摘要
为解决船舶综合监控系统中海量无序数据信息的有效利用问题,以中央冷却系统为研究对象,提出D-S证据理论结合粗糙集的信息融合方法对系统运行中的故障隐患状态进行识别.分别定义了等价属性和证据决策系数的概念,给出快速约简方法及基于证据信任度的证据合成方法,明确了证据基本可信度分配的求解过程.基于实际船舶典型状态数据的测试评估结果表明,所得结论与基于先验知识的判断基本一致,验证了所提方法对系统隐患故障状态评估的有效性.
In order to solve the problem of effective utilization of mass disorder data information in ship integrated monitoring system, taking the ship central cooling system as study objectives, and the information fusion method combining Dampster- Sharer evidence theory and Rough sets was proposed to identify the hidden trouble states in the operation of the system. The equivalence attribute and evidence-decision-coefficient were defined respectively, and a fast reduction algorithm and evidence synthesis method based on evidence confidence were derived. The solving procedure of basic probability assignment was specified. The testing evaluation results based on typical status data of real ship central cooling system show that the conclusions are basically consistent with those based on prior knowledge, which verifies the effectiveness of the proposed method in evaluating the state of hidden trouble in the system.
出处
《大连海事大学学报》
CAS
CSCD
北大核心
2017年第4期89-96,共8页
Journal of Dalian Maritime University
基金
国家自然科学基金资助项目(51479017)
中央高校基本科研业务费专项资金资助项目(3132015210
3132016368)
关键词
中央冷却系统
状态评估
粗糙集
D-S证据理论
信息融合
central cooling system
operating condition evaluation
rough sets
D-S evidence theory
information fusion
