基于相似性度量的氧化铝配料知识库在线维护  

Knowledge base on-line maintenance for blending expert system in production of alumina based on similarity measure

作　　者：王雅琳[1] 舒军[1] 阳春华[1] 桂卫华[1] 

机构地区：[1]中南大学信息科学与工程学院,长沙410083

出　　处：《计算机工程与应用》2007年第20期200-203,共4页Computer Engineering and Applications

基　　金：国家自然科学基金(the National Natural Science Foundation of China under Grant No.60574030);湖南省自然科学基金(the Natural Science Foundation of Hunan Province of China under Grant No.06FD026)

摘　　要：针对烧结法氧化铝优化配料专家系统知识库的结构特点,提出了一种基于相似性度量的专家知识库在线维护方法。构造规则的相似性度量函数,以此为基础进行规则不一致性判断,并遵循原有的知识组织策略在线实现规则的有序添加和修改,从而保证高效的专家推理,提出的方法已成功用于工业应用。In view of knowledge base＇s structure characteristics of blending optimization expert system in production of alumina with sintering process,an on-line maintenance method of expert knowledge base based on similarity measure is proposed.At first similarity measure function of rule is defined;then it is described in details how to take consistency check and orderly on-line add and modify rule following the original knowledge .organization strategies by similarity measure so as to ensure highly effective expert inference.The on-line maintenance method proposed is applied successfully to industries.

关 键 词：知识库维护 相似性度量 烧结法氧化铝 优化配料 专家系统 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]