Research On Six-Degree-of-Freedom Position Detection Method Based On Magnetic Tracer Particles

Multiphase flow is widely used in the modern energy and chemical industries,biomedicine,and other fields.However,the multiphase flow has time-varying,transient,and nonlinear characteristics,making the online detection of motion parameters and states in turbulent mixing,particle flow,and precipitation environments a challenging problem to solve.In this paper,we use magnetic tracer particles to simulate particles in multiphase flow,develop a high-precision magnetic positioning system,and use positioning inversion algorithms to achieve six-degree-of-freedom positional detection of magnetic tracer particles,thereby laying the groundwork for future engineering applications.The following is the primary research conducted for the thesis.Firstly,The magnetic dipole model is used to explore the magnetic field distribution of the studied magnetic tracer particles,and a mathematical model of magnetic field distribution is formulated in a straightforward and basic mathematical manner.Secondly,a variety of magnetic localization inversion algorithms,including linear algorithms and nonlinear algorithms,are studied and analyzed.The results indicate that the linear algorithm has the characteristics of fast localization speed and is easily perturbed by the magnetic field of the surrounding environment,while the LM algorithm has a weak global search capability and the genetic algorithm(GA)has a strong global search capability.On this premise,a hybrid Lin-GA search algorithm,which combines the benefits of linear algorithm and genetic algorithm,is presented to improve the global search precision of the inversion method.Subsequently,a magnetic positioning system based on a three-axis tunnel magnetoresistive(TMR)sensor array is designed and constructed to detect and preprocess the magnetic field strength of magnetic tracer particles in a certain spatial range,process and upload the signals to the host computer using an FPGA-based data processing system,invert the magnetic tracer particle position and angle information in six degrees of freedom using an optimization algorithm,and then output the magnetic tracer particle position and angle information in six degrees of freedom.Finally,an experimental evaluation of the performance of the inversion method in terms of localization error,orientation error,and execution time has been carried out.The results demonstrate that the Lin-GA-based hybrid search algorithm has greater localization and orientation accuracy as well as a shorter execution time,with the localization error of 0.519±0.116 mm,the orientation error of 0.014±0.006°,and an average execution time of 88.906±17.001 s,proving the method’s efficacy and superiority.