Motor deflection abnormal sound detection method and device, test equipment and storage medium

Abstract

The invention discloses a motor deflection abnormal sound detection method and device, test equipment and a storage medium, and the method comprises the steps: starting a target motor, collecting a sound signal transmitted by the target motor, and obtaining an audio sampling signal; based on a preset framing duration, performing short-time Fourier transform on the audio sampling signal to obtain first sound pressure frequency domain data of a plurality of audio clips; combining equivalent sound pressure level data obtained by carrying out equivalent sound pressure calculation and sound pressure level conversion on the first sound pressure frequency domain data of the plurality of audio clips based on a time sequence to obtain sound pressure level time domain data of which the sound pressure level changes along with time; fourier transform is carried out on the sound pressure level time domain data to obtain sound pressure level frequency domain data with the sound pressure level changing along with the modulation frequency; according to the sound pressure level frequency domain data and a preset deflection abnormal sound threshold, whether deflection abnormal sound exists in the target motor or not is determined. According to the invention, whether the motor has abnormal deflection noise or not can be accurately and effectively detected.

Description

Technical Field

[0001] The present invention relates to the technical field of audio detection, and in particular to a method, a device, a testing device and a storage medium for detecting abnormal noise caused by motor deflection. Background Art

[0002] Motors, such as the switch drive motor of a car tailgate, may emit a wobble noise during rotation due to an eccentricity problem in the shaft during the production process. The wobble noise can be detected by listening to the human ear, but this method is highly subjective, and the accuracy and efficiency of the detection are easily affected by human subjective factors. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention provides a motor yaw abnormal noise detection method, device, test equipment and storage medium, which can accurately and effectively detect whether the motor has yaw abnormal noise.

[0004] On the one hand, an embodiment of the present invention provides a method for detecting abnormal noise caused by motor deflection, comprising: Starting a target motor and collecting a sound signal emitted by the target motor to obtain an audio sampling signal; Based on a preset frame duration, performing a short-time Fourier transform on the audio sampling signal to obtain first sound pressure frequency domain data of a plurality of audio clips; Combining equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on time sequence to obtain sound pressure level time domain data of the sound pressure level changing with time; Performing Fourier transformation on the sound pressure level time domain data to obtain sound pressure level frequency domain data showing that the sound pressure level varies with the modulation frequency; It is determined whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and a preset yaw abnormal sound threshold.

[0005] According to some embodiments of the present invention, performing short-time Fourier transform on the audio sampling signal based on a preset frame duration to obtain first sound pressure frequency domain data of a plurality of audio segments includes: Based on a preset frame duration, the audio sampling signal is framed to obtain a plurality of audio segments; Perform Fourier transform on each of the audio segments to obtain first sound pressure frequency domain data of the multiple audio segments.

[0006] According to some embodiments of the present invention, performing short-time Fourier transform on the audio sampling signal based on a preset frame duration to obtain first sound pressure frequency domain data of a plurality of audio segments includes: Based on a preset frame duration and a frame overlap duration, the audio sampling signal is framed to obtain a plurality of audio segments; Perform Fourier transform on each of the audio segments to obtain first sound pressure frequency domain data of the multiple audio segments.

[0007] According to some embodiments of the present invention, combining the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time includes: Performing equivalent sound pressure calculation and sound pressure level conversion on the first sound pressure frequency domain data of each of the audio segments to obtain equivalent sound pressure level data; The equivalent sound pressure level data corresponding to the plurality of audio clips are combined based on a time sequence to obtain time-domain sound pressure level data showing a sound pressure level varying with time.

[0008] According to some embodiments of the present invention, combining the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time includes: Performing weighting processing on the first sound pressure frequency domain data of each of the audio segments to obtain second sound pressure frequency domain data; Performing equivalent sound pressure calculation and sound pressure level conversion on the second sound pressure frequency domain data of each of the audio segments to obtain equivalent sound pressure level data; The equivalent sound pressure level data corresponding to the plurality of audio clips are combined based on a time sequence to obtain time-domain sound pressure level data showing a sound pressure level varying with time.

[0009] According to some embodiments of the present invention, combining the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time includes: Performing one-third octave band A-weighting processing on the first sound pressure frequency domain data of each of the audio clips to obtain second sound pressure frequency domain data; Performing equivalent sound pressure calculation and sound pressure level conversion on the second sound pressure frequency domain data of each of the audio segments to obtain equivalent sound pressure level data; The equivalent sound pressure level data corresponding to the plurality of audio clips are combined based on a time sequence to obtain time-domain sound pressure level data showing a sound pressure level varying with time.

[0010] According to some embodiments of the present invention, determining whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and a preset yaw abnormal sound threshold includes: It is determined whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and the yaw abnormal sound thresholds at different frequencies.

[0011] In a second aspect, an embodiment of the present invention provides a motor deflection abnormal noise detection device, comprising: An audio sampling module, used for starting a target motor and collecting a sound signal emitted by the target motor to obtain an audio sampling signal; A first transformation module, configured to perform a short-time Fourier transform on the audio sampling signal based on a preset frame duration to obtain first sound pressure frequency domain data of a plurality of audio segments; A second transformation module is used to combine the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on time sequence to obtain sound pressure level time domain data of the sound pressure level changing with time; A third transformation module is used to perform Fourier transformation on the sound pressure level time domain data to obtain sound pressure level frequency domain data in which the sound pressure level changes with the modulation frequency; The determination module is used to determine whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and a preset yaw abnormal sound threshold.

[0012] In a third aspect, an embodiment of the present invention provides a testing device, including a processor and a memory, wherein the memory stores a computer program, and the processor is used to implement the above-mentioned motor runout abnormal sound detection method when running the computer program.

[0013] In a fourth aspect, an embodiment of the present invention provides a storage medium, wherein a computer program is stored in the storage medium, and when the computer program is executed, the above-mentioned motor runout abnormal sound detection method is implemented.

[0014] The embodiments of the present invention have at least the following beneficial effects: The audio sampling signal is framed by short-time Fourier transform, and the first sound pressure frequency domain data obtained by the frame processing is subjected to equivalent sound pressure calculation and sound pressure level transformation, so as to obtain sound pressure level time domain data representing the sound of the target motor. The sound pressure level time domain data is subjected to Fourier transform to obtain sound pressure level frequency domain data. Combined with the yaw abnormal sound threshold, it is possible to accurately and effectively detect whether the target motor has yaw abnormal sound.

[0015] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 A flowchart of the steps of a method for detecting abnormal noise caused by motor deflection according to an embodiment of the present invention; Figure 2 This is an energy spectrum diagram of a good motor according to an embodiment of the present invention; Figure 3 This is an energy spectrum diagram of a defective motor with abnormal yaw noise according to an embodiment of the present invention; Figure 4 A schematic diagram of a frame duration and a frame overlap duration according to an embodiment of the present invention; Figure 5 The motor abnormal noise modulation curve diagram of a good motor and a defective motor with abnormal yaw noise according to an embodiment of the present invention; Figure 6 It is a principle block diagram of a motor yaw abnormal noise detection device according to an embodiment of the present invention; Figure 7 The figure is a functional block diagram of a testing device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0018] In the description of the present invention, "several" means one or more, "multiple" means more than two, greater than, less than, and exceeding are understood as not including the number itself, and "above", "below", and "within" are understood as including the number itself. If there is a description of "first", "second", etc., it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0019] In the description of the present invention, unless otherwise clearly defined, words such as “setting”, “installation” and “connection” should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above words in the present invention in combination with the specific content of the technical solution.

[0020] Motor runout is a common technical problem in the motor production process. The main reason for motor runout is the eccentricity of the shaft, which causes friction between the shaft and the connecting structure. For the same motor, the eccentric position of the shaft is relatively fixed, so the abnormal noise of the runout during rotation occurs periodically. The study found that the sound signal emitted by a motor with a runout problem is equivalent to modulating a fixed frequency modulation signal in the sound signal of a normal motor operation. Therefore, by determining whether there is a modulation signal of a specific frequency in the sound signal of the motor operation, the motor can be quantitatively and qualitatively analyzed to detect whether the motor has a runout problem.

[0021] Please refer to Figure 1 This embodiment discloses a method for detecting abnormal noise caused by motor deflection, including steps S100 to S500. It should be noted that the steps in this embodiment are numbered only for the convenience of review and understanding, and do not limit the execution order of the steps. The contents of each step are described in detail below: S100, starting a target motor and collecting a sound signal emitted by the target motor to obtain an audio sampling signal; Exemplarily, the target motor is the motor to be tested. During the test, in order to minimize the impact of environmental noise on the test results, the target motor can be placed in a relatively quiet space (such as a silent room) for testing. The target motor is started under the condition of simulating the working condition of the target motor, and the sound signal emitted during the operation of the motor is collected through a microphone to achieve the collection of the sound signal and obtain an audio sampling signal.

[0022] S200, performing short-time Fourier transform on the audio sampling signal based on a preset frame duration to obtain first sound pressure frequency domain data of a plurality of audio clips; For example, short-time Fourier transform is a commonly used processing method in signal analysis. The main purpose of short-time Fourier transform is to divide the continuous time domain signal into multiple segments and perform Fourier transform respectively, so as to obtain the energy spectrum of time-frequency-amplitude of each segment of the signal, and perform signal analysis by observing the change of energy spectrum, such as Figure 2 The energy spectrum of a good motor is shown in Figure 3 The energy spectrum of defective products with abnormal yaw noise is shown. Figure 2 and Figure 3 The horizontal axis is time, the vertical axis is frequency, and the color changes represent the energy (amplitude). Figure 2 Compared with Figure 3 The middle frame shows that there are fluctuations in energy at position A, and vertical lines of varying depths can be seen, proving the regularity of energy changes over time. However, the observation and analysis of the waveform depends on the experience of the analyst, which requires a high level of professionalism from the analyst, and it is difficult to achieve quantitative and qualitative analysis.

[0023] In some application examples, step S200 includes: Based on a preset frame duration, the audio sampling signal is framed to obtain multiple audio clips; Fourier transform is performed on each audio segment to obtain first sound pressure frequency domain data of multiple audio segments.

[0024] Exemplarily, the short-time Fourier transform is to perform windowing processing on the continuous audio sampling signal, wherein the window size of the windowing processing is the frame duration, such as Figure 4 As shown in the time lengths t1, t2 and t3 in the figure, the continuous audio sampling signal is cut into multiple audio segments, and then Fourier transform is performed on each audio segment to obtain the sound pressure frequency domain data corresponding to each audio segment. In this way, the same processing is performed on the multiple audio segments to obtain the first sound pressure frequency domain data of the multiple audio segments.

[0025] Alternatively, in some other application examples, step S200 includes: Based on the preset frame duration and frame overlap duration, the audio sampling signal is framed to obtain multiple audio segments; Fourier transform is performed on each audio segment to obtain first sound pressure frequency domain data of the multiple audio segments.

[0026] For example, in order to ensure the continuity of the signal between the previous audio segment and the next audio segment, there is a certain overlap time between the previous audio segment and the next audio segment, that is, the frame overlap time (also called the frame shift time), such as Figure 4 As shown by the mark τ. For example, assuming that the total duration of the audio sampling signal is 1 second, the frame duration can be set to 0.1 second, and the frame overlap duration can be set to 0.05 second, so that the audio sampling signal can be divided into 20 audio segments. By performing Fourier transform on each audio segment, the sound pressure frequency domain data of each audio segment can be obtained. When all audio segments are subjected to the same processing, the first sound pressure frequency domain data of multiple audio segments can be obtained.

[0027] S300, combining equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of first sound pressure frequency domain data of multiple audio clips based on time sequence to obtain sound pressure level time domain data of sound pressure level changing with time; For example, sound pressure is the most basic physical quantity for quantitatively describing sound waves. The sound pressure usually referred to is the effective sound pressure, which is obtained by taking the root mean square of the instantaneous sound pressure over time within a certain period of time, as shown in Formula 1:

[0028] In the formula, p e is the effective sound pressure value, N is the number of sampling points in the time domain,x (n) is the time domain sampling point value, and n is the time domain variable.

[0029] The equivalent sound pressure of this embodiment refers to the concept and calculation method of effective sound pressure. The difference is that the equivalent sound pressure is obtained by calculating the root mean square of the frequency, as shown in Formula 2:

[0030] In the formula, p ' e is the equivalent sound pressure value, N is the number of sampling points in the frequency domain, x '(n) is the frequency domain sampling point value, and n is the frequency domain variable. The sound pressure level of this embodiment is the ratio of the equivalent sound pressure of the sound to the benchmark reference sound pressure, which is calculated by taking the logarithm with base 10 and multiplying it by 20. Through equivalent sound pressure calculation and sound pressure level transformation, multiple frequency-sound pressure data in the first sound pressure frequency domain data of each audio segment can be converted into a single sound pressure level data corresponding to the audio segment to characterize the sound pressure level within a preset time (such as the frame duration). In this way, by performing the same processing on multiple audio segments, sound pressure level data corresponding to multiple audio segments can be obtained, that is, sound pressure level time domain data of the sound pressure level changing with time.

[0031] In some application examples, step S300 includes: The equivalent sound pressure calculation and sound pressure level transformation are performed on the first sound pressure frequency domain data of each audio clip to obtain equivalent sound pressure level data; the equivalent sound pressure level data corresponding to multiple audio clips are combined based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time.

[0032] For example, as shown in the example above, for an audio sampling signal with a duration of 1 second, 20 audio segments are obtained after short-time Fourier transform, such as 0~0.1 seconds, 0.05~0.15 seconds, 0.1~0.2 seconds... The equivalent sound pressure calculation and sound pressure level transformation are performed on the first sound pressure frequency domain data of each audio segment to obtain the equivalent sound pressure data of each audio segment. The starting time of each audio segment is used as the time point of the audio segment, which can be recorded as (0, p ' e1 ), (0.05, p ' e2 ), (0.1, p ' e3 ) .... By combining the equivalent sound pressure level data corresponding to the multiple audio clips based on the time sequence, the time domain data of the sound pressure level changing with time can be obtained.

[0033] Alternatively, in some other application examples, step S300 includes: Performing weighting processing on the first sound pressure frequency domain data of each audio clip to obtain second sound pressure frequency domain data; The equivalent sound pressure is calculated and the sound pressure level is transformed for the second sound pressure frequency domain data of each audio clip to obtain equivalent sound pressure level data; the equivalent sound pressure level data corresponding to multiple audio clips are combined based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time.

[0034] For example, weighting processing is a processing method that simulates human hearing and corrects the sound pressure signal to a value that is close to the hearing perception. Common weighting processing methods include A-weighting, B-weighting, and C-weighting, among which A-weighting simulates the frequency characteristics of the human ear to low-intensity noise below 55dB, B-weighting simulates the frequency characteristics of medium-intensity noise between 55dB and 85dB, and C-weighting simulates the frequency characteristics of high-intensity noise. Since A-weighting is closer to the hearing characteristics of the human ear, this embodiment performs A-weighting processing on the first sound pressure frequency domain data of each audio segment to obtain second sound pressure frequency domain data, which can filter out sound signal components that the human ear is not sensitive to, retain sound signal components that the human ear is relatively sensitive to, and make the detection results more accurate. Among them, the relevant processing steps of the equivalent sound pressure level data and the sound pressure level time domain data are similar to the above-mentioned application examples and will not be repeated here.

[0035] Alternatively, in some other application examples, step S300 includes: The first sound pressure frequency domain data of each audio clip is subjected to one-third octave A-weighting processing to obtain second sound pressure frequency domain data; the second sound pressure frequency domain data of each audio clip is subjected to equivalent sound pressure calculation and sound pressure level transformation to obtain equivalent sound pressure level data; the equivalent sound pressure level data corresponding to multiple audio clips are combined based on the time sequence to obtain sound pressure level time domain data of the sound pressure level changing with time.

[0036] For example, the sound frequency that the human ear can hear is about 20~20000Hz, but the sensitivity to different frequencies is different. Generally speaking, low-frequency sounds are easy to distinguish, and the higher the frequency of the sound, the more difficult it is to distinguish. Therefore, through one-third octave band A-weighted processing, the signal components in the low-frequency band can be divided more finely, and the deviation and abnormal sound that affect the human ear's hearing can be analyzed more accurately.

[0037] S400, performing Fourier transform on the time-domain data of the sound pressure level to obtain frequency-domain data of the sound pressure level in which the sound pressure level varies with the modulation frequency; For example, for the sound pressure level time domain data, there is no significant difference between a good motor and a defective motor with motor runout noise, and as mentioned above, the defective motor with motor runout noise will have a modulation signal of a specific frequency during operation. Therefore, the sound pressure level time domain data needs to be converted into sound pressure level frequency domain data.

[0038] S500: Determine whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and a preset yaw abnormal sound threshold.

[0039] For example, please refer to Figure 5 The frequency domain data of sound pressure level can be presented as the change curve of sound pressure level-frequency (also known as the motor abnormal sound modulation curve). In the study, it was found that the change curve of the good motor (such as Figure 5 Compared with the defective motor with abnormal yaw noise (as shown in the figure marked Q1), the change curve of the defective motor with abnormal yaw noise (as shown in the figure marked Q1) is Figure 5 There will be obvious peaks, such as Figure 5 As shown by P1 in the figure. By setting a reasonable threshold of abnormal sound, such as Figure 5 As shown by the mark Y1, it is possible to determine whether the target motor has yaw abnormal sound in the sound pressure level frequency domain data. For example, when there is data exceeding the yaw abnormal sound threshold in the sound pressure level frequency domain data, it indicates that the target motor has yaw abnormal sound.

[0040] Since the human ear has different sensitivities to sounds of different frequencies, different thresholds may be set for different frequencies. Thus, step S500 includes: Determine whether the target motor has yaw noise based on the sound pressure level frequency domain data and the yaw noise thresholds at different frequencies.

[0041] For example, there are three very prominent peaks in the motor noise modulation curve shown in the figure (such as Figure 5 The horizontal coordinates of these peaks are the modulation frequency (i.e., the yaw frequency) of the target motor, and the vertical coordinates are the amplitude of the sound pressure level change at the modulation frequency. Different products may have one or more modulation frequencies. Therefore, the corresponding thresholds can be set according to different frequency bands (such as Figure 5 As shown in the figure with marks Y1, Y2 and Y3, the modulation frequency of the motor's yaw noise can be detected more accurately and effectively.

[0042] In this way, the audio sampling signal is framed through short-time Fourier transform, and the equivalent sound pressure calculation and sound pressure level transformation are performed on the first sound pressure frequency domain data obtained by the frame processing, so as to obtain the sound pressure level time domain data representing the sound of the target motor. The sound pressure level time domain data is Fourier transformed to obtain the sound pressure level frequency domain data. Combined with the yaw abnormal sound threshold, it is possible to accurately and effectively detect whether the target motor has yaw abnormal sound.

[0043] Please refer to Figure 6 This embodiment also provides a motor deflection abnormal sound detection device, including an audio sampling module 110, a first transformation module 120, a second transformation module 130, a third transformation module 140 and a determination module 150.

[0044] The audio sampling module 110 is used to start the target motor and collect the sound signal emitted by the target motor to obtain an audio sampling signal; The first transformation module 120 is used to perform short-time Fourier transform on the audio sampling signal based on a preset frame duration to obtain first sound pressure frequency domain data of multiple audio clips; The second transformation module 130 is used to combine the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the plurality of audio clips based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time; The third transformation module 140 is used to perform Fourier transformation on the sound pressure level time domain data to obtain sound pressure level frequency domain data in which the sound pressure level changes with the modulation frequency; The determination module 150 is used to determine whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and a preset yaw abnormal sound threshold.

[0045] It should be noted that the inventive concept of the embodiment of the motor yaw abnormal sound detection device is the same as the inventive concept of the above-mentioned motor yaw abnormal sound detection method embodiment. The contents not covered in the embodiment of the motor yaw abnormal sound detection device can be referred to the above-mentioned motor yaw abnormal sound detection method embodiment, which will not be repeated here. The audio sampling signal is framed by short-time Fourier transform, and the first sound pressure frequency domain data obtained by the frame processing is subjected to equivalent sound pressure calculation and sound pressure level transformation, so that the sound pressure level time domain data representing the target motor sound can be obtained. The sound pressure level time domain data is subjected to Fourier transform, so that the sound pressure level frequency domain data can be obtained. Combined with the yaw abnormal sound threshold, it can be accurately and effectively detected whether the target motor has yaw abnormal sound.

[0046] Please refer to Figure 7 This embodiment also provides a test device, including a processor 210 and a memory 220, wherein the memory 220 stores a computer program, and the processor 210 is used to implement the above-mentioned motor deflection abnormal sound detection method when running the computer program. Among them, the motor deflection abnormal sound detection method can be referred to above, and will not be repeated here. The audio sampling signal is subjected to frame processing by short-time Fourier transform, and the first sound pressure frequency domain data obtained by the frame processing is subjected to equivalent sound pressure calculation and sound pressure level transformation, so that the sound pressure level time domain data representing the target motor sound can be obtained, and the sound pressure level time domain data is subjected to Fourier transform to obtain the sound pressure level frequency domain data, and combined with the deflection abnormal sound threshold, it can accurately and effectively detect whether the target motor has deflection abnormal sound.

[0047] This embodiment also provides a storage medium, in which a computer program is stored, and when the computer program is executed, the above-mentioned motor deflection abnormal sound detection method is implemented. Among them, the motor deflection abnormal sound detection method can be referred to above, and will not be repeated here. The audio sampling signal is subjected to frame processing by short-time Fourier transform, and the first sound pressure frequency domain data obtained by the frame processing is subjected to equivalent sound pressure calculation and sound pressure level transformation, so that the sound pressure level time domain data representing the target motor sound can be obtained, and the sound pressure level time domain data is subjected to Fourier transform to obtain the sound pressure level frequency domain data, and combined with the deflection abnormal sound threshold, it can be accurately and effectively detected whether the target motor has deflection abnormal sound.

[0048] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A method for detecting abnormal noise caused by motor deflection. It is characterized in that include: Starting a target motor and collecting a sound signal emitted by the target motor to obtain an audio sampling signal; Based on a preset frame duration, performing a short-time Fourier transform on the audio sampling signal to obtain first sound pressure frequency domain data of a plurality of audio clips; Combining equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on time sequence to obtain sound pressure level time domain data of the sound pressure level changing with time; Performing Fourier transformation on the sound pressure level time domain data to obtain sound pressure level frequency domain data showing that the sound pressure level varies with the modulation frequency; It is determined whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and a preset yaw abnormal sound threshold.

2. The motor deflection abnormal noise detection method according to claim 1, It is characterized in that The step of performing short-time Fourier transform on the audio sampling signal based on a preset frame duration to obtain first sound pressure frequency domain data of a plurality of audio clips includes: Based on a preset frame duration, the audio sampling signal is framed to obtain a plurality of audio segments; Perform Fourier transform on each of the audio segments to obtain first sound pressure frequency domain data of the multiple audio segments.

3. The motor deflection abnormal noise detection method according to claim 1, It is characterized in that The step of performing short-time Fourier transform on the audio sampling signal based on a preset frame duration to obtain first sound pressure frequency domain data of a plurality of audio clips includes: Based on a preset frame duration and a frame overlap duration, the audio sampling signal is framed to obtain a plurality of audio segments; Perform Fourier transform on each of the audio segments to obtain first sound pressure frequency domain data of the multiple audio segments.

4. The motor deflection noise detection method according to claim 1, It is characterized in that The step of combining the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time includes: Performing equivalent sound pressure calculation and sound pressure level conversion on the first sound pressure frequency domain data of each of the audio segments to obtain equivalent sound pressure level data; The equivalent sound pressure level data corresponding to the plurality of audio clips are combined based on a time sequence to obtain time-domain sound pressure level data showing a sound pressure level varying with time.

5. The motor deflection noise detection method according to claim 1, It is characterized in that The step of combining the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time includes: Performing weighting processing on the first sound pressure frequency domain data of each of the audio segments to obtain second sound pressure frequency domain data; Performing equivalent sound pressure calculation and sound pressure level conversion on the second sound pressure frequency domain data of each of the audio segments to obtain equivalent sound pressure level data; The equivalent sound pressure level data corresponding to the plurality of audio clips are combined based on a time sequence to obtain time-domain sound pressure level data showing a sound pressure level varying with time.

6. The motor deflection abnormal noise detection method according to claim 1, It is characterized in that The step of combining the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on the time sequence to obtain the sound pressure level time domain data of the sound pressure level changing with time includes: Performing one-third octave band A-weighting processing on the first sound pressure frequency domain data of each of the audio clips to obtain second sound pressure frequency domain data; Performing equivalent sound pressure calculation and sound pressure level conversion on the second sound pressure frequency domain data of each of the audio segments to obtain equivalent sound pressure level data; The equivalent sound pressure level data corresponding to the plurality of audio clips are combined based on a time sequence to obtain time-domain sound pressure level data showing a sound pressure level varying with time.

7. The motor deflection abnormal noise detection method according to any one of claims 1 to 6, It is characterized in that The determining whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and a preset yaw abnormal sound threshold comprises: It is determined whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and the yaw abnormal sound thresholds at different frequencies.

8. A motor deflection noise detection device, It is characterized in that include: An audio sampling module, used for starting a target motor and collecting a sound signal emitted by the target motor to obtain an audio sampling signal; A first transformation module, configured to perform a short-time Fourier transform on the audio sampling signal based on a preset frame duration to obtain first sound pressure frequency domain data of a plurality of audio segments; A second transformation module is used to combine the equivalent sound pressure level data obtained by equivalent sound pressure calculation and sound pressure level transformation of the first sound pressure frequency domain data of the multiple audio clips based on time sequence to obtain sound pressure level time domain data of the sound pressure level changing with time; A third transformation module is used to perform Fourier transformation on the sound pressure level time domain data to obtain sound pressure level frequency domain data in which the sound pressure level changes with the modulation frequency; The determination module is used to determine whether the target motor has a yaw abnormal sound according to the sound pressure level frequency domain data and a preset yaw abnormal sound threshold.

9. A test device comprising a processor and a memory, wherein the memory stores a computer program. It is characterized in that When the processor runs the computer program, it is used to implement the motor runout abnormal sound detection method as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored therein. It is characterized in that When the computer program is executed, the motor runout abnormal noise detection method as described in any one of claims 1 to 7 is implemented.

Images