| Twisted and coiled polymer actuators(TCPAs)will contract when heated and extend during cooling down process.The heat-induced contraction makes TCPAs promising smart actuators.TCPAs show great advantages of low hysteresis,low cost,large load-bearing deformation,and long life cycles.In this research,a multi-scale model for predicting the displacement of the TCPA under different loads and temperatures is proposed.The precursor fiber of TCPA is considered as a bundle of monofilaments.Based on helical shape of monofilaments,the constitute equation of the twisted polymer actuator(TPA)was derived.The equivalent modulus of TPA was subsequently obtained.The restoring torque of TCPA was derived by considering the TPA’s thermal deformation and external load,which could explain TCPA’s large stroke under external load and temperature change.Finally,TCPA’s strain energy was calculated,and its thermal actuation was expressed according to Castigliano’s second theorem.The proposed model was validated and then used in parametric analysis of TCPA.Numerical results show that the amplitude of actuation increase linearly with pitch angle,and nonlinearly with spring index.The model can predict actuator’s thermal displacement at different temperatures based on its geometric characteristics and load conditions.However,it’s too much complicated for dynamic control of smart structures with TCPAs.For better force tracing and vibration control,a neural network model was proposed by training on experimental data.This neural-network model can describe the relationship of voltage and driving force more accurately.An inverse model was proposed to generate voltage sequence according to required control force.This model was experimentally validated,better force tracing performance was observed.These neural-network models were consequently used in actuation and vibration control of a cantilever beam.Experimental results shown that,the vibration amplitudes of the beam was reduced by 60% with using one TCPA. |
PDF Full Text Request