Headrest actuator defective product identification method and headrest actuator noise monitoring system

Abstract

The invention discloses a headrest actuator defective product identification method and a headrest actuator noise monitoring system. The headrest actuator defective product identification method comprises the following steps: S1, synchronously collecting a current signal, a shell vibration signal and a working noise signal when an electronic headrest actuator works; s2, processing the current signal to extract a motor commutation frequency and a fluctuation value thereof, processing the vibration signal to extract a vibration frequency peak value, and processing the noise signal to calculate an A-weighted sound pressure level; s3, executing multi-dimensional intelligent judgment based on the feature parameters extracted in the step S2; the judgment comprises the following steps: if the A weighted sound pressure level is greater than a preset noise threshold value, judging that the product is a defective product; if the A weighting sound pressure level is not greater than a preset noise threshold value, but the commutation frequency fluctuation value is greater than a preset fluctuation threshold value, determining that the product is a potential defective product; and if the A-weighted sound pressure level is not greater than a preset noise threshold value, the commutation frequency fluctuation value is not greater than a preset fluctuation threshold value, and the vibration frequency peak value is greater than a preset peak threshold value, determining that the product is a potential defective product. Interference factors are eliminated, the logic of noise value priority and associated parameter assistance is adopted in the judgment stage, the core noise requirement is met, and potential mechanical faults can be recognized.

Description

Technical Field

[0001] This invention relates to the field of headrest actuator technology, and in particular to a method for identifying defective headrest actuators and a noise monitoring system for headrest actuators. Background Technology

[0002] As a core adjustment component of modern car seats, the noise level of electronic headrest actuators during operation is a key indicator affecting driving and riding comfort. Currently, mainstream automakers have set stringent technical requirements for these actuators, stipulating that their operating noise level should not exceed 35 decibels (dB). However, in mass production, the efficient and accurate testing of actuator noise quality remains a significant technical challenge.

[0003] Traditional electronic headrest actuator noise detection relies on subjective human judgment, which is inaccurate and inefficient.

[0004] Therefore, a method for identifying defective headrest actuators and a noise monitoring system for headrest actuators are still needed to solve the above problems. Summary of the Invention

[0005] This invention provides a method for identifying defective headrest actuators and a noise monitoring system for headrest actuators to solve the above-mentioned problems.

[0006] The objective of this invention is achieved through the following technical solution: A method for identifying defective headrest actuators includes: S1: Synchronously acquire the current signal, housing vibration signal and working noise signal of the electronic headrest actuator during operation; S2: Process the current signal to extract the motor commutation frequency and current fluctuation value, process the housing vibration signal to extract the vibration frequency peak value, and process the working noise signal to calculate the A-weighted sound pressure level; S3: Based on the feature parameters extracted in step S2, perform multi-dimensional intelligent judgment; the judgment includes: If the A-weighted sound pressure level is greater than the preset noise threshold, it is determined to be a defective product; If the A-weighted sound pressure level is less than or equal to a preset noise threshold, but the current fluctuation value is greater than a preset fluctuation threshold, it is determined to be a potential defective product. If the A-weighted sound pressure level is not greater than a preset noise threshold and the current fluctuation value is not greater than a preset fluctuation threshold, but the peak value of the vibration frequency is greater than a preset peak value threshold, then it is determined to be a potential defective product.

[0007] Preferably, after step S3, step S4 is further included: repeating the test on the actuators determined to be potentially defective, and making a final determination based on the results of the repeat test.

[0008] Preferably, between steps S1 and S2, step S1a is further included: verifying the correlation between the current signal, housing vibration signal, and working noise signal of the electronic headrest actuator during operation; Step S1a includes: Calculate the first correlation coefficient r1 between the commutation frequency and the current fluctuation value, and the second correlation coefficient r2 between the current fluctuation value and the noise frequency; When r1 ≥ 0.8 and r2 ≥ 0.85, the signal is deemed valid and step S2 is executed; otherwise, the signal is reacquired.

[0009] Preferably, in step S1, the synchronous acquisition of the current signal, housing vibration signal and working noise signal of the electronic headrest actuator is performed during the process of the actuator completing one extension to retraction working cycle.

[0010] Preferably, in step S2, the vibration signal and noise signal are converted into a frequency domain spectrum by fast Fourier transform, and the vibration frequency peak and noise frequency features are extracted from it.

[0011] A second aspect of the present invention discloses a headrest actuator noise monitoring system, comprising: The multi-parameter acquisition module is used to synchronously acquire the current signal, vibration signal and noise signal when the actuator is working; The data processing and judgment module is connected to the multi-parameter acquisition module and is used to receive and process the current signal, vibration signal and noise signal.

[0012] Preferably, the multi-parameter acquisition module includes: A current acquisition unit is connected to the DC motor power supply circuit of the actuator via a differential probe to acquire the motor current waveform; A vibration acquisition unit, which is fixed to the actuator housing, is used to acquire vibration signals of the housing; A noise acquisition unit is positioned towards the actuator and is used to acquire working noise signals.

[0013] Preferably, the data processing and intelligent judgment module includes: an industrial control host, and communication interfaces respectively connected to the current acquisition unit, vibration acquisition unit and noise acquisition unit.

[0014] Preferably, it further includes a testing device, the testing device comprising: A fixing fixture is used to position the actuator; A soundproof cover is installed on the outside of the fixing clamp; The power supply unit is used to supply power to the actuator under test.

[0015] Compared with the prior art, the beneficial effects of the present invention include at least the following: The system synchronously collects multi-dimensional parameters of the actuator, including current, vibration, and noise, and establishes a correlation analysis model between these parameters to achieve accurate noise measurement and rapid identification of defective products. Through multi-dimensional data correlation analysis, it achieves accurate noise value assessment and defective product identification. In the data processing stage, it extracts feature parameters using professional algorithms to eliminate interference factors. In the judgment stage, it adopts a logic of prioritizing noise value and assisting with correlation parameters, which not only meets the core noise requirements but also identifies potential mechanical faults. Finally, it achieves full-process traceability through data storage and report generation. Attached Figure Description

[0016] Figure 1 This is a block diagram of a defective product identification method according to an embodiment of the present invention; Figure 2 This is a diagram of an electronic headrest actuator noise monitoring system according to an embodiment of the present invention; Figure 3 This is a multi-parameter feature map of a typical qualified actuator according to an embodiment of the present invention; Figure 4 The invention embodiment displays the correlation distribution of current commutation frequency, vibration frequency, and noise frequency in the form of a scatter plot, and marks qualified areas and defective areas. Detailed Implementation

[0017] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided to make the invention more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore repeated descriptions of them will be omitted.

[0018] The terms used to express position and direction in this invention are illustrated with reference to the accompanying drawings, but changes can be made as needed, and all such changes are included within the scope of protection of this invention.

[0019] This invention provides a method for identifying defective headrest actuators, comprising: S1: Synchronously acquire the current signal, housing vibration signal, and operating noise signal of the electronic headrest actuator during operation. The control system sends a command to start the actuator and simultaneously sends a trigger signal. The three sensors then begin to operate: the oscilloscope records the motor current waveform at a sampling rate of no less than 500 kS / s; the vibration sensor acquires the housing vibration acceleration signal in the frequency range of 50 Hz to 10 kHz; and the microphone records the sound pressure signal of the operating noise at a sampling frequency of 44.1 kHz. The three signals are synchronized through a synchronous triggering mechanism to ensure their timestamp alignment, with a time synchronization error of less than 1 millisecond, and are then transmitted to the data processing module in real time.

[0020] S2: Process the current signal to extract the motor commutation frequency and current fluctuation value, process the housing vibration signal to extract the vibration frequency peak, and process the operating noise signal to calculate the A-weighted sound pressure level. First, the raw current waveform acquired by the oscilloscope is digitally filtered, mainly using a 50Hz notch filter to eliminate power frequency interference, and a low-pass filter to smooth high-frequency noise. Since the DC brushed motor will generate characteristic current spikes or dips at the moment of commutation, the algorithm accurately locates the start time (T1, T2, …, Tn) of each commutation event by detecting the zero-crossing point of the first derivative (difference) of the current waveform or setting a threshold. Commutation period: Calculate the time interval between adjacent commutation points, such as C_i = T_{i+1} - T_i. Instantaneous commutation frequency: Calculated as the reciprocal of the period, i.e., F_i = 1 / C_i. Average commutation frequency (F_avg): Calculate the arithmetic mean of all instantaneous frequencies within a complete working cycle. Commutation frequency fluctuation (ΔF): A key indicator for measuring stability, it is usually calculated as the standard deviation of the instantaneous frequency relative to the average frequency, or directly as the difference between the maximum and minimum values ​​(peak-to-peak value). Its calculation formula can be expressed as: ΔF = max(F_i) - min(F_i) or ΔF = sqrt(Σ (F_i - F_avg)² / (n-1)).

[0021] Vibration Signal Processing and Peak Frequency Extraction: Vibration signal processing aims to identify the dominant frequency components of shell vibration, whose intensity is typically correlated with mechanical anomalies. Preprocessing: The vibration acceleration signal is de-trended and bandpass filtered (e.g., 50Hz - 10kHz) to eliminate DC offset and interference from irrelevant frequency bands. Frequency Domain Transformation: A Fast Fourier Transform (FFT) is applied to the preprocessed time-domain signal to convert it into a power spectral density (PSD) or amplitude spectrum in the frequency domain. Peak Extraction: Within the effective analysis frequency band (e.g., 100Hz - 5kHz), the frequency point corresponding to the global maximum value in the spectrum is searched. This frequency value is the vibration frequency peak (F_vib), which may correspond to the gear meshing frequency, bearing fault characteristic frequency, or its harmonics.

[0022] The goal of noise processing is to simulate the characteristics of human hearing and obtain an objective sound pressure level consistent with subjective perception. An A-weighted frequency response filter is applied to the original sound pressure time-domain signal acquired by the microphone in the digital domain. The transfer function of this filter is defined by international standards (such as IEC 61672-1), and it exhibits significant attenuation in the low-frequency range (e.g., below 100Hz) and high-frequency range (e.g., above 10kHz) to reflect the differences in human ear sensitivity to different frequencies.

[0023] S3: Based on the feature parameters extracted in step S2, perform multi-dimensional intelligent judgment; the judgment includes: If the A-weighted sound pressure level is greater than the preset noise threshold, it is determined to be a defective product.

[0024] If the A-weighted sound pressure level is not greater than a preset noise threshold, but the current fluctuation value is greater than a preset fluctuation threshold, it is determined to be a potentially defective product.

[0025] If the A-weighted sound pressure level is not greater than a preset noise threshold and the current fluctuation value is not greater than a preset fluctuation threshold, but the peak value of the vibration frequency is greater than a preset peak value threshold, then it is determined to be a potential defective product.

[0026] In step S3, the noise sound pressure level calculation uses an A-weighted network, which conforms to the GB / T 3241-2010 standard "Electroacoustic octave and fractional octave filters" to eliminate the interference of low-frequency noise on the measurement results.

[0027] The first level of acoustic evaluation directly determines whether there is an abnormality. This is typically set at 35 dB(A) according to product specifications. This is the ultimate user experience threshold. If the extracted A-weighted average sound pressure level exceeds 35 dB, the product is directly deemed defective. This is the most direct and final quality control point. Any noise exceeding the limit is considered non-compliant with product specifications and directly triggers a non-conformance conclusion.

[0028] The second-level judgment, if the extracted A-weighted average sound pressure level is less than 35dB, indicates that a normal DC motor should have a highly stable commutation frequency under constant load. If the commutation frequency fluctuation exceeds a preset value, and the current commutation frequency fluctuation threshold is ≤5Hz, then although the final noise level does not exceed the standard, there are clear signs of performance degradation. This judgment provides early warning for early faults or latent defects, enabling the identification of problems that cannot be traced by a single noise detection method. It points to the stability of the electrical system or mechanical load, thus determining and predicting the future noise sources of the actuator.

[0029] The third level of judgment determines whether the A-weighted sound pressure level is not greater than a preset noise threshold (within the same range as above), and whether the commutation frequency fluctuation is not greater than a preset fluctuation threshold (less than 5Hz), but the peak vibration frequency is greater than a preset peak threshold (greater than 2kHz). In this case, the fault may not have caused the total noise level to exceed the limit in the early stages due to low energy, but its characteristic frequency signal is already very significant. This judgment can specifically locate high-frequency mechanical faults, further refining the fault tracing capability. Accurate noise value assessment and defective product identification are achieved through multi-dimensional data correlation analysis. In the data processing stage, feature parameters are extracted using professional algorithms to eliminate interference factors. The judgment stage adopts a logic of noise value priority + correlation parameter assistance, which satisfies the core noise requirements while identifying potential mechanical faults.

[0030] Preferably, after step S3, step S4 is further included: repeating the test on the actuators determined to be potentially defective, and making a final determination based on the results of the repeat tests. This eliminates interference from accidental factors, confirms the stability and reproducibility of defects, thereby minimizing the risk of misjudgment while ensuring detection efficiency, and further enriching the fault diagnosis information.

[0031] Preferably, step S1a is further included between steps S1 and S2: verifying the correlation between the current signal, vibration signal and noise signal; Step S1a includes: Calculate the first correlation coefficient r1 between the commutation frequency and the vibration frequency, and the second correlation coefficient r2 between the vibration frequency and the noise frequency; When r1 ≥ 0.8 and r2 ≥ 0.85, the signal is deemed valid, and step S2 is executed; otherwise, signal re-acquisition is triggered. The Pearson correlation coefficient algorithm is used to calculate the correlation coefficient r1 between the current commutation frequency and the vibration frequency, and the correlation coefficient r2 between the vibration frequency and the noise frequency. When r1 ≥ 0.8 and r2 ≥ 0.85, a strong correlation is determined among the three signals, and the data is valid; otherwise, the signal re-acquisition process is triggered. This approach not only improves the reliability of the solution at the algorithm level but also significantly reduces the risk of false alarms or downtime caused by occasional sensor failures or human error during installation at the engineering level, demonstrating the high level of intelligence and practicality of the solution.

[0032] In step S1, the synchronous acquisition is performed during the actuator's completion of one full extension-retraction work cycle. It begins by moving at a set constant speed (e.g., 5 mm / s) to the maximum stroke position (fully extended), and then returns to the initial position at the same or a set speed. For example, for an actuator with a stroke of 100 mm, the total stroke of this complete cycle is 200 mm, taking approximately 40 seconds. The working speed, stroke, and cycle definition can be set during system initialization according to specific product specifications, and represent the entire process of the actuator completing one extension and retraction work cycle.

[0033] In one embodiment, in step S2, the vibration signal and noise signal are converted into a frequency domain spectrum using a Fast Fourier Transform (FFT), and the vibration frequency peak and noise frequency features are extracted from it. Before performing the FFT, the original signal needs to be preprocessed to ensure spectral quality. The DC component in the signal is removed to prevent it from generating false 0Hz peaks in the spectrum. While the amplitude change of a time-domain signal can reflect the overall level, it is difficult to accurately isolate and locate the mixed frequency components from different sources. Frequency domain analysis, on the other hand, can decompose complex waveforms into a series of single-frequency sine waves, intuitively showing the distribution of signal energy at different frequencies. Through the FFT processing and feature extraction process, the present invention can transform the original, mixed vibration and noise signals into a series of digital features with clear physical meaning and diagnostic value, laying a precise data foundation for subsequent intelligent judgment.

[0034] A second aspect of the present invention discloses a headrest actuator noise monitoring system, comprising: The multi-parameter acquisition module is used to synchronously acquire the current signal, vibration signal and noise signal when the actuator is working.

[0035] The data processing and judgment module, connected to the multi-parameter acquisition module, is used to receive and process the current signal, vibration signal, and noise signal. The multi-parameter acquisition module ensures that the data are acquired at the same time reference, guaranteeing a strict causal correspondence between vibration and noise events and motor current events. The data processing and judgment module then utilizes this synchronized data to execute an automated analysis chain from the raw signal to the quality conclusion.

[0036] Furthermore, the multi-parameter acquisition module includes: The current acquisition unit is connected to the DC motor power supply circuit of the actuator via a differential probe to acquire the motor current waveform. This unit preferably uses a high-bandwidth digital oscilloscope as the recording device, in conjunction with a high-precision differential probe. The differential probe effectively suppresses high-frequency common-mode noise (such as switching noise from PWM drives) in the motor drive circuit, accurately extracts weak differential-mode current signals, and ensures waveform purity.

[0037] The vibration acquisition unit is fixed to the actuator housing and is used to acquire the vibration signal of the housing; the piezoelectric sensor has an extremely wide frequency response range (up to 50Hz to 10kHz or higher), which can fully cover the vibrations generated from low-frequency mechanical resonance to high-frequency gear meshing or impact.

[0038] A noise acquisition unit is positioned towards the actuator and is used to acquire working noise signals.

[0039] Furthermore, the data processing and intelligent judgment module includes: an industrial control host and communication interfaces respectively connected to the current acquisition unit, vibration acquisition unit, and noise acquisition unit. The industrial control host is a ruggedized industrial computer with a multi-core processor, at least 8GB of RAM, and a solid-state drive. It runs dedicated data fusion and analysis software developed based on a real-time operating system or a general-purpose operating system (such as Windows / Linux). This software is configured to execute all algorithmic steps in the claims of this method, including signal preprocessing, FFT transformation, correlation coefficient calculation, feature comparison, and triple judgment logic.

[0040] Preferably, it further includes a testing device, the testing device comprising: A fixing fixture is used to position the actuator. A soundproof enclosure is installed outside the fixing fixture; a power supply unit is used to supply power to the actuator under test. During operation, the operator places the actuator into the fixing fixture, closes the soundproof enclosure, and then starts the test via control software. The power supply unit is powered on under control, driving the actuator to move. At this time, under the perception of the multi-parameter acquisition module, the actuator is in a fixed position, in a quiet background, and with ideal power supply. The testing device is the fundamental guarantee for achieving the high-precision, high-repeatability testing target of this invention.

[0041] In this example, the test conditions were: test actuator model HZ-2024, 12V power supply, and sampling rate of 44.1kHz. The results are as follows: Figure 3 Typical qualified performance multi-parameter characteristic spectrum, including current waveform (circular frequency), vibration frequency domain spectrum (peak value marked), noise frequency domain spectrum and pressure level value.

[0042] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the invention without departing from the principles and spirit of the invention, and all such changes should fall within the protection scope of the claims of the present invention.

Claims

1. A method for identifying defective headrest actuators, characterized in that, include: S1: Synchronously acquire the current signal, housing vibration signal and working noise signal of the electronic headrest actuator during operation; S2: Process the current signal to extract the motor commutation frequency and current fluctuation value, process the housing vibration signal to extract the vibration frequency peak value, and process the working noise signal to calculate the A-weighted sound pressure level; S3: Based on the feature parameters extracted in step S2, perform multi-dimensional intelligent judgment; the judgment includes: If the A-weighted sound pressure level is greater than the preset noise threshold, it is determined to be a defective product; If the A-weighted sound pressure level is less than or equal to a preset noise threshold, but the current fluctuation value is greater than a preset fluctuation threshold, it is determined to be a potential defective product. If the A-weighted sound pressure level is less than or equal to a preset noise threshold and the current fluctuation value is not greater than a preset fluctuation threshold, but the peak value of the vibration frequency is greater than a preset peak value threshold, then it is determined to be a potential defective product.

2. The method for identifying defective headrest actuators according to claim 1, characterized in that, The process includes step S4 after step S3: repeating the test on the actuators that are determined to be potentially defective, and making a final determination based on the results of the repeat test.

3. The method for identifying defective headrest actuators according to claim 1 or 2, characterized in that, Between steps S1 and S2, there is also step S1a: verifying the correlation between the current signal, housing vibration signal and working noise signal when the electronic headrest actuator is working; Step S1a includes: Calculate the first correlation coefficient r1 between the commutation frequency and the current fluctuation value, and the second correlation coefficient r2 between the current fluctuation value and the noise frequency; When r1 ≥ 0.8 and r2 ≥ 0.85, the signal is deemed valid and step S2 is executed; otherwise, the signal is reacquired.

4. The method for identifying defective headrest actuators according to claim 1, characterized in that, In step S1, the synchronous acquisition of the current signal, housing vibration signal and working noise signal of the electronic headrest actuator is performed during the process of the actuator completing one extension and retraction work cycle.

5. The method for identifying defective headrest actuators according to claim 1, characterized in that, In step S2, the vibration signal and noise signal are converted into a frequency domain spectrum by fast Fourier transform, and the vibration frequency peak and noise frequency features are extracted from it.

6. A headrest actuator noise monitoring system, used to execute the headrest actuator defective product identification method according to claim 1, characterized in that, include: The multi-parameter acquisition module is used to synchronously acquire the current signal, vibration signal and noise signal when the actuator is working; The data processing and judgment module is connected to the multi-parameter acquisition module and is used to receive and process the current signal, vibration signal and noise signal.

7. The headrest actuator noise monitoring system according to claim 6, characterized in that, The multi-parameter acquisition module includes: A current acquisition unit is connected to the DC motor power supply circuit of the actuator via a differential probe to acquire the motor current waveform; A vibration acquisition unit, which is fixed to the actuator housing, is used to acquire vibration signals of the housing; A noise acquisition unit is positioned towards the actuator and is used to acquire working noise signals.

8. The headrest actuator noise monitoring system according to claim 7, characterized in that, The data processing and intelligent judgment module includes an industrial control host and communication interfaces connected to the current acquisition unit, vibration acquisition unit and noise acquisition unit respectively.

9. The headrest actuator noise monitoring system according to claim 7, characterized in that, It also includes a testing device, which comprises: A fixing fixture is used to position the actuator; A soundproof cover is installed on the outside of the fixing clamp; The power supply unit is used to supply power to the actuator under test.

Images