基于液滴参数检测的涡街湿气过读预测模型

为提高两相涡街湿气测量精度，针对传统涡街过读公式预测精度差、适用范围受限问题，提出基于夹带液滴参数（夹带率和粒径）的涡街过读预测模型。为进行不同夹带液滴工况的实验研究，建立基于雾化混合的可调压环雾状流实验装置，并建立光学图像法测量系统，获得液滴直径及其分布信息。结合环雾状流型及涡街过读机制，考虑液滴-液膜传质和液滴-旋涡耦合作用，提出影响涡街过读的无量纲尺度参数。建立基于液相加载量、韦伯数和斯托克斯数的涡街过读预测模型，将夹带液滴参数和载气参数（密度和速度）的影响考虑在内，理论上可拓展公式适用范围。最后，评估现有过读关联式的预测性能，并结合实验和模型假设中夹带液滴参数的差异进行详细分析，进一步确认了夹带率和粒径对涡街过读特性的重要影响。结果表明：所提模型在不同液滴夹带条件下都给出了很好的预测，相对偏差在±1.0%以内，预测精度和可拓展性较其他公式有了较大提高。

A new overreading model for wet gas vortex metering considering entrained droplet flow parameters

To improve the wet gas metering accuracy by using a vortex flowmeter, aiming at the problems of low prediction accuracy and narrow application scope of the traditional meter overreading correlations, a meter overreading model based on flow parameters of the entrained droplets (entrainment fraction and droplet size) was proposed. To conduct experimental studies on different entrained droplet conditions, the annular mist flow loop based on atomizing mixing was developed, and the optical imaging system was built to measure the droplet diameter and distribution. The dimensionless scaling parameters were discovered by combining the annular mist flow pattern with the vortex meter overreading mechanism and taking into account the droplet-liquid film mass transfer and the droplet-vortex coupling mechanism. The meter overreading model was developed with liquid mass loading, Weber number and Stokes number, taking the effects of the entrained droplet parameters and the carried gas parameters (density and velocity) into consideration, thus, its application scope is expanded. Finally, the performance of other overreading correlations was evaluated and analyzed by the differences of the entrained droplet parameters in their experimental conditions and the model assumptions. The results indicate that the proposed model provides a uniform prediction for the meter overreading, the relative deviations are within ±1.0% error band, both the predicted accuracy and the model extensibility are greatly improved compared with other overreading correlations.