Research On Signal Demodulation Methods Of Wireless Passive Surface Acoustic Wave Torque Sensor

The torque load characteristics of the armored vehicles transmission system is one of the significant indicators reflecting the vehicle dynamic performance,but the corresponding test environment is usually narrow,high-speed rotating,and accompanied with noise,makes the traditional torque sensors difficult to achieve the installation,energy supply,and signal trans-mission,the demand of new type torque sensor is in urgent.Surface Acoustic Wave(SAW)torque sensor can cope with the challenges brought by the torque testing environment because it is able to work in a wireless passive state with small size and good stability.With the decades researches,the structure and sensitive mechanism of SAW torque sensor have become mature,but the SAW sensor echo signal in wireless passive state has the characteristics of high car-rier frequency,short effective length,and low signal-to-noise ratio(SNR),which makes the signal acquisition,processing,and fast analysis still deficient.Currently,the commonly used SAW sensor demodulation methods are direct detection and received signal strength indication(RSSI)detection,which have limitations in detection accuracy and detection speed,respec-tively.In this paper,taking the torque load test of armored vehicle transmission system as the background,two different wireless passive SAW sensor echo signal demodulation methods are studied and designed to cope with different test environments and requirements.The effec-tiveness of the demodulation methods is verified through experiments.The paper includes the following contents.(1)According to the nature of the torque quantity,the properties of SAW devices as wireless passive torque sensor are analyzed,and the single and differential type design of SAW torque sensitive elements is completed by coupling mode model and impedance model analysis.In order to study the demodulation method,the echo signal model of wireless passive SAW sensors under different excitations is established,after analysis,the echo signal is a frequency oscillation signal whose energy gradually decays.And the"over-excitation method"of the torque test system is developed,which lays the foundations for the follow-up researches.(2)To solve the problem of low SNR of SAW sensor echo signal,the signal processing idea of compressive sensing(CS)is introduced into the demodulation system of wireless pas-sive SAW sensor,which improves the system anti-noise ability.The sparsity of SAW sensor echo signal is analyzed,the CS framework based on random demodulation-analog informa-tion converter is formulated according to the signal features,and a complete CS demodulation system is designed by simulation.Combining the features of the observed signal,a secondary information prediction method based on the system transfer function and the low-speed ob-served signal is proposed.Based on the secondary information,an observed signal enhance-ment method is established,which effectively increases the recoverability of the observed signal and can further optimize the sampling rate of the system within the error allowance.The ef-fectiveness of the wireless passive SAW sensor demodulation system based on CS is verified by simulation experiments,the results show that the system can demodulate normally when the echo signal SNR is 5d B and the effective length is longer than 4.9?s,and other methods are compared to demonstrate the advantages of the CS method.(3)To address the RSSI detection method problem of poor dynamic characteristics,the SAW sensor excitation module and RSSI detector sweeping process output model are es-tablished through theoretical analysis and experimental generalization respectively,which can depict the characteristics of SAW sensor RSSI signal in the testing process and accurately predict the optimal sweeping frequency range.Based on the derived signal characteristics,a sliding window based SAW resonance peak search algorithm is proposed.The algorithm can pre-program the signal characteristics,with the window feature determination algorithm,it can quickly retrieve the global space and accurately estimate the coordinates of the resonance point according to the window range,which greatly improves the retrieval efficiency and applicabil-ity of the RSSI detection method.Finally,a simulation experimental study of wireless passive SAW sensor demodulation system based on RSSI detection is carried out,the system can work normally in the-3d Bm noise environment,and the detection efficiency is improved by more than 90%.the effectiveness and advantages of the improved RSSI detection method are verified through result analysis and comparative analysis.(4)The installation method and procedure of SAW torque sensor is studied.A novel electrical connection method for SAW sensors is proposed,and a matching installation and positioning tooling is designed,which can effectively improve the efficiency of SAW sensor welding,installation,and field commissioning.A static torque test platform using the armored vehicle universal axle as the test object is built to verify the performance of two SAW sensor signal demodulation systems.The static characteristics parameters of the two demodulation systems are obtained through static experiments and data analysis,the linearity,repeatability,hysteresis and fitting R²are 3.89%,5.00%,0.92%,0.996 and 0.81%,0.78%,0.62%,0.999,the effectiveness of the systems are verified.The performance advantages and disadvantages between the CS method and the improved RSSI detection method are compared,and the appli-cable scenarios of the two methods are analyzed.(5)The rotational interference and installation error interference problems of the SAW torque sensor are studied,the influence of the gyroscopic effect generated by rotation on the SAW torque sensor is analyzed,the theoretical model between the bias frequency value of the SAW torque sensor,the torque,and the rotation is established,and a least square support vector machine rotation error compensation model based on the optimization of chaotic distribution estimation is proposed,which effectively improves the SAW torque sensor in the rotation sit-uation,reduce the torque solution error caused by rotation interference to 10^-4.A clamp-on mechanical structure is proposed,which can reduce the generation and influence of installation errors.A mechanical model of the clamp-on mechanical structure is established,and the static performance of the clamp-on torque sensor is verified by experiments.It is confirmed that the structure has the advantages of easy arrangement of the SAW torque sensor,easy installation and disassembly,no damage to the shaft,and insensitive to installation errors.