基于二进制位缺失标识与改进D-S证据理论的数据融合技术研究

Data Fusion Based on Binary Identifier and Improved D-S Theory

近年来电网公司大力推进生产域信息化,目前已经建设并推广了一批信息化系统,涵盖了各专业领域,积累了大量业务数据和实时状态监测数据,但数据分散在各个系统中。如何充分挖掘海量数据潜在的价值,为电网的规划、建设、运营、改造、用户接入等提供有效的依据,是亟需解决的问题。基于此,研究了基于二进制位缺失标识的数据抽取技术,减少了50%以上的内存占用;提出了改进D-S证据理论的数据融合技术,实现高精度匹配(精度提高到85%以上),从而为优化电网生产运营模式、提升电网公司精益化管理水平提供技术支撑。

In recent years,the power grid has vigorously promoted the informatization,with a number of informatization systems,covering various professional fields.It has accumulated a large number of business data and real-time status data.However,the data is scattered in various systems.It is necessary to fully make use of the massive data,to provide effective basis for planning,construction,operation,transformation and user access of power grid.Therefore,data extraction based on binary identifier is researched to achieve less than 50%memory occupation,and data fusion aimed to improve D-S theory is proposed to achieve high-precision matching(the accuracy is improved to more than 85%),so as to optimize operation modes and improve lean management of power grid.