A method, system, device and storage medium for pneumatic data fusion

Abstract

A method, system, device, and storage medium for aerodynamic data fusion are disclosed. The method includes: acquiring aerodynamic data from different sources; using the measurements in the aerodynamic data as the identification framework of the D-S evidence theory; selecting a membership function based on the characteristics of the aerodynamic data, calculating the membership degrees between each group of aerodynamic data, and normalizing the membership degrees to construct a basic assignment function under the identification framework of the D-S evidence theory; calculating the distance and cosine value of the angle between each piece of evidence in the D-S evidence theory and the fused evidence, using the distance and cosine value of the angle as evidence adjustment parameters to correct the basic assignment function so that the basic assignment function after evidence fusion converges; using the converged basic assignment function as the weighting coefficient for data fusion to calculate the fused aerodynamic data. This invention improves and optimizes the D-S evidence theory, fully integrating the trend information provided by low-precision data and the precision information provided by high-precision data.

Description

Technical Field

[0001] This invention belongs to the field of aerodynamics technology and relates to an aerodynamic data fusion method, system, device and storage medium. Background Technology

[0002] Aerodynamics primarily studies the force characteristics of an aircraft, the flow patterns of air, and the accompanying chemical and physical changes when an aircraft is in relative motion with the air. It is an important branch of fluid mechanics and a crucial fundamental science and technology in the aerospace field. [1] Aerodynamic data is used to simulate and build aerodynamic models of aircraft, making the simulation results closer to reality, and is key to dynamic simulation. With technological advancements and increasing performance requirements for aircraft, the scale of aerodynamic data has increased, and its structure has become more complex. Aerodynamic data is reflected in the aircraft's equations of motion primarily in the form of three force coefficients (lift coefficient, drag coefficient, and side force coefficient) and three moment coefficients (roll moment coefficient, yaw moment coefficient, and pitch moment coefficient). These are functions of many variables, such as angle of attack, sideslip angle, aerodynamic control surface deflection angle, altitude, Mach number, angular velocity, ground effect, engine exhaust, landing gear configuration, flap configuration, and speed brake configuration. [2] .

[0003] There are generally three ways to obtain aerodynamic data: wind tunnel testing, numerical calculation, and flight testing. Each method has its own advantages and disadvantages: wind tunnel testing can obtain aerodynamic data under any operating conditions, but it cannot completely correct for the effects of tunnel wall interference, supports, Reynolds number, etc., so the reliability of the data is questionable; numerical calculation is low-cost and provides a lot of data, but the accuracy of the calculated results and the amount of calculation are mutually constrained, and one may lose sight of the other; flight testing can obtain real aerodynamic characteristics, however, its results are affected by sensor measurement accuracy, atmospheric disturbances and calculation errors, and it can only obtain limited aerodynamic data.

[0004] Improving the quality of aerodynamic data can reduce the amount of data required and effectively lower experimental costs. Obtaining more reliable aerodynamic data requires comprehensive consideration of uncertain information from multiple sources, such as information from multiple sensors and data of varying precision obtained from multiple measurements. The process of fusing this uncertain information is essentially an uncertain reasoning process. Due to its complete mathematical foundation, DS evidence theory can establish a trust function by constraining the probability of events without precisely specifying the probability that is difficult to obtain. It can conveniently handle the "uncertainty" caused by "not knowing" and has become one of the most commonly used methods in uncertainty reasoning. It is widely used in multi-sensor information fusion, but it also has some shortcomings. When the number of focal elements in the set is too large, it will cause the information explosion problem. It is also reflected in the counterintuitive paradoxes that occur when fusing conflicting evidence.

[0005] [1] Anderson. Fundamentals of Aerodynamics [M]. Translated by Yang Yong. Beijing: Aviation Industry Press, 2010: 109-110.

[0006] [2] Hu Mengquan, Zhang Dengcheng, Zhang Meizhong. Advanced Atmospheric Flight Mechanics: Aviation Industry Press, 2007. Summary of the Invention

[0007] The purpose of this invention is to address the problems in the prior art by providing a method, system, device, and storage medium for aerodynamic data fusion. This method assigns weights to aerodynamic data of different precision to reduce the computational load and complexity of fusion. By iteratively correcting the evidence and then utilizing the evidence combination method provided by the DS evidence theory, aerodynamic data fusion is achieved, thereby reducing the impact of error data.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] Firstly, a method for fusion of aerodynamic data is provided, including the following steps:

[0010] Acquire aerodynamic data from different sources and use the measurements in the aerodynamic data as the identification framework for DS evidence theory;

[0011] Based on the characteristics of aerodynamic data, a membership function is selected, the membership degree between each group of aerodynamic data is calculated, and the membership degree is normalized. Under the identification framework of DS evidence theory, a basic assignment function is constructed.

[0012] Calculate the distance and cosine of the angle between each piece of evidence in the DS evidence theory and the fused evidence. Use the distance and cosine of the angle as evidence adjustment parameters to modify the basic assignment function so that the basic assignment function after evidence fusion converges.

[0013] The convergent basic assignment function is used as the weighting coefficient for data fusion to calculate the fused aerodynamic data.

[0014] As a preferred embodiment, the step of acquiring aerodynamic data from different sources and using the measurements in the aerodynamic data as the identification framework for the DS evidence theory, wherein the aerodynamic data from different sources are {x i |i=1,2,…,N}, the identification frame is Θ={L i |i=1,2,...,N};

[0015] The steps of selecting a membership function based on the characteristics of aerodynamic data, calculating the membership degree between each group of aerodynamic data, and normalizing the membership degree to construct the basic assignment function within the identification framework of DS evidence theory include:

[0016] Choose a suitable membership function p, p ijTo reflect the differences between different sets of aerodynamic data, a membership matrix P consisting of N sets of aerodynamic data is obtained:

[0017]

[0018] The confidence coefficient R = {r1, r2, ..., r2} is obtained from the membership matrix P. N After normalization, the weight coefficients W = {w} are generated for the basic assignment. i The expression for calculating |i=1,2,…,N} is as follows:

[0019]

[0020]

[0021] The membership matrix P is normalized using the weight coefficients W to obtain the basic assignment function:

[0022]

[0023] As a preferred embodiment, the steps of calculating the distance and cosine value of the angle between each piece of evidence in the DS evidence theory and the fused evidence include:

[0024] The Jousselme distance is calculated to quantify the difference between different pieces of evidence m1 and evidence m2. The calculation expression is as follows:

[0025]

[0026]

[0027] In the formula, A i and A j It is any subset of the power set 2Θ; |A| represents the cardinality of set A, that is, the number of elements in set A; d 12 The larger the value, the stronger the conflict between the information contained in the two pieces of evidence;

[0028] Calculate the degree to which the i-th piece of evidence is supported by other evidence, v. i The generalization yields the total support parameters V = [v1, v2, ..., v]. N ], and normalize it, the calculation expression is as follows:

[0029]

[0030]

[0031] In the formula, v i The larger the value, the greater the degree to which the i-th piece of evidence is supported by other evidence;

[0032] Calculate the cosine value c of the angle between different pieces of evidence. ij The similarity s between the evidence i Generalization yields the similarity parameter S = [s1, s2, ..., s...] for all evidence. N ], and normalize it, the calculation expression is as follows:

[0033]

[0034]

[0035]

[0036] In the formula, Let s represent the evidence vector. i s represents the similarity between the i-th piece of evidence and other evidence. i The larger the value, the higher the similarity between the i-th piece of evidence and other evidence.

[0037] As a preferred embodiment, the step of using distance and the cosine of the included angle as evidence adjustment parameters to correct the basic assignment function includes:

[0038] The degree of support using the distance metric v i Similarity parameters representing the cosine index Together they constitute the discount factor κ i The discount factor K = [κ1,κ2,...,κ] used to generalize and obtain all the evidence is [κ1,κ2,...,κ]. N ], and normalize it, the calculation expression is as follows:

[0039]

[0040]

[0041] In the formula, a,b∈[0,1], a+b=1, are the adjustment parameters for the contribution of the distance index and the cosine index to the discount factor;

[0042] Using correction coefficients Modify the basic assignment function.

[0043] As a preferred embodiment, the step of bringing the basic assignment function after evidence fusion to converge includes:

[0044] Using correction coefficients After modifying the basic assignment function, the evidence for the modification is as follows:

[0045]

[0046] The updated fusion evidence is as follows:

[0047]

[0048] When|E k -E k-1 If |>ε, then the convergence requirement is not met;

[0049] Until the basic assignment function after evidence fusion converges, the basic assignment probability set is output:

[0050] E k ={m * (A1),m * (A2),…,m * (A N )}.

[0051] As a preferred embodiment, the step of using the convergent basic assignment function as the weighting coefficient for data fusion and calculating the fused aerodynamic data involves setting the basic assignment probability set E... k The elements in the data are used as weighting coefficients for data fusion. The measured values ​​in the fused aerodynamic data are then calculated according to the following formula:

[0052]

[0053] Secondly, a pneumatic data fusion system is provided, including:

[0054] The data acquisition module is used to acquire aerodynamic data from different sources and use the measurements in the aerodynamic data as the identification framework for the DS evidence theory.

[0055] The basic assignment function construction module is used to select the membership function based on the characteristics of aerodynamic data, calculate the membership degree between each group of aerodynamic data, and normalize the membership degree. It constructs the basic assignment function within the identification framework of DS evidence theory.

[0056] The basic assignment function correction module is used to calculate the distance and cosine value of the angle between each piece of evidence in the DS evidence theory and the fused evidence. The distance and cosine value of the angle are used as evidence adjustment parameters to correct the basic assignment function so that the basic assignment function after evidence fusion converges.

[0057] The data fusion module is used to calculate the fused aerodynamic data by using the convergent basic assignment function as the weighting coefficient for data fusion.

[0058] Thirdly, an electronic device is provided, comprising: a memory storing at least one instruction; and a processor executing the instruction stored in the memory to implement the pneumatic data fusion method.

[0059] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the pneumatic data fusion method.

[0060] Compared with the prior art, the present invention has at least the following beneficial effects:

[0061] To address the ever-increasing scale and complexity of aerodynamic data, this invention improves and optimizes upon the DS evidence theory. After acquiring aerodynamic data of varying precision, it avoids the phenomenon of "over-assimilation" of high-precision data information by low-precision data information in traditional DS evidence theory, fully integrating the trend information provided by low-precision data with the precision information provided by high-precision data. The aerodynamic data fusion method proposed in this invention reduces the computational load and complexity of fusion by scientifically allocating weights for aerodynamic data of different precisions, mitigating the "focal element explosion" problem. Regarding the issue of evidence conflict, this invention calculates the distance and cosine of the angle between each piece of evidence in the DS evidence theory and the fused evidence. Using these distances and cosines as evidence adjustment parameters, it corrects the basic assignment function, ensuring convergence of the basic assignment function after evidence fusion. By fusing multiple sets of experimental data, it effectively reduces the impact of large error data on the fusion results, providing a new method and approach for aerodynamic data fusion. Attached Figure Description

[0062] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 Flowchart of the pneumatic data fusion method according to an embodiment of the present invention;

[0064] Figure 2 Statistical chart of evidence iteration times under different levels of precision in embodiments of the present invention;

[0065] Figure 3 Schematic diagram of F-16 aerodynamic data fusion results according to an embodiment of the present invention:

[0066] (a) Force coefficient C x Fusion results; (b) Torque coefficient C L Fusion results; (c) Force coefficient C y Fusion results;

[0067] (d) Torque coefficient C m Fusion results; (e) Force coefficient C z Fusion results; (f) Torque coefficient C n Fusion results. Detailed Implementation

[0068] 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, those skilled in the art can obtain other embodiments without creative effort.

[0069] Please see Figure 1 The pneumatic data fusion method of this invention includes the following steps:

[0070] S1. Acquire aerodynamic data from different sources and use the measurements in the aerodynamic data as the identification framework for the DS evidence theory.

[0071] S2. Select a membership function based on the characteristics of aerodynamic data, calculate the membership degree between each group of aerodynamic data, normalize the membership degree, and construct a basic assignment function under the identification framework of DS evidence theory.

[0072] S3. Calculate the distance and cosine of the angle between each piece of evidence in the DS evidence theory and the fused evidence. Use the distance and cosine of the angle as evidence adjustment parameters to modify the basic assignment function so that the basic assignment function after evidence fusion converges.

[0073] S4. Using the convergent basic assignment function as the weighting coefficient for data fusion, the fused aerodynamic data is calculated.

[0074] In one possible implementation, the aerodynamic data from different sources mentioned in step S1 are {x} i The identification frame Θ = {L |i=1,2,…,N} is obtained from the measurements within it. i |i=1,2,...,N};

[0075] In one possible implementation, step S2 specifically includes the following steps:

[0076] Choose a suitable membership function p, p ij To reflect the differences between different sets of aerodynamic data, a membership matrix P consisting of N sets of aerodynamic data is obtained:

[0077]

[0078] The confidence coefficient R = {r1, r2, ..., r} is obtained from the membership matrix P. N After normalization, the weight coefficients W = {w} are generated for the basic assignment. i The expression for calculating |i=1,2,…,N} is as follows:

[0079]

[0080]

[0081] By normalizing the membership matrix P using the weight coefficients W, the basic assignment function is obtained:

[0082]

[0083] In one possible implementation, step S3, calculating the distance and cosine value of the angle between each piece of evidence in the DS evidence theory and the fused evidence, specifically includes:

[0084] The Jousselme distance is calculated to quantify the difference between different pieces of evidence m1 and evidence m2. The calculation expression is as follows:

[0085]

[0086]

[0087] In the formula, A i and A j It is any subset of the power set 2Θ; |A| represents the cardinality of set A, that is, the number of elements in set A; d 12 The larger the value, the stronger the conflict between the information contained in the two pieces of evidence;

[0088] Calculate the degree to which the i-th piece of evidence is supported by other evidence, v. i The generalization yields the total support parameters V = [v1, v2, ..., v]. N ], and normalize it, the calculation expression is as follows:

[0089]

[0090]

[0091] In the formula, v i The larger the value, the greater the degree to which the i-th piece of evidence is supported by other evidence;

[0092] Calculate the cosine value c of the angle between different pieces of evidence. ij The similarity s between the evidence i Generalization yields the similarity parameter S = [s1, s2, ..., s...] for all evidence. N ], and normalize it, the calculation expression is as follows:

[0093]

[0094]

[0095]

[0096] In the formula, Let s represent the evidence vector. i s represents the similarity between the i-th piece of evidence and other evidence. i The larger the value, the higher the similarity between the i-th piece of evidence and other evidence.

[0097] The degree of support using the distance metric v i Similarity parameters representing the cosine index Together they constitute the discount factor κ i The discount factor K = [κ1,κ2,...,κ] used to generalize and obtain all the evidence is [κ1,κ2,...,κ]. N ], and normalize it, the calculation expression is as follows:

[0098]

[0099]

[0100] In the formula, a,b∈[0,1], a+b=1, are the adjustment parameters for the contribution of the distance index and the cosine index to the discount factor;

[0101] Using correction coefficients The basic assignment function was modified, and the evidence for the modified version is as follows:

[0102]

[0103] The updated fusion evidence is as follows:

[0104]

[0105] When|E k -E k-1 If | > ε, then the convergence requirement is not met; return to iterating by calculating the distance and cosine of the angle between each piece of evidence in the DS evidence theory and the fused evidence, until the basic assignment function after evidence fusion converges, and output the basic assignment probability set:

[0106] E k ={m * (A1),m * (A2),…,m * (A N )}

[0107] Finally, the basic assignment probability set E k The elements in the data are used as weighting coefficients for data fusion. The measured values ​​in the fused aerodynamic data are then calculated according to the following formula:

[0108]

[0109] The core idea of ​​the method for generating basic probability assignments is to determine the mutual support based on the error distance between different pieces of evidence, and then generate basic probability assignments from the support. Since unequal support between pieces of evidence can lead to conflicting evidence, this invention employs a method of modifying conflicting evidence to adjust for these conflicts.

[0110] Using the Jousselme distance and cosine of the angle between the evidence as discount factors to modify the evidence, the probability value after fusion is continuously modified. Each time the discount factor is calculated, the Jousselme distance and cosine of the angle between the fused probability and the basic assignment probability are calculated. This can make full use of the reliability of the original evidence.

[0111] When generating basic probability assignments using aerodynamic data of different precisions, the basic assignment error between low-precision aerodynamic data is smaller, while the error between low-precision and high-precision data is larger. This results in low overall support of low-precision data for high-precision data. After evidence correction, the coefficients of high-precision data are smaller, and the fusion coefficients corresponding to high-precision data are also very small. It is difficult to fully integrate the information of high-precision data into the final aerodynamic data.

[0112] To better integrate aerodynamic data, this invention first uses an improved DS theory to fuse low-precision data, obtaining fused low-precision data. Then, it uses prior knowledge of the data's reliability to standardize it, and finally fuses data of different precision levels, making full use of the trend information provided by the low-precision data and the precision information provided by the high-precision data.

[0113] The verification example cited in reference [3] uses the data fusion method proposed in this invention and compares the results with those of the classic DS method, Yager method [4], Murphy method [5] and the method proposed by Deng et al. [6].

[0114] [3]Ali T,Dutta P,Boruah HA new combination rule for conflict problem of Dempster-Shafer evidence theory[J].International Journal of Energy,Information and Communications,2012,3(1):35-40.

[0115] [4]Yager R R.On the aggregation of prioritized belief structures[J].IEEE Transactions on Systems,Man,and Cybernetics-Part A:Systems and Humans,1996,26(6):708-717.

[0116] [5]Murphy C K.Combining belief functions when evidence conflicts[J].Decision Support Systems,2000,29(1):1-9.

[0117] [6]Deng Y, Shi WK, Zhu ZF. Efficient combination approach of conflictevidence[J]. Journal of Infrared and Millimeter Waves, 2004, 23(1):27-32.

[0118] Example framework: Θ = {A, B, C}, under this identification framework, there are 5 data points:

[0119]

[0120] Four out of the five pieces of original evidence strongly support element A, while the second piece of evidence conflicts significantly with the other evidence.

[0121] Table 1 Comparison of fusion results

[0122]

[0123] The results of various fusion methods in Table 1 show that the classic DS combination rule cannot handle extreme cases where evidence completely negates some elements. This means that no matter how much evidence is subsequently introduced, element A will be excluded first in the combination rule, leading to an inability to obtain a correct identification result. Although the Yager method improves the combination rule to some extent, when faced with clearly conflicting evidence, as more evidence is added, the probability of obtaining the uncertainty domain Θ only increases, again failing to correctly identify the result. The Murphy method uses average probability, which effectively solves the "one-vote veto" phenomenon in evidence. However, due to the influence of evidence conflict, a correct identification result can only be obtained after all five pieces of evidence are introduced. When there is little evidence, the identification result is not ideal. Deng's method uses relative entropy information to handle the identification difficulties caused by evidence conflict, significantly improving the identification effect. However, his method only processes and synthesizes the evidence once, so the probability allocation needs improvement. In the method of this invention, when only two conflicting pieces of evidence are introduced, a "one-vote veto" phenomenon will also occur. However, with the introduction of new evidence, conflicting information in the evidence can be quickly eliminated through iterative adjustment. It can be seen that when a third piece of evidence is introduced, the target can be identified very well. After introducing the same evidence, the difference in the converged probability allocation is more obvious, and the identification effect is better.

[0124] Under different threshold ε constraints, different numbers of evidence points were introduced, and the number of iterations and corrections to the evidence in the experiment was as follows: Figure 2 As shown, Figure 2 The horizontal axis represents the logarithm of the set threshold ε, denoted by -lg(ε); the vertical axis represents the number of iterations required for evidence convergence during the evidence revision process. Experimental results show that as the precision ε increases, the number of iterations required for the evidence combination result to converge also increases; when the precision reaches ε = e -5 Subsequently, the number of iterations required for various evidence combinations remained almost constant. Moreover, as more evidence was introduced, the number of iterations required to achieve the same level of accuracy increased.

[0125] In another embodiment, a wind tunnel simulation program was used to generate four sets of low-precision simulation data and one set of high-precision data for the F-16 under the same operating conditions. These data were used as the identification framework for the evidence theory to verify the practicality of the improved DS fusion method. The results were compared with those of the classic DS fusion method, and the errors between the different fusion results and the calculated values ​​were calculated.

[0126] To clearly demonstrate the fusion results, only local aerodynamic data fusion analysis was performed. At the same time, since the aerodynamic coefficient is determined by multiple independent variables, all other variables besides the plotting variables were set as constants when plotting. The specific parameter settings are shown in Table 2.

[0127] Table 2. Values ​​of experimental variables

[0128]

[0129] Based on the values ​​in Table 2, aerodynamic data were generated using the F-16 wind tunnel simulation program. The six aerodynamic force and moment coefficient data were then fused using the process described in the previous section. The results are as follows: Figure 3 As shown in Figures (a) to (f).

[0130] In the experiments of this embodiment, among the four sets of low-precision simulation data generated for each coefficient, one set had a relatively large simulation error. When performing DS fusion, the evidence generated by the other three sets of data showed lower support for this set of data; therefore, this set of data had a smaller impact on the final experimental results. Since the other three sets of simulation data had smaller errors, and the trends of aerodynamic data values ​​with changes in the independent variable were more similar, the three sets of experimental data with smaller low-precision errors had a greater impact on the low-precision aerodynamic data fusion results, while the data with larger errors had a smaller impact on the overall fusion results. Low-precision aerodynamic data fusion corresponds to... Figure 3 The low-precision fusion result; after fusing the high- and low-precision aerodynamic data, the following is obtained: Figure 3 The improved DS fusion results are presented. Experimental results show that the results of low-precision fusion differ significantly from those of the improved DS fusion. This is because low-precision data contains more data and can only provide the overall trend of data changes, having a smaller impact on the final result. In contrast, high-precision data measurement results are more accurate and therefore have a greater impact on the final fusion result. The classic DS fusion result in the figure uses DS evidence theory to obtain the fusion coefficients. The fusion results show that due to the large number of low-precision data sources, the final fusion result is closer to the low-precision fusion result.

[0131] With C x Taking parameters as an example, we established polynomial models for wind tunnel aerodynamic simulation data and two fused data respectively, and calculated the prediction accuracy of each model. The comparison results are shown in Table 3.

[0132] Table 3C x Model prediction results

[0133]

[0134] The data in Table 3 were obtained by comparing the predicted outputs of each model with the theoretical calculations; the results of the low-precision dataset are the average values ​​of the indicators of each low-precision data model. As shown in Table 3, the prediction accuracy of the model built using the improved DS evidence theory fusion of aerodynamic data is higher than that of the model based on the simulation dataset with arbitrary single precision and the model based on the classic DS evidence theory fusion. This indicates that the improved fusion method proposed in this invention can improve the quality of aerodynamic data and provides a new approach to aerodynamic data fusion.

[0135] Another embodiment of the present invention also proposes a pneumatic data fusion system, comprising:

[0136] The data acquisition module is used to acquire aerodynamic data from different sources and use the measurements in the aerodynamic data as the identification framework for the DS evidence theory.

[0137] The basic assignment function construction module is used to select the membership function based on the characteristics of aerodynamic data, calculate the membership degree between each group of aerodynamic data, and normalize the membership degree. It constructs the basic assignment function within the identification framework of DS evidence theory.

[0138] The basic assignment function correction module is used to calculate the distance and cosine value of the angle between each piece of evidence in the DS evidence theory and the fused evidence. The distance and cosine value of the angle are used as evidence adjustment parameters to correct the basic assignment function so that the basic assignment function after evidence fusion converges.

[0139] The data fusion module is used to calculate the fused aerodynamic data by using the convergent basic assignment function as the weighting coefficient for data fusion.

[0140] Another embodiment of the present invention provides an electronic device, comprising: a memory storing at least one instruction; and a processor executing the instructions stored in the memory to implement the pneumatic data fusion method.

[0141] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned aerodynamic data fusion method.

[0142] For example, the instructions stored in the memory can be divided into one or more modules / units. These modules / units are stored in a computer-readable storage medium and executed by the processor to complete the pneumatic data fusion method of this embodiment. The one or more modules / units can be a series of computer-readable instruction segments capable of performing a specific function, which describe the execution process of the computer program in the server.

[0143] The electronic device may be a smartphone, laptop, PDA, or cloud server, among other computing devices. It may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the electronic device may also include more or fewer components, or combinations of certain components, or different components; for example, it may also include input / output devices, network access devices, buses, etc.

[0144] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0145] The memory can be an internal storage unit of the server, such as a hard drive or RAM. Alternatively, it can be an external storage device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory can include both internal and external storage units. The memory is used to store computer-readable instructions and other programs and data required by the server. It can also be used to temporarily store data that has been output or will be output.

[0146] It should be noted that the information interaction and execution process between the above-mentioned module units are based on the same concept as the method embodiment. For details on their specific functions and technical effects, please refer to the method embodiment section. They will not be repeated here.

[0147] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0148] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0149] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0150] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for fusion of aerodynamic data, characterized in that, Includes the following steps: Acquire aerodynamic data from different sources and use the measurements in the aerodynamic data as the identification framework for DS evidence theory; Based on the characteristics of aerodynamic data, a membership function is selected, the membership degree between each group of aerodynamic data is calculated, and the membership degree is normalized. Under the identification framework of DS evidence theory, a basic assignment function is constructed. Calculate the distance and cosine of the angle between each piece of evidence in the DS evidence theory and the fused evidence. Use the distance and cosine of the angle as evidence adjustment parameters to modify the basic assignment function so that the basic assignment function after evidence fusion converges. The convergent basic assignment function is used as the weighting coefficient for data fusion to calculate the fused aerodynamic data. The step of acquiring aerodynamic data from different sources and using the measurements in the aerodynamic data as the identification framework for the DS evidence theory involves aerodynamic data from different sources. The identification framework is ; The steps of selecting a membership function based on the characteristics of aerodynamic data, calculating the membership degree between each group of aerodynamic data, and normalizing the membership degree to construct the basic assignment function within the identification framework of DS evidence theory include: Choose an appropriate membership function , To reflect the differences between different sets of aerodynamic data, we obtain Membership matrix composed of aerodynamic data : From the membership matrix Calculate the credibility coefficient After normalization, the weight coefficients of the basic assignment are generated. The calculation expression is as follows: Using weighting coefficients Membership matrix After normalization, the basic assignment function is obtained: 。 2. The aerodynamic data fusion method according to claim 1, characterized in that, The steps for calculating the distance and cosine value of the angle between each piece of evidence in the DS evidence theory and the fused evidence include: Calculate the Jousselme distance to quantify different pieces of evidence. With evidence The difference between them can be calculated using the following expression: In the formula, and It is a power set Any subset below; Represents a set The cardinality of the set The number of elements contained; The larger the value, the stronger the conflict between the information contained in the two pieces of evidence; Calculate the first The degree to which this piece of evidence is supported by other evidence The promotion obtained all support parameters. After normalization, the calculation expression is as follows: In the formula, The larger the value, the more significant the effect. The greater the degree to which a piece of evidence is supported by other evidence; Calculate the cosine of the angle between different pieces of evidence. Characterize the similarity between pieces of evidence The similarity parameter of all evidence is obtained by generalization. After normalization, the calculation expression is as follows: In the formula, Represents the evidence vector. Representing the The similarity between this piece of evidence and other evidence. The larger the value, the more... The higher the similarity between a piece of evidence and other evidence, the better.

3. The aerodynamic data fusion method according to claim 2, characterized in that, The step of using distance and the cosine of the included angle as evidence adjustment parameters to correct the basic assignment function includes: Use the support level represented by the distance index Similarity parameters representing the cosine index Together constitute the discount factor Discount factor that has been fully proven After normalization, the calculation expression is as follows: In the formula, , , which are the adjustment parameters for the contribution of the distance index and the cosine index to the discount factor; Using correction coefficients Modify the basic assignment function.

4. The aerodynamic data fusion method according to claim 3, characterized in that, The steps to bring the basic assignment function after evidence fusion to converge include: Using correction coefficients After modifying the basic assignment function, the evidence for the modification is as follows: The updated fusion evidence is as follows: when If the convergence requirement is not met, then the convergence requirement is not satisfied. Until the basic assignment function after evidence fusion converges, the basic assignment probability set is output: 。 5. The aerodynamic data fusion method according to claim 4, characterized in that, The step of using the convergent basic assignment function as the weighting coefficient for data fusion to calculate the fused aerodynamic data involves setting the basic assignment probability set... The elements in the data are used as weighting coefficients for data fusion. The measured values ​​in the fused aerodynamic data are then calculated according to the following formula: 。 6. A pneumatic data fusion system, characterized in that, include: The data acquisition module is used to acquire aerodynamic data from different sources and use the measurements in the aerodynamic data as the identification framework for the DS evidence theory. The basic assignment function construction module is used to select the membership function based on the characteristics of aerodynamic data, calculate the membership degree between each group of aerodynamic data, and normalize the membership degree. It constructs the basic assignment function within the identification framework of DS evidence theory. The basic assignment function correction module is used to calculate the distance and cosine value of the angle between each piece of evidence in the DS evidence theory and the fused evidence. The distance and cosine value of the angle are used as evidence adjustment parameters to correct the basic assignment function so that the basic assignment function after evidence fusion converges. The data fusion module is used to calculate the fused aerodynamic data by using the convergent basic assignment function as the weighting coefficient for data fusion. The step of acquiring aerodynamic data from different sources and using the measurements in the aerodynamic data as the identification framework for the DS evidence theory involves aerodynamic data from different sources. The identification framework is ; The steps of selecting a membership function based on the characteristics of aerodynamic data, calculating the membership degree between each group of aerodynamic data, and normalizing the membership degree to construct the basic assignment function within the identification framework of DS evidence theory include: Choose an appropriate membership function , To reflect the differences between different sets of aerodynamic data, we obtain Membership matrix composed of aerodynamic data : From the membership matrix Calculate the credibility coefficient After normalization, the weight coefficients of the basic assignment are generated. The calculation expression is as follows: Using weighting coefficients Membership matrix After normalization, the basic assignment function is obtained: 。 7. An electronic device, characterized in that, include: Memory, storing at least one instruction; The processor executes the instructions stored in the memory to implement the pneumatic data fusion method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the pneumatic data fusion method as described in any one of claims 1 to 5.

Images