斯米尔诺夫检验法在数据级信息融合数据关联中的应用

在信息融合系统中，为了确保进入融合中心的信息是满足融合需要的有效信息，在融合之前要对来自多个信息源的数据进行数据关联。作为信息融合前的信号预处理过程，利用数理统计中的斯米尔诺夫检验法对多个同类传感器测得的加速度响应信号进行数据关联，去掉故障传感器或受到严重干扰的通道数据，确保进入融合中心的数据能够有效反映被测对象的状态。通过某二级齿轮箱实验台实测数据的验证，证明该方法能够有效地检验出同类传感器信号是进行故障诊断的有效信号。

Application of Smirnov Test in Data Association for Data level Information Fusion

To ensure the informations entering into the fuse center to be effective to meet the demand of information fusion,they must be associated before fusion process.As a signal pre-process before information fusion Smirnov test is made to associate multiple acceleration response signals from homogeneous sensors to eliminate faulty or lnterfered channel data and to make sure that the data entering into the fuse center contain effective state information of UUT(Unit Under Test).Via measurements on a two-level gearbox testbed,the effectiveness of the method is verified.