摘要
为解决设计人员在问题分析阶段自身知识的局限性,辅助设计人员构建因果分析图,以设计问题为导向进行专利检索,并根据中文专利语言特点,构建专利因果词表,基于因果词表构建专利特定领域因果语句的提取规则;基于词性及词性组合构建因果语句中因果知识的提取规则,提出面向专利的因果知识提取方法;构建设计问题和因果知识间的匹配策略,采用余弦相似度算法匹配设计问题相关的因果知识,提出中文专利因果知识匹配方法;以卡堵的管道猪被清管器低损伤回收时解卡成功率低的问题为案例,应用该方法扩充因果图,验证该方法的有效性。研究结果表明:采用辅助因果图构建方法可增加原因节点数量,基于扩展后因果图的案例设计能有效解决提出的问题。
To address the limitations of designer’s own knowledge during the issue analysis stage,the cause-and-effect analysis graph was constructed to assist the designer.Patent index was oriented to design issues,and the cause-and-effect words list of patent were constructed according to the characteristics of Chinese patent language.The extraction rules of cause-effect statement in specific patent domain were constructed based on cause-effect word list.Based on the parts of speech and the combination of parts of speech,the extraction rules of causal knowledge in causal sentences are constructed,and a patent-oriented method of causal knowledge extraction was proposed.Matching strategies between design problems and causal knowledge was constructed.The cosine similarity algorithm was used to match the causal knowledge related to design problems.A causal knowledge matching method for Chinese patents was proposed.Taking the problem of the low success rate of unjamming when a jammed pipe pig recovered by a pipe cleaner with low damage as a case study,this method was applied to expand the causal graph,and verify the effectiveness of the method.The results showed that the number of cause nodes could be increased using auxiliary causal graph construction method.The case design based on the extended causal graph could effectively solve the problem.
作者
曹国忠
高钱钱
CAO Guozhong;GAO Qianqian(National Engineering Research Center for Technological Innovation Method and Tool,Hebei University of Technology,Tianjin 300401;School of Architecture&Art Design,Hebei University of Technology,Tianjin 300132)
出处
《机械设计》
北大核心
2025年第7期206-212,共7页
Journal of Machine Design
基金
河北省创新能力提升计划项目(21567657H)。
