基于人工神经网络的复杂海工装备项目工作结构分解  

Work structure decomposition of complex marine engineering equipment projects based on artificial neural network

作　　者：李敬花[1] 茆学掌 张涛 

机构地区：[1]哈尔滨工程大学船舶工程学院,黑龙江哈尔滨150001 [2]北京中船信息科技有限公司,北京100037

出　　处：《计算机集成制造系统》2017年第7期1511-1519,共9页Computer Integrated Manufacturing Systems

基　　金：工信部高技术船舶科研资助项目(2012[545])~~

摘　　要：为解决复杂海工装备项目工作结构分解困难、分解结果不满足生产要求等问题,采用人工神经网络方法分析项目数据和项目工作结构的关联性,建立项目特征参数与项目工作结构的关联关系,实现项目工作结构的智能化分解。针对海工装备结构和建造特点,采用有向无环法建立了海工装备项目工作结构分解模型;引入"域"的概念并基于项目数据建立了海工装备项目域;设计了分层模块化人工神经网络结构,并利用项目域数据进行了神经网络训练和测试,得到了稳定的神经网络;通过实例验证了该方法的有效性,解决了目前海工企业项目工作结构分解耗时耗力的问题。To solve the problem of decomposition difficulity and unfavorable result of complex marine equipment project,the artificial neural network method was used to analyze the relevance between project data and work structure and establish the correlations between characteristic parameter and work structure,which could achieve the intelligent decomposition of work structure.Aiming at the structure and construction features of marine equipment,the marine engineering equipment project decomposition model was established by directed acyclic.The concept of domain was introduced and the project domain of marine engineering equipment was build based on project data.Furthermore,the framework of artificial neural network was designed,and the training and testing of neural network based on project domain data were made to get the stable neural network.The validity of the proposed method was proved by an example,which effectively solved the problem of time-consuming and labor-intensive work structure decomposition in marine engineering enterprises.

关 键 词：工作分解结构 海工装备项目 人工神经网络 智能分解 项目管理 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]