Research On Intelligent Variant Design Based On CBR And Knowledge Recommendation

After the financial crisis,China’s manufacturing industry faces double pressure from developed countries and the backward other developing nations.The transformation and upgrading of manufacturing industry is imminent.The improvement of design ability is the only way for China to climb along the manufacturing value chain.The complexity and personalization of the product and the shortening of the life cycle increase the complexity and timeliness of the design.Under this background,variant design is not only a required course for the transformation and upgrading of China’s manufacturing industry,but also a trump card.The key of successful variant design is the effective reuse of previous design knowledge.Due to the natural similarity between case-based reasoning and variant design in general thinking,this paper takes it as the main body and combines knowledge recommendation technology to achieve reuse of design knowledge,so as to improve the efficiency of variant design.Aiming at the shortcomings of case acquirement in traditional case-based reasoning and the storage characteristics of design related data,a case acquiring method based on ontology population is proposed in this paper.Based on the semantic information from ontology,the design information stored in the PDM database and design documents is extracted and then organized as ontology instances.The ontology instances form the case base and provide the basis for the subsequent design knowledge reuse.A case adaptation rule acquisition method based on data mining is proposed in view of the shortage of knowledge acquisition in case adaptation.The cases are first pre-processed.Then,by comparing the case pairs,the changing events are extracted to represent the difference between cases.Then,the FP-Growth algorithm is applied to acquire the adaptation rules.Aiming at the existing problems of traditional case retrieval method,this paper proposes a case retrieval method considering reverse constraint,so as to ensure that the subsequent case adaptation can achieve the maximum degree of automation.First,the rule maturity of each case is calculated to measure the sufficiency and quality of the required adaptation rules.Then,the adaptability of each case is calculated to measure the adaptation workload.Finally,the suitability of the cases is obtained by synthesizing the maturity and the adaptability.The case with the largest suitability is selected as the starting point of the new design.In order to further improve the efficiency of knowledge reuse in variant design process,a knowledge recommendation method based on sequential pattern mining is proposed in view of the existing problems in knowledge recommendation.Firstly,the frequent knowledge sequences were found by analyzing historical knowledge usage data of the designers with sequential pattern mining technique.Then the knowledge was recommended by considering the designer’s knowledge usage behavior,the support degree of frequent knowledge sequences and the similarity between the current knowledge sequence and the frequent knowledge sequences.According to the intelligent variant design method based on CBR and knowledge recommendation proposed in this paper,an intelligent variant design system is developed based on the InforCenter platform of Hoteam Software and applied in an electric motor manufacturing enterprise.