Machine learning based method and system for optimizing production of acoustic foam

Abstract

The application discloses a sound insulation foam production optimization method and system based on machine learning, and relates to the technical field of building protection.The method comprises the following steps: based on an acoustic camera, sound insulation detection is performed to determine a sound wave spectrum, and a sound insulation coefficient is determined in combination with a building sound insulation standard; an optimization decision module is built, and the optimization decision module comprises an optimization decision module and an acoustic evaluation module; based on the sound wave spectrum and the sound insulation coefficient, product optimization decision and sound insulation evaluation are performed in combination with the optimization decision module to determine a product optimization strategy, wherein the optimization dimensions include sound insulation optimization and lightweight optimization, and the sound insulation optimization comprises the sound insulation coefficient; the product optimization strategy is subjected to process conversion based on an initial process flow to determine an optimized production process; and based on the optimized production process, optimized production management of the sound insulation foam is performed.The application solves the technical problem that the sound insulation foam production process is less intelligent in the prior art, and the sound insulation effect of the sound insulation foam product is poor in different frequency bands and complex environments.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of building protection, in particular to a sound insulation foam production optimization method and system based on machine learning. BACKGROUND

[0002] Currently, sound insulation foam has been widely used in the field of construction, especially in residential, commercial buildings and industrial facilities. However, most existing sound insulation foams use single materials or simple composite material structures, and the design and optimization of sound insulation foams have limitations, making it difficult to provide ideal sound insulation effects in different frequency bands and complex environments in practical applications.

[0003] Therefore, in the prior art, the sound insulation foam production process is less intelligent, resulting in poor sound insulation effect of sound insulation foam products in different frequency bands and complex environments. SUMMARY

[0004] The present application provides a sound insulation foam production optimization method and system based on machine learning, which solves the technical problem of low intelligence of sound insulation foam production process in the prior art, resulting in poor sound insulation effect of sound insulation foam products in different frequency bands and complex environments. The sound insulation performance of sound insulation materials is detected by an acoustic camera, and combined with an optimization decision module, accurate evaluation of sound insulation performance is realized, and based on the evaluation results, the sound insulation foam optimization design scheme is obtained, improving the sound insulation effect of sound insulation foam in different frequency bands and complex environments.

[0005] The present application provides a sound insulation foam production optimization method based on machine learning, which comprises: based on an acoustic camera, sound insulation detection is performed to determine a sound wave spectrum, wherein the sound insulation detection includes building self sound insulation and foam sound insulation. Based on the sound wave spectrum, combined with the building sound insulation standard, the sound insulation performance is evaluated to determine the sound insulation coefficient, and the sound insulation coefficient is subjected to attenuation processing based on the gap tolerance interval. An optimization decision module is built, which includes an optimization decision module and an acoustic evaluation module. Based on the sound wave spectrum and the sound insulation coefficient, combined with the optimization decision module, product optimization decision and sound insulation evaluation are performed to determine product optimization strategy, wherein the optimization dimension includes sound insulation optimization and lightweight optimization, and the sound insulation optimization includes sound insulation coefficient. Based on the initial process flow, the product optimization strategy is converted to determine the optimized production process. Based on the optimized production process, the optimized production management of sound insulation foam is performed.

[0006] In an implementation, the sound insulation performance evaluation determines a sound insulation coefficient, including: traversing the sound wave spectrum, performing spatial layer division, and determining a plurality of local sound insulation layers. Traversing the plurality of local sound insulation layers, based on the proportion of incident sound waves and absorbed sound waves, a plurality of sound insulation coefficients are calculated, which correspond one-to-one to the plurality of local sound insulation layers. The plurality of sound insulation coefficients are averaged to determine the sound insulation coefficient, which identifies an optimization amount based on the sound insulation coefficient and the building sound insulation standard.

[0007] In an implementation, the sound insulation coefficient packet has a gap tolerance interval-based attenuation processing, including: based on the construction fixed mode, analyzing the sound insulation attenuation degree based on the construction gap, determining the gap tolerance interval, which is the maximum value of the multi-mode attenuation degree that meets the construction standard. Based on the sound insulation attenuation degree, the sound insulation coefficient is attenuated and compensated.

[0008] In an implementation, before making product optimization decisions and sound insulation evaluations, including: determining the product characteristics of the sound insulation foam and the functional indicators based on the sound insulation performance. Interacting with sound insulation test records, by performing variable limitation clustering, mining influence relationships, the influence relationships are at least one product characteristic and the trend relationship of the related functional indicators. Taking the product characteristics as the matrix rows and the functional indicators as the matrix columns, a relationship matrix is established in combination with the influence relationship.

[0009] In an implementation, making product optimization decisions and sound insulation evaluations, including: the product characteristics include structure characteristics and morphology characteristics, and the structure characteristics include fiber structure and composite structure. Based on the product characteristics, the sound insulation foam is optimized based on finite element analysis with the relationship matrix as a constraint, and the fitness is corrected through sound insulation evaluation to determine the product optimization strategy.

[0010] In an implementation, optimizing the sound insulation foam based on finite element analysis, including: traversing the product characteristics, randomly determining a first number of optimization targets, the first number being a positive integer less than or equal to the total amount of structure characteristics. Based on the optimization target, the relationship matrix is used as a constraint to perform random disturbance based on the initial product characteristics of the sound insulation foam to determine a first strategy set. The optimization target number iteration and optimization iteration are performed until the convergence condition is met to determine the optimization strategy set, wherein the number iteration has no correlation.

[0011] In an implementation, the product optimization strategy is converted into a process, including: traversing the initial process flow, mapping the product optimization strategy to associated process nodes to determine a plurality of mapping groups, wherein each optimization strategy point has at least one associated process node. Based on the plurality of mapping groups, the process nodes of the optimization strategy points are allocated to determine the optimized production process, wherein the allocation ratio is the node correlation degree.

[0012] The application also provides a machine learning-based sound insulation foam production optimization system, which comprises:

[0013] An acoustic wave detection module is configured to determine an acoustic wave spectrum based on an acoustic camera for sound insulation detection, wherein the sound insulation detection includes building self sound insulation and foam sound insulation.

[0014] A sound insulation coefficient acquisition module is configured to determine a sound insulation coefficient based on the acoustic wave spectrum and in combination with a building sound insulation standard for sound insulation performance evaluation, wherein the sound insulation coefficient is subjected to attenuation processing based on a gap tolerance interval.

[0015] An optimization module is configured to build an optimization decision module, wherein the optimization decision module comprises an optimization decision module and an acoustic evaluation module.

[0016] An iteration module is configured to determine a product optimization strategy based on the acoustic wave spectrum and the sound insulation coefficient in combination with the optimization decision module for product optimization decision and sound insulation evaluation, wherein the optimization dimension comprises sound insulation optimization and lightweight optimization, and the sound insulation optimization includes a sound insulation coefficient.

[0017] A production process determination module is configured to convert the product optimization strategy into a process based on an initial process flow to determine an optimized production process.

[0018] A production management module is configured to perform optimized production management of sound insulation foam based on the optimized production process.

[0019] The sound insulation foam production optimization method and system based on machine learning provided in the present application are based on an acoustic camera to perform sound insulation detection to determine a sound wave spectrum, wherein the sound insulation detection includes building self sound insulation and foam sound insulation. Based on the sound wave spectrum, sound insulation performance evaluation is performed to determine a sound insulation coefficient in combination with a building sound insulation standard, and the sound insulation coefficient is subjected to attenuation processing based on a gap tolerance interval. An optimization decision module is built, which includes an optimization decision module and an acoustic evaluation module. Based on the sound wave spectrum and the sound insulation coefficient, product optimization decision and sound insulation evaluation are performed in combination with the optimization decision module to determine a product optimization strategy, wherein the optimization dimensions include sound insulation optimization and lightweight optimization, and the sound insulation optimization includes a sound insulation coefficient. The product optimization strategy is subjected to process conversion based on an initial process flow to determine an optimized production process. Based on the optimized production process, optimized production management of sound insulation foam is performed. The technical problem of low intelligence of the sound insulation foam production process in the prior art, which leads to poor sound insulation effect of the sound insulation foam product in different frequency bands and complex environments, is solved. The sound insulation performance of sound insulation materials is detected by an acoustic camera, and the optimization decision module is combined to realize accurate evaluation of the sound insulation performance. Based on the evaluation result, an optimized design scheme of sound insulation foam is obtained, and the sound insulation effect of the sound insulation foam in different frequency bands and complex environments is improved. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. In the present application, a flowchart is used to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can be added to these processes, or one or more steps of operations can be removed from these processes.

[0021] Figure 1 A sound insulation foam production optimization method based on machine learning provided by the embodiments of the present application is shown in the flowchart.

[0022] Figure 2 A sound insulation foam production optimization system based on machine learning provided by the embodiments of the present application is shown in the structural schematic diagram.

[0023] Legend of reference signs: sound wave detection module 11, sound insulation coefficient acquisition module 12, optimization module 13, iteration module 14, production process determination module 15, production management module 16. DETAILED DESCRIPTION

[0024] The above description is only a summary of the technical solutions of the present application. In order to enable the technical means of the present application to be more clearly understood, and to be implemented according to the content of the description, and in order to enable the above and other purposes, characteristics and advantages of the present application to be more apparent and easy to understand, the following specific embodiments of the present application are described.

[0025] In order to make the purposes, technical solutions and advantages of the present application more clear, the following will further describe the present application in combination with the drawings, and the described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those skilled in the art without making creative labor belong to the scope of protection of the present application.

[0026] In the following description, "some embodiments" are related to a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The term "first\second" is only to distinguish similar objects, and does not represent the specific order of the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0027] The embodiments of the present application provide a sound insulation foam production optimization method and system based on machine learning, as shown in Figure 1 The method comprises the following steps:

[0028] Based on the acoustic camera, sound insulation detection is performed to determine the sound wave spectrum, wherein the sound insulation detection includes building self sound insulation and foam sound insulation.

[0029] Based on the sound wave spectrum, the sound insulation performance is evaluated to determine the sound insulation coefficient in combination with the building sound insulation standard, and the sound insulation coefficient is subjected to attenuation processing based on the gap tolerance interval.

[0030] An optimization decision module is built, and the optimization decision module comprises an optimization decision module and an acoustic evaluation module.

[0031] The sound insulation of the building and the foam is detected by the acoustic camera, and a sound wave spectrum is obtained. The sound wave spectrum contains sound wave attenuation information of the building sound insulation and the foam sound insulation in different frequency bands, and is used to reflect the sound insulation effect of each material. Then, based on the sound wave spectrum, the sound insulation performance is evaluated and the sound insulation coefficient is determined based on the building sound insulation standard. The sound insulation coefficient is subjected to attenuation processing based on the gap tolerance interval, that is, the coefficient of the sound insulation coefficient after attenuation compensation processing. Further, an optimization decision module is built, which includes an optimization decision module and an acoustic evaluation module. The optimization decision module is used to obtain product optimization strategies according to product characteristics and the relationship matrix as constraints. The acoustic evaluation module is constructed based on a neural network model. Sample sound insulation foam product characteristics and corresponding sample sound insulation test results and sample weight test results are obtained through historical optimization records and experimental detection records. The sample sound insulation test results include sound insulation coefficients. The acoustic training data set is constructed based on the sample sound insulation foam product characteristics and the corresponding sample sound insulation test results and sample weight test results. The neural network model is supervised trained through the acoustic training data set, until the output accuracy of the model meets the preset accuracy requirement, and the acoustic evaluation module is obtained.

[0032] The method provided by the embodiments of the present application further includes:

[0033] The sound wave spectrum is traversed to divide the space layer and determine a plurality of local sound insulation layers.

[0034] The plurality of sound insulation coefficients are calculated based on the ratio of incident sound waves and absorbed sound waves, and the plurality of sound insulation coefficients correspond one-to-one to the plurality of local sound insulation layers.

[0035] The plurality of sound insulation coefficients are subjected to mean value calculation to determine the sound insulation coefficient, and the sound insulation coefficient is marked with an optimization amount, and the optimization amount is determined based on the sound insulation coefficient and the building sound insulation standard.

[0036] The sound insulation performance evaluation determines a sound insulation coefficient, including: traversing the sound wave spectrum, performing spatial layer division, and determining a plurality of local sound insulation layers, that is, acquiring the sound wave spectrum of the sound insulation foam at different frequency bands through the acoustic camera. The sound wave spectrum is traversed, and a plurality of local sound insulation layers are divided according to the spatial position, and each local sound insulation layer represents the sound insulation performance of a specific position. The plurality of local sound insulation layers are traversed, a plurality of sound insulation coefficients are calculated based on the ratio of incident sound waves to absorbed sound waves, the incident sound waves are the input sound waves during testing, and the absorbed sound waves are the sound waves collected after passing through the local sound insulation layer, the sound insulation coefficient is the ratio of the intensity of the incident sound waves to the intensity of the absorbed sound waves, and the plurality of sound insulation coefficients correspond to the plurality of local sound insulation layers one by one. Further, the plurality of sound insulation coefficients are averaged to determine the sound insulation coefficient, and the sound insulation coefficient is marked with an optimization amount, the optimization amount is determined based on the sound insulation coefficient and the building sound insulation standard, and the optimization amount is the difference between the building sound insulation standard and the sound insulation coefficient, so as to ensure that the sound insulation effect of the sound insulation foam in actual application meets the expectation.

[0037] The method provided by the embodiment of the application further includes:

[0038] Based on the construction fixing mode, the sound insulation attenuation degree based on the construction gap is analyzed, and the gap tolerance interval is determined, the gap tolerance interval being a maximum value of multi-mode attenuation degree meeting the construction standard.

[0039] Based on the sound insulation attenuation degree, the sound insulation coefficient is subjected to attenuation compensation processing.

[0040] According to the fixing mode that may occur in the actual construction process, the influence of the construction gap on the sound insulation performance is analyzed. Specifically, the influence of different gap sizes generated by the fixing mode in the construction process on the sound insulation effect can be measured through experiments or simulations, and the gap tolerance interval is determined. The gap tolerance interval is the maximum gap size that can be accepted in the construction, and the maximum value of the sound insulation attenuation degree under this size is the maximum value of the multi-mode attenuation degree, and the sound insulation attenuation degree is the influence parameter of the sound insulation coefficient. Further, based on the sound insulation attenuation degree, the sound insulation coefficient is subjected to attenuation compensation processing, that is, the sound insulation coefficient is compensated according to the sound insulation attenuation degree.

[0041] The method provided by the embodiment of the application further includes:

[0042] The product characteristics of the sound insulation foam are determined, and the functional indicators based on the sound insulation performance are determined.

[0043] The interactive sound insulation test record is subjected to variable limit clustering, and the influence relationship is mined, the influence relationship being the trend change relationship between at least one product characteristic and a related functional indicator.

[0044] The product characteristics are taken as matrix rows, the functional indicators are taken as matrix columns, and a relationship matrix is established in combination with the influence relationship.

[0045] Before making product optimization decision and sound insulation evaluation, product characteristics of sound insulation foam are determined, the product characteristics of sound insulation foam including its structural characteristics and morphological characteristics. The structural characteristics specifically include fiber structure and composite structure, and the morphological characteristics include parameters such as density and thickness of the foam. The functional index is the sound insulation coefficient of the sound insulation foam, which is measured by acoustic testing instruments, and the sound wave attenuation effects of different foam samples at different frequency bands are recorded. The interactive sound insulation test records the trend and change relationship between product characteristics and related functional indexes, specifically using various sound insulation test methods (such as standard test room, field test, etc.), and the sound insulation effects of sound insulation foam samples with different product characteristics are recorded. Through statistical software, the relationship between product characteristics (such as density, thickness) and sound insulation performance index sound insulation coefficient is analyzed. The influence relationship is the trend relationship between at least one product characteristic and the related functional index. The above analysis results are expressed in matrix form, with the product characteristics as the matrix rows and the functional indexes as the matrix columns, and a relationship matrix is established combining the influence relationship.

[0046] Based on the sound wave spectrum and the sound insulation coefficient, product optimization decision and sound insulation evaluation are made combining the optimization decision module to determine the product optimization strategy, wherein the optimization dimensions include sound insulation optimization and lightweight optimization, and the sound insulation optimization includes the sound insulation coefficient.

[0047] Based on the product optimization strategy, the process conversion is performed to determine the optimized production process.

[0048] Based on the optimized production process, the optimized production management of the sound insulation foam is performed.

[0049] Based on the sound wave spectrum and the sound insulation coefficient, the product characteristic trend relationship is obtained, and product optimization decision and sound insulation evaluation are made combining the optimization decision module to determine the product optimization strategy, wherein the optimization dimensions include sound insulation optimization and lightweight optimization, and the sound insulation optimization includes the sound insulation coefficient. Further, based on the initial process flow as the optimization benchmark, the process conversion is performed on the product optimization strategy to convert the product optimization strategy into a specific manufacturing process flow, and the optimized production process is determined. Finally, based on the optimized production process, the optimized production management of the sound insulation foam is performed. The technical problem of low intelligence of the sound insulation foam production process in the prior art, which leads to poor sound insulation effect of the sound insulation foam product in different frequency bands and complex environments, is solved. The sound insulation performance of the sound insulation material is detected by the acoustic camera, and the optimization decision module is combined to realize accurate evaluation of the sound insulation performance. Based on the evaluation result, the sound insulation foam optimization design scheme is obtained, and the sound insulation effect of the sound insulation foam in different frequency bands and complex environments is improved.

[0050] The method provided by the embodiment of the application further includes:

[0051] The product characteristics include structural characteristics and morphological characteristics, and the structural characteristics include fiber structure and composite structure.

[0052] The optimization of the soundproof foam based on finite element analysis is performed based on the product characteristics and the relationship matrix as constraints, and fitness correction is performed through sound insulation evaluation to determine the product optimization strategy.

[0053] The product optimization decision and sound insulation evaluation include: the product characteristics include structural characteristics and morphological characteristics, and the structural characteristics include fiber structure and composite structure. The fiber structure is, for example, the arrangement of fibers inside the foam, such as straight fibers, staggered fibers, etc. The composite structure is, for example, whether the foam includes a multi-layer structure, such as single-layer foam, double-layer foam, etc. Then, the optimization of the soundproof foam based on finite element analysis is performed based on the product characteristics and the relationship matrix as constraints, and fitness correction is performed through the acoustic evaluation module. The fitness is obtained by weighted summation based on the difference between the sound insulation result output by the acoustic evaluation module and the weight of the initial process flow and the weight result output by the acoustic evaluation module. Based on the fitness calculation result, the larger the fitness calculation result, the better the corresponding product strategy, and the product optimization strategy is determined.

[0054] The method provided by the embodiments of the present application further includes:

[0055] The product characteristics are traversed, and a first number of optimization targets are randomly determined. The first number is a positive integer less than or equal to the total amount of structural characteristics.

[0056] Based on the optimization targets, random perturbation of the initial product characteristics of the soundproof foam is performed based on the relationship matrix as constraints to determine a first strategy set.

[0057] The optimization target number iteration and optimization iteration are performed until the convergence condition is met to determine an optimization strategy set, and the number iteration has no correlation.

[0058] The optimization of the soundproof foam based on finite element analysis includes: traversing the product characteristics, randomly determining a first number of optimization targets in the product characteristics, the optimization targets being characteristic categories included in the product characteristics, such as the density and thickness characteristics of the morphological characteristics of the foam, and the first number being a positive integer less than or equal to the total amount of structural characteristics. Iterative analysis is performed for different target groups of different numbers each time to ensure randomness. Then, based on the optimization targets, random perturbation of the initial product characteristics of the soundproof foam is performed based on the relationship matrix as constraints. For example, the selected density is slightly adjusted in the preferred direction in the relationship matrix, such as adjusting the density from 100 kg / m 3 to 105 kg / m 3determining a first strategy set, performing sound insulation evaluation of the first strategy set by an acoustic evaluation module, performing iterative optimization based on the evaluation result, and the optimization direction is a direction of better sound insulation coefficient of the relationship matrix and a direction of lighter quality. Subsequently, the optimization target quantity iteration and the optimization iteration are performed until the convergence condition is met, the convergence condition is that the preset iteration number is met, the optimization is completed, the optimization strategy set is determined, and the optimization strategy set is the strategy set corresponding to the best sound insulation coefficient obtained after the iterative optimization, and the quantity iteration has no correlation.

[0059] The method provided by the embodiment of the application further includes:

[0060] The initial process flow is traversed, the product optimization strategy is mapped to the associated process nodes, and a plurality of mapping groups are determined, wherein each optimization strategy point has at least one associated process node.

[0061] Based on the plurality of mapping groups, the process nodes of the optimization strategy points are apportioned, and the optimized production process is determined, wherein the apportionment ratio is the node correlation degree.

[0062] The initial process flow is traversed, the product optimization strategy is mapped to the associated process nodes, that is, the strategy in the product optimization strategy that needs to be optimized for the initial process flow is obtained, and the associated process nodes corresponding to the strategy are determined, a plurality of mapping groups are determined, each mapping group corresponds to an optimization strategy category, wherein each optimization strategy point has at least one associated process node. Moreover, each strategy category in the optimization strategy and the associated process node are in a preset corresponding relationship, and the corresponding relationship is stored in an optimization corresponding database, in which database each optimization strategy category adjustment parameter corresponds to one or more associated process node adjustment parameters. Finally, based on the plurality of mapping groups, the process nodes of the optimization strategy points are apportioned, a plurality of process nodes will have an impact on the quality of a certain feature, the impact of the process node on the feature is a preset apportionment ratio, the node process parameters corresponding to the plurality of mapping groups are summed and multiplied by the preset apportionment ratio, and the preset apportionment ratio is a preset apportionment ratio based on the node correlation degree, so as to determine the specific node process parameters, and determine the optimized production process, wherein the apportionment ratio is the node correlation degree.

[0063] In the foregoing, with reference to Figure 1 The method for optimizing production of sound insulation foam based on machine learning according to the embodiment of the application is described in detail. Next, with reference to Figure 2 The system for optimizing production of sound insulation foam based on machine learning according to the embodiment of the application is described.

[0064] The machine learning-based sound insulation foam production optimization system according to the embodiment of the present application solves the technical problem that the sound insulation foam production process in the prior art is less intelligent, resulting in poor sound insulation effect of the sound insulation foam product in different frequency bands and complex environments. The sound insulation performance of the sound insulation material is detected by the acoustic camera, and the optimized decision module is combined to realize accurate evaluation of the sound insulation performance. Based on the evaluation result, the sound insulation foam optimization design scheme is obtained, and the sound insulation effect of the sound insulation foam in different frequency bands and complex environments is improved. The machine learning-based sound insulation foam production optimization system comprises: a sound wave detection module 11, a sound insulation coefficient acquisition module 12, an optimization module 13, an iteration module 14, a production process determination module 15, and a production management module 16.

[0065] The sound wave detection module 11 is configured to determine a sound wave spectrum based on the acoustic camera and sound insulation detection. The sound insulation detection includes building self sound insulation and foam sound insulation.

[0066] The sound insulation coefficient acquisition module 12 is configured to determine a sound insulation coefficient based on the sound wave spectrum and sound insulation performance evaluation combined with the building sound insulation standard. The sound insulation coefficient is subjected to attenuation processing based on a gap tolerance interval.

[0067] The optimization module 13 is configured to build an optimization decision module. The optimization decision module includes an optimization decision module and an acoustic evaluation module.

[0068] The iteration module 14 is configured to determine a product optimization strategy based on the sound wave spectrum and the sound insulation coefficient, combined with the optimization decision module for product optimization decision and sound insulation evaluation. The optimization dimension includes sound insulation optimization and lightweight optimization. The sound insulation optimization includes a sound insulation coefficient.

[0069] The production process determination module 15 is configured to convert the product optimization strategy based on the initial process flow to determine an optimized production process.

[0070] The production management module 16 is configured to perform optimized production management of the sound insulation foam based on the optimized production process.

[0071] The specific configuration of the sound insulation coefficient acquisition module 12 will be described in detail below. The sound insulation coefficient acquisition module 12 can further comprise: the sound insulation performance evaluation and the determination of the sound insulation coefficient, including: traversing the sound wave spectrum, dividing the space layer, and determining a plurality of local sound insulation layers. Based on the ratio of incident sound waves and absorbed sound waves, a plurality of sound insulation coefficients are calculated by traversing the plurality of local sound insulation layers. The plurality of sound insulation coefficients correspond to the plurality of local sound insulation layers one by one. The plurality of sound insulation coefficients are subjected to mean value calculation to determine the sound insulation coefficient. The sound insulation coefficient is marked with an optimization amount. The optimization amount is determined based on the sound insulation coefficient and the building sound insulation standard.

[0072] Next, the specific configuration of the sound insulation coefficient acquisition module 12 will be described in detail. The sound insulation coefficient acquisition module 12 further includes: based on the construction fixation mode, analyzing the sound insulation attenuation degree based on the construction gap, determining the gap tolerance interval, and the gap tolerance interval is the maximum value of the multi-mode attenuation degree that meets the construction standard. Based on the sound insulation attenuation degree, the sound insulation coefficient is subjected to attenuation compensation processing.

[0073] Next, the specific configuration of the optimization module 13 will be described in detail. The optimization module 13 can further include: before making product optimization decisions and sound insulation evaluation, including: determining the product characteristics of the sound insulation foam, and the functional indicators based on the sound insulation performance. Interactive sound insulation test records, by performing variable limit clustering, mining influence relationships, the influence relationship is the trend relationship of at least one product characteristic and related functional indicators. The product characteristics are used as matrix rows, and the functional indicators are used as matrix columns, and a relationship matrix is established in combination with the influence relationship.

[0074] Next, the specific configuration of the iteration module 14 will be described in detail. The iteration module 14 further includes: making product optimization decisions and sound insulation evaluation, including: the product characteristics include structure characteristics and morphology characteristics, and the structure characteristics include fiber structure and composite structure. Based on the product characteristics, the sound insulation foam is subjected to optimization based on finite element analysis, and the fitness is corrected through sound insulation evaluation, to determine the product optimization strategy, with the relationship matrix as a constraint.

[0075] Next, the specific configuration of the iteration module 14 will be described in detail. The iteration module 14 further includes: performing optimization based on finite element analysis on the sound insulation foam, including: traversing the product characteristics, randomly determining a first number of optimization targets, and the first number is a positive integer less than or equal to the total amount of structure characteristics. Based on the optimization target, the relationship matrix is used as a constraint to perform random disturbance based on the initial product characteristics of the sound insulation foam, to determine a first strategy set. The optimization target number iteration and optimization iteration are performed until the convergence condition is met, to determine the optimization strategy set, wherein the number iteration has no correlation.

[0076] Next, the specific configuration of the production process determination module 15 will be described in detail. The production process determination module 15 further includes: converting the product optimization strategy into a process, including: traversing the initial process flow, mapping the product optimization strategy to associated process nodes to determine a plurality of mapping groups, wherein each optimization strategy point has at least one associated process node. Based on the plurality of mapping groups, the process nodes of the optimization strategy points are apportioned to determine the optimized production process, wherein the apportionment ratio is the node correlation degree.

[0077] The machine learning based sound insulation foam production optimization system provided by the embodiment of the present application can execute the machine learning based sound insulation foam production optimization method provided by any embodiment of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0078] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, and the various units and modules are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized. In addition, the specific names of each functional unit are only for the convenience of mutual differentiation, and are not used to limit the protection scope of the present application.

[0079] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for optimizing the production of soundproof foam based on machine learning, characterized in that, The method comprises: Based on the acoustic camera, the sound insulation detection is carried out to determine the sound wave spectrum, wherein the sound insulation detection includes building self-sound insulation and foam sound insulation; Based on the sound wave spectrum, combined with the building sound insulation standard, the sound insulation performance evaluation is carried out to determine the sound insulation coefficient, and the sound insulation coefficient exists attenuation processing based on the gap tolerance interval; An optimization decision module is built, which includes an optimization decision module and an acoustic evaluation module; Based on the sound wave spectrum and the sound insulation coefficient, combined with the optimization decision module, product optimization decision and sound insulation evaluation are carried out to determine the product optimization strategy, wherein the optimization dimension includes sound insulation optimization and lightweight optimization, and the sound insulation optimization includes sound insulation coefficient; Based on the product optimization strategy, the process conversion is carried out to determine the optimized production process; Based on the optimized production process, the optimization production management of the sound insulation foam is carried out; The sound insulation performance evaluation to determine the sound insulation coefficient comprises: Traverse the sound wave spectrum to divide the space layer and determine a plurality of local sound insulation layers; Traverse the plurality of local sound insulation layers, calculate a plurality of sound insulation coefficients based on the ratio of incident sound wave and absorbed sound wave, and the plurality of sound insulation coefficients correspond to the plurality of local sound insulation layers one by one; The mean value of the plurality of sound insulation coefficients is calculated to determine the sound insulation coefficient, and the sound insulation coefficient is marked with an optimization amount, which is determined based on the sound insulation coefficient and the building sound insulation standard; The sound insulation coefficient packet exists attenuation processing based on the gap tolerance interval, which comprises: Based on the construction fixed mode, the sound insulation attenuation degree based on the construction gap is analyzed to determine the gap tolerance interval, which is the maximum value of the multi-mode attenuation degree meeting the construction standard; Based on the sound insulation attenuation degree, the attenuation compensation processing is carried out on the sound insulation coefficient. 2.The machine learning-based soundproof foam production optimization method of claim 1, wherein, Before the product optimization decision and sound insulation evaluation, it comprises: Determine the product characteristics of the sound insulation foam, and the functional indicators based on the sound insulation performance; Interactive sound insulation test records are obtained by variable limiting clustering to mine the influence relationship, which is the trend relationship between at least one product characteristic and related functional indicators; The relationship matrix is established based on the product characteristics as the matrix row, the functional indicators as the matrix column, and the influence relationship. 3.The machine learning-based soundproof foam production optimization method of claim 2, wherein, The product optimization decision and sound insulation evaluation comprises: The product characteristics include structure characteristics and morphology characteristics, and the structure characteristics include fiber structure and composite structure; Based on the product characteristics, the relationship matrix is used as a constraint to carry out optimization based on finite element analysis on the sound insulation foam, and the fitness is corrected through sound insulation evaluation to determine the product optimization strategy. 4.The method of claim 3, wherein, The optimization based on finite element analysis on the sound insulation foam comprises: Traverse the product characteristics to randomly determine a first number of optimization targets, and the first number is a positive integer less than or equal to the total amount of structure characteristics; Based on the optimization target, the relationship matrix is used as a constraint to carry out random disturbance on the initial product characteristics of the sound insulation foam to determine a first strategy set; The optimization target number iteration and optimization iteration are carried out until the convergence condition is met to determine the optimization strategy set, and the number iteration has no correlation. 5.The machine learning based soundproof foam production optimization method of claim 1, wherein, The product optimization strategy is converted into a process, including: Traverse the initial process flow, map the associated process nodes of the product optimization strategy, determine a plurality of mapping groups, wherein each optimization strategy point has at least one associated process node; Based on the plurality of mapping groups, the process nodes of the optimization strategy points are allocated to determine the optimized production process, wherein the allocation ratio is the node correlation degree.

6. A soundproofing foam production optimization system based on machine learning, characterized by, The system is used to execute the method of any one of claims 1-5, and the system comprises: An acoustic wave detection module for determining an acoustic wave spectrum based on an acoustic camera for sound insulation detection, wherein the sound insulation detection includes building self sound insulation and foam sound insulation; A sound insulation coefficient acquisition module for determining a sound insulation coefficient based on the acoustic wave spectrum and combining building sound insulation standards for sound insulation performance evaluation, wherein the sound insulation coefficient is subjected to attenuation processing based on a gap tolerance interval; An optimization module for building an optimization decision module, wherein the optimization decision module comprises an optimization decision module and an acoustic evaluation module; An iteration module for determining a product optimization strategy based on the acoustic wave spectrum and the sound insulation coefficient and combining the optimization decision module for product optimization decision and sound insulation evaluation, wherein the optimization dimension comprises sound insulation optimization and lightweight optimization, and the sound insulation optimization includes a sound insulation coefficient; A production process determination module for converting the product optimization strategy into a process based on an initial process flow to determine an optimized production process; A production management module for optimizing the production management of sound insulation foam based on the optimized production process.

Images