An aerodynamic simulation and damage assessment method based on ACMI technology

Abstract

The present application relates to the field of electronic digital data processing technology, and in particular to an aerodynamic simulation and damage assessment method based on ACMI technology. The method comprises: obtaining a historical data set D collected by ACMI; traversing D, and according to D i Get the i-th training sample and the i-th label B i ; Train the neural network model according to the i-th training sample and the i-th label to obtain the trained neural network model; obtain the label corresponding to the target data group according to the target data group collected by ACMI and the trained neural network model; obtain the real damage degree corresponding to the target data group according to the first collection parameter vector corresponding to the target data group, the label corresponding to the target data group and the damage assessment system. The present invention can improve the accuracy of damage degree assessment.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to an aerodynamic simulation and damage assessment method based on ACMI technology. Background Art

[0002] The Aerial Combat Maneuvering Instrumentation (ACMI) is an advanced data acquisition and analysis system used for air combat drills and training. It can collect and record various flight parameters and operating data of aircraft. Inputting the data of specific parameters collected by ACMI into the damage assessment system can realize the assessment of the degree of damage. These specific parameters usually include relevant parameters of the target (such as the location, size, protection level and structural strength of the target), relevant parameters of the device used (such as the type and strength of the device used) and relevant parameters of the environment (such as wind speed and weather visibility).

[0003] However, in some flight scenarios, some parameters collected by ACMI other than the above-mentioned specific parameters will also affect the actual degree of damage. For example, parameters related to the flight trajectory (flight path, speed, acceleration, etc.) can be used to evaluate the distance to the target. Parameters such as the delivery accuracy of the device used and the response time of the target will also have a certain impact on the degree of damage. This leads to a certain difference between the degree of damage predicted by the damage assessment system based only on the input data of specific parameters and the actual degree of damage, that is, the degree of damage predicted by the damage assessment system is not accurate. How to improve the accuracy of damage assessment based on the data collected by ACMI is an urgent problem to be solved. Summary of the invention

[0004] The object of the present invention is to provide an aerodynamic simulation and damage assessment method based on ACMI technology to improve the accuracy of damage assessment.

[0005] According to an embodiment of the present invention, an aerodynamic simulation and damage assessment method based on ACMI technology is provided, and the method comprises the following steps:

[0006] S100, obtaining a historical data set D collected by ACMI; D={D 1 ,D 2 ,…,D i ,…,D n}, D i is the i-th historical data group collected by ACMI, the value of i ranges from 1 to n, and n is the number of historical data groups collected by ACMI; D i =(d i,1 ,d i,2,…,d i,j ,…,d i,m ), d i,j D i The historical data of the j-th preset acquisition parameter is included, where the value range of j is 1 to m, and m is the number of preset acquisition parameters; any preset acquisition parameter belongs to the first category of acquisition parameters or the second category of acquisition parameters.

[0007] S200, traverse D, according to D i Get the i-th training sample and the i-th label B i ; The i-th training sample includes D i The corresponding first acquisition parameter vector F i and D i The corresponding second acquisition parameter vector R i , F i =(f i,1 ,f i,2 ,…,f i,x ,…,f i,u ), f i,x D i The historical data of the xth first-category acquisition parameter in R, where the value range of x is 1 to u, and u is the number of first-category acquisition parameters; i =(r i,1 ,r i,2 ,…,r i,y ,…,r i,m-u ), r i,y D i The historical data of the yth second-category acquisition parameter in B, where the value range of y is 1 to mu; i =(b i,1 ,b i,2 ,…,b i,x ,…,b i,u ), b i,x D i The adjustment amount of the historical data of the xth first-class acquisition parameter in (f i,1 +b i,1 ,f i,2 +b i,2 ,…,f i,x +b i,x ,…,f i,u +b i,u ) is input into the damage assessment system to obtain the predicted damage degree D i The corresponding actual degree of damage.

[0008] S300, training a neural network model according to the i-th training sample and the i-th label to obtain a trained neural network model; the trained neural network model is used to obtain labels corresponding to the input first acquisition parameter vector and the input second acquisition parameter vector.

[0009] S400, obtaining a label corresponding to the target data group according to the target data group collected by ACMI and the trained neural network model.

[0010] S500, acquiring a real damage degree corresponding to the target data group according to a first acquisition parameter vector corresponding to the target data group, a label corresponding to the target data group, and a damage assessment system.

[0011] Compared with the prior art, the present invention has at least the following beneficial effects:

[0012] The present invention divides the parameters collected by ACMI into at least two categories, namely, first-category collection parameters and second-category collection parameters, wherein the first-category collection parameters correspond to parameters that need to be input into the damage assessment system, and the second-category collection parameters are parameters that do not need to be input into the damage assessment system and may affect the degree of damage; the present invention constructs a first collection parameter vector corresponding to any historical data group according to the data of the first-category collection parameters in the historical data group, and the data in the first collection parameter vector are data that need to be input into the damage assessment system; the present invention constructs a second collection parameter vector corresponding to the historical data group according to the data of the second-category collection parameters in any historical data group, and the data in the second collection parameter vector are data that do not need to be input into the damage assessment system; the first collection parameter vector and the second collection parameter vector corresponding to any historical data group can jointly characterize the flight scene corresponding to the historical data group, and different flights The influence of the second type of acquisition parameters on the degree of damage is different in different scenarios. The present invention trains the neural network model by labeling each historical data group so that the trained neural network model has the ability to infer labels, that is, it has the ability to infer and correct the data of the first type of acquisition parameters (the damage degree predicted by inputting the corrected first type of acquisition parameter data into the damage assessment system is more accurate); the present invention obtains the corrected first acquisition parameter vector corresponding to the target data group based on the trained neural network model. Compared with directly inputting the first acquisition parameter vector corresponding to the target data group into the damage assessment system, the present invention inputs the corrected first acquisition parameter vector corresponding to the target data group into the damage assessment system. The predicted damage degree is closer to the actual damage degree corresponding to the target data group, which greatly improves the accuracy of the damage degree assessment corresponding to the target data group. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 A flow chart of an aerodynamic simulation and damage assessment method based on ACMI technology provided in an embodiment of the present invention;

[0015] Figure 2 A flowchart of a process for obtaining preset acquisition parameters according to an embodiment of the present invention;

[0016] Figure 3 The i-th label B in one embodiment of the present invention i Flow chart of the acquisition process;

[0017] Figure 4 A flowchart of obtaining a label corresponding to a target data group according to an embodiment of the present invention;

[0018] Figure 5 A flowchart of obtaining the actual damage degree corresponding to a target data group according to an embodiment of the present invention;

[0019] Figure 6 A flowchart of performing aerodynamic simulation according to a target data set collected by ACMI according to an embodiment of the present invention;

[0020] Figure 7 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] The ACMI-based aerodynamic simulation and damage assessment method provided in the embodiment of the present application can obtain the i-th training sample and the i-th label B by obtaining the historical data set set D collected by ACMI and then traversing the historical data set set D. i, then train the neural network model according to the i-th training sample and the i-th label to obtain the trained neural network model, and obtain the label corresponding to the target data group according to the target data group collected by ACMI and the trained neural network model. In this way, the actual damage degree corresponding to the target data group can be obtained according to the first acquisition parameter vector corresponding to the target data group, the label corresponding to the target data group and the damage assessment system, thereby greatly improving the accuracy of the damage degree assessment corresponding to the target data group.

[0023] According to the embodiments of the present application, Figure 1 As shown, an aerodynamic simulation and damage assessment method based on ACMI technology is provided, and the method comprises the following steps:

[0024] S100, obtaining a historical data set D collected by ACMI; D={D 1 ,D 2 ,…,D i ,…,D n}, D i is the i-th historical data group collected by ACMI, the value of i ranges from 1 to n, and n is the number of historical data groups collected by ACMI; D i =(d i,1 ,d i,2 ,…,d i,j ,…,d i,m ), d i,j D i The historical data of the j-th preset acquisition parameter is included, where the value range of j is 1 to m, and m is the number of preset acquisition parameters; any preset acquisition parameter belongs to the first category of acquisition parameters or the second category of acquisition parameters.

[0025] In this embodiment, different historical data groups are collected by ACMI in different flight scenarios. ACMI collects many parameters in different flight scenarios. It is possible to know whether any parameter is a preset collection parameter. It is also possible to know whether any parameter is a first-category collection parameter or a second-category collection parameter under the premise that it is a preset collection parameter.

[0026] As an optional specific implementation, any parameter collected by ACMI is determined as a preset collection parameter.

[0027] As a preferred specific embodiment, Figure 2 As shown, the process of obtaining the preset acquisition parameters includes:

[0028] S010, obtain the initial parameter set P collected by ACMI, P={p 1 ,p 2 ,…,p k ,…,p z}, pk is the kth initial parameter collected by ACMI, the value range of k is 1 to z, z is the number of initial parameters collected by ACMI, z>m.

[0029] In this embodiment, the "initial" in the initial parameter is only used to distinguish the preset acquisition parameters. Any parameter collected by ACMI is an initial parameter, but not all parameters collected by ACMI are preset acquisition parameters.

[0030] S020, traverse P and obtain p k Pearson correlation coefficient with damage degree k .

[0031] Those skilled in the art know that any method for obtaining the Pearson correlation coefficient in the prior art falls within the protection scope of the present invention.

[0032] S030, if |t k |≥w 1 , then determine p k is the preset acquisition parameter; otherwise, enter S040; w 1 is the first preset value, 0 <w 1 <1.

[0033] In this embodiment, w 1 is the empirical value. If |t k |≥w 1 , indicating p k There may be a greater or lesser impact on the extent of damage.

[0034] S040, judge p k Is it a preset parameter input to the damage assessment system? If so, determine whether p k is the preset acquisition parameter; otherwise, determine p k There are no preset collection parameters.

[0035] In this embodiment, the parameters input into the damage assessment system are preset by the damage assessment system and are knowable.

[0036] Based on S010-S040, parameters that have little effect on the degree of damage and do not need to be input into the damage assessment model can be screened out, reducing the dimension of the second acquisition parameter vector, which is beneficial to reducing the complexity of subsequent neural network model reasoning and improving the accuracy of neural network model reasoning.

[0037] In this embodiment, the process of obtaining the first type of acquisition parameters and the second type of acquisition parameters includes: determining the preset parameters input into the damage assessment system as the first type of acquisition parameters, and determining the parameters other than the first type of acquisition parameters in the preset acquisition parameters as the second type of acquisition parameters.

[0038] S200, traverse D, according to D i Get the i-th training sample and the i-th label B i ; The i-th training sample includes D i The corresponding first acquisition parameter vector F i and D i The corresponding second acquisition parameter vector R i , F i =(f i,1 ,f i,2 ,…,f i,x ,…,f i,u ), f i,x D i The historical data of the xth first-category acquisition parameter in R, where the value range of x is 1 to u, and u is the number of first-category acquisition parameters; i =(r i,1 ,r i,2 ,…,r i,y ,…,r i,m-u ), r i,y D i The historical data of the yth second-category acquisition parameter in B, where the value range of y is 1 to mu; i =(b i,1 ,b i,2 ,…,b i,x ,…,b i,u ), b i,x D i The adjustment amount of the historical data of the xth first-class acquisition parameter in (f i,1 +b i,1 ,f i,2 +b i,2 ,…,f i,x +b i,x ,…,f i,u +b i,u ) is input into the damage assessment system to obtain the predicted damage degree D i The corresponding actual degree of damage.

[0039] In this embodiment, the value range of i is 1 to n, so after executing S200, n training samples and n labels can be constructed.

[0040] As an optional specific implementation, B i Determined by experience, and the B determined by experience i With F i Add the summed vector and input it into the damage assessment system. If the predicted damage level output by the damage assessment system is D i The corresponding actual degree of damage is determined by experience. iis accurate, otherwise, adjust B again based on experience i , and the adjusted B i With F i Add the summed vector and input it into the damage assessment system. If the predicted damage level output by the damage assessment system is D i The corresponding actual degree of damage is then used to determine the adjusted B i To be accurate, the adjusted B i Determined as the final B i .

[0041] As a preferred specific embodiment, Figure 3 As shown, B i The acquisition process includes:

[0042] S210, traverse D, and D i The corresponding first acquisition parameter vector F i Input to the damage assessment system to obtain D i The corresponding predicted damage level.

[0043] Those skilled in the art are aware that any damage assessment system in the prior art falls within the protection scope of the present invention.

[0044] S220, traverse D, if there is a corresponding predicted damage level G in D i , then append the historical data set to the i-th set H i ; G i D i The corresponding actual damage level, H i is initialized to an empty collection.

[0045] In this embodiment, after executing S220, if there is no corresponding predicted damage level G in D, i historical data set, then H i is an empty set; if there is a corresponding predicted damage level G in D i historical data set, then H i Including all corresponding predicted damage levels in D is G i Historical data set.

[0046] S230, if H i is not an empty set, and H i If the number of historical data groups included is greater than or equal to 2, then H i The first acquisition parameter vector corresponding to each historical data group in D i The similarity of the first acquisition parameter vector corresponding to the maximum similarity is determined as D i The corresponding specified data group; if Hi is not an empty set, and H i If the number of historical data groups included is equal to 1, then H i The historical data group included is identified as D i The corresponding specified data group.

[0047] In this embodiment, if H i If it is an empty set, then determine B based on experience i , the B determined based on experience i With F i Add the summed vector and input it into the damage assessment system. If the summed vector is equal to D i The similarity of the corresponding first acquisition parameter vector is greater than or equal to the preset similarity threshold, and the predicted damage degree output by the damage assessment system is D i The corresponding actual degree of damage is determined by experience. i is accurate, otherwise, adjust B again based on experience i , and the adjusted B i With F i Add the summed vector and input it into the damage assessment system. If the summed vector is equal to D i The similarity of the corresponding first acquisition parameter vector is greater than or equal to the preset similarity threshold, and the predicted damage degree output by the damage assessment system is D i The corresponding actual degree of damage is then used to determine the adjusted B i To be accurate, the adjusted B i Determined as the final B i .

[0048] In this embodiment, H i When the number of historical data groups included is greater than or equal to 2, the historical data group corresponding to the maximum similarity is determined as D i The corresponding designated data group is conducive to constructing a label with a smaller correction amplitude for the first acquisition parameter vector for each training sample, which is conducive to accelerating the convergence process of the neural network model and improving the accuracy of the neural network model reasoning. Those skilled in the art know that any method of obtaining vector similarity in the prior art falls within the protection scope of the present invention.

[0049] S240, if D i The first acquisition parameter vector corresponding to the specified data group and D i If the similarity of the corresponding first acquisition parameter vector is greater than or equal to the preset similarity threshold, then D i The first acquisition parameter vector F' corresponding to the specified data group i .

[0050] In this embodiment, the preset similarity threshold is an empirical value, and the preset similarity threshold satisfies the condition of being greater than 0 and less than 1.

[0051] In this embodiment, different flight scenarios may correspond to the same actual damage degree. Although the second type of acquisition parameters have a certain influence on the actual damage degree, the influence is smaller than the influence of the first type of acquisition parameters on the actual damage degree. i The first acquisition parameter vector corresponding to the specified data group and D i If the similarity of the corresponding first acquisition parameter vector is greater than or equal to the preset similarity threshold, then it is determined that D i The flight scene corresponding to the specified data group is i The corresponding flight scenes are slightly different. i The first acquisition parameter vector F' corresponding to the specified data group i As D i The corrected vector of the corresponding first acquisition parameter vector is conducive to the neural network model to quickly extract the rules and characteristics of the relatively narrow range of values ​​of each value in the label, which is conducive to the rapid convergence of the neural network model and improve the accuracy of the neural network model reasoning.

[0052] In this embodiment, if D i The first acquisition parameter vector corresponding to the specified data group and D i The similarity of the corresponding first acquisition parameter vector is less than the preset similarity threshold, then it is determined that D i The flight scene corresponding to the specified data group is i The corresponding flight scenarios are quite different. In this case, if D i The first acquisition parameter vector F' corresponding to the specified data group i As D i The corrected vector of the corresponding first acquisition parameter vector, that is, D i The first acquisition parameter vector F' corresponding to the specified data group i With D i The difference of the first acquisition parameter vector corresponding to the i-th training sample will make the range of values ​​in the label wider, which is not conducive to the rapid convergence of the neural network model and the accuracy of the neural network model reasoning; in this case, B is determined based on experience i , the B determined based on experience i With F i Add the summed vector and input it into the damage assessment system. If the summed vector is equal to D i The similarity of the corresponding first acquisition parameter vector is greater than or equal to the preset similarity threshold, and the predicted damage degree output by the damage assessment system is D iThe corresponding actual degree of damage is determined by experience. i is accurate, otherwise, adjust B again based on experience i , and the adjusted B i With F i Add the summed vector and input it into the damage assessment system. If the summed vector is equal to D i The similarity of the corresponding first acquisition parameter vector is greater than or equal to the preset similarity threshold, and the predicted damage degree output by the damage assessment system is D i The corresponding actual degree of damage is then used to determine the adjusted B i To be accurate, the adjusted B i Determined as the final B i .

[0053] S250, F' i -F i Determined as B i .

[0054] Based on S210-S250, the automatic acquisition process of tags can be realized, thereby improving the efficiency of tag acquisition.

[0055] S300, training a neural network model according to the i-th training sample and the i-th label to obtain a trained neural network model; the trained neural network model is used to obtain labels corresponding to the input first acquisition parameter vector and the input second acquisition parameter vector.

[0056] In this embodiment, the value range of i is 1 to n, so the neural network model is trained using n training samples and n labels.

[0057] As an optional specific implementation, the neural network model is a multi-layer perceptron (MLP), and the D i The corresponding first acquisition parameter vector F i and D i The corresponding second acquisition parameter vector R i Perform head-to-tail splicing to obtain the input vector (f i,1 ,f i,2 ,…,f i,x ,…,f i,u ,r i,1 ,r i,2 ,…,r i,y ,…,r i,m-u), a similar method can be used to obtain the input vector corresponding to each training sample. During the training process, the input vectors corresponding to all training samples are organized into an array of shape (n, m), where the i-th row corresponds to the input vector of the i-th training sample; the labels of all training samples are organized into an array of shape (n, u), where the i-th row corresponds to the label of the i-th training sample; then, the fit method is used to train the model, that is, the parameters of the neural network model are continuously updated using the training data in an iterative manner, including the process of setting the number of training rounds (epochs), batch size (batch size) and validation set ratio (validation split). epochs, batch size and validation split are all empirical values, such as setting epochs=100, batch size=32, validation split=0.2; finally, the loss changes and other related indicators during the training process are monitored, where the loss of the neural network model is the average value of the difference between the label predicted by the neural network model and the true label, and the difference between any predicted label and the corresponding true label is the difference between 1 and the similarity between the predicted label and the corresponding true label. Optionally, the cosine similarity formula is used to obtain the similarity between any predicted label and the corresponding true label.

[0058] Those skilled in the art are aware that any method for training a neural network model in the prior art falls within the protection scope of the present invention.

[0059] S400, obtaining a label corresponding to the target data group according to the target data group collected by ACMI and the trained neural network model.

[0060] As an optional specific implementation manner, the target data group collected by ACMI includes data of all parameters collected by ACMI or only includes data of all preset collection parameters collected by ACMI.

[0061] As a specific implementation, Figure 4 As shown, step S400 includes:

[0062] S410, acquiring a first acquisition parameter vector and a second acquisition parameter vector corresponding to the target data group according to the target data group acquired by ACMI.

[0063] In this embodiment, a first acquisition parameter vector corresponding to the target data group is constructed based on the data of the first type of acquisition parameters in the target data group collected by ACMI, and a second acquisition parameter vector corresponding to the target data group is constructed based on the data of the second type of acquisition parameters in the target data group collected by ACMI.

[0064] S420: Input the first acquisition parameter vector and the second acquisition parameter vector corresponding to the target data group into the trained neural network model to obtain a label corresponding to the target data group.

[0065] Based on S410 - S420 , the label corresponding to the target data group can be obtained.

[0066] S500, acquiring a real damage degree corresponding to the target data group according to a first acquisition parameter vector corresponding to the target data group, a label corresponding to the target data group, and a damage assessment system.

[0067] As a specific implementation, Figure 5 As shown, step S500 includes:

[0068] S510: Determine the sum of the first acquisition parameter vector corresponding to the target data group and the label corresponding to the target data group as the modified first acquisition parameter vector corresponding to the target data group.

[0069] S520, inputting the corrected first acquisition parameter vector corresponding to the target data group into the damage assessment system, and determining the predicted damage degree output by the damage assessment system as the actual damage degree corresponding to the target data group.

[0070] Based on S510-S520, the actual damage degree corresponding to the target data group can be obtained.

[0071] The present embodiment divides the parameters collected by ACMI into at least two categories, namely, first-category collection parameters and second-category collection parameters, wherein the first-category collection parameters correspond to parameters that need to be input into the damage assessment system, and the second-category collection parameters are parameters that do not need to be input into the damage assessment system and may affect the degree of damage; the present embodiment constructs a first collection parameter vector corresponding to any historical data group based on the data of the first-category collection parameters in the historical data group, and the data in the first collection parameter vector are data that need to be input into the damage assessment system; the second collection parameter vector corresponding to any historical data group is constructed based on the data of the second-category collection parameters in the historical data group, and the data in the second collection parameter vector are data that do not need to be input into the damage assessment system; the first collection parameter vector and the second collection parameter vector corresponding to any historical data group can together characterize the flight scene corresponding to the historical data group, and different The impact of the second type of acquisition parameters on the degree of damage in the flight scenario is different. This embodiment trains the neural network model by labeling each historical data group so that the trained neural network model has the ability to infer labels, that is, it has the ability to infer and correct the data of the first type of acquisition parameters (the damage degree predicted by inputting the corrected first type of acquisition parameter data into the damage assessment system is more accurate); this embodiment obtains the corrected first acquisition parameter vector corresponding to the target data group based on the trained neural network model. Compared with directly inputting the first acquisition parameter vector corresponding to the target data group into the damage assessment system, this embodiment inputs the corrected first acquisition parameter vector corresponding to the target data group into the damage assessment system. The predicted damage degree is closer to the actual damage degree corresponding to the target data group, thereby improving the accuracy of the damage degree assessment corresponding to the target data group.

[0072] As an optional specific implementation, the method further includes the following steps:

[0073] S600, performing aerodynamic simulation based on the target data set collected by ACMI.

[0074] As an optional specific implementation, Figure 6 As shown, step S600 includes:

[0075] S610, acquiring data of parameters related to aerodynamic simulation in the target data group collected by ACMI.

[0076] In this embodiment, the parameters related to the aerodynamic simulation include airspeed, flight altitude, angle of attack, sideslip angle, aircraft attitude, acceleration, temperature and pressure, etc.

[0077] S620, inputting the data of the parameters related to the pneumatic simulation in the target data group collected by ACMI into the pneumatic simulation system to obtain the pneumatic simulation results.

[0078] See also Figure 7 , Figure 7 is a schematic diagram of the structure of a computer device provided in an embodiment of the present application, such as Figure 7 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 7 A processor 10 is taken as an example.

[0079] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0080] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the aerodynamic simulation and damage assessment method based on ACMI technology shown in the above embodiment.

[0081] The memory 20 may 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; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0082] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0083] like Figure 7 As shown, the computer device may further include an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 7 The example of connecting through bus is taken in the following.

[0084] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device may be a touch screen.

[0085] In addition, the embodiment of the present application also provides a computer-readable storage medium. The aerodynamic simulation and damage assessment method based on ACMI technology according to the embodiment of the present application can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be downloaded through a network and originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the aerodynamic simulation and damage assessment method based on ACMI technology shown in the above embodiment is implemented.

[0086] It is understandable that the embodiments of the present application may also provide a computer program product, which includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the aerodynamic simulation and damage assessment method based on ACMI technology described in any embodiment of the present application.

[0087] Those skilled in the art know that any data of parameters related to pneumatic simulation collected by ACMI and methods of performing pneumatic simulation by a pneumatic simulation system in the prior art fall within the protection scope of the present invention. As an optional specific embodiment, the pneumatic simulation system is CFD software, pneumatic analysis software, comprehensive simulation tool or professional aviation simulation tool.

[0088] Although some specific embodiments of the present invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. An aerodynamic simulation and damage assessment method based on ACMI technology, characterized in that: The method comprises the following steps: S100, obtain the historical data set D collected by ACMI; D={D1,D2,…,D i ,…,D n }, D i is the i-th historical data group collected by ACMI, the value range of i is 1 to n, and n is the number of historical data groups collected by ACMI; D i =(d i,1 ,d i,2 ,…,d i,j ,…,d i,m ), d i,j D i The historical data of the j-th preset acquisition parameter is included, where the value of j ranges from 1 to m, and m is the number of preset acquisition parameters; any preset acquisition parameter belongs to the first type of acquisition parameter or the second type of acquisition parameter; wherein the preset parameter input to the damage assessment system is determined to be the first type of acquisition parameter, and the parameters other than the first type of acquisition parameters in the preset acquisition parameters are determined to be the second type of acquisition parameters; S200, traverse D, according to D i Get the i-th training sample and the i-th label B i ; The i-th training sample includes D i The corresponding first acquisition parameter vector F i and D i The corresponding second acquisition parameter vector R i , F i =(f i,1 ,f i,2 ,…,f i,x ,…,f i,u ), f i,x D i The historical data of the xth first-category acquisition parameter in R, where the value range of x is 1 to u, and u is the number of first-category acquisition parameters; i =(r i,1 ,r i,2 ,…,r i,y ,…,r i,m-u ), r i,y D i The historical data of the yth second-category acquisition parameter in B, where the value range of y is 1 to mu; i =(b i,1 ,b i,2 ,…,b i,x ,…,b i,u ), b i,x D i The adjustment amount of the historical data of the xth first-class acquisition parameter in (f i,1 +b i,1 ,f i,2 +b i,2 ,…,f i,x +b i,x ,…,f i,u +b i,u ) is input into the damage assessment system to obtain the predicted damage degree D i The corresponding actual degree of damage; S300, training a neural network model according to the i-th training sample and the i-th label to obtain a trained neural network model; the trained neural network model is used to obtain labels corresponding to the input first acquisition parameter vector and the input second acquisition parameter vector; S400, obtaining a label corresponding to the target data group according to the target data group collected by ACMI and the trained neural network model; S500, acquiring a real damage degree corresponding to the target data group according to a first acquisition parameter vector corresponding to the target data group, a label corresponding to the target data group, and a damage assessment system; B i The acquisition process includes: S210, traverse D, and D i The corresponding first acquisition parameter vector F i Input to the damage assessment system to obtain D i The corresponding predicted damage level; S220, traverse D, if there is a corresponding predicted damage level G in D i , then append the historical data set to the i-th set H i ; G i D i The corresponding actual damage level, H i is initialized to an empty set; S230, if H i is not an empty set, and H i If the number of historical data groups included is greater than or equal to 2, then H i The first acquisition parameter vector corresponding to each historical data group in D i The similarity of the first acquisition parameter vector corresponding to the maximum similarity is determined as D i The corresponding specified data group; if H i is not an empty set, and H i If the number of historical data groups included is equal to 1, then H i The historical data group included is identified as D i The corresponding specified data group; S240, if D i The first acquisition parameter vector corresponding to the specified data group and D i If the similarity of the corresponding first acquisition parameter vector is greater than or equal to the preset similarity threshold, then D i The first acquisition parameter vector F' corresponding to the specified data group i ; S250, F' i -F i Determined as B i .

2. The ACMI-based aerodynamic simulation and damage assessment method according to claim 1, characterized in that: The process of obtaining preset acquisition parameters includes: S010, obtain the initial parameter set P collected by ACMI, P={p1,p2,…,p k ,…,p z }, p k is the kth initial parameter collected by ACMI, the value range of k is 1 to z, z is the number of initial parameters collected by ACMI, z>m; S020, traverse P and obtain p k Pearson correlation coefficient with damage degree k ; S030, if |t k |≥w1, then determine p k is the preset acquisition parameter; otherwise, enter S040; w1 is the first preset value, 0 <w1<1; S040, judge p k Is it a preset parameter input to the damage assessment system? If so, determine whether p k is the preset acquisition parameter; otherwise, determine p k There are no preset collection parameters.

3. The ACMI-based aerodynamic simulation and damage assessment method according to claim 1, characterized in that: S400 includes: S410, acquiring a first acquisition parameter vector and a second acquisition parameter vector corresponding to the target data group according to the target data group acquired by ACMI; S420: Input the first acquisition parameter vector and the second acquisition parameter vector corresponding to the target data group into the trained neural network model to obtain a label corresponding to the target data group.

4. The ACMI-based aerodynamic simulation and damage assessment method according to claim 1, characterized in that: S500 includes: S510, determining the sum of the first acquisition parameter vector corresponding to the target data group and the label corresponding to the target data group as the modified first acquisition parameter vector corresponding to the target data group; S520, inputting the corrected first acquisition parameter vector corresponding to the target data group into the damage assessment system, and determining the predicted damage degree output by the damage assessment system as the actual damage degree corresponding to the target data group.

5. The ACMI-based aerodynamic simulation and damage assessment method according to claim 1, characterized in that: The method further comprises the following steps: S600, performing aerodynamic simulation based on the target data set collected by ACMI.

6. The ACMI-based aerodynamic simulation and damage assessment method according to claim 5, characterized in that: S600 includes: S610, acquiring data of parameters related to aerodynamic simulation in the target data group collected by ACMI; S620, inputting the data of the parameters related to the pneumatic simulation in the target data group collected by ACMI into the pneumatic simulation system to obtain the pneumatic simulation results.

7. The ACMI-based aerodynamic simulation and damage assessment method according to claim 1, characterized in that: Parameters relevant to aerodynamic simulation include airspeed, flight altitude, angle of attack, sideslip angle, vehicle attitude, acceleration, temperature and pressure.

8. The ACMI-based aerodynamic simulation and damage assessment method according to claim 1, characterized in that: ACMI stands for Air Combat Maneuvering Instrument System.

Images