Research And Application On Equipment Resource Matching In Cloud Manufacturing Enviornment

Cloud Manufacturing is a new type of manufacturing model,which is based on the concept of cloud computing,combined with the actual situation of manufacturing industry and the basic idea of manufacturing as a service.Through a field survey conducted by a radiator manufacturing company in Shenzhen,found that the company often take a part of the manufacturing processing to outsourcing for focusing on design and R&D capabilities.Applying this business model to the cloud manufacturing environment can effectively avoid some problems such as inconsistent product quality and waiting time in outsourcing.The company's product manufacturing process requires a various of equipment resources,and the processing tasks also are complex and diverse.Therefore,the effective matching between equipment resources and manufacturing tasks is the key technology for successful application of cloud manufacturing.This paper proposes an optimized matching mechanism for equipment resources in the cloud environment.The main contents are as follows:Firstly,the equipment resource data in the cloud manufacturing platform is preprocessed to meet the requirements of the cluster analysis algorithm.That is,the k nearest neighbor filling algorithm is used to fill the missing data values and the Z-score standardization method is used for dimensionless processing.Based on this,the resource data is analyzed using the systematic method based on the sum of squared deviations to obtain the cluster pedigree map of the existing equipment resources in the cloud platform.the Bayesian discriminant method is used to generate the discriminant function for each category of equipment resources,the new resources are automatically classified on the cloud manufacturing platform.In view of the characteristics of the equipment resources,a unified description model for the cloud platform was established.Based on this model,an optimized matching mechanism for equipment resources in a cloud manufacturing environment was proposed.Including rough selection,preselection and optimization of equipment resources.Then the decision tree analysis method can be used to classify the categories based on the value range of the equipment resource's technical parameters.Matching the value range of the parameter with the resource parameter constraint in the user's publishing task can quickly select the corresponding resource category.The pre-selection of the device resource is to match the unified description model of the device resource and the manufacturing task.The preselected resource set that meets the processing requirements from the rough resource set.Resource optimization is based on an improved collaborative recommendation algorithm based on items,taking into account factors such as time,cost,quality,reliability,and credibility of the resources,calculating the user's interest in the resources of the device so as to recommend the most suitable resource to the user.Through the above research,it provides a theoretical basis for the research of cloud manufacturing,and realizes a rapid and effective resource matching of resources in a cloud manufacturing environment.Finally,by analyzing the data and manufacturing task of CNC lathe,the feasibility of the optimized matching mechanism is proved.Through the above research,it provides the theoretical basis for the research of cloud manufacturing,and realizes the fast and effective resource matching in the cloud manufacturing environment.