多传感器数据融合方法在加力燃烧室试验中的应用研究

在航空发动机加力燃烧室试验中,为实现同一截面真实气流状态参数的多传感器精确测量,提出了1种多传感器数据融合方法。该方法运用格拉布斯准则剔除多传感器测量数据中的无效数据,运用方差自适应加权融合方法对有效数据进行融合计算并获取真实参数,并分别利用算术平均值方法和自适应加权融合方法对试验数据进行处理分析。结果表明:通过多传感器融合方法获取的数据均方差明显优于算术平均值方法获取的方差,具有更高的测试精度和可靠性。

Research on Multi-Sensor Data Fusion Method for Afterburner Test

In order to effectively realize the precision measurement of actual air flow condition of the same section when using multisensor system in afterburner test. The method eliminates suspicious data first based on Grubbs, and then employs an adaptive weighted fusion estimated algorithm of multi-sensor to fuse effective data. A comparison between the effect of new multi-sensor fusion method and arithmetic mean value method was conducted. The comparison results reveal that new multi-sensor fusion method have obvious advantage on mean square deviation over arithmetic mean value method. Therefore, new multi-sensor fusion method has higher test accuracy and reliability.