Research Of Displacement Control Methodology For Magnetically Controlled Shape Memory Alloy Actuators Based On Krasnosel’skii-Pokrovskii Model

Micro-actuator is the core of the micro positioning equipment, looking for actuatorswith good performance is very significant for improving the accuracy of positioning.Magnetically controlled shape memory alloy is a new kind of shape memory material whichhas big deformation, high-frequency response and can be controlled precisely. Sincemagnetically controlled shape memory alloy has better performance than piezoelectricmaterials and magnetostrictive materials, It is expected to become the key materials of nextnew generation micro displacement actuator in the future.Hysteresis nonliearity phenomenon exists in magnetically controlled shape memoryalloy materials, this phenomenon can leads to slow responses and increases controllingdifficulties of the actutor which must be solved difficulty of controllability of the actutorwhich is the problem must be solved in research of high-accuracy control of magneticallycontrolled shape memory alloy actutors. In order to reduce and eliminate the influence ofhysteresis nonliearity phenomenon on the accuracy of actutor controlling, This paper studiedit from two aspects: firstly, the high-accuracy mathematical model was established forhysteresis phenomenon to describe the hysteresis characteristic of actutor; secondly,high-accuracy controller was designed to reduce and eliminate the influence of hysteresisphenomenon on the accuracy of actutor controlling.In this paper, a modified KP model was used to describe the complex hysteresisphenomenon of magnetically controlled shape memory alloy actutors. Considering theproblem of parameters identification in the model, weighted recursive least square methodand adaptive recursive algorithm were adopted for offline identification and onlineidentification respectively. The error of hysteresis model after identification were0.22%and1.18%respectively, which proved the efficiency and feasibility of the two proposedmodeling methods.In order to reduce and eliminate the influence of hysteresis nonlinearity on the accuracyof actutor controlling, this paper proposed two schemes. a PID hybrid control scheme based on inverse hysteresis model compensation was the first scheme. In this scheme, using amodifed differential evolutionary algorithm to set the parameters of PID controller, Throughthe above operation，the error of magnetically controlled shape memory alloy actutor wasreduced to0.49%. This scheme can meet the requirements of displacement controlling atnormal temperature but its control effect will become poor when the temperature increases tosome extent. In order to solve this problem, The second scheme was proposed, a modelreference adaptive control scheme based on KP model and inverse model compensationsignals with online adjusting was designed, and Simulation results showed that the error ofthe controlling system had been reduced to0.72%. The two controllers can both make theinput and output to linear relationship of one to one correspondence, eliminating thehysteresis phenomenon of the impact on the control accuracy.