Cooperative control method and system for noise reduction and fire prevention of transformer substation

Abstract

The invention discloses a cooperative control method and system for noise reduction and fire prevention of a transformer substation. The method comprises the following steps: determining the acoustic impedance mutation degree of an original composite board according to the frequency and amplitude distribution of target low-frequency noise of the transformer substation; when the acoustic impedance abrupt change degree exceeds an abrupt change threshold value, optimizing a first physical parameter of the original composite board; the heat flow density of the first composite board obtained through optimization is obtained, and if the heat flow density exceeds a fireproof threshold value, a second physical parameter of the first composite board is optimized; obtaining a noise penetration index of the target low-frequency-band noise in the second composite board obtained through optimization, and if the noise penetration index does not exceed a penetration threshold value, optimizing a third physical parameter of the second composite board; and determining final physical parameters corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution of the third composite board obtained through optimization so as to adjust the structure of the third composite board. According to the invention, collaborative optimization of sound absorption frequency band characteristics and fireproof performance can be realized, and both noise reduction and fire prevention can reach an ideal state at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of substations, and particularly to a collaborative control method and system for noise reduction and fire prevention of a substation. Background Art

[0002] The collaborative control of in-station noise reduction and fire prevention is a key measure for the environmental protection and safe operation of power substations. Its importance lies in that it can effectively reduce the impact of the low-frequency noise of the main transformer on the surrounding environment. At the same time, it ensures the high-temperature protection ability of the equipment, providing guarantee for the long-term stable operation of the substation.

[0003] Currently, the noise reduction and fire prevention control methods of substations usually adopt single-functional materials or simple composite materials. For example, single sound-absorbing boards or fireproof coatings. And the existing methods have obvious deficiencies in taking into account the optimization of the sound-absorbing frequency band and the improvement of fireproof performance. Especially when the structure of the composite material changes, the limitations of the existing methods are mainly reflected in the insufficient control of the interfacial characteristics of the composite material, resulting in the difficulty in accurately regulating the comprehensive influence of the interfacial characteristics on the acoustic performance and thermal performance. As a result, the imbalance between the noise reduction effect and the fire prevention effect occurs, and it is difficult for both to reach the ideal state at the same time. Therefore, the core challenge currently faced is how to accurately regulate the interfacial characteristics of the composite layer to achieve the collaborative optimization of the sound-absorbing frequency band and fireproof performance, so as to meet the dual requirements of the substation for low-frequency noise control and high-temperature protection. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a collaborative control method and system for noise reduction and fire prevention of a substation. By accurately regulating the comprehensive influence of the interfacial characteristics of the composite board on the acoustic performance and thermal performance, the collaborative optimization of the sound-absorbing frequency band characteristics and fireproof performance can be achieved, so that the noise reduction effect and the fire prevention effect reach the ideal state at the same time, thereby meeting the dual requirements of the substation for low-frequency noise control and high-temperature protection.

[0005] To achieve the above purpose, the embodiments of the present invention provide a collaborative control method for noise reduction and fire prevention of a substation, including: Obtain the original data of the substation, and obtain the frequency and amplitude distribution of the target low-frequency band noise according to the original data. The original data includes equipment operation load data and sound pressure level time series data; Determine the degree of acoustic impedance mutation between adjacent two layers of media of the original composite board of the substation according to the frequency and amplitude distribution. The original composite board is composed of a cenosphere board layer and a ceramic fiber cotton layer; When the degree of acoustic impedance mutation exceeds a preset mutation threshold, optimize the first physical parameter of the original composite board to obtain a first composite board; Obtain the heat flux density of the first composite board, and when the heat flux density exceeds a preset fire prevention threshold, optimize the second physical parameter of the first composite board to obtain a second composite board; Obtain the noise penetration index of the target low-frequency band noise in the second composite board, and when the noise penetration index does not exceed a preset penetration threshold, optimize the third physical parameter of the second composite board to obtain a third composite board; Obtain the acoustic impedance distribution and thermal conductivity distribution of the third composite board, and determine the final physical parameters of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution, so as to adjust the structure of the third composite board.

[0006] To achieve the above object, an embodiment of the present invention further provides a collaborative control system for noise reduction and fire prevention in a substation, including: A noise frequency and amplitude distribution acquisition module, configured to acquire the original data of the substation and obtain the frequency and amplitude distribution of the target low-frequency band noise according to the original data, where the original data includes equipment operation load data and sound pressure level time series data; An acoustic impedance mutation degree determination module, configured to determine the acoustic impedance mutation degree between adjacent two layers of media of the original composite board of the substation according to the frequency and amplitude distribution, where the original composite board is composed of a cenosphere board layer and a refractory fiber cotton layer; A first physical parameter optimization module, configured to optimize the first physical parameter of the original composite board to obtain a first composite board when the acoustic impedance mutation degree exceeds a preset mutation threshold; A second physical parameter optimization module, configured to obtain the heat flux density of the first composite board, and when the heat flux density exceeds a preset fire prevention threshold, optimize the second physical parameter of the first composite board to obtain a second composite board; A third physical parameter optimization module, configured to obtain the noise penetration index of the target low-frequency band noise in the second composite board, and when the noise penetration index does not exceed a preset penetration threshold, optimize the third physical parameter of the second composite board to obtain a third composite board; An acoustic-thermal performance balance control module, configured to obtain the acoustic impedance distribution and thermal conductivity distribution of the third composite board, and determine the final physical parameters of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution, so as to adjust the structure of the third composite board.

[0007] Compared with the prior art, the embodiments of the present invention provide a collaborative control method and system for noise reduction and fire prevention in a substation. By precisely regulating the comprehensive influence of the interface characteristics of the composite board on the acoustic performance and thermal performance, it is possible to achieve the collaborative optimization of the sound absorption frequency band characteristics and fire prevention performance, enabling the noise reduction effect and fire prevention effect to reach the ideal state simultaneously, thereby meeting the dual requirements of the substation for low-frequency noise control and high-temperature protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 FIG. 6 is a flowchart of a collaborative control method for noise reduction and fire prevention in a substation provided by an embodiment of the present invention; Figure 2 FIG. 9 is a structural block diagram of a collaborative control system for noise reduction and fire prevention in a substation provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0009] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art in the technical field of the present invention without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0010] It should be noted that the limitations of the existing noise reduction and fire prevention control methods for substations are mainly reflected in the insufficient control of the interface characteristics of composite materials. Traditional designs often ignore key factors such as the acoustic impedance mutation, the proportion of sound energy reflection and transmission, and the reconstruction of the heat conduction path at the interface of the cenosphere board and the aluminosilicate fiber cotton composite layer. These factors directly affect the breadth of the sound absorption frequency band and the stability of the fire prevention performance, but are not fully utilized due to the lack of a systematic collaborative control strategy. Among them, the degree of acoustic impedance mutation at the interface determines the proportion of sound energy reflection and transmission at the interface, thus directly affecting the effectiveness of the sound absorption frequency band. The reconstruction of the heat conduction path changes the fire prevention performance of the composite layer and increases the design complexity. It can be seen that these technical factors are mutually coupled. If they cannot be effectively coordinated and controlled, it will lead to an imbalance between the noise reduction effect and the fire prevention ability of the substation, making it difficult to meet the dual requirements of the substation for low-frequency noise control and high-temperature protection. Therefore, how to adjust the structure of the cenosphere board and the aluminosilicate fiber cotton composite layer, optimize the heat conduction path of the acoustic impedance at the interface, and thus achieve the collaborative improvement of the sound absorption frequency band characteristics and fire prevention performance has become a key issue in the collaborative control of noise reduction and fire prevention in substations.

[0011] To solve the above problems, the embodiments of the present invention provide a collaborative control method for noise reduction and fire prevention in a substation. Refer to Figure 1As shown in the figure, it is a flowchart of a collaborative control method for noise reduction and fire prevention of a substation provided by an embodiment of the present invention. The method includes steps S11 to S16: Step S11, obtain the original data of the substation, and obtain the frequency and amplitude distribution of the target low-frequency band noise according to the original data. The original data includes equipment operation load data and sound pressure level time series data.

[0012] It should be noted that a sensor array can be used to obtain the original data from the substation according to a fixed time series. The original data includes the equipment operation load data of the substation and the sound pressure level time series data when the substation is in the operating state. Among them, the equipment operation load data characterizes the working state of the equipment in the substation, and the sound pressure level time series data characterizes the change of the noise intensity generated by the equipment in the substation at different moments during operation.

[0013] Exemplarily, in practical applications, the sensor array can adopt 8 omnidirectional microphones, which are evenly arranged around the relevant equipment in the substation at a spacing of 4 meters. The sampling frequency can be set to 4096 Hz, and the sampling duration can be set to 30 min.

[0014] It should be noted that the low-frequency band noise generated during the operation of the substation has complex spectral characteristics. Through the collection and analysis of the original data of the substation in the embodiment of the present invention, a complete noise feature recognition process can be established.

[0015] In one optional embodiment, the obtaining the frequency and amplitude distribution of the target low-frequency band noise according to the original data specifically includes: Perform denoising and normalization processing on the original data to obtain the normalized equipment operation load data and the normalized sound pressure level time series data; Segment the normalized sound pressure level time series data according to the normalized equipment operation load data, and perform band-pass filtering on the segmented sound pressure level time series data according to a preset target frequency band to obtain a target low-frequency band signal; Perform multi-scale wavelet decomposition and Hilbert transform on the target low-frequency band signal in sequence to obtain the time-frequency distribution of the target low-frequency band signal; Extract the frequency components with amplitudes exceeding a preset noise reference value from the time-frequency distribution, and perform spectral analysis using fast Fourier transform to obtain the frequency and amplitude distribution of the target low-frequency band noise.

[0016] Specifically, when obtaining the frequency and amplitude distribution of the target low-frequency band noise based on the original data of the substation, the original data can be denoised first to remove outliers, and the denoised original data can be normalized (for example, normalized using the maximum and minimum value normalization method), and corresponding standardized equipment operation load data and standardized sound pressure level time series data are obtained; then, the standardized sound pressure level time series data is segmented according to the standardized equipment operation load data, and the segmented sound pressure level time series data is band-pass filtered according to the preset target frequency band to extract the target low-frequency band signal; thereafter, the extracted target low-frequency band signal can be subjected to multi-scale wavelet decomposition using a wavelet decomposition method, and detail coefficients at multiple scales are obtained (these detail coefficients represent signal components of different frequencies), and the detail coefficients at multiple scales are subjected to Hilbert transform. For the detail coefficient sequence x(t) at any scale, its Hilbert transform H[x(t)] can be defined as: , Represents the convolution operation. Furthermore, a complex analytical signal z(t) can be constructed based on x(t) and its Hilbert transform H[x(t)], and we have: , j represents the imaginary unit, and the instantaneous frequency ω(t) and instantaneous amplitude A(t) of the target low-frequency band signal are calculated based on the analytical signal z(t), and: , , represents the instantaneous phase, and: , according to the instantaneous frequency ω(t) and instantaneous amplitude A(t), the time-frequency distribution of the target low-frequency band signal can be obtained; finally, the frequency components whose amplitudes exceed the preset noise baseline value are extracted from the time-frequency distribution, and the fast Fourier transform (FFT) is used to perform spectral analysis on the extracted frequency components to obtain the frequency and amplitude distribution of the target low-frequency band noise accordingly.

[0017] It should be noted that in the preprocessing stage of the raw data, the raw data can be denoised to eliminate abnormal values that exceed the normal range. For example, when the collected instantaneous value of the sound pressure level exceeds 120dB or is lower than 30dB, it is judged as an abnormal value and eliminated. For the preprocessed raw data, the maximum and minimum value normalization method can be used for normalization, and the numerical range of the preprocessed raw data can be uniformly mapped to between 0 and 1 for subsequent analysis and processing.

[0018] It should be noted that when segmenting the standardized sound pressure level time series data based on the standardized device operating load data, different load rate intervals can be defined. For example, the interval with a load rate greater than 80% is divided into the heavy load segment, the interval with a load rate between 50% and 80% is divided into the medium load segment, and the interval with a load rate less than 50% is divided into the light load segment; for the sound pressure level data at each time point in the sound pressure level time series data, find the corresponding device operating load data (i.e., the load rate at that time point). Once the load rates at all time points are determined, the sound pressure level time series data can be segmented according to the above-defined load rate intervals. For example, if the load rate at a certain time point belongs to the heavy load segment, the sound pressure level data at that time point is classified as data in the heavy load segment. Similarly, classify the data in other load rate intervals.

[0019] Exemplarily, in the frequency analysis stage, the segmented sound pressure level time series data can be first band-pass filtered from 20 Hz to 200 Hz to extract the target low-frequency band signal, and then the wavelet decomposition method is used to perform 5-layer wavelet decomposition on the extracted target low-frequency band signal to obtain the detail coefficients of different frequency bands. Then, the instantaneous frequency and instantaneous amplitude are calculated through Hilbert transform to construct a time-frequency distribution diagram, which can intuitively display the variation characteristics of the low-frequency band noise frequency over time. After that, when the amplitude of a certain frequency component exceeds the environmental background noise reference value by more than 5 dB, it can be extracted as a characteristic frequency component, and the fast Fourier transform is used to perform spectrum analysis on the extracted characteristic frequency components to obtain the frequency-amplitude distribution.

[0020] It should be noted that in practical applications, through observation, it can be seen that the transformer in the substation mainly generates low-frequency noises of 100 Hz and 150 Hz during heavy load operation, which corresponds to the vibration frequency of the transformer core. The main frequency components can be identified through the peak detection algorithm, and feature extraction is combined with the environmental background noise. The energy ratio and signal-to-noise ratio of each frequency point can also be calculated. For example, in a certain measurement, the energy ratio of the 100 Hz frequency component reaches 35%, and the signal-to-noise ratio is 15 dB. The energy ratio of the 150 Hz frequency component is 25%, and the signal-to-noise ratio is 12 dB. These characteristic values constitute the low-frequency noise characteristic description vector under this working condition.

[0021] Step S12: Determine the degree of acoustic impedance mutation between adjacent two layers of media of the original composite board of the substation according to the frequency and amplitude distribution. The original composite board is composed of a cenosphere board layer and a silica aluminum fiber cotton layer.

[0022] It should be noted that the original composite board of the substation includes a cenosphere board layer and a silica aluminum fiber cotton layer. The cenosphere board layer and the silica aluminum fiber cotton layer are stacked to form a composite layer structure, and this composite layer structure has unique acoustic and thermal characteristics in terms of noise reduction and fire prevention.

[0023] In one of the optional embodiments, determining the degree of acoustic impedance mutation between adjacent two layers of media of the original composite board of the substation according to the frequency and amplitude distribution specifically includes: Constructing a three-dimensional structure of the original composite board according to the frequency and amplitude distribution and in combination with the original physical parameters of the original composite board of the substation; Calculating the relative change rate of the acoustic impedance between the perlite board layer and the aluminosilicate fiber cotton layer according to the interfacial acoustic impedance between the perlite board layer and the aluminosilicate fiber cotton layer in the three-dimensional structure; Determining the degree of acoustic impedance mutation between the perlite board layer and the aluminosilicate fiber cotton layer according to the relative change rate.

[0024] Specifically, in combination with the above embodiments, when determining the degree of acoustic impedance mutation between adjacent two layers of media of the original composite board according to the frequency and amplitude distribution of the target low-frequency band noise, first, according to the frequency and amplitude distribution data of the target low-frequency band noise and in combination with the original physical parameters of the original composite board of the substation (including the density, elastic modulus, and Poisson's ratio of the perlite board layer, and the density, elastic modulus, and Poisson's ratio of the aluminosilicate fiber cotton layer), adopt the tetrahedral mesh generation method to construct a three-dimensional structure of the original composite board, and set the pore diameter of the perlite board layer and the fiber arrangement density of the aluminosilicate fiber cotton layer; then, according to the interfacial acoustic impedance of the perlite board layer and the interfacial acoustic impedance of the aluminosilicate fiber cotton layer in the three-dimensional structure, calculate the relative change rate of the acoustic impedance between the perlite board layer and the aluminosilicate fiber cotton layer. Among them, the acoustic impedance Z can be defined as: Z = ρ×c, where ρ represents the medium density and c represents the propagation speed of sound waves in the medium (i.e., the speed of sound), reflecting the ability of the medium to impede the propagation of sound waves. Assuming the acoustic impedance of the perlite board layer is Z1 and the acoustic impedance of the aluminosilicate fiber cotton layer is Z2, then the relative change rate of the acoustic impedance between these two layers of media, the perlite board layer and the aluminosilicate fiber cotton layer, is: ; After that, the degree of acoustic impedance mutation between the perlite board layer and the aluminosilicate fiber cotton layer can be determined according to the calculated relative change rate of the acoustic impedance. It can be understood that the greater the relative change rate of the acoustic impedance, the greater the degree of acoustic impedance mutation between the perlite board layer and the aluminosilicate fiber cotton layer.

[0025] Exemplarily, the density of the perlite board layer can be set between 300 and 500 kg / m 3 ³, the elastic modulus of the perlite board layer can be set between 0.5 and 2 GPa, and the Poisson's ratio of the perlite board layer can be set to 0.25; the density of the aluminosilicate fiber cotton layer can be set between 80 and 150 kg / m 3 ³, the elastic modulus of the aluminosilicate fiber cotton layer can be set between 0.1 and 0.3 GPa, and the Poisson's ratio of the aluminosilicate fiber cotton layer can be set to 0.3; these original physical parameters jointly determine the acoustic performance and thermal performance of the original composite board.

[0026] It should be noted that when constructing the three-dimensional structure of the original composite board by using the tetrahedral mesh generation method, the selection of the mesh size has an important impact on the subsequent calculation accuracy. For example, for the cenosphere board layer, when its pore size distribution is in the range of 0.1 to 0.5 mm, the mesh size can be set to 1 / 6 of the minimum pore size, that is, about 0.017 mm. For the aluminosilicate fiber cotton layer, when the fiber arrangement density is 300 to 800 fibers per square centimeter, the mesh size can be set to 1 / 4 of the fiber diameter, about 0.008 mm. Determining the mesh size based on the characteristic sizes of the cenosphere board layer and the aluminosilicate fiber cotton layer ensures both calculation accuracy and avoids excessive calculation volume.

[0027] It should be noted that when sound waves propagate in the original composite board, the incident angle has a significant impact on the sound energy transfer effect. For example, when the incident angle is in the range of 0° to 30°, the sound waves mainly propagate along the normal direction. At this time, the acoustic impedance of the cenosphere board layer is about 0.8 MPa·s / m, and the acoustic impedance of the aluminosilicate fiber cotton layer is about 0.15 MPa·s / m. When the incident angle increases to the range of 30° to 60°, the propagation path of the sound waves extends, and the number of interface reflections increases. The equivalent acoustic impedances of the two media increase to 1.2 MPa·s / m and 0.25 MPa·s / m respectively. In the analysis of the interface acoustic impedance, the relative change rate of the acoustic impedance between adjacent media reflects the degree of obstruction of sound energy transfer. Taking the incident angle of 45° as an example, when the relative change rate of the acoustic impedance at the interface exceeds 0.5, the interface stress distribution shows obvious discontinuity, and the stress concentration coefficient reaches 2.5. In this case, the sound energy reflection ratio at the interface increases significantly.

[0028] Exemplarily, the propagation of sound energy in the original composite board follows the principle of energy conservation. The calculated sound energy distribution characteristics show that at a frequency of 100 Hz, when the incident sound pressure level is 80 dB, the reflection coefficient of the first layer interface is about 0.4, and the transmission coefficient is 0.6. The reflection coefficient of the second layer interface is about 0.3, and the transmission coefficient is 0.7. After considering the nonlinear characteristics of the medium material, when the frequency increases to 200 Hz, the degree of sudden change of the interface acoustic impedance increases, resulting in the reflection coefficient of the first layer interface increasing to 0.5 and the transmission coefficient decreasing to 0.5. The reflection coefficient of the second layer interface increases to 0.4 and the transmission coefficient decreases to 0.6. This frequency dependence indicates that the composite layer structure of the original composite board has a selective attenuation effect on different frequency noises.

[0029] Step S13: When the degree of sudden change of the acoustic impedance exceeds a preset sudden change threshold, optimize the first physical parameter of the original composite board to obtain the first composite board.

[0030] It should be noted that after obtaining the degree of acoustic impedance mutation between the cenosphere board layer and the aluminum silicate fiber cotton layer (i.e., the relative change rate of the acoustic impedance between the two layers of media), the degree of acoustic impedance mutation between the cenosphere board layer and the aluminum silicate fiber cotton layer can be compared with a preset mutation threshold. If it is determined that the degree of acoustic impedance mutation exceeds the preset mutation threshold, the first physical parameter of the original composite board can be optimized, and a composite board with optimized first physical parameters can be obtained accordingly, which is used as the first composite board.

[0031] In one optional embodiment, the first physical parameter includes the porosity of the cenosphere board layer and the density of the aluminum silicate fiber cotton layer; Then, optimizing the first physical parameter of the original composite board to obtain the first composite board specifically includes: According to the preset adjustment range and adjustment amplitude of the porosity of the cenosphere board layer and the preset adjustment range and adjustment amplitude of the density of the aluminum silicate fiber cotton layer, a first parameter set is obtained; Several groups of parameter combinations that meet the lower limit value of the compressive strength of the cenosphere board layer and the upper limit value of the elastic modulus of the aluminum silicate fiber cotton layer are screened out from the first parameter set to obtain a second parameter set; Several groups of parameter combinations that meet the total thickness range and the interlayer bonding area ratio of the original composite board are screened out from the second parameter set to obtain a third parameter set; For each group of parameter combinations in the third parameter set, the ratio of the adjacent interface acoustic impedance is calculated as the acoustic impedance matching degree; The parameter combinations in the third parameter set are iteratively optimized according to the acoustic impedance matching degree. During the iterative optimization process, the adjustment amplitude of the porosity of the cenosphere board layer and the adjustment amplitude of the density of the aluminum silicate fiber cotton layer decrease as the number of iterations increases; The porosity of the cenosphere board layer and the density of the aluminum silicate fiber cotton layer are optimized according to the final parameter combination to obtain the first composite board.

[0032] Specifically, in combination with the above embodiments, the first physical parameter of the original composite board includes the porosity of the cenosphere board layer and the density of the aluminum silicate fiber cotton layer. Correspondingly, when actually optimizing the first physical parameter of the original composite board, the preset adjustment range of the porosity of the cenosphere board layer (for example, floating 10% up and down from the original value) and the adjustment amplitude (i.e., the adjustment step, which is the initial value, for example, the initial adjustment step is 0.5%), the preset adjustment range of the density of the aluminum silicate fiber cotton layer (for example, floating 20% up and down from the original value) and the adjustment amplitude (i.e., the adjustment step, which is the initial value, for example, the initial adjustment step is 2 kg / m 3( ), a first parameter set is correspondingly obtained. The first parameter set includes multiple groups of parameter combinations, and each group of parameter combinations consists of a porosity of the cenosphere board layer (a certain porosity value within the adjusted range of porosity) and a density of the aluminum silicate fiber cotton layer (a certain density value within the adjusted range of density). For the first parameter set, several groups of parameter combinations that meet the following physical constraint conditions are selected therefrom: the compressive strength of the cenosphere board layer needs to be maintained above the lower limit value of the compressive strength, and the elastic modulus of the aluminum silicate fiber cotton layer needs to be maintained below the upper limit value of the elastic modulus. Correspondingly, a second parameter set is obtained. Among them, for each group of parameter combinations in the first parameter set, the actual compressive strength of the corresponding cenosphere board layer and the actual elastic modulus of the aluminum silicate fiber cotton layer are determined, and it is judged whether the actual compressive strength of the cenosphere board layer and the actual elastic modulus of the aluminum silicate fiber cotton layer corresponding to each group of parameter combinations meet the above physical constraint conditions. If they meet, the corresponding parameter combination is retained; otherwise, the corresponding parameter combination is excluded. Thus, the second parameter set is composed of all the retained parameter combinations. For the second parameter set, several groups of parameter combinations that meet the following geometric constraint conditions are selected therefrom: the total thickness of the original composite board should be within a certain range (for example, between 50 and 80 mm), and the interlayer bonding area ratio should exceed a certain ratio value (for example, exceed 85%). Correspondingly, a third parameter set is obtained. Among them, the total thickness of the original composite board is the sum of the thickness of the cenosphere board layer and the thickness of the aluminum silicate fiber cotton layer, and the interlayer bonding area ratio is the ratio of the contact area between the cenosphere board layer and the aluminum silicate fiber cotton layer to the total area. Correspondingly, for each group of parameter combinations in the second parameter set, the actual total thickness and the actual interlayer bonding area ratio of the corresponding composite board are calculated, and it is judged whether the actual total thickness and the actual interlayer bonding area ratio corresponding to each group of parameter combinations meet the above geometric constraint conditions. If they meet, the corresponding parameter combination is retained; otherwise, the corresponding parameter combination is excluded. Thus, the third parameter set is composed of all the retained parameter combinations. For each group of parameter combinations in the third parameter set, according to the density ρ1 and the sound velocity c1 of the cenosphere board layer, the acoustic impedance of the corresponding cenosphere board layer is calculated as: Z1 = ρ1 × c1. Similarly, according to the density and the sound velocity of the aluminum silicate fiber cotton layer, the acoustic impedance of the corresponding aluminum silicate fiber cotton layer is calculated as Z2. Then, according to the acoustic impedance Z1 of the cenosphere board layer and the acoustic impedance Z2 of the aluminum silicate fiber cotton layer corresponding to each group of parameter combinations, the ratio of the adjacent interface acoustic impedance can be calculated as: Ratio = Z2 / Z1, and the calculated ratio is used as the acoustic impedance matching degree. Then, the parameter combinations in the third parameter set are iteratively optimized. In each iteration, the parameter update direction can be calculated based on the acoustic impedance matching degree, and the adjustment amplitude of the porosity of the cenosphere board layer and the adjustment amplitude of the density of the aluminum silicate fiber cotton layer decrease with the increase of the iteration times. For example, the initial adjustment amplitude of the porosity of the cenosphere board layer is 0.5%, starting from 0.It starts to decrease as the number of iterations increases from 5%. After the iteration process ends, a set of final parameter combinations can be obtained. The final parameter combinations include the porosity of the cenosphere board layer and the density of the aluminosilicate fiber cotton layer determined through iterative optimization. Finally, the current porosity of the cenosphere board layer and the current density of the aluminosilicate fiber cotton layer are adjusted according to the obtained final parameter combinations, and the first composite board is correspondingly obtained.

[0033] It should be noted that in the actual iterative optimization process, for any set of parameter combinations in the third parameter set, the acoustic impedance matching degree under this set of parameter combinations can be calculated first, and then the influence of the changes of each parameter (that is, the porosity of the cenosphere board layer and the density of the aluminosilicate fiber cotton layer) on the acoustic impedance matching degree can be estimated. Assuming that the parameter vector composed of each parameter is X, the gradient of the acoustic impedance matching degree function f(X) That is, it represents the direction and rate of change of the acoustic impedance matching degree with the changes of each parameter. Therefore, the parameters can be updated according to the gradient information, and there is: , X old represents the parameter vector before update, η represents the learning rate, which is used to control the step size of each update, and X new represents the parameter vector after update; further, when any of the following conditions is met, the iteration process ends: (1) Reaching the maximum number of iterations: A maximum number of iterations is preset in advance. When the actual number of iterations reaches the maximum number of iterations, the iteration stops; (2) When the changes in the acoustic impedance matching degree in several consecutive rounds of iterations are all less than a certain threshold, it is considered that convergence has been achieved and the iteration stops; for example, if the changes in the acoustic impedance matching degree in five consecutive iterations are all less than 0.001, it is considered that the convergence criterion has been reached and the iteration stops; (3) When the change amplitude of the parameters is less than a certain threshold, the iteration stops; for example, if the changes in the porosity of the cenosphere board layer and the density of the aluminosilicate fiber cotton layer are both less than 0.01%, it is considered that the iterative optimization process is completed and the iteration stops.

[0034] Exemplarily, when the degree of acoustic impedance mutation exceeds the preset mutation threshold, the acoustic energy reflection at the surface interface is relatively strong, and it is necessary to optimize and adjust the structure of the original composite board; taking the low-frequency noise control of a certain substation as an example, the porosity of the cenosphere board layer in the original composite board is 45%, and the density of the aluminosilicate fiber cotton layer is 120 kg / m 3 , the degree of acoustic impedance mutation is 0.75, and the preset mutation threshold is 0.6. At this time, the degree of acoustic impedance mutation exceeds the preset mutation threshold, so it is necessary to adjust the porosity of the cenosphere board layer and the density of the aluminosilicate fiber cotton layer; during the parameter adjustment process, the adjustment range of the porosity of the cenosphere board layer is limited between 40.5% and 49.5%, and the adjustment step size each time is 0.5%; the adjustment range of the density of the aluminosilicate fiber cotton layer is between 96 and 144 kg / m3 between, the adjustment step size is 2 kg / m 3 ; In terms of physical parameter constraints, the compressive strength of the cenosphere board layer needs to be maintained above 2.5 MPa, and the elastic modulus of the aluminosilicate fiber cotton layer does not exceed 0.3 GPa. These physical constraints ensure that the cenosphere board layer and the aluminosilicate fiber cotton layer still have sufficient mechanical strength after adjusting the porosity and density; In terms of geometric parameter constraints, the total thickness of the original composite board is controlled within the range of 50 to 80 mm, and the interlayer bonding area ratio is maintained above 85%; When the porosity of the cenosphere board layer is adjusted to 47%, its compressive strength drops to 2.8 MPa, still meeting the compressive strength limit requirements. At the same time, the density of the aluminosilicate fiber cotton layer is increased to 130 kg / m 3 , the corresponding elastic modulus is 0.25 GPa, still meeting the elastic modulus limit requirements. And under this parameter combination, the acoustic impedance ratio of the adjacent interface is reduced from the original 2.8 to 1.8. The acoustic wave propagation calculation shows that for the optimized composite layer structure (i.e., the first composite board) at a frequency of 100 Hz, the acoustic energy reflection ratio is reduced from the original 0.45 to 0.35, and the transmission ratio is increased from the original 0.55 to 0.65.

[0035] Furthermore, assume that at the 10th iteration, the porosity of the cenosphere board layer is adjusted to 48%, and the density of the aluminosilicate fiber cotton layer is adjusted to 135 kg / m 3 , at this time, the degree of acoustic impedance mutation is reduced to 0.55, lower than the preset mutation threshold. At the same time, the total thickness of the composite board corresponding to this parameter combination is 75 mm, and the interlayer bonding area ratio is 88%, then all geometric parameter constraint requirements are met; For the determined physical parameter combination, while maintaining sufficient mechanical strength, the acoustic impedance matching degree of the optimized composite layer structure is increased by 35%. In the frequency band of 80 to 200 Hz, the average acoustic energy reflection ratio is reduced to 0.32, and the transmission ratio is increased to 0.68, indicating that the optimized composite layer structure can more effectively reduce low-frequency noise.

[0036] Step S14, obtain the heat flux density of the first composite board, and when the heat flux density exceeds the preset fire protection threshold, optimize the second physical parameters of the first composite board to obtain the second composite board.

[0037] It should be noted that after obtaining the first composite board through parameter adjustment, the heat flux density distribution inside the structure of the first composite board can be further obtained, and the obtained heat flux density is compared with the preset fire protection threshold to determine whether the first composite board meets the fire protection requirements, and output a fire protection performance evaluation result including the heat flux density distribution characteristics. If it is determined that the heat flux density exceeds the preset fire protection threshold, it means that the fire protection requirements are not met at this time, then the second physical parameters of the first composite board need to be optimized, and the composite board with optimized second physical parameters is obtained accordingly and used as the second composite board.

[0038] In one optional embodiment, obtaining the heat flux density of the first composite board specifically includes: Establish a three-dimensional grid cell of the first composite board, and add the thermal conductivity, density, and specific heat capacity of the cenosphere board layer and the aluminosilicate fiber cotton layer to the grid cell property table; Taking the external heat source temperature and the ambient temperature as the first-kind boundary conditions, and taking the convective heat transfer coefficient as the third-kind boundary conditions, establish a heat conduction equation in each grid cell; Use the finite volume method to discretely solve the heat conduction equation to obtain the temperature field distribution; According to the temperature field distribution, use the central difference method to obtain the temperature gradient; According to the temperature gradient, use the heat streamline tracing algorithm to identify the heat conduction path and obtain the heat flux density on the heat conduction path.

[0039] Specifically, in combination with the above embodiments, when obtaining the heat flux density of the first composite board, first, according to the composite layer structure characteristics of the first composite board, take the cenosphere board layer and the aluminosilicate fiber cotton layer as a whole, and use a finite element analysis software (such as ANSYS, COMSOL, etc.) to establish a three-dimensional grid cell including the cenosphere board layer and the aluminosilicate fiber cotton layer, and input the thermal conductivity, density, and specific heat capacity of the cenosphere board layer, as well as the thermal conductivity, density, and specific heat capacity of the aluminosilicate fiber cotton layer into the grid cell property table; then set the external heat source temperature and the ambient temperature as the first-kind boundary conditions, set the convective heat transfer coefficient as the third-kind boundary conditions, and establish a heat conduction equation in each grid cell of the three-dimensional grid cell, and the equation is: , ρ represents the medium density (kg / m 3 ), c p represents the specific heat capacity (J / kg·K), T represents the temperature (K), t represents the time (s), represents the gradient operator, k represents the thermal conductivity (W / m·K), Q represents the internal heat source per unit volume (W / m 3 ), if there is no internal heat source, then let Q = 0; then, use the finite volume method to discretely solve this heat conduction equation, and correspondingly obtain the temperature field distribution, where the entire calculation domain (i.e., the three-dimensional grid cell) is divided into multiple small control volumes (i.e., multiple grid cells), and for the heat conduction equation in any one control volume (denoted as the i-th control volume), after discretization, it can be expressed as: , V i represents the volume of the i-th control volume, T i nDenote the temperature of the \(i\)-th control volume at the \(n\)-th time step, \(\Delta t\) represents the time step (i.e., the time difference between the \((n + 1)\)-th time step and the \(n\)-th time step), \(f\) represents the adjacent control volume of the \(i\)-th control volume, \(k\) f and \(A\) f represent the thermal conductivity and the interface area between the \(i\)-th control volume and the adjacent control volume \(f\) respectively, \(T\) f denotes the temperature of the adjacent control volume \(f\), \(T\) i denotes the temperature of the \(i\)-th control volume, \(\Delta x\) f represents the characteristic length (such as the grid spacing) from the \(i\)-th control volume to the adjacent control volume \(f\), \(Q\) i denotes the heat source term within the \(i\)-th control volume. Further, set the relative change rate of the temperature field to be less than 0.001 as the convergence criterion, that is, stop the iteration when the maximum temperature change between two consecutive iterations satisfies the following condition: , after the iteration stops, a stable temperature field distribution \(T(x, y, z)\) can be obtained, where \((x, y, z)\) represents the spatial coordinates; then, based on the obtained temperature field distribution \(T(x, y, z)\), the corresponding temperature gradient can be obtained using the central difference method. For the three-dimensional case, the temperature gradient can be expressed as: ; finally, based on the calculated temperature gradient , the main heat conduction paths in the three-dimensional grid cells can be identified using the heat streamline tracing algorithm, and the heat flux density on the main heat conduction paths can be obtained. Among them, the heat streamline is a curve starting from a point and extending along the direction of the local temperature gradient. Starting from the initial point, it advances step by step along the direction of the local temperature gradient until it reaches the boundary or other specified end points; along the main heat conduction paths, the corresponding heat flux density \(q\) can be calculated at each point on the path according to the formula .

[0040] It should be noted that the heat conduction characteristics of the composite layer structure directly affect its fire protection performance. Taking a substation fireproof and sound insulation barrier (i.e., composite board) as an example, the thermal conductivity of the cenosphere board layer is 0.065 W / m·K, the density is 400 kg / m 3 , the specific heat capacity is 840 J / kg·K, the thermal conductivity of the aluminum silicate fiber cotton layer is 0.035 W / m·K, the density is 120 kg / m 3 , the specific heat capacity is 1200 J / kg·K; when establishing the three-dimensional grid cells, the grid size is set to 2 mm, and the total number of grids reaches 120,000; in the boundary condition setting, the external heat source temperature is set to 850 °C, the ambient temperature is 25 °C, as the first kind of boundary condition, the convective heat transfer coefficient of the outer surface is set to 25 W / (m 2·K) as the third type of boundary condition; for the calculation of the initial temperature field, a time step of 0.1 s is adopted. A heat conduction equation including heat conduction and convection terms is established in each grid cell. When solving by the finite volume method, the central difference scheme is used to discretize the spatial terms, and the implicit scheme is used to discretize the time terms. The relative change rate of the temperature field is set to be less than 0.001 as the convergence criterion; the calculation results show that under steady-state conditions, the highest temperature on the outer surface of the composite layer structure is 780 °C, and the highest temperature on the inner surface is 120 °C; through heat streamline tracing, it is found that there are three main heat transfer channels in the composite layer structure, which are located in the upper, middle, and lower regions respectively; the calculation of the heat flux density distribution shows that on the main heat conduction path, the maximum heat flux density on the outer surface reaches 15 kW / m 2 , while at the interface between the cenosphere board layer and the aluminosilicate fiber cotton layer, due to the existence of contact thermal resistance, the heat flux density is reduced to 8 kW / m 2 .

[0041] It should be noted that the fire performance evaluation adopts the heat flux density threshold method, and the preset fire threshold is 10 kW / m 2 . The analysis results show that under the action of an external heat source of 850 °C, only 15% of the area on the outer surface has a heat flux density exceeding the preset fire threshold, and these areas are mainly concentrated on the directly heated surface position. Inside the composite layer structure, due to the contact thermal resistance at the interface and the low thermal conductivity of the medium material itself, the heat flux density is lower than the preset fire threshold, indicating that the composite layer structure has good fire performance.

[0042] In one optional embodiment, the second physical parameter includes the porosity of the cenosphere board layer and the thickness of the aluminosilicate fiber cotton layer; Then, optimizing the second physical parameter of the first composite board to obtain the second composite board specifically includes: Mark the area where the heat flux density exceeds the preset fire threshold on the heat conduction path, and obtain the porosity adjustment amount of the cenosphere board layer and the thickness adjustment amount of the aluminosilicate fiber cotton layer corresponding to this area; According to the porosity adjustment amount and the thickness adjustment amount, update the three-dimensional grid cells of the first composite board. The grid size of the updated three-dimensional grid cells is determined by the minimum characteristic size of the cenosphere board layer and the aluminosilicate fiber cotton layer; Identify the new heat conduction path according to the updated three-dimensional grid cells, and obtain the new heat flux density on the new heat conduction path; Determine whether the new heat flux density exceeds a preset fire prevention threshold. If so, re-optimize the second physical parameters of the first composite board until the obtained new heat flux density does not exceed the preset fire prevention threshold. If not, optimize the porosity of the cenosphere board layer and the thickness of the aluminosilicate fiber cotton layer according to the currently obtained porosity adjustment amount and thickness adjustment amount to obtain a second composite board.

[0043] Specifically, in combination with the above embodiments, the second physical parameters of the first composite board include the porosity of the cenosphere board layer and the thickness of the aluminosilicate fiber cotton layer. Correspondingly, when actually optimizing the second physical parameters of the first composite board, on the identified heat conduction path, the heat streamline tracing method can be used to mark the area where the heat flux density exceeds the preset fire prevention threshold, and obtain the porosity adjustment amount of the cenosphere board layer and the thickness adjustment amount of the aluminosilicate fiber cotton layer corresponding to the marked area. For example, assuming that the initial porosity of the cenosphere board layer is P0, a reasonable adjustment ratio is determined through experiments or empirical data, preferably 1.2. Then, the porosity adjustment amount is: △P=(1.2 - 1)×P0 = 0.2×P0. Assuming that the initial thickness of the aluminosilicate fiber cotton layer is d0, in order to reduce the heat flux density below the preset fire prevention threshold, the thickness needs to be increased. The required increased thickness (i.e., the thickness adjustment amount) can be estimated through the thermal resistance formula R = d / (k×A) as: △d=(R target - R0)×k×A, where R target represents the target thermal resistance, R0 represents the initial thermal resistance, k represents the thermal conductivity, and A represents the heat transfer area; according to the obtained porosity adjustment amount of the cenosphere board layer and the thickness adjustment amount of the aluminosilicate fiber cotton layer, the new porosity of the cenosphere board layer and the new thickness of the aluminosilicate fiber cotton layer can be obtained. Based on the new porosity and new thickness, update the three-dimensional grid cells of the first composite board established in the above embodiments to correspondingly obtain updated three-dimensional grid cells. Among them, the grid size of the updated three-dimensional grid cells is determined by the minimum characteristic size of the cenosphere board layer and the aluminosilicate fiber cotton layer (for example, the pore diameter of the cenosphere board layer or the fiber diameter of the aluminosilicate fiber cotton layer) to ensure that the updated grid size can adapt to the change of the medium material properties. For example, for high-precision simulation, the grid size can usually be set to 1 / 4 to 1 / 6 of the minimum characteristic size; then, a new heat conduction path can be identified according to the updated three-dimensional grid cells, and the new heat flux density on the new heat conduction path can be obtained. For example, apply the steady-state heat conduction equation on the updated three-dimensional grid cells , perform numerical solution in combination with the boundary conditions, and again according to the formula Calculate the corresponding heat flux density q at each point on the path (the specific calculation process is similar to that in the above embodiment and will not be elaborated here); further, compare the newly obtained heat flux density with the preset fire prevention threshold to determine whether the first composite board after the update of the current second physical parameter meets the fire prevention requirements. If it is determined that the newly obtained heat flux density still exceeds the preset fire prevention threshold, it means that the first composite board after the update of the current second physical parameter still does not meet the fire prevention requirements, and it is necessary to return to re-optimize the second physical parameter of the first composite board (that is, re-execute the optimization process of the second physical parameter in this embodiment) until it is determined that the newly obtained heat flux density does not exceed the preset fire prevention threshold. Correspondingly, if it is determined that the newly obtained heat flux density does not exceed the preset fire prevention threshold, adjust the porosity of the perlite board layer and the thickness of the aluminosilicate fiber cotton layer of the first composite board according to the porosity adjustment amount of the perlite board layer and the thickness adjustment amount of the aluminosilicate fiber cotton layer obtained currently, and correspondingly obtain the second composite board.

[0044] Exemplarily, if the pore size distribution of the perlite board layer is between 0.1 and 0.5 mm, the grid size of the updated three-dimensional grid unit can be set to 0.025 mm (which is 1 / 4 of 0.1 mm). If the fiber diameter of the aluminosilicate fiber cotton layer is 0.03 mm, the grid size of the updated three-dimensional grid unit can be set to 0.0075 mm (which is 1 / 4 of 0.03 mm).

[0045] It should be noted that in the optimization of the fire prevention performance of the composite layer structure, the treatment of the heat flux density exceeding the preset fire prevention threshold involves multiple key links. Taking a substation fire prevention and sound insulation barrier as an example, when it is detected that the heat flux density on the outer surface reaches 12 kW / m 2 , exceeding the fire prevention threshold of 10 kW / m 2 , it is necessary to adjust and optimize the composite layer structure; through heat flow line tracing, it is found that the area with excessive heat flux density is mainly concentrated in the middle area, and the coverage area accounts for about 25% of the total area; for the excessive area, the partition thickening treatment method is adopted. In terms of the thickness optimization of the aluminosilicate fiber cotton layer, the thickness of the aluminosilicate fiber cotton layer in the excessive area is increased from the original 50 mm to 75 mm, and the thickening coefficient is 1.5. The calculation of the heat insulation performance after thickening shows that the thermal resistance of this area is increased from the original 1.4 K·m² / W to 2.1 K·m² / W, and the heat transfer conduction time is extended by 45%; in terms of the porosity optimization of the perlite board layer, the density gradient method is used to adjust the corresponding thickening area, and the porosity is increased from the original 45% to 54%, and the increase coefficient is 1.2. After the porosity is increased, the thermal conductivity of the perlite board layer is reduced to 0.052 W / m·K, and the compressive strength still remains above 2.8 MPa, meeting the structural compressive strength requirements; after the structure is adjusted, it is necessary to recalculate the interface heat transfer characteristics, and the contact heat transfer coefficient of the adjusted area is reduced from the original 80 W / (m2 ·K) is reduced to 65 W / (m 2 ·K), and this reduction is due to the change in the interfacial contact pressure of the dielectric material.

[0046] Furthermore, in the meshing of the new three-dimensional grid cells, the grid size of the thickened area can remain unchanged at 2 mm, but the number of grids increases to 150,000; the steady-state heat conduction calculation results show that under the action of an external heat source at 850 °C, the maximum heat flux density of the composite layer structure after parameter adjustment is reduced to 9.5 kW / m 2 , which is lower than the fire protection threshold of 10 kW / m 2 ; the temperature field distribution shows an obvious stepped shape, the highest temperature on the outer surface is reduced to 720 °C, and the highest temperature on the inner surface is reduced to 95 °C.

[0047] It can be understood that in practical applications, the parameter optimization of the composite layer structure often requires multiple rounds of iteration. If the heat flux density in some local areas still exceeds the fire protection threshold after the first round of optimization, a second round of optimization is required, and so on. As the number of iterations increases, the thickening coefficient of the thickness of the aluminosilicate fiber cotton layer and the improvement coefficient of the porosity of the cenosphere board layer are usually appropriately increased; through this step-by-step optimization method, the fire protection performance of the composite layer structure meets the design requirements; optimization experience shows that the thickening coefficient should not exceed 2.0, and the porosity improvement coefficient should not exceed 1.5, otherwise it will affect the overall performance of the composite layer structure.

[0048] Step S15: Obtain the noise penetration index of the target low-frequency band noise in the second composite board, and when the noise penetration index does not exceed the preset penetration threshold, optimize the third physical parameter of the second composite board to obtain the third composite board.

[0049] It should be noted that after obtaining the second composite board by parameter adjustment, the noise penetration index of the target low-frequency band noise inside the second composite board can be further obtained, and the obtained noise penetration index is compared with the penetration threshold to determine whether the second composite board meets the noise reduction requirements. If it is determined that the noise penetration index does not exceed the preset penetration threshold, it means that the noise reduction requirements are not met at this time, then the third physical parameter of the second composite board needs to be optimized, and the composite board with the optimized third physical parameter is obtained accordingly and used as the third composite board.

[0050] It should be noted that, as can be seen from the following embodiments, the finally obtained noise penetration index includes the noise penetration indexes at different incident angles. As long as the noise penetration index at any one incident angle does not exceed the preset penetration threshold, the third physical parameter of the second composite board needs to be optimized.

[0051] In one optional embodiment, the obtaining the noise penetration index of the target low-frequency band noise in the second composite board specifically includes: According to the structure of the second composite board, quarter-wavelength criterion is adopted to divide the acoustic grid units, and the acoustic parameters of the cenosphere board layer and the aluminum silicate fiber cotton layer are added. The acoustic parameters include density distribution, sound velocity distribution and acoustic impedance distribution; Set the incident angle range and frequency range of the sound wave. For each combination of incident angle and frequency, the sound wave tracking algorithm is used to simulate the propagation path of the target low-frequency noise in the acoustic grid units; Calculate the attenuation amount of the target low-frequency noise during propagation according to the material sound absorption coefficient, and the material sound absorption coefficient is determined by the density distribution and sound velocity distribution of the cenosphere board layer and the aluminum silicate fiber cotton layer; According to the acoustic impedance distribution of the cenosphere board layer and the aluminum silicate fiber cotton layer, the transfer matrix method is used to calculate the reflection coefficient and transmission coefficient of the target low-frequency noise at the interface; According to the attenuation amount, the reflection coefficient and the transmission coefficient, obtain the time-domain sound pressure signal on the propagation path; Perform fast Fourier transform on the time-domain sound pressure signal to obtain the frequency-domain sound pressure spectrum; According to the formula N P =|Pout| / |Pin|, calculate and obtain the noise penetration index of the target low-frequency noise in the second composite board. N P represents the noise penetration index, Pin and Pout respectively represent the input sound pressure on one surface of the second composite board and the output sound pressure on the other surface, and Pin and Pout are obtained according to the frequency-domain sound pressure spectrum.

[0052] Specifically, in combination with the above embodiments, when obtaining the noise penetration index of the target low-frequency noise in the second composite board, the acoustic grid units can be divided according to the quarter-wavelength criterion based on the structure of the second composite board. The grid size is set to one-fourth of the acoustic wavelength corresponding to the highest frequency, and the acoustic parameters of the perlite board layer and the aluminum silicate fiber cotton layer are input. The acoustic parameters include density distribution, sound velocity distribution, and acoustic impedance distribution. Then, the incident angle range of the sound wave (for example, from 0° to 60°) and the frequency range (for example, from 50 Hz to 200 Hz) are set. For each combination of the incident angle and frequency, the acoustic wave tracking algorithm is used to simulate the propagation path of the target low-frequency noise in the acoustic grid units of the second composite board, and the amplitude change of the sound pressure signal at each grid node is recorded. Then, the attenuation amount of the target low-frequency noise during propagation is calculated according to the material sound absorption coefficient. Among them, the material sound absorption coefficient (denoted as α) is the total sound absorption coefficient of the perlite board layer and the aluminum silicate fiber cotton layer, which is determined by the density distribution and sound velocity distribution of the perlite board layer and the density distribution and sound velocity distribution of the aluminum silicate fiber cotton layer. The calculation formula for the attenuation amount is: Attenuation amount = 10×log(1 - α). For example, considering the multi-layer effect, the material sound absorption coefficient is equal to the sum of the sound absorption coefficients of the perlite board layer and the aluminum silicate fiber cotton layer minus the product of the two. Assuming that at a frequency of 100 Hz, the sound absorption coefficient of the perlite board layer is 0.3 and the sound absorption coefficient of the aluminum silicate fiber cotton layer is 0.5, then the material sound absorption coefficient is: α = (0.3 + 0.5) - 0.3×0.5 = 0.65. Then, according to the acoustic impedance distribution of the perlite board layer and the acoustic impedance distribution of the aluminum silicate fiber cotton layer, the transfer matrix method is used to calculate the reflection coefficient and transmission coefficient of the target low-frequency noise at the interface. Assuming that the acoustic impedance of the perlite board layer is Z1 and the acoustic impedance of the aluminum silicate fiber cotton layer is Z2, then the calculation formula for the reflection coefficient is: , and the calculation formula for the transmission coefficient is: ; Then, update the sound pressure signal amplitude at each grid node according to the calculated attenuation amount (taking into account the attenuation effect to correct the energy loss of the sound pressure signal), and process the reflection and transmission of sound waves at the interface according to the calculated reflection coefficient and transmission coefficient to correct the sound pressure signal amplitude when the sound wave passes through the interface, and finally obtain the time-domain sound pressure signal on the entire propagation path (i.e., the corrected sound pressure signal); After that, the fast Fourier transform can be performed on the obtained time-domain sound pressure signal at each frequency point to obtain the corresponding frequency-domain sound pressure spectrum; Finally, based on the obtained frequency-domain sound pressure spectrum, the Pin and Pout corresponding to the target low-frequency band noise passing through the second composite board can be obtained. Pin represents the input sound pressure on one surface of the second composite board (usually referring to the front surface of the second composite board, where the target low-frequency band noise enters), and Pout represents the output sound pressure on the other surface of the second composite board (usually referring to the back surface of the second composite board, where the target low-frequency band noise exits). Based on the determined Pin and Pout, the noise penetration index N of the target low-frequency band noise in the second composite board can be calculated according to the formula N P = |Pout| / |Pin| P .

[0053] It should be noted that the acoustic performance evaluation of the composite layer structure requires accurate simulation of the sound wave propagation process. In the division of acoustic grid units, the division can be carried out according to the quarter-wavelength criterion, and the grid size can be set to one-fourth of the sound wave wavelength corresponding to the highest frequency. For example, if the highest frequency of the target low-frequency band noise is 200 Hz and its corresponding sound wave wavelength is 1.7 meters, the grid size can be set to 0.425 meters; for the perlite board layer, the material density is 400 kg / m 3 , the longitudinal wave speed is 2500 m / s, and the calculated acoustic impedance is 1 MPa·s / m; for the aluminum silicate fiber cotton layer, the material density is 120 kg / m 3 , the longitudinal wave speed is 1800 m / s, and the acoustic impedance is 0.216 MPa·s / m; In terms of setting the incident conditions of sound waves, the incident angle starts from 0° and takes a numerical point every 15° until 60°, forming 5 incident angles, and takes a frequency point every 25 Hz within the frequency range to obtain 7 characteristic frequencies. This parameter setting scheme ensures a comprehensive characterization of the sound wave propagation characteristics. When the target low-frequency band noise propagates in the composite layer structure, the change in sound pressure on the propagation path can be recorded by the sound wave tracking algorithm. Taking the case of a 90 dB, 100 Hz sound wave incident vertically as an example, during the propagation of the sound wave, when passing through the perlite board layer, the sound pressure level decreases by 15 dB, among which the material absorption loss accounts for 12 dB and the interface reflection loss accounts for 3 dB; when passing through the aluminum silicate fiber cotton layer, the sound pressure level further decreases by 22 dB, among which the material absorption loss accounts for 18 dB and the interface loss accounts for 4 dB.

[0054] Further, in the analysis of the acoustic characteristics of the interface, the acoustic impedance ratio of the two-layer material is 4.63. Through the transfer matrix method, the reflection coefficient at the interface is calculated to be 0.35, and the transmission coefficient is 0.65. This indicates that at this frequency, 35% of the acoustic energy is reflected at the interface, and 65% of the acoustic energy continues to propagate. The fast Fourier transform is used to perform frequency-domain analysis on the sound pressure signal. At different incident angles, the acoustic energy conversion characteristics show obvious differences. When the incident angle increases from 0° to 60°, the interface reflection coefficient gradually increases. At an incident angle of 45°, the reflection coefficient of the 100 Hz sound wave increases to 0.48, and the transmission coefficient decreases to 0.52. In the frequency range of 50 to 200 Hz, the average penetration coefficient at normal incidence is 0.42, while the average penetration coefficient at an incident angle of 60° decreases to 0.25. This shows that the composite layer structure has a better sound insulation effect on obliquely incident sound waves.

[0055] In one optional embodiment, the third physical parameter includes the porosity, pore diameter of the perlite board layer, and the density of the aluminum silicate fiber cotton layer; Then, optimizing the third physical parameter of the second composite board to obtain the third composite board specifically includes: According to the adjustment range and adjustment amplitude of the porosity of the perlite board layer, the adjustment range and adjustment amplitude of the pore diameter, and the adjustment range and adjustment amplitude of the density of the aluminum silicate fiber cotton layer, a set of third physical parameters is obtained; Using sensitivity analysis or partial derivative method, calculate the influence weights of the porosity, pore diameter of the perlite board layer, and the density of the aluminum silicate fiber cotton layer on the acoustic impedance; According to the influence weights, in accordance with the principle of acoustic impedance matching, classify and screen each third physical parameter in the set of third physical parameters, and establish the response relationship between the acoustic impedance and each third physical parameter; Using an iterative optimization algorithm to obtain the optimal parameter combination of each third physical parameter in the set of third physical parameters; According to the optimal parameter combination, optimize the porosity, pore diameter of the perlite board layer, and the density of the aluminum silicate fiber cotton layer to obtain the third composite board.

[0056] Specifically, in combination with the above embodiments, the third physical parameter of the second composite board includes the porosity, pore diameter of the perlite board layer, and the density of the aluminum silicate fiber cotton layer. Correspondingly, when actually optimizing the third physical parameter of the second composite board, it can be based on the preset adjustment range of the porosity P of the perlite board layer (i.e., the floating range that fluctuates up and down based on the porosity reference value P, such as [0.3, 0.6]) and the adjustment amplitude (i.e., the adjustment step), the adjustment range of the pore diameter Φ of the perlite board layer (i.e., based on the pore diameter reference value Φ b up and down floating range, such as [0.3, 0.6]) and the adjustment amplitude (i.e., the adjustment step), the adjustment range of the pore diameter Φ of the perlite board layer (i.e., at the pore diameter reference value Φ bThe floating range of up-and-down floating, such as [10, 100], with the unit of μm), the adjustment range (i.e., the adjustment step), and the density ρ of the aluminosilicate fiber cotton layer f The adjustment range (i.e., within the density reference value ρ fb The floating range of up-and-down floating, such as [150, 250], with the unit of kg / m 3 ) and the adjustment range (i.e., the adjustment step), and the third set of physical parameters is obtained accordingly; in order to quantify the influence of each third physical parameter (porosity P, pore diameter Φ, density ρ f ) on the acoustic impedance Z, sensitivity analysis or partial derivative method can be used to calculate the influence weights of the porosity P, pore diameter Φ of the cenosphere board layer and the density ρ of the aluminosilicate fiber cotton layer f on the acoustic impedance Z. Among them, assuming that the acoustic impedance Z is a function of each third physical parameter, and the function is Z = Z(P, Φ, ρ f ), which can be obtained by theoretical models (such as Biot model or acoustic model of porous materials, etc.) or experimental data fitting. Then, the partial derivative is calculated for each third physical parameter, and there is: , so as to obtain the influence weights of each third physical parameter on the acoustic impedance; then, according to the obtained influence weights corresponding to each third physical parameter, in accordance with the acoustic impedance matching principle, each third physical parameter in the third set of physical parameters is classified and screened (combining the adjustment range and adjustment amplitude corresponding to each third physical parameter) to establish the response relationship between the acoustic impedance and each third physical parameter; then, based on the established response relationship between the acoustic impedance and each third physical parameter, an iterative optimization algorithm is used to obtain the optimal parameter combination corresponding to each third physical parameter in the third set of physical parameters; finally, the porosity, pore diameter of the cenosphere board layer and the density of the aluminosilicate fiber cotton layer are adjusted according to the obtained optimal parameter combination, and the third composite board is obtained accordingly.

[0057] It should be noted that the acoustic impedance matching principle requires that the acoustic impedance of the composite layer structure be as close as possible to that of the surrounding medium (such as the acoustic impedance of air is Zair≈415Pa·s / m) to minimize the reflection coefficient , that is, the classification and screening is to adjust each third physical parameter (P, Φ, ρ f ) of the composite layer structure so that the acoustic impedance of the composite layer structure after parameter adjustment is close to the target value Z target (such as Z target = Zair). Specifically, according to the weight influence analysis, the third physical parameter with the greatest weight influence (such as porosity P) can be adjusted first, followed by the third physical parameter with medium weight influence (such as pore diameter Φ), and finally the third physical parameter with the smallest weight influence (such as density ρ f ). Then, first, fix Φ = Φ [[ID=]], b , ρ f = ρfb , within the range of \(P\in[0.3, 0.6]\), adjust \(P\) multiple times according to the adjustment step of \(P\), and calculate \(Z(P)\) after each adjustment. Screen the subset of \(P\) that makes \(|Z - Z target |\) the smallest. Then, for the subset of \(P\) screened out, within the range of \(\varPhi\in[10, 100]\), adjust \(\varPhi\) multiple times according to the adjustment step of \(\varPhi\), and calculate \(Z(P,\varPhi)\) after each adjustment. Further screen out the subset of \((P,\varPhi)\) that meets the conditions. Finally, for the subset of \((P,\varPhi)\) screened out, within the range of \(\rho f \in[150, 250]\), according to \(\rho f \)'s adjustment step, adjust \(\rho f \) multiple times, and calculate \(Z(P,\varPhi,\rho f )\) after each adjustment. Further screen out the \((P,\varPhi,\rho f )\) that meets the conditions as the candidate parameter set; for the obtained candidate parameter set, through numerical simulation or experimental data, fit the response relationship between the acoustic impedance and each third physical parameter: \(Z = Z(P,\varPhi,\rho f )\). Correspondingly, the fitted result is obtained as , where \(a_1\), \(a_2\), and \(a_3\) are the fitting coefficients of each third physical parameter respectively, and the values of the fitting coefficients can be determined through experiments or simulations (such as COMSOL Multiphysics).

[0058] It should be noted that when using the iterative optimization algorithm to obtain the optimal parameter combination of each third physical parameter in the third physical parameter set, the optimization goal is to minimize the deviation between the acoustic impedance \(Z\) and the target value \(Z target \): \(\min J=|Z(P,\varPhi,\rho f ) - Z target |\). For the third physical parameter set, the gradient descent method or genetic algorithm can be used to start iterative updates from the reference values \((P b ,\varPhi b ,\rho fb )\) of each third physical parameter, and in each iteration, \((P,\varPhi,\rho f )\) can be updated respectively according to the gradient information. For example, assuming \(i\) is the current iteration, then the values of each third physical parameter in the next iteration (i.e., the \((i + 1)\)-th iteration) can be updated as: , where \(\eta\) is the learning rate, preferably, \(\eta = 0.01\); set the change rate of the acoustic impedance between adjacent iteration cycles as the convergence criterion, and set a threshold as the convergence condition (such as \(\Delta Z rate <0.01\)). After each iteration, judge whether the convergence condition is met. If it is met, stop the iteration and obtain the optimal parameter combination accordingly.

[0059] Exemplarily, taking a fireproof and sound-insulating barrier of a substation as an example, in the initial state, the porosity of the cenosphere board layer is 45%, the pore diameter ranges from 0.2 to 0.5 mm, and the density of the aluminosilicate fiber cotton layer is 120 kg / m 3 , the measured noise penetration index is 0.35, which is lower than the preset penetration threshold of 0.4; in the parameter influence weight analysis, it is found by the single-factor variation method that for every 5% increase in the porosity of the cenosphere board layer, the acoustic impedance decreases by about 8%, and for every 0.1 mm increase in the pore diameter of the cenosphere board layer, the acoustic impedance decreases by about 5%. For every 20 kg / m increase in the density of the aluminosilicate fiber cotton layer 3 , the acoustic impedance increases by about 12%; based on these response relationships, the adjustment step of the porosity is set to 2%, the adjustment step of the pore diameter is set to 0.05 mm, and the adjustment step of the density of the aluminosilicate fiber cotton layer is set to 10 kg / m 3 , the acoustic impedance optimization process uses an iterative calculation method. In each round of iteration, the change rate of the acoustic impedance is calculated. When the change rate of the acoustic impedance in two consecutive rounds of iteration is less than 1% (i.e., satisfying ΔZ rate <0.01), it is determined to converge. The optimization calculation shows that when the porosity of the cenosphere board layer is increased to 52%, the pore diameter is adjusted to the range of 0.15 to 0.4 mm, and at the same time, the density of the aluminosilicate fiber cotton layer is increased to 145 kg / m 3 , the acoustic impedance matching of the composite layer structure reaches the optimum.

[0060] Step S16, obtain the acoustic impedance distribution and thermal conductivity distribution of the third composite board, and determine the final physical parameters of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution, so as to adjust the structure of the third composite board.

[0061] In one optional embodiment, the obtaining the acoustic impedance distribution and thermal conductivity distribution of the third composite board, and determining the final physical parameters of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution, so as to adjust the structure of the third composite board specifically includes: Establish an acoustic grid unit of the third composite board, calculate the acoustic impedance of each grid unit, and obtain the acoustic impedance distribution of the acoustic grid unit; According to the acoustic impedance distribution, obtain the thermal conductivity distribution of the acoustic grid unit through heat conduction simulation or experimental measurement; According to the acoustic impedance distribution and the thermal conductivity distribution, establish an acoustic-thermal performance coupling matrix; Set a normalized noise reduction index and a fire protection index, and establish an acoustic-thermal performance function according to the noise reduction index and the fire protection index. The acoustic-thermal performance function is f(N, F)=w N ×(1 - N)+w F ×F, N represents the noise reduction index, wN represents the weight of the noise reduction index, F represents the fire prevention index, w F represents the weight of the fire prevention index; Within a preset weight value range, an iterative optimization algorithm is used to determine the optimal weight combination of the noise reduction index and the fire prevention index in the acoustic-thermal performance function to determine the acoustic-thermal performance balance point. The acoustic-thermal performance coupling matrix is used to evaluate the comprehensive impact of the adjustment of the porosity, pore size of the perlite board layer, and the density of the aluminum silicate fiber cotton layer on the acoustic-thermal performance during the iterative optimization process; Obtain the porosity, pore size of the perlite board layer, and the density of the aluminum silicate fiber cotton layer corresponding to the acoustic-thermal performance balance point, so as to adjust the structure of the third composite board according to the porosity, pore size of the perlite board layer, and the density of the aluminum silicate fiber cotton layer corresponding to the acoustic-thermal performance balance point.

[0062] Specifically, in combination with the above embodiments, when obtaining the acoustic impedance distribution and thermal conductivity distribution of the third composite board, an acoustic grid unit of the third composite board can be established first. For example, an acoustic grid unit is established using the finite element method, and the grid size is set to 1mm×1mm×1mm. For each grid node, a corresponding third physical parameter (P, Φ, ρ f ) is assigned, and the acoustic impedance of each grid node is calculated. The acoustic impedance at the i-th grid node is expressed as: Z i =Z(P i , Φ i , ρ f_i ), so as to obtain the acoustic impedance distribution of the entire acoustic grid unit. Then, based on the obtained acoustic impedance distribution, the thermal conductivity distribution of the entire acoustic grid unit is obtained through heat conduction simulation or experimental measurement.

[0063] Furthermore, in combination with the above embodiments, when determining the final physical parameters of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and thermal conductivity distribution, an acoustic-thermal performance coupling matrix can be established first according to the obtained acoustic impedance distribution and thermal conductivity distribution. Among them, the acoustic impedance Z and the thermal conductivity λ are used as the row parameter and column parameter of the acoustic-thermal performance coupling matrix respectively, and the coupling coefficient =ΔZ / Δλ is defined to characterize the degree of mutual influence between the acoustic-thermal performances; then, a normalized noise reduction index N∈[0, 1] and a fire prevention index F∈[0, 1] are set, and an acoustic-thermal performance function is established according to the noise reduction index N and the fire prevention index F, and the acoustic-thermal performance function is f(N, F)=w N ×(1 - N)+w F ×F, w N represents the weight of the noise reduction index N, w F represents the weight of the fire prevention index F; then, when satisfying w N +w FOn the premise that = 1, preset w N The value range of and w F The value range of, and within the preset weight value range, use the iterative optimization algorithm to determine the optimal weight combination of the noise reduction index N and the fire protection index F in the acoustic-thermal performance function f(N, F), and use the acoustic-thermal performance coupling matrix to evaluate each third physical parameter (P, Φ, ρ f ) during the iterative optimization process. The comprehensive impact of the adjustment on the acoustic-thermal performance, that is, by adjusting w N and w F multiple times, and calculate f(N, F) corresponding to each set of adjusted w N and w F , select a set of w N and w F that makes the value of f(N, F) reach the maximum as the optimal weight combination, and denote it as (w N_m , w F_m ). It can be understood that when the value of f(N, F) reaches the maximum, the acoustic-thermal performance reaches the best balance point; finally, obtain each third physical parameter at the acoustic-thermal performance balance point, and denote it as (P m , Φ m , ρ f_m ), and adjust the porosity, pore diameter of the cenosphere board layer and the density of the aluminum silicate fiber cotton layer of the third composite board respectively according to (P m , Φ m , ρ f_m ) to adjust the structure of the third composite board.

[0064] It should be noted that the collaborative optimization of acoustic-thermal performance involves complex parameter balance problems. Taking a substation noise reduction and fire protection system as an example, analyze by establishing an acoustic-thermal performance coupling matrix. The noise reduction index includes the noise penetration index, sound pressure attenuation, and frequency response characteristics, and its value range is normalized to the interval from 0 to 1. The fire protection index includes heat flux density, temperature gradient, and flame retardant time, which are also normalized to the interval from 0 to 1; in the weight coefficient distribution, the optimal weight combination can be determined by the hierarchical progressive method: in the initial state, the noise reduction weight w N and the fire protection weight w F are both set to 0.5. Through iterative calculation, it is found that when the noise reduction weight w N is 0.6 and the fire protection weight w F is 0.4, the acoustic-thermal performance reaches the best balance point. At this balance point, the noise penetration index is 0.42, and the heat flux density is 8.5 kW / m 2 , and both indicators meet the design requirements.

[0065] In other alternative embodiments, the method further includes: After adjusting the structure of the third composite board, monitor the noise reduction and fire prevention operation status of the substation; Among them, the monitoring of the noise reduction and fire prevention operation status of the substation specifically includes: Collect the noise data and temperature data of the substation according to a preset sampling period; Adopt a multi-source data fusion method to obtain a sound-thermal performance evaluation index containing several acoustic features and several thermal features according to the noise data and the temperature data; Input the sound-thermal performance evaluation index into a trained deep neural network for prediction to obtain a sound-thermal performance early warning probability; Judge whether the sound-thermal performance early warning probability reaches the early warning standard according to a preset early warning level; If so, trigger an early warning signal of the corresponding level and output a real-time monitoring status report.

[0066] Specifically, in combination with the above embodiments, after adjusting the structure of the third composite board according to (P m , Φ m , ρ f_m ), the noise reduction and fire prevention operation status of the substation can be monitored in real time based on the adjusted third composite board. In specific implementation, the noise data and temperature data of the substation during operation can be collected first according to a preset sampling period. Among them, the data sampling period and storage format can be set according to the device characteristics to ensure the time alignment of noise data and temperature data from different sources; then, according to the currently collected noise data and temperature data, a multi-source data fusion method is used to extract several acoustic features and several thermal features to obtain the current sound-thermal performance evaluation index of the substation; then, the obtained current sound-thermal performance evaluation index of the substation is input into a trained deep neural network for prediction, and the current sound-thermal performance early warning probability of the substation is obtained accordingly; finally, according to the pre-set early warning level division standard, it is judged whether the predicted current sound-thermal performance early warning probability of the substation reaches the early warning standard. If it is determined that the early warning standard is reached, an early warning signal of the corresponding early warning level is triggered, and a real-time monitoring status report is output.

[0067] Exemplarily, the monitoring sensor network layout can adopt a grid layout, and 8 noise sensors, 12 temperature sensors, and 4 load sensors are arranged at key positions of the substation; the data sampling period can be set to 1 min, and each sensor generates 1440 data points per day.

[0068] It should be noted that the noise data includes at least the noise decibel value, the temperature data includes at least the surface temperature, and the acoustic features extracted by the multi-source data fusion method include at least: equivalent continuous sound level ( , where \(T\) is the total measurement time and \(p(t)\) is the instantaneous sound pressure at time \(t\)), the peak sound pressure, and the spectral centroid ( , where \(M\) is the total number of frequency points and \(f\) i is the frequency of the \(i\)-th frequency point, and \(A(f\) i ) is the sound pressure amplitude of the \(i\)-th frequency point), the thermal characteristics at least include: the highest surface temperature ( , where \(T_1, T_2, \cdots\) are the real-time readings of each temperature sensor), the temperature gradient ( , where \(T(x + \Delta x)\) is the temperature value at position \(x+\Delta x\), \(T(x-\Delta x)\) is the temperature value at position \(x-\Delta x\), \(\Delta x\) is a small distance step along the \(x\)-direction for discretizing space, and \(2\Delta x\) is the interval between position \(x+\Delta x\) and position \(x-\Delta x\)), and the heat flux density; further, embodiments of the present invention can pre-establish an acoustic-thermal performance evaluation index system for a substation based on a number of acoustic characteristics and a number of thermal characteristics. During the actual monitoring process, the corresponding operating characteristic values can be extracted from the noise data and temperature data collected in real time, and the real-time acoustic-thermal performance evaluation index can be obtained based on the acoustic-thermal performance evaluation index system.

[0069] It should be noted that for the trained deep neural network, the input dimension is the number of features included in the acoustic-thermal performance evaluation index. The hidden layer uses multiple fully connected layers, and each fully connected layer is followed by an activation function (such as ReLU, etc.). Taking the case where the hidden layer uses two fully connected layers as an example: the output vector of the first hidden layer is: \(H_1 = ReLU(W_1X + b_1)\), and the output vector of the second hidden layer is: \(H_2 = ReLU(W_2H_1 + b_2)\), where \(X\) is the feature vector corresponding to the acoustic-thermal performance evaluation index input into the deep neural network, \(W_1\) and \(W_2\) are the weight matrices of the first hidden layer and the second hidden layer respectively, \(b_1\) and \(b_2\) are the bias terms of the first hidden layer and the second hidden layer respectively. The output layer of the deep neural network finally outputs a scalar representing the acoustic-thermal performance warning probability, and the acoustic-thermal performance warning probability \(=\sigma(W\) o \(H_2 + b\) o ), where \(\sigma\) is the Sigmoid activation function to ensure that the output is between 0 and 1, \(W\) o is the weight matrix of the output layer, \(H_2\) is the output vector of the previous hidden layer (i.e., the second hidden layer), and \(b\) o is the bias term of the output layer; further, embodiments of the present invention can pre-collect a large amount of historical data to train the deep neural network model and mark the normal state and the abnormal state. For example, the label \(Y\) (\(Y = 1\) represents abnormal, \(Y = 0\) represents normal) can be used for marking, and the cross-entropy loss function (Binary Cross-Entropy Loss) can be used as the loss function during the model training process. The expression of the cross-entropy loss function is: , where N is the number of training samples, i is the sample index, yi is the true label of the i-th training sample, is the predicted probability of the i-th training sample.

[0070] It should be noted that the acoustic-thermal performance warning level in the embodiments of the present invention can adopt a three-level warning mechanism. The first-level warning is a yellow warning and satisfies that the acoustic-thermal performance warning probability is between 0.3 and 0.5. The second-level warning is an orange warning and satisfies that the acoustic-thermal performance warning probability is between 0.5 and 0.7. The third-level warning is a red warning and satisfies that the acoustic-thermal performance warning probability exceeds 0.7. Correspondingly, when judging whether the currently predicted acoustic-thermal performance warning probability of the substation reaches the warning standard, if it is determined that the currently predicted acoustic-thermal performance warning probability of the substation is between 0.3 and 0.5, it is determined that the warning standard is reached, and the warning signal corresponding to the first-level warning (i.e., the yellow warning) is triggered; if it is determined that the currently predicted acoustic-thermal performance warning probability of the substation is between 0.5 and 0.7, it is determined that the warning standard is reached, and the warning signal corresponding to the second-level warning (i.e., the orange warning) is triggered; if it is determined that the currently predicted acoustic-thermal performance warning probability of the substation exceeds 0.7, it is determined that the warning standard is reached, and the warning signal corresponding to the third-level warning (i.e., the red warning) is triggered.

[0071] It should be noted that the real-time monitoring status report of the substation can be set to be updated once an hour, and it can include information such as the trend chart of acoustic-thermal performance evaluation indicators, abnormal event statistics, and warning level distribution, so as to realize the real-time monitoring and warning of the operation status of the composite layer structure.

[0072] A collaborative control method for noise reduction and fire prevention of a substation provided by an embodiment of the present invention obtains the original data of the substation operation and extracts the low-frequency noise characteristics, constructs a network model of a composite layer structure of a cenosphere board layer and an aluminous silicate fiber cotton layer, optimizes the composite layer structure according to the acoustic impedance matching principle, adjusts the porosity of the cenosphere board and the density of the aluminous silicate fiber cotton, and realizes the effective attenuation of low-frequency noise. At the same time, the embodiment of the present invention also considers the fire prevention performance, analyzes the heat conduction path and adjusts the composite layer structure to ensure that the heat flux density meets the fire prevention requirements, and through acoustic simulation and parameter adjustment, seeks a balance point between noise reduction and fire prevention performance, and obtains the optimal physical parameters of the material. This solution realizes the collaborative optimization of low-frequency noise control and fire prevention performance control of the substation, enables the noise reduction effect and the fire prevention effect to reach the ideal state at the same time, thereby meeting the dual requirements of the substation for low-frequency noise control and high-temperature protection, and improving the safety and environmental friendliness of the substation operation.

[0073] The embodiment of the present invention also provides a collaborative control system for noise reduction and fire prevention of a substation, which is used to implement the collaborative control method for noise reduction and fire prevention of the substation described in any of the above embodiments. SeeFigure 2 As shown, it is a structural block diagram of a collaborative control system for noise reduction and fire prevention of a substation provided by an embodiment of the present invention. The system includes: A noise frequency and amplitude distribution acquisition module 11, configured to acquire the original data of the substation and obtain the frequency and amplitude distribution of the target low-frequency band noise according to the original data. The original data includes equipment operation load data and sound pressure level time series data; An acoustic impedance mutation degree determination module 12, configured to determine the acoustic impedance mutation degree between adjacent two layers of media of the original composite board of the substation according to the frequency and amplitude distribution. The original composite board is composed of a cenosphere board layer and a silica aluminum fiber cotton layer; A first physical parameter optimization module 13, configured to optimize the first physical parameter of the original composite board to obtain a first composite board when the acoustic impedance mutation degree exceeds a preset mutation threshold; A second physical parameter optimization module 14, configured to acquire the heat flux density of the first composite board and optimize the second physical parameter of the first composite board to obtain a second composite board when the heat flux density exceeds a preset fire prevention threshold; A third physical parameter optimization module 15, configured to acquire the noise penetration index of the target low-frequency band noise in the second composite board and optimize the third physical parameter of the second composite board to obtain a third composite board when the noise penetration index does not exceed a preset penetration threshold; An acoustic-thermal performance balance control module 16, configured to acquire the acoustic impedance distribution and thermal conductivity distribution of the third composite board and determine the final physical parameter of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution, so as to adjust the structure of the third composite board.

[0074] Preferably, the noise frequency and amplitude distribution acquisition module 11 specifically includes: A data denoising and standardization processing unit, configured to perform denoising and standardization processing on the original data to obtain standardized equipment operation load data and standardized sound pressure level time series data; A data filtering processing unit, configured to segment the standardized sound pressure level time series data according to the standardized equipment operation load data and perform band-pass filtering on the segmented sound pressure level time series data according to a preset target frequency band to obtain a target low-frequency band signal; A time-frequency distribution acquisition unit, configured to perform multi-scale wavelet decomposition and Hilbert transform on the target low-frequency band signal in sequence to obtain the time-frequency distribution of the target low-frequency band signal; A noise frequency and amplitude distribution acquisition unit, configured to extract frequency components with amplitudes exceeding a preset noise reference value from the time-frequency distribution, and perform spectral analysis using fast Fourier transform to obtain the frequency and amplitude distribution of the target low-frequency band noise.

[0075] Preferably, the acoustic impedance mutation degree determination module 12 specifically includes: A three-dimensional structure construction unit, configured to construct a three-dimensional structure of the original composite board according to the frequency and amplitude distribution and in combination with the original physical parameters of the original composite board of the substation; An acoustic impedance change rate calculation unit, configured to calculate the relative change rate of the acoustic impedance between the perlite board layer and the aluminosilicate fiber cotton layer according to the interfacial acoustic impedance between the perlite board layer and the aluminosilicate fiber cotton layer in the three-dimensional structure; An acoustic impedance mutation degree determination unit, configured to determine the acoustic impedance mutation degree between the perlite board layer and the aluminosilicate fiber cotton layer according to the relative change rate.

[0076] Preferably, the first physical parameters include the porosity of the perlite board layer and the density of the aluminosilicate fiber cotton layer; Then, the first physical parameter optimization module 13 specifically includes: A first parameter set acquisition unit, configured to obtain a first parameter set according to the preset adjustment range and adjustment amplitude of the porosity of the perlite board layer and the preset adjustment range and adjustment amplitude of the density of the aluminosilicate fiber cotton layer; A second parameter set acquisition unit, configured to screen out several groups of parameter combinations that meet the lower limit value of the compressive strength of the perlite board layer and the upper limit value of the elastic modulus of the aluminosilicate fiber cotton layer from the first parameter set to obtain a second parameter set; A third parameter set acquisition unit, configured to screen out several groups of parameter combinations that meet the total thickness range and the interlayer bonding area ratio of the original composite board from the second parameter set to obtain a third parameter set; An acoustic impedance matching degree calculation unit, configured to calculate the ratio of the corresponding adjacent interfacial acoustic impedance for each group of parameter combinations in the third parameter set as the acoustic impedance matching degree; A first physical parameter iteration unit, configured to iteratively optimize the parameter combinations in the third parameter set according to the acoustic impedance matching degree to obtain a final parameter combination. During the iterative optimization process, the adjustment amplitude of the porosity of the perlite board layer and the adjustment amplitude of the density of the aluminosilicate fiber cotton layer decrease as the number of iterations increases; A first physical parameter optimization unit, configured to optimize the porosity of the perlite board layer and the density of the aluminosilicate fiber cotton layer according to the final parameter combination to obtain a first composite board.

[0077] Preferably, the second physical parameter optimization module 14 specifically includes: A three-dimensional grid cell establishment unit, configured to establish three-dimensional grid cells of the first composite board, and add the thermal conductivity, density, and specific heat capacity of the cenosphere board layer and the aluminum silicate fiber cotton layer to the grid cell property table; A heat conduction equation establishment unit, configured to use the external heat source temperature and the ambient temperature as the first type of boundary condition, and use the convective heat transfer coefficient as the third type of boundary condition to establish a heat conduction equation in each grid cell; A temperature field distribution acquisition unit, configured to discretely solve the heat conduction equation by using the finite volume method to obtain the temperature field distribution; A temperature gradient acquisition unit, configured to obtain the temperature gradient by using the central difference method according to the temperature field distribution; A heat flux density acquisition unit, configured to identify the heat conduction path by using the heat streamline tracking algorithm according to the temperature gradient, and obtain the heat flux density on the heat conduction path.

[0078] Preferably, the second physical parameters include the porosity of the cenosphere board layer and the thickness of the aluminum silicate fiber cotton layer; Then, the second physical parameter optimization module 14 further includes: An adjustment amount acquisition unit, configured to mark the area where the heat flux density exceeds the preset fire prevention threshold on the heat conduction path, and obtain the porosity adjustment amount of the cenosphere board layer and the thickness adjustment amount of the aluminum silicate fiber cotton layer corresponding to the area; A three-dimensional grid cell update unit, configured to update the three-dimensional grid cells of the first composite board according to the porosity adjustment amount and the thickness adjustment amount, and the grid size of the updated three-dimensional grid cells is determined by the minimum feature size of the cenosphere board layer and the aluminum silicate fiber cotton layer; A heat conduction path reconstruction unit, configured to identify a new heat conduction path according to the updated three-dimensional grid cells, and obtain the new heat flux density on the new heat conduction path; A second physical parameter optimization unit, configured to determine whether the new heat flux density exceeds the preset fire prevention threshold. If so, re-optimize the second physical parameters of the first composite board until the obtained new heat flux density does not exceed the preset fire prevention threshold. If not, optimize the porosity of the cenosphere board layer and the thickness of the aluminum silicate fiber cotton layer according to the currently obtained porosity adjustment amount and thickness adjustment amount to obtain a second composite board.

[0079] Preferably, the third physical parameter optimization module 15 specifically includes: An acoustic grid cell division unit, which is used to divide acoustic grid cells according to the structure of the second composite board by using the quarter-wavelength criterion, and add the acoustic parameters of the cenosphere board layer and the aluminosilicate fiber cotton layer, where the acoustic parameters include density distribution, sound velocity distribution, and acoustic impedance distribution; A propagation path simulation unit, which is used to set the incident angle range and frequency range of sound waves, and for each combination of incident angle and frequency, simulate the propagation path of the target low-frequency noise in the acoustic grid cells by using the sound wave tracking algorithm; An attenuation amount calculation unit, which is used to calculate the attenuation amount of the target low-frequency noise during propagation according to the material sound absorption coefficient, and the material sound absorption coefficient is determined by the density distribution and sound velocity distribution of the cenosphere board layer and the aluminosilicate fiber cotton layer; A reflection and transmission coefficient calculation unit, which is used to calculate the reflection coefficient and transmission coefficient of the target low-frequency noise at the interface by using the transfer matrix method according to the acoustic impedance distribution of the cenosphere board layer and the aluminosilicate fiber cotton layer; A time-domain sound pressure signal acquisition unit, which is used to acquire the time-domain sound pressure signal on the propagation path according to the attenuation amount, the reflection coefficient, and the transmission coefficient; A frequency-domain sound pressure spectrum acquisition unit, which is used to perform a fast Fourier transform on the time-domain sound pressure signal to obtain a frequency-domain sound pressure spectrum; A noise penetration index calculation unit, which is used to calculate the noise penetration index of the target low-frequency noise in the second composite board according to the formula N P =|Pout| / |Pin|, where N P represents the noise penetration index, Pin and Pout respectively represent the input sound pressure on one surface and the output sound pressure on the other surface of the second composite board, and Pin and Pout are obtained according to the frequency-domain sound pressure spectrum.

[0080] Preferably, the third physical parameters include the porosity and pore diameter of the cenosphere board layer and the density of the aluminosilicate fiber cotton layer; Then, the third physical parameter optimization module 15 further includes: A third physical parameter set acquisition unit, which is used to obtain a third physical parameter set according to the adjustment range and adjustment amplitude of the porosity of the cenosphere board layer, the adjustment range and adjustment amplitude of the pore diameter, and the adjustment range and adjustment amplitude of the density of the aluminosilicate fiber cotton layer; A parameter influence weight calculation unit, which is used to calculate the influence weights of the porosity and pore diameter of the cenosphere board layer and the density of the aluminosilicate fiber cotton layer on the acoustic impedance by using the sensitivity analysis or partial derivative method; A parameter grading and screening unit, configured to perform grading and screening on each third physical parameter in the third physical parameter set according to the influence weight and in accordance with the acoustic impedance matching principle, and establish a response relationship between the acoustic impedance and each third physical parameter; A third physical parameter iteration unit, configured to obtain an optimal parameter combination of each third physical parameter in the third physical parameter set by using an iterative optimization algorithm; A third physical parameter optimization unit, configured to optimize the porosity, pore size of the perlite board layer and the density of the aluminosilicate fiber cotton layer according to the optimal parameter combination to obtain a third composite board.

[0081] Preferably, the acoustic-thermal performance balance control module 16 specifically includes: An acoustic impedance distribution acquisition unit, configured to establish an acoustic grid unit of the third composite board, calculate the acoustic impedance of each grid unit, and obtain the acoustic impedance distribution of the acoustic grid unit; A thermal conductivity distribution acquisition unit, configured to obtain the thermal conductivity distribution of the acoustic grid unit through heat conduction simulation or experimental measurement according to the acoustic impedance distribution; An acoustic-thermal performance coupling matrix establishment unit, configured to establish an acoustic-thermal performance coupling matrix according to the acoustic impedance distribution and the thermal conductivity distribution; An acoustic-thermal performance function establishment unit, configured to set a normalized noise reduction index and a fire prevention index, and establish an acoustic-thermal performance function according to the noise reduction index and the fire prevention index. The acoustic-thermal performance function is f(N, F)=w N ×(1 - N)+w F ×F, where N represents the noise reduction index, w N represents the weight of the noise reduction index, F represents the fire prevention index, and w F [[ID=2,3]]represents the weight of the fire prevention index; An acoustic-thermal performance balance point determination unit, configured to determine an optimal weight combination of the noise reduction index and the fire prevention index in the acoustic-thermal performance function by using an iterative optimization algorithm within a preset weight value range to determine an acoustic-thermal performance balance point. The acoustic-thermal performance coupling matrix is used to evaluate the comprehensive influence of the adjustment of the porosity, pore size of the perlite board layer and the density of the aluminosilicate fiber cotton layer on the acoustic-thermal performance during the iterative optimization process; A composite board parameter adjustment unit, configured to obtain the porosity, pore size of the perlite board layer and the density of the aluminosilicate fiber cotton layer corresponding to the acoustic-thermal performance balance point, and adjust the structure of the third composite board according to the porosity, pore size of the perlite board layer and the density of the aluminosilicate fiber cotton layer corresponding to the acoustic-thermal performance balance point.

[0082] It should be noted that a collaborative control system for noise reduction and fire prevention in a substation provided by an embodiment of the present invention can implement all processes of the collaborative control method for noise reduction and fire prevention in a substation described in any of the above embodiments. The functions and achieved technical effects of each module and unit in the system respectively correspond to those of the collaborative control method for noise reduction and fire prevention in a substation described in the above embodiments, and will not be elaborated here.

[0083] The above are only the preferred embodiments of the present invention. It should be pointed out that for those of ordinary skill in the art in this technical field, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A coordinated control method for noise reduction and fire prevention in a substation, characterized in that: Including: Obtain the original data of the substation, and obtain the frequency and amplitude distribution of the target low-frequency band noise according to the original data, where the original data includes equipment operation load data and sound pressure level time series data; Determine the degree of acoustic impedance mutation between adjacent two layers of media of the original composite board of the substation according to the frequency and amplitude distribution, where the original composite board is composed of a cenosphere board layer and a silica aluminum fiber cotton layer; When the degree of acoustic impedance mutation exceeds a preset mutation threshold, optimize the first physical parameter of the original composite board to obtain a first composite board; Obtain the heat flux density of the first composite board, and when the heat flux density exceeds a preset fire protection threshold, optimize the second physical parameter of the first composite board to obtain a second composite board; Obtain the noise penetration index of the target low-frequency band noise in the second composite board, and when the noise penetration index does not exceed a preset penetration threshold, optimize the third physical parameter of the second composite board to obtain a third composite board; Obtain the acoustic impedance distribution and thermal conductivity distribution of the third composite board, and determine the final physical parameter of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution, so as to adjust the structure of the third composite board.

2. The collaborative control method for noise reduction and fire prevention of a substation according to claim 1, characterized in that, The obtaining the frequency and amplitude distribution of the target low-frequency band noise according to the original data specifically includes: Perform denoising and normalization processing on the original data to obtain normalized equipment operation load data and normalized sound pressure level time series data; Segment the normalized sound pressure level time series data according to the normalized equipment operation load data, and perform band-pass filtering on the segmented sound pressure level time series data according to a preset target frequency band to obtain a target low-frequency band signal; Perform multi-scale wavelet decomposition and Hilbert transform on the target low-frequency band signal in sequence to obtain the time-frequency distribution of the target low-frequency band signal; Extract the frequency components with amplitudes exceeding a preset noise reference value from the time-frequency distribution, and perform spectrum analysis using fast Fourier transform to obtain the frequency and amplitude distribution of the target low-frequency band noise.

3. The collaborative control method for noise reduction and fire prevention of a substation according to claim 1, characterized in that, The determining the degree of acoustic impedance mutation between adjacent two layers of media of the original composite board of the substation according to the frequency and amplitude distribution specifically includes: Construct a three-dimensional structure of the original composite board according to the frequency and amplitude distribution and in combination with the original physical parameters of the original composite board of the substation; Calculate the relative change rate of the acoustic impedance between the cenosphere board layer and the silica aluminum fiber cotton layer according to the interface acoustic impedance between the cenosphere board layer and the silica aluminum fiber cotton layer in the three-dimensional structure; Determine the degree of acoustic impedance mutation between the cenosphere board layer and the silica aluminum fiber cotton layer according to the relative change rate.

4. The collaborative control method for noise reduction and fire prevention of a substation according to claim 1, characterized in that, The first physical parameter includes the porosity of the cenosphere board layer and the density of the silica aluminum fiber cotton layer; Then, the optimizing the first physical parameter of the original composite board to obtain a first composite board specifically includes: Obtain a first parameter set according to the preset adjustment range and adjustment amplitude of the porosity of the cenosphere board layer and the preset adjustment range and adjustment amplitude of the density of the silica aluminum fiber cotton layer; Screen out several groups of parameter combinations from the first parameter set that meet the lower limit of the compressive strength of the cenosphere board layer and the upper limit of the elastic modulus of the aluminosilicate fiber cotton layer to obtain a second parameter set; Screen out several groups of parameter combinations from the second parameter set that meet the total thickness range and the interlayer bonding area ratio of the original composite board to obtain a third parameter set; For each group of parameter combinations in the third parameter set, calculate the ratio of the adjacent interface acoustic impedance corresponding thereto as the acoustic impedance matching degree; Iteratively optimize the parameter combinations in the third parameter set according to the acoustic impedance matching degree to obtain the final parameter combination. During the iterative optimization process, the adjustment range of the porosity of the cenosphere board layer and the adjustment range of the density of the aluminosilicate fiber cotton layer decrease as the number of iterations increases; Optimize the porosity of the cenosphere board layer and the density of the aluminosilicate fiber cotton layer according to the final parameter combination to obtain a first composite board.

5. The collaborative control method for noise reduction and fire prevention of a substation according to claim 1, characterized in that, The obtaining of the heat flux density of the first composite board specifically includes: Establish a three-dimensional grid unit of the first composite board and add the thermal conductivity, density, and specific heat capacity of the cenosphere board layer and the aluminosilicate fiber cotton layer to the grid unit property table; Use the external heat source temperature and the ambient temperature as the first type of boundary conditions, and use the convective heat transfer coefficient as the third type of boundary conditions to establish a heat conduction equation in each grid unit; Discretize and solve the heat conduction equation by using the finite volume method to obtain the temperature field distribution; Obtain the temperature gradient by using the central difference method according to the temperature field distribution; According to the temperature gradient, use the heat streamline tracking algorithm to identify the heat conduction path and obtain the heat flux density on the heat conduction path.

6. The collaborative control method for noise reduction and fire prevention of a substation according to claim 5, characterized in that, The second physical parameters include the porosity of the cenosphere board layer and the thickness of the aluminosilicate fiber cotton layer; Then, the optimizing the second physical parameters of the first composite board to obtain a second composite board specifically includes: Mark the area where the heat flux density on the heat conduction path exceeds the preset fire protection threshold, and obtain the porosity adjustment amount of the cenosphere board layer and the thickness adjustment amount of the aluminosilicate fiber cotton layer corresponding to the area; Update the three-dimensional grid unit of the first composite board according to the porosity adjustment amount and the thickness adjustment amount. The grid size of the updated three-dimensional grid unit is determined by the minimum characteristic size of the cenosphere board layer and the aluminosilicate fiber cotton layer; Identify a new heat conduction path according to the updated three-dimensional grid unit and obtain the new heat flux density on the new heat conduction path; Judge whether the new heat flux density exceeds the preset fire protection threshold. If so, re-optimize the second physical parameters of the first composite board until the obtained new heat flux density does not exceed the preset fire protection threshold. If not, optimize the porosity of the cenosphere board layer and the thickness of the aluminosilicate fiber cotton layer according to the currently obtained porosity adjustment amount and thickness adjustment amount to obtain a second composite board.

7. The collaborative control method for noise reduction and fire prevention of a substation according to claim 1, characterized in that, The obtaining of the noise penetration index of the target low-frequency band noise in the second composite board specifically includes: According to the structure of the second composite board, quarter-wavelength criterion is adopted to divide the acoustic grid units, and the acoustic parameters of the cenosphere board layer and the aluminum silicate fiber cotton layer are added. The acoustic parameters include density distribution, sound velocity distribution and acoustic impedance distribution; Set the incident angle range and frequency range of the sound wave. For each combination of incident angle and frequency, the sound wave tracing algorithm is used to simulate the propagation path of the target low-frequency noise in the acoustic grid units; Calculate the attenuation amount of the target low-frequency noise during propagation according to the material sound absorption coefficient, which is determined by the density distribution and sound velocity distribution of the cenosphere board layer and the aluminum silicate fiber cotton layer; According to the acoustic impedance distribution of the cenosphere board layer and the aluminum silicate fiber cotton layer, the transfer matrix method is used to calculate the reflection coefficient and transmission coefficient of the target low-frequency noise at the interface; According to the attenuation amount, the reflection coefficient and the transmission coefficient, obtain the time-domain sound pressure signal on the propagation path; Perform fast Fourier transform on the time-domain sound pressure signal to obtain the frequency-domain sound pressure spectrum; According to the formula N P = |Pout| / |Pin|, the noise penetration index of the target low-frequency band noise in the second composite board is calculated, where N P represents the noise penetration index, Pin and Pout respectively represent the input sound pressure on one surface and the output sound pressure on the other surface of the second composite board, and Pin and Pout are obtained according to the frequency-domain sound pressure spectrum.

8. The collaborative control method for noise reduction and fire prevention of a substation according to claim 1, characterized in that, The third physical parameters include the porosity and pore size of the cenosphere board layer and the density of the aluminum silicate fiber cotton layer; Then, optimizing the third physical parameters of the second composite board to obtain the third composite board specifically includes: According to the adjustment range and adjustment amplitude of the porosity of the cenosphere board layer, the adjustment range and adjustment amplitude of the pore size, and the adjustment range and adjustment amplitude of the density of the aluminum silicate fiber cotton layer, obtain the third physical parameter set; Use sensitivity analysis or partial derivative method to calculate the influence weights of the porosity and pore size of the cenosphere board layer and the density of the aluminum silicate fiber cotton layer on the acoustic impedance; According to the influence weights, in accordance with the acoustic impedance matching principle, classify and screen each third physical parameter in the third physical parameter set, and establish the response relationship between the acoustic impedance and each third physical parameter; Use the iterative optimization algorithm to obtain the optimal parameter combination of each third physical parameter in the third physical parameter set; Optimize the porosity, pore size of the cenosphere board layer and the density of the aluminum silicate fiber cotton layer according to the optimal parameter combination to obtain the third composite board.

9. The collaborative control method for noise reduction and fire prevention of a substation according to claim 1, characterized in that, Obtain the acoustic impedance distribution and thermal conductivity distribution of the third composite board, and determine the final physical parameters of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution, so as to adjust the structure of the third composite board. Specifically include: Establish the acoustic grid units of the third composite board, and calculate the acoustic impedance of each grid unit to obtain the acoustic impedance distribution of the acoustic grid units; According to the acoustic impedance distribution, obtain the thermal conductivity distribution of the acoustic grid units through heat conduction simulation or experimental measurement; According to the acoustic impedance distribution and the thermal conductivity distribution, establish an acoustic-thermal performance coupling matrix; Set a normalized noise reduction index and a fire prevention index, and establish an acoustic-thermal performance function based on the noise reduction index and the fire prevention index. The acoustic-thermal performance function is f(N, F) = w N ×(1 - N) + w F ×F, where N represents the noise reduction index, w N represents the weight of the noise reduction index, F represents the fire prevention index, and w F represents the weight of the fire prevention index; Within a preset weight value range, an iterative optimization algorithm is used to determine the optimal weight combination of the noise reduction index and the fire prevention index in the acoustic-thermal performance function to determine the acoustic-thermal performance balance point. The acoustic-thermal performance coupling matrix is used to evaluate the comprehensive impact of the adjustment of the porosity, pore size of the perlite board layer, and the density of the aluminum silicate fiber cotton layer on the acoustic-thermal performance during the iterative optimization process; Obtain the porosity, pore size of the perlite board layer, and the density of the aluminum silicate fiber cotton layer corresponding to the acoustic-thermal performance balance point, and adjust the structure of the third composite board according to the porosity, pore size of the perlite board layer, and the density of the aluminum silicate fiber cotton layer corresponding to the acoustic-thermal performance balance point.

10. A collaborative control system for noise reduction and fire prevention in a substation, characterized in that, Including: A noise frequency and amplitude distribution acquisition module, configured to acquire the original data of the substation and obtain the frequency and amplitude distribution of the target low-frequency noise according to the original data. The original data includes equipment operation load data and sound pressure level time series data; An acoustic impedance mutation degree determination module, configured to determine the acoustic impedance mutation degree between adjacent two layers of media of the original composite board of the substation according to the frequency and amplitude distribution. The original composite board is composed of a perlite board layer and an aluminum silicate fiber cotton layer; A first physical parameter optimization module, configured to optimize the first physical parameter of the original composite board to obtain a first composite board when the acoustic impedance mutation degree exceeds a preset mutation threshold; A second physical parameter optimization module, configured to obtain the heat flux density of the first composite board and optimize the second physical parameter of the first composite board to obtain a second composite board when the heat flux density exceeds a preset fire prevention threshold; A third physical parameter optimization module, configured to obtain the noise penetration index of the target low-frequency noise in the second composite board and optimize the third physical parameter of the second composite board to obtain a third composite board when the noise penetration index does not exceed a preset penetration threshold; An acoustic-thermal performance balance control module, configured to obtain the acoustic impedance distribution and thermal conductivity distribution of the third composite board, and determine the final physical parameter of the third composite board corresponding to the acoustic-thermal performance balance point according to the acoustic impedance distribution and the thermal conductivity distribution, so as to adjust the structure of the third composite board.

Images