The Identification Of Phase Combination Based On Multisensor Data Fusion Of Oil-gas-water Three-phase Flow

At present, the major oil fields are in the stage of high water content production. Formation pressure and oil well flow pressure is very low. Oil well is generally in a state of degassing and mixed flow state of oil-gas-water three-phase. Multiphase flow is a complex flow characteristic, and it has the interface effect and relative velocity, which make it difficult to detect the parameters. At present, there is no mature high water content, low yield liquid three-phase flow parameters detection technology. It brings great measurement error due to the current multi-phase flow parameter detection is very popular and highly dependent on the precision of the sensor. In order to solve the above problems, this paper proposes a method based on on-line buffer separation, which does not depend on the absolute accuracy of the sensor and the flow rate of the flow regime. The specific research contents are as follows.The three-phase flow measurement method based on line separation and phase transformation is based on the principle of gravity differentiation, which controls the dynamic separation of the fluid and makes the oil gas and water phase separated by measuring the outlet end. We can get the flow of each phase by determining the phase transition point of the gas-water and oil-water. Sensors are used to make flow pattern judgments. Therefore, the absolute accuracy of the sensor is not high. The method reduces the influence of flow pattern on the measurement accuracy, and improves the accuracy of the measurement by through the tank buffer and gravity differentiation.In order to identify the flow pattern of fluid, a method based on multisensor feature extraction is proposed. This paper adopts wavelet technique for flow signal analysis, extraction of mean and sensitive signal variance, skewness coefficient and derivative composition phase recognition feature vector. On the basis of feature extraction, a Bayesian algorithm based on minimum error rate is established. The establishment of the model can realize the automatic identification of the oil-gas-water three-phase flow regime, and lay a foundation for the measurement of the phase velocity.To verify the correctness of the three phase flow measurement method on the phase state transition, the multiphase flow experiment was carried out on the measurement system. The experimental results show that the measurement method is suitable for the measurement of oil gas and water three-phase flow, the identification model can realize the identification of oil gas and water combined fluid, and achieve a higher accuracy of phase flow measurement.