采用Visual Studio.Net和Matlab混编的方式,结合ADO数据库技术,将整个烧结生产工艺过程作为诊断对象,设计开发出基于规则的烧结生产异常诊断专家系统。系统分为烧结全程诊断子系统、配料混合诊断子系统、布料点火诊断子系统和烧结过程诊断子系统。系统通过对特征参数的模糊处理,同时采用软测量模型,再结合诊断知识来实现异常诊断。通过现场仿真试验,系统诊断命中率达到95%以上,性能达到预定要求。通过系统的人机界面可以对异常信息、实时数据等进行查询,还可对知识规则库进行修改和添加等维护工作。
A rule-based abnormity diagnosis expert system for sintering process has been designed and developed by means of Visual Studio. Net, Matlab and ADO, whose diagnostic object is the whole sintering process. The system is composed of four subsystems which are whole sintering process subsystem, proportioning and blending subsystem, distributing and ignition subsystem, sintering process subsystem respectively. The system diagnoses the abnormal status of sintering process by managing characteristic parameters with fuzzy technique, soft-sensing models and diagnosis knowledge. The simulation shows that the accuracy rate of the system is over 95%, which has met the demand. The user can inquire about the abnormity information, real-time data, and consummate the knowledge base by modifying or adding knowledge on system interface.