在双截面电容层析成像(ECT)系统测量气力输送过程中栓塞流速度的应用中,利用动力学相关因子指数检测电容信号的均值突变,根据突变点设置时间窗以提取上下游信号,将所选信号映射成二值图像,并进行数学形态学闭运算处理,对处理后的图像进行二维互相关计算,估算出栓塞流速度。实验结果表明了该方法的有效性,和传统互相关法相比,改进互相关法可有效提取有明显波动特征的信号段,并通过二维图像互相关计算提高了互相关法的稳定性,从而提高了ECT技术在线测量两相流的准确性。
In the application of twin-plane electrical capacitance tomography (ECT) system to velocity measurement of plug flow in pneumatic conveying process, the dynamic lag correlation exponent was used to detect the average value abrupt changes of capacitance, and the detected abrupt change-points were used to determine the time windows for selecting proper upstream and downstream signals. The selected signals were thereafter mapped into binary images, which were then processed with mathematical morphology closed operation. Finally, the velocity of plug flow was estimated by implementing two dimensional cross correlation between the processed images. Experimental results demonstrated the effectiveness of the proposed method. Compared to the traditional cross-correlation method, the modified method could extract signals containing obvious fluctuations effectively, and improve the robustness of the cross correlation method by using two dimensional image correlation, enhancing the reliability of online measurement of two-phase flow using ECT.