Research On Resource Modeling And Decision Methods For The Cloud Service Of Intelligent Disassembly And Recycling For The Mechatronics Products

In recent years,the modern manufacturing information technology has developed and the product requirement has grown,the production quantity of mechatronics products has increased a lot.However,if the obsolete mechatronics product does not been efficiently recycled,the waste material of the product will harm the natural environment.In order to fully utilize social recycling resources,form a collaborative recycling system and efficiently recycle waste products,this paper adopts the concept of cloud manufacturing and proposes the cloud service of intelligent disassembly and recycling for mechatronics products.The key technologies are studied in this paper.The contributions include:(1)This paper defines the cloud service of intelligent disassembly and recycling for mechatronics products and analyzes the features which include virtualization,collaboration,real-time,knowledge-driven and openness.The system framework of the cloud service of intelligent disassembly and recycling for mechatronics products is designed.Based on the framework,the application model and the business model of the cloud-based recycling service in the recycling industry are discussed.(2)A semantic model of products is built in this research,which represents components,assembly structures,and indicates the semantic relationship among the instance of concepts.This paper describes the contact relationship of components,the geometric constraint and the knowledge of disassembly tools based on semantic rules.Disassembly-knowledge-based decision methods for deciding disassembly direction,disassembly tool and disassembly sequence are proposed to realize disassembly-knowledge-based selective disassembly planning.(3)An ontology model of the physical resource of the cloud-based recycling service is designed based on the semantic method,which describes the concepts such as the basic information,the service ability,the working status and the quality of the recycling service.The relationship of the concepts is also described by the ontology model.Then,a resource pool of multisource and heterogeneous recycling resources can be built.This paper adopts the computation method of semantic similarity,which is combined by concept matching and numeric parameter matching,to evaluate the similarity of the request and the recycling resource.Meanwhile,performance,reliability,availability,credibility and other QOS indicators,and recycling time,recycling cost and other numeric parameters are introduced to measure the recycling resource.(4)A combinatorial optimization mathematical model of cloud-based disassembly service is built in this research,the objective includes minimizing disassembly time and disassembly cost.A multi-objective genetic algorithm based on NSGA-II is designed to solve the NP-hard problem.The algorithm is improved by the proposed "adjustment operator".The algorithm generates a set of Pareto optimal solutions,and the disassembly service can be selected according to the user preference.(5)An ontology of smart products is designed to describe the concepts of product,component,material and usage information,and the relationship of the concepts.The ontology integrates the lifecycle information of smart products.A recycling choice decision method is designed based on the fuzzy rule to decide the recycling choice of components,which considers fuzzy variables including usage time,condition and working load,etc.Based on criteria such as service time,service cost,recycling revenue,service times,etc.,a grey-relational-analysis-based method for evaluating cloud-based recycling resources is proposed.A robot vacuum cleaner is taken as the case study to verify fuzzy-rule-based recycling choice decision method,grey-relational-analysis-based method for evaluating cloud-based recycling resources,and particle-swarm-optimization-based method for generating disassembly sequence.This paper studies the modeling and decision methods for the resources of recycling knowledge and recycling facility.The methods make the life cycle information of products knowledgeable and transparent.The proposed methods provide decision tools for product designers,manufactures,society recycling sites and related government agencies.The research methods can be extended to other domains such as product recycling,reuse,remanufacturing,etc.The methods also contribute to green design and design for maintenance.