Wireless telemetering method for torque and rotating speed of gear transmission system in limited space

Abstract

The invention provides a wireless telemetering method for torque and rotating speed of a gear transmission system in a limited space. According to the method, a torque signal and a rotating speed signal on a corresponding shaft can be extracted by measuring a torsional deformation signal and a bending deformation signal of an input shaft or an output shaft of a transmission gear and combining a strain signal analysis and processing method, so that indirect measurement of the torque and the rotating speed is realized; strain signals are collected, emitted, transmitted and stored through the wireless signal transmission technology, and the problem of signal transmission line winding caused by a wired signal transmission mode in the rotating process of a measured piece is solved. Compared with the traditional mode of measuring the rotating speed or the torque by adopting a rotating speed / torque sensor, the measuring method has the advantages of low requirement on the measuring space, low measuring cost, reliable measuring precision and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of gear transmission, and in particular to a wireless remote measurement method for torque and speed of a gear transmission system in a confined space. Background Art

[0002] Gear transmission systems offer numerous advantages, including compact structure, high transmission efficiency, high power density ratio, strong load-bearing capacity, and high reliability. They are widely used in the automotive, marine, and aviation industries. Torque and speed are crucial operating parameters of gear transmission systems, impacting their dynamic performance, lifespan, and safety. Accurate, timely, convenient, and reliable measurement of torque and speed is crucial for analyzing and studying the stress response, dynamic performance, and fatigue life of gear transmission systems, guiding their design and improving their performance. Therefore, torque and speed measurement is essential for gear transmission system design and development, testing and analysis, quality assurance, safety, and performance optimization.

[0003] The current torque measurement method mainly uses contact sensing technology, and its implementation methods can be divided into two categories: one is to integrate a torque sensor at the input / output shaft end of the gear transmission system. Although this method can achieve direct measurement, it is limited by the sensor volume and installation space conditions and is only suitable for laboratory environments or special equipment with reserved measurement interfaces; the second is an indirect measurement technology based on strain gauge measurement, which calculates the torque value by detecting the elastic torsional deformation of the drive shaft. Although this method breaks through spatial limitations, it faces fundamental constraints on the application of wired transmission methods on rotating parts.

[0004] Current methods for measuring rotational speed primarily rely on non-contact sensing technology, which can be categorized into two types: First, a photoelectric encoder or magnetoelectric tachometer is mounted on the surface of the rotating shaft in a gear transmission system. This method measures rotational speed by detecting periodic signals from surface markers. While achieving high measurement accuracy, this method is limited by the sensor's mounting accuracy and the dynamic balancing requirements of the rotating body, making it applicable only to test benches with precise assembly. Second, an indirect measurement technique based on the Doppler effect measures rotational speed by capturing the frequency offset of laser light reflected from the rotating surface. While this method overcomes the limitations of physical contact, it suffers from high cost and limited inter-axis space.

[0005] Therefore, it is of great significance to develop a wireless telemetry method for torque and speed of gear transmission systems in confined spaces. Summary of the Invention

[0006] The purpose of the present invention is to provide a wireless remote measurement method for torque and speed of a gear transmission system in a confined space, so as to solve the problems existing in the prior art.

[0007] The technical solution employed to achieve the objectives of the present invention is as follows: a method for wirelessly remotely measuring the torque and speed of a gear transmission system in a confined space. The output shaft of a drive motor is connected to the input shaft of the gear transmission system under test via a coupling. The input shaft of the gear transmission system under test is selected as the measured shaft. The wireless remote measurement method specifically includes the following steps:

[0008] 1) Four sets of strain gauges 1 are arranged circumferentially at intervals around the midsection of the measured shaft. Each set of strain gauges 1 includes orthogonally arranged bending strain gauges and torsional strain gauges, forming a dual-parameter synchronous sensing array. The bending strain gauges are arranged along the normal direction, while the torsional strain gauges are arranged along the tangential direction.

[0009] 2) Fixed installation of wireless strain collection nodes 2. Each group of strain units 1 is equipped with an independent wireless strain collection node 2, and a multi-node spatially redundant layout establishes a measurement data mutual verification mechanism. The wireless strain collection node 2 includes an installation box 201 and an acquisition terminal 202. The acquisition terminal is housed within the inner cavity of the installation box 201. The installation box 201 is fixedly connected to the outer wall of the coupling. Each installation box 201 is located at the corresponding position of the strain unit 1. The acquisition terminal 202 includes a full-bridge circuit, an ADS1220 ADC chip, an ESP32-S2 control chip, an SD card, a Wi-Fi radio module, and a battery pack power supply. The full-bridge circuit senses bending and torsional strain signals. The ADS1220 ADC chip converts analog strain signals into digital signals. The ESP32-S2 control chip serves as the main controller, controlling the scheduling of acquisition, storage, and wireless transmission functions. The SD card is used for data storage. The Wi-Fi radio module is used for wireless data transmission. The battery pack power supply provides power to each hardware module. To address the characteristics of rotating working conditions, acquisition terminal 202 utilizes a distributed dual-channel acquisition node architecture. Wireless acquisition nodes use dual-channel ADS1220 ADC chips to collect bending and torsional strain data from the measured shaft, store it on an SD card, and wirelessly transmit it to a computer for storage and analysis. The SD card is used to cache the raw data locally, and a time-division multiplexing protocol is used between nodes to prevent signal collisions.

[0010] 3) Data collection: When the gear transmission system is working, the strain unit 1 collects strain data. The wireless strain collection node 2 receives the real-time strain data from the corresponding strain unit 1 via the wireless network and uploads it to the computer 3 in real time.

[0011] 4) Process the strain data. Step 4) specifically includes the following sub-steps:

[0012] 4.1) Perform voltage domain conversion on the 24-bit raw AD sampling signal acquired by the wireless strain acquisition node. By setting a full-scale voltage range of ±5V, a linear mapping relationship between the AD sampling value and the actual voltage signal is established, achieving voltage quantization of the raw data.

[0013] 4.2) Eliminate the influence of chip gain.

[0014] 4.3) A baseline calibration strategy is used to eliminate the impact of installation errors on dynamic measurement data.

[0015] 4.4) A voltage-strain conversion model is constructed based on the mechanical properties of the material to convert the compensated voltage signal into a dimensionless strain variable.

[0016] 4.5) Use Butterworth low-pass filter to reduce noise of the decoupled strain signal.

[0017] 5) Calculate the measured shaft torque based on the strain data. Filter and demodulate the strain gauge voltage data to obtain the strain signal, and then calculate the effective torque value based on the shaft cross-sectional parameters.

[0018] 6) Calculate the measured shaft speed based on the strain data. Step 6) specifically includes the following sub-steps:

[0019] 6.1) A window function is selected based on the stationary characteristics of the bending and torsional strain signals, and the signals are framed based on the selected window function so that the signal in each window can be approximated as a stationary signal.

[0020] 6.2) Perform short-time Fourier transform on the framed data to construct a two-dimensional time-frequency energy matrix to form a time-frequency spectrum.

[0021] 6.3) The time-frequency ridge extraction method based on energy threshold is used to extract the transition frequency ridges.

[0022] 6.4) Calculate the shaft speed based on the corresponding relationship between frequency and speed.

[0023] Furthermore, the strain unit 1 comprises a base layer 101, a strain gauge 102, a dynamic compensation layer 103, and a protective layer 104, stacked in sequence. The base layer 101 is evenly coated with epoxy resin glue using a high-precision dispensing process to ensure reliable coupling between the strain gauge 102 and the substrate. The dynamic compensation layer 103 utilizes a temperature-compensating strain gauge and a thin-film temperature sensor to enable real-time temperature field monitoring. The protective layer 104 is constructed by stacking a carbon fiber composite protective cover and localized copper foil shielding tape to create a composite structure for electromagnetic shielding and physical protection.

[0024] Furthermore, in step 4), the AD quantization formula is as follows:

[0025]

[0026] Where V a Indicates the 24-bit AD chip sampling value, V q Indicates the quantization value, 10 indicates the maximum sampling output range of the AD chip, and 24 indicates the number of quantization bits of the AD chip.

[0027] The formula for the true quantization value after eliminating the influence of chip gain is as follows:

[0028] V q =V q / 128(2)

[0029] In step 4.3), the baseline calibration formula is as follows:

[0030] V q =V q -V init (3)

[0031]

[0032] In formula (3), V init It is obtained by collecting the strain signal of the measured shaft in the static reference state, and is taken as 1000 based on experience. Formula (4) is based on the previously calculated torsional strain signal V q Remove DC offset to further improve AC signal quality, where V val It represents the effective value after removing the DC component. The strain voltage signal is converted into the strain quantity, and the formula is as follows:

[0033]

[0034] In formula (5), ε is the strain, ΔR is the resistance change, R is the original resistance value, the voltage mV represents the strain voltage at a certain moment, and K is the strain gauge sensitivity coefficient. Since the collected bending and torsion signals are both subject to noise interference, the original strain signal ε needs to be filtered. A Butterworth low-pass filter is used, and its transfer function can be expressed as follows:

[0035]

[0036] In formula (7), f c represents the cutoff frequency of the Butterworth low-pass filter, f s represents the sampling frequency, w c represents the normalized cutoff frequency, where s represents the complex frequency variable and N represents the filter order.

[0037] Furthermore, in step 5), the torque calculation formula is:

[0038]

[0039] Where T is the measured shaft torque. τ is the shear stress. Zp is the anti-torque interface coefficient, as shown in Equation (9). τmax is the maximum shear stress. E is the longitudinal and transverse coefficient. μ is the Poisson coefficient.

[0040]

[0041] Where D is the outer diameter of the shaft being measured, and d is the inner diameter of the shaft being measured.

[0042] Furthermore, in step 6), a short-time Fourier transform is performed on the filtered bending and torsion signal. Assume that there is a discrete-time signal x[n], where n = 0, 1, 2, ..., N-1. A window function w[n] is selected with a length of M, and the signal x[n] is divided into multiple overlapping short-time windows. The length of each window is M, and the overlap length between adjacent windows is R. For the kth window, the starting point of the signal is n = kL, where L = MR is the step size of the window. Therefore, the window function is applied to each window to obtain the window signal X k [n]:

[0043] x k [n]=x[n+kL]·w[n](10)

[0044] In formula (10), n = 0, 1, 2, ..., M-1. Then, for each window signal x[n], its discrete Fourier transform is calculated:

[0045]

[0046] Where f = 0, 1, 2,…, M-1.

[0047] The Fourier transform result X of each window k [f], are combined into a two-dimensional matrix to form a time-frequency spectrum. The horizontal axis of the spectrum represents time, and the vertical axis represents frequency. After obtaining the time-frequency spectrum, the frequency ridges in the time-frequency spectrum can be extracted.

[0048] Furthermore, in step 6.3), the maximum value is first calculated based on the time-frequency matrix, and a threshold value is set based on the calculated maximum value. If the value in the time-frequency matrix is ​​less than the threshold value, it is set to 0. Then, the maximum value in each column of the time-frequency matrix is ​​calculated to obtain the maximum value of each column of the time-frequency matrix, and the position index of the maximum value in each column is recorded. The corresponding row position index represents the rotational frequency, and then the rotational frequency is converted into a rotational speed signal to obtain a rotational speed signal, which is expressed as follows:

[0049] v=60·f(12)

[0050] In formula (12), f represents the rotational frequency, and v represents the rotational speed signal.

[0051] Furthermore, the installation box 201 is prepared by 3D printing according to actual space.

[0052] Furthermore, the installation box 201 is fixedly mounted on the outer wall of the coupling using an anti-loosening threaded fastener 203.

[0053] The technical effects of the present invention are unquestionable:

[0054] A. It can measure the torque and speed of the gear transmission system when it is impossible to install a torque / speed sensor;

[0055] B. When the application of wired transmission methods on rotating parts is fundamentally restricted, wireless transmission can reduce costs and improve safety and reliability;

[0056] C. Under space-constrained conditions, the wireless acquisition module can be printed and installed according to the actual space, and the indirect measurement and analysis of torque and speed can be achieved through strain signal sensing technology;

[0057] D. It avoids the many design challenges that arise from the rotational motion of both the input and output shafts in the gear transmission system, such as complex support structures and high-precision speed control devices. This makes the test system structure simpler, easier to build and maintain, and shortens the test system development cycle.

[0058] E. It is practical, easy to operate and worthy of promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flow chart of a method for wireless remote measurement of torque and speed of a gear transmission system in a confined space;

[0060] Figure 2 Paste the schematic diagram for the strain gauge;

[0061] Figure 3 This is the working principle of the wireless strain dual-channel acquisition node;

[0062] Figure 4 This is the main functional module diagram of the wireless strain acquisition node;

[0063] Figure 5 To collect torsional strain signal diagram;

[0064] Figure 6 Comparison chart of torque signal calculated from wireless torsional strain signal and torque signal read by control system;

[0065] Figure 7 To collect bending strain signal diagram;

[0066] Figure 8 This is the result of short-time Fourier transform and refinement processing;

[0067] Figure 9 Comparison chart of the speed signal extracted from the wireless strain signal time-frequency and the speed signal read by the control system;

[0068] Figure 10 This is a schematic diagram of the wireless telemetry system;

[0069] Figure 11 This is a schematic diagram of the installation box location; Figure 11 a and 11b are different views of the schematic diagram.

[0070] In the figure: strain unit 1, base layer 101, strain gauge 102, dynamic compensation layer 103, protective layer 104, wireless strain acquisition node 2, installation box 201, acquisition terminal 202, anti-loosening threaded fastener 203, computer 3. DETAILED DESCRIPTION

[0071] The present invention will be further described below with reference to the following examples, but it should not be understood that the scope of the present invention is limited to the following examples. Without departing from the above technical ideas of the present invention, various substitutions and modifications can be made according to common technical knowledge and customary means in the art, and all should be included in the scope of protection of the present invention.

[0072] Example 1:

[0073] The limited installation space of the gear transmission system is manifested in that the shafts are fixed and cannot be removed, the limited space between the shafts prevents the connection of sensors, and the wiring of the shafts when they rotate is difficult. This embodiment provides a method for wireless remote measurement of torque and speed of a gear transmission system in a confined space. The output shaft of the drive motor is connected to the input shaft of the transmission system under test through a coupling. The input shaft of the transmission system under test is selected as the measured shaft. Figure 1 , the wireless telemetry method specifically includes the following steps:

[0074] 1) Lay out the strain gauges. Figure 2 First, four groups of strain units 1 are arranged at intervals along the circumference in the middle section of the measured shaft. Each group of strain units 1 includes orthogonally arranged bending strain gauges and torsional strain gauges to form a dual-parameter synchronous sensing array. The bending strain gauges are arranged along the normal direction, and the torsional strain gauges are arranged along the tangential direction. The strain unit 1 includes a base layer 101, a strain gauge 102, a dynamic compensation layer 103 and a protective layer 104 stacked in sequence. The base layer 101 is evenly coated with epoxy resin glue using a high-precision dispensing process to ensure reliable coupling between the strain gauge 102 and the substrate. The dynamic compensation layer 103 uses a temperature self-compensating strain gauge in the strain gauge and adds a thin film temperature sensor to realize real-time monitoring of the temperature field. The protective layer 104 is stacked with a carbon fiber composite material protective cover and a local copper foil shielding tape to construct an electromagnetic shielding and physical protection composite structure.

[0075] 2) Install wireless data transmission nodes. Figure 10 and Figure 11 Each group of strain units 1 is equipped with an independent wireless strain collection node 2, and a multi-node spatial redundant layout is used to establish a measurement data mutual verification mechanism. The wireless strain collection node 2 includes an installation box 201 and an acquisition terminal 202. The acquisition terminal is accommodated in the inner cavity of the installation box 201. The installation box 201 is fixedly connected to the outer wall of the coupling. Each installation box 201 is correspondingly installed at the position of the corresponding strain unit 1. Figure 3 and Figure 4 The acquisition terminal 202 includes a full-bridge circuit, an ADS1220 ADC chip, an ESP32-S2 control chip, an SD card, a Wi-Fi radio frequency module, and a battery pack power supply. The full-bridge circuit is used to sense bending deformation and torsional deformation strain signals. The ADS1220 ADC chip is used to convert strain analog signals into digital signals. The ESP32-S2 control chip serves as the main controller, controlling the scheduling of acquisition, storage, and wireless transmission functions. The SD card is used for data storage. The Wi-Fi radio frequency module is used for wireless data transmission. The battery pack power supply provides power for each hardware module. In view of the characteristics of the rotating working condition, the acquisition terminal 202 adopts a distributed dual-channel acquisition node architecture. The wireless acquisition node uses the dual-channel ADS1220 ADC chip to respectively collect the bending deformation and torsional deformation strain data of the measured shaft, stores them in the SD card, and wirelessly transmits them to the computer for storage and analysis. The SD card is used to cache the original data locally, and a time division multiplexing protocol is used between nodes to avoid signal collisions.

[0076] 3) Data Collection. When the gear transmission system is operating, strain unit 1 collects strain data. Wireless strain collection node 2 receives the real-time strain data from the corresponding strain unit 1 via the wireless network and uploads it to computer 3 in real time.

[0077] 4) Data conversion and cleaning. This includes the following sub-steps:

[0078] 4.1) Perform voltage domain conversion on the 24-bit AD raw sampling signal acquired by the wireless strain acquisition node. By setting a full-scale voltage range of ±5V, a linear mapping relationship between the AD sampling value and the actual voltage signal is established to achieve voltage quantization of the raw data. The AD quantization formula is as follows:

[0079]

[0080] Where V a Indicates the 24-bit AD chip sampling value, V q Indicates the quantization value, 10 indicates the maximum sampling output range of the AD chip, and 24 indicates the number of quantization bits of the AD chip.

[0081] 4.2) Eliminate the influence of chip gain. The formula for the true quantization value after eliminating the influence of chip gain is as follows:

[0082] V q =V q / 128(2)

[0083] 4.3) Use a baseline calibration strategy to eliminate the impact of installation errors on dynamic measurement data. The baseline calibration formula is as follows:

[0084] V q =V q -V init (3)

[0085]

[0086] In formula (3), V init It is obtained by collecting the strain signal of the measured shaft in the static reference state, and is taken as 1000 based on experience. Formula (4) is based on the previously calculated torsional strain signal V q Remove DC offset to further improve AC signal quality, where V val It represents the effective value after removing the DC component. The strain voltage signal is converted into the strain quantity, and the formula is as follows:

[0087]

[0088] In formula (5), ε is the strain, ΔR is the resistance change, R is the original resistance value, the voltage mV represents the strain voltage at a certain moment, and K is the strain gauge sensitivity coefficient.

[0089] 4.4) A voltage-strain conversion model is constructed based on the mechanical properties of the material to convert the compensated voltage signal into a dimensionless strain variable.

[0090] 4.5) Use a Butterworth low-pass filter to reduce the noise of the decoupled strain signal. Since both the collected bending and torsion signals are subject to noise interference, the original strain signal ε needs to be filtered. A Butterworth low-pass filter is used, and its transfer function can be expressed as follows:

[0091]

[0092] In formula (7), f c represents the cutoff frequency of the Butterworth low-pass filter, f s represents the sampling frequency, w c represents the normalized cutoff frequency, where s represents the complex frequency variable and N represents the filter order.

[0093] 5) Speed ​​calculation. Calculate the speed of the measured shaft based on the strain data. The specific steps are as follows:

[0094] 5.1) A window function is selected based on the stationary characteristics of the bending and torsional strain signals, and the signals are framed based on the selected window function so that the signal in each window can be approximated as a stationary signal.

[0095] 5.2) Perform a short-time Fourier transform on the framed data to construct a two-dimensional matrix of time-frequency energy and form a time-frequency spectrum. Perform a short-time Fourier transform on the filtered bending and torsion signal. Suppose there is a discrete-time signal x[n], where n = 0, 1, 2, ..., N-1. Select a window function w[n] with a length of M and divide the signal x[n] into multiple overlapping short-time windows. The length of each window is M, and the overlap length between adjacent windows is R. For the kth window, the starting point of the signal is n = kL, where L = MR is the step size of the window. Therefore, applying the window function to each window yields the window signal X k [n]:

[0096] x k [n]=x[n+kL]·w[n](8)

[0097] In formula (8), n = 0, 1, 2, ..., M-1, then, for each window signal x[n], calculate its discrete Fourier transform:

[0098]

[0099] Where f = 0, 1, 2,…, M-1.

[0100] The Fourier transform result X of each window k [f], are combined into a two-dimensional matrix to form a time-frequency spectrum. The horizontal axis of the spectrum represents time, and the vertical axis represents frequency.

[0101] 5.3) After obtaining the time-frequency graph, the frequency ridges in the time-frequency graph can be extracted. The frequency ridges are extracted using a time-frequency ridge extraction method based on an energy threshold. First, the maximum value is calculated based on the time-frequency matrix, and a threshold is set based on the calculated maximum value. If the value in the time-frequency matrix is ​​less than the threshold, it is set to 0. Then, the maximum value in each column of the time-frequency matrix is ​​calculated to obtain the maximum value of each column of the time-frequency matrix, and the position index of the maximum value in each column is recorded. The corresponding row position index represents the frequency, and then the frequency is converted into a speed signal to obtain a speed signal, which is expressed as follows:

[0102] v=60·f(10)

[0103] In formula (10), f represents the rotational frequency, and v represents the rotational speed signal.

[0104] 5.4) Calculate the shaft speed based on the corresponding relationship between frequency and speed.

[0105] 6) Load calculation. The torque of the measured shaft is calculated based on the strain data. First, the strain gauge voltage data needs to be filtered and demodulated to obtain the strain signal. Then, the effective torque value is calculated based on the cross-sectional parameters of the rotating shaft. The torque calculation formula is:

[0106]

[0107] Where T is the measured shaft torque. τ is the shear stress. Zp is the anti-torque interface coefficient, as shown in Equation (9). τmax is the maximum shear stress. E is the longitudinal and transverse coefficient. μ is the Poisson coefficient.

[0108]

[0109] Where D is the outer diameter of the shaft being measured, and d is the inner diameter of the shaft being measured.

[0110] It is worth noting that this embodiment uses a method that integrates wireless signal transmission and strain measurement to collect strain signals. The collected strain signals are processed and extracted to obtain the torque and speed corresponding to the test piece. Wireless signal transmission technology enables the collection, transmission, transmission, and storage of strain signals, avoiding the signal transmission line winding issues caused by wired signal transmission during the test piece's rotation. Because a single strain acquisition node is limited by signal synchronization and transmission stability, a distributed wireless acquisition system architecture is adopted. This solution establishes dedicated acquisition channels for different deformation types (bending / torsion) on a single wireless acquisition node, thereby ensuring the data integrity and measurement accuracy of anisotropic strain signals. This measurement system adopts a distributed architecture design, and its wireless acquisition nodes are equipped with dual-mode data management capabilities. Each node is stably powered by a 5000mAh universal 18650 battery pack and establishes a wireless communication link with the host system via a Wi-Fi network. Within the local area network environment established by a wireless router, the strain signals collected by the node in real time are digitally modulated and synchronously transmitted to a remote monitoring terminal for visualization and can also be stored locally on an embedded SD card for redundant backup.

[0111] Example 2:

[0112] The main contents of this embodiment are the same as those of embodiment 1, wherein the installation box 201 is prepared by 3D printing in actual space. The installation box 201 is fixedly mounted on the outer wall of the coupling using anti-loosening threaded fasteners 203. During installation, based on the geometric characteristics and spatial constraints of the wireless strain acquisition node, 3D printing technology is used to prepare a box body with a topologically optimized hollow installation structure with suitable shape and size. First, a polymer adhesive is applied to achieve the initial positioning of the installation box and the axial surface. Then, a belt-type mechanical constraint is implemented through the pre-set anchor holes on the installation box. Finally, anti-loosening threaded fasteners are used to complete the rigid connection of the node module. At the same time, when fixing, ensure that the battery is facing upwards for easy replacement.

[0113] Example 3:

[0114] The main contents of this embodiment are the same as those of embodiment 1 or 2, wherein this embodiment verifies the effectiveness of the method with a specific example. The torque signal obtained by calculating the torsional strain signal is as follows: Figure 5 As shown in the figure, the torque obtained by calculating the wireless torsional strain signal is the same as the torque read by the control system. Figure 6 As shown in the figure, it can be seen that when the test loading torque is small, the error between the torque extracted by the wireless torsional strain signal and the torque read by the control system is large, and the error between their average values ​​is 66.92%; when the test loading torque is large, the error between the torque extracted by the wireless torsional strain signal and the torque read by the control system is small, and the error between their average values ​​is 3.26%.

[0115] During the test, the change pattern of the measured shaft speed is set to 100r / min—500r / min—800r / min—1200r / min—1000r / min—700r / min—100r / min—0r / min, and the change pattern of the motor input torque is 6.7N·m—10N·m—20N·m—30N·m—20N·m—10N·m—0N·m. The bending strain signal is as follows Figure 7 As shown in the figure, the short-time Fourier transform and refinement processing of the strain signal are as follows: Figure 8 As shown, the wireless strain signal time-frequency extraction speed signal and the control system read the speed signal as shown Figure 9As shown in the figure, there is a certain error between the speed information extracted from the wireless strain signal and the speed signal read by the control system at 100 r / min, but the trend changes remain consistent under other variable speed conditions. At a theoretical speed of 500 r / min, the average speed error measured by the wireless strain node is 2.47%; at a theoretical speed of 800 r / min, the average speed error is 4.0%; at a theoretical speed of 1200 r / min, the average speed error is 1.45%; at a theoretical speed of 1000 r / min, the average speed error is 2.42%; and at a theoretical speed of 700 r / min, the average speed error is 0.65%.

[0116] Analysis of the results from this example demonstrates that the proposed method for short-range wireless telemetry of gear transmission torque in confined spaces can measure the torque and speed of a gear transmission system even when torque / speed sensors are not available. When the test load torque is high, the error between the average torque measurement and the torque and speed signals read by the control system is 3.26%. Speed ​​measurement accuracy is also high, with maximum and minimum errors of 4.0% and 0.65%, respectively.

Claims

1. A method for wireless remote measurement of torque and speed of a gear transmission system in a confined space, characterized by: The output shaft of the driving motor is connected to the input shaft of the gear transmission system under test through a coupling; the input shaft of the gear transmission system under test is selected as the measured shaft; the wireless telemetry method specifically includes the following steps: 1) Four groups of strain units (1) are arranged at intervals along the circumferential direction in the middle section of the measured shaft; wherein each group of strain units (1) comprises a bending strain gauge and a torsional strain gauge arranged orthogonally to form a dual-parameter synchronous sensing array; the bending strain gauges are arranged along the normal direction, and the torsional strain gauges are arranged along the tangential direction; 2) A wireless strain acquisition node (2) is fixedly installed; wherein, each group of strain units (1) is configured with an independent wireless strain acquisition node (2), and a measurement data mutual verification mechanism is constructed through a multi-node spatial redundant layout; the wireless strain acquisition node (2) includes an installation box (201) and an acquisition terminal (202); the acquisition terminal is accommodated in the inner cavity of the installation box (201); the installation box (201) is fixedly connected to the outer wall of the coupling; each installation box (201) is correspondingly arranged at a position of a corresponding strain unit (1); the acquisition terminal (202) includes a full-bridge circuit, an ADS1220 ADC chip, an ESP32-S2 control chip, an SD card, a Wi-Fi radio frequency module and a battery pack power supply; the full-bridge circuit is used to sense bending deformation and torsional deformation strain signals; the ADS1220 The ADC chip is used to convert the strain analog signal into a digital signal; the ESP32-S2 control chip is used as a main controller to control the scheduling of acquisition, storage and wireless transmission functions; the SD card is used for data storage; the Wi-Fi radio frequency module is used for wireless transmission of data; the battery pack power supply supplies power to each hardware module; in view of the characteristics of the rotating working condition, the acquisition terminal (202) adopts a distributed dual-channel acquisition node architecture; the wireless acquisition node uses the dual-channel ADS1220ADC chip to respectively acquire the bending deformation and torsional deformation strain data of the measured shaft, stores them in the SD card, and wirelessly transmits them to the computer for storage and analysis; the local SD card is used to cache the original data, and the time division multiplexing protocol is used between nodes to avoid signal collision; 3) Carrying out data collection: when the gear transmission system is working, the strain unit (1) collects strain data; the wireless strain collection node (2) receives the strain data measured in real time from the corresponding strain unit (1) through the wireless network, and uploads the data to the computer (3) in real time; 4) Processing the strain data; Step 4) specifically includes the following sub-steps: 4.1) Perform voltage domain conversion on the 24-bit AD raw sampling signal acquired by the wireless strain acquisition node. By setting a full-scale voltage range of ±5V, a linear mapping relationship between the AD sampling value and the actual voltage signal is established to achieve voltage quantization of the raw data. 4.2) Eliminate the influence of chip gain; 4.3) Use baseline calibration strategy to eliminate the impact of installation errors on dynamic measurement data; 4.4) Construct a voltage-strain conversion model based on the mechanical properties of the material to convert the compensated voltage signal into a dimensionless strain variable; 4.5) Using a Butterworth low-pass filter to reduce noise on the decoupled strain signal; 5) Calculate the measured shaft torque based on the strain data; filter and demodulate the strain gauge voltage data to obtain the strain signal, and then calculate the effective torque value based on the shaft cross-sectional parameters; 6) Calculating the measured shaft speed based on the strain data; Step 6) specifically includes the following sub-steps: 6.1) Selecting a window function based on the stationary characteristics of the bending and torsional strain signals, and performing frame processing on the signals based on the selected window function so that the signal within each window can be approximated as a stationary signal; 6.2) Perform short-time Fourier transform on the framed data to construct a two-dimensional time-frequency energy matrix to form a time-frequency spectrum; 6.3) Extract the time-frequency ridges using the energy threshold-based time-frequency ridge extraction method; 6.4) Calculate the shaft speed based on the corresponding relationship between frequency and speed.

2. The method for wireless remote measurement of torque and speed of a gear transmission system in a confined space according to claim 1, characterized in that: The strain unit (1) comprises a base layer (101), a strain gauge (102), a dynamic compensation layer (103) and a protective layer (104) which are stacked in sequence; the base layer (101) is evenly coated with epoxy resin glue using a high-precision dispensing process to ensure reliable coupling between the strain gauge (102) and the substrate; the dynamic compensation layer (103) uses a temperature self-compensating strain gauge in the strain gauge and adds a thin film temperature sensor to achieve real-time monitoring of the temperature field; the protective layer (104) is stacked with a carbon fiber composite material protective cover and a local copper foil shielding tape to construct an electromagnetic shielding and physical protection composite structure.

3. The method for wireless remote measurement of torque and speed of a gear transmission system in a confined space according to claim 1, characterized in that: In step 4), the AD quantization formula is as follows: Where V a Indicates the 24-bit AD chip sampling value, V q Indicates the quantization value, 10 indicates the maximum sampling output range of the AD chip, and 24 indicates the number of quantization bits of the AD chip; The formula for the true quantization value after eliminating the influence of chip gain is as follows: V q =V q / 128(2) In step 4.3), the baseline calibration formula is as follows: V q =V q -V init (3) In formula (3), V init It is obtained by collecting the strain signal of the measured shaft in the static reference state, and is taken as 1000 according to experience; Formula (4) is based on the previously calculated torsional strain signal V q Remove DC offset to further improve AC signal quality, where V val It represents the effective value after removing the DC component; the strain voltage signal is converted into the strain, and the formula is as follows: Unit: μm / m(5) In formula (5), ε is the strain, ΔR is the resistance change, R is the original resistance value, the voltage mV represents the strain voltage at a certain moment, and K is the strain gauge sensitivity coefficient. Since the collected bending and torsion signals are both subject to noise interference, the original strain signal ε needs to be filtered. A Butterworth low-pass filter is used, and its transfer function can be expressed as follows: In formula (7), f c represents the cutoff frequency of the Butterworth low-pass filter, f s represents the sampling frequency, w c represents the normalized cutoff frequency, where s represents the complex frequency variable and N represents the filter order.

4. The method for wireless remote measurement of torque and speed of a gear transmission system in a confined space according to claim 1, characterized in that: In step 5), the torque calculation formula is: Where, T is the measured shaft torque; τ is the shear stress; Zp is the anti-torque interface coefficient, as shown in formula (9); τmax is the maximum shear stress; E is the longitudinal and transverse coefficient; μ is the Poisson coefficient; Where D is the outer diameter of the shaft being measured, and d is the inner diameter of the shaft being measured.

5. The method for wireless remote measurement of torque and speed of a gear transmission system in a confined space according to claim 1, characterized in that: In step 6), a short-time Fourier transform is performed on the filtered bending and torsion signal. Assume that there is a discrete-time signal x[n], where n = 0, 1, 2, ..., N-1; a window function w[n] is selected with a length of M, and the signal x[n] is divided into multiple overlapping short-time windows; the length of each window is M, and the overlapping length between adjacent windows is R. For the kth window, the starting point of the signal is n = kL, where L = MR is the step size of the window. Therefore, the window function is applied to each window to obtain the window signal X k [n]: x k [n]=x[n+kL]·w[n](10) In formula (10), n = 0, 1, 2, ..., M-1. Then, for each window signal x[n], its discrete Fourier transform is calculated: Where, f = 0, 1, 2, ..., M-1; The Fourier transform result X of each window k [f], combined into a two-dimensional matrix to form a time-frequency spectrum diagram; the horizontal axis of the spectrum diagram represents time, and the vertical axis represents frequency; after obtaining the time-frequency diagram, the frequency ridge line in the time-frequency diagram can be extracted.

6. The method for wireless remote measurement of torque and speed of a gear transmission system in a confined space according to claim 5, characterized in that: In step 6.3), first calculate the maximum value based on the time-frequency matrix, and set a threshold value based on the calculated maximum value. If the value in the time-frequency matrix is ​​less than the threshold value, it is set to 0. Then, calculate the maximum value in each column of the time-frequency matrix to obtain the maximum value of each column of the time-frequency matrix, and record the position index of the maximum value in each column. The corresponding row position index represents the rotational frequency, and then convert the rotational frequency into a rotational speed signal to obtain a rotational speed signal, which is expressed as follows: v=60·f(12) In formula (12), f represents the rotational frequency, and v represents the rotational speed signal.

7. The method for wireless remote measurement of torque and speed of a gear transmission system in a confined space according to claim 1, characterized in that: The installation box (201) is prepared by 3D printing according to actual space.

8. The method for wireless remote measurement of torque and speed of a gear transmission system in a confined space according to claim 1, characterized in that: The installation box (201) is fixedly installed on the outer wall of the coupling by using an anti-loosening threaded fastener (203).

Images