在突发氯气泄漏时,及时充分地掌握泄漏源的源项信息,是科学预测氯气泄漏扩散事故的难点之一,也是控制和管理氯气泄漏事故的一项重要基础性工作。围绕泄漏源反演问题,对有障碍物环境下氯气管道泄漏扩散后的源项反演进行了数值模拟。采用基于Bayes推断理论的MCMC抽样方法,结合某氯乙酸生产厂区生产装置布局及氯气泄漏的扩散模型,对氯气泄漏源的空间位置和氯气泄漏强度等重要参数进行反演。结果展示了泄漏源项信息,通过结果检验得出参数的反演结果均在真实数值范围之内。

In a burst of toxic and hazardous substance spills, the quick acquisition of the leak source information is one of the difficult issues in the analysis of production safety and accident predictions. The inverse problem of the heavy gas leak source is an important step to establish a reliable heavy gas leakage diffusion model, for the heavy gas spill control and the heavy gas leakage management. The paper discusses the source term inversion after the chlorine gas pipeline leakage and diffusion in the obstacle environment. Combining the MCMC sampling method based on the Bayes inference theory with a Chloroacetic acid production factory production plant layout and a chlorine gas diffusion prediction model, the key parameters of the leak source such as the location and the intensity, are obtained by the inverse method, and it is shown that the results of these parameters are within the range of the true values.