摘要
应用基本粗糙集理论对改模方案进行改模知识归纳可以得到改模决策规则集。为解决其规则集规模大、实用性不强的问题,提出了一种基于变精度粗糙集的改模知识分层递阶归纳方法。首先,在研究改模知识的分层递阶层次表达模型的基础上,给出改模方案特征值具体化、完备化和层次化方法,并据此构建了基于分层递阶的改模方案特征决策表集;其次,在特征约简过程中,引入分类质量/误差矩阵,通过基于变精度粗糙集的改模方案特征层次约简,得到符合用户分类质量要求的核心特征集;然后,进一步利用基于分层递阶的规则层次生成和知识获取方法,得到规模小且实用性强的规则集;最后,通过实例验证了该方法的实用性和有效性。
Rule sets could be obtained by knowledge induction for injection mould repairs based on rough sets theory, but the rule sets were in large-scale and with poor usability. Aiming at these problems, a new method of knowledge hierarchical induction based on variable precision rough sets was proposed. Firstly, hierarchical representation model of injection mould repair knowledge was set up and decision table sets were generated by the process of specification, completion and hierarchy to feature value. Secondly, classification quality/error matrix was introduced into the process of knowledge induction. The core features sets satisfied classification quality requirements from different en- gineers were acquired by feature hierarchical reduction based on variable precision rough sets. Then, the rule sets were obtained according to rules hierarchical reduction and knowledge acquisition. Finally, a case study was provided to illustrate the feasibility and effectiveness of the proposed method.
出处
《计算机集成制造系统》
EI
CSCD
北大核心
2009年第11期2259-2265,共7页
Computer Integrated Manufacturing Systems
基金
国家自然科学基金资助项目(50775042)
国家863/CIMS主题资助项目(2007AA04Z1A8)
国家科技支撑计划资助项目(2006BAF01A43)
教育部博士点专项基金资助项目(20070562003)~~
关键词
变精度粗糙集
分层递阶表达
改模知识归纳
知识获取
模具
variable precision rough sets
hierarchical representation
mould repair knowledge induction
knowledge acquisition
mould
