Aero-engine torque and temperature parallel monitoring system and method based on surface acoustic wave resonator
Patent Information
- Application Number
- CN202611023544.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-10
- Publication Date
- 2026-10-16
AI Technical Summary
[0013]本发明针对目前航空发动机扭矩与温度监测时存在的能量供给与信号传输困难以及现有基于声表面波传感技术的监测方案存在的问题,提出一种基于声表面波谐振器的航空发动机扭矩与温度并行监测系统及方法,该监测系统不仅可以解决旋转部件上传感器供电和信号传输的瓶颈,而且能通过差动布置与双参量解耦以频分多址方式实现对扭矩与温度的并行监测,并通过线性调频信号激励和频谱细化算法大幅提升测量可靠性与精度
[0036] (1) A segmented loop rotor antenna with dual-feed ports is adopted, and a direct-plug mechanical and electrical connection with the sensing substrate is achieved through an RF coaxial connector, eliminating the traditional cable connection method or the radiator integrated sensing substrate method. This structure avoids the problem of loosening or falling off the connection due to centrifugal force and vibration under high speed conditions, optimizes the dynamic mass distribution of the rotor part, meets the strict requirements for dynamic balance, and reduces the influence of the metal shaft on the antenna radiation performance.
Smart Images

Figure CN122921630A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a parallel monitoring system and method for torque and temperature of aero-engines based on surface acoustic wave resonators, which belongs to the field of wireless passive sensing. Background Technology
[0002] Torque and temperature are important technical indicators for evaluating the performance and health of aero-engines, and torque is also a core parameter for analyzing the output power of the transmission system. By monitoring torque and temperature, effective guarantees can be provided for reliable engine operation, fault diagnosis, and safe maintenance.
[0003] However, monitoring the torque and temperature of aero engines faces complex operating conditions and challenges. These can be summarized as follows:
[0004] (1) Energy supply and signal transmission: Sensors on rotating parts are difficult to power and transmit data via wired means, and active sensors have power consumption and lifespan issues as well as flammability and explosion hazards under extreme environmental conditions.
[0005] (2) Extreme physical environment: The sensor needs to be installed on a high-speed rotating drive shaft, which results in high speed and high centrifugal force.
[0006] (3) Long-term stability of devices: The aerospace field has stringent requirements for the long-term stability of sensors, and any failure may lead to catastrophic consequences.
[0007] Surface Acoustic Wave (SAW) sensing technology offers a revolutionary solution to this problem. SAWR (SAW Resonator) devices are passive and wireless, operating only through electromagnetic excitation, thus overcoming the bottlenecks of power supply and signal transmission in rotating components. Made of piezoelectric materials such as quartz and lithium tantalate, they possess outstanding advantages such as high temperature resistance, small size, and long-term stability. By fixing the SAWR device to a rotating shaft in a specific manner, its resonant frequency changes with the strain and temperature rise caused by the shaft's torsional deformation. By monitoring the frequency changes with an external reader, torque and temperature values can be calculated with high precision.
[0008] The existing solutions for monitoring torque and temperature in aero-engines based on surface acoustic wave (SAW) sensing technology, and their problems, can be summarized as follows:
[0009] (1) Sensor antennas are usually selected as rod antennas or flat antennas and connected by coaxial cables. Under high speed conditions, the cables and their connectors will be subjected to continuous centrifugal force and vibration impact, which can easily lead to loosening or even falling off, seriously affecting the long-term reliability and safety of the system. In addition, there are also solutions that fold the sensor antenna radiator and directly print it on the sensor substrate. However, since the shaft is made of metal and is located in the strong induction field of the integrated printed antenna, the radiation characteristics of the antenna will change.
[0010] (2) For the excitation of surface acoustic wave resonant sensors, the transmitted signal is usually a high-frequency sinusoidal signal with a single frequency. This excitation method has inherent defects. When the resonant frequency of surface acoustic waves is deviated by a large amount due to changes in torque or temperature, the surface acoustic wave resonant sensor will become detuned. Observing the time-domain echo signal, we can see that its intensity is reduced, which leads to a shortening of the wireless monitoring distance of the sensor. When the resonant frequency deviates too much and exceeds a certain threshold, the reader will not be able to measure the echo signal.
[0011] (3) The transmitted signal and the echo signal will be reflected and scattered multiple times on the metal surface. The signals from different paths will superimpose on each other, resulting in drastic fluctuations in the echo amplitude. Specifically, the intensity of the time-domain echo signal fluctuates, and the peak amplitude of the spectrum is unstable, which causes peak detection jitter and greatly reduces the measurement reliability and anti-interference capability of the system.
[0012] (4) To achieve sufficient measurement accuracy, the system needs to acquire a sufficiently long time-domain signal to improve spectral resolution. However, the quality factor of the surface acoustic wave resonator is limited, resulting in a short effective duration of its free oscillation decay. Under the condition of fixed tail length, the frequency resolution that can be obtained by Fourier transform is fundamentally limited, making it difficult to balance measurement accuracy and response speed. Summary of the Invention
[0013] This invention addresses the challenges of energy supply and signal transmission in current aero-engine torque and temperature monitoring, as well as the problems existing in current monitoring schemes based on surface acoustic wave (SAW) sensing technology. It proposes a parallel monitoring system and method for aero-engine torque and temperature based on a SAW resonator. This monitoring system not only solves the bottlenecks in power supply and signal transmission for sensors on rotating components, but also achieves parallel monitoring of torque and temperature through differential arrangement and dual-parameter decoupling in a frequency division multiple access manner. Furthermore, it significantly improves measurement reliability and accuracy through linear frequency modulation signal excitation and spectral refinement algorithms.
[0014] This invention adopts the following technical solution: a parallel monitoring system for torque and temperature of an aero-engine based on a surface acoustic wave (SAW) resonator, comprising a SAW sensor, a sensing substrate, a sensor antenna, a reader antenna, a reader, and a rotating shaft. The SAW sensor includes four torque measuring devices and one temperature measuring device. The four torque measuring devices are differentially arranged on both sides of the rotating shaft, and the temperature measuring device is separately arranged on the surface of the sensing substrate. The sensor antenna is a through-hole dual-feed port segmented loop rotor antenna, fixed to the sensing substrate via an RF coaxial connector. The reader antenna is a single-feed port segmented loop antenna. The reader includes a transmit link, a receive link, and transmit / receive isolation. The system includes a switch and a main control module. The transceiver isolating switch is a single-pole double-throw switch used to switch between transmit and receive states. The first port a of the transceiver isolating switch is connected to the reader antenna, the second port b is connected to the transmit link, and the third port c is connected to the receive link. The reader adopts a multi-local oscillator architecture, is excited by a linear frequency modulated signal, and retains the image frequency that should be suppressed in the upmixing stage, which is also transmitted as an excitation signal. The echo signal radiated from the sensor antenna to the reader antenna is collected by downmixing and then processed through a signal processing flow that combines zero-padding fast Fourier transform, linear frequency modulated Z-transform spectrum refinement, and Quinn interpolation estimation to achieve high-resolution frequency calculation with a finite signal length.
[0015] Furthermore, the torque measuring device includes a piezoelectric substrate, an interdigital transducer, and a reflective grating array. Busbars are provided on opposite sides of the interdigital transducer. On a sensing substrate on which two torque measuring devices and one temperature measuring device are mounted, the radio frequency coaxial connector is connected to each torque measuring device and the temperature measuring device via a high-frequency impedance line. The piezoelectric substrate of the torque measuring device is an AT-cut quartz substrate with Euler angles of (0°, 126°, 0°), and the piezoelectric substrate of the temperature measuring device is a YX-cut quartz substrate with Euler angles of (0°, 90°, 0°).
[0016] Furthermore, the four torque measuring devices are a first torque measuring device, a second torque measuring device, a third torque measuring device, and a fourth torque measuring device. The initial resonant frequency of the first torque measuring device is 431MHz, the initial resonant frequency of the second torque measuring device is 433MHz, the initial resonant frequency of the third torque measuring device is 435MHz, the initial resonant frequency of the fourth torque measuring device is 437MHz, and the initial resonant frequency of the temperature measuring device is 439MHz. The four torque measuring devices and one temperature measuring device achieve parallel monitoring of the torque and temperature of the aero-engine in a frequency division multiple access manner.
[0017] Furthermore, the first torque measuring device and the fourth torque measuring device form a differential pair, which are connected in parallel with the temperature measuring device and arranged on one of the sensing substrates and then connected to the first feed port of the sensor antenna. The second torque measuring device and the third torque measuring device form a differential pair and are then connected to the second feed port of the sensor antenna.
[0018] Furthermore, the torque measuring device uses a single-component high-temperature resistant epoxy adhesive, which is directly pasted onto the surface of the rotating shaft through an opening in the sensing substrate. The temperature measuring element is bonded and fixed in the alumina ceramic and Kovar alloy composite package with high-temperature resistant elastic adhesive and then welded to the surface of the sensing substrate.
[0019] Furthermore, the sensing substrate uses an epoxy resin glass cloth laminate with a double-layer structure. High-frequency impedance lines and pads are printed on the upper layer of the sensing substrate. The pads are sprayed with immersion gold with a thickness of 2 microinches and are located on opposite sides of the torque measuring device. One side of the pad is bonded to the high-frequency impedance line by ball bonding, and the other side is connected to ground. Each pad is connected in parallel with the busbars on both sides of the interdigital transducer through gold wires and flows back to the lower layer of the sensing substrate.
[0020] Furthermore, the reader's transmission link generates a linear frequency modulated excitation signal, the starting frequency of which corresponds to the initial resonant frequency of the surface acoustic wave sensor. The excitation signal of the torque measuring device is generated by mixing the radio frequency signal and the image signal on the phase-locked loop PLL1 with DDS1, DDS2, and PLL1. The excitation signal of the temperature measuring device is generated by mixing the image signal on the phase-locked loop PLL1 with DDS3.
[0021] Furthermore, the signal acquisition section in the reader receiving link utilizes an asynchronous FIFO buffer to achieve cross-clock domain matching between the acquisition clock and the serial port transmission clock, and splits the ASCII code sampled value after analog-to-digital conversion into high and low bytes before transmitting it to the host computer in the original byte form for frame reassembly and signal processing.
[0022] Furthermore, the host computer sequentially performs DC removal, Blackman-Harris windowing, and zero-padding preprocessing on the surface acoustic wave echo signal. A rough frequency domain peak is obtained through fast Fourier transform. A 500-point linear frequency modulated Z-transform is performed within ±50 kHz of the peak to refine the spectrum. Then, Quinn interpolation is applied to the CZT peak point to calculate the frequency offset in order to obtain the accurate frequency of the resonator.
[0023] The present invention also adopts the following technical solution: a method for parallel monitoring of torque and temperature of aero-engines based on surface acoustic wave resonators, comprising the following steps:
[0024] Step A: After the system is powered on, the main control module controls the reader's transmit / receive isolation switch to connect with the transmit link and transmits an excitation signal containing 5 carrier waves. The excitation signal is transmitted to the sensor antenna in the form of electromagnetic waves through the combiner and the reader antenna. The carrier waves of the 5 excitation signals are linear frequency modulated signals, and their starting frequencies are the initial resonant frequencies of each surface acoustic wave sensor. The bandwidth is 750kHz and the linear frequency modulation time is 15μs.
[0025] Step B: After the sensor antenna receives the excitation signal, the interdigitated transducer converts the excitation signal component located in the resonant cavity into a surface acoustic wave through the inverse piezoelectric effect. After multiple coherent reflections and superpositions by the reflective grating array, a standing wave is formed in the resonant cavity, completing energy accumulation.
[0026] Step C: When the duration of the excitation signal is greater than the charging time of the resonant cavity of the surface acoustic wave sensor, the transceiver isolation switch of the reader is connected to the receiving link.
[0027] Step D: The standing waves of the five surface acoustic wave sensors are converted into electromagnetic waves by interdigital transducers through the positive piezoelectric effect. The echo signal is radiated from the sensor antenna to the reader antenna. The echo signal contains the resonant frequency information of each of the five sensors.
[0028] Step E: The echo signal is amplified by two stages of low-noise amplifiers, and then filtered out by two stages of RF bandpass filters to remove out-of-band interference;
[0029] Step F: The echo signal processed by the RF front-end is down-mixed with the 420MHz signal generated by the phase-locked loop to obtain an intermediate frequency signal in the 10~20MHz frequency band. The high-frequency image frequency is eliminated by a low-pass filter with a cutoff frequency of 20MHz, and then amplified by an intermediate frequency amplifier. After being acquired by a 12-bit resolution analog-to-digital converter, the echo signal is uploaded to the host computer via a serial port for signal processing. According to the length of the echo signal, the number of sampling points is set to 660, and each sampling point is represented in double-byte hexadecimal format.
[0030] Step G: In the signal preprocessing section, first remove the DC component from the ADC data, then add a Blackman-Harris window to suppress spectral leakage caused by data truncation;
[0031] Step H: Pad the data with 0 to 8192 points to make the spectrum smoother. After fast Fourier transform, obtain the coarse frequency domain peak information of the five surface acoustic wave sensors on the spectrum. The first four echoes correspond to the torque measurement device, and the fifth echo corresponds to the temperature measurement device.
[0032] Step 1: Within a narrow band of ±50kHz of five coarse frequency domain peaks, perform CZT spectrum refinement at 500 points. At the peaks after spectrum refinement, apply Quinn interpolation estimation. Calculate the precise frequency offset δ using the ratio of the peak and its left and right adjacent points to further optimize the frequency estimation accuracy.
[0033] Step J: Use a moving average filter with a length of 10 to suppress high-frequency random noise and transient disturbances;
[0034] Step K: Based on the relationship between torque, temperature, and the resonant frequency of the surface acoustic wave (SAW) sensor, a corresponding conversion is performed. The real-time temperature is measured using a SAW temperature measuring device. Based on the frequency torque coefficient at different temperatures, dual-parameter decoupling of torque and temperature is achieved. The torque value is obtained from the measurement results of four torque sensors, eliminating the influence of temperature. The measurement results are displayed on the display module to achieve real-time parallel monitoring of aero-engine torque and temperature.
[0035] The present invention has the following beneficial effects:
[0036] (1) A segmented loop rotor antenna with dual-feed ports is adopted, and a direct-plug mechanical and electrical connection with the sensing substrate is achieved through an RF coaxial connector, eliminating the traditional cable connection method or the radiator integrated sensing substrate method. This structure avoids the problem of loosening or falling off the connection due to centrifugal force and vibration under high speed conditions, optimizes the dynamic mass distribution of the rotor part, meets the strict requirements for dynamic balance, and reduces the influence of the metal shaft on the antenna radiation performance.
[0037] (2) In terms of excitation method, a linear frequency modulation signal is used instead of the traditional single-frequency sinusoidal excitation, so that the excitation signal continuously covers the resonant frequency of the surface acoustic wave sensor within a certain bandwidth. When the resonant frequency shifts due to torque or temperature changes, the sensor can still be effectively excited, fundamentally avoiding the problem of echo signal attenuation or even loss due to detuning, thereby significantly improving the detection stability and effective measurement distance of the system under complex working conditions.
[0038] (3) By adopting a dual-feed port symmetrical segmented loop antenna structure at the sensor end and a single-feed port segmented loop antenna structure at the reader end, the uniformity of signal radiation and reception is effectively improved, and the influence of directional change on the signal during rotation is reduced. At the same time, the corresponding placement of the radiating layers effectively suppresses the problem of echo amplitude fluctuation and spectrum peak jitter, thereby improving the anti-interference capability and measurement reliability of the system.
[0039] (4) In the signal processing, a method combining zero-padding fast Fourier transform, linear frequency modulated Z-transform spectrum refinement, and Quinn interpolation estimation algorithm is adopted to achieve refined spectrum analysis and peak location under the condition of finite length time domain signal. This method overcomes the problem of limited frequency resolution of traditional Fourier transform, and significantly improves frequency estimation accuracy while ensuring response speed, thereby meeting the dual requirements of high precision and real-time performance for dynamic measurement. Attached Figure Description
[0040] Figure 1 This is a schematic diagram of the parallel monitoring system for torque and temperature of an aero-engine based on a surface acoustic wave resonator, as presented in this invention.
[0041] Figure 2 This is a schematic diagram of a surface acoustic wave sensor and its sensing substrate.
[0042] Figure 3 This is a top view of the surface acoustic wave sensor and its antenna.
[0043] Figure 4 This is a side view schematic diagram of the surface acoustic wave sensor and its antenna.
[0044] Figure 5 This is a schematic diagram of a reader.
[0045] Figure 6 This is a schematic diagram of the reader antenna.
[0046] Figure 7 This is a schematic diagram of the reader's transmission link.
[0047] Figure 8 This is a schematic diagram of the spectrum of a linear frequency modulated signal.
[0048] Figure 9 This is a schematic diagram of the spectrum of the upmixer architecture.
[0049] Figure 10 This is a schematic diagram of the reader's receiving link.
[0050] Figure 11 This is a schematic diagram of the overall echo signal spectrum of the aero-engine torque and temperature parallel monitoring system based on surface acoustic wave resonators according to the present invention.
[0051] Figure 12 A schematic diagram of the spectrum for a spectrum refinement algorithm of linear frequency modulated Z-transform in signal processing.
[0052] Figure 13 A schematic diagram of the frequency estimation algorithm for Quinn interpolation in signal processing.
[0053] In the picture:
[0054] 1. Sensor substrate; 2. Sensor antenna; 3. Reader antenna; 4. Reader; 5. Shaft; 7. Temperature measuring device; 8. RF coaxial connector; 9. High-frequency impedance line; 10. First torque measuring device; 11. Second torque measuring device; 12. Third torque measuring device; 13. Fourth torque measuring device; 14. First feed port; 15. Second feed port; 16. Pad; 17. Gold wire; 60. Piezoelectric substrate; 61. Interdigital transducer; 62. Reflector grating array; 63. Busbar. Detailed Implementation
[0055] The invention will now be further described with reference to the accompanying drawings.
[0056] This invention relates to a parallel monitoring system for torque and temperature in aero-engines based on surface acoustic wave (SAW) resonators. It comprises a SAW sensor, a sensing substrate 1, a sensor antenna 2, a reader antenna 3, a reader 4, and a rotating shaft 5. The SAW sensor includes four torque measuring devices and one temperature measuring device 7. Two sensing substrates 1 are located on opposite sides of the rotating shaft 5. Each sensing substrate 1 has two spaced-apart openings. The four torque measuring devices are located in the openings on the two sensing substrates 1, allowing them to be directly attached to the surface of the rotating shaft 5 and arranged symmetrically and differentially along the shaft 5. These devices are used to sense and measure the strain caused by the torsion of the rotating shaft 5. The temperature measuring device 7 is encapsulated and directly soldered to the surface of the sensing substrate 1; it is sensitive only to temperature changes and unaffected by torque.
[0057] The sensor antenna 2 is a segmented loop rotor antenna with a dual-feed port and a through-hole structure. It is fixed to the sensing substrate 1 through the radio frequency coaxial connector 8. The substrate of the sensor antenna 2 is Rogers high-frequency microwave board, which utilizes its low dielectric loss and excellent thermal stability to ensure the signal transmission quality of the system under high temperature and high speed conditions.
[0058] The four torque measuring devices are designated as a first torque measuring device 10, a second torque measuring device 11, a third torque measuring device 12, and a fourth torque measuring device 13. The initial resonant frequency of the first torque measuring device 10 is 431 MHz, the initial resonant frequency of the second torque measuring device 11 is 433 MHz, the initial resonant frequency of the third torque measuring device 12 is 435 MHz, and the initial resonant frequency of the fourth torque measuring device 13 is 437 MHz. The initial resonant frequency of the temperature measuring device 7 is 439 MHz. The four torque measuring devices and one temperature measuring device achieve parallel monitoring of the torque and temperature of the aero-engine using a frequency division multiple access (FDMA) method.
[0059] The torque measuring device uses a single-component high-temperature resistant epoxy adhesive, which is directly bonded to the surface of the rotating shaft 5 through an opening in the sensing substrate 1 to maximize sensitivity and transmission of strain caused by torsion. The temperature measuring device 7 is bonded and fixed to the alumina ceramic and Kovar alloy composite package with high-temperature resistant elastic adhesive and then welded to the surface of the sensing substrate 1, making it almost unaffected by torque variables and only sensitive to and measuring temperature parameters.
[0060] The first torque measuring device 10 and the fourth torque measuring device 13 form a differential pair. After being connected in parallel with the temperature measuring device 7, they are arranged on one of the sensing substrates 1 and connected to the first feed port 14 of the sensor antenna 2. The second torque measuring device 11 and the third torque measuring device 12 form a differential pair and are connected to the second feed port 15 of the sensor antenna.
[0061] The sensing substrate 1 uses an epoxy resin glass cloth laminate with a double-layer structure. High-frequency impedance lines 9 and pads 16 are printed on the upper layer of the sensing substrate 1. The torque measuring device 6 includes a piezoelectric substrate 60, an interdigital transducer 61, and a reflective grating array 62. Busbars 63 are provided on opposite sides of the interdigital transducer 61. On the sensing substrate 1, which houses two torque measuring devices and one temperature measuring device 7, an RF coaxial connector 8 is connected to each torque measuring device 6 and temperature measuring device 7 via high-frequency impedance lines 9.
[0062] The pads 16 are plated with 2-microinch-thick immersion gold and are located on opposite sides of the torque measuring device 6. One side of the pad 16 is bonded to the high-frequency impedance line 9 using ball bonding. Each pad 16 is connected in parallel to the busbars 63 on both sides of the interdigital transducer 61 via gold wires 17 and flows back to the lower layer of the sensing substrate 1. The thickness of the sensing substrate 1 is slightly greater than the sum of the thicknesses of the torque measuring device, the temperature measuring device, and the adhesive to reduce the parasitic inductance caused by the height difference of the gold wire bonding 17.
[0063] The piezoelectric substrate 60 of the torque measuring device is an AT-cut quartz substrate with Euler angles of (0°, 126°, 0°), exhibiting high torque sensitivity but low temperature sensitivity. The piezoelectric substrate (not shown) of the temperature measuring device 7 is a YX-cut quartz substrate with Euler angles of (0°, 90°, 0°), exhibiting high temperature sensitivity.
[0064] The reader antenna 3 is a single-feed port segmented loop antenna with a center frequency in the 430~440MHz band. The reader antenna 3 and the sensor antenna 2 adopt an equivalent radiation structure and place the radiating layers accordingly, which improves the uniformity of signal radiation and reception, reduces the impact of directional changes on the signal during rotation, and effectively suppresses the problems of echo amplitude fluctuation and spectral peak jitter.
[0065] The reader 4 includes a transmit link, a receive link, a transmit / receive isolation switch, and a main control module. The transmit / receive isolation switch is a single-pole double-throw switch used to switch between transmit and receive states. The first port a of the transmit / receive isolation switch is connected to the reader antenna, the second port b is connected to the transmit link, and the third port c is connected to the receive link.
[0066] Please refer to Figure 7 As shown, the transmit link includes a direct digital frequency synthesizer, a phase-locked loop, an up-mixer, a power amplifier, and a combiner, used to generate and output multiple excitation signals. The direct digital frequency synthesizer integrates three independent output channels: DDS1, DDS2, and DDS3. The reader employs a multi-local oscillator architecture, using a linear frequency modulated (LFM) signal for excitation, overcoming the detuning problem caused by sensor resonant frequency offset, thus expanding the frequency measurement range and wireless testing distance. The spectrum of the LFM signal is shown below. Figure 8 As shown in the diagram. Simultaneously, the image frequency, which should have been suppressed during the upmixing stage, is also retained and transmitted as an excitation signal, achieving a low-cost excitation scheme for multiple local oscillators. Specifically, the excitation signal for the torque measurement device is generated from the upmixed RF signal and image signal of DDS1, DDS2, and PLL1; the excitation signal for the temperature measurement device is generated from the upmixed image signal of DDS3 and PLL1. The final upmixing architecture spectrum diagram of the excitation signal is shown in the diagram. Figure 9 As shown.
[0067] Please refer to Figure 10 As shown, the receiving link consists of an RF front-end and a baseband back-end. It includes a first low-noise amplifier, a first surface acoustic wave (SAW) bandpass filter, a second low-noise amplifier, a second SAW bandpass filter, a phase-locked loop (PLL), a downmixer, a baseband low-pass filter, a differential amplifier, and an analog-to-digital converter (ADC).
[0068] The output of the first low-noise amplifier is connected to the input of the first surface acoustic wave bandpass filter, the output of the first surface acoustic wave bandpass filter is connected to the input of the second low-noise amplifier, and the output of the second low-noise amplifier is connected to the input of the second surface acoustic wave bandpass filter, thus forming the radio frequency front end of the receiving link, and after down-mixing with PLL2, it is processed into intermediate frequency baseband.
[0069] The output of the baseband low-pass filter is connected to the input of the differential amplifier, and the output of the differential amplifier is connected to the input of the ADC. This forms the baseband back-end structure, which avoids the self-oscillation problem that is easy to occur when weak echo signals are amplified only in the radio frequency band. At the same time, it enables the measurement system to have a higher receiving gain and further improves the wireless detection distance.
[0070] The main control module is a field-programmable gate array (FPGA), which performs timing control on the transceiver isolating switches, drives each chip, and drives the transmission interface to transmit data to the host computer.
[0071] like Figure 11 As shown, in the host computer, the DC component of the acquired signal is first removed and windowed to suppress spectral leakage. Then, the data is padded with zeros and subjected to a fast Fourier transform to obtain a frequency domain signal containing multiple spectral peaks. The first four echoes correspond to the surface acoustic wave torsion measuring device, and the fifth echo corresponds to the surface acoustic wave temperature measuring device.
[0072] like Figure 12 As shown, after initially acquiring the coarse spectrum of the signal, a preset narrowband range is set near each coarse peak to further improve the accuracy of peak localization. Within this local interval, linear frequency modulated Z-transform (CZT) is applied for spectrum refinement. CZT can flexibly calculate the Z-transform of sampling points along a spiral curve. The figure clearly compares the spectra before and after refinement: the solid line represents the original coarsely estimated peak value, and the dashed line represents the spectrum curve after CZT processing. This significantly improves the frequency resolution of the local area without increasing the overall computational load.
[0073] like Figure 13 As shown, after refining the linear frequency modulated Z-transform spectrum, although the resolution within local frequency bands has been significantly improved, the discrete points still deviate from the true peak frequency. To further improve the frequency estimation accuracy, the Quinn interpolation estimation algorithm can be used to accurately calculate the frequency offset based on each refined peak and its adjacent frequency points. This algorithm uses the amplitude ratio of the two spectral lines near the peak to construct a frequency offset correction factor δ, and then performs interpolation correction on the discrete peak positions. The figure shows the comparison before and after Quinn interpolation: the solid line is the original spectrum, and the dashed line is the final frequency spectrum estimated after Quinn interpolation.
[0074] In the reader's receiving link, the signal acquisition section uses an asynchronous FIFO buffer to achieve cross-clock domain matching between the acquisition clock and the serial port transmission clock. It also splits the ASCII code sampled values after analog-to-digital conversion into high and low bytes and transmits them to the host computer in the original byte form for frame reassembly and signal processing, thereby increasing the data throughput by 150%.
[0075] The host computer sequentially preprocesses the surface acoustic wave (SAW) echo signal by removing DC, applying a Blackman-Harris window, and zero-padding. A coarse frequency domain peak is obtained through Fast Fourier Transform (FFT). A 500-point linear frequency-modulated Z-transform (CZT) is then performed within ±50 kHz of this peak to refine the spectrum. Finally, Quinn interpolation is applied to the CZT peak to calculate the frequency offset, thus obtaining the resonator's precise frequency. This signal processing flow is based on the MATLAB platform, with each measurement taking approximately 1–5 ms. Therefore, by using high-speed communication interfaces such as UDP, PCIe, and USB 3.0, the system's dynamic response will reach over 200 Hz.
[0076] This invention relates to a parallel monitoring method for torque and temperature in aero-engines based on surface acoustic wave resonators, comprising the following methods:
[0077] Step A: After the system is powered on, the main control module controls the reader's transmit / receive isolation switch to connect with the transmit link and transmits an excitation signal containing 5 carrier waves. The excitation signal is transmitted to the sensor antenna in the form of electromagnetic waves through the combiner and the reader antenna. The carrier waves of the 5 excitation signals are linear frequency modulated signals, and their starting frequencies are the initial resonant frequencies of each surface acoustic wave sensor. The bandwidth is 750kHz and the linear frequency modulation time is 15μs.
[0078] Step B: After the sensor antenna receives the excitation signal, the interdigital transducer converts the excitation signal component located in the resonant cavity into a surface acoustic wave through the inverse piezoelectric effect. After multiple coherent reflections and superpositions by the reflective grating array, a standing wave is formed in the resonant cavity, completing energy accumulation.
[0079] Step C: When the duration of the excitation signal is greater than the charging time of the resonant cavity of the surface acoustic wave sensor, the transceiver isolation switch of the reader is connected to the receiving link.
[0080] Step D: The standing waves from the five surface acoustic wave sensors are converted into electromagnetic waves via interdigital transducers using the piezoelectric effect, and the echo signals are radiated from the sensor antennas to the reader antenna. These echo signals contain the resonant frequency information of each of the five sensors.
[0081] Step E: The echo signal is amplified by two stages of low-noise amplifiers and then filtered out by two stages of RF bandpass filters to remove out-of-band interference.
[0082] Step F: The echo signal processed by the RF front-end is down-mixed with the 420MHz signal generated by the phase-locked loop to obtain an intermediate frequency (IF) signal in the 10-20MHz frequency band. This IF signal is then passed through a low-pass filter with a cutoff frequency of 20MHz to eliminate high-frequency image frequencies, amplified by the IF amplifier, and acquired by a 12-bit resolution analog-to-digital converter. The echo signal is then uploaded to the host computer via a serial port for signal processing. Based on the echo signal length, the number of sampling points is set to 660, with each sampling point represented in double-byte hexadecimal format.
[0083] Step G: In the signal preprocessing section, firstly, the DC component in the ADC data is removed to avoid the peak information at zero frequency affecting the measurement results. Then, a Blackman-Harris window is added to suppress spectral leakage caused by data truncation.
[0084] Step H: Pad the data with 0 to 8192 points to make the spectrum smoother and facilitate rough peak finding. After Fast Fourier Transform, the rough frequency domain peak information of the five surface acoustic wave sensors is obtained in the spectrum, where the first four echoes correspond to the torque measurement device and the fifth echo corresponds to the temperature measurement device.
[0085] Step 1: Within a narrow band of ±50kHz of five coarse frequency domain peaks, perform CZT spectrum refinement at 500 points. At the peaks after spectrum refinement, apply Quinn interpolation estimation. Calculate the precise frequency offset δ using the ratio of the peak and its left and right adjacent points to further optimize the frequency estimation accuracy.
[0086] Step J: Use a moving average filter with a length of 10 to suppress high-frequency random noise and transient disturbances.
[0087] Step K: Based on the relationship between torque, temperature, and the resonant frequency of the surface acoustic wave (SAW) sensor, perform the corresponding conversion. Measure the real-time temperature using a SAW temperature measuring device. Decouple the torque and temperature physical quantities based on the frequency-torque coefficient TOCF at different temperatures. Let the TOCF value at temperature T and torque M be TOCF(T, M), defined as:
[0088]
[0089] The above equation, after transformation, yields:
[0090]
[0091] in Let T be the resonant frequency at temperature T and torque M. At room temperature The initial value is 0 Nm. Using experimentally calibrated data, the TOCF at different temperatures is calculated using the same torque value M (≠ M0), and its first-order fitted curve with respect to T is plotted, denoted as... Different M values calculated and The torque value M can be fitted as an inverse proportional function, i.e. , ,in ~ Determined through calibration experiments.
[0092] The temperature sensing device is only affected by temperature changes. By calibrating the device in a constant temperature chamber, its resonant frequency at different temperatures can be obtained. When the ambient temperature changes, the frequency of the temperature sensing device will reflect the ambient temperature T. After calculating the temperature value, it can be substituted into... middle, and All of these can be measured using a vector network analyzer, and the torque value can be obtained by solving a linear equation in one variable.
[0093] The calculated temperature and torque results are output to the display module for real-time parallel monitoring of the aero-engine's operating status.
[0094] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be considered within the scope of protection of the present invention.
Claims
1. A parallel monitoring system for torque and temperature of an aero-engine based on a surface acoustic wave resonator, characterized in that: The system comprises a surface acoustic wave sensor, a sensing substrate (1), a sensor antenna (2), a reader antenna (3), a reader (4), and a rotating shaft (5). The surface acoustic wave sensor includes four torque measuring devices and one temperature measuring device (7). The four torque measuring devices are differentially arranged on both sides of the rotating shaft (5), and the temperature measuring device is arranged separately on the surface of the sensing substrate. The sensor antenna (2) is a dual-feed port segmented loop rotor antenna with a through-hole structure, which is fixed to the sensing substrate (1) through an RF coaxial connector (8). The reader antenna (3) is a single-feed port segmented loop antenna. The reader (4) includes a transmit link, a receive link, a transmit / receive isolation switch, and a main control module. The transceiver isolating switch is a single-pole double-throw switch used to switch between the transmit and receive states. The first port a of the transceiver isolating switch is connected to the reader antenna, the second port b is connected to the transmit link, and the third port c is connected to the receive link. The reader (4) adopts a multi-local oscillator architecture, is excited by a linear frequency modulated signal, and retains the image frequency that should be suppressed in the up-mixing stage, and transmits it as an excitation signal. The echo signal radiated by the sensor antenna (2) to the reader antenna (3) is collected by down-mixing, and then processed by a signal processing flow combining zero-padding fast Fourier transform, linear frequency modulated Z-transform spectrum refinement and Quinn interpolation estimation to achieve high-resolution frequency calculation under finite signal length.
2. The parallel monitoring system for aero-engine torque and temperature based on a surface acoustic wave resonator as described in claim 1, characterized in that: The torque measuring device (6) includes a piezoelectric substrate (60), an interdigital transducer (61), and a reflective grating array (62). Busbars (63) are provided on opposite sides of the interdigital transducer (61). On the sensing substrate (1) on which two torque measuring devices and one temperature measuring device (7) are mounted, the radio frequency coaxial connector (8) is connected to each torque measuring device (6) and temperature measuring device (7) through a high-frequency impedance line (9). The piezoelectric substrate (60) of the torque measuring device is an AT-cut quartz substrate with Euler angles of (0°, 126°, 0°). The piezoelectric substrate of the temperature measuring device (7) is a YX-cut quartz substrate with Euler angles of (0°, 90°, 0°).
3. The parallel monitoring system for aero-engine torque and temperature based on a surface acoustic wave resonator as described in claim 2, characterized in that: The four torque measuring devices are a first torque measuring device (10), a second torque measuring device (11), a third torque measuring device (12), and a fourth torque measuring device (13). The initial resonant frequency of the first torque measuring device (10) is 431MHz, the initial resonant frequency of the second torque measuring device (11) is 433MHz, the initial resonant frequency of the third torque measuring device (12) is 435MHz, the initial resonant frequency of the fourth torque measuring device (13) is 437MHz, and the initial resonant frequency of the temperature measuring device (7) is 439MHz. The four torque measuring devices and one temperature measuring device achieve parallel monitoring of the torque and temperature of the aero-engine in a frequency division multiple access manner.
4. The parallel monitoring system for aero-engine torque and temperature based on a surface acoustic wave resonator as described in claim 3, characterized in that: The first torque measuring device (10) and the fourth torque measuring device (13) form a differential pair. After being connected in parallel with the temperature measuring device (7), they are arranged on one of the sensing substrates (1) and connected to the first feed port (14) of the sensor antenna (2). The second torque measuring device (11) and the third torque measuring device (12) form a differential pair and are connected to the second feed port (15) of the sensor antenna.
5. The parallel monitoring system for aero-engine torque and temperature based on a surface acoustic wave resonator as described in claim 4, characterized in that: The torque measuring device uses a single-component high-temperature resistant epoxy adhesive, which is directly pasted onto the surface of the rotating shaft (5) through the opening on the sensing substrate (1). The temperature measuring element (7) is bonded and fixed in the alumina ceramic and Kovar alloy composite package with high-temperature resistant elastic adhesive and then welded to the surface of the sensing substrate (1).
6. The parallel monitoring system for aero-engine torque and temperature based on a surface acoustic wave resonator as described in claim 5, characterized in that: The sensing substrate (1) uses epoxy resin glass cloth laminate and adopts a double-layer structure. High-frequency impedance lines (9) and pads (16) are printed on the upper layer of the sensing substrate (1). The pads (16) are sprayed with immersion gold with a thickness of 2 microinches. They are located on opposite sides of the torque measuring device (6). One side of the pad (16) is bonded to the high-frequency impedance line (9) by ball bonding, and the other side is connected to ground. Each pad (16) is connected to the busbars (63) on both sides of the interdigital transducer (61) by gold wire (17) and flows back to the lower layer of the sensing substrate (1).
7. The parallel monitoring system for aero-engine torque and temperature based on a surface acoustic wave resonator as described in claim 6, characterized in that: The reader generates a linear frequency modulated excitation signal via a transmission link, the starting frequency of which corresponds to the initial resonant frequency of the surface acoustic wave sensor. The excitation signal of the torque measuring device is generated by mixing the radio frequency signal and the image signal on the phase-locked loop PLL1 with DDS1, DDS2, and PLL1. The excitation signal of the temperature measuring device is generated by mixing the image signal on the phase-locked loop PLL1 with DDS3.
8. The parallel monitoring system for aero-engine torque and temperature based on a surface acoustic wave resonator as described in claim 7, characterized in that: The signal acquisition part of the reader receiving link uses an asynchronous FIFO buffer to achieve cross-clock domain matching between the acquisition clock and the transmission clock, and splits the ASCII code sample value after analog-to-digital conversion into high and low bytes and transmits it to the host computer in the form of raw bytes for frame reassembly and signal processing.
9. The parallel monitoring system for torque and temperature of aero-engines based on surface acoustic wave resonators as described in claim 8, characterized in that: The host computer sequentially performs DC removal, Blackman-Harris windowing, and zero-padding preprocessing on the surface acoustic wave echo signal. It then obtains a rough frequency domain peak value through Fast Fourier Transform, performs a 500-point linear frequency modulated Z-transform within ±50 kHz of the peak value to refine the spectrum, and finally applies Quinn interpolation to the CZT peak point to calculate the frequency offset in order to obtain the accurate frequency of the resonator.
10. A method for parallel monitoring of torque and temperature in aero-engines based on surface acoustic wave resonators, characterized in that: Includes the following steps: Step A: After the system is powered on, the main control module controls the reader's transmit / receive isolation switch to connect with the transmit link and transmits an excitation signal containing 5 carrier waves. The excitation signal is transmitted to the sensor antenna in the form of electromagnetic waves through the combiner and the reader antenna. The carrier waves of the 5 excitation signals are linear frequency modulated signals, and their starting frequencies are the initial resonant frequencies of each surface acoustic wave sensor. The bandwidth is 750kHz and the linear frequency modulation time is 15μs. Step B: After receiving the excitation signal, the sensor antenna converts the excitation signal component located in the resonant cavity into a surface acoustic wave through the inverse piezoelectric effect. After multiple coherent reflections and superpositions by the reflective grating array, a standing wave is formed in the resonant cavity, completing energy accumulation. Step C: When the duration of the excitation signal is greater than the charging time of the resonant cavity of the surface acoustic wave sensor, the transceiver isolation switch of the reader is connected to the receiving link. Step D: The standing waves of the five surface acoustic wave sensors are converted into electromagnetic waves by interdigital transducers through the positive piezoelectric effect. The echo signal is radiated from the sensor antenna to the reader antenna. The echo signal contains the resonant frequency information of each of the five sensors. Step E: The echo signal is amplified by two stages of low-noise amplifiers, and then filtered out by two stages of RF bandpass filters to remove out-of-band interference; Step F: The echo signal processed by the RF front end is down-mixed with the 420MHz signal generated by the phase-locked loop to obtain an intermediate frequency signal in the 10~20MHz frequency band. The high-frequency image frequency is eliminated by a low-pass filter with a cutoff frequency of 20MHz, and then amplified by an intermediate frequency amplifier. After being acquired by a 12-bit resolution analog-to-digital converter, the echo signal is uploaded to the host computer for signal processing. According to the length of the echo signal, the number of sampling points is set to 660, and each sampling point is represented in double-byte hexadecimal format. Step G: In the signal preprocessing section, first remove the DC component from the ADC data, then add a Blackman-Harris window to suppress spectral leakage caused by data truncation; Step H: Pad the data with 0 to 8192 points to make the spectrum smoother. After fast Fourier transform, obtain the coarse frequency domain peak information of the five surface acoustic wave sensors on the spectrum. The first four echoes correspond to the torque measurement device, and the fifth echo corresponds to the temperature measurement device. Step 1: Within a narrow band of ±50kHz of five coarse frequency domain peaks, perform CZT spectrum refinement at 500 points. At the peaks after spectrum refinement, apply Quinn interpolation estimation. Calculate the precise frequency offset δ using the ratio of the peak and its left and right adjacent points to further optimize the frequency estimation accuracy. Step J: Use a moving average filter with a length of 10 to suppress high-frequency random noise and transient disturbances; Step K: Based on the relationship between torque, temperature and the resonant frequency of the surface acoustic wave sensor, perform the corresponding conversion, measure the real-time temperature through the surface acoustic wave temperature measuring device, decouple the torque and temperature physical quantities according to the frequency torque coefficient at different temperatures, obtain the torque value based on the measurement results of the four torque sensors and eliminate the influence of temperature, and display the measurement results on the display module to realize real-time parallel monitoring of the torque and temperature of the aero-engine.
