The Multi-Drill-Plan Inference System For The Emergency Rescue

With the rapid development of the modern petrochemical enterprises, the emergency drill has become more and more significant in traditional petrochemical industry. Nowadays, the use of virtual reality technology combined with expert system for emergency drill inference has become an important research direction in chemical safety drill. However, when dealing with the emergency plans of different environments, the traditional expert system needs to collect a lot of related plans to rebuild the knowledge base and it is unable to meet the requirements of customizing the simulation of emergency drill in short time. To overcome the drawback, this paper presents the multi-drill-plan inference system for the emergency rescue. And the concerned study is shown as follows.Firstly, the visualization system for text plan is researched. Its main task is to configure the relationship between the key operations and turn the text plan into the visual process.Secondly, the configurable NPC drill inference is proposed. It is responsible for dealing with the NPC drill situations. This paper mainly configures the entities and states in the drill to create the state graph, and infers the behaviors at different phases during the drill.Thirdly, the EFSM-based real-player drill inference is proposed. It is responsible for dealing with the real-player drill situations. Besides the configuration of the entities and states, every real player has one or more EFSM inference models. The transition algorithm of EFSM is used to infer the drill process of real players’.Finally, the multi-branch drill inference is proposed. Besides the configuration of the entities and states, the multi-branch drill needs to configure the expressions of every branch and insert the facts corresponding to the users’input at the key states of the drill, which will decide the next chosen branch.The inference system proposed in this paper has been widely used in the chemical enterprises. On the basis of the above inference system, the emergency drill process could be achieved by combining with the current three-dimensional rendering engine. It has been proved in reality that the proposed inference system can be quickly adapted to different emergency drill plans under different drill environments and has great practical value on staff training.