摘要
工业设备的大型化和工业过程的日趋复杂,使得对工业过程的状态监测和故障诊断显得尤为重要。文中在多元统计过程控制方法的基础上,将神经网络用于过程的监测,并在基于CORBA技术的多客户端工业过程中实时监测软件系统实现。考虑工业系统故障预知的滞后,软件设计中采用了特殊的复合神经网络结构以便于维护和拓展。FCCU的实验数据表明:系统能很好监测和判断过程运行状态,并且能够在客户端图形化显示,操作简单。
Status monitoring and fault diagnosis are the great importance in chemical engineering process,especially by now the equipment is becoming more and more complex. On the basis of the MSPC, this paper adopts the neural network for process monitoring and adds it to the multi - client distributed monitoring system designed based on CORBA. Considered the lag of the fault detection, the design introduces a specific compound network architecture that makes the software convenient for maintainability and extension. The experiment data shows that the graphical results could well - reflect the status of the FCCU system.
出处
《计算机技术与发展》
2006年第7期111-113,179,共4页
Computer Technology and Development
