Viscoelastic features based compensation technique for polymer optical fiber curvature sensors

Polymer optical fiber (POF) curvature sensors present some advantages over conventional technologies for angle assessment, which are compactness, lightweight, multiplexing capabilities and electromagnetic fields immunity. However, POF curvature sensors may present high hysteresis. Since the polymer is a viscoelastic material, which present a hysteretic response with stress or strain, this paper presents a compensation technique for the sensor hysteresis based on the POF viscoelastic features. The sensor response is modelled as a Prony series and the compensation is made by comparing the sensor response with a response without hysteresis obtained on the sensor analytical model. Results show that this technique not only reduces the hysteresis, but also generally reduces the sensor root mean squared error (RMSE) and increases the sensor linearity. The mean hysteresis obtained with the proposed technique is 0.09%, whereas the mean RMSE obtained is 0.80°.