应用数据融合方法辨识油气两相流流型

基于D-S证据理论和逐步回归方法,结合直管段差压波动信号的功率谱密度分布特征和电容传感器提供的电容信息,给出了一种油气两相流流型辨识的数据融合方法.通过差压传感器和电容传感器分别采集直管段差压波动信号和管道截面相分布信息,利用提取的特征参数和已经建立的与流型相关的电容关联式对各传感器的信任度函数进行分配,再利用D-S证据理论进行融合,得到油气两相流流型辨识结果.单传感器辨识结果与数据融合辨识结果的比较结果显示,数据融合方法具有一定的优越性.

Flow Pattern Identification of Oil-Gas Two Phase Flow Based on Data Fusion Technique

Based on D-S evidence theory and stepwise regression method, combining the power spectrum density function characteristics of differential pressure fluctuation signals and capacitance information from capacitance sensor, a new flow pattern identification method of oil-gas two phase flow was proposed. The eigenvalues extracted from differential pressure fluctuation signals and flow pattern based capacitance relation developed upon capacitance information were used to determine the belief function assignment, and then, the final identification result was obtained by using D-S evidence theory. Comparing the result of the flow pattern identification based on the separate original data with that based on the fused data, the result shows that the proposed method has certain advantage over the previous methods.