Salgado César M, Dam Roos S F, Puertas Eddie J A, Salgado William L
Instituto de Engenharia Nuclear, Divisão de Radiofármacos (IEN / DIRAD), Rua Hélio de Almeida, 75, Zip Code, 21941-906, Cidade Universitária, RJ, Brazil.
Universidade Federal do Rio de Janeiro, Programa de Engenharia Nuclear (UFRJ / PEN), Avenida Horácio Macedo, 2030, Bloco G - Sala 206, Zip Code, 21941-914, Cidade Universitária, RJ, Brazil.
Appl Radiat Isot. 2022 Jul;185:110215. doi: 10.1016/j.apradiso.2022.110215. Epub 2022 Apr 9.
During the production of oil and gas, barium sulfate (BaSO) scale occurs on the inner walls of the tubes leading the reduction of the internal diameter, making the fluid passage difficult and complicating the calculation of volume fractions of fluids. In this sense, this study presents a methodology for the development of volume fractions of fluids multiphase meters and the prediction of barium sulfate (BaSO) scale thickness. The spectra obtained by two NaI(Tl) detectors that record the transmitted and scattered beams are used as input data for the artificial neural network without the need of any parametrization method. Theoretical models for annular flow regime were developed using MCNP6 code. Different volume fractions and scale thickness values of oil-water-gas were generated as a data set to train and evaluate the neural network. The results indicate that it is possible to calculate the volume fraction regardless the scale thickness in offshore oil industry pipes. More than 88% of the results showed errors below 5% for all investigated samples.
在石油和天然气生产过程中，硫酸钡（BaSO）垢出现在管道内壁，导致内径减小，使流体通道变窄，并使流体体积分数的计算变得复杂。从这个意义上讲，本研究提出了一种用于开发多相流量计流体体积分数以及预测硫酸钡（BaSO）垢厚度的方法。由两个记录透射光束和散射光束的碘化钠（铊）探测器获得的光谱被用作人工神经网络的输入数据，无需任何参数化方法。使用MCNP6代码开发了环状流型的理论模型。生成了不同油水气体积分数和垢厚度值的数据集来训练和评估神经网络。结果表明，在海上石油工业管道中，无论垢厚度如何，都可以计算体积分数。对于所有研究样本，超过88%的结果误差低于5%。
