Research On Cloud Manufacturing Service Matching Optimization Method For Machining

With the new wave of changes in the global manufacturing,cloud manufacturing(CMfg)emerges as a new manufacturing mode,which has become one of the important links in the development of China's manufacturing industry towards information,service and network.At the same time,for the field of mechanical processing occupying an important position in the manufacturing industry,it has become a hot spot for cloud manufacturing research in specific application fields.However,in the face of massive heterogeneous manufacturing resources,CMfg platforms are difficult to achieve fast and accurate service matching,which has become the primary problem for cloud manufacturing machine processing service model applications.In addition,service trust evaluation,as a key factor in ensuring users to choose secure and credible manufacturing services in the cloud platform,requires more in-depth research.Furthermore,service composition optimization is an important issue to realize resource optimization configuration.The existing methods are difficult to guarantee the accuracy and stability of large-scale service composition results.Therefore,this paper takes the machiningoriented cloud manufacturing service as the research scope,and carries out corresponding research on the above problems.Firstly,the CMfg machine processing tasks and resource service types and characteristics were analyzed.Combined with the feature-based modeling method,the CMfg task requirements and resource service models for machining are constructed.The mathematical description forms were elaborated.In order to improve the matching efficiency,a cloud-oriented task and service matching model for machining was proposed,and the matching method of processing feature stage was also studied,which realized the fast and complete matching of tasks and services and the construction of candidate resource service sets in CMfg environment.Secondly,a satisfaction-based trust evaluation model for cloud manufacturing service is studied.The evaluation index system was built to quantize the satisfaction by monitoring the data of historical service evaluation.The concept of friends and platform were introduced to form a trust network.Meanwhile,the comprehensive trust was corrected by using the service satisfaction volatility,which was taken as a mechanism of reward and punishment in the long-term transaction process.The experiment results verified that the model reflected the subjective preference of consumers,and effectively suppressed the trust bias caused by malicious evaluation.Especially in a CMfg environment with massive machining services,it has high execution efficiency.Then,the construction process of cloud manufacturing service composition was analyzed,and the more realistic mathematical model of service combination optimization was established according to the characteristics of the machining industry.Aiming at the large-scale service composition optimization problem,an efficient dynamic ant colony genetic hybrid algorithm(DAAGA)was proposed,which introduced the optimal fusion evaluation strategy and iterative threshold to improve it.An algorithm comparison experiment with genetic algorithm,maximum and minimum ant algorithm,artificial bee colony algorithm and particle swarm optimization algorithm,as performed to validate the superiority of DAAGA algorithm.In addition,the feasibility and effectiveness of the algorithm were verified by an example.Finally,the prototype of the CMfg platform was introduced.Based on previous research and development,the tasks and service release,matching optimization and aftersales service of CMfg platform prototype were implemented.