A MOBA-Root-MUSIC-Based Demodulation Method for Surface Acoustic Wave Torque Sensor

To improve the measurement accuracy of surface acoustic wave (SAW) torque sensor, an ultrahigh precision SAW torque sensor demodulation method based on multi-objective bat algorithm (MOBA) and root-MUSIC algorithm was proposed. The root-MUSIC algorithm was adopted to construct the polynomial equation of SAW echo signals. The construction of the objective function was based on two principles: the least distance to unit circle and least phase angle to the mean roots. MOBA was used to optimize the objective function to improve the accuracy of the roots. The numerical simulation and actual interrogating experiments were carried out and comparisons with other methods were conducted. The demodulation results verified that the proposed method had a minimum estimation error with standard deviation of 1 Hz. The designed demodulation method was used to the SAW torque measurement system. The torque sensing SAW resonator (SAWR) was designed using an FEM software and manufactured with a frequency of 433.945 MHz and an unload
${Q}$
factor of 13 370. The SAWR was glued on a bearing to measure applied torque. The torque loading and unloading experiments demonstrated that the nonlinearity of the proposed method was 0.1402%, which is the minimum when compared to other methods. The hysteresis was 1.1791%, and the sensitivity was 1.3570 kHz/Nm. Numerical simulation and experimental results demonstrated the effectiveness and accuracy of the designed algorithm in SAW torque sensor demodulation system.