中文摘要：

用科氏质量流量计测量含气介质时测量结果重复性下降,测量数据的波动性变大.测量了大量的气液两相流瞬时质量流量波动数据,基于Hilbert-Huang变换对流量波动数据进行了经验模态分解,获得了各阶固有模态函数;对固有模态函数进行了Hilbert变换,得到了时间-频率-振幅谱;分析了固有模态函数的能量分布.研究结果表明,气液两相流瞬时质量流量的波动与含气率有关,对于不同的含气率工况,固有模态函数的能量分布不同.气泡分布越均匀,测量管受力越均匀,瞬时质量流量的IMF分量波动越小,标准差也越小.较高的含气率使得测量管受力变化较大,气塞的高速流动导致测量值波动变大,标准差增大,影响测量的稳定性.不同的含气率其每个IMF分量所占的能量比例不同.含气率较小其气泡分布均匀,能量集中在高频段,且分布均匀;随着含气率的增大,气泡大小和流动速度不均,能量分布也不均.若形成稳定的大气泡和小气泡,瞬时质量流量波动则呈现出一定的周期性,能量主要集中在高频分量上.

英文摘要：

When coriolis mass flow meter is used to measure gas-liquid two-phase flowrate,the measurement reproducibility decreases and the fluctuation increases.Amount of instantaneous gas-liquid two-phase flowrate data are measured using coriolis mass flow meter.Empirical mode decomposition （EMD） is adopted to process the fluctuating flowrate data based on Hilbert-Huang Transform,and the Intrinsic Mode Functions （IMFs） are obtained.Hilbert Transform of IMF is executed and the time-frequency-amplitude figures are plotted.The energy distributions of IMFs are analyzed.The results show that the fluctuation characteristics of instantaneous gas-liquid two-phase flowrate are related to voidage.Different voidage conditions correspond to different energy distribution of IMF.The more uniform the gas bubbles distribute,the more uniform force the measuring tube bears,so the IMF component is less fluctuant,and the standard deviation also has less error.The higher void fraction makes the tube bear much force; air plug＇s fast movement generates bigger measurement value and bigger errors,so affecting the measuring stability.Each IMF component of different void fractions has different percentage of energy.The smaller fraction produces more uniform distribution of the gas bubbles,and so the energy uniformly gathers in the higher frequency range.With the fraction increasing,the gas bubble sizes and flowrate are not becoming uniform,and so does the energy distribution.If there is stable big or small bubbles,the instantaneous flowrate fluctuation shows periodicity,and the energy gathers in the high frequency component.