CN103714050B - Resource modeling framework for manufacturing multi-layer cloud with multi-granularity characteristic - Google Patents

Abstract

The invention discloses a multi-layer cloud manufacturing resource modeling framework with multi-granularity characteristics, which provides a unified description specification for heterogeneous manufacturing resources in a cloud manufacturing environment, realizes global resource sharing and improves resource utilization. This resource modeling framework is divided into three layers, which are resource model layer, functional model layer and semantic metamodel layer from bottom to top. The resource model layer and the function model layer are associated through a many-to-many mapping relationship, which makes the resources and their corresponding functional characteristics loosely coupled together, and improves the flexibility of the modeling framework. The functional model layer and the semantic meta-model layer are associated through the conversion relationship, so that the resource description has accurate semantics.

Description

A multi-layer cloud manufacturing resource modeling framework with multi-granularity characteristics

technical field

The invention relates to a multi-layer cloud manufacturing resource modeling framework with multi-granularity characteristics implemented for resource semantic description in a cloud manufacturing environment, and belongs to the technical field of manufacturing resource description.

Background technique

Resource description is the basis of resource discovery and resource selection. The adequacy and accuracy of resource description determines the level of resource utilization. Therefore, resource description is one of the key issues in cloud manufacturing. The specification of resource description depends on an effective resource modeling framework. The current main resource modeling framework design technologies include standard-based resource modeling framework design and ontology-based resource modeling framework design.

The standard-based resource modeling framework design method mainly uses a series of standards related to web services, such as: using WSDL to describe web services, registering services through UDDI, and searching services based on keyword methods. However, this resource description method lacks machine-understandable resource semantics, and resource discovery is incomplete and inaccurate, resulting in low resource utilization.

Ontology-based resource modeling framework design method improves the deficiency of standard-based resource modeling framework design method from many aspects. For example: OWL-S provides semantic information of resource description, and provides a basis for automatic resource discovery and combination. WSMO uses a state machine to model resource behavior, and realizes capability orchestration between resource internal capability description and resource combination.

Existing resource modeling frameworks cannot be directly applied to the description of cloud manufacturing resources, because cloud manufacturing resources have characteristics such as large number, variety, heterogeneity, dynamics, and multi-granularity. From the adequacy of resource information, the standard-based resource modeling framework lacks the description of the multi-granularity functional characteristics of cloud manufacturing resources and the specific QoS information of cloud manufacturing resources. In terms of the accuracy of resource semantics, the ontology-based resource modeling framework only provides semantic annotations of input and output information of resource functional characteristics, and lacks semantic descriptions of resource context information and resource interoperability information. Therefore, based on the above deficiencies, the present invention designs a multi-layer cloud manufacturing resource modeling framework with multi-granularity characteristics.

Contents of the invention

Purpose of the invention: The present invention provides a multi-layer cloud manufacturing resource modeling framework with multi-granularity characteristics, which provides necessary information and accurate semantics for cloud manufacturing resource description. The multi-granularity characteristics of cloud manufacturing resources are displayed from different abstraction levels, and the multi-layer resource modeling framework can adapt to the dynamic characteristics of the required resources in the cloud manufacturing environment and the abstract level expression of different functions, so that resource sharing can be realized.

Technical solution: a multi-layer cloud manufacturing resource modeling framework with multi-granularity characteristics, which can provide resource abstraction and sharing for distributed resources in the cloud manufacturing environment. This model framework is divided into three layers, which are the resource model layer at the bottom layer, the functional model layer at the middle layer and the meta-model layer based on semantics at the high layer. The lowest layer describes various physical resources, which is the basis of cloud manufacturing, and is composed of heterogeneous resources owned by geographically distributed enterprises; the middle layer describes various functions of resources, and provides unified resource functions for the cloud manufacturing platform View; the highest layer defines resource-related concepts and the relationship between resources, provides sufficient semantic information for resource applications, and promotes semantic interoperability between collaborative enterprises.

The multi-layer cloud manufacturing resource modeling framework with multi-granularity features includes the following relationships:

The many-to-many mapping relationship, associating the resource model layer with the functional model layer, improves the flexibility of resource modeling by decoupling cloud manufacturing resources from their corresponding functional features. Resources can dynamically join or exit the cloud manufacturing system, which is in line with cloud manufacturing The dynamic characteristics of the system, a resource has multiple functional characteristics, and a functional characteristic can be provided by different resources, and the many-to-many mapping relationship reflects the multi-functional, multi-task, and multi-purpose characteristics of resources;

Semantic conversion relationship, associating the functional model layer with the semantic-based meta-model layer, and semantically annotating heterogeneous resource information through the concepts and relationships defined in the meta-model layer to ensure that heterogeneous resources can be interpreted correctly in different contexts , to ensure the semantic consistency of resources.

Furthermore, the resource model layer receives the heterogeneous resource information registered by distributed enterprises, designs the XML schema corresponding to the resource information according to the original resource description method of the enterprise, and converts the heterogeneous resource information into corresponding XML documents, so that heterogeneous Resource information isomorphism.

Further, the function model layer extracts resource function information from XML documents corresponding to resources, merges function information of similar resources, designs XML schemas corresponding to function information according to the result of function information merger, and establishes function information XML document views.

Furthermore, the meta-model layer extends the common ontology components, provides concepts, relationships, constraints, and axioms to mark resource-related semantics; defines cloud manufacturing resource semantic information, resource non-functional feature semantic information, and multiple Semantic information of resource function characteristics of granular characteristics, semantic information of cloud manufacturing context environment, and semantic information of supporting cloud manufacturing service level agreement mechanism.

Further, many-to-many relationship: For each registered resource in the resource model layer, search the functional information view sequentially, and link the resource with its corresponding functional characteristics. If there is no functional characteristic corresponding to the registered resource, then in the functional information Create a new functional characteristic item in the view, link the resource with the new characteristic item, and record the link relationship between the resource and the functional characteristic in the resource function mapping table; for each functional characteristic in the functional view, find the provided Link all resources of this functional characteristic with corresponding resources, and record the associated information in the functional characteristic mapping table; if there is no resource providing this functional characteristic in the resource registry, delete the functional characteristic item.

Further, semantic conversion relationship: according to the concepts defined in the semantic meta-model, mark the equivalent vocabulary involved in resources and functional characteristics; according to the relationship defined in the semantic meta-model, the relationship between resources and functional characteristics Marking; according to the constraints defined in the semantic metamodel, describe the constraints that resources and functional characteristics need to meet; according to the axioms defined in the semantic metamodel, mark the logical relationship between resources and functional characteristics, Provide machine-understandable reasoning information.

Beneficial effects: the present invention provides a description method of multi-granularity functional characteristics of cloud manufacturing resources, provides QoS information required for the application of cloud manufacturing resources, models the cloud manufacturing context environment, and provides description information supporting service level agreement mechanisms. The invention embodies the characteristics of cloud manufacturing resources, the model framework ensures the information interaction of cooperative heterogeneous enterprises, provides the dynamic relationship between manufacturing resources and manufacturing functions, ensures the flexibility of cloud manufacturing, and effectively distributes manufacturing resources to manage.

Description of drawings

Figure 1 is a conceptual framework diagram of the semantic metamodel;

Fig. 2 is a description diagram of multi-granularity functional characteristics;

Figure 3 is an example diagram for describing the functional characteristics of a multi-spindle lathe;

Figure 4 is an example diagram of cloud manufacturing context description based on historical experience;

Figure 5 is a framework diagram of multi-layer cloud manufacturing resource modeling with multi-granularity characteristics.

detailed description

Below in conjunction with specific embodiment, further illustrate the present invention, should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention, after having read the present invention, those skilled in the art will understand various equivalent forms of the present invention All modifications fall within the scope defined by the appended claims of the present application.

The multi-layer cloud manufacturing resource modeling framework with multi-granularity characteristics (as shown in Figure 5) is divided into three layers, namely, the bottom resource model layer, the middle layer functional model layer and the high layer semantic-based meta-model layer ( As shown in Figure 1). The lowest layer describes various physical resources, which is the basis of cloud manufacturing, and is composed of heterogeneous resources owned by geographically distributed enterprises; the middle layer describes various functions of resources, and provides unified resource functions for the cloud manufacturing platform View; the highest layer defines resource-related concepts and the relationship between resources, provides sufficient semantic information for resource applications, and promotes semantic interoperability between collaborative enterprises.

The multi-layer cloud manufacturing resource modeling framework with multi-granularity features includes the following relationships:

The many-to-many mapping relationship, associating the resource model layer with the functional model layer, improves the flexibility of resource modeling by decoupling cloud manufacturing resources from their corresponding functional features. Resources can dynamically join or exit the cloud manufacturing system, which is in line with cloud manufacturing The dynamic characteristics of the system, a resource has multiple functional characteristics, and a functional characteristic can be provided by different resources, and the many-to-many mapping relationship reflects the multi-functional, multi-task, and multi-purpose characteristics of resources;

Semantic conversion relationship, associating the functional model layer with the semantic-based meta-model layer, and semantically annotating heterogeneous resource information through the concepts and relationships defined in the meta-model layer to ensure that heterogeneous resources can be interpreted correctly in different contexts , to ensure the semantic consistency of resources.

The resource model layer receives the heterogeneous resource information registered by the distributed enterprise, designs the XML schema corresponding to the resource information according to the original resource description method of the enterprise, and converts the heterogeneous resource information into the corresponding XML document, so that the heterogeneous resource information isomorphism.

The function model layer extracts the resource function information from the XML document corresponding to the resource, merges the function information of similar resources, designs the XML schema corresponding to the function information according to the result of the function information merger, and establishes the function information XML document view.

The meta-model layer extends the common ontology components and provides concepts, relationships, constraints, and axioms to mark resource-related semantics; defines cloud manufacturing resource semantic information, resource non-functional feature semantic information, and multi-granularity features from four aspects. Resource functional feature semantic information (as shown in Figure 2), cloud manufacturing context semantic information and semantic information supporting cloud manufacturing service level agreement mechanism.

Many-to-many relationship: For each registered resource in the resource model layer, search the functional information view sequentially, and link the resource with its corresponding functional characteristics. If there is no functional characteristic corresponding to the registered resource, create it in the functional information view New functional feature items, link resources with new feature items, and record the link relationship between resources and functional features in the resource function mapping table; for each feature in the function view, find and provide this feature in the resource registry Link all the resources with the corresponding resources, and record the associated information in the feature mapping table; if there is no resource providing this feature in the resource registry, delete the feature item.

Semantic conversion relationship: according to the concepts defined in the semantic metamodel, mark the equivalent vocabulary involved in resources and functional characteristics; according to the relationship defined in the semantic metamodel, mark the relationship between resources and functional characteristics; According to the constraints defined in the semantic meta-model, describe the constraints that resources and functional characteristics need to meet; according to the axioms defined in the semantic meta-model, label the logical relationship between resources and functional characteristics, and provide machine-readable comprehension of reasoning information.

1. Example of functional characteristics description of multi-spindle lathe

An example of a multi-spindle lathe with finishing features is depicted in Figure 3. Based on the multi-level resource modeling framework of multi-granularity characteristics, the functional characteristics of the multi-spindle lathe can be extended to three different levels of multi-granularity functional characteristics according to the inclusion relationship, such as: drilling, turning, and finishing turning. The description granularity of these three functional levels is from coarse to fine, corresponding to three different levels of granularity, alliance layer, enterprise layer, and resource layer. Through the expansion of multi-granularity functional characteristics, multi-spindle lathes can be found and used at three different granularity levels, which improves resource utilization.

2. Example of cloud manufacturing context description based on historical experience

As shown in Figure 4, an example of cloud manufacturing context description based on historical experience. A multi-spindle lathe with the function of fine turning, its resource provider is NEV.Company@ChengDu, according to the historical experience of resource application, the resource consumer ISS.Company@Anchorage has sent a resource request, with the function of machining type, can be The lathe operation can be assisted by human resources. The resource location is in Shanghai, and the completion time is 3 hours. Based on the above requirements and historical experience, it can be found that this multi-spindle lathe is selected to provide services. The actual record of its services is no human assistance, and 2 According to the situation of this service, resource consumers can rate resources to represent the degree of support of resources for such needs. If the cloud platform receives a demand for similar resources, it can feedback the historical rating of this resource to the demander , the demander directly decides whether to choose this resource for service according to the rating, which increases resource utilization and resource discovery efficiency.

Claims (2)

1. A multi-layer cloud manufacturing resource modeling framework with multi-granularity features, characterized in that: it includes a bottom-level resource model layer, a middle-level functional model layer, and a high-level semantic-based meta-model layer; the resource model layer describes each A physical resource, which is composed of heterogeneous resources owned by enterprises in geographical distribution; the functional model layer describes various functions of resources, and provides a unified resource function view for the cloud manufacturing platform; the meta-model layer defines resource-related concepts and the relationship between resources; The multi-layer cloud manufacturing resource modeling framework with multi-granularity features includes many-to-many mapping and semantic transformation: The many-to-many mapping relationship, associating the resource model layer with the functional model layer, improves the flexibility of resource modeling by decoupling cloud manufacturing resources from their corresponding functional features, and dynamically joins or exits the cloud manufacturing system, which is in line with the cloud manufacturing system The dynamic characteristics of a resource has multiple functional characteristics, and a functional characteristic is provided by different resources, and the many-to-many mapping relationship reflects the multi-functional, multi-task, and multi-purpose characteristics of resources; Semantic conversion relationship, associating the functional model layer with the semantic-based meta-model layer, and semantically annotating heterogeneous resource information through the concepts and relationships defined in the meta-model layer; The resource model layer receives the heterogeneous resource information registered by the distributed enterprise, designs the XML schema corresponding to the resource information according to the original resource description method of the enterprise, and converts the heterogeneous resource information into the corresponding XML document, so that the heterogeneous resource information is the same structured; The function model layer extracts the resource function information from the XML document corresponding to the resource, merges the function information of similar resources, designs the XML schema corresponding to the function information according to the result of the function information merger, and establishes the function information XML document view; The metamodel layer extends the general ontology components, provides concepts, relationships, constraints, and axioms to mark resource-related semantics; defines the semantic information of cloud manufacturing resources and the semantics of non-functional features of resources from the four aspects of concepts, relationships, constraints, and axioms Information, resource functional feature semantic information with multi-granularity characteristics, cloud manufacturing context semantic information and semantic information supporting cloud manufacturing service level agreement mechanism; Many-to-many mapping relationship For each registered resource in the resource model layer, the resource function view is searched sequentially, and the resource is linked to the corresponding functional characteristics. If there is no functional characteristic corresponding to the registered resource, it is created in the resource function view The new functional feature item links the resource with the new feature item, and records the link relationship between the resource and the functional feature in the resource function mapping table; for each functional feature in the resource function view, look up the resource function mapping table to provide this Link all the resources of the functional characteristics with the corresponding resources, and record the associated information in the resource function mapping table; if there is no resource providing this functional characteristic in the resource function mapping table, delete the functional characteristic item. 2. The multi-layer cloud manufacturing resource modeling framework with multi-granularity characteristics as claimed in claim 1, wherein the semantic transformation relationship is based on the concept defined in the semantic meta-model, and the equivalent vocabulary involved in the resource and functional characteristics Labeling; according to the relationship defined in the semantic metamodel, label the relationship between resources and functional characteristics; according to the constraints defined in the semantic metamodel, describe the constraints that resources and functional characteristics need to meet; according to The axioms defined in the semantic meta-model mark the logical relationship between resources and functional characteristics, and provide machine-understandable reasoning information.