一种基于SDG用于危险分析的新型定性仿真技术

安全已经成为世界各国政府面临的一个重要问题。由于安全研究的实验无法在真实工厂中进行，因此仿真技术已经成为研究和解决安全问题的一个核心基础。然而，对于真实化工工厂的精确定量数学模型由于规模庞大和复杂而难以获得与计算。与此相反，定性模型相对容易建模，计算也相对较快。我们提出了一种基于符号有向图的定性仿真技术。它能够有效地针对复杂系统中各个组件间的原因与后果关系。通过网络拓朴和专家规则，SDG模型可以有效地找出系统中潜在的危险。结合目前最为广泛应用的过程危险分析方法，即HAZOP，提出一种新的SDG-HAZOP方法，以代替目前的人工安全分析。最后以65t／h锅炉为例，展示并讨论该新方法的优、缺点。

Novel Qualitative Simulation Technology Based on SDG for Hazard Analysis

Safety has become a serious problem in the front of governments around world. In order to study and solve safety problem, simulation technology is the key basis, since no safety experiment can be allowed in real factory. However, the exact quantitative mathematic model of chemical factory is too large and complicated to be created and calculated. On the contrary, qualitative simulation can be relative easily to model and quickly to calculate. A qualitative simulation method is proposed based on signed directed graph (SDG). It can effectively present cause and effect relationship between components in complex system. By iteration and expert rules, SDG model can efficiently find out potential hazard hidden in system. Inspired by the mechanism of most popular PHA method, e.g. Hazard ＆ Operability Study (HAZOP), a SDG-HAZOP method is recommended to substitute for manual hazard analysis. A 65t/h boiler's example is provided to demonstrate this novel hazard analysis method, and its advantages and disadvantages are also discussed.