针对传统方法难以实现对海量数据环境下的设备状态评价的问题,提出了一种基于海量数据融合的设备状态评价方法.首先,利用擅长处理海量数据的分布式聚类算法K-means对海量状态数据进行预处理为多个簇,并求出各个簇的质心作为该簇的代表信息;然后对代表信息进行加权处理;最后利用证据理论对加权的代表信息进行融合,从而决断出设备的状态.通过仿真实验结果表明,该方法能对海量信息进行有效融合,并能更合理地决断出设备的状态信息.%In view of the Traditional method is difficult to realize the evaluation of equipment state in massive data environment, this paper proposes a method of equipment state evaluation based on massive data fusion.First, Using the distributed clustering algorithm K-means, which is good at dealing with massive data, to pre-process the massive data., and calculating the centroid of each cluster as the cluster representative information;Then, it is imperative to weighted processing for representative information;finally, it uses the evidence theory to fuse the weighted representation information, and then the state of the device is determined.The simulation results show that the method can fuse massive information effectively, and can make more reasonable decision for the state information.
展开▼
