Optimization Design And Model Study Of Magnetically Controlled Shape Memory Alloy Sensor

Magnetically controlled shape memory alloy(MSMA)is a kind of new functional material.The advantage of MSMA compared to other smart materials is their ability to generate up to 10% strains,rapid response speed and high efficiency.It has broad prospects in the application area of the sensor.The MSMA mechanism is based on the martensite twin boundary motion driven by the external magnetic field when the material is in complete martensite state.After that,when the material is subjected to mechanical straining the twin variants reorient,which alters their magnetization and the surrounding magnetic field.This phenomenon which called MSMA inverse effect can be used for sensor applications.In this paper,MSMA sensor model based on the MSMA inverse effect was established.The quantum-chaos particle swarm optimization was applied to the multi-parameter identification of sensor model.The structure of sensor was optimized.Firstly,the magnetic circuit model of MSMA sensor which based on MSMA deformation mechanism was established and the formula for calculating the induced voltage was proposed.The output induced voltage depends on the press,complex magnetic field,frequency and amplitude of dynamic force.The MSMA sensor model provided the objective function for parameter identification algorithmSecondly,the global structure of MSMA sensor was optimized on the basis of the working conditions.By electromagnetic simulation,the structure of iron core,the permanent magnet size and number of turn was optimized successively.The MSMA sensor experimental platform was constructed.Finally,the quantum-chaos particle swarm optimization was applied to the multi-parameter identification of sensor model for induced voltage and it satisfied the applicability of the model under different experimental conditions.In order to obtain accurate experiment waveform,the experimental data was processed by the method of wavelet denoising.The comparison between simulation and experimental results verify the effectiveness of the identification algorithm.