的定义，得到了一个无限弹性体中受均布载荷分布时圆片裂纹前沿Ｉ型应力强度因子的精确表达式。所得到的结果与现有解完全相同。     

Study of the Damping Mechanisms of Loose Spring Skirts on Vibrating Pipes

Damping of the loose spring skirt (LSS) of a vibrating pipe under operating conditions was studied experimentally and numerically. Motion images made using a high speed charge-coupled device (CCD) camera as well as acceleration signals were analyzed in order to correlate the physical states of the LSS pipe system with the vibration transmissibility (TM) through the pipe. Experiments on the pipe undergoing harmonic excitation demonstrate that the damping of the LSS is caused by the same mechanism as with a conventional impact damper. Measurements of the acceleration TM as a function of excitation amplitude verified that the momentum exchange between the pipe and the auxiliary spring is the cause of the damping produced by the LSS pipe. Another finding of this research is that the experimental representation of the resonance shift phenomenon is related to two main operating conditions, namely: 1) the driving frequency and 2) the excitation amplitude. This research demonstrated the occurrence of the resonance shift phenomenon through experimental mapping of the system response at increasing excitation amplitudes using frequency sweep tests. Simulations of the LSS pipe system based on principal coordinate and impact damping analysis were performed in order to investigate and provide better explanations for LSS pipe system damping mechanisms. As a result of this research study, a new damping mechanism is proposed which is different from previous damping models. This research study has also addressed more practical boundary and excitation conditions for LSS pipe systems.     

矩形管道中微米级铝粉爆炸实验

在矩形管道粉尘爆炸装置中开展系列实验,系统研究了点火延迟时间、粉尘粒度及粉尘浓度对铝粉尘爆炸过程中最大爆炸压力和最大爆炸压力上升速率的影响。研究结果表明:不同的点火延迟时间对铝粉尘爆炸压力有显著影响,随着点火延迟时间由小变大,最大爆炸压力和最大爆炸压力上升速率呈现先增大后减小的趋势,且不同粒径的铝粉尘最大爆炸压力对应有不同点火延迟时间。随铝粉粒度的减小,最大爆炸压力和最大爆炸压力上升速率会呈现出先增大后减小的变化规律。铝粉最大爆炸压力和最大爆炸压力上升速率随浓度的增加均表现为先变大后减小的趋势,即铝粉浓度在特定数值时会使其爆炸威力最强。     

Investigation of a flat-plate oscillating heat pipe with Tesla-type check valves

Tesla-type check valves were integrated into a flat-plate oscillating heat pipe (FP-OHP) in order to promote and sustain a desired circulatory flow to increase overall thermal performance. Using neutron radiography, gray-scale images capturing the internal flow behavior within two bottom-heated copper FP-OHPs - one with Tesla-type valves (TV FP-OHP) and one without - both charged with water at a filling ratio of 70% - were collected. With the Tesla-type valves installed in the adiabatic section of the TV FP-OHP, it was found that circulation in the desired direction was promoted and that this promotion increased with heat input. The TV FP-OHP consistently possessed a lower thermal resistance than its counterpart without check valves. The percent-reduction in thermal resistance was on-the-order of 15-25% depending on the power input. Implementation of Tesla-type check valves is a promising means for circulatory flow rectification within an OHP, but future research is needed to further optimize valve design, quantity, and alignment.     

Long-Term Strength of a Thick-Walled Pipe Filled with an Aggressive Medium,with Account for Damageability of the Pipe Material and Residual Strength

This paper describes the study of scattered fracture of a thick-walled pipe filled with an aggressive medium, which creates uniform pressure on the inner surface of the pipe. It is assumed that the aggressive medium affects only the value of instantaneous strength. Damageability is described by an integral operator of the hereditary type. The problem is solved with allowance for residual strength of the pipe material behind the fracture front. Numerical calculation is carried out, and relationships between the fracture front coordinate and time for various concentrations of the aggressive medium and residual strength behind the fracture front are constructed.