Steering gear transmission noise detection device and detection method

Abstract

The steering gear transmission noise detection device comprises a transmission assembly, an adjusting assembly, a noise collecting device and a controller, the transmission assembly comprises a motor, a synchronous belt, a driving wheel, a driven wheel, a transmission shaft and a shell, the output end of the motor is coaxially and fixedly connected with the driving wheel, and the driven wheel is coaxially and fixedly connected with the transmission shaft; the synchronous belt is wound on the peripheral walls of the driving wheel and the driven wheel, the transmission shaft is rotationally arranged in the shell, and the shell is fixedly installed on the bottom plate; the adjusting assembly comprises a driving part, an adjusting block and a tension sensor, the adjusting block is fixedly connected with the motor, the adjusting block is connected with the driving part, the tension sensor is used for detecting the tension value of the synchronous belt, and the controller is connected with the tension sensor, the driving part and the motor; the noise collecting device comprises a microphone connected with the controller, the controller processes audio signals of the microphone in real time, and abnormal sound characteristics are recognized through a built-in spectrum analysis unit.

Description

Technical Field

[0001] This invention relates to the field of steering gear noise detection technology, and more specifically to a steering gear transmission noise detection device and detection method. Background Technology

[0002] R-EPS steering systems rely on synchronous belts to transmit motor torque, and the tooth profile accuracy of the synchronous belt directly affects its NVH (noise, vibration, and harshness) performance. During the manufacturing process of synchronous belts, factors such as mold wear and vulcanization process fluctuations often lead to tooth profile deviations, such as uneven tooth pitch or tooth deformation. These defects can cause abnormal meshing during transmission, resulting in two typical types of abnormal noise problems:

[0003] Machine gun noise: It manifests as an intermittent "ticking" sound, mainly caused by periodic impacts during transmission due to uneven tooth pitch.

[0004] Scraping noise: This manifests as a continuous frictional whistling sound, mainly caused by the deformation of the tooth profile, which causes the synchronous belt to deviate to one side during operation, squeezing the pulley edge and generating friction.

[0005] Currently, the detection of such abnormal noises mainly relies on manual listening, which suffers from low efficiency, high subjectivity, poor consistency, and easy omissions. Therefore, there is an urgent need for an automated, high-precision, and quantifiable steering gear transmission noise detection device and method to improve product quality and production efficiency. Summary of the Invention

[0006] In view of the technical problems of low efficiency, strong subjectivity, poor consistency and easy omission caused by relying on manual listening inspection in the existing technology, the present invention proposes a steering gear transmission noise detection device and detection method.

[0007] The technical solution adopted by this invention is as follows: A steering gear transmission noise detection device includes a transmission assembly, an adjustment assembly, a noise acquisition device, and a controller. The transmission assembly includes a motor, a synchronous belt, a drive wheel, a driven wheel, a transmission shaft, and a housing. The output end of the motor is coaxially and fixedly connected to the drive wheel. The driven wheel is coaxially and fixedly connected to the transmission shaft. The synchronous belt is wound around the outer peripheral walls of the drive wheel and the driven wheel. The transmission shaft is rotatably disposed inside the housing, and the housing is fixedly mounted on a base plate. The adjustment assembly includes a drive component, an adjustment block, and a tension sensor. The adjustment block is fixedly connected to the motor and connected to the drive component. The tension sensor is used to detect the tension value of the synchronous belt. The controller is connected to the tension sensor, the drive component, and the motor respectively. The noise acquisition device includes a microphone connected to the controller. The controller processes the audio signal from the microphone in real time and identifies abnormal noise characteristics through a built-in spectrum analysis unit.

[0008] Optionally, the controller controls the drive component according to the formula Tt=(4a*T) / (Dd) based on the feedback from the tension sensor, so as to adjust the tension of the synchronous belt to the target tension value Tt, where: D is the driven pulley pitch diameter, d is the drive pulley pitch diameter, a is the center distance between the drive pulley and the driven pulley, and T is the preset reference tension value.

[0009] Optionally, the controller is connected to a CAN signal transmitting device, and sends stepped speed control commands to the motor through the CAN signal transmitting device.

[0010] This invention also discloses a method for detecting steering gear transmission noise, using the steering gear transmission noise detection device described above, comprising the following steps:

[0011] (1) Fix the housing to the base plate, and the controller adjusts the tension of the synchronous belt to the preset value Tt based on the feedback from the tension sensor;

[0012] (2) The controller controls the motor to run according to the stepped speed command, which includes multiple incremental speed steps;

[0013] (3) When the motor reaches the steady state stage of each speed step, the microphone collects the audio signal and sends it to the controller;

[0014] (4) The controller performs spectrum analysis on the received audio signal and identifies abnormal noise characteristics through the spectrum analysis unit: if periodic pulse peaks are identified in the 200~400Hz frequency band, it is determined that machine gun noise exists; if a continuous wideband energy band is identified in the 350~2200Hz frequency band, it is determined that scraping noise exists.

[0015] (5) The controller outputs the defect judgment result.

[0016] Optionally, the spectrum analysis unit dynamically separates noise and valid signals using an adaptive threshold algorithm. The steps of the adaptive threshold algorithm are as follows: dynamically calculate the noise power estimate, calculate the dynamic threshold T based on the estimate, and compare the spectral component Y(f) with the threshold T to filter valid spectral components; wherein, the filtering condition for valid spectral components is: Y(f) > T.

[0017] Optionally, the dynamic threshold T is calculated as follows:

[0018] T = λ(f) * P noise (f) + β(f),

[0019] Where λ(f) is the frequency domain sensitivity coefficient, β(f) is the offset, and P noise( f) is the noise power estimate at frequency point f.

[0020] The noise power P noise The real-time calculation formula for (f) is:

[0021] P noise (f) = min(P) current (t, f), α * P noise (f) + (1 - α) * P current (t, f)

[0022] Among them, P noise (f) represents the power of frequency point f at the current time t, and α is the forgetting factor.

[0023] Optionally, the method for identifying periodic pulse peaks in step (4) includes the following steps: calculating the frequency difference Δf between adjacent local maxima in the spectrum, and confirming the existence of periodicity when the difference between multiple consecutively occurring Δf is ≤5%.

[0024] Optionally, step (4) also includes a method for verifying the periodic pulse peak: performing cepstral analysis on the spectrum to obtain a cepstral graph C(τ). If the cepstral graph shows a significant peak at a time delay τ, and τ satisfies τ ≈ 1 / Δf, where Δf is the periodic frequency difference confirmed in step 7, then the periodicity is confirmed to exist.

[0025] Optionally, the stepped speed command in step (2) includes four speed jumps, and the duration of each speed jump after reaching a steady state is not less than 5 seconds.

[0026] Optionally, the criteria for determining the broadband energy band in step (4) are: within the target frequency band of 350~2200Hz, the power value of more than 60% of the frequency points is higher than the background noise power value of the corresponding frequency point by more than 10dB.

[0027] The beneficial effects of this invention are: (1) By integrating the transmission components, the actual synchronous belt drive working environment and load conditions of the R-EPS steering gear are accurately simulated. The motor provides power, the drive wheel, driven wheel, and synchronous belt constitute the core transmission chain, the transmission shaft transmits torque, and the housing provides the installation reference and affects the acoustic environment. This ensures that the testing environment is highly consistent with the actual operating conditions of the product, and the test results have authenticity and reference value. The tension sensor provides real-time feedback, and the drive component changes the position of the motor, i.e. the center distance a, through the adjustment block, which provides the necessary hardware support for subsequent precise control of the tension force. It overcomes the defect that manual control cannot accurately control and quantify the tension force. By automatically controlling the tension force adjustment, the motor's stepped speed operation, the noise signal acquisition and analysis, and the abnormal noise characteristic identification through the controller, the whole process is automated, which significantly improves the testing efficiency.

[0028] (2) A preset tension formula is used to ensure that the tension of the synchronous belt is consistent in each test; a spectrum analysis combined with an adaptive threshold algorithm and clear abnormal noise characteristics is used for quantitative judgment, eliminating the subjectivity of manual listening and ensuring the consistency and reliability of the test results. The identification of machine gun noise is combined with the periodic analysis of the difference between adjacent peak frequencies to ensure the accuracy of periodic judgment; a quantitative standard of broadband energy band proportion is used for scraping noise to improve the identification accuracy. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the steering gear transmission noise detection device proposed in an embodiment of the present invention;

[0030] Figure 2 This is a cross-sectional view of the transmission assembly proposed in an embodiment of the present invention;

[0031] Figure 3 This is a flowchart of the steering gear transmission noise detection method proposed in an embodiment of the present invention;

[0032] Figure 4 The spectrum analysis unit proposed in this embodiment of the invention identifies the normal spectrum in the 200~400Hz frequency band;

[0033] Figure 5 The periodic pulse peaks identified by the spectrum analysis unit in the 200-400Hz frequency band in the embodiments of the present invention;

[0034] Figure 6 The spectrum analysis unit proposed in this embodiment of the invention identifies the normal spectrum in the 350~2200Hz frequency band;

[0035] Figure 7 This refers to the continuous broadband energy band identified by the spectrum analysis unit proposed in this embodiment of the invention in the 350~2200Hz frequency band.

[0036] The labels in the attached figures are as follows: 1. Motor; 2. Synchronous belt; 3. Drive wheel; 4. Driven wheel; 5. Drive shaft; 6. Housing; 7. Base plate; 8. Power cord; 9. Adjusting block; 10. Power supply; 11. Microphone; 12. CAN signal transmitting device. Detailed Implementation

[0037] The present application will now be described in further detail with reference to the accompanying drawings and embodiments.

[0038] like Figure 1-7As shown, the steering gear transmission noise detection device of this embodiment mainly includes: a transmission assembly including a motor 1, a synchronous belt 2, a drive wheel 3, a driven wheel 4, a transmission shaft 5, and a housing 6. The output shaft of the motor 1 is coaxially and fixedly connected to the drive wheel 3. The motor 1 is connected to a power supply 10 via a power cable 8, and the driven wheel 4 is coaxially and fixedly connected to the transmission shaft 5. The synchronous belt 2 is fitted onto the outer peripheral walls of the drive wheel 3 and the driven wheel 4. The transmission shaft 5 is rotatably mounted inside the housing 6 via bearings. The housing 6 is fixedly mounted on a base plate 7 by bolts or other means.

[0039] The adjustment assembly includes a drive component (such as a cylinder or a servo motor-driven lead screw and nut mechanism), an adjustment block 9, and a tension sensor. The adjustment block 9 is fixedly connected to the housing of the motor 1 (non-rotating part). The push rod or slider of the drive component is connected to the adjustment block 9. By extending or moving the drive component, the adjustment block 9 and the motor 1 fixed thereon can be moved, thereby changing the center distance between the drive wheel 3 and the driven wheel 4, thus adjusting the tension of the synchronous belt 2. The tension sensor (such as an S-type tension sensor) is installed between the adjustment block 9 and the fixed base, or directly integrated into the tensioning mechanism, to detect the tension value of the synchronous belt 2 in real time and transmit the signal to the controller.

[0040] The noise acquisition device includes a microphone 11 (preferably a directional microphone). The microphone 11 is positioned near the engagement area of ​​the synchronous belt to acquire noise signals generated during transmission. The microphone 11 is connected to the controller via a signal line.

[0041] The controller is typically an industrial control computer or a high-performance PLC. The controller is connected to the tension sensor, the control terminals of the drive unit, the driver of motor 1, and microphone 11. The controller runs control software, whose core functional modules include:

[0042] The tension control module receives signals from the tension sensor and calculates the direction and distance the drive component needs to move based on the preset target tension Tt and the mechanical formula Tt = (4a * T) / (D - d) (where D is the 4-section diameter of the driven wheel, d is the 3-section diameter of the drive wheel, a is the current center distance, and T is the preset reference tension value, which needs to be determined according to the belt type, length, etc.). It then sends commands to control the drive component's action, thus achieving closed-loop control of the tension.

[0043] Motor control module: Sends stepped speed commands (such as 500 rpm, 1000 rpm, 1500 rpm, 2500 rpm) to the driver of motor 1 via a built-in or external CAN signal transmitting device 12, and ensures that the motor maintains a steady state for at least 5 seconds after each speed step reaches a steady state.

[0044] Signal processing and analysis module: Acquires and processes audio signals input from microphone 11 in real time. Built-in spectrum analysis unit: Performs FFT transformation on the audio signal in the steady-state phase to obtain the power spectrum. Adaptive threshold algorithm: This algorithm dynamically estimates the noise power P at each frequency point f. noise (f). Various methods can be used, for example:

[0045] Method 1 (based on minimum statistics and recursive average: using formula P) noise (f) = min(P) current (t, f), α* P noise (f) + (1 - α) * P current (t, f) continuously tracks the minimum power (i.e., noise baseline) at each frequency point, with a forgetting factor α typically ranging from 0.95 to 0.99. Then, according to T = λ(f) * P noise The dynamic threshold T is calculated using λ(f) + β(f). Here, λ(f) is the frequency domain sensitivity coefficient, ranging from 1.5 to 3.0, with a larger value taken for high-frequency bands (e.g., >1kHz) to better suppress broadband noise; β(f) is the offset, set according to the required signal-to-noise ratio (e.g., 3dB). Finally, the spectral component Y(f) is compared with the threshold T, and effective components where Y(f) > T are selected.

[0046] Method 2 (MCRA2 and Quantile Method): The MCRA2 (Minimum Controlled Recursive Averaging) algorithm is used to track the minimum spectral value, and combined with the quantile method (e.g., calculating the 10th percentile of the current frame's spectral power) as the noise power P. noise Estimate (f) and then calculate the threshold T.

[0047] Abnormal noise feature identification:

[0048] Machine gun noise identification: In the effective spectral components of the 200~400Hz frequency band, look for the existence of periodic pulse peaks.

[0049] Step 1: Find all local maxima (peak points) within this frequency band;

[0050] Step 2: Calculate the frequency difference Δf between adjacent peak points;

[0051] Step 3: Check multiple consecutive (e.g., ≥3) Δf values. If the maximum difference between them does not exceed 5% of the minimum Δf, it is preliminarily determined that there is periodicity, and its fundamental frequency is approximately Δf.

[0052] Verification steps: Perform cepstral analysis on the frequency spectrum of the 200~400Hz band (calculate C(τ) = |IFFT(log|FFT(S(f))|)|^2, where S(f) is the frequency spectrum of this band). If the cepstral C(τ) shows a significant peak at a time delay τ≈1 / Δf, the periodicity is further confirmed, and it is determined to be machine gun noise.

[0053] Scraping noise identification: In the 350–2200 Hz frequency band, check for the existence of a continuous broadband energy band. Step 1: Estimate the background noise power P at each frequency point f within this band using an adaptive threshold algorithm. noise (f);

[0054] Step 2: Calculate the measured power P at each frequency point f. current (f) and background noise power P noise (f) difference (dB);

[0055] Step 3: Statistical analysis was conducted within the 350~2200Hz frequency band, where the measured power was more than 10dB higher than the background noise (P0). current (f) >P noise The number N of frequency points (f) + 10dB) above ;

[0056] Step 4: Calculate the proportion Ratio = N above / N total , where N total This represents the total number of frequency points within this frequency band.

[0057] Step 5: If Ratio > 60%, it is determined that there is a continuous wideband energy band, that is, there is scraping noise.

[0058] The controller also has a built-in result judgment and output module: it integrates the identification results under various speed steps and outputs the final defect judgment conclusion, such as "qualified", "machine gun noise exists", "scraping noise exists", "compound abnormal noise exists", etc., which can be output through the display screen, indicator lights, IO signals or upper computer communication.

[0059] This embodiment also discloses a method for detecting steering gear transmission noise. The method for detecting steering gear transmission noise using the above-mentioned device includes the following steps:

[0060] (1) Securely fix the housing 6 of the steering gear to be tested onto the base plate 7. The controller starts the tension control process: read the real-time value of the tension sensor, calculate the required action command of the drive component according to the preset target tension Tt and the formula Tt = (4a * T) / (D - d), control the action of the drive component, and make the feedback value of the tension sensor stable within the range of Tt ± allowable error;

[0061] (2) The controller sends step speed commands to the motor 1 through the CAN signal transmitting device 12. For example, the target speed is set to 500 rpm, 1000 rpm, 1500 rpm, and 2500 rpm in sequence. The controller monitors the actual speed of the motor 1. When the actual speed reaches the target speed and runs stably for at least 5 seconds, it is considered to have entered the steady state stage of the speed step.

[0062] (3) When the motor enters the steady state phase of each speed step (maintained for ≥5s), the controller triggers the microphone 11 to collect the audio signal during the steady state phase (usually for several seconds) and transmits the collected audio signal to the controller.

[0063] (4) The controller processes and analyzes the audio signals acquired in each steady-state phase: performs FFT transformation on the audio signals to calculate the power spectral density; uses the aforementioned adaptive threshold algorithm to calculate the dynamic threshold T, and filters out the effective spectral components Y(f) with power higher than the threshold T; in the effective spectral components of the 200~400Hz frequency band, the method of periodic judgment of adjacent peaks Δf and cepstral verification is applied to identify whether there are periodic pulse peak characteristics. If such characteristics are identified, the speed point is marked as having a machine gun noise risk; in the effective spectral components of the 350~2200Hz frequency band, the broadband energy band judgment standard is applied to identify whether there are continuous broadband energy band characteristics. If such characteristics are identified, the speed point is marked as having a scraping noise risk.

[0064] (5) Based on the analysis results of all speed steps: If no machinegun noise or scraping noise characteristics are detected under all speed steps, the transmission noise of the tested steering gear is deemed to be qualified; if machinegun noise characteristics are detected under any speed step, a machinegun noise defect is determined to exist; if scraping noise characteristics are detected under any speed step, a scraping noise defect is determined to exist; if both characteristics are detected simultaneously, a combined abnormal noise defect is determined to exist. The controller outputs the final judgment result to the PLC through preset methods, such as screen display, report printing, or sending OK / NG signals.

[0065] It is understood that the specific embodiments described above are merely for explaining the relevant invention and not for limiting the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Multiple technical solutions in the same embodiment, as well as multiple technical solutions in different embodiments, can be arranged and combined to form new technical solutions that do not contradict or conflict with each other. Any equivalent structural transformations made based on the content of this specification and drawings, whether directly or indirectly applied to other related technical fields, are similarly included within the scope of protection of this invention.

Claims

1. A steering gear transmission noise detection device, characterized in that, Includes transmission components, adjustment components, noise acquisition devices, and controllers. The transmission assembly includes a motor, a synchronous belt, a drive wheel, a driven wheel, a transmission shaft, and a housing. The output end of the motor is coaxially and fixedly connected to the drive wheel. The driven wheel is coaxially and fixedly connected to the transmission shaft. The synchronous belt is wound around the outer peripheral walls of the drive wheel and the driven wheel. The transmission shaft is rotatably disposed inside the housing. The housing is fixedly mounted on a base plate. The adjustment assembly includes a drive unit, an adjustment block, and a tension sensor. The adjustment block is fixedly connected to the motor and connected to the drive unit. The tension sensor is used to detect the tension value of the synchronous belt. The controller is connected to the tension sensor, the drive unit, and the motor respectively. The noise acquisition device includes a microphone connected to a controller, which processes the microphone's audio signal in real time and identifies abnormal noise characteristics through a built-in spectrum analysis unit.

2. The steering gear transmission noise detection device according to claim 1, characterized in that, The controller, based on feedback from the tension sensor, controls the drive component according to the formula Tt=(4a*T) / (Dd) to adjust the tension of the synchronous belt to the target tension value Tt, where: D is the driven pulley pitch diameter, d is the drive pulley pitch diameter, a is the center distance between the drive pulley and the driven pulley, and T is the preset reference tension value.

3. The steering gear transmission noise detection device according to claim 1, characterized in that, The controller is connected to a CAN signal transmitter and sends stepped speed control commands to the motor through the CAN signal transmitter.

4. A method for detecting steering gear transmission noise, using the steering gear transmission noise detection device according to any one of claims 1-3, characterized in that, Includes the following steps: (1) Fix the housing to the base plate, and the controller adjusts the tension of the synchronous belt to the preset value Tt based on the feedback from the tension sensor; (2) The controller controls the motor to run according to the stepped speed command, which includes multiple incremental speed steps; (3) When the motor reaches the steady state stage of each speed step, the microphone collects the audio signal and sends it to the controller; (4) The controller performs spectrum analysis on the received audio signal and identifies abnormal noise characteristics through the spectrum analysis unit: if periodic pulse peaks are identified in the 200~400Hz frequency band, it is determined that there is machine gun noise; If a continuous wideband energy band is detected in the 350~2200Hz frequency range, it is determined that scraping noise exists; (5) The controller outputs the defect judgment result.

5. The steering gear transmission noise detection method according to claim 4, characterized in that, The spectrum analysis unit dynamically separates noise and valid signals using an adaptive threshold algorithm. The steps of the adaptive threshold algorithm are as follows: dynamically calculate the noise power estimate, calculate the dynamic threshold T based on the estimate, and compare the spectrum component Y(f) with the threshold T to filter valid spectrum components; wherein, the filtering condition for valid spectrum components is: Y(f) > T.

6. The steering gear transmission noise detection method according to claim 4, characterized in that, The dynamic threshold T is calculated as follows: T = λ(f) * P noise (f)+ β(f), Where λ(f) is the frequency domain sensitivity coefficient, β(f) is the offset, and P noise( f) is the noise power estimate at frequency point f. The noise power P noise The real-time calculation formula for (f) is: P noise (f) = min( P current (t, f), α * P noise (f) + (1 - α) * P current (t, f) ) Among them, P noise (f) represents the power of frequency point f at the current time t, and α is the forgetting factor.

7. The steering gear transmission noise detection method according to claim 4, characterized in that, The method for identifying periodic pulse peaks in step (4) includes the following steps: calculating the frequency difference Δf between adjacent local maxima in the spectrum; when the difference between multiple consecutively occurring Δf is ≤5%, periodicity is confirmed.

8. The steering gear transmission noise detection method according to claim 4, characterized in that, Step (4) also includes a method for verifying the periodic pulse peak: performing cepstral analysis on the spectrum to obtain a cepstral graph C(τ). If the cepstral graph shows a significant peak at a time delay τ, and τ satisfies τ ≈ 1 / Δf, where Δf is the periodic frequency difference confirmed in step 7, then the periodicity is confirmed to exist.

9. The steering gear transmission noise detection method according to claim 4, characterized in that, The stepped speed command in step (2) includes four speed jumps, and each speed jump is maintained for no less than 5 seconds after reaching a steady state.

10. The method for detecting steering gear transmission noise according to claim 4, characterized in that, The criteria for determining the broadband energy band in step (4) are: within the target frequency band of 350~2200Hz, the power value of more than 60% of the frequency points is higher than the background noise power value of the corresponding frequency point by more than 10dB.

Images