| Petrochemical production and processing is the pillar industry to our national economy. It makes contribution to the economy of country and providing services for many people. As we all know, it is difficult to avoid any accident for petrochemical companies during the operation, which not only affects the operation of enterprises, but also makes some negative impact to the enterprise. So the petrochemical enterprises focus on the emergency plan for safety accident, the general way is to organize employees to perform the training of emergency drill on the real field. This kind of model needs to pre-organize staff in different departments, and to establish the standard process, which always makes the process too single to enhance the quality of staff effectively. Therefore, it’s urgent for enterprises to find a new emergency drill training model, which can not only improve the ability of dealing with accidents, but also save time and other natural resources for businesses.According to the actual needs of enterprises, the researchers designed different kinds of simulation system, including the case expert system, the fuzzy expert system, and reasoning systems which based on plans. To some extent, those simulation systems has already solved some business problems, but because of different kinds of safety accidents’ types enterprise must take a lot of energy to establish various exercise depending on the actual demand. On the other hand, some kinds of inference simulation system can only be carried out in accordance with the standard process drill, which will not be able to meet the purpose of the enterprise to enhance the ability of their employees when they are dealing with different types of accidents.In this background, this paper presents a reasoning mechanism of the simulation which is based an objective object node for reasoning. The main idea of this mechanism is independent on the standard procedure, this mechanism will encapsulated this key equipment into independent objective object, the states of objective object which changes itself that can lead to other states’changes. Therefore, this can produce a series of chain reaction which can train employees to deal with an emergency in a real accident situation.The specific process of this mechanism can be divided into the following four steps. Firstly, according to the real equipment operations of petrochemical factory, we can create a reasoning logic tree of different objective objects, and put their state nodes into the XML configuration file; secondly, objective object inference system will build object logical relations of operating equipment by reading the configuration file information, including forward inference, expression inference, reversible inference and conditional branching inference; thirdly, we can connect objective inference system and script framework through luabind engine. And then we can write the script file of the equipment of perform process based on the states of objects; finally, the results of logical inference will be packaged into a message based on the communication module, and then sent to the virtual reality simulation client,to achieve the goal that inference results can be displayed dynamically.The simulation mechanism in this article has been applied in some petrochemical enterprise and achieved a remarkable results.In the process of practical application, real person can select virtual role in the system and perform the task of commander. His action will directly change the state of some key devices. According to the variation of device states, objective object inference system can lead to a series of different training processes. Objective reasoning mechanism can improve the adaptability and diversity of emergency drills. At the same time, the results can be displayed based on virtual reality technique which increased the immersive experience of participating personnel and improve the petrochemical enterprise staffs ability to deal with sudden accidents. |
PDF Full Text Request