Fuselage turbulence micro-rib drag reduction prediction method based on Bayesian probability model

Abstract

The invention discloses a fuselage turbulence micro-rib resistance reduction prediction method based on a Bayesian probability model, and belongs to the technical field of micro-rib resistance reduction. According to the method, the problems of high calculation cost and high difficulty caused by large calculation grid quantity and lack of calculation methods when the resistance reduction characteristics of the fuselage micro-ribs are solved by a traditional numerical simulation method are solved. According to the method, parameterization extraction is carried out on related features of fuselage micro-ribs, shape parameters, environment parameters, flight state parameters and the like of the micro-ribs are obtained, parameter dimension reduction is carried out, then modeling is carried out on historical data through a Bayesian neural network method, and finally accurate and efficient prediction of resistance increment of unknown micro-rib structures is achieved. And uncertainty evaluation is carried out on a prediction result. According to the method, the problems that a traditional CFD calculation method is high in calculation cost are effectively solved, the test efficiency is further improved, and accurate and efficient resistance reduction income evaluation of the fuselage micro-rib can be completed.

Description

Technical Field

[0001] The invention relates to a method for predicting drag reduction of fuselage turbulence micro-ribs, and in particular to a method for predicting drag reduction of fuselage turbulence micro-ribs based on a Bayesian probability model, and belongs to the technical field of micro-rib drag reduction. Background Art

[0002] As a new drag reduction technology with broad application prospects in the field of transportation, the research on micro-rib technology has long received extensive attention from relevant research institutions at home and abroad. Different from traditional drag reduction technologies such as vortex generators, as a micron-scale characteristic configuration, micro-ribs are difficult to simulate, test, and prepare, especially for the turbulent drag reduction of civil aircraft fuselage walls with smaller scales in the boundary layer. The rib spacing on the surface of the micro-ribs used in civil aircraft fuselages is usually The fuselage of a civil aircraft is usually 30m, and the difference between the characteristic dimensions of the two is As mentioned above, for civil aircraft fuselages with micro-ribbed surfaces as a whole, there are two problems in directly meshing the micro-ribbed surface features during the numerical simulation of drag reduction characteristics: first, the amount of meshes will reach billions, which is unbearable; second, there is a lack of suitable calculation methods, among which conventional turbulence models cannot accurately simulate the boundary layer characteristics of the micro-ribbed surface, and direct numerical simulation or LES and other calculation methods are obviously not applicable. Therefore, it is necessary to combine the micro-ribbed drag reduction mechanism and establish a more efficient method suitable for the overall performance evaluation of the fuselage with the support of typical micro-ribbed surface drag reduction data. Summary of the invention

[0003] In order to overcome the problems of high computational cost and difficulty caused by large computational grid and lack of computational methods in solving the drag reduction characteristics of fuselage micro-ribs in traditional numerical simulation methods, the present invention provides a method for predicting the drag reduction of fuselage turbulent micro-ribs based on a Bayesian probability model. A brief overview of the present invention is given below to provide a basic understanding of certain aspects of the present invention. It should be understood that this overview is not an exhaustive overview of the present invention. It is not intended to identify the key or important parts of the present invention, nor is it intended to limit the scope of the present invention.

[0004] The technical solution of the present invention: A method for predicting drag reduction of fuselage turbulent micro-ribs based on a Bayesian probability model comprises the following steps: S1. Using the turbulent micro-rib surface drag reduction calculation data as support, the data was sorted and the principal component analysis method was used to reduce the data dimension; S2, using the Bayesian neural network method to establish a prediction probability model for the data after dimensionality reduction in step S1, and perform multiple rounds of iterations until the loss function of the prediction probability model converges; The specific implementation process is: the data after dimension reduction in S1 is used for model training and construction. Assuming that the parameter of the neural network is W and the training set is , where X is the input data and Y is the label; then given a new input data , new tags , the prior distribution of neural network weights is expressed as: ; in, represents the distribution of predicted values, Represents the modeling result of likelihood ratio, represents the posterior probability; To obtain the posterior probability , using the variational inference method, that is, assuming , each weight in the neural network Normal distribution Sampling in the hope and Close; at the same time, the KL divergence method is used to measure the distance between distributions, and its expression is: ; ; ; in, represents the model parameters, represents the mean, represents the standard deviation, represents the mean of the i-th parameter, represents the standard deviation of the i-th parameter, represents the maximum likelihood estimation of model parameters, represents the distribution of model estimates, represents the approximate true posterior distribution The variational distribution of represents the prior distribution, represents the likelihood function; The loss function of the Bayesian neural network is expressed as: ; in, represents the logarithm of the variational distribution, represents the logarithm of the prior distribution, represents the logarithm of the likelihood function; Perform gradient descent on the current loss function to continuously update value to obtain the optimal parameters, so that the optimal solution As the weight for model training, build a neural network; S3. Perform prediction analysis and uncertainty quantitative evaluation on the prediction probability model obtained in step S2, and provide an uncertainty range for the prediction value.

[0005] Furthermore, the specific implementation method of step S1 includes the following steps: S1.1. In the turbulent micro-rib control data, the input variables include control components, flight status, reference quantities, and control parameters; among which, the control components include: vertical tail, wing, horizontal tail; the flight status includes: speed, altitude, angle of attack, and sideslip angle; the reference quantities include: reference area, reference length, reference span, moment reference point, and speed pressure; the control parameters include: micro-rib shape parameters and environmental parameters, among which the micro-rib shape includes: dimensionless size and micro-structure shape; the output variables include drag increment, drag distribution, pitch moment increment, and pressure distribution; S1.2, perform feature dimensionality reduction on the data involved in S1.1; assuming that the total amount of sample data is m, the sample has n features, and the sample data is represented by K; first, K is formed into ; secondly, decentralize each row in the matrix, that is, subtract the mean of each row from the value of the row; secondly, obtain the covariance matrix , the n-dimensional variables in K are simplified to: ;

[0006] The covariance matrix expansion is expressed as: ; Among them, Cov is the covariance. Assume that the covariance between the pth and qth variables is calculated. After decentralization, The expression is: ; when hour, = , is the variance. After decentralization, assuming that the variance of the i-th variable is calculated, its expression is simplified to: ; Secondly, the covariance matrix is ​​decomposed into eigenvalues ​​according to , calculate Eigen(H) and get the eigenvalue and the corresponding eigenvector , arrange the eigenvectors into a matrix by row from top to bottom according to the corresponding eigenvalue size, and take the first The rows form the matrix R, which is reduced to The sample data expression for the dimension is: .

[0007] Further: the uncertainty assessment quantification method of the prediction data in step S3, by analyzing the distribution of weight coefficients in the Bayesian neural network, obtains the distribution of the prediction state points, thereby establishing an uncertainty prediction model; assuming that the distribution of the current prediction state points obeys the mean , the variance is The total sample size is n. By obtaining the z-score in the standard normal distribution table, ) is expressed as: ; in, represents the quantile of the standard normal distribution, represents the sample mean; The resulting confidence interval is the final uncertainty quantification range.

[0008] The beneficial effects of the present invention are embodied in: Compared with the prior art, the present invention performs parameter extraction of the related features of the fuselage micro-ribs, obtains the micro-rib shape parameters, environmental parameters, flight state parameters, etc., and performs parameter dimension reduction, and then models the historical data through the Bayesian neural network method, and finally achieves accurate and efficient prediction of the drag increment of the unknown micro-rib configuration, and performs uncertainty assessment on the prediction results. The present invention effectively solves the problems of high computational cost and high computational cost of the traditional CFD calculation method, and further improves the test efficiency, and can complete the accurate and efficient drag reduction benefit evaluation of the fuselage micro-ribs. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 This is a flow chart of a method for predicting fuselage turbulent micro-rib drag reduction based on a Bayesian probability model as described in Example 1. DETAILED DESCRIPTION

[0010] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is described below by the specific embodiments shown in the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In addition, in the following description, the description of known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention. It should also be understood that, unless otherwise specified or indicated, the terms "first", "second", "third" and the like in the specification are only used to distinguish between the various components, elements, steps, etc. in the specification, rather than to represent the logical relationship or sequential relationship between the various components, elements, steps, etc.

[0011] Embodiment 1, combination Figure 1 This embodiment describes a method for predicting drag reduction of fuselage turbulent micro-ribs based on a Bayesian probability model, comprising the following steps: S1. Using the turbulent micro-rib surface drag reduction calculation data as support, the data was sorted and the principal component analysis method was used to reduce the data dimension; The specific implementation method includes the following steps: S1.1. In the turbulent micro-rib control data, the input variables include control components, flight status, reference quantities, and control parameters; among which, the control components include: vertical tail, wing, horizontal tail; the flight status includes: speed, altitude, angle of attack, and sideslip angle; the reference quantities include: reference area, reference length, reference span, moment reference point, and speed pressure; the control parameters include: micro-rib shape parameters and environmental parameters, among which the micro-rib shape includes: dimensionless size and micro-structure shape; the output variables include drag increment, drag distribution, pitch moment increment, and pressure distribution; S1.2, perform feature dimensionality reduction on the data involved in S1.1. Assume that the total amount of sample data is m, the sample has n features, and the sample data is represented by K; first, K is formed into ; secondly, decentralize each row in the matrix, that is, subtract the mean of each row from the value of the row; secondly, obtain the covariance matrix , the n-dimensional variables in K are simplified to: ; The covariance matrix expansion is expressed as: ; Among them, Cov is the covariance. Assume that the covariance between the pth and qth variables is calculated. After decentralization, The expression is: ; when hour, = , is the variance. After decentralization, assuming that the variance of the i-th variable is calculated, its expression is simplified to: ; Secondly, the covariance matrix is ​​decomposed into eigenvalues ​​according to , calculate Eigen(H) and get the eigenvalue and the corresponding eigenvector , arrange the eigenvectors into a matrix by row from top to bottom according to the corresponding eigenvalue size, and take the first The rows form the matrix R, which is reduced to The sample data expression for the dimension is: .

[0012] S2, using the Bayesian neural network method to establish a prediction probability model for the data after dimensionality reduction in step S1, and perform multiple rounds of iterations until the loss function of the prediction probability model converges; The specific implementation process is: the data after dimension reduction in S1 is used for model training and construction. Assuming that the parameter of the neural network is W and the training set is , where X is the input data and Y is the label; then given a new input data , new tags , the prior distribution of neural network weights is expressed as:

[0013] in, represents the distribution of predicted values, Represents the modeling result of likelihood ratio, represents the posterior probability; To obtain the posterior probability , using the variational inference method, that is, assuming , each weight in the neural network Normal distribution Sampling in the hope and Close; at the same time, the KL divergence method is used to measure the distance between distributions, and its expression is: ; ; ; in, represents the model parameters, represents the mean, represents the standard deviation, represents the mean of the i-th parameter, represents the standard deviation of the i-th parameter, represents the maximum likelihood estimation of model parameters, represents the distribution of model estimates, represents the approximate true posterior distribution The variational distribution of represents the prior distribution, represents the likelihood function; The loss function of the Bayesian neural network is expressed as: ; in, represents the logarithm of the variational distribution, represents the logarithm of the prior distribution, represents the logarithm of the likelihood function; Perform gradient descent on the current loss function to continuously update value to obtain the optimal parameters, so that the optimal solution As the weight for model training, build a neural network; S3, performing prediction analysis and uncertainty quantitative evaluation on the prediction probability model obtained in step S2, and providing the uncertainty range of the prediction value; The uncertainty assessment quantitative method of the prediction data is to analyze the distribution of the weight coefficients in the Bayesian neural network to obtain the distribution of the prediction state points, thereby establishing an uncertainty prediction model; assuming that the distribution of the current prediction state points follows the mean , the variance is The total sample size is n. By obtaining the z-score in the standard normal distribution table, ) is expressed as: ; in, and represents the quantile of the standard normal distribution, represents the sample mean; The resulting confidence interval is the final uncertainty quantification range, and the default true value is ) will fall within this range.

[0014] Embodiment 2: An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for predicting fuselage turbulent micro-rib drag reduction based on a Bayesian probability model described in Example 1 are implemented.

[0015] The computer device of the present invention may be a device including a processor and a memory, such as a single chip microcomputer including a central processing unit. Furthermore, the processor is used to implement the steps of the above-mentioned method for predicting the drag reduction of fuselage turbulent micro-ribs based on the Bayesian probability model when executing the computer program stored in the memory.

[0016] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0017] The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, 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), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0018] Embodiment 3: A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of a method for predicting fuselage turbulent micro-rib drag reduction based on a Bayesian probability model described in Example 1 are implemented.

[0019] The computer-readable storage medium of the present invention can be any form of storage medium that can be read by a processor of a computer device, including but not limited to non-volatile memory, volatile memory, ferroelectric memory, etc. The computer-readable storage medium stores a computer program. When the processor of the computer device reads and executes the computer program stored in the memory, the steps of the above-mentioned method for predicting fuselage turbulent micro-rib drag reduction based on a Bayesian probability model can be implemented.

[0020] The computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer readable media do not include electric carrier signals and telecommunication signals.

[0021] The embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for predicting drag reduction of fuselage turbulent micro-ribs based on a Bayesian probability model, characterized in that: The following steps are involved: S1. Using the turbulent micro-rib surface drag reduction calculation data as support, the data was sorted and the principal component analysis method was used to reduce the data dimension; S2, using the Bayesian neural network method to establish a prediction probability model for the data after dimensionality reduction in step S1, and perform multiple rounds of iterations until the loss function of the prediction probability model converges; The specific implementation process is: the data after dimension reduction in S1 is used for model training and construction. Assuming that the parameter of the neural network is W and the training set is , where X is the input data and Y is the label; then given a new input data , new tags , the prior distribution of neural network weights is expressed as: ； in, represents the distribution of predicted values, Represents the modeling result of likelihood ratio, represents the posterior probability; To obtain the posterior probability , using the variational inference method, that is, assuming , each weight in the neural network Normal distribution Sampling in the hope and Close; at the same time, the KL divergence method is used to measure the distance between distributions, and its expression is: ； ； ； in, represents the model parameters, represents the mean, represents the standard deviation, represents the mean of the i-th parameter, represents the standard deviation of the i-th parameter, represents the maximum likelihood estimation of model parameters, represents the distribution of model estimates, represents the approximate true posterior distribution The variational distribution of represents the prior distribution, represents the likelihood function; The loss function of the Bayesian neural network is expressed as: ； in, represents the logarithm of the variational distribution, represents the logarithm of the prior distribution, represents the logarithm of the likelihood function; Perform gradient descent on the current loss function to continuously update value to obtain the optimal parameters, so that the optimal solution As the weight for model training, build a neural network; S3. Perform prediction analysis and uncertainty quantitative evaluation on the prediction probability model obtained in step S2, and provide an uncertainty range for the prediction value.

2. The method for predicting fuselage turbulent micro-rib drag reduction based on a Bayesian probability model according to claim 1, characterized in that: The specific implementation method of step S1 includes the following steps: S1.

1. In the turbulent micro-rib control data, the input variables include control components, flight status, reference quantities, and control parameters; among which, the control components include: vertical tail, wing, horizontal tail; the flight status includes: speed, altitude, angle of attack, and sideslip angle; the reference quantities include: reference area, reference length, reference span, moment reference point, and speed pressure; the control parameters include: micro-rib shape parameters and environmental parameters, among which the micro-rib shape includes: dimensionless size and micro-structure shape; the output variables include drag increment, drag distribution, pitch moment increment, and pressure distribution; S1.2, perform feature dimensionality reduction on the data involved in S1.1; assuming that the total amount of sample data is m, the sample has n features, and the sample data is represented by K; first, K is formed into ; secondly, decentralize each row in the matrix, that is, subtract the mean of each row from the value of the row; secondly, obtain the covariance matrix , the n-dimensional variables in K are simplified to: ； The covariance matrix expansion is expressed as: ； Among them, Cov is the covariance. Assume that the covariance between the pth and qth variables is calculated. After decentralization, The expression is: ； when hour, = , is the variance. After decentralization, assuming that the variance of the i-th variable is calculated, its expression is simplified to: ； Secondly, the covariance matrix is ​​decomposed into eigenvalues ​​according to , calculate Eigen(H) and get the eigenvalue and the corresponding eigenvector , arrange the eigenvectors into a matrix by row from top to bottom according to the corresponding eigenvalue size, and take the first The rows form the matrix R, which is reduced to The sample data expression for the dimension is: 。 3. The method for predicting fuselage turbulent micro-rib drag reduction based on a Bayesian probability model according to claim 1 or 2, characterized in that: The uncertainty evaluation and quantification method of the prediction data in step S3 is to obtain the distribution of the prediction state points by analyzing the distribution of the weight coefficients in the Bayesian neural network, thereby establishing an uncertainty prediction model; assuming that the distribution of the current prediction state points obeys the mean , the variance is The total sample size is n. By obtaining the z-score in the standard normal distribution table, ) is expressed as: ； in, represents the quantile of the standard normal distribution, represents the sample mean; The resulting confidence interval is the final uncertainty quantification range.

4. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of a method for predicting fuselage turbulence micro-rib drag reduction based on a Bayesian probability model as described in any one of claims 1 to 3 when executing the computer program.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for predicting fuselage turbulent micro-rib drag reduction based on a Bayesian probability model as described in any one of claims 1 to 3 are implemented.

Images