Optimal microphone layout method for fault diagnosis of air door actuator of automobile air conditioner

Abstract

The invention provides an optimal microphone layout method for fault diagnosis of an automobile air conditioner air door actuator, and relates to the technical field of automobile air conditioner air door actuators, the automobile air conditioner air door actuator is taken as a circle center in advance, one-way microphones are arranged on concentric circles with different diameters, and an acquisition port of each one-way microphone faces outwards in the radial direction; comprising the following steps: S1, acquiring sound signal time domain data collected by a one-way microphone; s2, performing empirical mode decomposition on the signal time domain data at different positions; s3, solving a reconstructed signal envelope line through Hilbert transform; and S4, determining the optimal microphone layout according to the characteristic parameters of the envelope. The sound is collected through the one-way microphone, the collected sound is relatively clearer, the reliability of the optimal microphone layout is improved through empirical mode decomposition and Hilbert transform and finally analysis according to characteristic parameters, and then the optimal microphone layout is confirmed.

Description

Technical Field

[0001] The invention relates to the technical field of automobile microphone layout, and in particular to an optimal microphone layout method for fault diagnosis of an automobile air-conditioning damper actuator. Background Art

[0002] The automobile air conditioner damper actuator comprises a housing, which is provided with a mounting cavity with an upper opening, and has a gear set and a driving motor inside. A worm for driving the gear set to operate is connected to the output shaft of the driving motor, and a worm sleeve is installed.

[0003] In the prior art, there is a method of judging whether a car air-conditioning damper actuator is faulty by listening to the sound it makes when it is working. However, it is not easy to directly collect vibration information from the internal structure of the car air-conditioning damper actuator to judge whether a fault occurs. Therefore, it is necessary to collect the sound from the outside.

[0004] However, when acquiring sound from the outside, the sound position information acquired at different positions is also different, and it is impossible to accurately obtain the best sound collection position, and thus it is impossible to obtain the best feature ratio; and if multiple positions are set for sound collection, the acquired sound is often unreliable, and these sound signals will increase the difficulty of fault diagnosis of automobile air-conditioning damper actuators. Summary of the invention

[0005] Technical problem to be solved by the present invention In the prior art, obtaining sound information through multiple positions is often unreliable, so these sound signals will increase the difficulty of diagnosing automobile air-conditioning damper actuator faults. In order to overcome the above defects of the prior art, the present invention provides an optimal microphone layout method for automobile air-conditioning damper actuator fault diagnosis.

[0006] The present invention provides an optimal microphone layout method for fault diagnosis of an automobile air-conditioning damper actuator. The method preliminarily takes the automobile air-conditioning damper actuator as the center, arranges unidirectional microphones on concentric circles of different diameters, and the collection port of each unidirectional microphone faces radially outward, and comprises the following steps:

[0007] S1. Obtain the time domain data of the sound signals collected by all unidirectional microphones;

[0008] S2, decomposing the signal time domain data at different positions into multiple intrinsic mode functions and a residual based on the empirical mode, and reconstructing the signal by summarizing a certain number of the most relevant intrinsic mode functions by determining the threshold of the correlation value;

[0009] S3, obtaining the envelope of the reconstructed signal through Hilbert transform;

[0010] S4. Determine the optimal microphone layout according to the time domain characteristics and frequency domain characteristic parameters of the envelope.

[0011] Compared with the prior art, the optimal microphone layout method for automobile air-conditioning damper actuator fault diagnosis in the present application has the following advantages: using a unidirectional microphone to collect sound in one direction, reducing noise and interference from other directions, and through empirical mode decomposition and Hilbert transform, and finally analyzing according to characteristic parameters, the sound obtained is relatively clearer, thereby improving the reliability of the optimal microphone layout and confirming the optimal microphone layout.

[0012] In a possible implementation, step S2 specifically includes the following steps:

[0013] S21, decomposition based on the principle of empirical mode decomposition;

[0014] Find all the maximum points of the sound signal time domain data X(t), and use the cubic spline interpolation function to fit the upper envelope of the original data; find all the minimum points of the sound signal time domain data X(t), and fit all the minimum points through the cubic spline interpolation function to form the lower envelope of the data; the average value of the upper envelope and the lower envelope is m1(t), and the difference between the sound signal time domain data X(t) and the average value m1(t) is h1(t);

[0015] X(t)-m1(t)=h1(t)

[0016] S22, according to the empirical mode decomposition principle of step S21, the signal time domain data X(t) at different positions is decomposed into multiple intrinsic mode functions c based on the empirical mode decomposition. i (t) and a residual r n (t);

[0017] The initial residual r1(t) is:

[0018] X(t)-c1(t)=r1(t)

[0019] Wherein, c1(t) is the component with the highest evaluation rate of the sound signal time domain data X(t);

[0020] The residual is screened multiple times by setting the decomposition level n, and the result is:

[0021] r n-1 (t)-c n (t) = r n (t)

[0022]

[0023] S23, propose and calculate the original sound signal time domain data X(t) and the related intrinsic mode function c i (t) and the unrelated intrinsic mode functions c i(t), determine the threshold of the correlation value, and use the threshold to summarize a certain number of the most relevant intrinsic mode functions c i (t) to reconstruct the signal:

[0024] x(t)=Σc i (t).

[0025] Compared with existing technologies, EMD has unique advantages in processing real-time signals.

[0026] In a possible implementation, in step S21, if the waveform of the sound signal time domain data obtained in step S1 is asymmetrical, then after executing step S21, the average value of the upper envelope and the lower envelope obtained is used as the new sound signal time domain data, and step S21 is repeated k times until the waveform of the average value of the upper envelope and the lower envelope obtained is symmetrical, until h 1k (t) becomes an eigenmode function:

[0027] h 1(k-1) (t)-m 1k (t) = h 1k (t).

[0028] In a possible implementation, the standard deviation SD calculated from two consecutive iterations is:

[0029]

[0030] Among them, the standard deviation SD value is between 0.2 and 0.3.

[0031] Compared with the prior art, the mean area of ​​the upper envelope and the lower envelope is prevented from being infinitesimal, thereby eliminating the amplitude fluctuation in the physical sense.

[0032] In a possible implementation manner, step S3 specifically includes the following steps:

[0033] S31, calculating the analytical signal of the reconstructed signal:

[0034]

[0035] Where x(t) is the reconstructed signal;

[0036] The analytical signal y(t) of the reconstructed signal x(t) is:

[0037]

[0038] in is the instantaneous amplitude, is the instantaneous phase;

[0039] 32. Take the modulus of the analytical signal to get the envelope:

[0040] z(t)=|y(t)|=|a(t)exp[jθ(t)]|.

[0041] In a possible implementation manner, step S4 specifically includes the following steps:

[0042] S41, calculating the time domain characteristic and frequency domain characteristic parameters of the envelope, and sorting them from large to small respectively;

[0043] Time domain characteristics:

[0044] Mean:

[0045]

[0046] Standard Deviation:

[0047]

[0048] RMS:

[0049]

[0050] Frequency domain characteristics:

[0051] Frequency domain amplitude average:

[0052]

[0053] Center of gravity frequency:

[0054]

[0055] S42, summing the rankings obtained from the time domain features and frequency domain feature parameters of each position;

[0056] S43. Sort the positions in ascending order of sum to determine the best microphone layout. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A structural diagram of an optimal microphone layout method for fault diagnosis of an automobile air-conditioning damper actuator according to the present invention;

[0058] Figure 2 The present invention is a flow chart of the optimal microphone layout method for automobile air conditioning damper actuator fault diagnosis. DETAILED DESCRIPTION

[0059] First, those skilled in the art should understand that these implementations are only used to explain the technical principles of the embodiments of the present application, and are not intended to limit the protection scope of the embodiments of the present application. Those skilled in the art can make adjustments to them as needed to adapt to specific application scenarios.

[0060] The present application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0061] See also Figure 1-2 As shown, an embodiment of the present application discloses an optimal microphone layout method for fault diagnosis of an automobile air-conditioning damper actuator. Unidirectional microphones are pre-arranged on concentric circles of different diameters with the automobile air-conditioning damper actuator as the center, and the collection port of each unidirectional microphone faces radially outward.

[0062] Among them, a unidirectional microphone, also known as a directional microphone, is a type of microphone used to mainly receive sound from one direction. This type of microphone is usually used to reduce noise and interference from other directions.

[0063] Specifically: Consider the entire automobile air-conditioning damper actuator as a whole, and place unidirectional microphones around the automobile air-conditioning damper actuator. The automobile air-conditioning damper actuator is placed at a hypothetical center of a circle, radiating in a circular shape, with a different radiation radius each time, that is, each time the radiation directional microphone is arranged differently and the sound is collected as a set of data multiple times, all pointing to the center of the circle, to ensure that each set of sound data at each position is recorded.

[0064] In this embodiment, an optimal microphone layout method for fault diagnosis of an automobile air conditioner damper actuator specifically includes the following steps:

[0065] S1. Obtain the time domain data X(t) of the sound signals collected by all unidirectional microphones.

[0066] S2. The signal time domain data X(t) at different positions is decomposed into multiple intrinsic mode functions and a residual based on the empirical mode, and a certain number of the most relevant intrinsic mode functions are aggregated by determining the threshold of the correlation value to reconstruct the signal.

[0067] The specific steps include:

[0068] S21, decomposition based on the principle of empirical mode decomposition;

[0069] Find all the maximum points of the sound signal time domain data X(t), and use the cubic spline interpolation function to fit the upper envelope of the original data; find all the minimum points of the sound signal time domain data X(t), and fit all the minimum points through the cubic spline interpolation function to form the lower envelope of the data; the average value of the upper envelope and the lower envelope is m1(t), and the difference between the sound signal time domain data X(t) and the average value m1(t) is h1(t);

[0070] X(t)-m1(t)=h1(t)

[0071] If the waveform of the sound signal time domain data obtained in step S1 is asymmetric, it does not meet the condition of decomposing into an IMF. After executing step S21, the average value of the upper envelope and the lower envelope is taken as the new sound signal time domain data, and step S21 is repeated k times until the waveform of the average value of the upper envelope and the lower envelope is symmetric. 1k (t) becomes an eigenmode function:

[0072] h 1(k-1) (t)-m 1k (t) = h 1k (t)

[0073] When the mean of the upper and lower envelopes tends to infinity, the amplitude fluctuations in the physical sense may be eliminated. Therefore, the screening process should be used with caution. Because when the mean of the upper and lower envelopes tends to infinity, the resulting intrinsic mode function (IMF) may become a constant amplitude signal of pure frequency modulation. In order to ensure that the modal function (IMF) retains sufficient physical meaning in amplitude and frequency modulation, it is necessary to determine an iterative stopping criterion, which can be achieved by limiting the size of the standard deviation SD calculated from the results of two consecutive iterations:

[0074]

[0075] The standard deviation SD value is between 0.2 and 0.3. For screening programs, an SD value of 0.2-0.3 is a very tight limit on the variability between screenings.

[0076] S22, according to the empirical mode decomposition principle of step S21, the signal time domain data X(t) at different positions is decomposed into multiple intrinsic mode functions c based on the empirical mode decomposition. i (t) and a residual r n (t).

[0077] The initial residual r1(t) is:

[0078] X(t)-c1(t)=r1(t)

[0079] Among them, c1(t) is the component with the highest rate of the sound signal time domain data X(t); c1(t) is separated from the remaining data in the above manner.

[0080] That is, using the residual r n (t) is used to replace the difference h1(t), and c1(t) is used to replace the average value m1(t).

[0081] Since the residual r1(t) still contains more information about the frequency components, it is regarded as a new signal to be decomposed and the screening process is repeated again. This procedure can be repeated for all subsequent r j (t), the result is:

[0082] r n-1 (t)-c n (t) = r n (t)

[0083] The above screening process can be stopped by setting the decomposition level n. "X(t)" is first decomposed into multiple intrinsic mode functions "c i (t)” and residual “r n (t)”:

[0084]

[0085] Among them, c i (t) is the eigenmode function, r n (t) is the residual.

[0086] S23. Considering that the original signal has a better correlation with the correlated IMF than with the uncorrelated IMF, it is proposed that there is a direct relationship between correlation and relevance, that is, the higher the correlation value, the higher the relevance.

[0087] A threshold value of the correlation value is determined, and then the relevant and irrelevant IMFs are distinguished according to the threshold value. Then, a certain number of the most relevant IMFs are summed up by the threshold value to reconstruct the signal.

[0088] x(t)=∑c i (t)

[0089] S3. Obtain the envelope of the reconstructed signal through Hilbert transform.

[0090] The specific steps include:

[0091] S31, calculating the analytical signal of the reconstructed signal:

[0092] The Hilbert transform is equivalent to an orthogonal filter

[0093]

[0094] Where x(t) is the reconstructed signal;

[0095] The analytical signal y(t) of the reconstructed signal x(t) is:

[0096]

[0097] in is the instantaneous amplitude, is the instantaneous phase;

[0098] 32. Take the modulus of the analytical signal to get the envelope:

[0099] z(t)=|y(t)|=|a(t)exp[jθ(t)]|.

[0100] S4. Determine the optimal microphone layout according to the time domain characteristics and frequency domain characteristic parameters of the envelope.

[0101] The specific steps include:

[0102] S41, calculating the time domain characteristic and frequency domain characteristic parameters of the envelope, and sorting them from large to small respectively;

[0103] Time domain characteristics:

[0104] Mean:

[0105]

[0106] Standard Deviation:

[0107]

[0108] RMS:

[0109]

[0110] Frequency domain characteristics:

[0111] Frequency domain amplitude average:

[0112]

[0113] Center of gravity frequency:

[0114]

[0115] S42, summing the rankings obtained from the time domain features and frequency domain feature parameters of each position;

[0116] S43. Sort the positions in ascending order of sum to determine the best microphone layout.

[0117] A unidirectional microphone is used to collect sound from one direction, reducing noise and interference from other directions. Empirical mode decomposition and Hilbert transform are used, and finally analysis is performed based on characteristic parameters. The resulting sound is relatively clearer, which improves the reliability of the optimal microphone layout and thus confirms the optimal microphone layout.

[0118] For example, if three unidirectional microphones are set around the car air conditioner damper actuator, each unidirectional microphone has two characteristic parameters a and b. Calculated by the above formula:

[0119] The a and b of the sound signal collected by the unidirectional microphone at the first position are 0.2 and 0.3 respectively;

[0120] The a and b of the sound signal collected by the unidirectional microphone at the second position are 0.1 and 0.2 respectively;

[0121] The a and b of the sound signal collected by the unidirectional microphone at the third position are 0.3 and 0.4 respectively;

[0122] It can be seen that the parameter a is ranked from large to small: 0.3, 0.2, 0.1

[0123] Parameter b from large to small: 0.4, 0.3, 0.2

[0124] Then the rankings of a and b of the unidirectional microphone in the first position are 2 and 2 respectively; the sum of the rankings is 4.

[0125] The rankings of a and b of the unidirectional microphone in the second position are 3 and 3 respectively; the sum of the rankings is 6.

[0126] The rankings of a and b of the unidirectional microphone in the third position are 1 and 1 respectively; the sum of the rankings is 2.

[0127] Finally, the rankings are in ascending order of the sum: 2, 4, and 6. That is, the smallest sum is 2, which ranks first. It can be determined that the microphone in the third position is the best microphone layout.

[0128] By analogy, unidirectional microphones can also be set up in other locations. Through multiple calculations and sorting, the optimal microphone layout position can be obtained. The sound signal collected by this optimal microphone layout position is more convenient, has richer information, and has more obvious characteristics than the signal collected from the inside, which can significantly reduce the difficulty of fault diagnosis of the air conditioning damper actuator.

[0129] This application proposes a method for determining the optimal microphone layout by using time domain characteristics and frequency domain characteristic parameters of the reconstructed signal envelope based on empirical mode decomposition (EMD) and Hilbert transform, which can improve the reliability of the optimal microphone layout. EMD can decompose complex signals into several intrinsic mode functions (IMFs), which represent different frequency components of the signal. Compared with other signal processing methods (such as Fourier transform, wavelet transform, etc.), EMD has unique advantages in processing real-time signals.

[0130] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "in the present embodiment", "specific example", or "some examples" etc. means that the specific features, mechanisms, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, mechanisms, materials or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0131] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. An optimal microphone layout method for fault diagnosis of an automobile air-conditioning damper actuator, which pre-arranges unidirectional microphones on concentric circles of different diameters with the automobile air-conditioning damper actuator as the center, and the collection port of each unidirectional microphone faces radially outward, comprising the following steps: S1. Obtain the time domain data of the sound signals collected by all unidirectional microphones; S2, decomposing the signal time domain data at different positions into multiple intrinsic mode functions and a residual based on the empirical mode, and reconstructing the signal by summarizing a certain number of the most relevant intrinsic mode functions by determining the threshold of the correlation value; S3, obtaining the envelope of the reconstructed signal through Hilbert transform; S4. Determine the optimal microphone layout according to the time domain characteristics and frequency domain characteristic parameters of the envelope.

2. The optimal microphone layout method for automobile air-conditioning damper actuator fault diagnosis according to claim 1 is characterized in that: In step S2, the following steps are specifically included: S21, decomposition based on the principle of empirical mode decomposition; Find all the maximum points of the sound signal time domain data X(t), and use the cubic spline interpolation function to fit the upper envelope of the original data; find all the minimum points of the sound signal time domain data X(t), and fit all the minimum points through the cubic spline interpolation function to form the lower envelope of the data; the average value of the upper envelope and the lower envelope is m1(t), and the difference between the sound signal time domain data X(t) and the average value m1(t) is h1(t); X(t)-m1(t)=h1(t) S22, according to the empirical mode decomposition principle of step S21, the signal time domain data X(t) at different positions is decomposed into multiple intrinsic mode functions c based on the empirical mode decomposition. i (t) and a residual r n (t); The initial residual r1(t) is: X(t)-c1(t)=r1(t) Wherein, c1(t) is the component with the highest evaluation rate of the sound signal time domain data X(t); The residual is screened multiple times by setting the decomposition level n, and the result is: r n-1 (t)-c n (t)=r n (t) S23, propose and calculate the original sound signal time domain data X(t) and the related intrinsic mode function c i (t) and the unrelated intrinsic mode functions c i (t), determine the threshold of the correlation value, and use the threshold to summarize a certain number of the most relevant intrinsic mode functions c i (t) to reconstruct the signal: x(t)=∑c i (t)。 3. The optimal microphone layout method for automobile air-conditioning damper actuator fault diagnosis according to claim 2 is characterized in that: If the waveform of the sound signal time domain data obtained in step S1 is asymmetrical, then after executing step S21, the average value of the upper envelope and the lower envelope is taken as the new sound signal time domain data, and step S21 is repeated k times until the waveform of the average value of the upper envelope and the lower envelope is symmetrical, h 1k (t) becomes an eigenmode function: h 1(k-1) (t)-m 1k (t)=h 1k (t)。 4. The optimal microphone layout method for automobile air conditioner damper actuator fault diagnosis according to claim 3 is characterized in that: The standard deviation SD is calculated from two consecutive iterations: Among them, the standard deviation SD value is between 0.2 and 0.

3.

5. The optimal microphone layout method for automobile air-conditioning damper actuator fault diagnosis according to claim 2 is characterized in that: In step S3, the following steps are specifically included: S31, calculating the analytical signal of the reconstructed signal: Where x(t) is the reconstructed signal; The analytical signal y(t) of the reconstructed signal x(t) is: in is the instantaneous amplitude, is the instantaneous phase; 32. Take the modulus of the analytical signal to get the envelope: z(t)=|y(t)|=|a(t)exp[jθ(t)]|.

6. The optimal microphone layout method for fault diagnosis of automobile air conditioner damper actuator according to claim 5 is characterized in that: In step S4, the following steps are specifically included: S41, calculating the time domain characteristic and frequency domain characteristic parameters of the envelope, and sorting them from large to small respectively; Time domain characteristics: Mean: Standard Deviation: RMS: Frequency domain characteristics: Frequency domain amplitude average: Center of gravity frequency: S42, summing the rankings obtained from the time domain features and frequency domain feature parameters of each position; S43. Sort the positions in ascending order of sum to determine the best microphone layout.

Images