Road pavements in tunnels are usually made of asphalt mixtures,which,unfortunately,are flammable materials.Hence,this type of pavement could release heat,and more specifically smoke,in the event of a tunnel fire,there...Road pavements in tunnels are usually made of asphalt mixtures,which,unfortunately,are flammable materials.Hence,this type of pavement could release heat,and more specifically smoke,in the event of a tunnel fire,thereby worsening the environmental conditions for human health.Extensive research has been conducted in recent years to enhance the fire reaction of traditional asphalt mixtures for the road pavements used in tunnels.The addition of the Flame Retardants(FRs)in conventional asphalt mixtures appears to be promising.Nevertheless,the potential effects of the FRs in terms of the reduction in consequences on tunnel users in the event of a large fire do not seem to have been sufficiently investigated by using fluid dynamics analysis as a computational tool.Given this gap of knowledge,this article aims to quantitatively evaluate whether the use of flame-retarded asphalt mixtures,as opposed to traditional ones without FRs,might mitigate the adverse effects on the safety of evacuees and fire brigade by performing numerical analyses in the case of a tunnel fire.To achieve this goal,3D Computational Fluid Dynamics(CFD)models,which were executed using the Fire Dynamics Simulator(FDS)tool,were established in the case of a major fire of a Heavy Goods Vehicle(HGV)characterized by a maximum Heat Release Rate(HRRmax)of 100 MW.The people evacuation process was also simulated,and the Evac tool was used.Compared to the traditional asphalt pavements without FRs,the simulation findings indicated that the addition of the FRs causes a reduction in CO and CO_(2)levels in the tunnel during the aforementioned fire,with a minor number of evacuees being exposed to the risk of incapacity to self-evacuate,as well as certain safety benefits for the operability of the firefighters entering the tunnel downstream of the fire when the tunnel is naturally ventilated.展开更多
Oil spills continue to generate various issues and concerns regarding their effect and behavior in the marine environment,owing to the related potential for detrimental environmental,economic and social implications.I...Oil spills continue to generate various issues and concerns regarding their effect and behavior in the marine environment,owing to the related potential for detrimental environmental,economic and social implications.It is essential to have a solid understanding of the ways in which oil interacts with the water and the coastal ecosystems that are located nearby.This study proposes a simplified model for predicting the plume-like transport behavior of heavy Bunker C fuel oil discharging downward from an acutely-angled broken pipeline located on the water surface.The results show that the spill overall profile is articulated in three major flow areas.The first,is the source field,i.e.,a region near the origin of the initial jet,followed by the intermediate or transport field,namely,the region where the jet oil flow transitions into an underwater oil plume flow and starts to move horizontally,and finally,the far-field,where the oil re-surface and spreads onto the shore at a significant distance from the spill site.The behavior of the oil in the intermediate field is investigated using a simplified injection-type oil spill model capable of mimicking the undersea trapping and lateral migration of an oil plume originating from a negatively buoyant jet spill.A rectangular domain with proper boundary conditions is used to implement the model.The Projection approach is used to discretize a modified version of the Navier-Stokes equations in two dimensions.A benchmark fluid flow issue is used to verify the model and the results indicate a reasonable relationship between specific gravity and depth as well as agreement with the aerial data and a vertical temperature profile plot.展开更多
The airflow and dispersion of a pollutant in a complex urban area of Beijing, China, were numerically examined by coupling a Computational Fluid Dynamics （CFD） model with a mesoscale weather model. The models used w...The airflow and dispersion of a pollutant in a complex urban area of Beijing, China, were numerically examined by coupling a Computational Fluid Dynamics （CFD） model with a mesoscale weather model. The models used were Open Source Field Operation and Manipulation （OpenFOAM） software package and Weather Research and Forecasting （WRF） model. OpenFOAM was firstly validated against wind-tunnel experiment data. Then, the WRF model was integrated for 42 h starting from 0800 LST 08 September 2009, and the coupled model was used to compute the flow fields at 1000 LST and 1400 LST 09 September 2009. During the WRF-simulated period, a high pressure system was dominant over the Beijing area. The WRF-simulated local circulations were characterized by mountain valley winds, which matched well with observations. Results from the coupled model simulation demonstrated that the airflows around actual buildings were quite different from the ambient wind on the boundary provided by the WRF model, and the pollutant dispersion pattern was complicated under the influence of buildings. A higher concentration level of the pollutant near the surface was found in both the step-down and step-up notches, but the reason for this higher level in each configurations was different： in the former, it was caused by weaker vertical flow, while in the latter it was caused by a downward-shifted vortex. Overall, the results of this study suggest that the coupled WRF-OpenFOAM model is an important tool that can be used for studying and predicting urban flow and dispersions in densely built-up areas.展开更多
A CFD code has been developed based on the conservation principles describing gas and solid flow in fluidized beds. This code is employed to simulate not only the spatiotemporal gas and solid phase velocities and v...A CFD code has been developed based on the conservation principles describing gas and solid flow in fluidized beds. This code is employed to simulate not only the spatiotemporal gas and solid phase velocities and voidage profiles in a two dimensional bed but also fluid dynamics in the jet region. The computational results show that gas flow direction is upward in the entire bed accompanied with random local circulations, whilst solid flow direction is upward at the center and downward near the wall. The radical reason of strong back mixing of solid particles and good transfer behavior between two phases is that the jet entrains solid particles. Numerical calculation indicates that gas velocity, solid velocity and pressure profile have a significant change when the voidage is 0 8. The simulated time averaged voidage profiles agree with the experimental results and simulated data reported by Gidaspow and Ettehadieh(1983). Therefore, CFD model can be regarded as a useful tool to study the jet characteristics in dense gas solid fluidized beds.展开更多
The full alumina dissolution process in aluminum electrolysis cells was investigated using an improved computational fluid dynamics(CFD)model based on the previous researches by consideration of agglomerate formation....The full alumina dissolution process in aluminum electrolysis cells was investigated using an improved computational fluid dynamics(CFD)model based on the previous researches by consideration of agglomerate formation.The results show that the total mass of alumina agglomerate and its maximum size are mainly dependent on the feeding amount and increase with increasing it.Higher superheat can effectively inhibit the agglomerate formation and thus promote the full alumina dissolution behavior.The full alumina dissolution process mainly includes a fast stage and a slow stage,with an average dissolution rate of 17.24 kg/min and 1.53 kg/min,respectively.About 50%(mass percentage)of the total alumina particles,almost all of which are the well-dispersed alumina fine grains,dissolve within the fast dissolution stage of about 10 s.The maximum values of the average dissolution rate and final percentage of the cumulative dissolved alumina mass are obtained with a feeding amount of 1.8 kg for a superheat of 12℃.The formation of the alumina agglomerates and slow dissolution characteristics play a dominant role in the full dissolution of alumina particles.展开更多
CFD models have been developed to investigate the Iongwall goaf gas flow patternsunder different mining and geological control conditions.The Iongwall goaf wastreated as porous regions and gas flow was modelled as a m...CFD models have been developed to investigate the Iongwall goaf gas flow patternsunder different mining and geological control conditions.The Iongwall goaf wastreated as porous regions and gas flow was modelled as a momentum sink added to themomentum equation.Gas desorption from the caved goaf and destressed coal seamswithin the mining disturbed area was modelled as additional mass sources in the continuityequation.These CFD models were developed according to specific Iongwall layoutsand calibrated against field monitoring data.Two case studies were presented demonstratingthe application of CFD modelling of goaf gas flow characteristics for improved goafgas capture and the reduction of oxygen ingress into the goaf areas for self-heating prevention.Results from the case studies indicate that the optimum goaf drainage strategywould be a combination of shallow (near the face) and deep holes to improve the overalldrainage efficiency and gas purity.For gassy Iongwall faces retreating against the seam dip,it is recommended to conduct cross-measure roof hole drainage targeting the fracturedzones overlying the return corner,rather than high capacity surface goaf drainage deep inthe goaf.展开更多
In this study a mathematical model of a small scale single pellet for the oxidative coupling of methane（OCM）over titanite pervoskite is developed.The method is based on a computational fluid dynamics（CFD）code whic...In this study a mathematical model of a small scale single pellet for the oxidative coupling of methane（OCM）over titanite pervoskite is developed.The method is based on a computational fluid dynamics（CFD）code which known as Fluent may be adopted to model the reactions that take place inside the porous catalyst pellet.The steady state single pellet model is coupled with a kinetic model and the intra-pellet concentration profiles of species are provided.Subsequent to achieving this goal,a nonlinear reaction network consisting of nine catalytic reactions and one gas phase reaction as an external program is successfully implemented to CFD-code as a reaction term in solving the equations.This study is based on the experimental design which is conducted in a differential reactor with a Sn/BaTiO3 catalyst（7-8 mesh） at atmospheric pressure,GHSV of 12000 h-1,ratio of methane to oxygen of 2,and three different temperatures of 1023,1048 and 1073 K.The modeling results such as selectivity and conversion at the pellet exit are in good agreement with the experimental data.Therefore,it is suggested that to achieve high yield in OCM process the modeling of the single pellet should be considered as the heart of catalytic fixed bed reactor.展开更多
A modified particle bed model derived from the two-fluid momentum balance equations was employed to predict the gas-fluidised bed behaviour. Additional terms are included in both the fluid and the particle momentum ba...A modified particle bed model derived from the two-fluid momentum balance equations was employed to predict the gas-fluidised bed behaviour. Additional terms are included in both the fluid and the particle momentum balance equations to take into account the effect of the dispersed solid phase. This model has been extended to two-dimensional formulations and has been implemented in the commercial code CFX 4.3. The model correctly simulates the homogeneous fluidisation of Geldart Group A and the bubbling fluidisation of Geldart Group B in gas-solid fluidised beds.展开更多
It is highly important in Japan to choose a good site for wind turbines, because the spatial distribution of wind speed is quite complicated over steep complex terrain. We have been developing the unsteady numerical m...It is highly important in Japan to choose a good site for wind turbines, because the spatial distribution of wind speed is quite complicated over steep complex terrain. We have been developing the unsteady numerical model called the RIAM-COMPACT (Research Institute for Applied Mechanics, Kyushu University, Computational Prediction of Airflow over Complex Terrain). The RIAM-COMPACT is based on the LES (Large-Eddy Simulation). The object domain of the RIAM-COMPACT is from several m to several km, and can predict the airflow and gas diffusion over complex terrain with high precision. In the present paper, the design wind speed evaluation technique in wind turbine installation point by using the mesoscale meteorological model and RIAM-COMPACT CFD model was proposed. The design wind speed to be used for designing WTGs can be calculated by multiplying the ratio of the mean wind speed at the hub-height to the mean upper-air wind speed at the inflow boundary, i.e., the fractional increase of the mean hub-height wind speed, by the reduction ratio, R. The fractional increase of the mean hub-height wind speed was evaluated using the CFD simulation results. This method was proposed as Approach 1 in the present paper. A value of 61.9 m/s was obtained for the final design wind speed, Uh, in Approach 1. In the evaluation procedure of the design wind speed in Approach 2, neither the above-mentioned reduction rate, R, nor an upper-air wind speed of 1.7 Vo, where Vo is the reference wind speed, was used. Instead, the value of the maximum wind speed which was obtained from the typhoon simulation for each of the investigated wind directions was adopted. When the design wind speed was evaluated using the 50-year recurrence value, the design wind speed was 48.3 m/s. When a somewhat conservative safety factor was applied, that is, when the 100 year recurrence value was used instead, the design wind speed was 52.9 m/s.展开更多
This paper presents the fundamentals of direct inverse modeling using CFD simulations to detect air pollution sources in urban areas. Generally, there are four techniques used for detecting pollution sources: the anal...This paper presents the fundamentals of direct inverse modeling using CFD simulations to detect air pollution sources in urban areas. Generally, there are four techniques used for detecting pollution sources: the analytical technique, the optimization technique, the probabilistic technique, and the direct technique. The study discusses the potentialities and limits of each technique, where the direct inverse technique is focused. Two examples of applying the direct inverse technique in detecting pollution source are introduced. The difficulties of applying the direct inverse technique are investigated. The study reveals that the direct technique is a promising tool for detecting air pollution source in urban environments. However, more efforts are still needed to overcome the difficulties explained in the study.展开更多
为提升低空风切变预报精度,本文综合运用欧洲中期天气预报中心第五代再分析资料[European Centre for Medium-Range Weather Forecasts(ECMWF)fifth-generation reanalysis data,ERA5]和美国国家环境预报中心(National Centers for Envi...为提升低空风切变预报精度,本文综合运用欧洲中期天气预报中心第五代再分析资料[European Centre for Medium-Range Weather Forecasts(ECMWF)fifth-generation reanalysis data,ERA5]和美国国家环境预报中心(National Centers for Environmental Prediction,NCEP)的FNL全球再分析资料(Final Operational Global Analysis)、先进星载热发射和反射辐射仪全球数字高程模型以及兰州中川机场的实况观测资料,采用中尺度数值天气预报模式(Weather Research and Forecasting Model,WRF)、WRF结合计算流体动力学(Computational Fluid Dynamics,CFD)方法、长短期神经网络(Long Short-Term Memory,LSTM)方法,对2021年4月15-16日兰州中川机场的两次风切变过程进行模拟分析。结果表明:(1)在小于1 km的网格中使用大涡模拟,WRF模式在单个站点风速模拟任务中表现更好,但在近地面水平风场风速模拟效果上,不如WRF模式结合计算流体力学模型方案;(2)对于飞机降落过程中遭遇的两次低空风切变的模拟,WRF-LES和WRF-CFD两种模式都可以模拟出第一次低空风切变,而第二次受传入模式的WRF风速数据值较小的影响,两种模式风速差都没有达到阈值,需要在后续工作中进一步验证;(3)低风速条件(6 m·s^(-1))下,基于LSTM的单变量风速预测模型平均绝对误差基本维持在0.59 m·s^(-1),能较好地把握不同地形与环流背景条件下风速变化的非线性关系,虽然受到WRF误差和观测要素不全的限制,多变量风速预测能在保证平均绝对百分比误差小于6.60%的情况下,以更高的计算效率和泛化能力实现风速预测。本文不仅验证了WRF-CFD和WRF-LES耦合方案在风场和低空风切变预报中的差异,还探讨了基于LSTM的风速预测的可行性和准确性,期望为提高风场模拟精度,缩短精细风场模拟时间提供新的视角和方法。展开更多
文摘Road pavements in tunnels are usually made of asphalt mixtures,which,unfortunately,are flammable materials.Hence,this type of pavement could release heat,and more specifically smoke,in the event of a tunnel fire,thereby worsening the environmental conditions for human health.Extensive research has been conducted in recent years to enhance the fire reaction of traditional asphalt mixtures for the road pavements used in tunnels.The addition of the Flame Retardants(FRs)in conventional asphalt mixtures appears to be promising.Nevertheless,the potential effects of the FRs in terms of the reduction in consequences on tunnel users in the event of a large fire do not seem to have been sufficiently investigated by using fluid dynamics analysis as a computational tool.Given this gap of knowledge,this article aims to quantitatively evaluate whether the use of flame-retarded asphalt mixtures,as opposed to traditional ones without FRs,might mitigate the adverse effects on the safety of evacuees and fire brigade by performing numerical analyses in the case of a tunnel fire.To achieve this goal,3D Computational Fluid Dynamics(CFD)models,which were executed using the Fire Dynamics Simulator(FDS)tool,were established in the case of a major fire of a Heavy Goods Vehicle(HGV)characterized by a maximum Heat Release Rate(HRRmax)of 100 MW.The people evacuation process was also simulated,and the Evac tool was used.Compared to the traditional asphalt pavements without FRs,the simulation findings indicated that the addition of the FRs causes a reduction in CO and CO_(2)levels in the tunnel during the aforementioned fire,with a minor number of evacuees being exposed to the risk of incapacity to self-evacuate,as well as certain safety benefits for the operability of the firefighters entering the tunnel downstream of the fire when the tunnel is naturally ventilated.
文摘Oil spills continue to generate various issues and concerns regarding their effect and behavior in the marine environment,owing to the related potential for detrimental environmental,economic and social implications.It is essential to have a solid understanding of the ways in which oil interacts with the water and the coastal ecosystems that are located nearby.This study proposes a simplified model for predicting the plume-like transport behavior of heavy Bunker C fuel oil discharging downward from an acutely-angled broken pipeline located on the water surface.The results show that the spill overall profile is articulated in three major flow areas.The first,is the source field,i.e.,a region near the origin of the initial jet,followed by the intermediate or transport field,namely,the region where the jet oil flow transitions into an underwater oil plume flow and starts to move horizontally,and finally,the far-field,where the oil re-surface and spreads onto the shore at a significant distance from the spill site.The behavior of the oil in the intermediate field is investigated using a simplified injection-type oil spill model capable of mimicking the undersea trapping and lateral migration of an oil plume originating from a negatively buoyant jet spill.A rectangular domain with proper boundary conditions is used to implement the model.The Projection approach is used to discretize a modified version of the Navier-Stokes equations in two dimensions.A benchmark fluid flow issue is used to verify the model and the results indicate a reasonable relationship between specific gravity and depth as well as agreement with the aerial data and a vertical temperature profile plot.
基金supported by the Public Welfare Special Fund Program(Meteorology)of the Chinese Ministry of Finance under Grant No.GYHY201106033
文摘The airflow and dispersion of a pollutant in a complex urban area of Beijing, China, were numerically examined by coupling a Computational Fluid Dynamics （CFD） model with a mesoscale weather model. The models used were Open Source Field Operation and Manipulation （OpenFOAM） software package and Weather Research and Forecasting （WRF） model. OpenFOAM was firstly validated against wind-tunnel experiment data. Then, the WRF model was integrated for 42 h starting from 0800 LST 08 September 2009, and the coupled model was used to compute the flow fields at 1000 LST and 1400 LST 09 September 2009. During the WRF-simulated period, a high pressure system was dominant over the Beijing area. The WRF-simulated local circulations were characterized by mountain valley winds, which matched well with observations. Results from the coupled model simulation demonstrated that the airflows around actual buildings were quite different from the ambient wind on the boundary provided by the WRF model, and the pollutant dispersion pattern was complicated under the influence of buildings. A higher concentration level of the pollutant near the surface was found in both the step-down and step-up notches, but the reason for this higher level in each configurations was different： in the former, it was caused by weaker vertical flow, while in the latter it was caused by a downward-shifted vortex. Overall, the results of this study suggest that the coupled WRF-OpenFOAM model is an important tool that can be used for studying and predicting urban flow and dispersions in densely built-up areas.
文摘A CFD code has been developed based on the conservation principles describing gas and solid flow in fluidized beds. This code is employed to simulate not only the spatiotemporal gas and solid phase velocities and voidage profiles in a two dimensional bed but also fluid dynamics in the jet region. The computational results show that gas flow direction is upward in the entire bed accompanied with random local circulations, whilst solid flow direction is upward at the center and downward near the wall. The radical reason of strong back mixing of solid particles and good transfer behavior between two phases is that the jet entrains solid particles. Numerical calculation indicates that gas velocity, solid velocity and pressure profile have a significant change when the voidage is 0 8. The simulated time averaged voidage profiles agree with the experimental results and simulated data reported by Gidaspow and Ettehadieh(1983). Therefore, CFD model can be regarded as a useful tool to study the jet characteristics in dense gas solid fluidized beds.
基金financial supports from the National Natural Science Foundation of China (No. 51704126)the Natural Science Foundation of Jiangsu Province, China (No. BK20170551)Jiangsu Planned Projects for Postdoctoral Research Funds, China (No. 2019K046)。
文摘The full alumina dissolution process in aluminum electrolysis cells was investigated using an improved computational fluid dynamics(CFD)model based on the previous researches by consideration of agglomerate formation.The results show that the total mass of alumina agglomerate and its maximum size are mainly dependent on the feeding amount and increase with increasing it.Higher superheat can effectively inhibit the agglomerate formation and thus promote the full alumina dissolution behavior.The full alumina dissolution process mainly includes a fast stage and a slow stage,with an average dissolution rate of 17.24 kg/min and 1.53 kg/min,respectively.About 50%(mass percentage)of the total alumina particles,almost all of which are the well-dispersed alumina fine grains,dissolve within the fast dissolution stage of about 10 s.The maximum values of the average dissolution rate and final percentage of the cumulative dissolved alumina mass are obtained with a feeding amount of 1.8 kg for a superheat of 12℃.The formation of the alumina agglomerates and slow dissolution characteristics play a dominant role in the full dissolution of alumina particles.
文摘CFD models have been developed to investigate the Iongwall goaf gas flow patternsunder different mining and geological control conditions.The Iongwall goaf wastreated as porous regions and gas flow was modelled as a momentum sink added to themomentum equation.Gas desorption from the caved goaf and destressed coal seamswithin the mining disturbed area was modelled as additional mass sources in the continuityequation.These CFD models were developed according to specific Iongwall layoutsand calibrated against field monitoring data.Two case studies were presented demonstratingthe application of CFD modelling of goaf gas flow characteristics for improved goafgas capture and the reduction of oxygen ingress into the goaf areas for self-heating prevention.Results from the case studies indicate that the optimum goaf drainage strategywould be a combination of shallow (near the face) and deep holes to improve the overalldrainage efficiency and gas purity.For gassy Iongwall faces retreating against the seam dip,it is recommended to conduct cross-measure roof hole drainage targeting the fracturedzones overlying the return corner,rather than high capacity surface goaf drainage deep inthe goaf.
文摘In this study a mathematical model of a small scale single pellet for the oxidative coupling of methane（OCM）over titanite pervoskite is developed.The method is based on a computational fluid dynamics（CFD）code which known as Fluent may be adopted to model the reactions that take place inside the porous catalyst pellet.The steady state single pellet model is coupled with a kinetic model and the intra-pellet concentration profiles of species are provided.Subsequent to achieving this goal,a nonlinear reaction network consisting of nine catalytic reactions and one gas phase reaction as an external program is successfully implemented to CFD-code as a reaction term in solving the equations.This study is based on the experimental design which is conducted in a differential reactor with a Sn/BaTiO3 catalyst（7-8 mesh） at atmospheric pressure,GHSV of 12000 h-1,ratio of methane to oxygen of 2,and three different temperatures of 1023,1048 and 1073 K.The modeling results such as selectivity and conversion at the pellet exit are in good agreement with the experimental data.Therefore,it is suggested that to achieve high yield in OCM process the modeling of the single pellet should be considered as the heart of catalytic fixed bed reactor.
基金Supported by EU Comm ission(No.ENK5 - CT2 0 0 0 - 0 0 314 )
文摘A modified particle bed model derived from the two-fluid momentum balance equations was employed to predict the gas-fluidised bed behaviour. Additional terms are included in both the fluid and the particle momentum balance equations to take into account the effect of the dispersed solid phase. This model has been extended to two-dimensional formulations and has been implemented in the commercial code CFX 4.3. The model correctly simulates the homogeneous fluidisation of Geldart Group A and the bubbling fluidisation of Geldart Group B in gas-solid fluidised beds.
文摘It is highly important in Japan to choose a good site for wind turbines, because the spatial distribution of wind speed is quite complicated over steep complex terrain. We have been developing the unsteady numerical model called the RIAM-COMPACT (Research Institute for Applied Mechanics, Kyushu University, Computational Prediction of Airflow over Complex Terrain). The RIAM-COMPACT is based on the LES (Large-Eddy Simulation). The object domain of the RIAM-COMPACT is from several m to several km, and can predict the airflow and gas diffusion over complex terrain with high precision. In the present paper, the design wind speed evaluation technique in wind turbine installation point by using the mesoscale meteorological model and RIAM-COMPACT CFD model was proposed. The design wind speed to be used for designing WTGs can be calculated by multiplying the ratio of the mean wind speed at the hub-height to the mean upper-air wind speed at the inflow boundary, i.e., the fractional increase of the mean hub-height wind speed, by the reduction ratio, R. The fractional increase of the mean hub-height wind speed was evaluated using the CFD simulation results. This method was proposed as Approach 1 in the present paper. A value of 61.9 m/s was obtained for the final design wind speed, Uh, in Approach 1. In the evaluation procedure of the design wind speed in Approach 2, neither the above-mentioned reduction rate, R, nor an upper-air wind speed of 1.7 Vo, where Vo is the reference wind speed, was used. Instead, the value of the maximum wind speed which was obtained from the typhoon simulation for each of the investigated wind directions was adopted. When the design wind speed was evaluated using the 50-year recurrence value, the design wind speed was 48.3 m/s. When a somewhat conservative safety factor was applied, that is, when the 100 year recurrence value was used instead, the design wind speed was 52.9 m/s.
文摘This paper presents the fundamentals of direct inverse modeling using CFD simulations to detect air pollution sources in urban areas. Generally, there are four techniques used for detecting pollution sources: the analytical technique, the optimization technique, the probabilistic technique, and the direct technique. The study discusses the potentialities and limits of each technique, where the direct inverse technique is focused. Two examples of applying the direct inverse technique in detecting pollution source are introduced. The difficulties of applying the direct inverse technique are investigated. The study reveals that the direct technique is a promising tool for detecting air pollution source in urban environments. However, more efforts are still needed to overcome the difficulties explained in the study.
文摘为提升低空风切变预报精度,本文综合运用欧洲中期天气预报中心第五代再分析资料[European Centre for Medium-Range Weather Forecasts(ECMWF)fifth-generation reanalysis data,ERA5]和美国国家环境预报中心(National Centers for Environmental Prediction,NCEP)的FNL全球再分析资料(Final Operational Global Analysis)、先进星载热发射和反射辐射仪全球数字高程模型以及兰州中川机场的实况观测资料,采用中尺度数值天气预报模式(Weather Research and Forecasting Model,WRF)、WRF结合计算流体动力学(Computational Fluid Dynamics,CFD)方法、长短期神经网络(Long Short-Term Memory,LSTM)方法,对2021年4月15-16日兰州中川机场的两次风切变过程进行模拟分析。结果表明:(1)在小于1 km的网格中使用大涡模拟,WRF模式在单个站点风速模拟任务中表现更好,但在近地面水平风场风速模拟效果上,不如WRF模式结合计算流体力学模型方案;(2)对于飞机降落过程中遭遇的两次低空风切变的模拟,WRF-LES和WRF-CFD两种模式都可以模拟出第一次低空风切变,而第二次受传入模式的WRF风速数据值较小的影响,两种模式风速差都没有达到阈值,需要在后续工作中进一步验证;(3)低风速条件(6 m·s^(-1))下,基于LSTM的单变量风速预测模型平均绝对误差基本维持在0.59 m·s^(-1),能较好地把握不同地形与环流背景条件下风速变化的非线性关系,虽然受到WRF误差和观测要素不全的限制,多变量风速预测能在保证平均绝对百分比误差小于6.60%的情况下,以更高的计算效率和泛化能力实现风速预测。本文不仅验证了WRF-CFD和WRF-LES耦合方案在风场和低空风切变预报中的差异,还探讨了基于LSTM的风速预测的可行性和准确性,期望为提高风场模拟精度,缩短精细风场模拟时间提供新的视角和方法。
