基于UML、构件技术的DPIDS的研究与实现
|
摘要
本文为解决智能农业诊断系统重复开发的问题,研究开发了智能农业诊断平台。为能够高效率地开发该珍断平台,并保证该平台具有良好的扩展性能,本文从UML建模方法、面向对象设计模式、基于框架元模型的构件接口技术等三个方面对该平台的设计与实现进行了研究,取得了以下成果:
1、基于UML的平台建模
UML是目前最常用的一种面向对象统一的建模语言。在智能农业诊断平台的开发中统一采用UML进行建模,使得分析人员、设计人员、开发人员在一个一致的平台上进行交互,有利于团队的沟通和信息共享,大大提高了智能农业平台的开发效率。在设计中广泛采用了UML用例图、时序图、类图、构件图、状态图等强有力的表示手法,逐步明晰了平台的体系结构,辅助设计出了扩展性能良好、耦合度小、内聚性强的模块和结构。
2、基于框架元模型的构件接口描述
框架元模型是一种用于描述应用框架构造和组成的元模型。框架的内部元素包括扩展点、设计模式、构件、控制流。框架中支持灵活扩展和定制的机制称为框架的扩展点,支持符合应用系统特定需求的实现。在平台构件的设计中采用框架元模型对扩展点进行的分类,从数据扩展点、功能扩展点、界面扩展点、业务扩展点四个方面对构件的接口进行管理,确保了构件具有满足需求的扩展性能。
3、模块数据与界面的定制策略
为了实现平台的功能,能够开发出适应于不同农业领域的诊断系统,设计了模块数据定制模块和界面定制模块。通过这两个模块的实现,使用户可以通过定制向导加载适应于不同领域的模块数据,并能在运行时对界面风格进行设置,使系统适用于不同的应用环境。
4、智能农业诊断平台体系结构
智能农业诊断平台设计了基于三个构件层的框架。最底层是农业诊断专家系统的各个数据处理构件,封装了专家系统中知识库处理、解释处理、推理处理等基本的数据处理过程。第二层构件是面向用户的构件,通过胶合底层构件的方式来实现专家系统的基本功能,包括智能决策、智能查询等构件。第三层是平台定制级的构件,包括两个方面的定制,一个是数据定制,一个是界面定制,采用元
This paper addresses some core technologies in developing a flexible platform of intelligent agriculture diagnosis system, which includes UML modeling method, object-oriented design patterns, component technologies based on frame meta-mode to love the problem of repeat development in intelligent agriculture diagnosis field. The work can be summarized as the following: 1,platform modeling based on UMLUML is the most popular standard modeling language at present. Modeling with UML provides a coherent interface and makes the communication of analysers, designers,developers more convenient, which is in favor of communication and information share among the group and improves the efficiency of developing intelligent agriculture platform greatly. This paper adopts use case frame, sequence frame, class frame, component frame, state frame to analyze the achitecture of the platform,improve the expansibility and reduce coupling of the modules. 2,component interface description based on frame meta-modelFrame meta-model is used to describe the structure and combination of apply frame. Frame includes inner elements of hot-spot, design pattern, component, control flow etc. Hot-spot in the frame is a method for supporting flexible expanse and customization.This paper adopts frame meta-model to manage the hot-spots, classify the hot-spots into four classes-data hot-spots, function hot-spots,interface hot-spots and business hot-spots, which insures the expansibility of the components. 3,customization of module data and user interfaceWe design module date customizing module and user interface customizing module to realize the main function of this platform to develop diagnosis systems of different algriculture fields. Users of this platform can use these two modules to upload module data of different fields and set interface style to make the system adapt to different applicaiton entironment. 4,the architecture of intelligent agriculture diagnosis platformIntelligent agriculture diagnosis platform includes a frame based on three
component layers. The lowest is data processing component layer, which encapsulate the modules of knowledge base process, explaining process, reasoning process. The second layer includes the components which communicate with users directly, for example, intelligent decision-making module,intelligent search module,etc. The third layer includes platform customizing components, which are used to customize data and user interface and adopt meta database to manage these interface elements.
1、基于UML的平台建模
UML是目前最常用的一种面向对象统一的建模语言。在智能农业诊断平台的开发中统一采用UML进行建模,使得分析人员、设计人员、开发人员在一个一致的平台上进行交互,有利于团队的沟通和信息共享,大大提高了智能农业平台的开发效率。在设计中广泛采用了UML用例图、时序图、类图、构件图、状态图等强有力的表示手法,逐步明晰了平台的体系结构,辅助设计出了扩展性能良好、耦合度小、内聚性强的模块和结构。
2、基于框架元模型的构件接口描述
框架元模型是一种用于描述应用框架构造和组成的元模型。框架的内部元素包括扩展点、设计模式、构件、控制流。框架中支持灵活扩展和定制的机制称为框架的扩展点,支持符合应用系统特定需求的实现。在平台构件的设计中采用框架元模型对扩展点进行的分类,从数据扩展点、功能扩展点、界面扩展点、业务扩展点四个方面对构件的接口进行管理,确保了构件具有满足需求的扩展性能。
3、模块数据与界面的定制策略
为了实现平台的功能,能够开发出适应于不同农业领域的诊断系统,设计了模块数据定制模块和界面定制模块。通过这两个模块的实现,使用户可以通过定制向导加载适应于不同领域的模块数据,并能在运行时对界面风格进行设置,使系统适用于不同的应用环境。
4、智能农业诊断平台体系结构
智能农业诊断平台设计了基于三个构件层的框架。最底层是农业诊断专家系统的各个数据处理构件,封装了专家系统中知识库处理、解释处理、推理处理等基本的数据处理过程。第二层构件是面向用户的构件,通过胶合底层构件的方式来实现专家系统的基本功能,包括智能决策、智能查询等构件。第三层是平台定制级的构件,包括两个方面的定制,一个是数据定制,一个是界面定制,采用元
This paper addresses some core technologies in developing a flexible platform of intelligent agriculture diagnosis system, which includes UML modeling method, object-oriented design patterns, component technologies based on frame meta-mode to love the problem of repeat development in intelligent agriculture diagnosis field. The work can be summarized as the following: 1,platform modeling based on UMLUML is the most popular standard modeling language at present. Modeling with UML provides a coherent interface and makes the communication of analysers, designers,developers more convenient, which is in favor of communication and information share among the group and improves the efficiency of developing intelligent agriculture platform greatly. This paper adopts use case frame, sequence frame, class frame, component frame, state frame to analyze the achitecture of the platform,improve the expansibility and reduce coupling of the modules. 2,component interface description based on frame meta-modelFrame meta-model is used to describe the structure and combination of apply frame. Frame includes inner elements of hot-spot, design pattern, component, control flow etc. Hot-spot in the frame is a method for supporting flexible expanse and customization.This paper adopts frame meta-model to manage the hot-spots, classify the hot-spots into four classes-data hot-spots, function hot-spots,interface hot-spots and business hot-spots, which insures the expansibility of the components. 3,customization of module data and user interfaceWe design module date customizing module and user interface customizing module to realize the main function of this platform to develop diagnosis systems of different algriculture fields. Users of this platform can use these two modules to upload module data of different fields and set interface style to make the system adapt to different applicaiton entironment. 4,the architecture of intelligent agriculture diagnosis platformIntelligent agriculture diagnosis platform includes a frame based on three
component layers. The lowest is data processing component layer, which encapsulate the modules of knowledge base process, explaining process, reasoning process. The second layer includes the components which communicate with users directly, for example, intelligent decision-making module,intelligent search module,etc. The third layer includes platform customizing components, which are used to customize data and user interface and adopt meta database to manage these interface elements.
引文
1、胡文蕙,赵文,张世琨等。基于构件技术的应用框架元模型的研究。软件学报,2004,15(1):1～8。
2、青润。软件工程之全程建模实现。电子工业出版社,2004。
3、Hans-Erik Eriksson Magnus Penker著。夏昕,何克清译。UML业务建模。机械工业出版社,中信出版社,2003。
4、何清法,李国杰,焦丽梅等。基于关系结构的轻量级工作流引擎。计算机研究与发展,2001年2月,第38卷。
5、Alan W. Brown, Kurt C. Wallnau. The Current State of CBSE. IEEE Software[J]1998
6、Joseph Giarratano, Gary Riley. Expert Systems Principles and Programming.1998
7、胡文蕙+,赵文,张世琨,王立福。基于构件技术的应用框架元模型的研究。软件学报[J],2004
8、Maria Taboada, Julio Des and Jose Mira, Roque Marin。Diagnosis Systems in Medicine With Reusable Knowledge Components. IEEE INTELLIGENT SYSTEMS[J] 2001
9、赵春江。农业智能专家系统平台。兰州大学学报。
10、Erich Gamma, Richard Helm, Ralph Johnson, John Vlissides. Design Patterns 2003
11, Johnson RE. Frameworks=(Components+Patterns). Communications of the ACM, 1997,40(10):39-42.
12, Rumbaugh J, Jacobon I, Booch G. The UML Reference Manual. New York: Addison-Wesley, 1999.
13, Andert G. Object frameworks in the Taligent OS. In: Proc. of the Compcon'94. Los Alamitos: IEEE CS Press, 1994. 112～121.
14, Architecture of the San Francisco frameworks. 1998. http://www.research.ibm.com
15, Cotter S, Potel M. Inside Taligent Technology. New York: Addison-Wesley, 1995.
16, Carnegie Mellon University. The Software Product Line Practice Version 4.1, 1997. http://www.sei.cmu.edu
17, Braga RTV, Masiero PC. An approach for frameworks construction and instantiation using pattern languages. In: Proc. of the Int'l. Conf. on Computer Science, Software Engineering, Information Technology, e-Business, and Applications. ACIS, Foz do Iguacu, 2002. 305~310.
18,Gamma E, Helm R, Johnson R, Vlissides J. Design Patterns: Elements of Reusable Object-Oriented Software. New York: Addison-Wesley, 1995. 175~184; 315~324.
19,Fayad M, Schmidt DC. Object-Oriented application frameworks. Communications of the ACM, 1997,40(10):43~ 54.
2、青润。软件工程之全程建模实现。电子工业出版社,2004。
3、Hans-Erik Eriksson Magnus Penker著。夏昕,何克清译。UML业务建模。机械工业出版社,中信出版社,2003。
4、何清法,李国杰,焦丽梅等。基于关系结构的轻量级工作流引擎。计算机研究与发展,2001年2月,第38卷。
5、Alan W. Brown, Kurt C. Wallnau. The Current State of CBSE. IEEE Software[J]1998
6、Joseph Giarratano, Gary Riley. Expert Systems Principles and Programming.1998
7、胡文蕙+,赵文,张世琨,王立福。基于构件技术的应用框架元模型的研究。软件学报[J],2004
8、Maria Taboada, Julio Des and Jose Mira, Roque Marin。Diagnosis Systems in Medicine With Reusable Knowledge Components. IEEE INTELLIGENT SYSTEMS[J] 2001
9、赵春江。农业智能专家系统平台。兰州大学学报。
10、Erich Gamma, Richard Helm, Ralph Johnson, John Vlissides. Design Patterns 2003
11, Johnson RE. Frameworks=(Components+Patterns). Communications of the ACM, 1997,40(10):39-42.
12, Rumbaugh J, Jacobon I, Booch G. The UML Reference Manual. New York: Addison-Wesley, 1999.
13, Andert G. Object frameworks in the Taligent OS. In: Proc. of the Compcon'94. Los Alamitos: IEEE CS Press, 1994. 112～121.
14, Architecture of the San Francisco frameworks. 1998. http://www.research.ibm.com
15, Cotter S, Potel M. Inside Taligent Technology. New York: Addison-Wesley, 1995.
16, Carnegie Mellon University. The Software Product Line Practice Version 4.1, 1997. http://www.sei.cmu.edu
17, Braga RTV, Masiero PC. An approach for frameworks construction and instantiation using pattern languages. In: Proc. of the Int'l. Conf. on Computer Science, Software Engineering, Information Technology, e-Business, and Applications. ACIS, Foz do Iguacu, 2002. 305~310.
18,Gamma E, Helm R, Johnson R, Vlissides J. Design Patterns: Elements of Reusable Object-Oriented Software. New York: Addison-Wesley, 1995. 175~184; 315~324.
19,Fayad M, Schmidt DC. Object-Oriented application frameworks. Communications of the ACM, 1997,40(10):43~ 54.