复杂装备状态监测实时流数据处理框架  

Real-time stream data processing framework for complex equipment condition monitoring

作　　者：庄雪吟 张力[1] 翁晓奇[1] 李虎斌[1] 刘英博[1] 

机构地区：[1]清华大学软件学院,北京100084

出　　处：《计算机集成制造系统》2013年第12期2929-2939,共11页Computer Integrated Manufacturing Systems

基　　金：国家863计划资助项目(2012AA040911);国家科技支撑计划资助项目(2012BAF12B23)~~

摘　　要：为了解决数据处理层的高通量实时流处理、数据传输协议定制、海量监测数据存储、动态弹性计算等问题,提出一种基于物联网的复杂装备状态监测实时流数据处理框架。采用模型驱动与构件化设计方式,框架可分为数据层、运行层、模型层、工具层、应用层以及其他辅助模块。运行层采用流水线并行计算的方式进行状态监测流数据处理;模型层通过定义处理单元、拓扑模型和协议模型等,对系统数据处理业务及数据传输协议进行定制。数据层采用非结构化数据库自由表方式,进行海量监测数据存储。通过在工程机械领域状态监测中的应用,验证了该框架的有效性。To solve the problems of data processing layer such as high-throughput real-time stream data processing, data transfer protocol, massive data storage and elastic computing, a real-time stream data processing framework based on Internet of Things （IoT） was proposed. By using model driven and modular software design approaches, the framework was divided into data layer, runtime layer, model layer, instrument layer, application layer and other assistant modules. The runtime layer monitored the framework processes stream data with pipeline computing tech nology; the model layer customized data process procedures and data transfer protocols by defining process unit, to pology model and protocol model; the data layer applied free tables of unstructured database to store massive data of device conditions. The application of condition monitoring for construction machinery showed the effectiveness of the proposed framework.

关 键 词：复杂装备 状态监测 物联网 流数据处理 拓扑模型 协议模型 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论] TP311[自动化与计算机技术—计算机科学与技术]