石油天然气作为人类世界的主要能源之一，对其生产和输送过程中油、气、水流量的计量是监测、控制油井和油藏动态特性的主要依据。而发展实时在线多相流流量检测技术对提高能源行业生产管理水平、开发海洋边际油气田、降低生产管理成本等具有重要意义。目前多相流在线检测技术的研究主要存在以下问题：(1)传感器结构设计目标主要是提高成像精度，而对提高多相流过程参数测量精度的研究较少；(2)难以实现测量精度和传感器成本、适用范围之间的平衡。需要发展安全性高、成本低、测量精度满足工业需求的多相流检测传感器及计算模型；(3)对不同工况下流体流动状态还没有清楚的认知；(4)大多集中于对单个传感器有效工作范围内多相流过程参数的检测，而无法实现全量程范围内相含率的计量；(5)对油-气-水三相流的研究较少。基于国家自然科学基金项目“油-气-水多相混输管内流水合物生-消过程可视化识别”和深圳市孔雀项目“海洋高含气井间原油采出物智能化测量技术”，本文基于多传感器数据融合，选用了安全性高、成本低的电学传感器和文丘里管，建立了多相流过程参数在线检测模型。具体工作包括以下几项：(1) 基于ECT系统敏感场的软场特性，设计新型电容传感器结构实现对环状流含水率和含气率的检测。并通过仿真数据对传感器结构参数进行了优化设计。(2) 融合ECT和ERT传感器测量数据建立了油-水两相流全量程相含率计算模型，结合文丘里管差压信号实现油-水两相流总体积流量和各分相体积流量的计算。在测量过程参数的同时，利用图像重建算法分析了不同工况下油-水两相流的流动特性。(3) 利用ECT测量数据建立不依赖于流型识别的相含率计算模型，并考虑相间滑差对结果的影响。(4) 分析ECT和ERT测量数据随油-气-水三相流相含率的变化规律，给出有效工作范围，并建立相含率计算模型。(5) 建立了基于虚高系数修正的混合流量模型，实现气-液两相流各分相体积流量的计量。
Petroleum and natural gas is one of the main sources of energy for people. The measurement of multiphase flow which widely existed in the production processes of petroleum industry is very important for monitoring and controlling the dynamic characteristics of oil wells and reservoirs. Developing online measurements of flow parameters such as phase fraction and flow rate of each individual phase without separating the mixture, are crucial to improve the management level of industry, reduce the cost and develop marine oil and gas.At present, the research of multiphase flow on-line measurement mainly has the following problems: (1) The design of sensors mainly aims at improving the accuracy of reconstructed image, and focuses little on the accurate measurement of multiphase flow. (2) It is difficult to balance measurement accuracy and cost. So it is very desirable to develop sensor and model with high safety and low cost to measure multiphase flow. (3) The fluid in different conditions has not be well understood. (4) Most of the single sensor can only measure limited range of phase fraction and be invalid in some conditions of multiphase flow. (5) The research focuses little on gas-oil-water three-phase flow. Based on the National Natural Science Fund “ Identification of hydrate generation-elimination process for gas-oil-water mixture” and Peacock Project of Shenzhen “intelligent measurement technology for crude oil extraction from marine high gas well”, this paper establishes the model of multiphase flow measurement by multi-sensor data fusion technology which includes electrical tomography sensor and venturi tube. The main work is as follows: (1) Based on the “soft field” attribution of sensitivity map for ECT system, a capacitance sensor has been developed to obtain the phase fraction of annular flow pattern. And the parameters of sensor are optimized through simulation data. (2) A new method based on data fusion of ECT and ERT has achieved the full range phase fraction measurement. And then, by combining differential pressure obtained from venturi tube, the volume flow rate of each phase can be calculated.(3) Based on the reconstructed image of ECT, a phase fraction calculation model which does not depend on flow pattern recognition is established. In this method, the effect of slip ratio between different phase are also considered to modify the void fraction. (4) For gas-oil-water three-phase flow, this work analyses the change rule of measurement data with the phase fraction. Based on large number of dynamic experiments, the work rang of ECT and ERT are investigated, and the calculation model of phase fraction is established.(5) To obtain the flow rate of each phase for gas-liquid mixture, a mixed model based on virtual high coefficient correction is established.
