多元可燃气体爆炸极限理论预测模型研究*

为准确掌握和预测多元可燃气体的爆炸极限,开展2种多元可燃气体爆炸极限的理论预测模型研究。第1种模型针对“多种可燃气体+多种惰性气体”在空气中或氧气中混合,基于求解可燃气体绝热火焰温度的总比热特性方法以及化学平衡反应中的贫燃料(富氧)反应,提出该多元可燃气体的爆炸下限预测模型;第2种模型针对“可燃气体+惰性气体+氧气”混合,基于热平衡方程及混合气体的各组分浓度、淬灭电势及燃烧潜热,提出该多元可燃气体的爆炸极限预测模型。结果表明:在预测多元可燃气体的爆炸极限时,第1种模型具有较广泛的应用性,且表现出较高的准确度;第2种模型具有使用简单的特点,且扩展了LCR（勒夏特列原理）的应用范围。     

Vessel outflow sensitivity to composition

The vessel composition is important when considering vessel outflow because changes in composition change the density and potentially more importantly change the shape and location of the boundary of the two-phase envelope. The influence on the phase boundary can be significant on vessel outflow as the liquid phase density is typically two orders of magnitude larger than the gas phase density for pressures and temperatures remote from the thermodynamic critical point. In this article two issues are addressed, outflow sensitivity to composition of hydrocarbon systems with a large component of methane, and outflow from systems where a petroleum fraction representation of composition is used. It is shown that approximating multi-component systems with a large methane content by methane can result in significant discrepancies in the predicted mass flow rate and vessel pressure. When a vessels composition is characterised by petroleum fractions, and the Kesler–Lee equations are used to predict thermodynamic properties, the predicted outflow is almost comparable to the predicted outflow calculated using a mole fraction representation of composition.     

基于空位溶液与DA理论的多组分气体吸附方法及验证*

为准确预测煤层气中多组分气体的吸附性能,将空位溶液与Dubinbin-Astakhov（DA）理论相结合,提出1种适用于煤层气吸附系统的多组分混合吸附模型。在模型中,吸附体系被视为气体与假设的“空位溶质”的多元混合物,吸附体系视为气相和吸附相空位溶质之间的平衡。结果表明:根据单组分气体吸附等温线,采用D-A方程计算分析了生成二元（吸附质+空位）混合物中纯组分气体的活度系数,并优化吸附参数;模型能够根据在单一温度下收集的纯组分吸附数据预测不同温度下的多组分气体吸附;模型计算结果与实测结果吻合良好,误差在10%以内,充分说明模型是切实可靠的。     

Numerical study on the initial expansion of two-phase cloud from an instantaneous release

In industries some dangerous liquefied gases may accidentally release and it may form a flammable or toxic mixture after mixing with air. One tool that is being developed in industry for two-phase cloud dispersion modeling is computational fluid dynamics (CFD). In this paper, the dispersion processes of different dangerous materials including liquefied chlorine, liquefied ammonia and liquefied petroleum gas were simulated in the same condition to analyze the characteristics of the initial expansion processes by CFD tool. The heat and mass transfer between droplets and the vapor after an instantaneous release event was calculated by using the Eulerian–Lagrangian method. The results from a number of 3-D CFD based studies were compared with the available small-scale experimental results. The results show that the present model and numerical simulation are reliable.     

A simplified steady-state model for air,water and steam curtains

Curtain mitigation systems are modeled here since they have experimentally shown their efficiency in reducing the concentration of certain toxic gases within dense gas clouds. Air, water and steam are analyzed in a model as the physical barriers to decrease the gas concentration. The model, developed for a steady-state mitigation process, is based on the mass, energy and momentum conservation laws. Concentration estimations during the dispersion before and after the mitigation are performed with a SLAB type model. A sensitivity analysis for each model is given to detect which variables have bigger effects. A release of chlorine is used as an example and the results are calculated in a prototype developed in Visual C++, where the model is solved using the Runge–Kutta 4th order method. The results include the effects of composition, speed, temperature and height of the releasing point as well as a comparison with CFD simulations. The proposed model is simplified and it cannot reproduce eddy effects but it is fast and robust enough. The model provides a set of equations that can be used in numerical problems where explicit derivatives are required, e.g. optimizations procedures.     

Steady-state critical two-phase flashing flow with possible multiple choking phenomenon: Part 1: Physical modelling and numerical procedure

In this paper, the modelling of pure fluid steady-state adiabatic flashing flow through a pipe involving an abrupt enlargement is presented. The thermal non-equilibrium two-phase flow model DEM has been adjusted for subcooled inlet conditions close to saturation and/or for inlet two-phase mixture state. The multichoking flow phenomenon occurs when two basic criteria are simultaneously fulfilled. A general procedure taking into account this possible occurrence is developed on the basis of two iterative algorithms. The first algorithm is applied to the mass flow rate upstream from the enlargement and the second algorithm is based on the length of the pipe downstream from the enlargement. The proposed methodology involves the improved physical one-dimensional two-phase flow model DEM and the global non-equilibrium model through the abrupt enlargement.     

Modelling and control of the dispersion of hazardous heavy gases

Dispersion of several common `heavy' gases (ethylene, propylene, ammonia, and chlorine) has been modelled on the basis of modifications in plume path theory. The model takes into account, among other things, the variations in temperature, density, and specific heat during the movement of the heavy gas plume. The effects of wind speed, density of the gas, and venting speed on the plume dispersion have been simulated. Based on the simulations a set of empirical equations has been developed. The equations have been validated by theoretical as well as experimental studies.Studies have also been carried out to simulate the effect of venting speed (manipulated by injecting hot air with the released gas) on the plume dispersion. The study reveals that the effect of venting speed on dispersion is very pronounced and can be used to reduce the risk posed by the accidental luxurious release of toxic/flammable gases. For example an increase of 20% in venting speed of chlorine (54.1 m/s) can reduce the distance up to which toxic concentration would occur by about 1100 meters.     

Experiments and numerical simulation on methane flame quenching by water mist

The study of extinguishment using water mist has been motivated due to the phase-out of the use of halens and the search for alternative means that preserve all of the benefits of a clean total flooding agent without adverse environmental impact. With the numerical simulation, we analyzed a gas–liquid two-phase problem including water (liquid), air and methane (gas) using Eulerian equations for the liquid phase and the full Navier–Stokes equations for the gas phase. Gaseous mass, momentum and energy equations are integrated simultaneously by a Harten–Yee explicit non-MUSCL modified-flux type TVD scheme for the convective terms and a central difference scheme for the viscous terms. Liquid phase conservation equations are solved with an application of a flux-vector-splitting scheme. In the experiments in an open room (500×500×500 mm) we observed an interaction of the diffusion flame with the water mists. The results show remarkable flame quenching and a good agreement between the numerical and experimental results.     

A New Homogeneous Non-Equilibrium Model to Compute Vapor-Liquid Two-Phase Critical Pressure Ratios of Multicomponent Hydrocarbon Mixtures

The critical pressure ratio (η_c) is an essential parameter for computing the vapor-liquid two-phase critical pressure and mass flow rate of multicomponent hydrocarbon mixtures flowing through valves and leakage orifices. The Homogeneous Non-Equilibrium Diener-Schmidt (HNE-DS) model widely used to calculate η_c assumes that the fluid's volume linearly changes with the pressure (using the Clausius-Clapeyron equation), which is not suitable for multicomponent gas mixtures. In this paper, a new Homogeneous Non-Equilibrium (new-HNE) model is proposed to calculate η_c of gas mixtures. Firstly, a new critical flow compressibility factor (ω_c) is developed from its thermodynamic definition and the Peng-Robinson equation of state (EOS), overcoming the inherent limitations of the Clausius-Clapeyron equation. Then, η_c is correlated to the newly derived ω_c by fitting experimental data at various pressures and gas mass fractions of both single-component and multicomponent gas mixtures, yielding the new HNE-DS model. Results show that, for the water-steam and air-water two-phase flow, the average relative deviations (ARD) between the calculated critical pressure ratios and experimental values are equal to 2.8% and 4.93%, respectively, which represents a significant improvement in comparison with the original HNE-DS model. Moreover, this new model is extended to the applications of Liquefied natural gas (LNG)/liquefied petroleum gas (LPG) fluids, and will further contribute to the calculation of the leakage mass flow rate of fluid flowing through the orifices/valves.     

Numerical study of choked two-phase flow of hydrocarbons fluids through orifices

Valves and orifices are the most widely devices of flow control used in oil and gas industry. In particular, they are installed in relief piping system in order to control the discharge flow during potential plant overpressuring scenarios, thus ensuring plant safety. It is a common practice to flow liquid and gas mixtures through such restriction devices.Rigorous models are available to precisely size pressure relief devices operating in single phase flow; however for two-phase flow, no models are considered sufficiently reliable for predicting the relevant flow conditions.In the present paper, two-phase flow of hydrocarbons fluids through an orifice under critical conditions has been numerically investigated.The existing literature has been analyzed and data on two-phase flow of highly volatile mixtures of hydrocarbons through openings have been collected. A comparison has been carried out with numerical simulations carried out by the multiphase flow simulation tool OLGA by SPT.The Henry–Fauske model has been used as orifice choke model and the orifice discharge flow coefficient, required as input by OLGA, has been calculated by Chisholm's model.Comparison between OLGA's results and experimental data shows that Henry–Fauske model markedly underestimates the mass flow rate through the orifice, if Chisholm's model is used to calculate discharge coefficient. It was found that the error of the model could be minimized using different values of orifice discharge coefficient (C_d).A new discharge flow coefficient model, suitable for choked two-phase flow across orifices, is proposed in this study and it has been determined to match the above mentioned experimental measurements.     

Small-scale experimental study of vaporization flux of liquid nitrogen released on ice

One of the LNG accident scenarios is the collision of an LNG carrier on an iceberg during marine transportation. A collision can result in damages to the vessel and lead to the leakage of the contents on ice or an ice-water mixture. When cryogenic liquid comes in contact with ice, it undergoes rapid vaporization due to the difference in temperature between the ice and cryogenic liquid. This process is different from the heat transfer between water and cryogenic liquid as ice is a solid and thus heat transfer to the pool occurs primarily through conduction. In this paper, the heat transfer phenomenon between ice and cryogenic liquid was studied through a small-scale experiment and the resulting vaporization mass fluxes were reported. The experiment involved six spills with varying amount of liquid nitrogen on different ice temperature to determine its effect on vaporization mass flux. The vaporization mass fluxes were determined by direct measurement of the mass loss during the experiment. The results indicated that the vaporization mass flux was a function of release rate and ice temperature. When the release rate and ice temperature was high, the vaporization mass flux follows a decreasing trend. With further reduction in release rate and ice temperature, the vaporization mass flux was found to be independent with time. The one dimensional conduction model was validated against experimental results. The predicted temperatures and heat flux were found to be in good agreement with the experimental data.     

多元混合气爆炸极限的非线性预测研究

根据混合气的爆炸极限与混合气各成分的体积浓度之间具有非线性关系的特点,笔者提出采用神经网络非线性方法来计算含有H2,CH4和CO的多元混合气体的爆炸极限。在模型中,H2,CH4和CO的体积浓度作为输入,爆炸上限和下限作为输出。计算结果表明,该非线性模型预测混合气爆炸下限和上限的最大相对误差为3.90%,3.57%,而模型预测值与计算值的相关系数分别为0.971,0.981;非线性模型的预测结果要好于偏最小二乘回归的预测结果。当H2,CO,CH4在混合气中的体积浓度给定时,非线性模型能够准确预测混合气的爆炸极限。     

Thermodynamic investigation and hydrate inhibition of real gas flow through orifice during depressurization

A thermodynamic procedure has been proposed which can be used to predict the gas pressure, temperature and flow rate through orifice upon chock flow condition, using equation of state (EOS). The procedure applied for emergency depressurization operation incorporating the Peng-Robinson EOS and validated by comparing flow rates of a multi-component hydrocarbon gas mixture for thirteen experimental cases. The average absolute deviations of the predicted flow rates for orifice discharge coefficients of 0.85 and 0.9, are 7.36% and 2.03%, respectively. The corresponding error for API 520 (American Petroleum Institute Recommendation Practice 520) method is 6.91%. In this work, the hydrate formation temperature and hydrate inhibitor type and its required weight fraction for preventing the hydrate formation upon orifice and its upstream conditions are evaluated by the EZ-Thermo software using the Moshfeghian–Maddox method. The results qualitatively show that the hydrate prevention is essential for the safety of the operation due to low temperature condition.     

液氮泡沫发生器内流动特性研究

低温液氮与泡沫混合液直接接触产生氮气泡沫是一种新型的掺混形式,利用液氮高汽化比的特点,搭建液氮泡沫可视化实验装置,进行氮气-水两相流及液氮泡沫流动特性的研究。结果表明,液氮相变产生大量氮气,其与泡沫液混合产生泡沫,温度有所回升,最终趋于泡沫混合液温度;管路沿程压降较小;液氮射流破碎及流动过程可分为6个区域:低温液氮区、向上循环翻滚区、滞留区、泡沫与泡沫混合液混合区、致密泡沫区、泡沫混合液区。流体向下游流动过程中持续发泡;为防止管路结冰,需合理控制泡沫混合液与液氮流量。     

Development of a mitigation system against hydrogen-air deflagrations in nuclear power plants

A novel mitigation system against hydrogen-air deflagrations in nuclear power plant buildings is proposed and developed through a series of field experiments using explosion vessels of different volume sizes. The mitigation system is installed on the outer surface of the vessels, and it comprises flame arrester and explosion air bag. The flame arrester is made by stacking 10–20 sheets of fine-mesh wire screens, and the air bag is connected for holding explosion gas. The successful mitigation mechanism is the sequence of pressure-rise reduction by the air bag expansion, flame quenching by the flame arrester, and the slow burning of the gas mixture sucked from the air bag back into the vessel due to the negative pressure caused by the rapid condensation of water vapor inside the vessel. Necessary conditions for the successful mitigation system are discussed, and the practical unit size of flame arrester sheet is recommended.     

Revised emergency vacuum relief device sizing for atmospheric distillation systems

Conventional vacuum relief methodologies are usually protective responses; that is, they accomplish their purpose by substitution of an inert gas (usually nitrogen) for the process gases removed by an external vacuum source, or for condensable vapour collapsed by an internal process mechanism (e.g. condensation). While this approach is theoretically possible for all potential vacuum scenarios, it becomes practically impossible to implement for installations where a rapid phase change can impart near-instantaneous system pressure reductions. The procedure outlined in this paper takes a preventive approach: eliminate the source of vacuum generation before the safe lower system pressure limit is reached. For distillation and other refluxing systems, this vacuum source is usually the main overhead condenser, which is designed to collapse large volumes of condensable vapour. To eliminate the vacuum source requires elimination of the system's ability to rapidly condense vapour. This goal is accomplished by introduction of inert gas directly into the condensing system to ‘blanket’ the heat transfer surface and stop condensation. The procedure determines the rate, amount and location for introduction of inert gas. The required design data include: (i) system starting pressure, (ii) maximum allowable system vacuum, (iii) volume of the condensing system, and (iv) normal system condensing rate. By determining the rate at which the condenser removes vapour volume from the system, and designing an inert gas delivery system to meet or exceed this rate, the vacuum generation potential of the system is effectively eliminated using a much smaller quantity of inert gas than with the more traditional volume substitution methods.     

Fast and tiny: A model for the flame propagation of nanopowders

To avoid the influence of external parameters, such as the vessel volume or the initial turbulence, the explosion severity should be determined from intrinsic properties of the fuel-air mixture. Therefore, the flame propagation of gaseous mixtures is often studied in order to estimate their laminar burning velocity, which is both independent of external factors and a useful input for CFD simulation. Experimentally, this parameter is difficult to evaluate when it comes to dust explosion, due to the inherent turbulence during the dispersion of the cloud. However, the low inertia of nanoparticles allows performing tests at very low turbulence without sedimentation. Knowledge on flame propagation concerning nanoparticles may then be modelled and, under certain conditions, extrapolated to microparticles, for which an experimental measurement is a delicate task. This work focuses on a nanocellulose with primary fiber dimensions of 3 nm width and 70 nm length. A one-dimensional model was developed to estimate the flame velocity of a nanocellulose explosion, based on an existing model already validated for hybrid mixtures of gas and carbonaceous nanopowders similar to soot. Assuming the fast devolatilization of organic nanopowders, the chemical reactions considered are limited to the combustion of the pyrolysis gases. The finite volume method was used to solve the mass and energy balances equations and mass reactions rates constituting the numerical system. Finally, the radiative heat transfer was also considered, highlighting the influence of the total surface area of the particles on the thermal radiation. Flame velocities of nanocellulose from 17.5 to 20.8 cm/s were obtained numerically depending on the radiative heat transfer, which proves a good agreement with the values around 21 cm/s measured experimentally by flame visualization and allows the validation of the model for nanoparticles.