Method and device for sound insulation prediction optimization of wood skeleton supporting body of high-speed train floor structure

Abstract

The embodiment of the invention discloses a sound insulation prediction optimization method and device for a wood bone supporting body of a high-speed train floor structure. The method comprises the following steps: acquiring main influence factors of a to-be-tested floor; determining the weighted sound reduction index of the to-be-detected floor; calculating a significant feature subset based on an mRMR feature screening method according to the main influence factors and the weighted sound reduction index; standardizing the significant feature subsets, inputting the standardized significant feature subsets into a prediction model, and performing sound insulation characteristic prediction and optimization in combination with constraint conditions to obtain an optimization scheme; wherein the prediction model is obtained by training a decision tree model by taking a plurality of standardized significant feature subsets as a sample set; and outputting an optimization scheme. By implementing the method provided by the embodiment of the invention, sound insulation characteristic prediction and scheme optimization can be carried out on the wooden skeleton support body in the high-speed train floor structure, the experiment cost is reduced, and the prediction efficiency and accuracy are improved.

Description

Technical Field

[0001] The invention relates to the technical field of vibration reduction and noise reduction in rail transit, and in particular to a method and device for predicting and optimizing sound insulation of a wooden frame support of a floor structure of a high-speed train. Background Art

[0002] As the speed of high-speed trains continues to increase, the noise inside and outside the train has increased significantly, and a large amount of noise is transmitted into the train through the body structure, seriously affecting the riding comfort of passengers and drivers. Among them, the floor structure of high-speed trains, especially the wheel-rail area close to the bottom of the train, is the main path for noise to propagate from the outside to the inside. Therefore, improving the sound insulation properties of the floor structure to reduce the impact of the noise under the train on the inside of the train has become a direct and effective way to improve the acoustic environment inside the train.

[0003] In the composite floor of high-speed trains, the wooden frame support is an indispensable component, and its proportion in the overall structure cannot be ignored. However, there are relatively few studies on this structure. Traditional floor sound insulation prediction methods usually rely on two methods: experiments and simulations. Experimental methods often require multiple on-site assembly and testing to obtain reliable data, which places high demands on laboratories and materials, resulting in a significant increase in experimental costs. Simulation methods are mainly based on finite element analysis or statistical energy methods for modeling and calculation, which requires the model to have high accuracy and detailed investigation of material parameters.

[0004] Therefore, it is necessary to design a new method to predict the sound insulation characteristics and optimize the scheme of the wooden frame support in the floor structure of high-speed trains, reduce the experimental cost, and improve the efficiency and accuracy of the prediction. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the prior art and provide a method and device for predicting and optimizing the sound insulation of a wooden frame support of a high-speed train floor structure.

[0006] To achieve the above object, the present invention adopts the following technical solution: a method for predicting and optimizing sound insulation of a wooden frame support of a high-speed train floor structure, comprising: Obtain the main influencing factors of the floor to be tested; Determine the weighted sound insulation of the floor to be tested; Calculate a significant feature subset based on the main influencing factors and the weighted sound insulation value based on the mRMR feature screening method; The significant feature subset is standardized and then input into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain an optimization scheme; wherein the prediction model is obtained by training the decision tree model using the standardized plurality of significant feature subsets as sample sets; The optimization solution is output.

[0007] A further technical solution is: determining the weighted sound insulation value of the floor to be tested includes: Obtaining the relevant sound source and received sound pressure levels of the floor to be tested obtained by performing sound pressure level tests in a sound source room and a receiving room; use Calculate the sound insulation of the floor to be tested; where L1 is the average sound pressure level in the source room, L2 is the average sound pressure level in the receiving room, S 1 is the surface area of ​​the floor to be tested, A is the sound absorption coefficient of the receiving room, , V is the volume of the receiving room, T is the reverberation time of the receiving room; Combined with the sound insulation frequency curve and the sound insulation of the floor to be tested, the weighted sound insulation is calculated according to the specified standard.

[0008] A further technical solution is: the significant feature subset is calculated based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation, including: The main influencing factors are analyzed and preprocessed to obtain original characteristic parameters; wherein the original characteristic parameters include basic material property parameters, acoustic parameters and structural layout properties, the basic material property parameters include density, elastic modulus, Poisson's ratio and shear modulus, the acoustic parameters include weighted sound insulation of a single vibration damping pad and a single wooden frame; the structural layout properties include the weight, length, width, thickness, quantity, contact area of ​​the vibration damping pad and the wooden frame, the number of wooden frames and the number of vibration damping pads on the wooden frames; Calculating the corresponding maxφ(D, R) value of the original characteristic parameter based on the mRMR characteristic screening method to obtain a calculation result; The original feature parameters are screened according to the calculation results to obtain a significant feature subset.

[0009] A further technical solution is: the prediction model is obtained by training a decision tree model using a number of standardized significant feature subsets as sample sets, including: Obtain the main influencing factors of the floor and determine the weighted sound insulation of the floor; According to the main influencing factors of the floor and the weighted sound insulation of the floor, the relevant significant feature subsets are calculated based on the mRMR feature screening method; The Min-max normalization method is used to linearly transform and map the relevant salient feature subsets to obtain a sample set; Build a decision tree model; The decision tree model is trained using the sample set to obtain a prediction model.

[0010] A further technical solution is that the hyperparameters of the decision tree model are searched and finally determined within their respective value ranges through an optimization algorithm.

[0011] A further technical solution is: the significant feature subset is standardized and then input into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain an optimization solution, including: The significant feature subset is standardized and then input into the prediction model, which uses the CART algorithm to predict the sound insulation characteristics, and determines the optimization scheme of the wooden frame support of the high-speed train floor structure in combination with the constraint conditions and the prediction results.

[0012] A further technical solution is as follows: the constraint conditions include the overall thickness and overall density of the floor structure.

[0013] The present invention provides a device for predicting and optimizing sound insulation of a wooden frame support of a high-speed train floor structure, comprising: A factor acquisition unit is used to obtain the main influencing factors of the floor to be tested; A weighted sound insulation value determination unit, used to determine the weighted sound insulation value of the floor to be tested; A significant feature subset calculation unit, used for calculating a significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation; A prediction optimization unit, used for inputting the standardized significant feature subset into a prediction model to predict and optimize the sound insulation characteristics in combination with constraint conditions to obtain an optimization solution; wherein the prediction model is obtained by training a decision tree model using a number of standardized significant feature subsets as sample sets; An output unit is used to output the optimization solution.

[0014] A further technical solution is as follows: the weighted sound insulation determination unit comprises: The sound pressure level acquisition subunit is used to acquire the relevant sound source and receiving sound pressure levels of the floor to be tested obtained by performing sound pressure level tests in the sound source room and the receiving room; Sound insulation determination subunit, used to adopt Calculate the sound insulation of the floor to be tested; where L1 is the average sound pressure level in the source room, L2 is the average sound pressure level in the receiving room, S 1 is the surface area of ​​the floor to be tested, A is the sound absorption coefficient of the receiving room, , V is the volume of the receiving room, and T is the reverberation time of the receiving room.

[0015] The weighted sound insulation calculation subunit is used to calculate the weighted sound insulation according to the specified standard by combining the sound insulation frequency curve and the sound insulation of the floor to be tested.

[0016] A further technical solution is as follows: the significant feature subset calculation unit comprises: A preprocessing subunit is used to analyze and preprocess the main influencing factors to obtain original characteristic parameters; wherein the original characteristic parameters include basic material property parameters, acoustic parameters and structural layout properties, the basic material property parameters include density, elastic modulus, Poisson's ratio and shear modulus, the acoustic parameters include weighted sound insulation of a single vibration damping pad and a single wooden frame; the structural layout properties include the weight, length, width, thickness, number, contact area of ​​the vibration damping pad and the wooden frame, the number of wooden frames and the number of vibration damping pads on the wooden frames; A calculation subunit, used for calculating the corresponding maxφ(D, R) value of the original characteristic parameter based on the mRMR characteristic screening method to obtain a calculation result; The screening subunit is used to screen the original feature parameters according to the calculation results to obtain a significant feature subset.

[0017] Compared with the prior art, the present invention has the following beneficial effects: the present invention identifies the main influencing factors of the floor to be tested and determines its weighted sound insulation, calculates the significant feature subset by the mRMR feature screening method, and inputs the subset into the prediction model after standardization to predict and optimize the sound insulation characteristics; finally, outputs an optimization scheme based on the prediction result, realizes the prediction of the sound insulation characteristics and scheme optimization of the wooden frame support in the floor structure of the high-speed train, reduces the experimental cost, and improves the efficiency and accuracy of the prediction.

[0018] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0020] Figure 1 A schematic flow chart of a method for predicting and optimizing sound insulation of a wooden frame support of a high-speed train floor structure provided by an embodiment of the present invention; Figure 2 A schematic diagram of a composite floor structure provided by an embodiment of the present invention; Figure 3 Schematic diagram of the test provided by the embodiment of the present invention Figure 1 ; Figure 4 Schematic diagram of the test provided by the embodiment of the present invention Figure 2 ; Figure 5 A schematic diagram of the maxφ(D, R) value calculated based on the mRMR algorithm provided in an embodiment of the present invention; Figure 6 A schematic diagram of prediction and measured results of a training set of a decision tree model provided in an embodiment of the present invention; Figure 7 A schematic diagram of prediction and measured results of a test set of a decision tree model provided in an embodiment of the present invention; Figure 8 An optimized schematic diagram of the maximum sound insulation solution provided by an embodiment of the present invention; Fig. 9 A schematic block diagram of a device for predicting and optimizing sound insulation of a wooden frame support of a high-speed train floor structure provided by an embodiment of the present invention; Fig.10 A schematic block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0023] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0024] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0025] See also Figure 1 , Figure 1A schematic flow chart of a method for predicting and optimizing the sound insulation of a wooden frame support for a floor structure of a high-speed train provided in an embodiment of the present invention. The method for predicting and optimizing the sound insulation of a wooden frame support for a floor structure of a high-speed train is applied to a server. First, the main influencing factors of the floor to be tested and their related sound pressure level data are collected to calculate the sound insulation; the sound pressure level test results of the source room and the receiving room are used, combined with the sound absorption coefficient and reverberation time of the receiving room, to calculate the weighted sound insulation of the floor to be tested; the mRMR feature screening method is used to analyze the original feature parameters (such as material properties and structural layout) and screen out a significant feature subset; the significant feature subset is Min-max standardized to prepare a training data set; the decision tree model is trained using the standardized sample set, and the hyperparameters are optimized to improve the model performance; the standardized significant features are input into the decision tree model, the sound insulation characteristics are predicted using the CART algorithm, and the optimization scheme is obtained in combination with the structural constraints, thereby reducing the experimental cost and improving the prediction efficiency and accuracy.

[0026] Figure 1 1 is a flow chart of a method for predicting and optimizing the sound insulation of a wooden frame support of a high-speed train floor structure provided by an embodiment of the present invention. Figure 1 As shown, the method includes the following steps S110 to S150.

[0027] S110. Obtain the main influencing factors of the floor to be tested.

[0028] In this example, see Figure 2 , the method of this embodiment is aimed at composite flooring.

[0029] The main influencing factors of the floor to be tested usually include the following aspects: Material properties: The basic physical properties of floor materials such as density, elastic modulus, Poisson's ratio and shear modulus will affect its sound insulation performance. Different materials have different absorption and reflection characteristics in sound wave propagation.

[0030] Structural layout: The design and structural configuration of the floor, including the number, layout, size (such as length, width, thickness) of wooden frame supports and the location and number of vibration damping pads, will affect the transmission and isolation of sound waves.

[0031] Contact area: The contact area between the floor and other structures (such as walls and supporting columns) will affect the propagation path and energy transfer of sound waves, thereby affecting the sound insulation effect.

[0032] Acoustic parameters: such as the weighted sound insulation of a single vibration damping pad and wooden frame of the floor, directly affect the floor's ability to absorb and attenuate sound waves.

[0033] Environmental factors: Environmental parameters such as the sound absorption coefficient, volume and reverberation time of the receiving room will also affect the sound insulation test results.

[0034] In the structural composition of the floor to be tested, the aluminum profiles, floor cloths, interlayer materials, etc. that are not mentioned will not change. By comprehensively considering these major influencing factors, the sound insulation characteristics of the floor can be more accurately predicted and the corresponding optimization design can be carried out.

[0035] S120: Determine the weighted sound insulation value of the floor to be tested.

[0036] In this embodiment, the weighted sound reduction index (usually represented by Rw) is an indicator used to measure the sound wave insulation ability of building components (such as walls, floors, etc.). It reflects the sound attenuation effect of the component within a specific frequency range, usually in decibels (dB).

[0037] The calculation of weighted sound insulation takes into account the sensitivity of the human ear to sounds of different frequencies, usually using the A-weighting or C-weighting method. A-weighting is mainly used to evaluate the impact on human voice, while C-weighting is used to consider the impact of low-frequency noise. The higher the Rw value, the better the sound insulation effect of the component, and the more effectively it can isolate external noise.

[0038] In practical applications, weighted sound insulation is widely used in architectural acoustic design, indoor environmental noise control and other fields to help engineers and designers choose appropriate materials and structures to meet sound insulation requirements.

[0039] In this embodiment, the reverberation method is used to obtain the original data of the floor to determine the weighted sound insulation.

[0040] In one embodiment, the above step S120 may include steps S121 to S123; S121. Obtain relevant sound source and receiving sound pressure levels of the floor to be tested obtained by performing sound pressure level tests in a sound source room and a receiving room.

[0041] In this embodiment, the sound insulation of the composite floor is preferably tested in the sound source room and the receiving room according to the standard, and the weighted sound insulation is calculated. Figure 3 and Figure 4 The weighted sound insulation of the composite floor is obtained by the reverberation test method, which may include: the sensors, sound source locations and test methods in the sound source room and the receiving room of the test are in accordance with the standard ISO10140-2-2021; before the method is implemented, the structural parameters of the composite floor are known. Figure 3 and Figure 4As shown, the sound source is a 12-sided omnidirectional sound source that can emit sound evenly in 12 different directions (i.e., from 12 sides) without specific directionality. The characteristic of this sound source is that the intensity of sound propagation in all directions is the same, without directional bias. Sound propagation is carried out with the help of a microphone so that the receiving room can receive the sound pressure.

[0042] Carry out sound pressure level tests in the source room and receiving room, and record the source and received sound pressure levels.

[0043] Testing should be performed in accordance with standards (such as ISO 10140-2-2021) to ensure accuracy and comparability of data.

[0044] Obtaining accurate sound pressure level data is the basis for calculating sound insulation. Ensuring that the test process complies with the standard improves the credibility and validity of the results. The sound insulation performance of the floor to be tested can be judged by comparing the difference in sound pressure levels between the source room and the receiving room.

[0045] S122, use Calculate the sound insulation of the floor to be tested; where L1 is the average sound pressure level in the source room, L2 is the average sound pressure level in the receiving room, S 1 is the surface area of ​​the floor to be tested, A is the sound absorption coefficient of the receiving room, , V is the volume of the receiving room, and T is the reverberation time of the receiving room.

[0046] In this embodiment, R is the sound insulation of the floor to be tested. By comprehensively considering multiple factors (such as sound pressure level, sound absorption coefficient and reverberation time), a more accurate sound insulation can be obtained. This calculation can quantify the sound insulation performance of the floor to be tested and provide data support for subsequent analysis.

[0047] S123. Calculate the weighted sound insulation value according to the specified standard by combining the sound insulation frequency curve and the sound insulation value of the floor to be tested.

[0048] In this embodiment, the sound insulation frequency curve is a graph showing the sound insulation performance of a material or structure at different frequencies. It usually shows the sound insulation (in dB) at a range of frequencies (such as from 20 Hz to 20 kHz), reflecting the material's ability to isolate sound waves.

[0049] The weighted sound insulation can reflect the sound insulation effect at different frequencies, especially the performance difference between low and high frequencies. It provides designers and engineers with a more intuitive evaluation of sound insulation performance, which facilitates the selection of materials and structures in practical applications. This step helps ensure that the product meets the relevant sound insulation standards and requirements, and provides support for subsequent market promotion.

[0050] Through these three steps, the sound insulation performance of the floor to be tested can be systematically evaluated, from data collection to calculation, and then to standardized evaluation, to ensure the accuracy and reliability of the results. This process not only provides a theoretical basis for engineering design, but also lays the foundation for subsequent product improvement and market application.

[0051] S130, calculating a significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation.

[0052] In this embodiment, the significant feature subset refers to a part of the features that have the greatest impact on the model performance selected from the original feature set during the feature selection process. These features can effectively improve the accuracy of the model, reduce computational complexity, and prevent overfitting.

[0053] The features in the salient feature subset usually have a greater impact on the results in tasks such as classification, regression or clustering. By selecting these features, the performance of the model can be improved; through feature selection, the salient feature subset can help reduce the dimension of the data, making data processing more efficient and model training faster; excluding irrelevant or redundant features can reduce the noise in the data, thereby improving the generalization ability of the model.

[0054] In one embodiment, the above-mentioned step S130 may include steps S131 to S133.

[0055] S131. Analyze and preprocess the main influencing factors to obtain original characteristic parameters; wherein the original characteristic parameters include basic material property parameters, acoustic parameters and structural layout properties. The basic material property parameters include density, elastic modulus, Poisson's ratio and shear modulus. The acoustic parameters include weighted sound insulation of a single vibration damping pad and a single wooden frame. The structural layout properties include the weight, length, width, thickness, number, contact area of ​​the vibration damping pad and the wooden frame, as well as the number of wooden frames and the number of vibration damping pads on the wooden frames.

[0056] In this example, based on the sound insulation test results of the high-speed train floor structure and the experimental data of the composite floor, the main influencing factors of the wooden frame support were analyzed. Under the laboratory conditions and the material setting restrictions of the acoustic software, the following main parameters were determined: Basic material properties: including density, elastic modulus, Poisson's ratio and shear modulus, etc.

[0057] Acoustic parameters: mainly include the weighted sound insulation Rw of a single vibration damping pad and a single wooden frame.

[0058] Structural layout attributes: covering the weight, length, width, thickness, quantity of vibration-damping pads and wooden frames, the contact area between vibration-damping pads and wooden frames, the number of wooden frames, and the contact area between wooden frames and profiles, etc.

[0059] By analyzing these 23 original characteristic parameters, we can have a more comprehensive understanding of the relationship between materials, acoustic properties and structural layout, thereby optimizing the sound insulation effect of high-speed train floors.

[0060] By identifying the key parameters that affect sound insulation performance, we can guide floor design and improve the overall sound insulation effect. Enhance the sound insulation characteristics, improve the comfort and safety of passengers, and meet the high standards of modern transportation. Through effective material selection and structural layout, reduce unnecessary material use, thereby reducing production and maintenance costs. Provide data support for subsequent research and promote further exploration and technological progress in related fields.

[0061] Among them, the main influencing factors of the wooden frame support are shown in Table 1.

[0062] Table 1. Main influencing factors of wooden frame support

[0063] S132, calculating the corresponding maxφ(D, R) value of the original characteristic parameter based on the mRMR characteristic screening method to obtain a calculation result.

[0064] In this embodiment, the calculation result refers to the maxφ(D, R) value corresponding to each original characteristic parameter.

[0065] This implementation uses the maximum relevance minimum redundancy (mRMR) regression data feature selection algorithm, which uses mutual information to measure the correlation between variables. Given two discrete random variables X and Y, their marginal probability distribution functions are p(x), p(y), and their joint probability distribution function is p(x, y), then the mutual information of X and Y is: ; Assume that the target feature subset of mRMR is S, the number of features to be selected is m, xi, xj (i, j = 1, 2, ..., m, i ≠ j) represent any two features in S, and c represents the target variable. Then, the m features with the greatest correlation with c can be calculated by formula (4), and the redundancy among the m features can be eliminated by formula (5). ; ; Combine the maximum correlation D with the minimum redundancy R, define an operator to combine D and R, and consider the simplest combination method: ; Finally, the feature set S with maximum relevance and minimum redundancy is obtained: .

[0066] S133: Filter the original feature parameters according to the calculation results to obtain a significant feature subset.

[0067] In this embodiment, the programming software is used to input the code, and the maxφ(D, R) values ​​corresponding to the 23 original feature parameters in Table 1 are calculated. Here, maxφ(D, R)=1 is used as the dividing line for distinguishing whether the feature is suitable or not. Figure 5 The feature parameters are sorted out and the feature parameters of F20, F15 and F16 are manually selected and stored in the significant feature subset, and a new number is given as shown in Table 2.

[0068] Table 2. Subset of significant features

[0069] S140, standardizing the significant feature subset and inputting it into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain an optimization solution; wherein the prediction model is obtained by training a decision tree model using a number of standardized significant feature subsets as sample sets.

[0070] In this embodiment, the optimization scheme refers to the result of predicting the sound insulation characteristics through a prediction model (such as a decision tree model) based on a standardized significant feature subset, and determining the optimal scheme of the support structure after combining the constraint conditions.

[0071] Standardize the subset of significant features to eliminate the dimensional differences between different features and ensure the effectiveness of model training. Use the standardized subset of significant features as a sample set to train the decision tree model to form a model that can predict sound insulation characteristics. Input new features into the model to predict sound insulation characteristics and evaluate the performance of existing designs.

[0072] Based on the model output, adjustments are made based on the design requirements and constraints in actual applications (such as cost, material restrictions, etc.). Based on the model prediction results, the configuration of vibration damping pads and wooden frames, including quantity, size, material, etc., is optimized to improve the sound insulation performance. After optimization, the sound insulation characteristics are re-evaluated to ensure that the expected goals are achieved. Based on the prediction results and actual test data, the optimization plan is iterated to gradually improve the sound insulation effect of the design.

[0073] The significant feature subset is standardized and then input into the prediction model, which uses the CART algorithm to predict the sound insulation characteristics, and determines the optimization scheme of the wooden frame support of the high-speed train floor structure in combination with the constraint conditions and the prediction results.

[0074] Specifically, the constraints include the overall thickness and overall density of the floor structure.

[0075] The overall sound insulation is the core goal of the floor structure acoustic design. In order to ensure the feasibility and effectiveness of the design, "constraints" were formulated in combination with the actual engineering situation and project design goals, mainly including overall thickness and overall density.

[0076] Overall thickness: The floor structure must follow certain space and material restrictions when assembled. As an important part of the floor structure, the overall thickness of the wooden frame support is directly affected by the assembly conditions. Reasonable thickness design can ensure the stability of the structure and the sound insulation effect.

[0077] Overall density: In the design of high-speed trains, lightweighting is an important strategy to improve operating efficiency. Reducing the overall density can not only reduce the demand for traction power, but also effectively improve the energy efficiency of the train. Therefore, the project sets a limit of no more than 1000 kg / m³ on the overall density of the wooden frame support to support the lightweight goal.

[0078] like Figure 8 As shown in the figure, by calculating the permutations and combinations of the significant feature subsets, we initially obtained 27,512,540 possible design solutions. However, not all solutions meet the actual assembly requirements, so we further screened them to ensure that the following conditions were met: The product of the number of vibration-damping pads on a single wooden frame and its length shall not exceed the length of the single wooden frame.

[0079] The width of the vibration damping pad cannot be greater than the width of the wooden frame.

[0080] After eliminating unreasonable solutions, the decision tree prediction model was used to calculate the remaining solutions, and finally the best solution that met the feasibility of the project was screened out. A total of 283,541 solutions met all constraints and assembly requirements.

[0081] After detailed calculations, the maximum sound insulation solution identified has the following characteristics: Weighted sound insulation: 50.5 dB (verified by the reverberation method test, the actual measured value is 50.55 dB, with an error of only 0.05 dB, indicating the rationality and accuracy of the optimization result).

[0082] The specific values ​​of the significant characteristic parameters are given in Table 3.

[0083] Table 3. Structural parameters of the maximum sound insulation solution

[0084] The overall density of the maximum sound insulation solution meets the weight requirement of no more than 1000 kg / m3; the thickness of the wooden frame is 41.4 mm, the thickness of the vibration damping pad is 12 mm, and the overall thickness of the wooden frame support is 53.4 mm, which can meet the project requirements. Based on the above analysis, the solution obtained meets all the limiting conditions and is a feasible solution. In addition, it is verified by the reverberation method test.

[0085] For the optimization process, the optimization goal is to determine the optimization design target of the wooden frame support of the high-speed train floor structure based on the decision tree model as the maximum sound insulation.

[0086] Constraints: Under project limitations or planning requirements, set the following constraints: Overall thickness: The floor structure is subject to conditions during assembly, and the overall thickness of the wooden frame support must comply with these limitations.

[0087] Overall density: In order to achieve lightweight high-speed trains, reduce traction power requirements and improve operating efficiency, the overall density of the wooden frame support must be controlled within a certain range.

[0088] Optimization implementation: Based on the decision tree machine learning model, the optimal design scheme for the wooden frame support of the high-speed train floor structure is determined: First, the newly generated sample library is arranged and combined to calculate various design solutions.

[0089] Then, those solutions that are unreasonable or do not meet the actual assembly requirements are eliminated.

[0090] Finally, based on the set constraints, the decision tree learning model is used to predict and determine the maximum sound insulation and the best optimization solution.

[0091] For the prediction model, the entire training process is as follows, including steps S141 to S145.

[0092] S141. Obtain the main influencing factors of the floor and determine the weighted sound insulation of the floor; S142. According to the main influencing factors of the floor and the weighted sound insulation of the floor, the relevant significant feature subset is calculated based on the mRMR feature screening method.

[0093] In this embodiment, the specific implementation process of the above steps S141-S142 is consistent with that of steps S110-S130, and will not be repeated here.

[0094] S143. Use the Min-max normalization method to perform linear transformation and mapping on the relevant significant feature subset to obtain a sample set.

[0095] In this embodiment, at this stage, the subset of significant features formed after screening is standardized to make the data dimensionless to eliminate the dimensional differences between different features. The Min-max standardization method is used to map the values ​​of all features to the interval [0, 1]. For the sample sequence x1, x2, ..., xn, the transformed sequence is y1, y2, ..., yn, and the formula is as follows: .

[0096] Different features may have different dimensions. After standardization, the features are kept within the same range to avoid undue influence of certain features on the model due to dimensionality issues.

[0097] Normalization can accelerate model convergence, especially when using gradient descent, which helps improve training efficiency and prediction accuracy.

[0098] S144. Construct a decision tree model.

[0099] In this embodiment, the hyperparameters of the decision tree model are searched and finally determined within their respective value ranges through an optimization algorithm.

[0100] In the decision tree model involved, the main hyperparameters are optimized by the optimal selection algorithm. After searching the respective value ranges, they are finally determined as follows: Maximum tree depth: set to 4. This parameter controls the depth of the tree, that is, the longest path from the root node to the leaf node. Limiting the depth helps avoid model complexity and reduce the risk of overfitting.

[0101] Maximum number of splitting features: Set to "Auto". This parameter determines the maximum number of features considered at each node split. Automatically selecting the number of features can increase the flexibility and adaptability of the model, ensuring that the model makes full use of the available features.

[0102] Minimum number of samples required for leaf nodes: Set to 1. This parameter limits the minimum number of samples required for each leaf node to avoid overfitting the model on the training data and ensure the generalization ability of the model.

[0103] Split criteria: Select the Gini index as the measure of split quality. The Gini index is used to evaluate the purity of the data set and help the model select the optimal features during the splitting process, thereby improving the accuracy of classification.

[0104] After completing the hyperparameter setting, the next steps include dividing the dataset into training and test sets. The model is trained on the training set so that it can learn the patterns in the data; then, the model is validated on the test set to evaluate its performance and generalization ability. This ensures the effectiveness and accuracy of the model in real-world applications.

[0105] S145. Use the sample set to train the decision tree model to obtain a prediction model.

[0106] During the training process, mean square error (MSE) and square correlation coefficient (R²) are used as evaluation criteria for model training and validation. The sample data is randomly divided into 70% training set and 30% test set, which are used for model training and validation respectively. The calculation formulas of MSE and R² are as follows: ; .

[0107] MSE provides an absolute measure of the prediction error, while R² reflects the model's ability to explain data variation. The combination of the two can comprehensively evaluate the performance of the model. Through these indicators, the degree of fit of the model can be effectively judged, thus providing a basis for subsequent model improvement and tuning.

[0108] In this stage, the CART algorithm is used to build a decision tree model for the training set. The input variables and output variables of the model are X and Y respectively. The algorithm selects the optimal split point by traversing each feature variable and its value, and divides the sample data into different subspaces. The loss function is optimized using square error, and the optimal value is finally output. The split attribute and split value are selected for data partitioning.

[0109] Specifically, assuming that X and Y are input and output variables respectively, a training data set is formed. Each feature variable and its corresponding value are traversed. Suppose the current segmentation variable is the jth variable and the corresponding segmentation value is s, then two regions can be divided and defined.

[0110] Then, the input space will be continuously divided into L subspaces , , …, , each subspace Both contain some sample data and output values , the solution of the current model can be expressed as: ; The error size of the loss function is calculated by square error To compare, use the square error minimization criterion to determine the optimal split point and prediction value. Then according to the least squares, subspace All output on The mean of is the optimal value of , which can be expressed as: ; Select the optimal segmentation attribute j and attribute value s to divide the input space, expressed as: ; Traverse all input feature variables and their values ​​to find the current optimal split point , and then divide the current space into two sub-areas according to the split point. If the two sub-areas cannot be divided any further, the corresponding optimal output value can be obtained, which is expressed as: ; The CART algorithm can process various types of data (numerical and categorical) and does not require excessive preprocessing of the data, making it suitable for diverse data sets.

[0111] The tree structure generated by the decision tree model is easy to understand and interpret, which helps domain experts analyze the impact of features on the output.

[0112] By selecting the optimal segmentation attributes, the model can automatically identify and select the most important features for prediction, improving the efficiency and accuracy of the model.

[0113] In the result comparison, the measured and predicted results of the training set and the test set show good similarity. The MSE of the training set is 0.1105668 and R² is 0.8905, and the MSE of the test set is 0.1412402 and R² is 0.8213, which shows that the model performs well in both the training and validation stages. Figure 6 He Ru Figure 7 shown.

[0114] S150: Output the optimization solution.

[0115] The method of this embodiment is based on experimental data, adopts a decision tree machine learning method, and screens feature parameters through the mRMR feature selection algorithm. Subsequently, the trained decision tree model is used to predict sound insulation. Under the set constraints, with the help of this model, the optimal design scheme of the wooden frame support can be determined, thereby effectively saving manpower, time and economic costs.

[0116] The above-mentioned sound insulation prediction and optimization method for the wooden frame support of the floor structure of high-speed trains identifies the main influencing factors of the floor to be tested and determines its weighted sound insulation, calculates the significant feature subset through the mRMR feature screening method, and inputs it into the prediction model after standardization to predict and optimize the sound insulation characteristics; finally, based on the prediction results, an optimization scheme is output to realize the prediction of the sound insulation characteristics and scheme optimization of the wooden frame support in the floor structure of high-speed trains, reduce the experimental cost, and improve the efficiency and accuracy of the prediction.

[0117] Fig.10 1 is a schematic block diagram of a high-speed train floor structure wood frame support body sound insulation prediction and optimization device 300 provided by an embodiment of the present invention. Fig.10As shown, corresponding to the above high-speed train floor structure wood frame support body sound insulation prediction optimization method, the present invention also provides a high-speed train floor structure wood frame support body sound insulation prediction optimization device 300. The high-speed train floor structure wood frame support body sound insulation prediction optimization device 300 includes a unit for executing the above high-speed train floor structure wood frame support body sound insulation prediction optimization method, and the device can be configured in a server. Specifically, please refer to Fig.10 The high-speed train floor structure wood frame support body sound insulation prediction and optimization device 300 includes a factor acquisition unit 301, a weighted sound insulation determination unit 302, a significant feature subset calculation unit 303, a prediction optimization unit 304 and an output unit 305.

[0118] The factor acquisition unit 301 is used to acquire the main influencing factors of the floor to be tested; the weighted sound insulation determination unit 302 is used to determine the weighted sound insulation of the floor to be tested; the significant feature subset calculation unit 303 is used to calculate the significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation; the prediction optimization unit 304 is used to standardize the significant feature subset and input it into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain an optimization plan; wherein the prediction model is obtained by training the decision tree model with a number of standardized significant feature subsets as sample sets; the output unit 305 is used to output the optimization plan.

[0119] In one embodiment, the weighted sound insulation determination unit 302 includes: The sound pressure level acquisition subunit is used to acquire the relevant sound source and receiving sound pressure levels of the floor to be tested obtained by performing sound pressure level tests in the sound source room and the receiving room; Sound insulation determination subunit, used to adopt Calculate the sound insulation of the floor to be tested; where L1 is the average sound pressure level in the source room, L2 is the average sound pressure level in the receiving room, S 1 is the surface area of ​​the floor to be tested, A is the sound absorption coefficient of the receiving room, , V is the volume of the receiving room, and T is the reverberation time of the receiving room.

[0120] The weighted sound insulation calculation subunit is used to calculate the weighted sound insulation according to the specified standard by combining the sound insulation frequency curve and the sound insulation of the floor to be tested.

[0121] In one embodiment, the salient feature subset calculation unit 303 includes: A preprocessing subunit is used to analyze and preprocess the main influencing factors to obtain original characteristic parameters; wherein the original characteristic parameters include basic material property parameters, acoustic parameters and structural layout properties, the basic material property parameters include density, elastic modulus, Poisson's ratio and shear modulus, the acoustic parameters include weighted sound insulation of a single vibration damping pad and a single wooden frame; the structural layout properties include the weight, length, width, thickness, number, contact area of ​​the vibration damping pad and the wooden frame, the number of wooden frames and the number of vibration damping pads on the wooden frames; A calculation subunit, used for calculating the corresponding maxφ(D, R) value of the original characteristic parameter based on the mRMR characteristic screening method to obtain a calculation result; The screening subunit is used to screen the original feature parameters according to the calculation results to obtain a significant feature subset.

[0122] In one embodiment, the above-mentioned device further includes a prediction model generation unit, and the prediction model generation unit includes: Obtain subunits to obtain the main influencing factors of the floor and determine the weighted sound insulation of the floor; A processing subunit, used for calculating relevant significant feature subsets based on the mRMR feature screening method according to the main influencing factors of the floor and the weighted sound insulation of the floor; A mapping subunit, used to linearly transform and map the relevant salient feature subsets using a Min-max normalization method to obtain a sample set; Construct subunits for building decision tree models; The training subunit is used to train the decision tree model using the sample set to obtain a prediction model.

[0123] In one embodiment, the prediction optimization unit 304 is used to standardize the significant feature subset and input it into the prediction model, and the prediction model uses the CART algorithm to predict the sound insulation characteristics, and determines the optimization scheme of the wooden frame support of the high-speed train floor structure in combination with the constraint conditions and the prediction results.

[0124] It should be noted that technicians in the relevant field can clearly understand that the specific implementation process of the above-mentioned high-speed train floor structure wooden frame support body sound insulation prediction and optimization device 300 and each unit can refer to the corresponding description in the aforementioned method embodiment, and for the convenience and brevity of description, it will not be repeated here.

[0125] The above-mentioned high-speed train floor structure wood frame support body sound insulation prediction and optimization device 300 can be implemented in the form of a computer program. The computer program can be used in Fig.10 Runs on the computer device shown.

[0126] See also Fig.10 , Fig.10 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 is a server, which may be an independent server or a server cluster composed of multiple servers.

[0127] See also Fig.10 The computer device 500 includes a processor 502 , a memory and a network interface 505 connected via a system bus 501 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .

[0128] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions, which, when executed, can cause the processor 502 to execute a method for predicting and optimizing sound insulation of a wooden frame support of a high-speed train floor structure.

[0129] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0130] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a method for predicting and optimizing sound insulation of a wooden frame support of a high-speed train floor structure.

[0131] The network interface 505 is used to communicate with other devices over the network. Fig.10 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0132] The processor 502 is used to run the computer program 5032 stored in the memory to implement the following steps: Obtain the main influencing factors of the floor to be tested; determine the weighted sound insulation of the floor to be tested; calculate the significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation; standardize the significant feature subset and input it into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain an optimization plan; wherein the prediction model is obtained by training the decision tree model by using several standardized significant feature subsets as sample sets; and output the optimization plan.

[0133] In one embodiment, when the processor 502 implements the step of determining the weighted sound insulation value of the floor to be tested, the processor 502 specifically implements the following steps: Obtain the relevant sound source and received sound pressure levels of the floor to be tested by performing sound pressure level tests in the sound source room and the receiving room; Calculate the sound insulation of the floor to be tested; where L1 is the average sound pressure level in the source room, L2 is the average sound pressure level in the receiving room, S 1 is the surface area of ​​the floor to be tested, A is the sound absorption coefficient of the receiving room, , V is the volume of the receiving room, T is the reverberation time of the receiving room; combined with the sound insulation frequency curve and the sound insulation of the floor to be tested, the weighted sound insulation is calculated according to the specified standard.

[0134] In one embodiment, when the processor 502 implements the step of calculating a significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation, the processor 502 specifically implements the following steps: The main influencing factors are analyzed and preprocessed to obtain original characteristic parameters; wherein the original characteristic parameters include basic material property parameters, acoustic parameters and structural layout properties, the basic material property parameters include density, elastic modulus, Poisson's ratio and shear modulus, the acoustic parameters include weighted sound insulation of a single vibration damping pad and a single wooden frame; the structural layout properties include the weight, length, width, thickness, quantity, contact area of ​​the vibration damping pad and the wooden frame, the number of wooden frames and the number of vibration damping pads on the wooden frames; the corresponding maxφ(D, R) values ​​of the original characteristic parameters are calculated based on the mRMR feature screening method to obtain calculation results; the original characteristic parameters are screened according to the calculation results to obtain a significant feature subset.

[0135] In one embodiment, when the processor 502 implements the step of training a decision tree model by using a plurality of standardized significant feature subsets as sample sets, the prediction model specifically implements the following steps: The main influencing factors of the floor are obtained to determine the weighted sound insulation of the floor; the relevant significant feature subsets are calculated based on the mRMR feature screening method according to the main influencing factors of the floor and the weighted sound insulation of the floor; the relevant significant feature subsets are linearly transformed and mapped using the Min-max standardization method to obtain a sample set; a decision tree model is constructed; and the decision tree model is trained using the sample set to obtain a prediction model.

[0136] The hyperparameters of the decision tree model are searched and finally determined within their respective value ranges through an optimization algorithm.

[0137] In one embodiment, when the processor 502 implements the step of normalizing the significant feature subset and inputting it into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain the optimization solution, the following steps are specifically implemented: The significant feature subset is standardized and then input into the prediction model, which uses the CART algorithm to predict the sound insulation characteristics, and determines the optimization scheme of the wooden frame support of the high-speed train floor structure in combination with the constraint conditions and the prediction results.

[0138] The constraints include the overall thickness and overall density of the floor structure.

[0139] It should be understood that in the embodiment of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0140] It can be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment can be completed by instructing the relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiment of the above method.

[0141] Therefore, the present invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps: Obtain the main influencing factors of the floor to be tested; determine the weighted sound insulation of the floor to be tested; calculate the significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation; standardize the significant feature subset and input it into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain an optimization plan; wherein the prediction model is obtained by training the decision tree model by using several standardized significant feature subsets as sample sets; and output the optimization plan.

[0142] In one embodiment, when the processor executes the computer program to implement the step of determining the weighted sound insulation value of the floor to be tested, the processor specifically implements the following steps: Obtain the relevant sound source and received sound pressure levels of the floor to be tested by performing sound pressure level tests in the sound source room and the receiving room; Calculate the sound insulation of the floor to be tested; where L1 is the average sound pressure level in the source room, L2 is the average sound pressure level in the receiving room, S 1 is the surface area of ​​the floor to be tested, A is the sound absorption coefficient of the receiving room, , V is the volume of the receiving room, T is the reverberation time of the receiving room; combined with the sound insulation frequency curve and the sound insulation of the floor to be tested, the weighted sound insulation is calculated according to the specified standard.

[0143] In one embodiment, when the processor executes the computer program to implement the step of calculating a significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation, the processor specifically implements the following steps: The main influencing factors are analyzed and preprocessed to obtain original characteristic parameters; wherein the original characteristic parameters include basic material property parameters, acoustic parameters and structural layout properties, the basic material property parameters include density, elastic modulus, Poisson's ratio and shear modulus, the acoustic parameters include weighted sound insulation of a single vibration damping pad and a single wooden frame; the structural layout properties include the weight, length, width, thickness, quantity, contact area of ​​the vibration damping pad and the wooden frame, the number of wooden frames and the number of vibration damping pads on the wooden frames; the corresponding maxφ(D, R) values ​​of the original characteristic parameters are calculated based on the mRMR feature screening method to obtain calculation results; the original characteristic parameters are screened according to the calculation results to obtain a significant feature subset.

[0144] In one embodiment, when the processor executes the computer program to implement the step that the prediction model is obtained by training the decision tree model by using the standardized plurality of significant feature subsets as sample sets, the processor specifically implements the following steps: The main influencing factors of the floor are obtained to determine the weighted sound insulation of the floor; the relevant significant feature subsets are calculated based on the mRMR feature screening method according to the main influencing factors of the floor and the weighted sound insulation of the floor; the relevant significant feature subsets are linearly transformed and mapped using the Min-max standardization method to obtain a sample set; a decision tree model is constructed; and the decision tree model is trained using the sample set to obtain a prediction model.

[0145] The hyperparameters of the decision tree model are searched and finally determined within their respective value ranges through an optimization algorithm.

[0146] In one embodiment, when the processor executes the computer program to implement the step of normalizing the significant feature subset and inputting it into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain the optimization solution, the following steps are specifically implemented: The significant feature subset is standardized and then input into the prediction model, which uses the CART algorithm to predict the sound insulation characteristics, and determines the optimization scheme of the wooden frame support of the high-speed train floor structure in combination with the constraint conditions and the prediction results.

[0147] The constraints include the overall thickness and overall density of the floor structure.

[0148] The storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which are computer-readable storage media that can store program codes.

[0149] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0150] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0151] The steps in the method of the embodiment of the present invention can be adjusted in order, combined and deleted according to actual needs. The units in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs. In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0152] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention.

[0153] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure, characterized in that: include: Obtain the main influencing factors of the floor to be tested; Determine the weighted sound insulation of the floor to be tested; Calculate a significant feature subset based on the main influencing factors and the weighted sound insulation value based on the mRMR feature screening method; The significant feature subset is standardized and then input into the prediction model to predict and optimize the sound insulation characteristics in combination with the constraint conditions to obtain an optimization scheme; wherein the prediction model is obtained by training the decision tree model using the standardized plurality of significant feature subsets as sample sets; The optimization solution is output.

2. The method for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure according to claim 1 is characterized in that: Determining the weighted sound insulation value of the floor to be tested includes: Obtaining the relevant sound source and received sound pressure levels of the floor to be tested obtained by performing sound pressure level tests in a sound source room and a receiving room; use Calculate the sound insulation of the floor to be tested; where L1 is the average sound pressure level in the source room, L2 is the average sound pressure level in the receiving room, S1 is the surface area of ​​the floor to be tested, A is the sound absorption coefficient of the receiving room, , V is the volume of the receiving room, T is the reverberation time of the receiving room; Combined with the sound insulation frequency curve and the sound insulation of the floor to be tested, the weighted sound insulation is calculated according to the specified standard.

3. The method for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure according to claim 1 is characterized in that: The calculating of the significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation value includes: The main influencing factors are analyzed and preprocessed to obtain original characteristic parameters; wherein the original characteristic parameters include basic material property parameters, acoustic parameters and structural layout properties, the basic material property parameters include density, elastic modulus, Poisson's ratio and shear modulus, the acoustic parameters include weighted sound insulation of a single vibration damping pad and a single wooden frame; the structural layout properties include the weight, length, width, thickness, quantity, contact area of ​​the vibration damping pad and the wooden frame, the number of wooden frames and the number of vibration damping pads on the wooden frames; Calculating the corresponding maxφ(D, R) value of the original characteristic parameter based on the mRMR characteristic screening method to obtain a calculation result; The original feature parameters are screened according to the calculation results to obtain a significant feature subset.

4. The method for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure according to claim 1 is characterized in that: The prediction model is obtained by training the decision tree model using several standardized significant feature subsets as sample sets, including: Obtain the main influencing factors of the floor and determine the weighted sound insulation of the floor; According to the main influencing factors of the floor and the weighted sound insulation of the floor, the relevant significant feature subsets are calculated based on the mRMR feature screening method; The Min-max normalization method is used to linearly transform and map the relevant salient feature subsets to obtain a sample set; Build a decision tree model; The decision tree model is trained using the sample set to obtain a prediction model.

5. The method for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure according to claim 4 is characterized in that: The hyperparameters of the decision tree model are searched and finally determined within their respective value ranges through an optimization algorithm.

6. The method for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure according to claim 1, characterized in that: The standardization of the significant feature subset is input into the prediction model, and the sound insulation characteristics are predicted and optimized in combination with the constraint conditions to obtain an optimization scheme, including: The significant feature subset is standardized and then input into the prediction model, which uses the CART algorithm to predict the sound insulation characteristics, and determines the optimization scheme of the wooden frame support of the high-speed train floor structure in combination with the constraint conditions and the prediction results.

7. The method for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure according to claim 6 is characterized in that: The constraints include the overall thickness and overall density of the floor structure.

8. Sound insulation prediction and optimization device for wooden frame support of high-speed train floor structure, characterized in that: include: A factor acquisition unit is used to obtain the main influencing factors of the floor to be tested; A weighted sound insulation value determination unit, used to determine the weighted sound insulation value of the floor to be tested; A significant feature subset calculation unit, used for calculating a significant feature subset based on the mRMR feature screening method according to the main influencing factors and the weighted sound insulation; A prediction optimization unit, used for inputting the standardized significant feature subset into a prediction model to predict and optimize the sound insulation characteristics in combination with constraint conditions to obtain an optimization solution; wherein the prediction model is obtained by training a decision tree model using a number of standardized significant feature subsets as sample sets; An output unit is used to output the optimization solution.

9. The device for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure according to claim 8, characterized in that: The weighted sound insulation determination unit comprises: The sound pressure level acquisition subunit is used to acquire the relevant sound source and receiving sound pressure levels of the floor to be tested obtained by performing sound pressure level tests in the sound source room and the receiving room; Sound insulation determination subunit, used to adopt Calculate the sound insulation of the floor to be tested; where L1 is the average sound pressure level in the source room, L2 is the average sound pressure level in the receiving room, S1 is the surface area of ​​the floor to be tested, A is the sound absorption coefficient of the receiving room, , V is the volume of the receiving room, T is the reverberation time of the receiving room; The weighted sound insulation calculation subunit is used to calculate the weighted sound insulation according to the specified standard by combining the sound insulation frequency curve and the sound insulation of the floor to be tested.

10. The device for predicting and optimizing sound insulation of wooden frame support of high-speed train floor structure according to claim 8, characterized in that: The salient feature subset calculation unit comprises: A preprocessing subunit is used to analyze and preprocess the main influencing factors to obtain original characteristic parameters; wherein the original characteristic parameters include basic material property parameters, acoustic parameters and structural layout properties, the basic material property parameters include density, elastic modulus, Poisson's ratio and shear modulus, the acoustic parameters include weighted sound insulation of a single vibration damping pad and a single wooden frame; the structural layout properties include the weight, length, width, thickness, number, contact area of ​​the vibration damping pad and the wooden frame, the number of wooden frames and the number of vibration damping pads on the wooden frames; A calculation subunit, used for calculating the corresponding maxφ(D, R) value of the original characteristic parameter based on the mRMR characteristic screening method to obtain a calculation result; The screening subunit is used to screen the original feature parameters according to the calculation results to obtain a significant feature subset.

Images