Research On Algorithm Of Free Magnetic Field Point In Ferromagnetic Resonance Temperature Measurement

The emerging ferromagnetic resonance temperature measurement technology has many advantages in theory,so it has been widely concerned by scholars at home and abroad.The free field point(FFP)needs to be solved in this technique.It traditionally uses particle swarm optimization.However,with the recent years,the optimization algorithm has developed rapidly,especially the meta-heuristic algorithm.The calculation accuracy and convergence speed of the algorithm greatly affect the accuracy and reaction speed of the temperature measurement.Therefore,applying the current algorithm research results to the FFP,or trying physics theory,helps to promote the development of this technology.This paper is aimed at solving the problem of FFP.This paper discusses three aspects of improving the existing meta heuristic algorithm,proposing a new meta heuristic algorithm and theoretical derivation.On the improved meta heuristic algorithm,based on the bat optimization algorithm with time-varying inertia weight learning mechanism(wBA),the reverse learning and Levy perturbation mechanism are introduced to improve it.An improved optimized bat algorithm(wLevyOLBA)based on reverse learning and Levy perturbation is proposed.In the proposed new heuristic algorithm,the Island Algorithm is proposed.In terms of theoretical derivation,based on Biot-Savar's law,an analytical method for calculating the current of each coil when any point on the XY plane is FFP is obtained.The wLevyOLBA algorithm uses six standard functions for verification and compares it with the BA algorithm and the wBA algorithm.The island algorithm uses the CEC13 test set for verification and comparison with the PSO algorithm.The results show that the wLevyOLBA algorithm is superior to the wBA algorithm and the BA algorithm in terms of convergence performance and robustness.The island algorithm is worse than the PSO algorithm when computing functions with certain characteristics,but its computational accuracy and robustness are significantly better than those of the PSO algorithm in most functions and multiple dimensions.This verifies the analysis of the algorithm in this paper.The wLevyOLBA algorithm,Island algorithm and analysis method are respectively used to solve FFP,and the results are brought into the simulation software for simulation experiments.The results show that the three methods are superior to the traditional PSO algorithm.The position error of FFP calculated by analytical method is within 3 mm,which meets the needs of magnetic signal measurement.Compared with the meta-heuristic method,the analytical method not only overcomesthe shortcomings of the traditional iteration method such as slow convergence and uncertain results,but also has theoretical basis for its solution.It has absolute advantages in accuracy and calculation speed.It is helpful for the development of temperature measurement,heating control,concentration imaging and other technologies of magnetic nanoparticles.