Air conditioner abnormal sound positioning method, air conditioner and storage medium

Abstract

The invention provides an air conditioner abnormal sound positioning method, an air conditioner and a storage medium. The method comprises the steps that abnormal sound signals collected by a microphone array are obtained; performing noise reduction and enhancement processing on the abnormal sound signal collected by each microphone in the microphone array to obtain a pure abnormal sound signal corresponding to each microphone; performing abnormal sound source positioning operation according to the pure abnormal sound signal; wherein the step of carrying out noise reduction and enhancement processing on the abnormal sound signals collected by each microphone in the microphone array to obtain the pure abnormal sound signals corresponding to each microphone comprises the following steps of: carrying out framing processing on time domain signals of the abnormal sound signals, and converting the time domain signals into Hankel matrixes; performing iterative calculation on the Hankel matrix by using a preset bilateral random projection theory method to obtain a low-rank matrix, and obtaining an estimated pure abnormal sound signal matrix; and performing time domain conversion on the pure abnormal sound signal matrix to obtain a pure abnormal sound signal. According to the invention, the abnormal sound detection precision under complex background noise can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioners, and in particular to a method for locating abnormal sounds of air conditioners, an air conditioner using the method for locating abnormal sounds of air conditioners, and a computer-readable storage medium using the method for locating abnormal sounds of air conditioners. Background Art

[0002] The primary goal of the air conditioner abnormal noise detection system, acoustic signal enhancement system, and abnormal noise location system is to achieve efficient detection, enhancement, and location of abnormal noise during air conditioner operation through the coordinated operation of software and hardware. Abnormal noise is an abnormal sound produced by the air conditioner during operation due to mechanical failure, component aging, or other abnormal conditions. These abnormal sounds not only affect the user experience but may also indicate potential equipment failures, requiring timely detection and resolution.

[0003] Traditional methods for detecting abnormal noise rely primarily on manual auscultation or simple acoustic sensors. However, due to the strong background noise (such as wind, compressor vibration, and electrical current) during air conditioner operation, abnormal noise is often masked, making it difficult to accurately detect and locate. Furthermore, air conditioners only operate for a short time under specific operating conditions (such as defrost mode), requiring long wait times before testing, which is inefficient. Furthermore, when abnormal noise occurs, manual investigation is required to gradually identify the source of the noise and determine if it is a structural issue. However, each investigation requires redoing the experiment, which is inefficient.

[0004] In an existing method for sound source localization, an audio signal received by a microphone array is obtained, wherein the microphone array includes multiple microphones and the audio signal includes a speech signal and a noise signal; a relative time difference between the audio signals received by any two microphones among the multiple microphones is obtained based on the audio signal and a preset parameter, wherein the preset parameter is set according to the signal-to-noise ratio of the audio signal; and the audio signal is located based on the relative time difference between the audio signal reaching any two microphones among the multiple microphones and the position of each microphone.

[0005] However, this solution solves the problem of voice control but does not solve the problem of locating abnormal noises from the air conditioner.

[0006] Therefore, a more optimized method for locating abnormal noises of air conditioners needs to be considered. Summary of the Invention

[0007] The first object of the present invention is to provide a method for locating abnormal noise of an air conditioner, which improves the detection accuracy of abnormal noise under complex background noise.

[0008] A second object of the present invention is to provide an air conditioner that improves the accuracy of detecting abnormal sounds under complex background noise.

[0009] A third object of the present invention is to provide a computer-readable storage medium that improves the accuracy of abnormal sound detection under complex background noise.

[0010] In order to achieve the above-mentioned first purpose, the method for locating abnormal noise of an air conditioner provided by the present invention includes: obtaining abnormal noise signals collected by a microphone array; performing noise reduction and enhancement processing on the abnormal noise signals collected by each microphone in the microphone array to obtain a pure abnormal noise signal corresponding to each microphone; and performing an abnormal noise source location operation based on the pure abnormal noise signal; wherein, the step of performing noise reduction and enhancement processing on the abnormal noise signals collected by each microphone in the microphone array to obtain a pure abnormal noise signal corresponding to each microphone includes: performing frame processing on the time domain signal of the abnormal noise signal and converting the time domain signal of each frame into a Hankel matrix; using a preset bilateral random projection theory method to iteratively calculate a low-rank matrix of the Hankel matrix to obtain an estimated pure abnormal noise signal matrix; and performing time domain conversion on the pure abnormal noise signal matrix to obtain a pure abnormal noise signal.

[0011] As can be seen from the above scheme, the air conditioner abnormal sound localization method of the present invention uses the abnormal sound signal as a low-rank matrix and adaptively obtains the rank value of each frame. Using the bilateral random projection theory method, the low-rank matrix is ​​further approximated to a pure speech matrix and converted into a pure speech signal matrix. This effectively removes background noise, extracts a pure abnormal sound signal, and improves the accuracy of abnormal sound detection in complex background noise environments.

[0012] In a further solution, the low-rank matrix L is obtained by the following formula: , where Y1 and Y2 are the bilateral projection matrices of the Hankel matrix X, , A1 and A2 are random matrices.

[0013] It can be seen that the bilateral random projection theory method is used to perform bilateral projection processing on the Hankel matrix, which can retain the main structural information of the abnormal sound signal and improve the detection accuracy.

[0014] In a further solution, the steps of iteratively calculating the low-rank matrix of the Hankel matrix using the preset bilateral random projection theory method include: optimizing the bilateral projection to a single-sided projection, and the optimized low-rank matrix Obtained by the following formula: .

[0015] It can be seen from this that by optimizing the bilateral projection to a single-sided projection, the optimized low-rank matrix only needs to calculate the single-sided projection matrix, which greatly reduces the computational complexity of the matrix and improves the computational efficiency of the algorithm.

[0016] In a further embodiment, the microphone array includes at least three microphones, with one microphone provided on at least one of the upper and lower sides in the vertical direction of the air conditioner, and one microphone provided on at least one of the left and right sides in the horizontal direction of the air conditioner; the step of locating the source of the abnormal noise based on the pure abnormal noise signal includes: calculating a first phase difference between the abnormal noise source and the two microphones based on the pure abnormal noise signals from one microphone in the horizontal direction, and calculating a horizontal angle of incidence of the abnormal noise source in the coordinate system based on the first phase difference; calculating a second phase difference between the abnormal noise source and the two microphones based on the pure abnormal noise signals from one microphone in the vertical direction, and calculating a horizontal angle of incidence of the abnormal noise source in the coordinate system based on the second phase difference; determining a three-dimensional direction vector of the abnormal noise source based on the horizontal angle of incidence and the vertical angle of incidence; obtaining a time difference between the sound propagating from the abnormal noise source to any two microphones, and obtaining a distance from the abnormal noise source to the origin of the coordinate system based on the time difference and the speed of sound; and obtaining the spatial coordinates of the abnormal noise source based on the distance from the abnormal noise source to the origin of the coordinate system and the three-dimensional direction vector.

[0017] It can be seen that when locating the source of abnormal noise based on pure abnormal noise signals, the phase difference of the abnormal noise frequency propagating to different microphones is calculated, and then the horizontal incident angle and vertical incident angle of the abnormal noise source are calculated. The distance from the abnormal noise source to the origin of the coordinate system is obtained based on the time difference of the sound source signal being transmitted to different microphones, and the spatial point coordinates of the abnormal noise source are determined, so that the position of the abnormal noise of the air conditioner can be located, thereby improving the efficiency of air conditioner troubleshooting.

[0018] In a further solution, the microphone array includes four microphones, one microphone is respectively arranged on the upper and lower sides of the air conditioner in the vertical direction, and one microphone is respectively arranged on the left and right sides of the air conditioner in the horizontal direction; the step of performing an abnormal sound source locating operation based on the pure abnormal sound signal includes: calculating a first phase difference between the abnormal sound source and the two microphones according to the pure abnormal sound signals of the two microphones in the horizontal direction, and calculating the horizontal incident angle of the abnormal sound source in the coordinate system according to the first phase difference; calculating a second phase difference between the abnormal sound source and the two microphones according to the pure abnormal sound signals of the two microphones in the vertical direction, and calculating the vertical incident angle of the abnormal sound source in the coordinate system according to the second phase difference; determining the three-dimensional direction vector of the abnormal sound source according to the horizontal incident angle and the vertical incident angle; obtaining the time difference between the sound propagating from the abnormal sound source to any two microphones, obtaining the distance from the abnormal sound source to the origin of the coordinate system according to the time difference and the speed of sound, and obtaining the spatial coordinates of the abnormal sound source according to the distance from the sound source to the microphones and the three-dimensional direction vector.

[0019] It can be seen that by calculating the horizontal incident angle using two microphones in the horizontal direction and calculating the vertical incident angle using two microphones in the vertical direction, the detection accuracy can be further improved.

[0020] In a further solution, the step of calculating the first phase difference between the abnormal sound source and the two microphones in the horizontal direction based on the pure abnormal sound signals of the two microphones includes: calculating the phases corresponding to the two microphones in the horizontal direction based on the frequencies corresponding to the maximum amplitude points in the spectrum diagram of the pure abnormal sound signals of the two microphones in the horizontal direction, and subtracting the phases corresponding to the two microphones in the horizontal direction to obtain the first phase difference.

[0021] It can be seen that the detection efficiency can be improved by calculating the phase difference between the two microphones using the maximum amplitude point in the spectrum of the pure abnormal sound signal.

[0022] In a further embodiment, the horizontal incident angle is obtained by the following formula: , where ∆Ψ LR is the first phase difference, v is the speed of sound, f is the frequency of the sound wave, and D1 is the distance between the two microphones in the horizontal direction.

[0023] In a further solution, the step of calculating the second phase difference between the abnormal sound source and the two microphones in the vertical direction based on the pure abnormal sound signals of the two microphones includes: calculating the phases corresponding to the two microphones in the vertical direction based on the frequencies corresponding to the maximum amplitude points in the spectrum diagram of the pure abnormal sound signals of the two microphones in the vertical direction, and subtracting the phases corresponding to the two microphones in the vertical direction to obtain the second phase difference.

[0024] It can be seen that the detection efficiency can be improved by calculating the phase difference between the two microphones using the maximum amplitude point in the spectrum of the pure abnormal sound signal.

[0025] In a further embodiment, the vertical incidence angle is obtained by the following formula: , where ∆Ψ UD is the second phase difference, v is the speed of sound, f is the frequency of the sound wave, and D2 is the distance between the two microphones in the vertical direction.

[0026] In order to achieve the second purpose of the present invention, the present invention provides an air conditioner including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the steps of the above-mentioned air conditioner abnormal sound locating method are implemented.

[0027] In order to achieve the third purpose of the present invention, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above-mentioned method for locating abnormal noise of an air conditioner when the computer program is executed by a controller. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of an embodiment of a method for locating abnormal noise of an air conditioner according to the present invention.

[0029] Figure 2Schematic diagram of the position distribution of four microphones in an embodiment of the method for locating abnormal noise of an air conditioner according to the present invention.

[0030] Figure 3 This is a flowchart of the steps of performing noise reduction and enhancement processing on the abnormal sound signal collected by each microphone in the microphone array in the embodiment of the air conditioner abnormal sound localization method of the present invention to obtain the pure abnormal sound signal corresponding to each microphone.

[0031] The present invention will be further described below with reference to the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0032] The method for locating abnormal sounds of an air conditioner of the present invention is an application program applied in an air conditioner, and is used for locating abnormal sounds of the air conditioner.

[0033] Example of method for locating abnormal sound of air conditioner: like Figure 1 As shown, in this embodiment, when the method for locating abnormal sound of an air conditioner is working, step S1 is first executed to obtain the abnormal sound signal collected by the microphone array. In order to collect the abnormal sound signal of the air conditioner, it is necessary to set up a microphone array for signal collection. At the same time, in order to realize the location of the abnormal sound source, the microphone array needs to be equipped with at least three microphones. In the vertical direction of the air conditioner, at least one microphone is set on the upper and lower sides, and in the horizontal direction of the air conditioner, at least one microphone is set on the left and right sides. Preferably, as Figure 2 As shown, a spatial rectangular coordinate system is established with the center of the air conditioner as the coordinate origin. The microphone array includes four microphones, one microphone is set on the upper and lower sides of the air conditioner in the vertical direction, and one microphone is set on the left and right sides of the air conditioner in the horizontal direction. The microphones on the left and right sides of the air conditioner are located on the x-axis, the middle position of the air conditioner in the front direction is the y-axis, and the z-axis is vertically upward. In the horizontal direction, the coordinates of the microphones on the left and right sides can be expressed as x L =(-d, 0, 0) and x R =(d, 0, 0), in the vertical direction, the coordinates of the two microphones on the upper and lower sides can be expressed as x U =(0, 0, d) and x D =(0, 0, -d), the microphone can be set according to the position that the air conditioner needs to monitor. For example, it can be set on the side of the air conditioner indoor unit to locate the abnormal sound of the air conditioner indoor unit; or it can be set on the side of the air conditioner outdoor unit to locate the abnormal sound of the air conditioner outdoor unit.

[0034] After obtaining the abnormal sound signal collected by the microphone array, step S2 is executed to perform noise reduction and enhancement processing on the abnormal sound signal collected by each microphone in the microphone array to obtain a pure abnormal sound signal corresponding to each microphone. In order to reduce the noise in the abnormal sound signal and enhance the abnormal sound, it is necessary to perform noise reduction and enhancement processing on the abnormal sound signal.

[0035] In this embodiment, see Figure 3 When performing noise reduction and enhancement processing on the abnormal sound signal collected by each microphone in the microphone array to obtain the pure abnormal sound signal corresponding to each microphone, step S11 is first executed to perform frame processing on the time domain signal of the abnormal sound signal and convert the time domain signal of each frame into a Hankel matrix. For the collected time domain signal x(t) of the abnormal sound signal, it is framed to obtain the time domain signal of each frame. , and convert it into the Hankel matrix X. Determine whether the noise of the abnormal sound signal is Gaussian noise. If it is not Gaussian noise, let , Pre-whitening is performed, i.e., singular value decomposition is performed on the Hankel matrix X to obtain an orthogonal matrix Q and an upper triangular matrix R for subsequent energy threshold method (ETM) calculation. The ETM is used to calculate the preliminary rank value r of the current frame matrix, which is a well-known technique to those skilled in the art and will not be described in detail here.

[0036] After obtaining the Hankel matrix, step S12 is executed to iteratively calculate the low-rank matrix of the Hankel matrix using a preset bilateral random projection theory method to obtain an estimated pure abnormal noise signal matrix. After performing singular value decomposition on the Hankel matrix X, to ensure that the main structural information is retained when solving the low-rank matrix L of the Hankel matrix X, the preset bilateral random projection theory method is used to perform bilateral projection processing on the Hankel matrix X to calculate the low-rank matrix L of the current frame matrix when the rank is r. In this embodiment, the low-rank matrix L is obtained by the following formula: , where Y1 and Y2 are the bilateral projection matrices of the Hankel matrix X, , A1 and A2 are random matrices.

[0037] Since the bilateral random projection theory method has the problem of slow singular value decay and low computational efficiency when solving low-rank matrices, it is necessary to optimize the bilateral random projection calculation. In this embodiment, the steps of iteratively calculating the low-rank matrix on the Hankel matrix using the preset bilateral random projection theory method include: optimizing the bilateral projection to a single-sided projection, and the optimized low-rank matrix Obtained by the following formula: .

[0038] Specifically, ,but , since A1 and A2 are random matrices, we can let In summary, we can Substitution in Chinese Computing , thus After conversion, we can get the optimized low-rank matrix The calculation formula is: , the optimized low-rank matrix Only the single-sided projection matrix Y2 needs to be calculated, which greatly reduces the computational complexity of the matrix and improves the computational efficiency of the algorithm.

[0039] When using the above formula to solve the low-rank matrix of the Hankel matrix X, the inverse of the Hankel matrix X needs to be solved. Tikhonov regularization is often used to solve the inverse operation of the matrix. After the Hankel matrix X is decomposed using the optimized bilateral random projection theory method, the Hankel matrix X is truncated according to the number of singular values, so that the rank of the projection matrix Y2 may be less than the rank of the original Hankel matrix X, resulting in an insufficient rank matrix. Since Tikhonov regularization can only perform inverse operations on full-rank matrices, Tikhonov regularization is modified to expand it to solve the inverse operation of insufficient rank matrices. The modified formula is as follows: , where m is the row of the Hankel matrix X, d is the column of the Hankel matrix X, α is a positive number between 0 and 1, with a value of 0.2, and I m and I d represent the m-order identity matrix and the d-order identity matrix respectively.

[0040] When using the preset bilateral random projection theory method to iteratively calculate the low-rank matrix of the Hankel matrix, iteratively calculate the low-rank matrix starting from k=1 , select a hard threshold To update the number of non-zero elements in the sparse matrix so that the number of non-zero elements in the sparse matrix is ​​less than or equal to the hard threshold , and calculate the sparse matrix , the calculation formula is as follows: ,in, The following relationship needs to be satisfied: . Among them, the sparse matrix Represents the matrix Before extraction A collection of elements, is the Hankel matrix at the current iteration number, is the low-rank matrix at the current number of iterations, where The value of does not exceed the given value q, the extracted element value must be guaranteed to be a non-zero element, and the extracted element value must be guaranteed to be greater than the unextracted element value.

[0041] The iteration is terminated by detecting whether the change in the low-rank matrix after update is reduced to below a certain threshold or whether the number of iterations k meets the set requirements. The threshold judgment calculation formula is as follows: Where, Represents the solution matrix The square of the Frobenius norm of is the low-rank matrix under the previous iteration number, Represents the solution matrix The square of the Frobenius norm, is a given threshold.

[0042] If the iteration does not terminate, then , update the Hankel matrix and then perform the next iterative calculation.

[0043] Through such iterative operations, the low-rank matrix L is gradually optimized to achieve the best approximation of the pure abnormal sound signal matrix, thereby achieving the enhancement of the abnormal sound signal. , get the estimated pure noise signal matrix. If the noise is non-Gaussian noise, it needs to be de-whitened: let .

[0044] After obtaining the clean abnormal sound signal matrix, step S13 is executed to transform the clean abnormal sound signal matrix into the time domain to obtain the clean abnormal sound signal. Diagonal average is performed on the low-rank matrix L and converted into the enhanced time domain clean abnormal sound signal. .

[0045] After obtaining the pure abnormal sound signal corresponding to each microphone, step S3 is executed to perform abnormal sound source localization operation based on the pure abnormal sound signal. In order to quickly and accurately determine the specific location of the abnormal sound of the air conditioner, it is necessary to perform abnormal sound source localization operation on the pure abnormal sound signal.

[0046] In this embodiment, the steps of locating the source of an abnormal sound based on a pure abnormal sound signal include: calculating a first phase difference between the propagation of the abnormal sound source to one microphone and another microphone in the horizontal direction based on the pure abnormal sound signals from the two microphones, and calculating the horizontal incident angle of the abnormal sound source in the coordinate system based on the first phase difference; calculating a second phase difference between the propagation of the abnormal sound source to the two microphones based on the pure abnormal sound signals from the one microphone and another microphone in the vertical direction, and calculating the horizontal incident angle of the abnormal sound source in the coordinate system based on the second phase difference; determining a three-dimensional direction vector of the abnormal sound source based on the horizontal incident angle and the vertical incident angle; obtaining a time difference between the propagation of sound from the abnormal sound source to any two microphones, obtaining a distance from the abnormal sound source to the microphones based on the time difference and the speed of sound, and obtaining the spatial coordinates of the abnormal sound source based on the distance from the sound source to the microphones and the three-dimensional direction vector. When locating the source of abnormal noise based on pure abnormal noise signals, the phase difference of the abnormal noise frequency propagating to different microphones is calculated, and then the horizontal and vertical incident angles of the abnormal noise source are calculated. The distance from the sound source to the microphone is obtained based on the time difference of the sound source signal being transmitted to different microphones, and the spatial point coordinates of the abnormal noise source are determined. This can locate the position of the abnormal noise in the air conditioner and improve the efficiency of air conditioner troubleshooting.

[0047] Preferably, the microphone array includes four microphones, one microphone disposed on each of the upper and lower sides of the air conditioner in the vertical direction, and one microphone disposed on each of the left and right sides of the air conditioner in the horizontal direction. The step of locating the source of the abnormal noise based on the pure abnormal noise signals includes: calculating a first phase difference between the propagation of the abnormal noise source to the two microphones based on the pure abnormal noise signals from the two horizontal microphones, and calculating the horizontal angle of incidence of the abnormal noise source in the coordinate system based on the first phase difference; calculating a second phase difference between the propagation of the abnormal noise source to the two microphones based on the pure abnormal noise signals from the two vertical microphones, and calculating the vertical angle of incidence of the abnormal noise source in the coordinate system based on the second phase difference. Calculating the horizontal angle of incidence using the two horizontal microphones and the vertical angle of incidence using the two vertical microphones can further improve detection accuracy.

[0048] In this embodiment, the step of calculating the first phase difference between the abnormal sound source and the two microphones in the horizontal direction based on the pure abnormal sound signals from the two microphones includes: calculating the phases corresponding to the two microphones in the horizontal direction based on the frequencies corresponding to the maximum amplitude points in the spectrum graphs of the pure abnormal sound signals from the two microphones in the horizontal direction, and subtracting the phases corresponding to the two microphones in the horizontal direction to obtain the first phase difference. Specifically, when calculating the phases corresponding to the two microphones in the horizontal direction, a fast Fourier transform is performed on the pure abnormal sound signals to find the frequency f corresponding to the maximum amplitude points in the two spectrum graphs. The frequencies fmax corresponding to the maximum amplitudes of the two are the same, but the phases are different. The phase can be obtained by calculating the inverse tangent value of the frequency signal, that is, , where Im represents the imaginary part of the signal at the fmax frequency after the fast Fourier transform, and Re represents the real part, which can be directly extracted from the fast Fourier transform output parameters. The square root of the sum of the squares of the two is the amplitude, i.e., the amplitude at the fmax frequency. Detection efficiency can be improved by calculating the phase difference between the two microphones at the maximum amplitude point in the spectrum of the pure abnormal sound signal.

[0049] In this embodiment, the horizontal incident angle is obtained by the following formula: , where ∆Ψ LR is the phase difference between the two microphones in the horizontal direction, that is, the first phase difference, v is the speed of sound, f is the frequency of the sound wave, and D1 is the distance between the two microphones in the horizontal direction.

[0050] In this embodiment, the step of calculating the second phase difference between the abnormal noise source and the two microphones based on the clean abnormal noise signals from the two vertical microphones includes calculating the phases corresponding to the two vertical microphones based on the frequencies corresponding to the maximum amplitude points in the spectrogram of the clean abnormal noise signals from the two vertical microphones, and subtracting the phases corresponding to the two vertical microphones to obtain the second phase difference. The method for calculating the phases corresponding to the two vertical microphones is the same as the method for calculating the phases corresponding to the two horizontal microphones and is not further described here.

[0051] In this embodiment, the vertical incident angle is obtained by the following formula: , where ∆Ψ UD is the phase difference between the two microphones in the vertical direction, that is, the second phase difference, v is the speed of sound, f is the frequency of the sound wave, and D2 is the distance between the two microphones in the vertical direction.

[0052] Of course, when calculating the horizontal incident angle and the vertical incident angle based on the pure abnormal sound signals of a microphone in the horizontal direction and a microphone in the vertical direction, the same calculation method as above can be used, which will not be repeated here.

[0053] After obtaining the horizontal incident angle and the vertical incident angle, the three-dimensional direction vector r of the abnormal sound source is determined according to the horizontal incident angle and the vertical incident angle. s =(sinθcosΦ, sinθsinΦ, cosθ). Next, obtain the time difference ∆t of the sound propagating from the abnormal sound source to any two microphones. Obtain the point where the peak value of the pure abnormal sound signal of the two microphones is the largest. Since the peak value occurs at different times in each group of signals, calculate the difference between the two to obtain the time difference ∆t. After obtaining the time difference, according to formula d s =v∆t, where v is the speed of sound. By calculating the time difference between the sound source and any pair of microphones, the distance d from the source of the abnormal sound to the origin of the coordinate system can be calculated. sThe spatial coordinates of the abnormal noise source are obtained based on the distance from the abnormal noise source to the origin of the coordinate system and the three-dimensional direction vector. The spatial coordinates of the abnormal noise source can be expressed as: s =d s r s , thereby completing the positioning of the abnormal sound source.

[0054] In an optional embodiment, the step of locating the source of an abnormal sound based on a pure abnormal sound signal includes: taking each pair of microphones as a group, obtaining the time difference between the sound source reaching each pair of microphones, and constructing a system of equations based on the time differences of at least three groups of microphones, the position coordinates of the microphones, and the speed of sound. For example, taking three groups of microphones as an example, the system of equations is constructed as follows: , where x i 、y i 、z i represents the coordinates of the i-th microphone in the x, y, and z directions, ∆t ij represents the time difference between the sound source reaching microphones i and j. The above equation contains three unknowns, so at least three sensor data points are required to solve it. Alternatively, multiple sensors can be deployed to obtain multiple sets of sound source coordinates and then averaged to improve location accuracy.

[0055] As can be seen from the above, the air conditioner noise localization method of the present invention uses the noise signal as a low-rank matrix and adaptively obtains the rank value of each frame. Using bilateral random projection theory, the low-rank matrix is ​​further approximated to a pure speech matrix and converted into a pure speech signal matrix. This effectively removes background noise, extracts a pure noise signal, and improves noise detection accuracy in complex background noise environments.

[0056] Air conditioner embodiment: The air conditioner of this embodiment includes a controller, and when the controller executes a computer program, the steps in the embodiment of the method for locating abnormal sounds of an air conditioner are implemented.

[0057] For example, a computer program may be divided into one or more modules, one or more of which are stored in a memory and executed by a controller to implement the present invention. One or more modules may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the air conditioner.

[0058] The air conditioner may include, but is not limited to, a controller and a memory. Those skilled in the art will appreciate that the air conditioner may include more or fewer components, or a combination of certain components, or different components. For example, the air conditioner may also include input and output devices, network access devices, buses, etc.

[0059] For example, the controller can be a central processing unit (CPU), other general-purpose controllers, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose controller can be a microcontroller or any conventional controller. The controller is the control center of the air conditioner and connects the various parts of the entire air conditioner using various interfaces and lines.

[0060] The memory can be used to store computer programs and / or modules. The controller implements various functions of the air conditioner by running or executing the computer programs and / or modules stored in the memory and accessing data stored in the memory. For example, the memory may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (e.g., a voice reception function, a voice-to-text function, etc.); the data storage area may store data generated based on the use of the mobile phone (e.g., audio data, text data, etc.). Furthermore, the memory may include high-speed random access memory (RAM) and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a SmartMediaCard (SMC), a Secure Digital (SD) card, a flash memory card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0061] Computer readable storage medium embodiment: If the air conditioner integrated module of the above-mentioned embodiment is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the process described in the above-mentioned air conditioner abnormal sound locating method embodiment can also be implemented by a computer program instructing the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a controller, the computer program can implement the steps of the above-mentioned air conditioner abnormal sound locating method embodiment. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. Storage media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, removable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signals, telecommunication signals, and software distribution media. It should be noted that the content of the computer-readable medium can be appropriately expanded or reduced based on the requirements of legislation and patent practice within a jurisdiction. For example, in some jurisdictions, legislation and patent practice do not require that computer-readable media include electric carrier signals and telecommunication signals.

[0062] It should be noted that the above are only preferred embodiments of the present invention, but the design concept of the invention is not limited thereto. Any non-substantial modifications made to the present invention using this concept also fall within the scope of protection of the present invention.

Claims

1. A method for locating abnormal sound of an air conditioner, applied to an air conditioner, characterized in that: include: Acquire abnormal sound signals collected by the microphone array; Performing noise reduction and enhancement processing on the abnormal sound signal collected by each microphone in the microphone array to obtain a pure abnormal sound signal corresponding to each microphone; performing an abnormal sound source location operation according to the pure abnormal sound signal; The step of performing noise reduction and enhancement processing on the abnormal sound signal collected by each microphone in the microphone array to obtain a pure abnormal sound signal corresponding to each microphone includes: Performing frame processing on the time domain signal of the abnormal sound signal, and converting the time domain signal of each frame into a Hankel matrix; Iteratively calculating a low-rank matrix of the Hankel matrix using a preset bilateral random projection theory method to obtain an estimated pure abnormal sound signal matrix; The pure abnormal sound signal matrix is ​​transformed into the time domain to obtain the pure abnormal sound signal.

2. The method for locating abnormal sound of an air conditioner according to claim 1, characterized in that: The low-rank matrix L is obtained by the following formula: , where Y1 and Y2 are the bilateral projection matrices of the Hankel matrix X, , A1 and A2 are random matrices.

3. The method for locating abnormal sound of an air conditioner according to claim 2, characterized in that: The step of iteratively calculating a low-rank matrix from the Hankel matrix using a preset bilateral random projection theory method includes: Optimize the bilateral projection to a single-sided projection, and the optimized low-rank matrix Obtained by the following formula: 。 4. The method for locating abnormal sound of an air conditioner according to any one of claims 1 to 3, characterized in that: The microphone array includes at least three microphones, one of which is arranged on at least one of the upper and lower sides in the vertical direction of the air conditioner, and one of which is arranged on at least one of the left and right sides in the horizontal direction of the air conditioner; The step of performing an abnormal sound source location operation according to the pure abnormal sound signal comprises: Calculating a first phase difference between the abnormal sound source and the two microphones based on the pure abnormal sound signals from one microphone and the other microphone in the horizontal direction, and calculating a horizontal incident angle of the abnormal sound source in the coordinate system based on the first phase difference; Calculating a second phase difference between the abnormal sound source and the two microphones based on the pure abnormal sound signals from one microphone and the other microphone in the vertical direction, and calculating a horizontal incident angle of the abnormal sound source in the coordinate system based on the second phase difference; Determine the three-dimensional direction vector of the abnormal sound source according to the horizontal incident angle and the vertical incident angle; Obtaining a time difference between the time it takes for the sound to propagate from the abnormal sound source to any two of the microphones, and obtaining a distance from the abnormal sound source to the origin of the coordinate system based on the time difference and the speed of sound; The spatial coordinates of the abnormal sound source are obtained according to the distance from the abnormal sound source to the origin of the coordinate system and the three-dimensional direction vector.

5. The method for locating abnormal sound of an air conditioner according to claim 4, characterized in that: The microphone array includes four microphones, one microphone is respectively arranged on the upper and lower sides along the vertical direction of the air conditioner, and one microphone is respectively arranged on the left and right sides along the horizontal direction of the air conditioner; The step of locating the source of the abnormal sound according to the pure abnormal sound signal comprises: Calculating the first phase difference between the abnormal sound source and the two microphones according to the pure abnormal sound signals from the two microphones in the horizontal direction, and calculating the horizontal incident angle of the abnormal sound source in the coordinate system according to the first phase difference; A second phase difference between the abnormal noise source and the two microphones is calculated based on the pure abnormal noise signals from the two microphones in the vertical direction, and the vertical incident angle of the abnormal noise source in the coordinate system is calculated based on the second phase difference.

6. The method for locating abnormal sound of an air conditioner according to claim 5, characterized in that: The step of calculating a first phase difference between the abnormal sound source and the two microphones according to the pure abnormal sound signals from the two microphones in the horizontal direction comprises: Calculating the phases corresponding to the two microphones in the horizontal direction according to the frequencies corresponding to the maximum amplitude points in the spectrum diagram of the pure abnormal sound signals of the two microphones in the horizontal direction; The phases corresponding to the two microphones in the horizontal direction are subtracted to obtain the first phase difference.

7. The method for locating abnormal sound of an air conditioner according to claim 6, characterized in that: The horizontal incident angle is obtained by the following formula: , where ∆Ψ LR is the first phase difference, v is the speed of sound, f is the frequency of the sound wave, and D1 is the distance between the two microphones in the horizontal direction.

8. The method for locating abnormal sound of an air conditioner according to claim 5, characterized in that: The step of calculating a second phase difference of the abnormal sound source propagating to the two microphones according to the pure abnormal sound signals of the two microphones in the vertical direction comprises: Calculate the phases corresponding to the two microphones in the vertical direction according to the frequencies corresponding to the maximum amplitude points in the spectrum graph of the pure abnormal sound signals of the two microphones in the vertical direction; The second phase difference is obtained by subtracting the phases corresponding to the two microphones in the vertical direction.

9. The method for locating abnormal sound of an air conditioner according to claim 8, characterized in that: The vertical incidence angle is obtained by the following formula: , where ∆Ψ UD is the second phase difference, v is the speed of sound, f is the frequency of the sound wave, and D2 is the distance between the two microphones in the vertical direction.

10. An air conditioner comprising a processor and a memory, characterized in that: The memory stores a computer program, and when the computer program is executed by the processor, the steps of the method for locating abnormal noise of an air conditioner are implemented as described in any one of claims 1 to 9.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the controller, the steps of the method for locating abnormal noise of an air conditioner as claimed in any one of claims 1 to 9 are implemented.

Images