Kazuki Okabayashi

Development of a low-rise industrial source dispersion model

In Japan, with amendment of the Air Pollution Control Law in May 1996, various substances, including benzene and trichloroethylene, were newly designated as hazardous air pollutants, and environmental standards were established. In this situation, it is necessary to develop a dispersion model that is applicable to environmental impact assessment of industrial areas with a complex of factory buildings. To overcome this problem, modification of the ISC downdraught model was undertaken based on datasets from wind tunnel experiments by the Ministry of International Trade and Industries and Japan Environmental Management Association for Industry. This new model is called the METI-LIS model, and comparison shows that the performance of the model is better than that of the original ISC model.

A new wind tunnel technique for investigating gas diffusion behind a structure

Abstract We predicted the diffusion of exhaust gas without thermal buoyancy from the top of a structure in down-wash under the influence of a structure by means of wind tunnel. We reproduced a wind directional fluctuation σ A of about 10° in a wind tunnel. The present experimental method and experimental results are introduced here. Wind directional fluctuation was simulated by means of turntable equipment which rotated the model according to probability of wind direction in the field. In measuring concentration, NH 3 as tracer gas was used and a sampling method was adopted. The tracer gas was continuously sampled while rotating the model. As a result the sampled gas concentration was integrated with the weight according to the probability of each wind direction. We compared results of this wind tunnel experiment with the field test and the reliability of this prediction method was confirmed.

Development of overlapping modelling for atmospheric diffusion

Abstract Diffusion modelling in wind tunnels is well-known as a very useful technique to predict pollutant gas concentrations released from continuous sources in complex terrain for environmental assessment. The similarity laws for atmospheric diffusion modelling entail simulation of turbulent fluctuations of wind direction σA, observed over the concentration averaging time, in the wind tunnel. The new concentration overlapping system with the turntable of 12 m diameter in the wind-tunnel test section has been developed and shown to be useful to simulate longer, hourly averaged concentration and to predict the TAF (terrain amplification factor) and the TAP (terrain affected peak) in complex terrain.

Prediction of gas diffusion in complicated terrain by a potential flow model

Abstract A numerical model was developed to simulate gaseous diffusion in complicated terrain. This model calculates the air flow as a potential flow by the Boundary Element Method, and gaseous diffusion by an analytical Gaussian equation in the potential flow. Plume spreads σy and σz are modified by multiregression equations derived from wind tunnel experiments, and the terrain height is elongated depending on the atmospheric stability. First, tracer data from Cinder Cone Butte in the U.S. measured by the U.S.-EPA were predicted by the model in order to examine the prediction accuracy under stable conditions. The averaged ratio of the observed concentration to predicted concentration for 12 runs was better than a factor of 10. Next, tracer data from the Geysers area in the U.S. measured by the U.S.-DOE were used to examine the prediction accuracy under neutral conditions. The ratio of the observed concentration to predicted concentration for two runs under neutral conditions was better than factor of two at most locations, but prediction capability is poor in blocked or separated flow conditions.

Development of numerical model for dispersion over complicated terrain in the convective boundary layer

Complicated terrain and atmospheric stability are important factors in the predicting the dispersion of air pollutants. The aim of our study is to develop practical dispersion model for unstable conditions for regulatory use. The numerical model we adopted was a combination of the potential flow model and Lagrangian stochastic dispersion model. In this model, time-mean flow field is solved by the potential flow model and concentration field is calculated by Lagrangian stochastic model by using the flow field from potential model. Wind tunnel experiments simulating gas dispersion in the convective boundary layer were also done. The data sets of turbulent properties and concentrations obtained in the wind tunnel were used for the modification of dispersion model and model validation. Based on the developed model, user-friendly software applicable to dispersion over complicated terrain was also developed.

Effects of wind directional fluctuations on gas diffusion over a model terrain

Abstract Wind tunnel experiments were carried out to study the effects of wind directional fluctuations on gas diffusion, under a neutral stability condition, over terrain models using the overlapping method. Ground-level concentrations were measured under 3 conditions with standard deviations of wind directional fluctuation, σA, of 5°, 10° and 20°. From these experimental results, an attempt was made to reduce the general rule of the relation between σA and the maximum ground-level concentration, Cmax over a terrain. The verification and limitation in the application of this rule are discussed. Furthermore, it is reported that large σA has a strong influence on the ground-level concentration patterns and so unexpected concentration patterns appear under some conditions.

Integrated Numerical Prediction of Atomization Process of Liquid Hydrogen Jet

The 3-D structure of the liquid atomization behavior of an LH2 jet flow through a pinhole nozzle is numerically investigated and visualized by a new type of integrated simulation technique. The present Computational Fluid Dynamics (CFD) analysis focuses on the heat transfer effect on the consecutive breakup of a cryogenic liquid column, the formation of a liquid film, and the generation of droplets in the outlet section of the pinhole nozzle. Utilizing the governing equations for a high-speed turbulent cryogenic jet flow through a pinhole nozzle based on the thermal nonequilibrium LES-VOF model in conjunction with the CSF model, an integrated parallel computation is performed to clarify the detailed atomization process of a high-speed LH2 jet flow through a pinhole nozzle and to acquire data, which is difficult to confirm by experiment, such as atomization length, liquid core shape, droplet-size distribution, spray angle, droplet velocity profiles, and thermal field surrounding the atomizing jet flow. According to the present computation, the cryogenic atomization rate and the LH2 droplets-gas two-phase flow characteristics are found to be controlled by the turbulence perturbation upstream of the pinhole nozzle, hydrodynamic instabilities at the gasliquid interface and shear stress between the liquid core and the periphery of the LH2 jet. Furthermore, calculation of the effect of cryogenic atomization on the jet thermal field shows that such atomization extensively enhances the thermal diffusion surrounding the LH2 jet flow. (Translation of the article originally published in Cryogenics 48 (2008) 238–247)