Design Of Dynamic Experiment Platform And Model Identification Based On Filamentary Shape Memory Alloy

In the design of actuators using Shape Memory Alloy(SMA),the theoretical model is often difficult to be used in practical applications due to many influencing factors and nonlinearity of the shape memory alloy.In order to explore the various properties and parameters of shape memory alloy wires,a dynamic experiment platform needs to be set up,which should have characteristics such as multiparameter acquisition,dynamic load setting,ambient temperature control and online modification of experimental parameters.In this thesis,a dynamic experimental platform of shape memory alloy wire was designed under MATLAB and Simulink environment.First of all,the hardware and software of the platform was configured,then the corresponding hardware code generation module was configured in Simulink.A series of research on the properties of shape memory alloy was carried out by using the experimental platform.According to the input and output results of the system,the relationship of Ni-Ti shape memory alloy wire among current,stress and strain was determined and the influence of heat dissipation obtained.Finally,the model was set up by the method of system identification and compared with the actual system to verify its correctness.The main contents of the thesis are as follows:1.The design of experimental platform.Firstly,the whole electrical system was designed and assembled according to the experimental requirements.The FAULHABER linear motor was chosen as the stress and strain sensor.Secondly,the DAC expansion board based on BeagleBone Black was manufactured according to the problems in the debugging process,which laid a foundation for following work.2.The design of software platform.First of all,the idea of the whole software system and the use of the experimental platform were introduced,and the drive programs such as sensors were developed.Secondly,the S-Function Builder was used to make the drive program a code generation module under Simulink.The method combined the hardware platform with Simulink,and the algorithm was downloaded into the controller through Rapid Control Prototype,which facilitates the real-time observation of the effect of the algorithm and modifying the control parameters online.3.Characteristic experiment and data acquisition.The experiment was focused on studying the hysteresis property of shape memory alloys and the data of Ni-Ti shape memory alloy wire,containing stress,strain and current,were collected.The relationship among the three variables were obtained and a brief analysis was made.Secondly,the different performance of shape memory alloys under air and transformer oil conditions was compared.4.Modeling and verification.Firstly,sinusoidal signals with different frequencies were used as input of the system.The position parameters under constant load force were obtained.Using MATLAB,the linear transfer function and the nonlinear model of the shape memory alloy were identified.Finally,a simple control method was used to verify the system identification model by experiment and simulation,which lay a foundation for the rapid and accurate design of shape memory alloy actuator.