Most factories use toxic or flammable chemicals in their industrial processes; this poses a risk of leakage due to accidents, which can sometimes result in mass casualties and considerable property damage. Therefore, quantitative risk prediction and assessment are necessary to protect the public and minimize losses in the event of a chemical release. Several methods can be used to evaluate chemical dispersion in the atmosphere, but most are based on the assumption of neutral buoyancy and far-field wind conditions. With this assumption, a model is valid only after a significant amount of time has elapsed from the moment chemicals are released or dispersed from a source. However, the most dangerous locations are typically close to a leak source. Therefore, a dispersion model for the initial phase of a leak is required to quantify the risk and predict the extent of damage. This study developed a dispersion model for initial consequence analysis using a three-dimensional unsteady convective diffusion equation. A continuous point source was assumed, and ethane was used as the leak material to minimize the effects of buoyancy. The unsteady concentration field developed rapidly as the diffusion coefficient and wind velocity increased, but the steady-state concentration field was not affected by wind velocity. This dispersion model also allows for the simple consideration of ground adsorption, making it scalable to real-world situations. The time to reach a steady state required for an effective emergency response was predicted, and the results were interpreted in terms of mathematical methods and physical characteristics.

中文翻译：

---

基于扩散方程的初始后果分析的扩散模型

大多数工厂在其工业流程中使用有毒或易燃化学品；这会带来因事故而泄漏的风险，有时会导致大规模人员伤亡和重大财产损失。因此，有必要进行定量风险预测和评估，以保护公众并最大程度地减少化学品泄漏时的损失。有多种方法可用于评估化学物质在大气中的扩散，但大多数方法都是基于中性浮力和远场风条件的假设。根据这一假设，只有在化学物质从源头释放或扩散的那一刻起经过相当长的时间后，模型才有效。然而，最危险的位置通常靠近泄漏源。因此，需要泄漏初始阶段的扩散模型来量化风险并预测损坏程度。本研究使用三维非定常对流扩散方程开发了用于初始后果分析的弥散模型。假设连续点源，并使用乙烷作为泄漏材料，以尽量减少浮力的影响。随着扩散系数和风速的增大，非稳态浓度场迅速发展，但稳态浓度场不受风速影响。该分散模型还可以简单地考虑地面吸附，使其可扩展到现实世界的情况。预测了有效应急响应所需的达到稳定状态的时间，并用数学方法和物理特性解释了结果。