Modeling and Simulation Study for Solid Particles Detection in Gas-Oil Pipelines Using Field Free Line in Magnetic Particle Imaging and Its Hardware Realization

• Published in: IEEE transactions on magnetics 2017-07, Vol.53 (7), p.1-8
• Main Authors: Kalra, Mohit, Meribout, Mahmoud
• Format: Article
• Language: English
• Subjects: Atoms & subatomic particles; Coils; Computer simulation; Cross-sections; Direct current; Finite element method; Finite-element method (FEM) simulation; Gas pipelines; Hardware; Image reconstruction; Imaging; Iron oxides; Kaczmarz method; Magnetic domains; Magnetic hysteresis; magnetic particle imaging (MPI); Magnetic resonance imaging; Magnetism; Magnetization; Modelling; Nanoparticles; Natural gas; Nonlinearity; Optimization techniques; Particle swarm optimization; particle swarm optimization (PSO) technique; Permanent magnets; Petroleum pipelines; Pipelines; Powders; real-time black powder detection; Saturation magnetization; Scanning; Sizing; superparamagnetic iron oxide nanoparticles (SPIONs)
• ISSN: 0018-9464; 1941-0069
• DOI: 10.1109/TMAG.2016.2627526

This paper suggests a new magnetic particle imaging (MPI) technique for 2-D tomographic imaging to detect the presence of flowing solid particles (i.e., black powder, which consist of paramagnetic nanoparticles) passing through gas-oil pipelines. The hardware consists of a ring of a modified Halbach array of 24 permanent magnets, in addition to two surrounding Helmholtz coil pairs, and a receiving coil, which are evenly distributed across a cross section of the probe. With the application of static and dynamically moving drive fields, the MPI utilizes the full benefit of superparamagnetic iron oxide nanoparticles (SPIONs) by providing a linear response when they are exposed to a relatively low magnetic field and no response when they are saturated. The scanning of region of interest with the field-free line (FFL) instead of the field-free point (FFP) is motivated by the fact that a high-intensity dc current is required for FFP, which is not tolerated in oil-gas pipelines where the maximal current should not exceed few amperes. With the presence of an oscillatory magnetic field, SPIONs react with a nonlinear magnetization response, which is further measured by the receiving coil. A 2-D image reconstruction is then performed using the frequency-based image reconstruction process. The hardware design, specifically the sizing of permanent magnets with regard to their relative position and dimensions, was refined following an optimization technique based on the particle swarm optimization technique. The assessment of the system using finite-element method indicates that the system can reconstruct the 2-D profile of the SPION with a mean absolute error of prediction of less than 4% and 12% using FFL and FFP methods, respectively.