One-dimensional acoustic diffuse reflection metasurface optimization arrangement method and device and storage medium

Abstract

The application discloses a one-dimensional acoustic diffuse reflection metasurface optimization arrangement method, which comprises the following steps: L1, initializing a metasurface unit and a corresponding unit arrangement, calculating the sound pressure distribution and obtaining the standard deviation of the scattered sound pressure distribution; L2, adopting MCMC annealing optimization to realize global search of the unit arrangement scheme, and obtaining the optimal arrangement scheme of the metasurface unit as the initial state of Q-Learning; L3, further optimizing the arrangement by using Q-Learning reinforcement learning, making the metasurface scattering of the optimal unit arrangement scheme more uniform, and obtaining the optimized metasurface arrangement; L4, calculating the scattered sound pressure of the optimized metasurface, and drawing the polar coordinate diagram of the optimized sound field distribution. The application can globally search in the initial optimization stage, quickly converges to a relatively optimal unit arrangement structure, and avoids falling into a local optimum; and the initial result is locally fine-tuned in the later optimization stage, so that the maximum uniformity of the diffuse reflection scattering energy in each direction is realized.

Description

Technical Field

[0001] This invention relates to the field of acoustic metamaterials technology, specifically to a method, device, and storage medium for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces that integrates MCMC and Q-Learning. Background Technology

[0002] Acoustic metasurfaces are a novel type of metamaterial capable of achieving noise control, stealth technology, and ultrasonic imaging. By designing intricate unit structures within the subwavelength scale, acoustic metasurfaces can achieve anomalous manipulation of sound waves, such as negative refraction, focusing, wave attenuation, and diffuse reflection. Diffuse reflection metasurfaces, in particular, have the advantage of uniformly dispersing incident sound energy in all directions, effectively reducing the impact of concentrated sound energy on noise control and echo suppression. However, achieving efficient diffuse reflection hinges on the arrangement and design of the metasurface unit structures.

[0003] In order to combine the various units in a reasonable way and thus make the scattered sound field uniformly distributed in space, existing technologies generally use traditional optimization methods, such as genetic algorithms, simulated annealing algorithms, and particle swarm optimization. Although these methods can explore the solution space of metasurface arrangement to a certain extent, due to the discrete and highly nonlinear characteristics of the arrangement problem itself, these methods are often prone to getting stuck in local optima. At the same time, a single global search strategy is difficult to take into account both global and local optimization needs, resulting in an unsatisfactory final diffuse reflection effect.

[0004] Furthermore, the Monte Carlo Markov Chain (MCMC) method, due to its strong global search capability, has been introduced into the field of metasurface optimization. It can quickly find a better initial permutation through random perturbation and acceptance criteria. However, the MCMC algorithm has certain shortcomings in local fine-tuning. On the other hand, although reinforcement learning (such as Q-Learning) can be used as an adaptive learning method in the existing technology, it can gradually learn the optimal policy through interaction with the environment and achieve local fine-tuning and fine-grained adjustment. However, reinforcement learning alone is easily affected by the large dimension of the state space and insufficient training samples, making it difficult to converge stably to the global optimum.

[0005] To address these technical problems, this application proposes a method for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces. Summary of the Invention

[0006] The main objective of this invention is to provide a method for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces. By first using the MCMC algorithm to perform a global search and obtain a preliminary, relatively optimal arrangement, and then using the Q-Learning algorithm to perform local fine-tuning of the preliminary arrangement, the standard deviation of the scattered sound field distribution is further reduced, thereby achieving maximum uniformity of diffuse reflection scattered energy in all directions, thus solving the technical problems mentioned in the background art.

[0007] The present invention solves the above-mentioned technical problems by adopting the following technical solutions:

[0008] One-dimensional acoustic diffuse reflection metasurface optimization arrangement method, including:

[0009] L1. Initialize the metasurface units and their corresponding unit arrangements, calculate their sound pressure distribution, and obtain the standard deviation of the scattered sound pressure distribution;

[0010] L2. MCMC annealing optimization is used to achieve a global search for the cell arrangement scheme, and the optimal arrangement scheme of the metasurface cells is obtained as the initial state of Q-Learning;

[0011] L3. Q-Learning reinforcement learning is used to further optimize the arrangement, making the metasurface scattering of the optimal unit arrangement more uniform, and obtaining the optimized metasurface arrangement.

[0012] L4. Calculate the scattered sound pressure of the optimized metasurface and plot the polar coordinate diagram of the optimized sound field distribution.

[0013] Preferably, the specific operation flow of the L1 step includes:

[0014] L11. Define N metasurface elements, and each element reflects light from K predefined reflection angles {θ1, θ2, ..., θ...}. K Select from};

[0015] L12. Randomly initialize the metasurface unit arrangement S0 and calculate its sound pressure distribution. Set the objective function J as the standard deviation of the scattered sound pressure distribution, and we have:

[0016]

[0017] Among them, P i Normalized sound pressure level for each angle, It is the average sound pressure at all angles, N represents the number of discrete sampling points for the scattering angle, and J represents the scattering uniformity error. The smaller the value, the more uniform the scattering.

[0018] Preferably, the specific operation process of the L2 step includes:

[0019] L21. Set the optimization hyperparameters: initial temperature T, annealing coefficient α, and maximum number of iterations MaxIter_MCMC;

[0020] L22. Using the Metropolis-Hastings algorithm, a unit i is randomly selected from 1 to N, and its reflection angle is randomly selected from K angles to modify, generating a new permutation S_new;

[0021] L23. After each iteration, the temperature is reduced by calculating T = α * T, and the optimal solution is recorded;

[0022] L24. Throughout the iteration process, continuously update the optimal permutation S_best until T approaches 0 or reaches the maximum number of iterations MaxIter_MCMC.

[0023] Preferably, the specific operation flow for generating the new permutation S_new in step L22 includes:

[0024] a1. Calculate the change in the objective function ΔJ using the following formula:

[0025] ΔJ=J new -J old

[0026] J new and J old These represent the variances of the metasurface scattering acoustic pressure before and after the change.

[0027] a2. Decide whether to accept S_new based on the Metropolis criterion:

[0028] If ΔE < 0, meaning the new structural arrangement S_new makes the scattering more uniform, then it is accepted directly; if ΔE ≥ 0, then the decision on whether to accept S_new is made by probability P = e - ΔJ / T, where T is the annealing parameter.

[0029] Preferably, the specific operation process of step L3 includes:

[0030] L31. Set up the Q-value table Q(S,A), initialized to zero, with learning rate μ, discount factor γ, exploration rate ε, and Q-Learning training rounds MaxIter_Q;

[0031] L32. Select the current state S_t. In each iteration round, for each unit in the current state S_t, the agent selects an action using an ε-greedy policy:

[0032] The reflection angle of a specified unit is randomly changed from its current value to another candidate angle with probability ε.

[0033] Select the action with the highest Q value in the corresponding state from the Q-table with probability 1-ε;

[0034] After the L33 selected action, the corresponding unit in the arrangement is adjusted to generate a new state S_t+1 and the corresponding optimal arrangement S_best.

[0035] Preferably, the specific operation process for generating S_best in the L33 step includes:

[0036] b1. Calculate and obtain the scattering uniformity J(S_t+1) of the current arrangement S_t+1, and then calculate the reward R_t. The calculation formula is:

[0037] R_t = J_best - J(S_t+1)

[0038] where J_best is the optimal metasurface scattering sound pressure;

[0039] If J(S_t+1) < J_best, then R_t is positive, that is, the uniformity of the new state is improved, encouraging optimization;

[0040] b2. Update the Q value using the Bellman update formula: Update the optimal solution S_best. If J(S_t+1) < J_best, then S_best = S_t+1, and continue to iterate using the following formula:

[0041] Q(S t , A t ) = Q(S t , A t ) + μ·(R t + γmax a' Q(S t+1 , a') - Q(S t , A t ))

[0042] where (S t , A t ) is a set of state-action pairs, Q(S t , A t ) is the current Q value of the state-action pair (S t , A t ); R t is the immediate reward of the current action; γ is the discount factor, and 0 ≤ γ ≤ 1, which is used to control the importance of future rewards, and μ is the learning rate, which is used to determine the update amplitude of the Q value; max a' Q(S t+1 , a') represents taking the maximum value of the Q values corresponding to all possible actions a' in the next state S_t+1; a' represents any one of all possible actions in the next state S_t+1; R t + γmax a' Q(S t+1,a') is the target value, R t +γmax a' Q(S t+1 ,a')-Q(S t A t The time-series difference error represents the deviation between the current estimate and the "target".

[0043] The maximum Q value obtained after the optimal action is selected as the updated Q value;

[0044] b3. Repeat the Q-value update operation until the predetermined number of rounds is reached or the change in the Q-table is less than the preset value. Then the training ends and the system outputs the final optimized hypersurface arrangement S_best.

[0045] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0046] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0047] As can be seen from the above technical solution, the present invention provides a method for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces. Compared with the prior art, the present invention has the following advantages:

[0048] 1. This invention uses a Markov chain Monte Carlo (MCMC) algorithm in the early stages of optimization to perform a global search, thereby quickly converging to a better unit arrangement structure and avoiding getting trapped in local optima. Furthermore, by introducing a Q-Learning reinforcement learning algorithm in the later stages of optimization, the initial results are locally fine-tuned to further reduce the standard deviation of the scattered sound field distribution and achieve maximum uniformity of diffuse reflection scattered energy in all directions.

[0049] 2. This invention calculates the reward by combining the scattering uniformity of the current arrangement during the iterative selection of the unit arrangement structure state using the Q-value table, and iteratively updates the Q-value. This enables stable convergence to the global optimal solution during reinforcement learning, facilitating the output of the final optimized metasurface arrangement.

[0050] It should be understood that the descriptions in this section are not intended to identify key or essential features of embodiments of the invention, nor are they intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Of course, implementing any product of the invention does not necessarily require achieving all of the advantages described above simultaneously. Attached Figure Description

[0051] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0052] Figure 1 This is a schematic diagram of the overall method flow of the present invention;

[0053] Figure 2 This is a schematic diagram of the initial arrangement of metasurface acoustic pressure distribution in this invention;

[0054] Figure 3 This is a schematic diagram of the metasurface acoustic pressure distribution after MCMC optimization according to the present invention;

[0055] Figure 4 This is a schematic diagram of the metasurface acoustic pressure distribution after MCMC and Q-Learning optimization according to the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] For details in the embodiments, please refer to Figures 1 to 4 .

[0058] like Figure 1 As shown. The one-dimensional acoustic diffuse reflection metasurface optimization arrangement method proposed in this embodiment of the invention includes the following steps:

[0059] Step 1: Start (Initialization):

[0060] (1) Define N metasurface units;

[0061] (2) Each unit receives K predefined reflection angles {θ1, θ2, ..., θ...} K Select from};

[0062] (3) Randomly initialize the metasurface unit arrangement S0;

[0063] (4) Calculate its sound pressure distribution and set the objective function J as the standard deviation of the scattered sound pressure distribution:

[0064]

[0065] Where: P iNormalized sound pressure level (dB) for each angle This represents the average sound pressure level across all angles, N represents the number of discrete sampling points for the scattering angle, and J represents the scattering uniformity error; a smaller J value indicates more uniform scattering. The optimization hyperparameters are set as follows: initial temperature T, annealing coefficient α, and maximum number of iterations MaxIter_MCMC.

[0066] At this time, the metasurface acoustic pressure distribution is as follows: Figure 2 As shown.

[0067] Step 2: MCMC Annealing Optimization (Global Search):

[0068] Its main objective is to explore different unit arrangement schemes and find a better solution S_best as the initial state for Q-Learning.

[0069] (5) Using the Metropolis-Hastings algorithm, randomly select a unit i (from 1 to N), modify its reflection angle (randomly selected from K angles), and generate a new permutation S_new.

[0070] (a) Calculate the change in the objective function ΔJ: ΔJ = J new -J old ;

[0071] J new and J old These represent the variances of the metasurface scattering sound pressure before and after the change.

[0072] (b) Decide whether to accept S_new based on the Metropolis criteria:

[0073] If ΔE < 0 (the new solution is better), that is, the new structural arrangement S_new makes the scattering more uniform, then it is accepted directly.

[0074] If ΔE≥0, then the decision on whether to accept S_new is based on the probability P=e-ΔJ / T (T is the annealing parameter).

[0075] (6) After each iteration, reduce the temperature T: T = α * T; record the optimal solution, and continuously update the optimal permutation S_best throughout the entire iteration process until T approaches 0 or reaches the maximum number of iterations MaxIter_MCMC.

[0076] At this time, the metasurface acoustic pressure distribution is as follows: Figure 3 As shown.

[0077] Step 3: Q-Learning Reinforcement Learning (Local Optimization):

[0078] Objective: Based on S_best, further optimize the arrangement using reinforcement learning to make the scattering more uniform.

[0079] (7) Set up the Q-value table Q(S,A) (initialized to zero), the learning rate μ (controlling the Q-value update speed), the discount factor γ (measuring the importance of future rewards), the exploration rate ε (used to control the balance between random exploration and greedy selection), and the number of Q-Learning training rounds MaxIter_Q.

[0080] (8) Select the current state S_t (initialized from the MCMC result S_best); in each iteration round, for each cell in the current state S_t, the agent uses the ε-greedy policy to select an action: with probability ε, randomly change the reflection angle of a certain cell from the current value to another candidate angle (the action set is all possible reflection angles (there are K angles here)). With probability 1-ε, select the action with the highest Q-value in the corresponding state in the Q-table, that is, select the action that maximizes the expected reward.

[0081] (9) After selecting the action, adjust the corresponding cell in the permutation to generate a new state S_t+1.

[0082] (a) Calculate the reward R_t: Calculate the scattering uniformity J(S_t+1) of the current permutation S_t+1; calculate the reward R_t = J_best - J(S_t+1). If J(S_t+1) < J_best, then R_t is positive, that is, the uniformity of the new state is improved (the standard deviation is reduced), encouraging optimization;

[0083] (b) Update the Q-value using the Bellman update formula: Update the optimal solution S_best. If J(S_t+1) < J_best, then S_best = S_t+1, and continue to iterate using the following formula: <00002​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​γ is the immediate reward for the current action; γ is the discount factor, where 0 ≤ γ ≤ 1, used to control the importance of future rewards; μ is the learning rate, used to determine the magnitude of Q-value updates; max a' Q(S t+1 ,a') means taking the maximum Q value for all possible actions a' in the next state S_t+1; a' means any one of all possible actions in the next state S_t+1; R t +γmax a' Q(S t+1 ,a') is the target value, R t +γmax a' Q(S t+1 ,a')-Q(S t A t The time-series difference error represents the deviation between the current estimate and the "target".

[0086] The maximum Q-value obtained after selecting the optimal action is used as the updated Q-value.

[0087] (c) After completing one Q-value update, proceed to the next step. Repeat this process until the predetermined number of rounds is reached or the Q-table change is very small, at which point training ends, and the system outputs the final optimized hypersurface arrangement S_best.

[0088] Step 4: Calculate the optimized scattered sound pressure:

[0089] (10) Output the optimal arrangement S_best and calculate the scattering characteristics of the optimized metasurface.

[0090] (11) Plot the polar coordinate diagram of the optimized metasurface acoustic field distribution, such as Figure 4 As shown.

[0091] In summary, the algorithm adopted in this application employs a two-stage optimization strategy:

[0092] The first stage: Using the Markov chain Monte Carlo (MCMC) method, based on the Metropolis-Hastings rule, a global search is performed on the initial random arrangement to quickly converge to a better unit arrangement structure and avoid getting trapped in local optima;

[0093] The second stage: Based on the MCMC results, Q-Learning reinforcement learning is used to further optimize the arrangement and perform local fine-tuning on the preliminary results in order to further reduce the standard deviation of the scattered sound field distribution and achieve maximum uniformity of diffuse reflection scattered energy in all directions.

[0094] In one specific embodiment, in order to optimize the 1D metasurface arrangement, the following parameters are substituted into the specific operation steps of the above embodiment:

[0095] Setting parameters:

[0096] The metasurface is set to contain N = 40 elements; the reflection angle is predefined, k = 5, which are 15°, 30°, 45°, 60°, and 80° respectively;

[0097] The initial temperature was set to T = 1.0, the annealing coefficient α = 0.95, and the number of MCMC iterations MaxIter_MCMC = 800.

[0098] Set the Q-Learning parameters as follows: learning rate α = 0.05, discount factor γ = 0.9, exploration rate ε = 0.3. The maximum number of Q-Learning iterations, MaxIter_Q = 2000.

[0099] The optimization process at this point is as follows: initially randomize S0, run MCMC sampling to obtain the preliminary optimized permutation S_init; run Q-Learning reinforcement learning to obtain the final optimized result S_best;

[0100] The final experimental results are as follows:

[0101] The standard deviation of the initial random permutation is 5.69, and the corresponding sound pressure distribution is as follows: Figure 2 As shown; the standard deviation of MCMC after optimization is 4.11, and its corresponding sound pressure distribution is as follows. Figure 3 As shown; after MCMC and 2000 training rounds, the optimized standard deviation is 2.44, and the corresponding sound pressure distribution is as follows. Figure 4 As shown, the uniformity is improved by 57.12% compared to the initial randomly arranged metasurface, and by 40.63% compared to the MCMC optimization.

[0102] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0103] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0104] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the one-dimensional acoustic diffuse reflection metasurface optimization arrangement methods described in the above embodiments.

[0105] It is understood that the system provided in the embodiments of the present invention corresponds to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0106] This application also provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, communication interface, and memory communicate with each other via the communication bus.

[0107] Memory, used to store computer programs;

[0108] When the processor executes the program stored in the memory, it implements the above-mentioned one-dimensional acoustic diffuse reflection metasurface optimization arrangement method.

[0109] The communication bus mentioned in the above-mentioned electronic devices can be a standard bus for interconnecting peripheral components or an extended industrial standard structure bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc.

[0110] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0111] The memory may include random access memory or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0112] The processors mentioned above can be general-purpose processors, including central processing units, network processors, etc.; they can also be digital signal processors, application-specific integrated circuits, field-programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0113] It should also be noted that electronic devices include terminal devices, which can also be called terminals, user equipment, mobile stations, mobile terminals, etc. Terminal devices can be mobile phones, smart TVs, wearable devices, tablets, computers with wireless transceiver capabilities, virtual reality terminal devices, augmented reality terminal devices, wireless terminals in industrial control, wireless terminals in autonomous driving, wireless terminals in remote surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, and so on. The embodiments of this application do not limit the specific technologies or device forms used in the terminal devices.

[0114] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium, or a semiconductor medium (e.g., solid-state drive), etc.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0116] Furthermore, it should be noted that if any directional indication (such as up, down, left, right, front, back, etc.) is involved in the embodiments of the present invention, the directional indication is only used to explain the relative positional relationship and movement of each component in a specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0117] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the meaning of "and / or" throughout the text includes three parallel solutions; for example, "A and / or B" includes solution A, solution B, or a solution where both A and B are satisfied simultaneously. Furthermore, in the embodiments of this invention, "multiple" refers to two or more. Moreover, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

Claims

1. A method for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces, characterized in that, include: L1. Initialize the metasurface units and their corresponding unit arrangements, calculate their sound pressure distribution, and obtain the standard deviation of the scattered sound pressure distribution; L2. MCMC annealing optimization is used to achieve a global search for the cell arrangement scheme, and the optimal arrangement scheme of the metasurface cells is obtained as the initial state of Q-Learning; L3. Q-Learning reinforcement learning is used to further optimize the arrangement, making the metasurface scattering of the optimal unit arrangement more uniform, and obtaining the optimized metasurface arrangement. L4. Calculate the scattered sound pressure of the optimized metasurface and plot the polar coordinate diagram of the optimized sound field distribution; The specific operation process of the L3 step includes: L31. Setting the Q-value table Its initialization is set to zero, and the learning rate is... Discount factor Exploration rate And the number of training rounds in Q-Learning, MaxIter_Q; L32. Select the current state S_t. In each iteration round, for each unit in the current state S_t, the agent selects an action using an ε-greedy policy: The reflection angle of a specified unit is randomly changed from its current value to another candidate angle with probability ε. With probability 1 ε selects the action with the highest Q value in the corresponding state from the Q table; L33. After selecting an action, adjust the corresponding unit in the arrangement to generate a new state S_t+1 and the corresponding optimal arrangement S_best.

2. The method for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces as described in claim 1, characterized in that, The specific operation procedure for the L1 step includes: L11. Settings There are K metasurface units, and each unit reflects light from K predefined angles. Select from; L12. Randomly initialize the metasurface unit arrangement S0 and calculate its sound pressure distribution, setting the objective function. As the standard deviation of the scattered sound pressure distribution, we have: in, Normalized sound pressure level for each angle, It is the average sound pressure level across all angles. To represent the number of discrete sampling points for the scattering angle, This represents the scattering uniformity error; the smaller the value, the more uniform the scattering.

3. The method for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces as described in claim 2, characterized in that, The specific operation process of the L2 step includes: L21. Set optimization hyperparameters, initial temperature Annealing coefficient Maximum number of iterations MaxIter_MCMC; L22. Using the Metropolis-Hastings algorithm, from 1 to... Randomly select a unit Randomly select from K angles and modify their reflection angles to generate a new permutation S_new; L23. After each iteration, through calculation Lower the temperature and record the optimal solution; L24. Throughout the entire iteration process, continuously update the optimal permutation S_best until... Approaching 0 or reaching the maximum number of iterations, MaxIter_MCMC.

4. The method for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces as described in claim 3, characterized in that, The specific operation process for generating the new permutation S_new in step L22 includes: a1. Calculate the change in the objective function The calculation formula is as follows: in and These represent the variances of the metasurface scattering acoustic pressure before and after the change. a2. Decide whether to accept S_new based on the Metropolis criterion: like If the new structural arrangement S_new makes the scattering more uniform, then it can be directly accepted; if Then through probability Decide whether to accept S_new, where These are the annealing parameters.

5. The method for optimizing the arrangement of one-dimensional acoustic diffuse reflection metasurfaces as described in claim 1, characterized in that, The specific operation process for generating S_best in step L33 includes: b1. Calculate the scattering uniformity of the current arrangement S_t+1. Then calculate the reward. The calculation formula is as follows: in, The optimal metasurface scattering acoustic pressure; like ,but A positive value indicates improved uniformity of the new state, encouraging optimization. b2. Update Q-value: Update the optimal solution S_best, if... ,but Continuing the iteration using the following formula, we have: in, For a set of state-action pairs, State-action pairs The current Q value; An immediate reward for the current action; As a discount factor, and , is used to control the importance of future rewards, and μ is the learning rate, which determines the magnitude of Q-value updates; Indicates the next state In the middle, for all possible actions The corresponding Q value is the maximum value; Indicates the next state Any one of the following possible actions; For the target value, It is the time-series difference error, which represents the deviation between the current estimate and the "target". The maximum Q value obtained after the optimal action is selected as the updated Q value; b3. Repeat the Q-value update operation until the predetermined number of rounds is reached or the change in the Q-table is less than the preset value. Then the training ends and the system outputs the final optimized hypersurface arrangement S_best.

6. A computer-readable storage medium, characterized in that, The device stores a computer program that, when executed by a processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 5.

7. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 5.

Images