An intelligent gamma-ray technique for determining wax thickness in pipelines

Measuring the wax deposition inside pipelines is one of the critical parameters in the oil, gas and petrochemical industries to control the flow through the pipelines. This paper presents a novel method using artificial neural networks to measure the thickness of the wax. This method was based on counting the backscattered gamma-ray from different thicknesses of the wax inside the pipes with different diameters. For this purpose, the system was simulated by MCNPX code and the designed setup was optimized. The main analyses were based on the simulation results but the verification was performed using a real experimental setup. The results showed a good agreement between the simulation results and the experimental data with a root mean square error less than 1%. Response of the detector was simulated for a standard industrial nominal pipe ranged from 2 to 4.5 inches and for radiation sources Cs-137 and Co-60. Using these data, a multilayer perceptron for different energy sources was trained. The best prediction of the wax thickness was obtained for the case of using two radiation sources, simultaneously. The output of the trained neural network showed that the proposed method is capable of measuring the wax thickness inside the pipe with a good accuracy.