Driving Force Tracing of Twisted and Coiled Actuators Based on the Neural Network Model

Due to the hysteresis properties, in this paper, the neural network model of twisted and coiled actuators describing the relationship between voltage and driving force are established. Experimental data were recorded and then used as the training data of the neural network model. The plant model and the inverse model of the neutral network were established later, after which the thermal driving force tracing of the actuators was investigated through the neural network. For driving force at 0.05Hz, 0.0625Hz and 0.1 Hz, the tracing errors are 5.0%, 6.0% and 4.0%, respectively. Thus, the neural network model has a good prediction accuracy, which lays a foundation for the further application of TCPA materials.