Study On Modeling And Control Of Electro-Active Smart Material IPMC

Ion-exchange Polymer-Metal Composites (IPMC), known as artificial muscle, is a type of Electro-Active smart material. With the similar performance as biologic muscle, IPMC is very interesting polymer because of its capability to transform electrical energy into mechanical energy and viceversa, making it particularly attractive for possible applications in different fields, such as Micro-Electro-Mechanical System(MEMS), biological medical, bionic mechanism and so on. However, its response has complex mechanism, and takes on nonlinear characteristic, and also is impressible to surroundings condition and material parameters, which blocks its development and utilization.Firstly a complete set of experimental platform is developed for IPMC. Then, the response characteristics are studied experimentally with respect to different types, frequencies, amplitude of stimulation signals and different size of material, environmental humidity, so as to understand its response characterization and establish a detailed IPMC database, and also its response mechanisms are analyzed which lays a good foundation in modeling and controlling for IPMC.Based on the prior information, the model order is determined by correlation analysis and curve fiting of frequency response data, and a most suitable ARX(AutoRegressive exogenous) model structure is chosen with third order for IPMC through off-line identification in the set of candidate models; Based on the model structure, the least squares recursive identification method is utilized to estimate the unknown parameters on line based on Lab VIEW so as to avoid the impact of surroundings change on validity of model result.Aiming at IPMC’s hysteresis, two kinds of preisach hysteresis models are established. The one is a classic preisach parametric model which is achieved by discretization and online identification, and further more the the model error is analyzed to improve it. The other is a generalized Pl(Prandtl-lshlinskii) model which is derived based on LMS(Least Mean Square) in terms of first-order and second-order reversal datasets by reversal segmentation of preisach plane. The reason is also explained why the creep phenomenon happened in IPMC, the recursive weighted least squares algorithm is selected to identify the parameter of creep model. Finally, the hysteresis PI model and liner creep model are combined to build the mix model of hysteresis and creep of IPMC, and testing results show that the identification curve generated by the mix model is quite similar to the output curve of IPMC.Then the fuzzy PID and inverse compensation controller are designed, and simulation results validate that PID controller can control linear characteristic of IPMC effectively while inverse compensation controller can eliminate hysteresis of IPMC to guarantee IPMC’s tracking to desired trajectory. Two adaptive inverse control structures are built to compensate the hysteresis and creep property of IPMC. Finally, the control structure is simulated and good simulation results are got. The piezoelectric smart materials cantilever beam based PVDF sensor and IPMC actuator is also investigated in this paper. The piezoelectric sensor, actuator equation and state equation of the smart cantilever beam are established. The active vibration optional control method based on LQR is designed, and the control system is simulated. The simulation results show that the control method is effective and applicable.At last, a new model for IPMC actuator based on non-integer order models is proposed. An experimental setup has been realized to study the IPMCs non-integer behavior and an algorithm based on Marquardt of least squares recursive identification has been developed in order to identify a fractional order model of IPMC actuator. And the fractional order PID controllers are designed compared with the integral order PID controllers. Simulation results demonstrate that the behavior of fractional order PID controller is much better than the integral order PID controllers, and fractional order PID controllers can achieve better control performance.