CAJ | 학술논문

流程工厂设计涉及不同的工程应用背景及专业知识, 大量隐含的设计规则尚未提炼, 因此建模过程存在效率低、模型质量不高等问题. 结合流程工厂模型的领域特性, 提出一种用于挖掘工厂设计规则的频繁类型树模型. 该模型以大量实例工厂模型中各构件的属性及拓扑连接关系为依据构造频繁类型树; 在后续模型的构建过程中, 通过对该树的查询为当前工厂设计提供实时的智能预测. 实验结果表明, 文中模型能够挖掘到不同工程应用背景下的隐含建模知识, 在未来模型的构建过程中, 利用这些知识所提供的智能预测可以大幅缩小构件选择范围, 有效地提升流程工厂模型的构建效率及设计质量.
류정공엄설계섭급불동적공정응용배경급전업지식, 대량은함적설계규칙상미제련, 인차건모과정존재효솔저、모형질량불고등문제. 결합류정공엄모형적영역특성, 제출일충용우알굴공엄설계규칙적빈번류형수모형. 해모형이대량실례공엄모형중각구건적속성급탁복련접관계위의거구조빈번류형수; 재후속모형적구건과정중, 통과대해수적사순위당전공엄설계제공실시적지능예측. 실험결과표명, 문중모형능구알굴도불동공정응용배경하적은함건모지식, 재미래모형적구건과정중, 이용저사지식소제공적지능예측가이대폭축소구건선택범위, 유효지제승류정공엄모형적구건효솔급설계질량.
Process plant design involves a variety of engineering backgrounds and specialized knowledge. With abundant latent design rules not being extracted, the existing modeling methods present the disadvan-tages of low efficiency and deficient model quality. In that sense, a Frequent-Type-Tree model for mining process plant design rules is proposed. The model constructs a Frequent Type Tree by analyzing component attributes and topological correlations in plant models and provides real-time intelligent predictions by que-rying the tree for future design. Experimental results show that the proposed model can mine valuable latent modeling knowledge in various application backgrounds. The mined knowledge is then used to shrink ranges of component selection by a large margin while future modeling efficiency and design quality is improved significantly.