Abstract
Single-layer water cut measurement is of great significance for identifying and shutting off the unwanted water, analyzing oil remained and optimizing production. Currently, however, only the water cut of multilayer mixture can be measured by testing samples taken from wellhead, a way which is widely used in oilfields. That of single-layer fluid cannot be determined yet
To address the problem, this paper puts forward a new impedance sensor that offers long-term online monitoring of single-layer water cut. This sensor is based on the different electrical conductivity of oil and water. It has two layers. The inner one contains three electrodes - two at both sides sending sinusoidal excitation signals and one at the middle receiving signals that have been attenuated by the water-oil medium. With the Maxwell's model of oil-water mixed fluid, the receiver then can measure the water cut online. The outer layer of the sensor is made of PEEK, an insulative protection. In front of the electrodes lies a static mixer which makes the measurement more accurate by fully blending the two media when they flow through the electrodes.
Laboratory tests are carried out with the prototype of the sensor at various oil-water mixing ratios, fluid flow rates, and temperatures. Results show that the average margin of error is within ± 3%. Higher accuracy is seen when high water cut and flow rate enable oil globules to disperse more evenly and the space in between to get wider and the RMS error is less than 2%. If the water cut drops below 80%, the aggregation of the droplets will cause wild fluctuation and more errors in the measurement. In addition, the mineralization of the mixture directly changes its conductivity, which largely impacts the result. Meanwhile, temperature can influence the ionic movement intensity and then alter the conductivity of the medium. Therefore, in practice, the sensor calibration needs to be performed according to the range of medium salinity, and the temperature of the medium is collected in real time for temperature compensation. It is shown that after the adjustment, the water cut measurement results have higher accuracy and consistency.
The impedance sensor can realize online water cut monitoring for a single-layer, indicated by tests. It is more suitable for the increasing high water cut oilfields in that it is more accurate as the water cut grows.
Neural Network for Metal Detection Based on Magnetic Impedance Sensor
