Frequency based hysteresis compensation for piezoelectric tube scanner using artificial neural networks

• Main Authors: Othman, Yahya Sheriff, Mahmood, Iskandar Al-Thani, Aibinu, Abiodun Musa, Alang Md Rashid, Nahrul Khair
• Format: Article
• Language: English
• Published: Elsevier 2012
• Subjects: Q Science (General); TJ212 Control engineering
• Online Access: http://irep.iium.edu.my/25836/1/Frequency_Based_Hysteresis_Compensation_for_Piezoelectric_Tube_Scanner_using_Artificial_Neural_Networks.pdf; http://irep.iium.edu.my/25836/; http://www.sciencedirect.com/science/article/pii/S1877705812026409

Piezoelectric positioners are widely used as a nano technology tools. A common application of piezoelectric positioners is the piezoelectric tube scanner. This is because they guarantee a precise nanoscale positioning. However, there are some losses of positioning precession due to vibration in fast scanning, and some nonlinear drawbacks represented by creep and hysteresis. Hysteresis problem appears at wide scanning range. For such a nonlinear problem, it is impossible to get an exact model for control system design. Therefore, in this paper a neural network scheme has been used to generate a proper control signal based on open-loop input/output (I/O) data. Control signal is then calculated in frequency domain based on I/O spectrum. Neural network is trained offline using set of reference signal harmonics to produce the required control signal harmonics. An inverse Fourier transform is performed to obtain the time domain control signal. Experimental results show that the developed control scheme improves the performance of the system by minimizing the effect of hysteresis.