Determination of eccentric deposition thickness on offshore horizontal pipes by gamma-ray densitometry and artificial intelligence technique.

The extraction of oil is accompanied by water and sediments that, mixed with the oil, cause the formation of scale depositions in the pipelines walls promoting the reduction of the inner diameter of the pipes, making it difficult for the fluids to pass through interest. In this sense, there is a need to control the formation of these depositions to evaluate preventive and corrective measures regarding the waste management of these materials, as well as the optimization of oil extraction and transport processes. Noninvasive techniques such as gamma transmission and scattering can support the determination of the thickness of these deposits in pipes. This paper presents a novel methodology for prediction of scale with eccentric deposition in pipes used in the offshore oil industry and its approach is based on the principles of gamma densitometry and deep artificial neural networks (DNNs). To determine deposition thicknesses, a detection system has been developed that utilizes a 1 mm narrow beam geometry of collimation aperture comprising a source of 137Cs and three properly positioned 2″×2″ NaI(Tl) detectors around the system, pipe-scale-fluid. Crude oil was considered in the study, as well as eccentric deposits formed by barium sulfate, BaSO4. The theoretical models adopted a static flow regime and were developed using the MCNPX mathematical code and, secondly, used for the training and testing of the developed DNN model, a 7-layers deep rectifier neural network (DRNN). In addition, the hyperparameters of the DRNN were defined using a Baysian optimization method and its performance was validated via 10 experiments based on the K-Fold cross-validation technique. Following the proposed methodology, the DRNN was able to achieve, for the test sets (untrained samples), an average mean absolute error of 0.01734, mean absolute relative error of 0.29803% and R2 Score of 0.9998813 for the scale thickness prediction and an average accuracy of 100% for the scale position prediction. Therefore, the results show that the 7-layers DRNN presents good generalization capacity and is able to predict scale thickness with great precision, regardless of its position inside the tube.

A new application of radioactive particle tracking using MCNPX code and artificial neural network.

Stirrers and mixers are highly used in chemical, food, pharmaceutical, cosmetic, concrete industries and others. During the fabrication process, the equipment may fail to appropriately stir or mix the solution. Besides that, it is also important to determine when the right homogeneity of the mixture is attained. Thus, it is very important to have a diagnosis tool for these industrial units to assure the quality of the product and maintain market competitiveness. Nuclear techniques, such as gamma densitometry, are widely used in industry to overcome a sort of difficulties, as they are minimally non-invasive techniques. This paper presents a method based on the principles of the radioactive particle tracking technique to predict the instantaneous position of a radioactive particle to monitor a concrete mixture inside an industrial unit by means of Monte Carlo method and artificial neural network. Counts obtained by an array of detectors properly positioned around the mixing canister will be correlated to each other, by means of an appropriate mathematical search location algorithm, in order to predict the instantaneous positions occupied by an inserted radioactive particle. The simulation consists of a detection geometry of eight NaI(Tl) scintillator detectors, a 662 keV 137Cs point source with isotropic emission of gamma-rays and a polyvinyl chloride tank. At first, the tank is air filled and, afterwards, filled with concrete made with Portland cement. The modeling of the detection system is performed using the MCNPX code. For both medium, the correlation coefficient was 0.99 for all coordinates, which indicates that this methodology could be a good tool to evaluate industrial mixers.

Inorganic scale thickness prediction in oil pipelines by gamma-ray attenuation and artificial neural network.

Scale can be defined as chemical compounds that are inorganic, initially insoluble, and precipitate accumulating on the internal walls of pipes, surface equipment, and/or parts of components involved in the production and transport of oil. These compounds, when precipitating, cause problems in the oil industry and consequently result in losses in the optimization of the extraction process. Despite the importance and impact of the precipitation of these compounds in the technological and economic scope, there remains difficulty in determining the methods that enable the identification and quantification of the scale at an initial stage. The use of gamma transmission technique may provide support for a better understanding of the deposition of these compounds, making it a suitable tool for the noninvasive determination of their deposition in oil transport pipelines. The geometry used for the scale detection includes a 280-mm diameter steel tube containing barium sulphate (BaSO4) scale ranging from 0.5 to 6 cm, a gamma radiation source with divergent beam, and a NaI(Tl) 2 × 2″ scintillation detector. The opening size of the collimated beam was also evaluated (2-7 mm) to quantify the associated error in calculating the scale. The study was done with computer simulation, using the MCNP-X code, and the results were validated using analytical equations. Data obtained by the simulation were used to train an artificial neural network (ANN), thereby making the study system more complex and closer to the real one. The input data provided for the training, testing, and validation of the network consisted of pipes with 4 different internal diameters (D1, D2, D3, and D4) and 14 different scale thicknesses (0.5 to 7 cm, with steps of 0.5 cm). The network presented generalization capacity and good convergence, with 70% of cases with less than 10% relative error and a linear correlation coefficient of 0.994, which indicates the possibility of using this study for this purpose.