Motor vibration sound detection method and device and storage medium

Abstract

The invention relates to a motor vibration sound detection method and device and a storage medium. The motor vibration sound detection method comprises the following steps: acquiring an audio signal of motor vibration, and acquiring a frequency domain feature of the audio signal; harmonic characteristics, frequency stability and slope stability of the audio signals are determined according to the frequency domain characteristics, the harmonic characteristics are used for judging harmonic components of the audio signals, and the frequency stability and the slope stability are used for evaluating the frequency change trend of the audio signals; and detecting vibration noise generated in the vibration process of the motor according to at least one of the harmonic characteristics, the frequency stability and the slope stability. According to the invention, detection of motor vibration noise can be realized.

Description

Technical Field

[0001] The present disclosure relates to the field of audio processing, and in particular to a motor vibration sound detection method, device and storage medium. Background Art

[0002] In many industrial and mechanical applications, motors are key working parts, and their vibration status and performance are critical to the normal operation of the machine. However, there may be abnormal conditions such as vibration noise in the motor vibration, which may indicate motor failure or problems.

[0003] In the related art, the vibration state of the motor is generally detected by collecting vibration signals through special equipment such as vibration sensors, and then analyzing the vibration intensity and frequency of the collected vibration signals, but the noise in the motor vibration cannot be analyzed.

[0004] In addition, motor noise detection is currently basically determined by human listening, which is too subjective, and there is no solution to objectively detect motor vibration noise. Summary of the invention

[0005] In order to overcome the problems existing in the related art, the present disclosure provides a motor vibration sound detection method, device and storage medium.

[0006] According to a first aspect of an embodiment of the present disclosure, a motor vibration sound detection method is provided, comprising:

[0007] Acquire an audio signal of motor vibration, and acquire frequency domain characteristics of the audio signal; determine the harmonic characteristics, frequency stability, and slope stability of the audio signal according to the frequency domain characteristics, wherein the harmonic characteristics are used to determine the harmonic components of the audio signal, and the frequency stability and the slope stability are used to evaluate the frequency change trend of the audio signal; and detect vibration noise generated during the motor vibration process according to at least one of the harmonic characteristics, the frequency stability, and the slope stability.

[0008] In one embodiment, determining the harmonic features of the audio signal based on the frequency domain features includes: normalizing the frequency domain features to obtain frequency domain quantization values; performing a differential operation on the frequency domain quantization values ​​to obtain a differential feature sequence, wherein the differential feature sequence includes error-free frequency domain quantization values; and extracting the harmonic features in the differential feature sequence.

[0009] In one embodiment, the normalization processing of the frequency domain features to obtain frequency domain quantization values ​​includes: intercepting a frequency band within the effective frequency range of the motor vibration signal of the frequency domain features; performing normalization processing based on the frequency average value of the spectrum within the frequency band to obtain a frequency domain quantization value, and the frequency domain quantization value has a one-to-one correspondence with the spectrum energy corresponding to the frequency.

[0010] In one embodiment, the extracting of harmonic features in the differential feature sequence includes: determining a first confidence interval according to a first confidence level, and taking the maximum value of the first confidence interval as a broad-spectrum reference value of broad-spectrum noise, wherein the broad-spectrum noise is steady-state noise in the audio signal that is independent of the harmonic features; determining the difference between each value in the differential feature sequence and the broad-spectrum reference to obtain multiple differences; and taking the sum of the multiple differences as the harmonic feature value.

[0011] In one implementation, determining the frequency stability of the audio signal based on the frequency domain features includes: converting the frequency domain features into a time domain spectrum; normalizing the time domain spectrum to obtain a time domain spectrum quantization value; determining the variance of the time domain spectrum quantization value, and using the variance as the frequency stability.

[0012] In one embodiment, determining the slope stability of the audio signal based on the frequency domain features includes: converting the frequency domain features into a time domain spectrum; normalizing the time domain spectrum to obtain a time domain spectrum quantization value; performing a differential operation on the time domain spectrum quantization value to obtain a differential feature sequence; determining the variance of the differential feature sequence, and using the variance as the slope stability.

[0013] In one embodiment, the vibration noise generated during the vibration of the motor is detected according to at least one of the harmonic characteristic, the frequency stability and the slope stability, including: setting thresholds of the harmonic characteristic value, the frequency stability and the slope stability according to the harmonic characteristic value, the frequency stability and the slope stability, respectively; if the harmonic characteristic value is less than the threshold or at least one of the frequency stability and the slope stability is greater than the threshold, it is determined that the vibration noise exists.

[0014] According to a second aspect of an embodiment of the present disclosure, there is provided a motor vibration sound detection device, comprising:

[0015] An acquisition unit is used to acquire an audio signal of motor vibration and acquire frequency domain characteristics of the audio signal; a processing unit is used to determine the harmonic characteristics, frequency stability and slope stability of the audio signal according to the frequency domain characteristics, the harmonic characteristics are used to determine the harmonic components of the audio signal, and the frequency stability and the slope stability are used to evaluate the frequency change trend of the audio signal; a detection unit is used to detect vibration noise generated during the vibration of the motor according to at least one of the harmonic characteristics, the frequency stability and the slope stability.

[0016] In one embodiment, the processing unit determines the harmonic features of the audio signal based on the frequency domain features in the following manner: normalizing the frequency domain features to obtain frequency domain quantization values; performing a differential operation on the frequency domain quantization values ​​to obtain a differential feature sequence, wherein the differential feature sequence includes error-free frequency domain quantization values; and extracting the harmonic features in the differential feature sequence.

[0017] In one embodiment, the processing unit normalizes the frequency domain features to obtain frequency domain quantization values ​​in the following manner: intercepting a frequency band within the effective frequency range of the motor vibration signal of the frequency domain features; performing normalization based on the frequency average value of the spectrum within the frequency band to obtain a frequency domain quantization value, and the frequency domain quantization value has a one-to-one correspondence with the spectrum energy corresponding to the frequency.

[0018] In one embodiment, the processing unit extracts the harmonic features in the differential feature sequence in the following manner: determining a first confidence interval according to a first confidence level, and taking the maximum value of the first confidence interval as a broad-spectrum reference value of broad-spectrum noise, wherein the broad-spectrum noise is steady-state noise in the audio signal that is independent of the harmonic features; determining the difference between each numerical value in the differential feature sequence and the broad-spectrum reference to obtain multiple differences; and taking the sum of the multiple differences as the harmonic feature value.

[0019] In one embodiment, the processing unit determines the frequency stability of the audio signal according to the frequency domain features in the following manner: converting the frequency domain features into a time domain spectrum; normalizing the time domain spectrum to obtain a time domain spectrum quantization value; determining the variance of the time domain spectrum quantization value, and using the variance as the frequency stability.

[0020] In one embodiment, the processing unit determines the slope stability of the audio signal according to the frequency domain features in the following manner: converting the frequency domain features into a time domain spectrum; normalizing the time domain spectrum to obtain a time domain spectrum quantization value; performing a differential operation on the time domain spectrum quantization value to obtain a differential feature sequence; determining the variance of the differential feature sequence, and using the variance as the slope stability.

[0021] In one embodiment, the detection unit detects the vibration noise generated during the vibration process of the motor according to at least one of the harmonic characteristic, the frequency stability and the slope stability in the following manner: thresholds of the harmonic characteristic value, the frequency stability and the slope stability are set respectively according to the harmonic characteristic value, the frequency stability and the slope stability; if the harmonic characteristic value is less than the threshold or at least one of the frequency stability and the slope stability is greater than the threshold, it is determined that the vibration noise exists.

[0022] According to a third aspect of an embodiment of the present disclosure, there is provided a motor vibration sound detection device, comprising:

[0023] A processor; a memory for storing processor executable instructions; wherein the processor is configured to: execute the motor vibration and sound detection method described in the first aspect or any one of the embodiments of the first aspect.

[0024] According to a fourth aspect of an embodiment of the present disclosure, a storage medium is provided, in which instructions are stored. When the instructions in the storage medium are executed by a processor of a terminal, the terminal is enabled to execute the motor vibration and sound detection method described in the first aspect or any one of the embodiments of the first aspect.

[0025] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: by collecting the sound signal of the motor vibration and applying the signal processing and analysis technology, it is not only possible to measure the motor vibration intensity, frequency and other indicators, but also possible to convert the subjective vibration noise into objective calculation and analysis. Through the method in the present disclosure, an objective and accurate evaluation of the motor vibration state is achieved, the influence of subjective factors is reduced, and the accuracy of judgment is improved.

[0026] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0028] Figure 1 The figure is a schematic diagram showing a motor vibration sound detection environment according to an exemplary embodiment.

[0029] Figure 2 It is a schematic diagram showing a motor vibration noise characteristic according to an exemplary embodiment.

[0030] Figure 3 It is a schematic diagram showing a motor vibration noise characteristic according to an exemplary embodiment.

[0031] Figure 4 The figure is a flow chart of a motor vibration sound detection method according to an exemplary embodiment.

[0032] Figure 5 The invention is a flow chart showing a method for determining harmonic characteristics of an audio signal according to an exemplary embodiment.

[0033] Figure 6It is a schematic diagram showing the overall frequency of a motor starting to oscillate according to an exemplary embodiment.

[0034] Figure 7 The figure is a flow chart of a method for normalizing frequency domain features according to an exemplary embodiment.

[0035] Figure 8 The figure is a flow chart of a method for extracting harmonic features according to an exemplary embodiment.

[0036] Fig. 9 The invention is a flow chart showing a method for determining frequency stability of an audio signal according to an exemplary embodiment.

[0037] Fig.10 The invention is a flow chart showing a method for determining slope stability of an audio signal according to an exemplary embodiment.

[0038] Fig.11 The invention is a flow chart showing a method for detecting vibration noise generated during motor vibration according to an exemplary embodiment.

[0039] Fig.12 The figure is a flow chart of a motor vibration sound detection method according to an exemplary embodiment.

[0040] Fig.13 is a schematic diagram of a normalized spectrum according to an exemplary embodiment.

[0041] Fig.14 is a schematic diagram of a spectrum after differential calculation according to an exemplary embodiment.

[0042] Fig.15 is a schematic diagram of a spectrum after harmonic delta feature extraction according to an exemplary embodiment.

[0043] Fig.16 is a schematic diagram showing frequency stability according to an exemplary embodiment.

[0044] Fig.17 is a schematic diagram showing slope stability according to an exemplary embodiment.

[0045] Fig.18 is a schematic diagram of vibration signal comparison according to an exemplary embodiment.

[0046] Fig.19 The figure is a schematic diagram of a harmonic delta characteristic calculation result according to an exemplary embodiment.

[0047] Fig. 20The figure is a schematic diagram showing a frequency stability calculation result according to an exemplary embodiment.

[0048] Fig.21 is a schematic diagram of a slope stability calculation result according to an exemplary embodiment.

[0049] Fig. 22 The figure is a block diagram of a motor vibration and sound detection device according to an exemplary embodiment.

[0050] Fig.23 is a block diagram of a motor vibration and sound detection device 300 according to an exemplary embodiment. DETAILED DESCRIPTION

[0051] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure.

[0052] In the accompanying drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limitations on the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present disclosure. The embodiments of the present disclosure are described in detail below in conjunction with the accompanying drawings.

[0053] The motor vibration sound detection method provided by the embodiment of the present disclosure can be applied to the scene of detecting the vibration state of the motor. For example, it is applied to the scene of detecting the motor vibration noise detection of devices such as watches, mobile phones, and tablet computers. Among them, a motor is a device that converts electrical energy into mechanical energy, and can use the mechanical energy to generate kinetic energy to drive other devices. For example, the motor can be various types of electric motors, engines, or mechanical vibration generators.

[0054] In many industrial and mechanical applications, motors are key working parts, and their vibration status and performance are critical to the normal operation of the machine. However, there may be abnormal conditions such as vibration noise in the motor vibration, which may indicate motor failure or problems.

[0055] In the related art, the vibration state of the motor is generally detected by collecting vibration signals through special equipment such as vibration sensors, and then analyzing the vibration intensity and frequency of the collected vibration signals, but the noise in the motor vibration cannot be analyzed. In addition, the detection of motor noise is currently basically determined by human listening, which is too subjective, and there is no solution to objectively detect motor vibration noise.

[0056] Figure 1 The figure is a schematic diagram showing a motor vibration sound detection environment according to an exemplary embodiment.

[0057] The present disclosure embodiment below takes a terminal device with computing capability as the detection device 100, a sound card as the signal conversion device 101, a watch device with a vibration motor as the motor vibration device 102, and a microphone as the sound receiving device 103 as an example to illustrate the test environment of the motor vibration sound detection method involved in the above embodiment. First, the watch motor can be controlled to start vibrating in any way, and then the motor vibration signal is collected by the microphone and transmitted back to the terminal device via the sound card, and the terminal device calculates and analyzes the recorded signal.

[0058] In the disclosed embodiment, the motor vibration signal is collected mainly for the calculation and analysis of the motor vibration noise characteristics. By judging the motor vibration noise by hearing, the characteristics of the vibration noise are summarized as follows:

[0059] Figure 2 FIG. 1 is a schematic diagram showing a motor vibration noise characteristic according to an exemplary embodiment. Figure 2 As shown, the characteristics of motor vibration noise are the instability of signal frequency, which changes over time, resulting in large fluctuations in the sound.

[0060] Figure 3 FIG. 1 is a schematic diagram showing a motor vibration noise characteristic according to an exemplary embodiment. Figure 3 As shown, the characteristics of motor vibration noise are also manifested in that the harmonic characteristics are not obvious or even absent, and the sound is chaotic.

[0061] In view of this, a motor vibration sound detection method is proposed in the embodiment of the present disclosure. In the motor vibration sound detection method provided by the present disclosure, the analysis of three characteristic indicators, namely harmonic characteristics, frequency stability and slope stability, is integrated. If abnormal harmonic characteristics, unstable frequency or abnormal frequency change trend appear in the motor vibration audio signal, it can be detected that the motor vibration generates vibration noise.

[0062] Figure 4 is a flow chart of a motor vibration sound detection method according to an exemplary embodiment. Figure 4 As shown, the motor vibration sound detection method includes the following steps.

[0063] In step S11 , an audio signal of motor vibration is obtained, and frequency domain features of the audio signal are obtained.

[0064] The audio signal of the motor vibration is collected through a vibration signal collection device. The vibration signal may propagate differently at different locations, and the signal strength may weaken with increasing distance. To ensure the signal-to-noise ratio of the vibration signal, a collection device such as a microphone can be placed as close to the vibration motor as possible, or an Electret Condenser Microphone (ECM) can be used to fit the vibration motor to collect the audio signal. Placing a standard microphone as close to the vibration motor as possible, or using an ECM for signal collection, can enhance the signal strength and improve the signal quality, which helps to better capture the characteristics of the vibration in subsequent analysis.

[0065] At the same time, the algorithm requires a higher vibration frequency measurement range, and a higher sampling rate needs to be set. Using a higher sampling rate (for example, 48,000 samples / second) can ensure that more frequency information is captured in the signal, thereby more accurately analyzing the characteristics of the signal. Especially when analyzing motor vibration, high-frequency harmonic components may be involved, so a higher sampling frequency is required to ensure that these frequency components are not lost.

[0066] After obtaining the audio signal of the motor vibration, the audio signal is overlapped and framed and then windowed and converted into frequency domain features. For example, the audio signal is subjected to Fast Fourier Transform (FFT) to calculate the overall time-frequency spectrum data.

[0067] Among them, in order to ensure the acquisition of more precise time-frequency spectrum data, set a reasonable sliding window and overlapping sampling points, as well as a higher FFT size. For example, set the frame length to 2048 sample points. Slide 2048 sample points each time, and there is an overlap of 1024 sample points between adjacent windows. This ensures that the analysis will not miss the instantaneous changes of the signal, and also makes the transition between frames smoother. In addition, after dividing the signal into frames, apply the window function to reduce the influence of the frame edge. The window function can make the frame boundary smoother and avoid unnecessary spectrum leakage.

[0068] In step S12, the harmonic characteristics, frequency stability and slope stability of the audio signal are determined according to the frequency domain characteristics.

[0069] Among them, the harmonic characteristics are used to determine the harmonic components of the audio signal, and the frequency stability and slope stability are used to evaluate the frequency change trend of the audio signal.

[0070] In vibration signals, harmonics are integer multiples of the original frequency. This characteristic is normal mechanical vibration behavior. The audio signal generated when the motor vibrates normally will contain these harmonic components. However, the harmonic characteristics in noise are not obvious or even non-existent. By analyzing the differences or changes between harmonics, the noise in motor vibration can be detected.

[0071] Frequency stability and slope stability are indicators used to evaluate the trend of vibration frequency changes in audio signals. Frequency stability measures the stability of the vibration frequency in the sound signal. However, when the frequency of a signal containing noise undergoes a stable and regular change, simple frequency stability may be insensitive to such changes and lead to misjudgment. Therefore, slope stability can be introduced. Slope stability measures the trend of frequency changes in the sound signal. By combining frequency stability and slope stability, the trend of vibration frequency changes in the audio signal can be judged more comprehensively, reducing the possibility that a single frequency stability test may be insensitive to a certain change, thereby leading to misjudgment and missed judgment of vibration noise.

[0072] In step S13, vibration noise generated during the vibration of the motor is detected based on at least one of harmonic characteristics, frequency stability, and slope stability.

[0073] Based on the analysis of the three characteristic indicators of harmonic characteristics, frequency stability and slope stability, if abnormal harmonic characteristics, unstable frequency or abnormal frequency change trend appear in the motor vibration audio signal, it can be determined that the motor vibration produces vibration noise.

[0074] In the disclosed embodiment, by collecting and processing the sound signal of motor vibration, not only can the motor vibration intensity, frequency and other indicators be measured, but also the subjective vibration noise can be converted into objective calculation and analysis. Through the method in the disclosed embodiment, the motor vibration noise is objectively and accurately detected, the influence of subjective factors is reduced, and the accuracy of judgment is improved.

[0075] Figure 5 is a flow chart of a method for determining harmonic characteristics of an audio signal according to an exemplary embodiment. Figure 5 As shown, the method includes the following steps.

[0076] In step S21, the frequency domain features are normalized to obtain frequency domain quantization values.

[0077] The amplitude values ​​of different frequency points in the frequency domain features may vary greatly, which may affect the subsequent analysis. Through normalization, these amplitude differences can be eliminated, so that the amplitude values ​​of different frequency points can be compared within the same scale range, so as to better perform feature analysis later.

[0078] In step S22, a differential operation is performed on the frequency domain quantization value to obtain a differential feature sequence.

[0079] Differentiation of frequency domain quantized values ​​is an operation used to process spectrum data in order to highlight changes and features in the spectrum. In vibration signal analysis, frequency domain quantized values ​​are normalized values ​​that represent the amplitude or energy distribution of different frequencies. Differentiation of frequency domain quantized values ​​is performed to calculate the amplitude difference between adjacent frequency points in order to eliminate errors and better capture frequency changes or features.

[0080] The overall frequency will slowly increase between the time the motor starts to vibrate and the time the problem occurs. This frequency variation is normal and expected and should not be considered abnormal. Figure 6 FIG. 1 is a schematic diagram showing the overall frequency of a motor starting to vibrate according to an exemplary embodiment. Figure 6 As shown in the figure, the overall frequency of the motor starts to increase slowly. In order to prevent this normal change from being misjudged as abnormal, and to remove the low-frequency envelope trend, the frequency domain quantization value is differentially calculated to obtain the differential feature sequence. The low-frequency envelope trend is not due to the abnormality of the motor itself, but the frequency is slowly increased due to normal operation.

[0081] Among them, the frequency domain quantization value can be differentially calculated as follows:

[0082] ΔF (x) =abs(F (x) -F(x-1))

[0083] F(x)-F(x-1) means subtracting the amplitude corresponding to each frequency from the amplitude corresponding to its adjacent frequency, and taking the absolute value of the difference.

[0084] In step S23, the harmonic features in the differential feature sequence are extracted.

[0085] In vibration signal analysis, the data in the differential feature sequence is the amplitude difference between adjacent frequency points that eliminates the low-frequency envelope trend error, reflecting the change of harmonics. However, not all data in the differential feature sequence are related to harmonic features. In order to extract the harmonic features in the signal, it is necessary to separate the noise in the signal from the specific harmonic components. For example, calculate the broad spectrum reference value of the broad spectrum noise, and subtract the broad spectrum reference value from the data in the differential feature sequence. This can highlight the harmonic features and make them easier to analyze and detect.

[0086] In the disclosed embodiment, the frequency domain features are normalized, the frequency domain quantized values ​​are differentially calculated to obtain a differential feature sequence, and finally the harmonic features are extracted from the differential feature sequence. Abnormal harmonic features can indicate the noise of motor vibration. By analyzing the harmonic features, the motor vibration noise can be better detected to reduce misjudgment and improve the accuracy of judgment.

[0087] Figure 7 is a flow chart of a method for normalizing frequency domain features according to an exemplary embodiment. Figure 7 As shown, the method includes the following steps.

[0088] In step S31 , a frequency band within the effective frequency range of the motor vibration signal of the frequency domain characteristics is intercepted.

[0089] In the analysis of motor vibration signals, the effective frequency range of motor vibration signals is usually concentrated in a specific frequency band rather than covering the entire spectrum. By intercepting a specific frequency range in the frequency domain characteristics, in order to more centrally analyze and detect information related to vibration noise detection, unnecessary analysis of the entire spectrum is avoided, thereby improving detection efficiency.

[0090] Different effective frequency ranges can be set for motors of different devices. For example, in the application scenario of vibration noise detection of watch motors, the intercepted frequency range can be [800, 15000] Hz. In the application scenario of vibration noise detection of motors such as mobile phones, the intercepted frequency range can be [400, 15000] Hz.

[0091] In step S32, normalization processing is performed based on the frequency average value of the spectrum within the frequency band to obtain a frequency domain quantization value.

[0092] For each single-frame spectrum, the amplitude or energy distribution average of each frequency within the effective frequency range is calculated, and this average is used as the quantization reference 1 to adjust the frequency domain values ​​of all frequencies to obtain the frequency domain quantization value.

[0093] In the disclosed embodiment, the frequency domain quantization value is obtained by intercepting the effective frequency band of the frequency domain feature and performing a benchmark quantization on the amplitude or energy distribution of each frequency. Through the normalization processing method in the disclosed embodiment, the amplitude difference can be eliminated, and the amplitude values ​​of different frequency points can be compared within the same scale range. The numerical range during calculation can also be reduced, the calculation complexity can be reduced, and it is helpful to perform feature analysis and calculation more effectively.

[0094] Figure 8 is a flow chart of a method for extracting harmonic features according to an exemplary embodiment. Figure 8 As shown, the method includes the following steps.

[0095] In step S41, a first confidence interval is determined according to a first confidence level, and a maximum value of the first confidence interval is used as a broad-spectrum reference value of broad-spectrum noise.

[0096] Among them, broad spectrum noise is steady-state noise in the audio signal that is unrelated to the harmonic features. Not all acquired audio signals are related to harmonic features. In order to extract the harmonic features in the signal, it is necessary to separate the noise in the signal from the specific harmonic components. This can highlight the harmonic features and make them easier to analyze and detect.

[0097] Since the proportion of points with harmonic features is extremely low, the confidence interval of all frequency domain quantization values ​​in the differential feature sequence can be calculated with a higher confidence level. This confidence interval indicates that within a given frequency range, data values ​​with a certain confidence level will fall within this interval. The maximum value of the confidence interval data is intercepted as the broad spectrum benchmark. Within the confidence interval, find the data point with the maximum amplitude value, that is, the maximum value of the confidence interval data. This maximum value is considered to be the broad spectrum benchmark, which represents the prominent harmonic features in the signal, while suppressing other low amplitude or noise components.

[0098] In step S42, the difference between each value in the differential feature sequence and the broad spectrum reference is determined to obtain a plurality of difference values.

[0099] Subtract the broad spectrum reference value from each value in the differential feature sequence. The purpose of this operation is to eliminate the broad spectrum noise component in the signal, thereby making the harmonic features more prominent.

[0100] In step S43, the sum of the multiple differences is taken as the harmonic characteristic value.

[0101] In the data of harmonic features after removing broad spectrum noise, all these highlighted data are summed up, and the value obtained represents the total intensity of all harmonic components. By adding all these differences, a harmonic feature value is obtained to represent the quantification of all harmonic components in the signal.

[0102] In the disclosed embodiment, the harmonic features without broad-spectrum noise are obtained by intercepting the maximum value of the confidence interval data as the broad-spectrum benchmark, and then subtracting this benchmark value from the differential feature sequence. Since the broad-spectrum benchmark is defined as the data point with the largest amplitude, this means that the harmonic components are highlighted and other irrelevant components are suppressed. The effect of this is to enhance the visibility of the harmonic features, making them easier to identify and analyze in the data. In this way, harmonic features can be extracted and quantified more accurately, providing a more powerful tool for the detection and analysis of vibration noise.

[0103] After completing the quantification and analysis of the harmonic characteristics of the vibration signal, the frequency stability and slope stability of the audio signal can be further determined. Frequency stability and slope stability are indicators used to evaluate the trend of the vibration frequency in the audio signal over time. Therefore, based on the conversion of the acquired motor vibration audio signal into frequency domain features after overlapping frame division and windowing in the above disclosed embodiment, it is necessary to convert the frequency domain features into time domain spectrum.

[0104] Fig. 9 is a flow chart showing a method for determining frequency stability of an audio signal according to an exemplary embodiment. Fig. 9 As shown, the method includes the following steps.

[0105] In step S51, the frequency domain features are converted into a time domain spectrum.

[0106] For frequency stability and slope stability measurements, since the motor vibration frequency has a large fluctuation range, a smaller Fourier transform size can be used, for example, the size of the short-time Fourier transform (STFT) can be set to 256. In this way, a shorter time window can be provided to capture frequency changes faster and more accurately.

[0107] In step S52, the time domain spectrum is normalized to obtain a quantized value of the time domain spectrum.

[0108] The amplitude values ​​at different frequency points in the time domain spectrum may vary greatly, which may affect subsequent analysis. Through normalization, these amplitude differences can be eliminated, so that the amplitude values ​​at different frequency points can be compared within the same scale range, so as to better analyze the frequency stability in the future.

[0109] In step S53, the variance of the time domain spectrum quantization value is determined, and the variance is used as the frequency stability.

[0110] After intercepting the effective frequency range, the time domain spectrum is normalized, and the variance of the normalized time domain spectrum quantization value is directly calculated to obtain the frequency stability of each frequency. Finally, the average value of all frequency stabilities is calculated to obtain the frequency stability value of this frame.

[0111] Among them, the frequency stability value of this frame can be calculated by the following formula:

[0112]

[0113] x i is the frequency stability value of a certain frequency in a certain frame, and mean(x) is the average value of the frequency stability value of a certain frequency at all times.

[0114] In the disclosed embodiment, frequency stability measures the stability of the frequency components in the signal. By calculating the variance between the frequency components, it is possible to understand whether the frequency of the signal varies within a relatively stable range. Frequency stability can be used to detect frequency fluctuations. For example, in motor vibration, if the frequency fluctuations are large, it may indicate that the motor may have vibration noise.

[0115] Fig.10 is a flow chart showing a method for determining the slope stability of an audio signal according to an exemplary embodiment. Fig.10 As shown, the method includes the following steps.

[0116] In step S61, the frequency domain features are converted into a time domain spectrum.

[0117] For frequency stability and slope stability measurements, since the motor vibration frequency has a large fluctuation range, a smaller Fourier transform size can be used, for example, the size of the short-time Fourier transform (STFT) can be set to 256. In this way, a shorter time window can be provided to capture frequency changes faster and more accurately.

[0118] In step S62, the time domain spectrum is normalized to obtain a quantized value of the time domain spectrum.

[0119] The amplitude values ​​at different frequency points in the time domain spectrum may vary greatly, which may affect the subsequent analysis. Through normalization, these amplitude differences can be eliminated, so that the amplitude values ​​at different frequency points can be compared within the same scale range, so as to better analyze the slope stability in the subsequent analysis.

[0120] In step S63, a differential operation is performed on the time domain spectrum quantization value to obtain a differential feature sequence.

[0121] In step S64, the variance of the differential feature sequence is determined, and the variance is used as the slope stability.

[0122] After intercepting the effective frequency range, the time domain spectrum is normalized, and the variance is calculated after the normalized time domain spectrum quantization value is differentially operated to obtain the slope stability of each frequency. Finally, the slope stability of all frequencies is averaged to obtain the slope stability value of this frame.

[0123] In the disclosed embodiment, slope stability focuses on the rate of frequency change. By analyzing the slope change of the spectrum, it is determined whether the change trend of the signal frequency is stable. In motor vibration, slope stability can reveal whether the rate of frequency change tends to be stable, thereby helping to determine whether there is noise in the vibration.

[0124] By comprehensively analyzing frequency stability and slope stability, we can gain a more comprehensive understanding of the frequency characteristics and changing trends of the signal, helping to detect abnormal conditions in the signal, such as the presence of vibration noise. The comprehensive analysis of the two can increase the understanding of the dynamic properties of the signal and provide more accurate vibration and sound analysis results.

[0125] Fig.11 is a flow chart showing a method for detecting vibration noise generated during motor vibration according to an exemplary embodiment. Fig.11 As shown, the method includes the following steps.

[0126] In step S71, thresholds of the harmonic characteristic value, frequency stability and slope stability are set respectively according to the harmonic characteristic value, frequency stability and slope stability.

[0127] In the detection of vibration noise, the method in the embodiment of the present disclosure includes three key characteristic indicators, namely harmonic characteristic value, frequency stability and slope stability. Each indicator provides different aspects of information about the vibration signal, such as harmonic components, frequency change stability and slope change stability.

[0128] In step S72, if the harmonic characteristic value is less than the threshold or at least one of the frequency stability and the slope stability is greater than the threshold, it is determined that vibration noise exists.

[0129] In order to convert these characteristic indicators into quantifiable judgment criteria, a suitable threshold can be set for each characteristic. If the harmonic characteristic value is less than the threshold or at least one of the frequency stability and slope stability is greater than the corresponding threshold, it means that this indicator is abnormal to some extent. Among them, the threshold can be adjusted according to the actual situation to ensure that the vibration noise is judged as accurately as possible.

[0130] In the disclosed embodiment, the comprehensive use of multiple feature indicators and threshold settings can improve the reliability and accuracy of vibration noise detection. Different features can play a complementary role in different situations, and the threshold settings can be flexibly adjusted according to actual needs to achieve the best detection effect.

[0131] The present disclosure embodiment will now illustrate the motor vibration sound detection method mentioned above with examples, using watch motor vibration noise detection as an application scenario of the method, and using harmonic delta features as harmonic features in the method.

[0132] Fig.12 The figure is a flow chart of a motor vibration sound detection method according to an exemplary embodiment.

[0133] like Fig.12As shown, firstly, the vibration signal is collected. During the collection process, in order to ensure that the vibration signal-to-noise ratio is sufficient, the standard microphone is as close to the vibration motor as possible or the ECM is used to fit the vibration motor, and a higher vibration frequency measurement range is set. For example, the recording is performed according to 48,000 samples.

[0134] After collecting the audio signal of motor vibration, perform time domain signal analysis, overlap frame division and windowing, and perform FFT, transformation, and calculate the overall time-frequency spectrum data. For example, to ensure the fine acquisition of time-frequency spectrum data, the sliding window is set to 2048, the overlapping window is set to 1024, and the FFT size of the motor vibration harmonic delta feature is set to 4096.

[0135] Because the effective analysis frequency band of motor vibration is not full-band, it is necessary to perform bandpass data capture before measuring the harmonic Delta component. For example, use the frequency range [800,15000] to capture the frame and perform FFT transformation on the frequency domain data.

[0136] In order to ensure that the variance measurement of the final output data is quantized to a uniform order of magnitude, the time-frequency spectrum data needs to be normalized. For example, the average value of the single-frame spectrum in the frequency range of [800,15000] is calculated and used as the quantization benchmark 1 to adjust all frequency domain values. Fig.13 is a schematic diagram of a normalized spectrum according to an exemplary embodiment. The normalized spectrum is as follows: Fig.13 shown.

[0137] In most cases, there is a process of slowly increasing frequency from the start of vibration to the occurrence of problems during motor vibration, but this process is normal and not a feature of unstable noise. Therefore, to prevent misjudgment, the low-frequency envelope trend is removed, the spectrum difference calculation is performed, and the absolute value of the difference data is taken. The formula is as follows:

[0138] ΔF (x) =abs(F (x) -F(x-1))

[0139] Fig.14 FIG. 1 is a schematic diagram of a spectrum after differential calculation according to an exemplary embodiment. Fig.14 shown.

[0140] In order to extract the harmonic delta feature in the signal, the broad spectrum curve in the signal needs to be separated to highlight the harmonic delta. One example is that because the proportion of points with harmonic delta features is extremely low, the confidence interval of all data in the curve can be calculated according to the confidence level of 0.8, and the maximum value of the confidence interval data is intercepted as the broad spectrum benchmark, and the above differential feature sequence is subtracted from the broad spectrum benchmark. Fig.15FIG. 1 is a schematic diagram of a spectrum after harmonic delta feature extraction according to an exemplary embodiment. The spectrum after highlighting the harmonic delta is as follows: Fig.15 After all harmonic delta features are highlighted, the sum of all data in this array represents the harmonic delta component.

[0141] Based on the frequency domain characteristics obtained by the above harmonic delta component measurement frame, FFT conversion is performed again to the time-frequency spectrum. Since the motor vibration frequency fluctuation range is large, the FFT size must be reduced when measuring frequency stability and slope stability. For example, when using STFT, the size is set to 256.

[0142] Fig.16 is a schematic diagram showing a frequency stability according to an exemplary embodiment, such as Fig.16 As shown, the effective frequency band is intercepted by referring to the harmonic delta method and the data is normalized. The variance of the normalized spectrum is directly calculated to obtain the stability of each frequency. Finally, the average value of all frequency stabilities is calculated as the frequency stability value of this frame. The calculation formula is as follows:

[0143]

[0144] Fig.17 is a schematic diagram showing a slope stability according to an exemplary embodiment. The normalized spectrum data is differentiated by referring to the above-mentioned method of calculating harmonic delta, and the variance of the differential feature sequence is calculated to obtain the slope stability of each frequency, such as Fig.17 The average value of all frequency slope stability is calculated as the slope stability of this frame.

[0145] In the disclosed embodiment, the motor vibration noise can be comprehensively judged by the above three characteristic indicators, and a threshold can be set. If any indicator fails, vibration noise exists. The threshold value can be adjusted according to the actual situation to ensure that the vibration noise is judged as accurately as possible.

[0146] The present disclosure embodiment will now illustrate the motor vibration sound detection method mentioned above with examples, using watch motor vibration noise detection as an application scenario of the method, and using harmonic delta features as harmonic features in the method.

[0147] Fig.18 FIG. 1 is a schematic diagram showing a vibration signal comparison according to an exemplary embodiment. Fig.18 As shown, the left picture shows the subjective perception of normal motor vibration sound, and the right picture shows the subjective perception of abnormal motor vibration signal.

[0148] The harmonic delta characteristics, frequency stability and slope stability of the above vibration signal are calculated. Fig.19FIG. 1 is a schematic diagram showing a harmonic delta characteristic calculation result according to an exemplary embodiment. Fig.19 The larger the harmonic delta component, the less noise there is in the signal.

[0149] Fig. 20 FIG. 1 is a schematic diagram showing a frequency stability calculation result according to an exemplary embodiment. Fig. 20 The smaller the frequency stability, the more stable the listening experience.

[0150] Fig.21 FIG. 1 is a schematic diagram showing a slope stability calculation result according to an exemplary embodiment. Fig.21 The smaller the slope stability, the more stable the listening sound is over time.

[0151] In the embodiment of the present disclosure, by calculating and quantifying the characteristics of a normal vibration signal of a vibration signal with noise, it is verified that the motor vibration sound detection method of the present disclosure has the effect of objectively and accurately detecting motor vibration noise.

[0152] Based on the same concept, an embodiment of the present disclosure also provides a motor vibration and sound detection device.

[0153] It is understandable that the motor vibration and sound detection device provided in the embodiment of the present disclosure includes hardware structures and / or software modules corresponding to the execution of each function in order to realize the above functions. In combination with the units and algorithm steps of each example disclosed in the embodiment of the present disclosure, the embodiment of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiment of the present disclosure.

[0154] Fig. 22 FIG. 1 is a block diagram of a motor vibration sound detection device according to an exemplary embodiment. Fig. 22 The device includes an acquisition unit 201, a processing unit 202 and a detection unit 203.

[0155] An acquisition unit 201 is used to acquire an audio signal of motor vibration and acquire frequency domain characteristics of the audio signal;

[0156] A processing unit 202, configured to determine the harmonic characteristics, frequency stability, and slope stability of the audio signal according to the frequency domain characteristics, wherein the harmonic characteristics are used to determine the harmonic components of the audio signal, and the frequency stability and slope stability are used to evaluate the frequency change trend of the audio signal;

[0157] The detection unit 203 is used to detect vibration noise generated during the vibration of the motor according to at least one of harmonic characteristics, frequency stability and slope stability.

[0158] In one embodiment, the processing unit 202 determines the harmonic features of the audio signal based on the frequency domain features in the following manner: normalizing the frequency domain features to obtain frequency domain quantization values; performing a differential operation on the frequency domain quantization values ​​to obtain a differential feature sequence, wherein the differential feature sequence includes the frequency domain quantization values ​​with errors removed; and extracting the harmonic features in the differential feature sequence.

[0159] In one embodiment, the processing unit 202 normalizes the frequency domain features to obtain frequency domain quantization values ​​in the following manner: intercepting a frequency band within the effective frequency range of the motor vibration signal of the frequency domain features; performing normalization based on the frequency average value of the spectrum within the frequency band to obtain a frequency domain quantization value, and the frequency domain quantization value and the spectrum energy corresponding to the frequency have a one-to-one correspondence.

[0160] In one embodiment, the processing unit 202 extracts the harmonic features in the differential feature sequence in the following manner: determining a first confidence interval according to a first confidence level, and taking the maximum value of the first confidence interval as a broad-spectrum reference value of broad-spectrum noise, where the broad-spectrum noise is steady-state noise in the audio signal that is unrelated to the harmonic features; determining the difference between each value in the differential feature sequence and the broad-spectrum reference to obtain multiple differences; and taking the sum of the multiple differences as the harmonic feature value.

[0161] In one embodiment, the processing unit 202 determines the frequency stability of the audio signal according to the frequency domain features in the following manner: converting the frequency domain features into a time domain spectrum; normalizing the time domain spectrum to obtain a quantized value of the time domain spectrum;

[0162] The variance of the quantized values ​​of the time-domain spectrum is determined and used as a measure of frequency stability.

[0163] In one embodiment, the processing unit 202 determines the slope stability of the audio signal according to the frequency domain features in the following manner: converting the frequency domain features into a time domain spectrum; normalizing the time domain spectrum to obtain a quantized value of the time domain spectrum;

[0164] The time domain spectrum quantization value is subjected to a differential operation to obtain a differential feature sequence; the variance of the differential feature sequence is determined, and the variance is used as the slope stability.

[0165] In one embodiment, the detection unit 203 detects the vibration noise generated during the motor vibration process according to at least one of the harmonic characteristics, frequency stability and slope stability in the following manner: thresholds of the harmonic characteristic value, frequency stability and slope stability are set according to the harmonic characteristic value, frequency stability and slope stability, respectively; if the harmonic characteristic value is less than the threshold or at least one of the frequency stability and the slope stability is greater than the threshold, it is determined that vibration noise exists.

[0166] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0167] Fig.23 is a block diagram of a motor vibration sound detection device 300 according to an exemplary embodiment. For example, the device 300 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0168] Reference Fig.23 , the device 300 may include one or more of the following components: a processing component 302 , a memory 304 , a power component 306 , a multimedia component 308 , an audio component 310 , an input / output (I / O) interface 312 , a sensor component 314 , and a communication component 316 .

[0169] The processing component 302 generally controls the overall operation of the device 300, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 302 may include one or more processors 320 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 302 may include one or more modules to facilitate the interaction between the processing component 302 and other components. For example, the processing component 302 may include a multimedia module to facilitate the interaction between the multimedia component 308 and the processing component 302.

[0170] The memory 304 is configured to store various types of data to support operations on the device 300. Examples of such data include instructions for any application or method operating on the device 300, contact data, phone book data, messages, pictures, videos, etc. The memory 304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0171] The power component 306 provides power to the various components of the device 300. The power component 306 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 300.

[0172] The multimedia component 308 includes a screen that provides an output interface between the device 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 308 includes a front camera and / or a rear camera. When the device 300 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0173] The audio component 310 is configured to output and / or input audio signals. For example, the audio component 310 includes a microphone (MIC), and when the device 300 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 304 or sent via the communication component 316. In some embodiments, the audio component 310 also includes a speaker for outputting audio signals.

[0174] I / O interface 312 provides an interface between processing component 302 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0175] The sensor assembly 314 includes one or more sensors for providing various aspects of the status assessment of the device 300. For example, the sensor assembly 314 can detect the open / closed state of the device 300, the relative positioning of components, such as the display and keypad of the device 300, the sensor assembly 314 can also detect the position change of the device 300 or a component of the device 300, the presence or absence of user contact with the device 300, the orientation or acceleration / deceleration of the device 300, and the temperature change of the device 300. The sensor assembly 314 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 314 can also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 314 can also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0176] The communication component 316 is configured to facilitate wired or wireless communication between the device 300 and other devices. The device 300 can access a wireless network based on a communication standard, such as WiFi, 3G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 316 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 316 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0177] In an exemplary embodiment, the apparatus 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0178] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 304 including instructions, and the instructions can be executed by the processor 320 of the device 300 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0179] It is to be understood that in the present disclosure, "plurality" refers to two or more than two, and other quantifiers are similar. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. The singular forms "a", "the" and "the" are also intended to include plural forms, unless the context clearly indicates other meanings.

[0180] It is further understood that the terms "first", "second", etc. are used to describe various information, but such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other, and do not indicate a specific order or degree of importance. In fact, the expressions "first", "second", etc. can be used interchangeably. For example, without departing from the scope of the present disclosure, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information.

[0181] It will be further understood that the terms “center”, “longitudinal”, “lateral”, “lower”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present embodiment and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation.

[0182] It can be further understood that, unless otherwise specified, “connection” includes a direct connection without other components between the two, and also includes an indirect connection with other components between the two.

[0183] It is further understood that, although the operations are described in a specific order in the drawings in the embodiments of the present disclosure, it should not be understood as requiring the operations to be performed in the specific order shown or in a serial order, or requiring the execution of all the operations shown to obtain the desired results. In certain environments, multitasking and parallel processing may be advantageous.

[0184] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modifications, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure.

[0185] It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the scope of the appended claims.

Claims

1. A motor vibration sound detection method, characterized in that: include: Acquire an audio signal of motor vibration, and acquire frequency domain characteristics of the audio signal; Determining a harmonic feature, a frequency stability, and a slope stability of the audio signal according to the frequency domain feature, wherein the harmonic feature is used to determine the harmonic component of the audio signal, and the frequency stability and the slope stability are used to evaluate the frequency change trend of the audio signal; Vibration noise generated during vibration of the motor is detected according to at least one of the harmonic characteristics, the frequency stability, and the slope stability.

2. The motor vibration sound detection method according to claim 1, characterized in that: The determining the harmonic feature of the audio signal according to the frequency domain feature comprises: Normalizing the frequency domain features to obtain frequency domain quantization values; Performing a differential operation on the frequency domain quantization value to obtain a differential feature sequence, wherein the differential feature sequence includes the frequency domain quantization value with error removed; Extracting harmonic features from the differential feature sequence.

3. The motor vibration sound detection method according to claim 2, characterized in that: The normalizing the frequency domain features to obtain frequency domain quantization values ​​includes: Intercepting a frequency band within an effective frequency range of the motor vibration signal having the frequency domain characteristics; Normalization processing is performed based on the frequency average value of the spectrum within the frequency band to obtain a frequency domain quantization value, and the frequency domain quantization value has a one-to-one correspondence with the spectrum energy corresponding to the frequency.

4. The method according to claim 2, characterized in that: The extracting harmonic features from the differential feature sequence comprises: determining a first confidence interval according to a first confidence level, and taking a maximum value of the first confidence interval as a broad-spectrum reference value of broad-spectrum noise, wherein the broad-spectrum noise is steady-state noise in the audio signal that is unrelated to the harmonic feature; Determine the difference between each value in the differential feature sequence and the broad spectrum reference to obtain a plurality of difference values; The sum of the multiple differences is used as the harmonic characteristic value.

5. The motor vibration sound detection method according to claim 1, characterized in that: The determining the frequency stability of the audio signal according to the frequency domain feature comprises: Converting the frequency domain features into a time domain spectrum; Normalizing the time domain spectrum to obtain a time domain spectrum quantization value; The variance of the time domain spectrum quantization value is determined, and the variance is used as the frequency stability.

6. The motor vibration sound detection method according to claim 1, characterized in that: The determining the slope stability of the audio signal according to the frequency domain feature comprises: Convert the frequency domain features into a time domain spectrum; Normalizing the time domain spectrum to obtain a time domain spectrum quantization value; Performing a differential operation on the time domain spectrum quantization value to obtain a differential feature sequence; The variance of the differential feature sequence is determined, and the variance is used as the slope stability.

7. The motor vibration sound detection method according to claim 1, characterized in that: The detecting the vibration noise generated during the vibration of the motor according to at least one of the harmonic characteristics, the frequency stability and the slope stability comprises: According to the harmonic characteristic value, frequency stability and slope stability, threshold values ​​of the harmonic characteristic value, frequency stability and slope stability are set respectively; If the harmonic characteristic value is smaller than a threshold value or at least one of the frequency stability and the slope stability is greater than a threshold value, it is determined that the vibration noise exists.

8. A motor vibration sound detection device, characterized in that: include: An acquisition unit, used to acquire an audio signal of motor vibration and acquire frequency domain characteristics of the audio signal; a processing unit, configured to determine a harmonic feature, a frequency stability, and a slope stability of the audio signal according to the frequency domain feature, wherein the harmonic feature is used to determine the harmonic component of the audio signal, and the frequency stability and the slope stability are used to evaluate the frequency change trend of the audio signal; The detection unit is used to detect the vibration noise generated during the vibration of the motor according to at least one of the harmonic characteristics, the frequency stability and the slope stability.

9. The motor vibration sound detection device according to claim 8, characterized in that: The processing unit determines the harmonic feature of the audio signal according to the frequency domain feature in the following manner: Normalizing the frequency domain features to obtain frequency domain quantization values; Performing a differential operation on the frequency domain quantization value to obtain a differential feature sequence, wherein the differential feature sequence includes the frequency domain quantization value with error removed; Extracting harmonic features from the differential feature sequence.

10. The motor vibration sound detection device according to claim 9, characterized in that: The processing unit performs normalization processing on the frequency domain feature to obtain a frequency domain quantization value in the following manner: Intercepting a frequency band within an effective frequency range of the motor vibration signal having the frequency domain characteristics; Normalization processing is performed based on the frequency average value of the spectrum within the frequency band to obtain a frequency domain quantization value, and the frequency domain quantization value has a one-to-one correspondence with the spectrum energy corresponding to the frequency.

11. The device according to claim 9, characterized in that The processing unit extracts the harmonic features in the differential feature sequence in the following manner: determining a first confidence interval according to a first confidence level, and taking a maximum value of the first confidence interval as a broad-spectrum reference value of broad-spectrum noise, wherein the broad-spectrum noise is steady-state noise in the audio signal that is unrelated to the harmonic feature; Determine the difference between each value in the differential feature sequence and the broad spectrum reference to obtain a plurality of difference values; The sum of the multiple differences is used as the harmonic characteristic value.

12. The motor vibration sound detection device according to claim 8, characterized in that: The processing unit determines the frequency stability of the audio signal according to the frequency domain feature in the following manner: Converting the frequency domain features into a time domain spectrum; Normalizing the time domain spectrum to obtain a time domain spectrum quantization value; The variance of the time domain spectrum quantization value is determined, and the variance is used as the frequency stability.

13. The motor vibration sound detection device according to claim 8, characterized in that: The processing unit determines the slope stability of the audio signal according to the frequency domain feature in the following manner: Convert the frequency domain features into a time domain spectrum; Normalizing the time domain spectrum to obtain a time domain spectrum quantization value; Performing a differential operation on the time domain spectrum quantization value to obtain a differential feature sequence; The variance of the differential feature sequence is determined, and the variance is used as the slope stability.

14. The motor vibration sound detection device according to claim 8, characterized in that: The detection unit detects the vibration noise generated during the vibration of the motor according to at least one of the harmonic characteristics, the frequency stability and the slope stability in the following manner: According to the harmonic characteristic value, frequency stability and slope stability, threshold values ​​of the harmonic characteristic value, frequency stability and slope stability are set respectively; If the harmonic characteristic value is smaller than a threshold value or at least one of the frequency stability and the slope stability is greater than a threshold value, it is determined that the vibration noise exists.

15. A motor vibration sound detection device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to: execute the method described in any one of claims 1 to 7.

16. A storage medium, characterized in that: The storage medium stores instructions, and when the instructions in the storage medium are executed by a processor of the terminal, the terminal is enabled to execute the method according to any one of claims 1 to 7.

Images