Analysis and experimental investigation of a modular differential dielectric sensor (DDS) for use in multiphase separation, process measurement and control

In subsea processing systems, oil industry increasingly demands accurate and stable continuous measurement of the percent water in crude oil production streams (watercut) over the entire 0 to 100% range. High accuracy and stability are also required for surface measurement to support niche applications such as control of processes which remove trace amounts of oil and particulates from produced water prior to disposal.; Differential Dielectric Sensors (DDS) have been developed by Chevron as independent tools connected with multiphase meters for process management and composition measurement. DDS is unique in its use of very low noise and high sensitivity differential measurements between two identical sensors and the use of physics based models for multiphase flow characterization. In a process control system, DDS shows good measurement stability and is adaptive to composition measurement compensating for changes in oil composition, gas fraction, emulsion state, water NaCl concentration, temperature and flow rate. Because of its auto calibration capability, DDS can also conduct real time calibration for sensor configuration changes caused by factors such as corrosion and erosion. Existing watercut tools predominantly depend on empirical data and correlations that are sensitive to fluid properties and therefore are limited in their general applicability. The scope of this study is to expand the capability of the current DDS for real time characterization of dielectric property of flowing oil/water mixture and make it more accurate and predictable.; The main objective of this dissertation is to develop appropriate mathematical models for the DDS which characterize the microwave attenuation and phase shift as a function of fluid properties, sensor geometry and operational conditions. Forward models based on the analysis of microwave propagation have been developed for sensors configured as rectangular and circular waveguides. Finite Element Analysis (FEA) and experimental investigations have been conducted to validate the sensor analytical models. A dedicated closed-loop experimental facility is used to obtain in-line real-time measurement of DD sensor data in a controlled configuration. Results of this project will be useful for optimization and refinement of multiphase meters.