汽车发动机零部件DVP辅助生成方法研究

Research on Aided Generation Mehtod of Automobile Engine Parts'DVP

针对汽车发动机零部件设计验证计划(Design Verification Plan,DVP)编制工作中存在的知识多源异构,人工编制效率低的问题,研究了汽车发动机零部件DVP辅助生成方法。首先建立了零部件的设计验证案例表达模型和推理规则的表达模型,构建了汽车发动机零部件设计验证知识库;提出设计验证案例属性的分类及其相似度计算方法,在此基础上给出了案例局部相似度计算方法;提出了两阶段的案例检索与规则推理修正融合的集成推理模型,实现零部件DVP的辅助生成。最后通过实例计算和性能分析表明,该方法减少了汽车发动机零部件DVP编制的工作量,验证了该方法的有效性。

Aiming at the problems of multi-source heterogeneous knowledge and low manual compilation efficiency in the preparation of the automotive engine components's design and verification plan(DVP),the aided generation method of automobile engine components'DVP is studied.First of all,in order to build a knowledge base for component design verification,a case expression model for component design verification and an expression model for reasoning rules are established.The classfication and the similarity calculation method of design verfication case attributes are proposed for the local similarity calulation method of the case.An integrated reasoning model combining two-stage case retrieval and rule-based reasoning modification is proposed to realize the auxiliary generation of components's DVP.Finally,calculations and performance analysis of examples show that this method reduces the workload of DVP compilation of automobile engine parts and verifies the effectiveness of this method.