A parameter design method of an acoustic superstructure unit

Abstract

This invention discloses a parameter design method for acoustic superstructure units. The parameter design method includes the following steps: constructing a theoretical and numerical simulation model of thermo-solid-acoustic multi-field coupling; obtaining a set of design parameters for several acoustic superstructure units as input features of the theoretical and numerical simulation model, calculating the sound absorption coefficients corresponding to the design parameter sets, and constructing a training dataset based on the several design parameter sets and sound absorption coefficients; training a pre-constructed forward acoustic performance surrogate model based on Res-MLP using the training dataset to obtain a fully trained forward acoustic performance surrogate model; and performing inverse parameter optimization of the acoustic superstructure units based on the fully trained forward acoustic performance surrogate model, combined with the global heuristic search capability of a genetic algorithm, to obtain the optimal design parameters. Compared with existing technologies, this parameter design method significantly reduces the computational cost of high-dimensional nonlinear design spaces.

Description

Technical Field

[0001] This invention relates to the field of noise control technology, and more specifically to a parameter design method for an acoustic superstructure unit. Background Technology

[0002] With the rapid development of aerospace equipment and high-speed aircraft, low-frequency broadband noise control in complex service environments has become a core bottleneck for improving equipment performance and meeting environmental noise standards. According to the classical acoustic quality law, for traditional porous sound-absorbing materials to achieve effective low-frequency sound absorption, their thickness must be comparable to the target sound wave wavelength. This size requirement fundamentally contradicts the stringent space and weight constraints in aerospace and other fields, making it unsuitable for lightweight integrated scenarios such as aircraft cabins and engine nacelles. The emergence of acoustic metastructures provides a new paradigm for breaking through the limitations of traditional acoustic theory. Through the design of subwavelength-scale local resonant structures, extraordinary sound wave modulation capabilities can be achieved, enabling efficient low-frequency sound absorption at subwavelength thicknesses.

[0003] However, most existing passive acoustic superstructures have inherent defects: their acoustic response characteristics are fixed after the design and manufacturing are completed, their operating bandwidth is locked, and they cannot adapt to the dynamic shift of the noise spectrum caused by changes in the flight state of the aircraft (such as altitude and Mach number) and the operating conditions of the equipment, making it difficult to meet the adaptive noise control requirements in complex variable operating conditions.

[0004] Traditional acoustic design methods, such as the transfer matrix method and the finite element method, suffer from high computational costs, long solution cycles, and a tendency to get trapped in local optima when dealing with such high-dimensional design spaces. They cannot achieve fast, on-demand reverse design, which seriously restricts the engineering application of acoustic metastructures. Summary of the Invention

[0005] The purpose of this invention is to overcome the above-mentioned technical deficiencies and propose a parameter design method for acoustic superstructure units. This method solves the technical problems of existing acoustic superstructures having fixed operating bandwidths, which cannot adapt to the adaptive control requirements of dynamic noise spectrum in aerospace temperature-varying environments, and the highly nonlinear mapping relationship between the geometric parameters and acoustic performance of acoustic superstructures under thermo-solid-acoustic multi-field coupling, resulting in extremely low computational efficiency, high difficulty in reverse design, and easy entrapment in local optima by traditional numerical methods.

[0006] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a parameter design method for an acoustic superstructure unit, the acoustic superstructure unit comprising an inner tube, a front micro-perforated plate, a rear micro-perforated plate, a piston, a first elastic element, a second elastic element, a sound-absorbing layer, and a shell. The front and rear micro-perforated plates are fixedly connected to both ends of the inner tube, the piston is slidably disposed within the inner tube, the first elastic element connects the piston and the front micro-perforated plate, the second elastic element connects the piston and the rear micro-perforated plate, the elastic modulus of the first elastic element changes with temperature, the elastic modulus of the second elastic element remains constant during temperature changes, the sound-absorbing layer covers the outer surface of the inner tube, and the shell covers the sound-absorbing layer. The method is characterized by comprising the following steps: Construct a theoretical and numerical simulation model of thermo-solid-acoustic multi-field coupling; A set of design parameters for several acoustic superstructure units is obtained as input features of the theoretical and numerical simulation model. After calculating the sound absorption coefficient corresponding to the set of design parameters, a training dataset is constructed based on the set of design parameters and the sound absorption coefficient. The training dataset is used to train the pre-built Res-MLP-based forward acoustic performance proxy model to obtain a fully trained forward acoustic performance proxy model. Based on the well-trained forward acoustic performance proxy model, and combined with the global heuristic search capability of the genetic algorithm, the acoustic superstructure unit is subjected to inverse parameter optimization to obtain the optimal design parameters of the acoustic superstructure unit.

[0007] In some embodiments, the construction of the theoretical and numerical simulation model for thermo-solid-acoustic multi-field coupling includes: The acoustic theoretical model of the acoustic superstructure unit is established by using the transfer matrix method. The viscous heat dissipation characteristics of the sound-absorbing layer are characterized by the Johnson-Champoux-Allard equivalent fluid model. The phase change mechanical behavior of the first elastic element is characterized by the one-dimensional Brinson constitutive model. The theoretical mapping relationship between temperature and piston displacement is established for the selection of initial design parameters and boundary determination. A two-dimensional axisymmetric finite element multi-field coupled simulation model is established. Based on the established theoretical mapping relationship between temperature and piston displacement, an acoustic-solid-thermal coupled computational domain is constructed. The Johnson-Champoux-Allard equivalent fluid model is set for the sound-absorbing layer, and boundary layer meshes are set at the interface between air and the piston and the interface between air and the inner wall of the inner tube.

[0008] In some embodiments, the design parameters include the outer diameter of the housing, the outer diameter of the inner tube, the wall thickness of the housing, the flow resistance scaling factor of the sound-absorbing layer, the distance between the piston and the rear micro-perforated plate, and the ambient temperature.

[0009] In some embodiments, obtaining the design parameter set of the plurality of acoustic superstructure units includes: The design parameters are uniformly sampled using the Latin hypercube sampling method that minimizes central deviation. Based on pre-defined hard geometric compatibility constraints, with the goal of ensuring the geometric manufacturability of all sampled samples, multiple sets of design parameters that are geometrically valid and physically feasible are generated.

[0010] In some embodiments, the forward acoustic performance proxy model sequentially includes an input layer, several cascaded residual blocks, and an output layer. The input layer is used to receive the input features. Each residual block includes two fully connected layers, a batch normalization layer, a ReLU activation function, and a Dropout layer. The output layer uses a Sigmoid activation function to output the predicted sound absorption coefficient value at the corresponding frequency point within a preset range.

[0011] In some embodiments, when the parameter design objective is to maximize sound absorption performance, the inverse parameter optimization of the acoustic superstructure unit based on the fully trained forward acoustic performance surrogate model, combined with the global heuristic search capability of the genetic algorithm, to obtain the optimal design parameters of the acoustic superstructure unit, includes: Based on the preset sound absorption frequency band [f1, f2], an initial population is randomly generated within the preset design parameter space; each individual in the initial population is input into the fully trained forward acoustic performance proxy model to calculate the predicted sound absorption coefficient values ​​of each individual at multiple frequency points within the target sound absorption frequency band; the single-objective evaluation function is:

[0012] Where, α i The sound absorption coefficient at the i-th frequency point predicted by the forward acoustic performance proxy model is used as the convergence criterion. The genetic algorithm is used to perform a global heuristic search on the initial population to iteratively obtain the optimal design parameter set of the acoustic superstructure unit. The optimal design parameter set obtained by optimization is input into the two-dimensional axisymmetric finite element multi-field coupled simulation model for high-fidelity independent verification. After confirming that the consistency error between the measured sound absorption curve and the design target is within a preset range, the parameter design of the acoustic superstructure unit is completed.

[0013] In some embodiments, when the parameter design objectives are to ensure sound absorption performance, lightweight structure, and compact size, the inverse parameter optimization of the acoustic superstructure unit is performed based on the fully trained forward acoustic performance surrogate model and combined with the global heuristic search capability of the genetic algorithm to obtain the optimal design parameters of the acoustic superstructure unit, including: Based on the preset sound absorption frequency band [f1, f2], an initial population is randomly generated within the preset design parameter space. This initial population contains multiple sets of individual design parameters for acoustic superstructure units. The initial population is then input into the fully trained forward acoustic performance proxy model to predict the sound absorption coefficient spectrum of each individual design parameter within the target sound absorption frequency band. Based on the sound absorption coefficient spectrum, the comprehensive fitness value of each individual design parameter is calculated using a multi-objective evaluation function, which is:

[0014] in, ω SAC , ω mass , ω thick The weighting coefficients are used as follows: maximizing the multi-objective evaluation function value serves as the convergence criterion for the genetic algorithm iteration, quantitatively characterizing the overall performance of the acoustic superstructure unit within the preset sound absorption frequency band; and using the sound absorption coefficient spectrum, the average sound absorption coefficient within the target frequency band is calculated. f SAC And combined with standardized structural quality evaluation items f mass and radial thickness evaluation item f thick A weighted summation is performed; the parameter optimization process uses a genetic algorithm to perform a global heuristic search on the initial population, and iteratively generates the optimal design parameter set with the goal of maximizing the multi-objective evaluation function value; the optimal design parameter set obtained by optimization is input into the two-dimensional axisymmetric finite element multi-field coupled simulation model for high-fidelity independent verification, and after confirming that the consistency error between the measured sound absorption curve and the design target is within a preset range, the parameter design of the acoustic superstructure unit is completed.

[0015] Secondly, the present invention also provides a parameter design device for acoustic superstructure units, comprising: The multi-field coupling modeling module constructs theoretical and numerical simulation models of thermo-solid-acoustic multi-field coupling; The training data construction module obtains several design parameter sets of the acoustic superstructure units as input features of the theoretical and numerical simulation models, calculates the sound absorption coefficients corresponding to the design parameter sets, and constructs a training dataset based on the several design parameter sets and the sound absorption coefficients. The proxy model training module uses the training dataset to train the pre-built Res-MLP-based forward acoustic performance proxy model to obtain a fully trained forward acoustic performance proxy model. The reverse optimization verification module, based on the fully trained forward acoustic performance proxy model and combined with the global heuristic search capability of the genetic algorithm, performs reverse parameter optimization on the acoustic superstructure unit to obtain the optimal design parameters of the acoustic superstructure unit.

[0016] Thirdly, the present invention also provides an electronic device, characterized in that the memory is used to store a program; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the above-described method for designing the parameters of the acoustic superstructure unit.

[0017] Fourthly, the present invention also provides a readable storage medium for storing a computer-readable program or instructions, which, when executed by a processor, can implement the steps in the above-described method for parameter design of acoustic superstructure units.

[0018] Fifthly, the present invention also provides an acoustic system, characterized in that it comprises: Mounting substrate; An acoustic superstructure unit, wherein the acoustic superstructure unit is designed using the above-mentioned parameter design method for acoustic superstructure units; The acoustic superstructure unit is acoustically coupled to the mounting substrate at one end where the rear micro-perforated plate is located, and multiple acoustic superstructure units are arranged in an array, wherein the spacing between two adjacent acoustic superstructure units is not exactly the same in at least one direction.

[0019] In some embodiments, the mounting base includes planar portions and / or curved portions; the acoustic superstructure unit is a cylindrical structure, wherein the axis is perpendicular to the portion of the mounting base to which it is connected.

[0020] In some embodiments, the distance between the central axes of two adjacent acoustic superstructure units is 1 / 4 to 1 / 2 of the center wavelength corresponding to the upper limit frequency of the target noise reduction band; the array has an initial resonant frequency with a gradient distribution in each acoustic superstructure unit under the same ambient temperature in at least one direction.

[0021] Compared with the prior art, the present invention has the following beneficial effects: The parameter design method provided by this invention proposes an intelligent reverse design framework based on a forward acoustic performance proxy model and a genetic algorithm using Res-MLP. This reduces the single iteration cycle of acoustic reverse design from several hours to 25ms, significantly reducing the computational cost of high-dimensional nonlinear design space. It solves the problems of high computational cost, easy getting trapped in local optima, and inability to meet the needs of multi-constraint engineering design in traditional design methods. The acoustic system provided by this invention utilizes the principle of temperature-induced phase change to convert changes in ambient temperature into piston displacement, achieving passive adaptive acoustic control without external power supply or additional control system. It has a simple structure, high reliability, and is perfectly suited for extreme and energy-constrained service environments such as aerospace. Attached Figure Description

[0022] Figure 1 It is an exploded diagram of an acoustic superstructure unit; Figure 2 This is a cross-sectional view of an acoustic superstructure unit; Figure 3 This is a flowchart of a parameter design method for an acoustic superstructure unit provided in an embodiment of the present invention; Figure 4 This is a logical framework diagram of the parametric design method; Figure 5 This is the sound absorption coefficient curve for bidirectional incidence; Figure 6 This is a schematic diagram of the Res-MLP network architecture; Figure 7 A schematic diagram of a parametric design apparatus for an acoustic superstructure unit according to another embodiment of the present disclosure is shown. Figure 8 A schematic diagram of an electronic device according to another embodiment of the present disclosure is shown.

[0023] Figure 9 This is a schematic diagram of the structure of an acoustic system provided in an embodiment of the present invention; Figure 10 This is a structural diagram when the mounting base is a plane; Figure 11 This is a structural diagram when the mounting base is a curved surface. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0025] To address the challenges of passive acoustic superstructures having a fixed operating bandwidth, which cannot adapt to the adaptive control requirements of dynamic noise spectrum in aerospace temperature environments, and the highly nonlinear mapping relationship between the geometric parameters and acoustic performance of acoustic superstructures under thermo-solid-acoustic multi-field coupling, which leads to extremely low computational efficiency, high reverse design difficulty, and susceptibility to local optima using traditional numerical methods, this invention provides a parameter design method for acoustic superstructure units. This method enables passive adaptive acoustic control without external power supply or additional control system. The parametric design method for this acoustic superstructure element is used to design acoustic superstructure elements and determine their core design parameters. Please refer to [link / reference]. Figure 1 and Figure 2 , Figure 1 It is an exploded diagram of an acoustic superstructure unit; Figure 2 This is a cross-sectional view of an acoustic superstructure unit.

[0026] The acoustic superstructure unit includes an inner tube 11, a front micro-perforated plate 12, a rear micro-perforated plate 13, a piston 14, a first elastic element 15, a second elastic element 16, a sound-absorbing layer 17, and an outer shell 18.

[0027] The front micro-perforated plate 12 and the rear micro-perforated plate 13 are fixedly connected to both ends of the inner tube 11, respectively. The piston 14 is disposed inside the inner tube 11 and is slidably connected to the inner tube 11 in a sealed manner. A front resonant cavity is formed between the piston 14 and the front micro-perforated plate 12, and a rear resonant cavity is formed between the piston 14 and the rear micro-perforated plate 13. A first elastic element 15 is disposed in the front resonant cavity and connects the piston 14 and the front micro-perforated plate 12, and a second elastic element 16 is disposed in the rear resonant cavity and connects the piston 14 and the rear micro-perforated plate 13. The elastic modulus of the first elastic element 15 changes with temperature, while the elastic modulus of the second elastic element 16 remains constant when the temperature changes. A sound-absorbing layer 17 covers the outer surface of the inner tube 11, and an outer shell 18 covers the sound-absorbing layer 17.

[0028] The piston 14 divides the inner cavity of the inner tube 11 into two independent cavities: a front resonant cavity and a rear resonant cavity. The equivalent depths of the front and rear resonant cavities dynamically change with temperature, thus forming continuously tunable Helmholtz resonators whose resonant frequencies satisfy the core physical relationship. (V is the cavity volume). By changing the cavity volume through the displacement of piston 14, the sound absorption frequency can be continuously adjusted.

[0029] Based on the above embodiments, the inner tube 11 is made of a material with good thermal conductivity to enable the first elastic element 15 to respond quickly to temperature changes. The inner tube 11 can be made of a high thermal conductivity aluminum alloy, such as 6061-T6 aluminum alloy tubing. The piston 14 can be a rubber piston to obtain good sliding sealing performance.

[0030] The first elastic element 15 can be a nickel-titanium-based shape memory alloy spring, i.e., an SMA spring. The austenite initiation temperature of the SMA spring is 30°C, and the austenite termination temperature is 50°C. The second elastic element 16 is an offset spring, providing a stable offset force. The first elastic element 15 and the second elastic element 16 are symmetrically arranged on both sides of the piston 14 to form a force balance, realizing precise and continuous control of the axial displacement of the piston 14.

[0031] When the ambient temperature varies within the range of 30℃ to 50℃, the SMA spring undergoes a reversible martensitic-austenitic phase transformation. Its elastic modulus and axial restoring force continuously increase with rising temperature, overcoming the preload of the bias spring and pulling the piston 14 to produce axial displacement. When the temperature decreases, the elastic modulus and restoring force of the SMA spring decrease, and the bias spring pushes the piston 14 to return to its original position. The axial displacement of the piston 14 dynamically changes the volume of the front and rear resonant cavities: as the temperature increases, the depth of the front resonant cavity decreases, and the depth of the rear resonant cavity increases; as the temperature decreases, the depth of the front resonant cavity increases, and the depth of the rear resonant cavity decreases. This continuous change in cavity volume causes a continuous shift in the resonant frequency of the Helmholtz resonator, ultimately achieving dynamic control of the absorption peak to adapt to the target noise spectrum at different temperatures.

[0032] Based on the above embodiments, the sound-absorbing layer 17 adopts an axial gradient laminated structure of "loose-dense-loose", consisting of a melamine foam layer 171, a glass fiber mat layer 172, and another melamine foam layer 171 from the outside to the inside. The core material parameters are a melamine foam porosity of 0.98 and a flow resistance of 15000 Pa. s / m²; glass fiber mat porosity 0.92, flow resistance 80000Pa s / m². By adjusting the flow resistance scaling factor (0.5~2.0) of the "loose layer", acoustic impedance matching over a wide frequency range is achieved, maximizing the viscous dissipation and heat dissipation of wide frequency acoustic energy, forming a wide frequency sound absorption complement with the internal Helmholtz resonator, and solving the problem of narrow sound absorption bandwidth of a single resonant structure.

[0033] Based on the above embodiments, the outer shell 18 is made of carbon fiber reinforced composite (CFRP) tubing, which is the main load-bearing body of the superstructure, and at the same time provides rigid mounting boundaries and acoustic rigid boundaries for the internal acoustic structure.

[0034] The entire acoustic superstructure unit uses the outer shell 18 to bear the main structural load, the sound-absorbing layer 17 to achieve mid-to-high frequency broadband sound absorption, and the internal Helmholtz resonator to achieve low-frequency adjustable sound absorption. The three work together to achieve the integrated structure and function of load bearing, broadband sound absorption and adaptive frequency tuning.

[0035] Please see Figure 3 and Figure 4 , Figure 3This is a flowchart of a parameter design method for an acoustic superstructure unit provided in an embodiment of the present invention; Figure 4 This is a logical framework diagram of the parametric design method.

[0036] This parametric design method is used to design the aforementioned acoustic superstructure unit, and includes the following steps: S101. Construct a theoretical and numerical simulation model for thermo-solid-acoustic multi-field coupling; S102. Obtain a set of design parameters for several acoustic superstructure units as input features of the theoretical and numerical simulation model. Calculate the sound absorption coefficient corresponding to the set of design parameters and construct a training dataset based on the set of design parameters and the sound absorption coefficient. S103. The pre-built forward acoustic performance proxy model based on Res-MLP is trained using the training dataset to obtain a fully trained forward acoustic performance proxy model. S104. Based on the fully trained forward acoustic performance proxy model, and combined with the global heuristic search capability of the genetic algorithm, perform inverse parameter optimization on the acoustic superstructure unit to obtain the optimal design parameters of the acoustic superstructure unit.

[0037] In some embodiments, the method for constructing the theoretical and numerical simulation model of the thermo-solid-acoustic multi-field coupling in step S101 includes: establishing an acoustic theoretical model of the acoustic superstructure unit using the transfer matrix method; characterizing the viscous heat dissipation characteristics of the porous sound-absorbing layer 17 using the Johnson-Champoux-Allard (JCA) equivalent fluid model; and characterizing the phase change mechanical behavior of the first elastic element 15, i.e., the SMA spring, using a one-dimensional Brinson constitutive model, and establishing a theoretical mapping relationship between temperature, piston displacement, resonant frequency, and sound absorption coefficient for the selection of initial design parameters and boundary determination.

[0038] A two-dimensional axisymmetric finite element multi-field coupled simulation model was established, and an acoustic-solid-thermal coupled computational domain was constructed using the two-dimensional axisymmetric finite element multi-field coupled simulation model in COMSOL Multiphysics software. A JCA equivalent fluid model was set for the sound-absorbing layer 17, and boundary layer meshes were set at the interface between air and piston 14 and the interface between air and the inner wall of inner tube 11 to accurately analyze the subtle acoustic field response within the viscous-thermal boundary layer.

[0039] In step S102, the selected design parameters are the core design parameters that play a leading role in the mechanical properties, thermal adaptive frequency modulation characteristics, and broadband sound absorption performance of the acoustic superstructure unit 1. In some embodiments, these include the outer diameter of the outer shell 18, the outer diameter of the inner tube 11, the wall thickness of the outer shell 18, the flow resistance scaling factor of the sound absorption layer 17, the distance between the piston 14 and the rear micro-perforated plate 13, and the ambient temperature.

[0040] Based on the above embodiments, the design parameters are as follows: the outer diameter of the outer shell 18 is 40-60 mm, the outer diameter of the inner tube 11 is 20-35 mm, the wall thickness of the outer shell 18 is 1.5-4.0 mm, the sound-absorbing layer 17 has a flow resistance scaling factor of 0.5-2.0 for its loose portion, the distance between the piston 14 and the rear micro-perforated plate 13 is 10-50 mm, and the ambient temperature is 30℃-50℃.

[0041] Based on the above embodiments, in step S102, the method for screening design parameter sets includes using a Latin hypercube sampling method with minimum central deviation to uniformly sample the design parameters; simultaneously, introducing hard geometric compatibility constraints to ensure the geometric manufacturability of all sampled samples and avoid "ill-conditioned samples" with structural interference, ultimately generating multiple sets of geometrically valid and physically feasible samples as input features. In this embodiment, 8000 sets of design parameter sets are generated.

[0042] Based on the above embodiments, step S102, the method for constructing the training dataset includes inputting the aforementioned 8000 sets of samples as input features into the theoretical and numerical simulation model, and calculating the sound absorption coefficient curves for a predetermined frequency range and frequency interval, such as... Figure 5 As shown. In this embodiment, the frequency range is 100–3000 Hz, with a frequency interval of 10 Hz. The sound absorption coefficients corresponding to each design parameter group are obtained, which in this embodiment are 291-dimensional sound absorption coefficient output vectors. The design parameter groups and the sound absorption coefficients are used to construct a training dataset. The dataset is then randomly divided into a training set (6400 sets), a validation set (800 sets), and a test set (800 sets) in an 8:1:1 ratio.

[0043] Based on the above embodiments, the method for calculating the dataset further includes performing Z-score standardization on all input features to eliminate the differences in the dimensions of different parameters, alleviate gradient instability, and improve the convergence speed and stability of model training.

[0044] Regarding step S103, in some embodiments, such as Figure 6 As shown, Figure 6This is a schematic diagram of the Res-MLP network architecture. The Res-MLP network architecture consists of an input layer, several cascaded residual blocks, and an output layer. The input layer receives the input features, namely the aforementioned 6-dimensional standardized design parameters. In this embodiment, there are 8 cascaded residual blocks. Each residual block includes 2 fully connected layers, a batch normalization (BN) layer, a ReLU activation function, and a dropout layer, and uses skip connections to achieve identity mapping. The width of the hidden layer neurons in a single residual block is set to 1024, which effectively alleviates the gradient vanishing problem in deep network training and significantly improves the ability to capture strongly nonlinear acoustic features such as formant shift and multi-peak coupling. The output layer contains 291 neurons, uses the Sigmoid activation function, and outputs the predicted absorption coefficient values ​​corresponding to 291 frequency points in the range of 100Hz to 3000Hz, to ensure that the output values ​​are strictly constrained within the physically valid range of [0,1].

[0045] Based on the above embodiments, the method for training a Res-MLP-based forward acoustic performance proxy model using a dataset is as follows: Mean Squared Error (MSE) is used as the loss function; the optimizer employs the Adaptive Moment Estimation (Adam) optimizer; and an early stopping mechanism is introduced during training, where the coefficient of determination R between the training set and the validation set is lowered. 2 Training is automatically terminated when the difference remains below 0.001 to prevent overfitting. The model's generalization ability is validated using an independent test set. The final trained forward acoustic performance proxy model achieves a coefficient of determination R0 between the predicted absorption coefficient and the simulated FEM value. 2 A value greater than 0.99 indicates the completion of a high-fidelity, millisecond-level response forward acoustic performance proxy model.

[0046] In some embodiments, when the parameter design objective is to maximize sound absorption performance, the inverse parameter optimization of the acoustic superstructure unit is performed based on the fully trained forward acoustic performance surrogate model, combined with the global heuristic search capability of the genetic algorithm, to obtain the optimal design parameters of the acoustic superstructure unit, including: First, based on the preset sound absorption frequency band [f1, f2], an initial population is randomly generated within the preset design parameter space. This initial population consists of multiple sets of individuals representing six-dimensional design parameters, including the outer diameter of the outer shell, the outer diameter of the inner tube, the wall thickness, the flow resistance scaling factor, the piston distance, and the ambient temperature. Second, each individual parameter in the initial population is input into the fully trained forward acoustic performance proxy model to predict the sound absorption coefficient at multiple frequency points within the preset sound absorption frequency band. Subsequently, the fitness value of each individual parameter is calculated using a single-objective evaluation function, which is:

[0047] Where, αi The absorption coefficient at the i-th frequency point predicted by the forward acoustic performance surrogate model is given. In this process, the forward acoustic performance surrogate model is used as a surrogate evaluation function, completely replacing the traditional time-consuming finite element simulation, achieving millisecond-level performance prediction and fitness evaluation. Finally, with maximizing the single-objective evaluation function value as the optimization objective, a global iterative search is performed using a genetic algorithm to obtain the optimal design parameter set. The optimal design parameter set is then input into the two-dimensional axisymmetric finite element multi-field coupled simulation model for independent verification. After confirming that the consistency error between the measured absorption curve and the design target is within a preset range, the optimal design parameter set is obtained.

[0048] In some embodiments, when the parameter design objectives are to ensure sound absorption performance, lightweight structure, and compact size, the method of performing inverse parameter optimization on the acoustic superstructure unit based on the fully trained forward acoustic performance surrogate model and combined with the global heuristic search capability of the genetic algorithm to obtain the optimal design parameters of the acoustic superstructure unit includes: Based on the preset sound absorption frequency bands [f1, f2], an initial population is randomly generated within the preset six-dimensional design parameter space. Next, the initial population is input into the fully trained forward acoustic performance surrogate model to predict the sound absorption coefficient spectrum of each individual design parameter within the target sound absorption frequency band. Subsequently, the comprehensive fitness value of each individual design parameter is calculated using a multi-objective evaluation function, which is:

[0049] Among them, obj w As a convergence criterion for the genetic algorithm iteration, it is used to quantitatively characterize the overall performance of the superstructure unit within a preset frequency band; f SAC The average sound absorption coefficient within the target sound absorption frequency band is the core optimization objective. f mass This is a standardized structural quality evaluation item; the smaller the structural quality, the higher the evaluation value. f thick The radial thickness is the standardized evaluation item; the smaller the radial dimension of the structure, the higher the evaluation value. ω SAC , ω mass , ω thick This is a weighting coefficient, which can be flexibly adjusted according to project requirements.

[0050] Finally, a genetic algorithm is used to perform multi-generation evolutionary search to generate the optimal design parameter set by maximizing the multi-objective evaluation function value. The optimal design parameter set is then input into the two-dimensional axisymmetric finite element multi-field coupled simulation model for independent verification. After confirming that its acoustic characteristics meet the design requirements, the optimal design parameter set of the superstructure unit is obtained.

[0051] Figure 7 A schematic diagram of a parametric design apparatus for an acoustic superstructure unit according to another embodiment of the present disclosure is shown. Figure 7 As shown, the parameter design device 400 for acoustic superstructure units includes a multi-field coupling modeling module 410, a training data construction module 420, a surrogate model training module 430, and a reverse optimization verification module 440.

[0052] The multi-field coupling modeling module 410 constructs theoretical and numerical simulation models of thermo-solid-acoustic multi-field coupling; The training data construction module 420 acquires several design parameter sets of the acoustic superstructure units as input features of the theoretical and numerical simulation models, calculates the sound absorption coefficients corresponding to the design parameter sets, and constructs a training dataset based on the several design parameter sets and the sound absorption coefficients. The proxy model training module 430 uses the training dataset to train the pre-built Res-MLP-based forward acoustic performance proxy model to obtain a fully trained forward acoustic performance proxy model. The reverse optimization verification module 440, based on the fully trained forward acoustic performance proxy model and combined with the global heuristic search capability of the genetic algorithm, performs reverse parameter optimization on the acoustic superstructure unit to obtain the optimal design parameters of the acoustic superstructure unit.

[0053] The parameter design device 400 for the aforementioned acoustic superstructure unit can be implemented as a computer program, which can be used in, for example... Figure 8 It runs on the electronic device shown.

[0054] Please see Figure 8 , Figure 8 This is a schematic block diagram of an electronic device provided in an embodiment of the present invention. The electronic device 500 can be a host computer or a server.

[0055] See Figure 8 The electronic device 500 includes a processor 502, a memory, and a network interface 505 connected via a device bus 501. The memory may include a storage medium 503 and internal memory 504.

[0056] The storage medium 503 may store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, it enables the processor 502 to execute the parameter design method for the acoustic superstructure unit.

[0057] The processor 502 provides computing and control capabilities to support the operation of the entire electronic device 500.

[0058] The internal memory 504 provides an environment for the operation of the computer program 5032 in the storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the parameter design method of the acoustic superstructure unit.

[0059] This network interface 505 is used for network communication, such as providing data transmission. Those skilled in the art will understand that... Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the electronic device 500 to which the present invention is applied. The specific electronic device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0060] The processor 502 is used to run a computer program 5032 stored in a memory to implement the parameter design method of the acoustic superstructure unit disclosed in the embodiments of the present invention.

[0061] Those skilled in the art will understand that Figure 8 The embodiments of the computer device shown do not constitute a limitation on the specific configuration of the computer device. In other embodiments, the computer device may include more or fewer components than shown, or combine certain components, or have different component arrangements. For example, in some embodiments, the computer device may include only memory and a processor. In such embodiments, the structure and function of the memory and processor are consistent with those described above, and will not be repeated here.

[0062] It should be understood that, in this embodiment of the invention, the processor 502 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0063] In another embodiment of the invention, a computer-readable storage medium is provided. This computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when executed by a processor, the computer program implements the parameter design method for the acoustic superstructure unit disclosed in the embodiments of the invention.

[0064] Those skilled in the art will readily understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. 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 foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0065] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Units with the same function may be grouped into one unit. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, or may be electrical, mechanical, or other forms of connection.

[0066] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0067] Furthermore, 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0068] If the integrated unit is implemented as 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, in essence, 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. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, a backend server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks.

[0069] Please see Figures 9 to 11 , Figure 9 This is a schematic diagram of the structure of an acoustic system provided in an embodiment of the present invention; Figure 10 This is a structural diagram when the mounting base is a plane; Figure 11 This is a structural diagram when the mounting base is a curved surface.

[0070] The acoustic system includes an acoustic superstructure unit 1 and a mounting base 2.

[0071] The acoustic superstructure unit 1 is acoustically coupled to the mounting substrate 2 at one end where the micro-perforated plate 13 is located. The connection method can be integrated co-curing molding, bolt fastening, high-temperature adhesive bonding, etc. To ensure pressure exchange between the rear resonant cavity and the external sound field, an acoustic flow gap is reserved at the contact interface between the rear micro-perforated plate 13 and the mounting substrate 2, or an acoustic through hole is provided at the corresponding position on the mounting substrate 2.

[0072] Furthermore, multiple acoustic superstructure units 1 are arranged in an array, and the spacing between adjacent acoustic superstructure units 1 is not exactly the same in at least one direction. That is, the array formed by the acoustic superstructure units is a non-uniform array, and in at least one direction, at least two sets of adjacent acoustic superstructure units 1 have different spacings. It is easy to understand that the array can also have multiple sets of adjacent acoustic superstructure units 1 with different spacings in multiple directions. Through the non-uniform spacing design, the spatial periodicity of the sound field can be broken, enhancing the acoustic robustness of the array during dynamic frequency modulation.

[0073] In some embodiments, the acoustic superstructure unit 1 is a cylindrical structure, with the inner tube 11, the sound-absorbing layer 17 and the outer shell 18 arranged coaxially and having a common central axis, which is perpendicular to the part of the mounting base 2 where the acoustic superstructure unit 1 is installed.

[0074] The mounting base 2 is the carrier on which the acoustic superstructure unit 1 is installed, provided by the equipment that actually installs the acoustic superstructure unit 1. Examples include aircraft bulkheads, engine nacelle inner walls, equipment bay partitions, and rail transit vehicle inner walls. Therefore, the mounting base 2 can be planar, curved, or a combination of both. However, regardless of whether it is planar or curved, the central axis of the structural unit 1 is perpendicular to the portion of the mounting base 2 to which it is connected. Figure 4 As shown, when the mounting base 2 is planar, each acoustic superstructure unit 1 is arranged in a row-column matrix, and the central axes of each acoustic superstructure unit 1 are parallel to each other. Figure 5 As shown, when the mounting base 2 is a curved surface, each acoustic superstructure unit 1 is conformally arranged along the surface of the mounting base 2.

[0075] In some embodiments, the distance between the central axes of two adjacent acoustic superstructure units 1 is 1 / 4 to 1 / 2 of the center wavelength corresponding to the upper limit frequency of the target noise reduction band. This ensures the coordinated control effect of the array and avoids the problem of acoustic interference cancellation. When the mounting substrate 2 is curved, the central axes of two adjacent acoustic superstructure units 1 are not parallel. In this case, the distance between the central axes at half the height of the acoustic superstructure unit 1 is used for calculation.

[0076] In some embodiments, the initial compression of the second elastic element 16 of the acoustic superstructure unit 1 in the array formed by the acoustic superstructure unit 1 is gradient-distributed in at least one direction. By adjusting the initial compression of the second elastic element 16, each acoustic superstructure unit 1 is at a different initial equilibrium position at the same ambient temperature, thereby obtaining a gradient-distributed initial resonant frequency. For example, along the left-to-right direction, the initial compression of the second elastic element 16 gradually increases or decreases. In other embodiments, the gradient distribution may also occur in multiple directions. For example, along the left-to-right and front-to-back directions, the initial compression of the second elastic element 16 varies with a gradient; or the acoustic superstructure unit 1 with a median initial frequency is located in the center of the array and diffuses outwards in a frequency gradient distribution.

[0077] By setting adjacent acoustic superstructure units 1 with different center-to-center spacings and coordinating them with different initial compression amounts resulting in initial frequency gradients, different acoustic superstructure units 1 can exhibit differentiated initial resonant frequencies and thermal frequency modulation response characteristics under the same ambient temperature. This allows the resonant absorption peaks of adjacent acoustic superstructure units 1 to overlap within the target frequency band, forming multiple complementary absorption peaks. This effectively fills the absorption bandwidth gaps of a single acoustic superstructure unit, achieving full-band coverage control of multi-peak, broadband noise. Simultaneously, when the ambient temperature dynamically changes, the absorption peaks of all acoustic superstructure units 1 in the array can synchronously and continuously shift, maintaining a gradient peak distribution. Throughout the entire operating temperature range of 30℃ to 50℃, it maintains an ultra-wide effective absorption bandwidth, perfectly adapting to non-stationary, broadband dynamic noise environments in scenarios such as aerospace and high-speed rail transportation.

[0078] Acoustic systems and their parameter design methods possess the following outstanding technical advantages and beneficial effects: 1. This invention achieves ultra-wide-range thermal adaptive frequency modulation, solving the problem of fixed bandwidth in traditional passive superstructures. Within an operating temperature range of 30℃ to 50℃, the resonant absorption peak of the acoustic superstructure can continuously shift from 620Hz to 1860Hz, achieving an ultra-wide continuous frequency modulation bandwidth of 1240Hz. Furthermore, throughout the entire frequency modulation range, the absorption coefficient at the resonant peak remains stably maintained at approximately 0.99, demonstrating excellent adaptive noise control capabilities and perfectly adapting to the dynamic noise spectrum of aerospace vehicles under different flight conditions.

[0079] 2. The invention achieves integrated structural and functional design, balancing high-efficiency sound absorption, structural load-bearing capacity, and lightweight construction. The superstructure of this invention uses a CFRP outer shell as the main load-bearing component, with a measured peak axial compression failure load reaching 116kN, significantly exceeding the specific strength of traditional aluminum alloy shells of the same size. Simultaneously, through the synergistic effect of the gradient sound-absorbing layer and the adjustable resonant core, under single-objective optimization, the average sound absorption coefficient in the target frequency band can reach 0.91–0.92, truly realizing multi-functional integration of "load-bearing capacity, sound absorption, and adaptive control," solving the industry problem of the inability to simultaneously achieve performance and strength in traditional sound-absorbing structures.

[0080] 3. This invention enables highly efficient and intelligent reverse engineering of multi-field coupled acoustic structures, significantly improving design efficiency. The forward acoustic performance proxy model based on Res-MLP constructed in this invention achieves high-fidelity prediction of sound absorption performance, with a determination coefficient R0. 2 The efficiency reaches 0.99; combined with the reverse design framework of genetic algorithm, a single reverse design iteration only takes 25ms, and the computational efficiency is more than 3 orders of magnitude higher than the traditional FEM-driven optimization method, which greatly reduces the design cost.

[0081] 4. It achieves multi-objective collaborative optimization, perfectly adapting to the stringent engineering constraints in the aerospace field. Through the weighted multi-objective optimization design of this invention, the total structural mass can be reduced by 14.23% and the radial total thickness by 33.3% while ensuring that the average sound absorption coefficient in the target frequency band is not less than 0.88, and excellent structural load-bearing capacity is maintained. This solves the problem that existing design methods cannot take into account multi-dimensional engineering constraints.

[0082] 5. Employing a passive thermal self-driven mechanism, this invention boasts extremely high engineering practicality and reliability. Its adaptive control requires no external power supply or additional sensing and control systems; it passively achieves continuous adjustment of the sound absorption frequency solely through changes in ambient temperature. With its simple structure, convenient assembly, and strong anti-interference capabilities, it can be directly adapted to complex service environments with variable temperatures and limited energy, such as aerospace, high-speed rail, and shipbuilding, possessing extremely high engineering promotion value.

[0083] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A parameter design method for an acoustic superstructure unit, the acoustic superstructure unit comprising an inner tube, a front micro-perforated plate, a rear micro-perforated plate, a piston, a first elastic element, a second elastic element, a sound-absorbing layer, and a shell, wherein the front and rear micro-perforated plates are fixedly connected to both ends of the inner tube, the piston is slidably disposed within the inner tube, the first elastic element connects the piston and the front micro-perforated plate, the second elastic element connects the piston and the rear micro-perforated plate, the elastic modulus of the first elastic element changes with temperature, the elastic modulus of the second elastic element remains constant during temperature changes, the sound-absorbing layer covers the outer surface of the inner tube, and the shell covers the sound-absorbing layer, characterized in that... Includes the following steps: Construct a theoretical and numerical simulation model of thermo-solid-acoustic multi-field coupling; A set of design parameters for several acoustic superstructure units is obtained as input features of the theoretical and numerical simulation model. After calculating the sound absorption coefficient corresponding to the set of design parameters, a training dataset is constructed based on the set of design parameters and the sound absorption coefficient. The training dataset is used to train the pre-built Res-MLP-based forward acoustic performance proxy model to obtain a fully trained forward acoustic performance proxy model. Based on the well-trained forward acoustic performance proxy model, and combined with the global heuristic search capability of the genetic algorithm, the acoustic superstructure unit is subjected to inverse parameter optimization to obtain the optimal design parameters of the acoustic superstructure unit.

2. The parameter design method for the acoustic superstructure unit according to claim 1, characterized in that, The theoretical and numerical simulation model for constructing a thermo-solid-acoustic multi-field coupling includes: The acoustic theoretical model of the acoustic superstructure unit is established by using the transfer matrix method. The viscous heat dissipation characteristics of the sound-absorbing layer are characterized by the Johnson-Champoux-Allard equivalent fluid model. The phase change mechanical behavior of the first elastic element is characterized by the one-dimensional Brinson constitutive model. The theoretical mapping relationship between temperature and piston displacement is established for the selection of initial design parameters and boundary determination. A two-dimensional axisymmetric finite element multi-field coupled simulation model is established. Based on the established theoretical mapping relationship between temperature and piston displacement, an acoustic-solid-thermal coupled computational domain is constructed. The Johnson-Champoux-Allard equivalent fluid model is set for the sound-absorbing layer, and boundary layer meshes are set at the interface between air and the piston and the interface between air and the inner wall of the inner tube.

3. The parameter design method for the acoustic superstructure unit according to claim 1, characterized in that, The design parameters include the outer diameter of the outer shell, the outer diameter of the inner tube, the wall thickness of the outer shell, the flow resistance scaling factor of the sound-absorbing layer, the distance between the piston and the rear micro-perforated plate, and the ambient temperature.

4. The parameter design method for the acoustic superstructure unit according to claim 1, characterized in that, The acquisition of design parameter sets for several acoustic superstructure units includes: The design parameters are uniformly sampled using the Latin hypercube sampling method that minimizes central deviation. Based on pre-defined hard geometric compatibility constraints, with the goal of ensuring the geometric manufacturability of all sampled samples, multiple sets of design parameters that are geometrically valid and physically feasible are generated.

5. The parameter design method for the acoustic superstructure unit according to claim 1, characterized in that, The forward acoustic performance proxy model sequentially includes an input layer, several cascaded residual blocks, and an output layer. The input layer is used to receive the input features. Each residual block includes two fully connected layers, a batch normalization layer, a ReLU activation function, and a Dropout layer. The output layer uses a Sigmoid activation function to output the predicted sound absorption coefficient values ​​at corresponding frequency points within a preset range.

6. The parameter design method for the acoustic superstructure unit according to claim 1, characterized in that, When the parameter design objective is to maximize sound absorption performance, the acoustic superstructure unit is subjected to inverse parameter optimization based on the fully trained forward acoustic performance surrogate model and combined with the global heuristic search capability of the genetic algorithm, in order to obtain the optimal design parameters of the acoustic superstructure unit, including: Based on the preset sound absorption frequency band [f1, f2], an initial population is randomly generated within the preset design parameter space; each individual in the initial population is input into the fully trained forward acoustic performance proxy model to calculate the predicted sound absorption coefficient values ​​of each individual at multiple frequency points within the target sound absorption frequency band; the single-objective evaluation function is: wherein, a i The sound absorption coefficient of the i-th frequency point predicted by the forward acoustic performance proxy model, taking the maximum of the single-objective evaluation function value as the convergence criterion, the initialization population is globally heuristically searched by the genetic algorithm, and an optimal design parameter group of the acoustic superstructure unit is iteratively obtained; the optimal design parameter group obtained by optimization is input into the two-dimensional axisymmetric finite element multi-field coupling simulation model for high-fidelity independent verification, and after confirming that the consistency error between the measured sound absorption curve and the design target is within the preset range, the parameter design of the acoustic superstructure unit is completed.

7. The parameter design method for the acoustic superstructure unit according to claim 1, characterized in that, Under the multiple constraints of ensuring sound absorption performance, lightweight structure, and compact size in parameter design, the acoustic superstructure unit is optimized by inverse parameter search based on the fully trained forward acoustic performance surrogate model and the global heuristic search capability of the genetic algorithm, in order to obtain the optimal design parameters of the acoustic superstructure unit, including: Based on the preset sound absorption frequency band [f1, f2], an initial population is randomly generated within the preset design parameter space. This initial population contains multiple sets of individual design parameters for acoustic superstructure units. The initial population is then input into the fully trained forward acoustic performance proxy model to predict the sound absorption coefficient spectrum of each individual design parameter within the target sound absorption frequency band. Based on the sound absorption coefficient spectrum, the comprehensive fitness value of each individual design parameter is calculated using a multi-objective evaluation function, which is: in, ω SAC , ω mass , ω thick The weighting coefficients are used as follows: maximizing the multi-objective evaluation function value serves as the convergence criterion for the genetic algorithm iteration, quantitatively characterizing the overall performance of the acoustic superstructure unit within the preset sound absorption frequency band; and using the sound absorption coefficient spectrum, the average sound absorption coefficient within the target frequency band is calculated. f SAC And combined with standardized structural quality evaluation items f mass and radial thickness evaluation item f thick A weighted summation is performed; the parameter optimization process uses a genetic algorithm to perform a global heuristic search on the initial population, and iteratively generates the optimal design parameter set with the goal of maximizing the multi-objective evaluation function value; the optimal design parameter set obtained by optimization is input into the two-dimensional axisymmetric finite element multi-field coupled simulation model for high-fidelity independent verification, and after confirming that the consistency error between the measured sound absorption curve and the design target is within a preset range, the parameter design of the acoustic superstructure unit is completed.

8. A parameter design device for an acoustic superstructure unit, characterized in that, include: The multi-field coupling modeling module constructs theoretical and numerical simulation models of thermo-solid-acoustic multi-field coupling; The training data construction module obtains several design parameter sets of the acoustic superstructure units as input features of the theoretical and numerical simulation models, calculates the sound absorption coefficients corresponding to the design parameter sets, and constructs a training dataset based on the several design parameter sets and the sound absorption coefficients. The proxy model training module uses the training dataset to train the pre-built Res-MLP-based forward acoustic performance proxy model to obtain a fully trained forward acoustic performance proxy model. The reverse optimization verification module, based on the fully trained forward acoustic performance proxy model and combined with the global heuristic search capability of the genetic algorithm, performs reverse parameter optimization on the acoustic superstructure unit to obtain the optimal design parameters of the acoustic superstructure unit.

9. An electronic device, characterized in that, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the parameter design method of the acoustic superstructure unit according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the parameter design method of the acoustic superstructure unit according to any one of claims 1 to 7.

11. An acoustic system, characterized in that, include: Mounting substrate; An acoustic superstructure unit, wherein the acoustic superstructure unit is designed using the parameter design method of the acoustic superstructure unit according to any one of claims 1 to 7; The acoustic superstructure unit is acoustically coupled to the mounting substrate at one end where the rear micro-perforated plate is located, and multiple acoustic superstructure units are arranged in an array, wherein the spacing between two adjacent acoustic superstructure units is not exactly the same in at least one direction.

12. The acoustic system according to claim 11, characterized in that, The mounting base includes a planar portion and / or a curved portion; the acoustic superstructure unit is a cylindrical structure, wherein the axis is perpendicular to the portion of the mounting base to which it is connected.

13. The acoustic system according to claim 12, characterized in that, The distance between the central axes of two adjacent acoustic superstructure units is 1 / 4 to 1 / 2 of the center wavelength corresponding to the upper limit frequency of the target noise reduction band; in at least one direction, each of the acoustic superstructure units has an initial resonant frequency with a gradient distribution under the same ambient temperature.

Images