Sound field prediction method and device based on RBF neural network, equipment and medium

Abstract

The invention relates to a sound field prediction method and device based on an RBF neural network, equipment and a medium, and belongs to the technical field of vibration noise control, and the sound field prediction method based on the RBF neural network comprises the steps: building a sound field structure model, carrying out the sound field vibration simulation of the sound field structure model, and obtaining the sound field vibration data under different working conditions; the built RBF neural network is trained based on the sound field vibration data, parameters of the RBF neural network are adjusted, and the RBF neural network comprises an input layer, a hidden layer and an output layer; the to-be-predicted sound field vibration data is predicted based on the completely trained RBF neural network, the total sound pressure level predicted value is obtained, and the sound field prediction precision and prediction efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration noise control, and in particular to a sound field prediction method, device, equipment and medium based on RBF neural network. Background Art

[0002] Acoustic field prediction is an important area in vibration and noise control research. It involves the accurate prediction of sound propagation and distribution characteristics. However, the complexity of the acoustic field often exceeds the scope of traditional prediction methods, especially when dealing with the correlation between near-field and far-field acoustic fields. These differences and dynamic changes make it extremely challenging to build an accurate prediction model.

[0003] The near-field sound field usually contains rich sound wave details, such as interference, diffraction and reflection, which make the distribution of sound waves extremely complex; while the far-field sound field is relatively simple, and sound waves mainly propagate in the form of spherical waves; however, the correlation between the two is crucial, because even slight changes in the near-field sound field may have a significant impact on the far-field sound field.

[0004] Traditional sound field prediction methods, such as physical models and numerical calculations, can simulate the propagation characteristics of the sound field to a certain extent, but these methods are not only computationally intensive and time-consuming, but also have limited prediction accuracy when dealing with complex sound field environments. Summary of the invention

[0005] In view of this, it is necessary to provide a sound field prediction method, device, equipment and medium based on RBF neural network to solve the technical problems of low accuracy and low efficiency of sound field prediction.

[0006] In order to solve the above problems, the present invention provides a sound field prediction method based on RBF neural network, comprising: Establishing an acoustic field structure model, performing acoustic field vibration simulation on the acoustic field structure model, and obtaining acoustic field vibration data under different working conditions; Training the constructed RBF neural network based on the sound field vibration data, and adjusting the parameters of the RBF neural network, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; The sound field vibration data to be predicted is predicted based on the fully trained RBF neural network to obtain a predicted value of the total sound pressure level.

[0007] In a possible implementation, establishing the sound field structure model includes: The acoustic field structure model is established with aluminum sheet as the acoustic field structure and air domain as the acoustic field domain.

[0008] In a possible implementation, performing sound field vibration simulation on the sound field structure model to obtain sound field vibration data under different working conditions includes: A frequency range and a frequency step are determined, and a point load is applied to the aluminum sheet based on the frequency range and the frequency step to obtain sound field vibration data under different working conditions, wherein the sound field vibration data includes vibration data and noise data.

[0009] In a possible implementation, the frequency range is 5 Hz to 8000 Hz, and the frequency step is 5 Hz.

[0010] In a possible implementation, the training of the constructed RBF neural network based on the sound field vibration data and adjusting the parameters of the RBF neural network include: Constructing an RBF neural network and determining parameters of the RBF neural network; Dividing the sound field vibration data to obtain training set data and test set data; Inputting the training set data into the input layer, transforming the training set data through the input layer to obtain a numerical vector; Inputting the digitized vector into the hidden layer, performing nonlinear transformation and feature extraction on the digitized vector through the hidden layer to obtain a feature vector, wherein the neuron with radial basis function as the activation function of the hidden layer; Inputting the feature vector into the output layer, performing linear combination on the feature vector through the output layer to obtain a prediction result; Performing a denormalization process on the prediction result to obtain prediction data, performing error calculation on the prediction data to obtain an error, and adjusting the parameters of the RBF neural network based on the error to obtain a fully trained RBF neural network; The test set data is input into the fully trained RBF neural network to obtain the predicted data of the test set, and the fully trained RBF neural network is evaluated based on the predicted data of the test set.

[0011] In a possible implementation, the parameters of the RBF neural network include a beta value, a center point, and a dimension of an output layer.

[0012] In a possible implementation, the sound field vibration data to be predicted is predicted based on the fully trained RBF neural network to obtain a predicted total sound pressure level value, including: Inputting vibration data in the sound field vibration data to be predicted into the well-trained RBF neural network, predicting the vibration data based on the well-trained RBF neural network, and obtaining noise data with the same frequency as the vibration data; A total sound pressure level prediction value is obtained based on the noise data.

[0013] On the other hand, the present invention also provides a sound field prediction device based on RBF neural network, comprising: A data acquisition module is used to establish a sound field structure model, perform sound field vibration simulation on the sound field structure model, and obtain sound field vibration data under different working conditions; A training module, used for training the constructed RBF neural network based on the sound field vibration data, and adjusting the parameters of the RBF neural network, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; The prediction module is used to predict the sound field vibration data to be predicted based on the fully trained RBF neural network to obtain a predicted value of the total sound pressure level.

[0014] In another aspect, the present invention further provides an electronic device, comprising: a processor and a memory; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, the steps in the sound field prediction method based on the RBF neural network as described above are implemented.

[0015] On the other hand, the present invention also provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the sound field prediction method based on RBF neural network as described above.

[0016] The beneficial effects of the present invention are as follows: based on the sound field vibration data, the constructed RBF neural network is trained, and the parameters of the RBF neural network are adjusted; based on the fully trained RBF neural network, the sound field vibration data to be predicted is predicted to obtain a total sound pressure level prediction value, thereby improving the sound field prediction accuracy and prediction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of an embodiment of a sound field prediction method based on RBF neural network provided by the present invention; Figure 2 A schematic diagram of an RBF neural network model of a sound field prediction method based on an RBF neural network provided by the present invention; Figure 3A schematic diagram showing a comparison between a predicted value and an actual value of the sound field prediction method based on the RBF neural network provided by the present invention; Figure 4 A schematic diagram showing a comparison between another predicted value and an actual value of the sound field prediction method based on the RBF neural network provided by the present invention; Figure 5 A schematic diagram of the structure of an embodiment of a sound field prediction device based on RBF neural network provided by the present invention; Figure 6 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0018] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0019] The present invention discloses a sound field prediction method, device, equipment and medium based on RBF neural network, which can be used in a computer. The method, device or computer-readable storage medium involved in the present invention can be integrated with the above-mentioned device or can be relatively independent.

[0020] A specific embodiment of the present invention discloses a sound field prediction method based on RBF neural network, which can be executed by a computer, specifically by one or more processors of the computer. Figure 1 As shown, the sound field prediction method based on RBF neural network includes: S101, establishing a sound field structure model, performing sound field vibration simulation on the sound field structure model, and obtaining sound field vibration data under different working conditions; S102, training the constructed RBF neural network based on the sound field vibration data, and adjusting the parameters of the RBF neural network, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; S103 , predicting the sound field vibration data to be predicted based on the fully trained RBF neural network to obtain a predicted value of the total sound pressure level.

[0021] Among them, the mapping relationship between vibration data and noise data is established through the RBF (Radial Basis Function) neural network; the RBF neural network is used to quickly and accurately predict the sound wave propagation and distribution characteristics in various complex sound field environments without the need for cumbersome physical models and numerical calculations, saving a lot of manpower costs while reducing the possibility of errors due to repetitive work.

[0022] Compared with the prior art, the sound field prediction method based on RBF neural network provided in this embodiment establishes a sound field structure model, performs sound field vibration simulation on the sound field structure model, and obtains sound field vibration data under different working conditions; trains the constructed RBF neural network based on the sound field vibration data, and adjusts the parameters of the RBF neural network, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; predicts the sound field vibration data to be predicted based on the well-trained RBF neural network, obtains the predicted value of the total sound pressure level, and improves the sound field prediction accuracy and prediction efficiency.

[0023] In some embodiments, in step S101, a sound field structure model is established, and the sound field structure model is established with the aluminum plate as the sound field structure and the air domain as the sound field domain. In the Multiphysics finite element software, an aluminum sheet is selected as the acoustic field structure, and the air domain is used as the acoustic field domain to establish an acoustic field structure model; an acoustic field vibration simulation is performed on the acoustic field structure model to obtain acoustic field vibration data under different working conditions, and the frequency range and frequency step are determined. The frequency range is 5Hz~8000Hz, and the frequency step is 5Hz. Based on the frequency range and frequency step, a point load is applied to the aluminum sheet to obtain acoustic field vibration data under different working conditions, where the acoustic field vibration data includes vibration data and noise data. A point load is applied to the aluminum sheet, and the research frequency is (5,5,8000), that is, the frequency is 5Hz~8000Hz, and the frequency step is 5Hz. The acoustic vibration simulation of the structure under different working conditions is performed to obtain multiple groups of acoustic field vibration data under different working conditions. For each working condition, one-third octave band data in the same frequency band is selected, and a large part of the working condition data is used as a training set, and the remaining small amount of working condition data is used as a test set.

[0024] In some embodiments, in step S102, the constructed RBF neural network is trained based on the sound field vibration data, and the parameters of the RBF neural network are adjusted, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; first, the RBF neural network is constructed, and the schematic diagram of the RBF neural network model is shown in FIG. Figure 2 ,like Figure 2 As shown, is the input of the neural network, that is, the corresponding sound field vibration data, is the output of the neural network, It is the activation function, which maps the input of the neuron of the neural network to the output. is the weight matrix, i.e., the influencing factor of the neural network; the parameters of the RBF neural network are determined, including the beta value (self.beta), the center point (num_cebters), and the dimension of the output layer (out_dim); the sound field vibration data is divided to obtain the training set data and the test set data; secondly, the training set data is input into the input layer, and the training set data is transformed through the input layer to obtain a numerical vector; the numerical vector is input into the hidden layer, and the numerical vector is nonlinearly transformed and feature extracted through the hidden layer to obtain a feature vector, wherein the radial basis function is used as the basis. The function is the neuron of the activation function of the hidden layer; the feature vector is input into the output layer, and the feature vector is linearly combined through the output layer to obtain the prediction result; finally, the prediction result is denormalized to obtain the prediction data, the error is calculated on the prediction data to obtain the error, and the parameters of the RBF neural network are adjusted based on the error to obtain a fully trained RBF neural network; the test set data is input into the fully trained RBF neural network to obtain the prediction data of the test set, and the trained RBF neural network is evaluated based on the prediction data of the test set to evaluate the generalization ability of the RBF neural network and the performance of practical applications.

[0025] The RBF neural network has only three layers. The first layer is the input layer, the second layer is the hidden layer, and the third layer is the output layer. The weights between the neurons from the input layer to the hidden layer are all 1. The hidden layer is a neuron that uses the radial basis function as the activation function. The connection relationship between the hidden layer and the output layer is an ordinary neural network. The weights between them can be changed through training. Define the beta value, center point and expansion range of the RBF neural network. Different beta values ​​will produce different effects on the function. A smaller beta value will make the function smoother and cover a wider input space. Import the sorted training set data and standardize the data to ensure that the neural network can receive the operation. Denormalize the results predicted by the neural network to obtain the predicted data, and calculate the error of the predicted data. Finally, draw a comparison curve between the predicted value and the true value, and export the trained neural physical model rbf_model.joblib. By comparing the curves, continuously adjust the neural network model parameters. Finally, when self.beta=0.35, num_centers=85, out_dim=1, a better fitting model is obtained, that is, a fully trained RBF neural network.

[0026] In some embodiments, in step S103, the sound field vibration data to be predicted is predicted based on a well-trained RBF neural network to obtain a total sound pressure level prediction value, the vibration data in the sound field vibration data to be predicted is input into the well-trained RBF neural network, the vibration data is predicted based on the well-trained RBF neural network to obtain noise data with the same frequency as the vibration data; and the total sound pressure level prediction value is obtained based on the noise data.

[0027] Comparing the predicted results with the actual values, the predicted total sound pressure level of test set 1 is 144.226dB, and the actual total sound pressure level is 145.167dB. The difference in total sound pressure is 0.941dB. For the comparison diagram, please refer to Figure 3 The predicted total sound pressure level of test set 2 is 135.047dB, and the actual total sound pressure level is 136.229dB. The difference in total sound pressure is 1.182dB. For the comparison diagram, please refer to Figure 4 ,Through comparison, it can be seen that in complex and changeable sound field environments such as ,different working conditions, the sound field can be predicted more efficiently and a certain ,prediction accuracy can be guaranteed, which can save a lot of time and manpower and ,material resources.

[0028] In order to better implement the sound field prediction method based on RBF neural network in the embodiment of the present invention, based on the sound field prediction method based on RBF neural network, correspondingly, Figure 5 As shown, the embodiment of the present invention further provides a sound field prediction device based on RBF neural network. The sound field prediction device 500 based on RBF neural network includes: The data acquisition module 501 is used to establish a sound field structure model, perform sound field vibration simulation on the sound field structure model, and obtain sound field vibration data under different working conditions; A training module 502 is used to train the constructed RBF neural network based on the sound field vibration data and adjust the parameters of the RBF neural network, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; The prediction module 503 is used to predict the sound field vibration data to be predicted based on the fully trained RBF neural network to obtain a predicted value of the total sound pressure level.

[0029] like Figure 6 As shown, the present invention also provides an electronic device 600 , which can be a computing device such as a mobile terminal, a desktop computer, a notebook, a palmtop computer, a server, etc. The electronic device 600 includes a processor 601 , a memory 602 , and a display 603 . Figure 6 Only some components of the electronic device 600 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0030] In some embodiments, the memory 602 may be an internal storage unit of the electronic device 600, such as a hard disk or memory of the electronic device 600. In other embodiments, the memory 602 may also be an external storage device of the electronic device 600, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 600. Further, the memory 602 may also include both an internal storage unit of the electronic device 600 and an external storage device. The memory 602 is used to store application software and various types of data installed in the electronic device 600, such as program codes installed in the electronic device 600. The memory 602 may also be used to temporarily store data that has been output or is to be output. In one embodiment, a sound field prediction program based on an RBF neural network is stored on the memory 602, and the sound field prediction program based on the RBF neural network can be executed by the processor 601, thereby realizing the sound field prediction method based on the RBF neural network of each embodiment of the present invention.

[0031] In some embodiments, the processor 601 may be a central processing unit (CPU), a microprocessor or other data processing chip, and is used to run program codes or process data stored in the memory 602, such as a sound field prediction method based on an RBF neural network.

[0032] In some embodiments, the display 603 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 603 is used to display the identification information of the sound field prediction program based on the RBF neural network and to display a visual user interface. The components 601-603 of the electronic device 600 communicate with each other through a system bus.

[0033] In some embodiments, when the processor 601 executes the sound field prediction program based on the RBF neural network in the memory 602, the various steps in the sound field prediction method based on the RBF neural network described in the above embodiments are implemented. Since the sound field prediction method based on the RBF neural network has been described in detail above, it will not be repeated here.

[0034] Accordingly, the present invention also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, the steps or functions of the sound field prediction method based on the RBF neural network provided in the above-mentioned method embodiments can be implemented.

[0035] In summary, the sound field prediction method, device, equipment and medium based on RBF neural network provided by the present invention establish a sound field structure model, perform sound field vibration simulation on the sound field structure model, and obtain sound field vibration data under different working conditions; train the constructed RBF neural network based on the sound field vibration data, and adjust the parameters of the RBF neural network, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; predict the sound field vibration data to be predicted based on the well-trained RBF neural network, obtain the total sound pressure level prediction value, and improve the sound field prediction accuracy and prediction efficiency.

[0036] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0037] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A sound field prediction method based on RBF neural network, characterized in that: include: Establishing an acoustic field structure model, performing acoustic field vibration simulation on the acoustic field structure model, and obtaining acoustic field vibration data under different working conditions; Training the constructed RBF neural network based on the sound field vibration data, and adjusting the parameters of the RBF neural network, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; The sound field vibration data to be predicted is predicted based on the fully trained RBF neural network to obtain a predicted value of the total sound pressure level.

2. The sound field prediction method based on RBF neural network according to claim 1 is characterized in that: The step of establishing the sound field structure model comprises: The acoustic field structure model is established with aluminum sheet as the acoustic field structure and air domain as the acoustic field domain.

3. The sound field prediction method based on RBF neural network according to claim 2 is characterized in that: The performing of sound field vibration simulation on the sound field structure model to obtain sound field vibration data under different working conditions includes: A frequency range and a frequency step are determined, and a point load is applied to the aluminum sheet based on the frequency range and the frequency step to obtain sound field vibration data under different working conditions, wherein the sound field vibration data includes vibration data and noise data.

4. The sound field prediction method based on RBF neural network according to claim 3 is characterized in that: The frequency range is 5 Hz to 8000 Hz, and the frequency step is 5 Hz.

5. The sound field prediction method based on RBF neural network according to claim 3 is characterized in that: The step of training the constructed RBF neural network based on the sound field vibration data and adjusting the parameters of the RBF neural network includes: Constructing an RBF neural network and determining parameters of the RBF neural network; Dividing the sound field vibration data to obtain training set data and test set data; Inputting the training set data into the input layer, transforming the training set data through the input layer to obtain a numerical vector; Inputting the digitized vector into the hidden layer, performing nonlinear transformation and feature extraction on the digitized vector through the hidden layer to obtain a feature vector, wherein the neuron with radial basis function as the activation function of the hidden layer; Inputting the feature vector into the output layer, performing linear combination on the feature vector through the output layer to obtain a prediction result; Performing a denormalization process on the prediction result to obtain prediction data, performing error calculation on the prediction data to obtain an error, and adjusting the parameters of the RBF neural network based on the error to obtain a fully trained RBF neural network; The test set data is input into the fully trained RBF neural network to obtain the predicted data of the test set, and the fully trained RBF neural network is evaluated based on the predicted data of the test set.

6. The sound field prediction method based on RBF neural network according to claim 5 is characterized in that: The parameters of the RBF neural network include beta value, center point and dimension of output layer.

7. The sound field prediction method based on RBF neural network according to claim 5, characterized in that: The method of predicting the sound field vibration data to be predicted based on the fully trained RBF neural network to obtain a predicted value of the total sound pressure level includes: Inputting vibration data in the sound field vibration data to be predicted into the well-trained RBF neural network, predicting the vibration data based on the well-trained RBF neural network, and obtaining noise data with the same frequency as the vibration data; A total sound pressure level prediction value is obtained based on the noise data.

8. A sound field prediction device based on RBF neural network, characterized in that: include: A data acquisition module is used to establish a sound field structure model, perform sound field vibration simulation on the sound field structure model, and obtain sound field vibration data under different working conditions; A training module, used for training the constructed RBF neural network based on the sound field vibration data, and adjusting the parameters of the RBF neural network, wherein the RBF neural network includes an input layer, a hidden layer, and an output layer; The prediction module is used to predict the sound field vibration data to be predicted based on the fully trained RBF neural network to obtain a predicted value of the total sound pressure level.

9. An electronic device, characterized in that: including memory and processor; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, the steps in the sound field prediction method based on the RBF neural network as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the sound field prediction method based on RBF neural network as described in any one of claims 1-7.

Images