Method, terminal and storage medium for determining equivalent model of transformer multi-source noise

Abstract

The present invention provides a method, terminal, and storage medium for determining an equivalent model of transformer multi-source noise. The method includes: obtaining the spatial coordinates of multiple preset detection points around the transformer and the actual sound pressure levels in preset octave bands; obtaining the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source; constructing a single-variable linear regression model corresponding to the preset octave band based on the spatial coordinates of the multiple preset detection points and the actual sound pressure levels in the preset octave bands, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source; solving the single-variable linear regression model corresponding to the preset octave band to obtain the sound pressure levels of the multiple equivalent sound sources in the preset octave bands; and obtaining an equivalent model of transformer multi-source noise corresponding to the preset octave band based on the number of equivalent sound sources, the sound pressure levels of each equivalent sound source, and the spatial coordinates of each equivalent sound source. The present invention provides a convenient and simple equivalence process with minimal computational effort.

Description

Technical Field

[0001] The present invention relates to the technical field of noise equivalent models, and in particular to a method for determining a transformer multi-sound source noise equivalent model, a terminal, and a storage medium. Background Art

[0002] As substations are located closer to residential areas, noise pollution from these stations is receiving increasing attention. The transformer is the largest single piece of equipment within a substation and the primary source of noise. Constructing an accurate transformer multi-source noise equivalent model based on acoustic equivalent source theory is crucial for predicting substation noise.

[0003] At present, near-field acoustic holography technology is usually used to construct a transformer multi-source noise equivalent model. However, this method is computationally complex and requires a lot of calculations. Summary of the Invention

[0004] The embodiments of the present invention provide a method, a terminal, and a storage medium for determining a transformer multi-source noise equivalent model to solve the problems of complex calculations and large amount of calculations in the prior art.

[0005] In a first aspect, an embodiment of the present invention provides a method for determining a transformer multi-source noise equivalent model, comprising:

[0006] Obtaining the spatial coordinates of multiple preset detection points around the transformer and the actual sound pressure levels in preset octave bands;

[0007] Obtain the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source;

[0008] Constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source, and solving the univariate linear regression model corresponding to the preset octave band to obtain the sound pressure levels of the multiple equivalent sound sources in the preset octave band;

[0009] According to the number of equivalent sound sources, the sound pressure level of each equivalent sound source in a preset octave band and the spatial coordinates of each equivalent sound source, an equivalent model of transformer multi-sound source noise corresponding to the preset octave band is obtained.

[0010] In one possible implementation, a univariate linear regression model corresponding to the preset octave band is constructed based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source. The univariate linear regression model corresponding to the preset octave band is solved to obtain the sound pressure levels of the multiple equivalent sound sources in the preset octave band, including:

[0011] Constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source;

[0012] According to the predicted sound pressure level of each preset detection point in the preset octave band and the actual sound pressure level of each preset detection point in the preset octave band, the univariate linear regression model corresponding to the preset octave band is solved, and the sound pressure levels of multiple equivalent sound sources in the preset octave band are obtained through double-layer optimization.

[0013] In one possible implementation, based on the predicted sound pressure levels of each preset detection point in the preset octave band and the actual sound pressure levels of each preset detection point in the preset octave band, a single-variable linear regression model corresponding to the preset octave band is solved, and the sound pressure levels of multiple equivalent sound sources in the preset octave band are obtained through double-layer optimization, including:

[0014] The average distribution method is used to generate the first batch of random variables as input variables for the first optimization of the single variable linear regression model corresponding to the preset octave band;

[0015] According to the MSE loss function, the first optimization is performed, and the error values calculated by the MSE loss function are selected from the first batch of random variables in ascending order, and the first preset number of random variables are arranged;

[0016] Based on the mathematical expectation and variance of the first preset number of random variables, a second batch of random variables is generated using the normal distribution method as input variables for the second optimization of the single variable linear regression model corresponding to the preset octave band;

[0017] According to the cross-entropy loss function, a second optimization is performed, and the sound pressure level of each equivalent sound source in the preset octave band is finally solved from the second batch of random variables.

[0018] In one possible implementation, the MSE loss function is:

[0019]

[0020] Among them, MSE is the error value calculated by the MSE loss function; m is the number of predicted detection points; Lm j Lw is the actual sound pressure level of the jth preset detection point in the preset octave band; j is the predicted sound pressure level of the j-th preset detection point in the preset octave band, n is the number of equivalent sound sources; Lw ij Lp is the sound pressure level in the preset octave band generated by the i-th equivalent sound source at the j-th preset detection point; iis the sound pressure level of the ith equivalent sound source in the preset octave band, which is an unknown quantity; d ij is the distance between the i-th equivalent sound source and the j-th preset detection point; α is the atmospheric absorption attenuation coefficient of noise during propagation.

[0021] In one possible implementation, the cross entropy loss function is:

[0022]

[0023] Among them, L is the value calculated by the cross entropy loss function; P(Lp i ) is Lp i The probability of taking a value in the normal distribution function.

[0024] In one possible implementation, the number of equivalent sound sources is 24;

[0025] The distribution of the equivalent sound sources is as follows: the long box walls are arranged at equal intervals according to the 4*2 specification, and the short box walls are arranged at equal intervals according to the 2*2 specification.

[0026] In a second aspect, an embodiment of the present invention provides a device for determining a transformer multi-source noise equivalent model, comprising:

[0027] A first acquisition module is used to obtain the spatial coordinates of multiple preset detection points around the transformer and the actual sound pressure level in a preset octave band;

[0028] A second acquisition module is used to obtain the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source;

[0029] a solution module for constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source, and solving the univariate linear regression model corresponding to the preset octave band to obtain the sound pressure levels of the multiple equivalent sound sources in the preset octave band;

[0030] The model determination module is used to obtain the transformer multi-source noise equivalent model corresponding to the preset octave band according to the number of equivalent sound sources, the sound pressure level of each equivalent sound source in the preset octave band and the spatial coordinates of each equivalent sound source.

[0031] In one possible implementation, the solution module is specifically used to:

[0032] Constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source;

[0033] According to the predicted sound pressure level of each preset detection point in the preset octave band and the actual sound pressure level of each preset detection point in the preset octave band, the univariate linear regression model corresponding to the preset octave band is solved, and the sound pressure levels of multiple equivalent sound sources in the preset octave band are obtained through double-layer optimization.

[0034] In a third aspect, an embodiment of the present invention provides a terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for determining a transformer multi-source noise equivalent model as described in the first aspect or any possible implementation of the first aspect are implemented.

[0035] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the method for determining the transformer multi-source noise equivalent model as described in the first aspect or any possible implementation of the first aspect.

[0036] An embodiment of the present invention provides a method, terminal and storage medium for determining an equivalent model of multi-sound source noise of a transformer, which obtains the spatial coordinates of multiple preset detection points around the transformer and the actual sound pressure levels in a preset octave band; obtains the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source; constructs a single-variable linear regression model corresponding to the preset octave band according to the spatial coordinates of the multiple preset detection points and the actual sound pressure levels in the preset octave band, the number of equivalent sound sources and the spatial coordinates of each equivalent sound source, and solves the single-variable linear regression model corresponding to the preset octave band to obtain the sound pressure levels of the multiple equivalent sound sources in the preset octave band; obtains the equivalent model of multi-sound source noise of the transformer corresponding to the preset octave band according to the number of equivalent sound sources, the sound pressure levels of each equivalent sound source and the spatial coordinates of each equivalent sound source, which can overcome the problems of complex calculation and large amount of calculation in near-field acoustic holography technology. It only needs to measure the sound pressure levels of a small number of detection points near the transformer, and obtains the equivalent model of multi-sound source noise of the transformer based on the single-variable linear regression method. The equivalence process is convenient and simple, and the amount of calculation is small. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 This is a flowchart of a method for determining a transformer multi-source noise equivalent model provided by an embodiment of the present invention;

[0039] Figure 2 Schematic diagram of a transformer multi-source noise equivalent model provided by an embodiment of the present invention;

[0040] Figure 3 is a plan view of equivalent source position distribution provided by an embodiment of the present invention;

[0041] Figure 4 1 is a schematic structural diagram of a device for determining a transformer multi-source noise equivalent model provided by an embodiment of the present invention;

[0042] Figure 5 is a schematic diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In the following description, specific details such as particular system structures and techniques are provided for purposes of illustration, not limitation, to facilitate a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

[0044] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below with reference to the accompanying drawings.

[0045] See also Figure 1 , which shows a flowchart of the method for determining the equivalent model of transformer multi-source noise provided by an embodiment of the present invention. The method for determining the equivalent model of transformer multi-source noise may be executed by a terminal.

[0046] See also Figure 1 , the method for determining the equivalent model of the transformer multi-source noise includes:

[0047] In S101 , the spatial coordinates of a plurality of preset detection points around the transformer and the actual sound pressure levels in preset octave bands are obtained.

[0048] In this embodiment, an actual typical substation can be selected to obtain a plane distribution diagram and a three-dimensional space model of the substation. A spatial rectangular coordinate system is established with a certain point on the ground of the substation as the coordinate origin. The sound pressure level of several preset detection points (denoted as m) in the free space around the substation is measured, and the spatial coordinates (x Rj ,y Rj ,z Rj ) The actual sound pressure level Lm corresponding to the preset octave band j Among them, the detection point can also be called field point.

[0049] In a possible implementation, the coordinate origin is used as the reference point, the east direction is used as the positive direction of the X axis, the north direction is used as the positive direction of the Y axis, and the upward direction is used as the positive direction of the Z axis. According to the spatial position relationship between the detection point and the coordinate origin, the spatial position coordinates of the detection point are established, which are recorded as (x Rj ,y Rj ,z Rj ), the unit is "meter".

[0050] In this embodiment, a noise real-time signal analyzer can be used to measure the actual sound pressure level at each preset detection point. The noise real-time signal analyzer is a pocket-sized real-time analyzer using digital signal processing technology, which can perform spectrum and amplitude analysis on noise, vibration or other electrical signals.

[0051] When measuring the sound pressure level at the detection point, a clear, windless day should be selected, and the temperature, air humidity, and atmospheric pressure of the day should be recorded. When measuring the sound pressure level at the detection point using a real-time noise signal analyzer, each sampling point should be sampled continuously for 30 to 60 seconds, sampled five times in a row, and the average value should be taken. The sampling results of the sampling points should be subjected to spectrum analysis to extract the sound pressure levels of the eight octave bands of 63 Hz, 125 Hz, 250 Hz, 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz, and 8000 Hz.

[0052] The position of each preset detection point can be selected according to actual needs, and the preset octave band can be any one of the eight octave bands mentioned above.

[0053] In S102 , the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source are obtained.

[0054] Among them, the equivalent sound source can also be called an equivalent point sound source.

[0055] This embodiment can be based on the multi-point equivalent source theory. According to the obtained spatial position parameters and geometric parameters of the transformer, a related algorithm is used to obtain the geometric spatial information of the transformer equivalent point sound source, including the number of equivalent point sound sources and the spatial distribution of the equivalent point sound sources. Based on this, a transformer multi-point equivalent source noise radiation model (i.e., a transformer multi-source noise equivalent model) is established. The spatial position coordinates of n equivalent sources are marked as (x NSi ,y NSi ,z NSi ).

[0056] The transformer may be a 500KV three-phase main transformer, and the geometric parameters of the transformer are 16.0m in length, 5.0m in width, and 5.0m in height.

[0057] In some embodiments, the number of the equivalent sound sources is 24;

[0058] The distribution of the equivalent sound sources is as follows: the long box walls are arranged at equal intervals according to the 4*2 specification, and the short box walls are arranged at equal intervals according to the 2*2 specification.

[0059] Based on the geometric parameter information of the above transformer, when constructing the transformer multi-source noise equivalent model, the number of equivalent sound sources is set to 24, and the arrangement method is that the long box wall is equivalent to the equivalent sound source arranged at 4x2 equal intervals, and the short box wall is equivalent to the equivalent sound source arranged at 2x2 equal intervals, as shown in the following example: Figure 2 As shown, Figure 2 The middle cuboid is a virtual model of the transformer, and the black dots are equivalent sound sources. The equivalent sound sources are distributed on the four sides of the cuboid, and there are no equivalent sound sources on the upper and lower sides. Figure 2 Clearly indicate the equivalent sound source, Figure 2 Only the equivalent sound sources of one long box wall and one short box wall are drawn. However, in the actual model, both the two opposite long box walls have equivalent sound sources, and both the two opposite short box walls have equivalent sound sources.

[0060] According to the spatial position of the equivalent sound source, the spatial position coordinates of the equivalent sound source are recorded as (x NSi ,y NSi ,z NSi ).

[0061] In S103, a single-variable linear regression model corresponding to the preset octave band is constructed based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of equivalent sound sources and the spatial coordinates of each equivalent sound source, and the single-variable linear regression model corresponding to the preset octave band is solved to obtain the sound pressure levels of multiple equivalent sound sources in the preset octave band.

[0062] This embodiment constructs a univariate linear regression model corresponding to a preset octave band and solves the univariate linear regression model corresponding to the preset octave band to obtain the sound pressure levels of multiple equivalent sound sources in the preset octave band.

[0063] In this embodiment, to obtain the sound pressure levels of the eight octave bands of the equivalent sound source, the eight octave bands can be used as preset octave bands, and steps S101-S103 described above can be executed. Finally, the sound pressure levels of the equivalent sound source of the transformer multi-source noise equivalent model obtained in S104 can include the sound pressure levels of the eight octave bands. In other words, different univariate linear regression models must be constructed and solved for different octave bands to obtain the sound pressure levels of the equivalent sound source in that octave band.

[0064] In some embodiments, the above S103 may include:

[0065] Constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source;

[0066] According to the predicted sound pressure level of each preset detection point in the preset octave band and the actual sound pressure level of each preset detection point in the preset octave band, the univariate linear regression model corresponding to the preset octave band is solved, and the sound pressure levels of multiple equivalent sound sources in the preset octave band are obtained through double-layer optimization.

[0067] This embodiment solves the univariate linear regression model corresponding to the preset octave band through double-layer optimization, which can improve accuracy.

[0068] In some embodiments, the single-variable linear regression model corresponding to the preset octave band is solved based on the predicted sound pressure level of each preset detection point in the preset octave band and the actual sound pressure level of each preset detection point in the preset octave band, and the sound pressure levels of multiple equivalent sound sources in the preset octave band are obtained through double-layer optimization, including:

[0069] The average distribution method is used to generate the first batch of random variables as input variables for the first optimization of the single variable linear regression model corresponding to the preset octave band;

[0070] According to the MSE loss function, the first optimization is performed, and the error values calculated by the MSE loss function are selected from the first batch of random variables in ascending order, and the first preset number of random variables are arranged;

[0071] Based on the mathematical expectation and variance of the first preset number of random variables, a second batch of random variables is generated using the normal distribution method as input variables for the second optimization of the single variable linear regression model corresponding to the preset octave band;

[0072] The formula for the normal distribution is as follows:

[0073]

[0074] Among them, μ is the mathematical expectation value of the first preset number of random variables, σ 2 is the variance of the first preset number of random variables.

[0075] According to the cross-entropy loss function, a second optimization is performed, and the sound pressure level of each equivalent sound source in the preset octave band is finally solved from the second batch of random variables.

[0076] A uniform function may be used to generate a first batch of evenly distributed random variables. Each random variable includes the sound pressure levels of 24 equivalent sound sources in a preset octave band, and each random variable may be different.

[0077] During the first optimization, this embodiment calculates the global error between the predicted value and the actual measured value using the MSE loss function. A preset number of random variables with smaller errors are selected. Based on the mathematical expectation and variance of these random variables, a second batch of random variables is generated using a normal distribution. During the second optimization, a cross-entropy loss function is used to ultimately solve for the sound pressure level of each equivalent sound source in a preset octave band from the second batch of random variables. The sound pressure level of each equivalent sound source in the preset octave band is obtained through the second optimization.

[0078] In some embodiments, the above MSE loss function is:

[0079]

[0080] Among them, MSE is the error value calculated by the MSE loss function; m is the number of predicted detection points; Lm j Lw is the actual sound pressure level of the jth preset detection point in the preset octave band; j is the predicted sound pressure level of the j-th preset detection point in the preset octave band, n is the number of equivalent sound sources; Lw ij Lp is the sound pressure level in the preset octave band generated by the i-th equivalent sound source at the j-th preset detection point; i is the sound pressure level of the ith equivalent sound source in the preset octave band, which is an unknown quantity; d ij is the distance between the i-th equivalent sound source and the j-th preset detection point; α is the atmospheric absorption attenuation coefficient of noise during propagation.

[0081] in, like Figure 3 As shown, Figure 3 4 field points (field point 1, field point 2, field point 3 and field point 4) and an equivalent source plane are shown. There are 4 equivalent sound sources in the equivalent source plane. Figure 3 The d in represents the distance between one of the field points (detection point) and one of the equivalent sound sources.

[0082] The MSE loss function is a globally sensitive loss function used to locate Lp during the second optimization. i candidate intervals.

[0083] In this embodiment, the sound pressure level of the equivalent sound source in space is the quantity to be determined, which is first set as Lp i According to the attenuation formula of noise propagation in the free field, during the propagation process from the transformer to the detection point, each equivalent sound source only experiences geometric divergence attenuation and atmospheric absorption attenuation. The sound pressure level generated at the detection point is: Lw ij =Lp i-20lg(d ij )-0.001*α*d ij -11, where α can be obtained by looking up Table 1. When measuring the sound pressure level at the preset detection point, the temperature and relative humidity will be recorded. ij It determines the geometric divergence attenuation and atmospheric absorption attenuation during the propagation process.

[0084] The sound pressure levels generated by all equivalent sound sources at the detection point are superimposed to obtain the predicted sound pressure level at the detection point. The superposition formula is as follows:

[0085] Table 1 Atmospheric absorption attenuation coefficient

[0086]

[0087] In some embodiments, the cross entropy loss function is:

[0088]

[0089] Among them, L is the value calculated by the cross entropy loss function; P(Lp i ) is Lp i The probability of taking a value in the normal distribution function.

[0090] Cross entropy is used as a loss function and provides data for gradient descent in the iterative convergence process. Through small batch stochastic gradient descent, Lp is converged. i The value range of Lp is finally solved. i The value of is the sound pressure level of the equivalent sound source of the transformer multi-source noise equivalent model in the preset octave band.

[0091] The above calculation only calculates the sound pressure level of a single octave band of the equivalent sound source. If you want to obtain the sound pressure levels of eight octave bands of 63HZ, 125HZ, 250HZ, 500HZ, 1000HZ, 2000HZ, 4000HZ and 8000HZ, you need to build an 8-fold model and use the double-layer optimization method to perform 8 optimal solutions to obtain the sound pressure levels of the 8 octave bands of the equivalent sound source.

[0092] In S104 , a transformer multi-sound source noise equivalent model corresponding to the preset octave band is obtained according to the number of equivalent sound sources, the sound pressure level of each equivalent sound source in the preset octave band, and the spatial coordinates of each equivalent sound source.

[0093] After solving the sound pressure level of each equivalent sound source in the preset octave band, the transformer multi-source noise equivalent model corresponding to the preset octave band can be obtained based on the number of equivalent sound sources, the sound pressure level of each equivalent sound source in the preset octave band and the spatial coordinates of each equivalent sound source.

[0094] If the sound pressure levels of each equivalent sound source in eight octave bands are obtained, the final transformer multi-source noise equivalent model can be obtained.

[0095] This embodiment obtains the spatial coordinates of multiple preset detection points around the transformer and the actual sound pressure levels in the preset octave band; obtains the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source; constructs a single-variable linear regression model corresponding to the preset octave band according to the spatial coordinates of the multiple preset detection points and the actual sound pressure levels in the preset octave band, the number of equivalent sound sources and the spatial coordinates of each equivalent sound source, and solves the single-variable linear regression model corresponding to the preset octave band to obtain the sound pressure levels of multiple equivalent sound sources in the preset octave band; obtains the transformer multi-sound source noise equivalent model corresponding to the preset octave band according to the number of equivalent sound sources, the sound pressure level of each equivalent sound source and the spatial coordinates of each equivalent sound source, which can overcome the problems of complex calculation and large amount of calculation in near-field acoustic holography technology. It only needs to measure the sound pressure levels of a small number of detection points near the transformer, and obtain the transformer multi-sound source noise equivalent model based on the single-variable linear regression method. The equivalence process is convenient and simple, and the amount of calculation is small.

[0096] This example uses noise radiation attenuation characteristics to construct a univariate linear regression model. Using MSE and cross-entropy as loss functions, a two-layer optimization approach is used to determine the sound pressure level of a transformer's multi-source equivalent model. This process is simple and convenient, requiring minimal computation, and the two-layer optimization approach provides a more accurate result.

[0097] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0098] The following are device embodiments of the present invention. For details not fully described therein, reference may be made to the corresponding method embodiments described above.

[0099] Figure 4 The following is a schematic diagram showing the structure of a device for determining a transformer multi-source noise equivalent model provided by an embodiment of the present invention. For ease of explanation, only the parts related to the embodiment of the present invention are shown, which are described in detail as follows:

[0100] like Figure 4 As shown, the device 30 for determining the equivalent model of transformer multi-source noise includes: a first acquisition module 31 , a second acquisition module 32 , a solution module 33 and a model determination module 34 .

[0101] A first acquisition module 31 is configured to acquire the spatial coordinates of a plurality of preset detection points around the transformer and the actual sound pressure levels in preset octave bands;

[0102] A second acquisition module 32 is used to obtain the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source;

[0103] A solution module 33 is configured to construct a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of the plurality of preset detection points and the actual sound pressure levels in the preset octave band, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source, and solve the univariate linear regression model corresponding to the preset octave band to obtain the sound pressure levels of the plurality of equivalent sound sources in the preset octave band;

[0104] The model determination module 34 is used to obtain a transformer multi-source noise equivalent model corresponding to a preset octave band according to the number of equivalent sound sources, the sound pressure level of each equivalent sound source in a preset octave band, and the spatial coordinates of each equivalent sound source.

[0105] In a possible implementation, the solution module 33 is specifically configured to:

[0106] Constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of equivalent sound sources, and the spatial coordinates of each equivalent sound source;

[0107] According to the predicted sound pressure level of each preset detection point in the preset octave band and the actual sound pressure level of each preset detection point in the preset octave band, the univariate linear regression model corresponding to the preset octave band is solved, and the sound pressure levels of multiple equivalent sound sources in the preset octave band are obtained through double-layer optimization.

[0108] In a possible implementation, the solution module 33 is specifically configured to:

[0109] The average distribution method is used to generate the first batch of random variables as input variables for the first optimization of the single variable linear regression model corresponding to the preset octave band;

[0110] According to the MSE loss function, the first optimization is performed, and the error values calculated by the MSE loss function are selected from the first batch of random variables in ascending order, and the first preset number of random variables are arranged;

[0111] Based on the mathematical expectation and variance of the first preset number of random variables, a second batch of random variables is generated using the normal distribution method as input variables for the second optimization of the single variable linear regression model corresponding to the preset octave band;

[0112] According to the cross-entropy loss function, a second optimization is performed, and the sound pressure level of each equivalent sound source in the preset octave band is finally solved from the second batch of random variables.

[0113] In one possible implementation, the MSE loss function is:

[0114]

[0115] Among them, MSE is the error value calculated by the MSE loss function; m is the number of predicted detection points; Lm j Lw is the actual sound pressure level of the jth preset detection point in the preset octave band; j is the predicted sound pressure level of the j-th preset detection point in the preset octave band, n is the number of equivalent sound sources; Lw ij Lp is the sound pressure level in the preset octave band generated by the i-th equivalent sound source at the j-th preset detection point; i is the sound pressure level of the ith equivalent sound source in the preset octave band, which is an unknown quantity; d ij is the distance between the i-th equivalent sound source and the j-th preset detection point; α is the atmospheric absorption attenuation coefficient of noise during propagation.

[0116] In one possible implementation, the cross entropy loss function is:

[0117]

[0118] Among them, L is the value calculated by the cross entropy loss function; P(Lp i ) is Lp i The probability of taking a value in the normal distribution function.

[0119] In one possible implementation, the number of equivalent sound sources is 24;

[0120] The distribution of the equivalent sound sources is as follows: the long box walls are arranged at equal intervals according to the 4*2 specification, and the short box walls are arranged at equal intervals according to the 2*2 specification.

[0121] Figure 5 Schematic diagram of a terminal provided by an embodiment of the present invention. Figure 5 As shown, the terminal 4 of this embodiment includes: a processor 40, a memory 41, and a computer program 42 stored in the memory 41 and executable on the processor 40. When the processor 40 executes the computer program 42, the steps of the above-mentioned method for determining the equivalent model of multiple sound sources of transformer noise are implemented, for example Figure 1 Alternatively, when the processor 40 executes the computer program 42, the functions of the modules / units in the above-mentioned device embodiments are realized, for example, Figure 4 Functions of the modules / units 31 to 34 are shown.

[0122] Exemplarily, the computer program 42 may be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program 42 in the terminal 4. For example, the computer program 42 may be divided into Figure 4 Modules / units 31 to 34 are shown.

[0123] The terminal 4 can be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal 4 can include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will understand that Figure 5 It is only an example of terminal 4 and does not constitute a limitation on terminal 4. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0124] The processor 40 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0125] The memory 41 may be an internal storage unit of the terminal 4, such as a hard disk or memory of the terminal 4. The memory 41 may also be an external storage device of the terminal 4, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash memory card, etc. equipped on the terminal 4. Furthermore, the memory 41 may include both an internal storage unit of the terminal 4 and an external storage device. The memory 41 is used to store the computer program and other programs and data required by the terminal. The memory 41 may also be used to temporarily store data that has been output or is about to be output.

[0126] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0127] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0128] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0129] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical functional division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices or units, and can be electrical, mechanical, or other forms.

[0130] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0131] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0132] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can also implement all or part of the processes in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the above-mentioned method embodiments for determining the equivalent model of multiple sound sources of transformer noise. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium.

[0133] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A method for determining a transformer multi-source noise equivalent model, characterized in that: include: Obtaining the spatial coordinates of multiple preset detection points around the transformer and the actual sound pressure levels in preset octave bands; Obtain the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source; Constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure levels in the preset octave band, the number of the equivalent sound sources, and the spatial coordinates of each equivalent sound source, and solving the univariate linear regression model corresponding to the preset octave band to obtain the sound pressure levels of the multiple equivalent sound sources in the preset octave band; According to the number of equivalent sound sources, the sound pressure level of each equivalent sound source in the preset octave band and the spatial coordinates of each equivalent sound source, an equivalent model of transformer multi-sound source noise corresponding to the preset octave band is obtained.

2. The method for determining the transformer multi-source noise equivalent model according to claim 1, characterized in that: The method comprises: constructing a univariate linear regression model corresponding to the preset octave band according to the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of the equivalent sound sources, and the spatial coordinates of each equivalent sound source, and solving the univariate linear regression model corresponding to the preset octave band to obtain the sound pressure levels of the multiple equivalent sound sources in the preset octave band, including: Constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of the equivalent sound sources, and the spatial coordinates of each equivalent sound source; According to the predicted sound pressure level of each preset detection point in the preset octave band and the actual sound pressure level of each preset detection point in the preset octave band, the univariate linear regression model corresponding to the preset octave band is solved, and the sound pressure levels of multiple equivalent sound sources in the preset octave band are obtained through double-layer optimization.

3. The method for determining the transformer multi-source noise equivalent model according to claim 2, characterized in that: The method solves a single-variable linear regression model corresponding to the preset octave band based on the predicted sound pressure level of each preset detection point in the preset octave band and the actual sound pressure level of each preset detection point in the preset octave band, and obtains the sound pressure levels of multiple equivalent sound sources in the preset octave band through double-layer optimization, including: Generate a first batch of random variables using an average distribution method as input variables for the first optimization of the univariate linear regression model corresponding to the preset octave band; Performing a first optimization based on the MSE loss function, selecting from the first batch of random variables a preset number of random variables whose error values calculated by the MSE loss function are arranged in ascending order; According to the mathematical expectation and variance of the first preset number of random variables, a second batch of random variables is generated using a normal distribution method as input variables for the second optimization of the univariate linear regression model corresponding to the preset octave band; A second optimization is performed based on the cross entropy loss function, and the sound pressure level of each equivalent sound source in the preset octave band is finally obtained from the second batch of random variables.

4. The method for determining the transformer multi-source noise equivalent model according to claim 3, characterized in that: The MSE loss function is: Among them, MSE is the error value calculated by the MSE loss function; m is the number of preset detection points; Lm j Lw is the actual sound pressure level of the jth preset detection point in the preset octave band; j is the predicted sound pressure level of the j-th preset detection point in the preset octave band, Lw ij =Lp i -20lg(d ij )-0.001*α*d ij -11; n is the number of equivalent sound sources; Lw ij Lp is the sound pressure level in the preset octave band generated by the i-th equivalent sound source at the j-th preset detection point; i is the sound pressure level of the ith equivalent sound source in the preset octave band, which is an unknown quantity; d ij is the distance between the i-th equivalent sound source and the j-th preset detection point; α is the atmospheric absorption attenuation coefficient of noise during propagation.

5. The method for determining the transformer multi-source noise equivalent model according to claim 4, characterized in that: The cross entropy loss function is: Among them, L is the value calculated by the cross entropy loss function; P(Lp i ) is Lp i The probability of taking a value in the normal distribution function.

6. The method for determining a transformer multi-source noise equivalent model according to any one of claims 1 to 5, characterized in that: The number of equivalent sound sources is 24; The distribution of the equivalent sound sources is as follows: the long box walls are arranged at equal intervals according to a 4*2 specification, and the short box walls are arranged at equal intervals according to a 2*2 specification.

7. A device for determining a transformer multi-source noise equivalent model, characterized in that: include: A first acquisition module is used to obtain the spatial coordinates of multiple preset detection points around the transformer and the actual sound pressure level in a preset octave band; A second acquisition module is used to obtain the number of equivalent sound sources of the transformer and the spatial coordinates of each equivalent sound source; a solving module, configured to construct a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of the equivalent sound sources, and the spatial coordinates of each equivalent sound source, and solve the univariate linear regression model corresponding to the preset octave band to obtain the sound pressure levels of the multiple equivalent sound sources in the preset octave band; The model determination module is used to obtain the transformer multi-source noise equivalent model corresponding to the preset octave band according to the number of equivalent sound sources, the sound pressure level of each equivalent sound source in the preset octave band and the spatial coordinates of each equivalent sound source.

8. The device for determining the transformer multi-source noise equivalent model according to claim 7, characterized in that: The solution module is specifically used for: Constructing a univariate linear regression model corresponding to the preset octave band based on the spatial coordinates of multiple preset detection points and the actual sound pressure level in the preset octave band, the number of the equivalent sound sources, and the spatial coordinates of each equivalent sound source; According to the predicted sound pressure level of each preset detection point in the preset octave band and the actual sound pressure level of each preset detection point in the preset octave band, the univariate linear regression model corresponding to the preset octave band is solved, and the sound pressure levels of multiple equivalent sound sources in the preset octave band are obtained through double-layer optimization.

9. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for determining the transformer multi-source noise equivalent model as described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for determining the transformer multi-source noise equivalent model as described in any one of claims 1 to 6 are implemented.

Images