Study On Service-oriented Sharing And Reuse Of Geo-analysis Model

Geo-analysis models are the abstracts and simplifications of the real world.Many researches have been proved that geo-analysis models are the effective methods for simulating the geographical phenomenon and processes.To date,geography and related domains have accumulated massive geo-analysis models with different purposes,research area,and spatiotemporal scale.For models from different domains,these models are heterogeneous in forms and structures,which leads to the obstacles in sharing and reuse.These models finally become “model islands”.The “model islands”lead to model re-developing,which cause massive waste of related resource and restrict the development of geography.For the geo-analysis models following the specific standards,they can not be reused by other standards,which also causes the problem for modeling sharing and reuse in solving complex geographic issues.With the development of web,the service-oriented sharing of geo-analysis models becomes research trend of geography.Related research progresses can reduce the cost of models applied in geographical researches and avoid the waste of related resource.And it would contribute to the geographical simulation tasks with different domains on the web.So,service-oriented geo-analysis models have become effective methods for model sharing and reuse on the web.This research focus on the service-oriented model,analyze the features of geoanalysis models,explore the usage requirement,and summary the significance of model sharing and reuse.This research design the standard of model description and standardlized service-oriented encapsulation and packaging for sharing.Then,based on the model encapsulation and model interaction design,this research publish model services to make the heterogenous geo-analysis model as web services for model reusing.And,aiming the standardlized models,this research designs and develops the interoperation engines to interoperate models with other standards,which can bridge the gap among different standards.Finally,by the analysis of distributed network used by geographical simulation,this research designs the services running strategies of geoanalysis model in distributed network,which can involve simulation resource into geographical simulation.The key points of this research are listed below.(1)Geo-analysis model summary for model sharing and reuse.At first,based on the the user's understanding and usage,this research summarized the types and attributes of geo-analysis model.Secondly,the application requirements of geographic analysis model are analyzed,and the application requirements of single model simulation and multi model process simulation are summarized.Finally,this research analyzes the connotation of model sharing and reuse from the two processes of geoanalysis model sharing and reuse.(2)Service-oriented method design for geo-analysis model sharing.At first,considering different domains,this research design the methods for model description with several domains.Based on the methods,this research designs a set of structural fields for model description for sharing.Then,with the help of standardlized description of model,this research designs the encapsulation method for hetereogeneous geoanalysis models.Based on standardlized component design and universal protocol,this research designs and develops the method for standardlized encapsulating and wrapping.(3)Geo-analysis model reusing on the web.This research designs and develops the model service container and service-oriented interfaces for model reusing on the web.Then,this research summaries the features of different standards,and design a universal standard by community sharing.And,based on the universal standard,this research designs and develops several interoperation engines among different standards.The interoperation engines have three modules,including fields mapping,functions converting,and components re-organzing,to help users interoperate models among different standards.(4)Distributed model running for geographical simulation.Based on the requirement of model running,this research design the strategies of simulation resource access,configuration,and management.For the requirement of task running,this research design the configuration method for geographical simulation.And,in the simulation task running,this research designs the controlling strategies and methods to balance the task loading among different computing nodes,and address the users' requirement.This research focus on the model knowledge analysis,sharing,reusing,and distributed simulation,which help users share and resue geo-analysis models through the requirement of geographical simulations.Based on the content above,this research designs a set of systems and tools.With the development of such systems and tools,this research validates the effectiveness and practicability.Based on the developement of sharing and reusing geo-analysis models on the web,this research promotes the sharing and reuse of geo-analysis models in different domains,thus contribute to comprehensive geographic simulation in complex scenes.