Microphone array optimization methods, equipment, electronic terminals and storage media

Abstract

This invention discloses a microphone array optimization method, device, electronic terminal, and storage medium. The microphone array optimization method of this invention adopts the following technical solution: Pre-collected typical abnormal noise acoustic signal parameters of GIS, and the acoustic monitoring distance and azimuth parameters of GIS are input into an optimized microphone array model. The performance of the microphone array model is simulated and analyzed, and the main lobe and side lobe parameters in the array response spectrum are evaluated. Based on the evaluation results, it is determined whether to iteratively optimize the microphone array model until an ideal result is achieved, and then the microphone array model parameters are output. Since the pre-collected typical abnormal noise acoustic signal parameters of GIS, and the acoustic monitoring distance and azimuth parameters of GIS are derived from actual GIS operating data, they have strong practical relevance, making the final optimized microphone array model suitable for abnormal noise monitoring of GIS power station equipment. This highly targeted approach improves the monitoring effect.

Description

Technical Field

[0001] This invention relates to the field of abnormal noise detection and location technology for power equipment, specifically a microphone array optimization method, device, electronic terminal, and computer-readable storage medium for monitoring abnormal noise in GIS. Background Technology

[0002] Gas-insulated metal-enclosed equipment (GIS) has been widely used in power systems both domestically and internationally due to its advantages such as high voltage, high current, compact structure, flexible layout, stable operation, long service life, superior technical specifications, and immunity to external influences. However, GIS malfunctions often produce abnormal noises. If these malfunctions are not detected and resolved promptly, they may lead to serious safety accidents.

[0003] For detecting and locating abnormal noises from equipment caused by mechanical defects, the most effective method is obviously to directly measure the acoustic signal. However, due to the complex noise environment at substation sites, abnormal noises can be masked by background noise from the equipment, making it difficult to capture and locate the noises using only the human ear or individual acoustic sensors. Detecting and locating abnormal noises from power equipment involves sound source identification methods, requiring the use of microphone arrays (acoustic arrays) to pinpoint the sound source. The relevant parameters of the microphone array, such as its size and the spacing of its elements, significantly impact the sound source localization results. Therefore, it is necessary to optimize the design of the microphone arrays used in GIS (Gas Sensing System) to improve their monitoring effectiveness. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide a method, device, electronic terminal and computer-readable storage medium for optimizing the microphone array for monitoring abnormal noises in GIS, thereby improving the monitoring effect of the microphone array.

[0005] To achieve the above and other related objectives, the present invention adopts the following technical solution: a microphone array optimization method for monitoring abnormal noises in GIS, comprising the following steps:

[0006] 1) The microphone array model is simplified;

[0007] 2) Optimize the microphone array to obtain an optimized microphone array model;

[0008] 3) Input the pre-collected typical abnormal acoustic signal parameters of GIS, the acoustic monitoring distance of GIS, and the azimuth parameters into the optimized microphone array model, perform simulation analysis on the performance of the microphone array model, and evaluate the main lobe and side lobe parameters in the array response spectrum.

[0009] 4) If the evaluation result of step 3) is not up to standard, repeat steps 1) to 3); if the evaluation result of step 3) is up to standard, output the microphone array model parameters and end.

[0010] Preferably, step 1) includes:

[0011] 1.1) Initialize the radius range of the microphone array rings: Limit the minimum radius interval between two adjacent rings of the microphone array to ρ. min Set the number of rings to M1, and then obtain the radius range of the microphone array model;

[0012] 1.2) Initialize the microphone array element spacing range: Limit the minimum spacing d between adjacent elements in each ring of the microphone array. min In the process of generating new array elements for each ring, the number of array elements on the ring is set to M2, and the array element spacing range of the microphone array model is obtained.

[0013] Preferably, step 2) includes:

[0014] 2.1) Population initialization: Initialize the parameters of the radius of each microphone array ring and / or the element spacing on each ring, and generate the corresponding target vector and mutation vector;

[0015] 2.2) Cross: The test vector is obtained by performing a binary cross on the target vector and variation vector of the radius of each microphone array ring and / or the element spacing on each ring;

[0016] 2.3) Boundary condition handling.

[0017] Preferably, the evaluation of the main lobe and side lobe parameters in the array response spectrum in step 3) includes: evaluating the width and amplitude of the main lobe and evaluating the width and amplitude of the side lobes.

[0018] Preferably, the typical acoustic signal parameters of abnormal noise in GIS in step 3) include the sound source frequency and the signal-to-noise ratio.

[0019] Preferably, the typical abnormal noise acoustic signal parameters of the GIS also include acoustic signal parameters of abnormal noise modes such as loose shielding, loose contacts, loose bolts, internal foreign objects, and transformer noise.

[0020] Preferably, in step 4), if the difference between the evaluation results obtained from repeating steps 1) to 3) twice is within a set range, the microphone array model parameters are output and the process ends.

[0021] Corresponding to the microphone array optimization method for monitoring abnormal noises in GIS of the present invention, the present invention also provides a microphone array optimization device for implementing the above-mentioned microphone array optimization method for monitoring abnormal noises in GIS.

[0022] Corresponding to the microphone array optimization method for monitoring abnormal noises in GIS of the present invention, the present invention also provides an electronic terminal, including a storage unit and a processing unit. The storage unit is used to store pre-collected acoustic signal parameters of typical abnormal noises in GIS. The processing unit is used to execute steps 1) to 4) of the above technical solution. The storage unit is also used to store the microphone array model parameters output by the processing unit.

[0023] Corresponding to the microphone array optimization method for monitoring abnormal noises in GIS according to the present invention, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the microphone array optimization method for monitoring abnormal noises in GIS as described above.

[0024] As described above, the microphone array optimization method for monitoring abnormal noises in GIS according to the present invention has the following beneficial effects: It uses pre-collected typical abnormal noise acoustic signal parameters of GIS, the acoustic monitoring distance of GIS, and azimuth parameters as input into an optimized microphone array model. The performance of the microphone array model is simulated and analyzed, and the main lobe and side lobe parameters in the array response spectrum are evaluated. Based on the evaluation results, it is determined whether to iteratively optimize the microphone array model until an ideal result is achieved, and then the microphone array model parameters are output. Since the pre-collected typical abnormal noise acoustic signal parameters of GIS, the acoustic monitoring distance of GIS, and the azimuth parameters are all derived from actual GIS operating data, they have strong practical relevance. This makes the final optimized microphone array model suitable for monitoring abnormal noises in GIS power station equipment, highly targeted, and able to improve monitoring effectiveness.

[0025] The microphone array optimization device, electronic terminal, and computer-readable storage medium of the present invention for monitoring abnormal noises in GIS are used to execute the microphone array optimization method for monitoring abnormal noises in GIS of the present invention, and of course also have the above-mentioned beneficial effects, which will not be repeated here. Attached Figure Description

[0026] Figure 1 This is a flowchart of a microphone array optimization method for monitoring abnormal noises in GIS according to the present invention;

[0027] Figure 2 and Figure 3 The figures are respectively the directivity function diagrams of the microphone array in polar coordinates and rectangular coordinates in a specific embodiment of the present invention;

[0028] Figure 4 This is a schematic diagram of an electronic terminal according to the present invention.

[0029] Component labeling explanation: 1-Storage unit, 2-Processing unit. Detailed Implementation

[0030] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0031] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the illustrations only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0032] Please see Figure 1 This invention provides a method for optimizing a microphone array for monitoring abnormal noises in GIS, comprising the following steps:

[0033] 1) The microphone array model is simplified;

[0034] 2) Optimize the microphone array to obtain an optimized microphone array model;

[0035] 3) Input the pre-collected typical abnormal acoustic signal parameters of GIS, the acoustic monitoring distance of GIS, and the azimuth parameters into the optimized microphone array model to perform simulation analysis on the performance of the microphone array model, and evaluate the main lobe and side lobe parameters in the array response spectrum.

[0036] 4) If the evaluation result of step 3) is not up to standard, repeat steps 1) to 3); if the evaluation result of step 3) is up to standard, output the microphone array model parameters and end.

[0037] In a microphone array optimization method for monitoring abnormal noises in GIS (Gas-Insulated Switchgear) according to the present invention, pre-collected typical abnormal noise acoustic signal parameters of GIS, acoustic monitoring distance of GIS, and azimuth parameters are input into an optimized microphone array model to simulate and analyze the performance of the microphone array model, and the main lobe and side lobe parameters in the array response spectrum are evaluated. Based on the evaluation results, it is determined whether to iteratively optimize the microphone array model until the ideal result is achieved, and then the microphone array model parameters are output. Since the pre-collected typical abnormal noise acoustic signal parameters of GIS, acoustic monitoring distance of GIS, and azimuth parameters are derived from actual GIS operating data, they have strong practical relevance. This makes the final optimized microphone array model suitable for abnormal noise monitoring of GIS power station equipment, with strong relevance, and can improve the monitoring effect.

[0038] The directivity of a microphone array refers to the characteristic that the amplitude of the sound signal received by the array varies with the azimuth angle. When the microphone array is focused in a certain direction, the ratio between the array output in any direction and the array output in the direction of the principal maximum is the directivity function of the array. Figure 2 and Figure 3 The image shows the directivity function of the microphone array, often simply called the radiation pattern. In the array radiation pattern, the beam containing the maximum directivity function along the reference direction is called the main lobe. Beams in other directions where the directivity function equals the maximum of the main beam are called grating lobes. A series of beams with directivity functions smaller than the maximum of the directivity function along the reference direction are called side lobes. For a microphone array to have high resolution and good anti-interference capability, it should have a small main lobe width and small side lobes.

[0039] Under constraints such as microphone array aperture and number of elements, optimized microphone design can achieve the desired radiation pattern by optimizing parameters such as the number of array elements, array position, and element excitation phase and amplitude. Optimized microphone array design can not only effectively reduce the cost of array microphones by using as few elements as possible, but also achieve the requirements of low sidelobe performance and high resolution.

[0040] Differential evolution is a global optimization method based on population intelligence. It does not rely on the feature information of the problem, uses floating-point real number encoding, and searches the entire population space to guide population evolution through the difference information between individuals in the population. After the corresponding operations and multiple iterations of the differential evolution algorithm, the optimal solution to the problem is finally obtained.

[0041] In a microphone array optimization method for monitoring abnormal noises in GIS according to the present invention, the microphone array model is first simplified. In the synthesis process of the sparsely distributed concentric ring array, the constraints are limited to the array aperture R and the minimum radius interval between adjacent rings is limited to ρ. minThe spacing between the array elements on each ring is limited to the range [d]. min ,d max And ensure that the adjacent array elements of each ring are spaced at the same interval.

[0042] In a preferred embodiment of the microphone array optimization method for monitoring abnormal noises in GIS according to the present invention, step 1) includes:

[0043] 1.1) Initialize the radius range of the microphone array rings: Limit the minimum radius interval between two adjacent rings of the microphone array to ρ. min Set the number of rings to M1, and then obtain the radius range of the microphone array model;

[0044] 1≤j<i≤M1,ρ min If ρ is a constant, then min{ρ} i -ρ j}≥ρ min The radius constraint vector is then constructed as follows:

[0045] ρ″=[ρ min ,2·ρ min ,…(M1-1)ρ min ,0]

[0046] This reduces the search space for the radius of the annulus from the original R to SR, where SR is represented as:

[0047] SR=R-M1·ρ min

[0048] During the process of initializing the radius of the annulus to generate new individuals, M1 random numbers are randomly generated in the search space [0,SR], and these M1 numbers are arranged in ascending order to obtain the indirect radius vector. Adding it to the radius constraint variable ρ″ forms the radius vector ρ, that is:

[0049] ρ=ρ′+ρ″

[0050] 1.2) Initialize the microphone array element spacing range: Limit the minimum spacing d between adjacent elements in each ring of the microphone array. min In the process of generating new array elements for each ring, the number of array elements on the ring is set to M2, and the array element spacing range of the microphone array model is obtained.

[0051] The constraint condition for the element spacing is to limit the minimum spacing d between adjacent elements. min Introduce the array element spacing constraint vector:

[0052]

[0053] During the generation of new individuals at the array element intervals, in the search space [0,d]max -d min M² random numbers are generated on the array, and these M² numbers are arranged in ascending order to obtain the indirect array element spacing vector. Adding it to the element spacing constraint vector d″ forms the element spacing vector d, that is:

[0054] d=d′+d″

[0055] Through the initialization of the annular radius and element spacing described above, a new individual was generated, and the constraints were well satisfied. In the simulation of this embodiment, the present invention divides the optimization variables into annular radius, element spacing on each annular ring, and annular radius and element spacing. When only the annular radius is optimized, the optimization variable is expressed as v = ρ; when only the element spacing is optimized, the optimization variable is expressed as v = d; when both the annular radius and element spacing are jointly optimized, the optimization variable is expressed as v = [ρ, d].

[0056] The effect of constraint vectors when simultaneously optimizing the annular radius and element spacing using an improved differential evolution algorithm is explained. During the optimization process of the improved differential algorithm, due to the introduction of constraint vectors ρ″ and d″, the search space for the annular radius decreases from R to SR, and the search space for the element spacing decreases from [0, d... max Decrease to [0, d max -d min During the initialization process of the difference algorithm, it is only necessary to randomly generate indirect individuals v′=[ρ′,d″] in the reduced search space. This indirect individual v′ is formed by subtracting the constraint vector v″=[ρ″,d″] from each individual v=[ρ,d] in the population. This indirect individual v′ is used in the mutation, crossover and selection processes of the difference algorithm.

[0057] There are two situations in which the constraint vector v″ needs to be added to the indirect individual v′: First, when calculating the fitness function, the constraint vector needs to be added to the indirect individual to form a real individual and obtain the corresponding peak sidelobe level before the selection operation is performed; Second, when obtaining the final optimization result, the peak sidelobe level of the final optimal individual needs to be calculated and the corresponding radiation pattern needs to be drawn.

[0058] In a microphone array optimization method for monitoring abnormal noises in GIS according to the present invention, an improved differential evolution algorithm is applied to optimize the microphone array. Preferably, step 2) includes:

[0059] 2.1) Population initialization: Initialize the parameters of the radius of each microphone array ring and / or the element spacing on each ring, and generate the corresponding target vector and mutation vector.

[0060] When applying the improved differential evolution algorithm to a sparsely distributed concentric ring array to optimize only the radius of the microphone array rings, the initial population consists of a vector of real-valued parameters with NP individuals and dimension M1 (i.e., the number of rings). The i-th individual in the initial population...

[0061] ρ i (0)={ρ i1 (0),ρ i2 (0)…ρ iM Element initialization of (0)}(i=1,2…NP,j=1,2…M1-1):

[0062] ρ ij (0)=(R-M1ρ min -0)·rand+0

[0063] From the above equation, ρ ij (0) constitutes a value in [0, R-M1ρ] min Given M1-1 random numbers, these M1-1 values ​​need to be sorted in ascending order. ρ ij The value of the Mth dimension of (0) is ρ iM (0) = R, which is the array aperture. In fact, when only the radius is optimized, only the first M1-1 dimensions of the individual need to be optimized, and the value of the last dimension is a constant.

[0064] When applying the improved differential evolution algorithm to a sparsely distributed concentric ring array to optimize only the element spacing, an initial population is set up, where the number of individuals in the population is NP, and each individual contains M2 variables. The i-th individual in the initial population...

[0065] d i (0)={d i1 (0),d i2 (0)…d iM The elements of (0)} (i = 1, 2…NP, j = 1, 2…M2) are initialized as follows:

[0066] d ij (0)=(d max -d min )·rand+0

[0067] d ij (0) constitutes a value in [0, d max -d min The M2 random numbers do not need to be sorted in ascending order.

[0068] When jointly optimizing the radius of the annulus and the element spacing on each annulus, the dimension of each individual is initialized from 1 to M1 as the radius of the optimized annulus, and the element spacing of the optimized annulus is initialized from 1 to M2 as the dimension of each individual, that is:

[0069] X i (0)={x ij (0)}={ρ i (0),d i The initialization of the value of (0)} is as described above, and will not be repeated here.

[0070] For each X i (0)={x ij For individuals (t) (i = 1, 2…NP, j = 1, 2…N), the mutation vector is generated as follows:

[0071] V i (t+1)=X p (t)+F·[X j (t)-X k (t)]

[0072] The randomly selected numbers i, j, p, and k are all distinct and are integers within the range [1, NP]. Therefore, NP must be greater than 4. F is the scaling factor, which is a real constant factor.

[0073] 2.2) Crossover: The test vector is obtained by performing a binary crossover on the target vector and variation vector of the radius of each microphone array ring and / or the element spacing on each ring.

[0074] Using binary crossover, the test vector U i (t+1)={U ij (t+1)} is derived from the target vector X i (0)={x ij (t)} and the mutation vector V i (t+1)={v ij The result is obtained by performing a crossover operation on (t)}, when rand≤CR or j=j rand At that time, U i The j-th dimension component of (t+1) is given by V i The j-th dimension component of (t+1) is provided, otherwise U i The j-th dimension component of (t+1) is provided. rand Let be a random number in the interval [l, N], and CR represent the crossover probability factor. The equation for the binomial crossover operation is:

[0075]

[0076] 2.3) Boundary Condition Handling

[0077] When the optimization variable is only the radius of the annulus, the radius of the annulus must satisfy the minimum spacing ρ between the annulus. min The search space for the radius of the annulus is reduced from R to SR, U i (t+1)={uij (t+1)}(i∈[1,NP]j∈[1,M1]) represents individuals generated after mutation, crossover, and other operations, where t is the current generation number and u is the element. ij (t+1) If it exceeds the range [0,SR], u ij (t+1) is generated as follows:

[0078] u ij (t+1)=SR·rand(1) i=1,2…NP,j=1,2…M1-1

[0079] When the optimization variable is the spacing between the array elements on each ring, if u ij (t+1) exceeds [0, d max -d min The range of ], u ij (t+1) is generated as follows:

[0080] u ij (t+1)=(d max -d min )·rand(1) i=1,2…NP,j=1,2…M

[0081] When the optimization variable is the radius of the ring, after boundary processing, each element of the individual needs to be arranged from smallest to largest. When the optimization variable is the spacing between the array elements on each ring, it is not necessary to arrange them from smallest to largest.

[0082] The radius vector of the optimal sparse concentric ring array obtained by jointly optimizing the concentric circle radius and the element spacing through the improved differential evolution algorithm is r = (0.0455, 0.0985, 0.155), and the optimal number of elements on each ring is N = (20, 26, 26).

[0083] In a microphone array optimization method for monitoring abnormal noises in GIS according to the present invention, the optimization effect of the microphone array is determined by evaluating the main lobe and side lobe parameters in the microphone array response spectrum. Preferably, the evaluation of the main lobe and side lobe parameters in the array response spectrum in step 3) includes: evaluating the width and amplitude of the main lobe and evaluating the width and amplitude of the side lobe.

[0084] The acoustic characteristics of abnormal noises in GIS power station equipment are highly specific. These abnormal noises generally occur within the 100Hz to 800Hz range, classifying them as low-frequency noise sources with strong frequency characteristics. Preferably, the typical acoustic signal parameters for abnormal noises in GIS equipment in step 3) include the source frequency and signal-to-noise ratio. The sources of abnormal noises in GIS substations are generally transformer noise, loose parts, and internal foreign objects. Preferably, the typical acoustic signal parameters for abnormal noises in GIS equipment include acoustic signal parameters related to loose shielding, loose contacts, loose bolts, internal foreign objects, and transformer noise abnormal noise patterns.

[0085] In a microphone array optimization method for monitoring abnormal noises in GIS according to the present invention, pre-collected typical abnormal noise acoustic signal parameters of GIS, acoustic monitoring distance of GIS, and azimuth parameters are input into an optimized microphone array model. After simulating and analyzing the performance of the microphone array model, the main lobe and side lobe parameters in the microphone array response spectrum are evaluated to determine whether the optimization meets the standards. However, extreme cases or certain factors may prevent the optimization model from achieving the evaluation standard. In such cases, the program needs to obtain a reasonable result before ending the operation. Therefore, as a preferred embodiment, in step 4), if the difference between the evaluation results obtained from two consecutive repetitions of steps 1) to 3) is within a set range, the microphone array model parameters are output and the process ends. If the difference between two consecutive evaluation results is very small, it can be considered that no further significant optimization effect can be obtained, and the program can end and output the optimized microphone array model parameters.

[0086] Corresponding to the microphone array optimization method for monitoring abnormal noises in GIS according to the present invention, the present invention also provides an electronic terminal, such as... Figure 4 As shown, it includes a storage unit and a processing unit. The storage unit is used to store pre-collected typical abnormal acoustic signal parameters of GIS. The processing unit is used to execute steps 1) to 4) of the above technical solution. The storage unit is also used to store the microphone array model parameters output by the processing unit.

[0087] Corresponding to the microphone array optimization method for monitoring abnormal noises in GIS according to the present invention, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the microphone array optimization method for monitoring abnormal noises in GIS as described above.

[0088] Based on the technical solutions of the above specific embodiments, the microphone array optimization method for monitoring abnormal noises in GIS of the present invention can optimize the microphone array model to be suitable for monitoring abnormal noises in GIS power station equipment, which is highly targeted and can improve the monitoring effect.

[0089] An electronic terminal and a computer-readable storage medium of the present invention are used to execute the microphone array optimization method for monitoring abnormal noises in GIS, and of course, also have the above-mentioned beneficial effects, which will not be repeated here.

[0090] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method of optimizing a microphone array, characterized by, Includes the following steps: 1) The microphone array model is simplified; 1.1) Limit the minimum radius interval between adjacent rings, set the number of rings, and construct a ring radius constraint vector; 1.2) Limit the minimum spacing between adjacent array elements, set the number of each ring array element and construct the array element spacing constraint vector; and generate indirect individuals of the microphone array model based on the above constraint vector; 2) The indirect individuals are optimized using the differential evolution algorithm. During mutation, crossover, and selection, they are kept in the constraint space. The constraint vector is added during fitness calculation to form the real individuals, thereby obtaining the optimized microphone array model. 3) Input the pre-collected typical abnormal acoustic signal parameters of GIS, the acoustic monitoring distance and azimuth parameters of GIS into the optimized microphone array model, and evaluate the array main lobe width, main lobe amplitude, side lobe width and side lobe amplitude. 4) If the evaluation result of step 3) is not up to standard, repeat steps 1) to 3); if the evaluation result of step 3) is up to standard, output the microphone array model parameters and end.

2. The microphone array optimization method of claim 1, wherein, Step 1) includes: 1.1) initialization of the microphone array annular radius range: define the minimum radius interval of the adjacent two annular rings of the microphone array as , set the number of annular rings as M 1, and obtain the radius range of the microphone array model; 1.2) Initialize the microphone array element interval range: define the minimum interval of adjacent elements of each ring of the microphone array In the process of generating new element individuals for each ring, the number of elements on the ring is set to M 2, the element interval range of the microphone array model is obtained.

3. The microphone array optimization method as described in claim 1, characterized in that, Step 2) includes: 2.1) Population initialization: Initialize the parameters of the radius of each microphone array ring and / or the element spacing on each ring, and generate the corresponding target vector and mutation vector; 2.2) Cross: The test vector is obtained by performing a binary cross on the target vector and variation vector of the radius of each microphone array ring and / or the element spacing on each ring; 2.3) Boundary condition handling.

4. The microphone array optimization method as described in claim 1, characterized in that, The typical acoustic signal parameters of abnormal noise in GIS in step 3) include the sound source frequency and the signal-to-noise ratio.

5. The microphone array optimization method as described in claim 1, characterized in that, The acoustic signal parameters of typical abnormal noises in GIS include acoustic signal parameters of abnormal noise modes such as loose shielding, loose contacts, loose bolts, internal foreign objects, and transformer noise.

6. The microphone array optimization method as described in claim 1, characterized in that, in In step 4), if the difference between the evaluation results obtained from repeating steps 1) to 3) twice is within the set range, the microphone array model parameters are output and the process ends.

7. A microphone array optimization device, characterized in that, It is used to implement the microphone array optimization method according to any one of claims 1-6.

8. An electronic terminal, comprising a storage unit and a processing unit, characterized in that, The storage unit is used to store pre-collected typical abnormal acoustic signal parameters of GIS, and the processing unit is used to execute steps 1) to 4) as described in claim 1. The storage unit is also used to store the microphone array model parameters output by the processing unit.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the microphone array optimization method according to any one of claims 1-6.

Images