An aircraft airborne component heating prediction method, device, and storage medium

Abstract

The present application relates to a method, device, and storage medium for predicting the heat generation of aircraft airborne components, belonging to the field of aerospace technology. By collecting the original historical data set of airborne components; using an improved Latin hypercube sampling method to sample the original historical data set to obtain training data samples; performing computational fluid dynamics simulations based on the training data samples to obtain the calculation results of the thermal characteristics of airborne components under different working conditions; using the attribute information and flight condition information of aircraft airborne components as input data, and using the calculation results of the thermal characteristics as output data to train a convolutional neural network model to generate a thermal prediction neural network model; this thermal prediction neural network model is used to perform thermal prediction according to new input data and generate the temperature distribution data of aircraft airborne components at the prediction moment. The present application improves the uniformity and comprehensiveness of sampling and can achieve real-time and full-process prediction of the thermal characteristics of aircraft airborne components.

Description

Technical Field

[0001] This application belongs to the field of aerospace technology, and more particularly relates to a method, device, and storage medium for predicting the heat generation of aircraft-borne components. Background Art

[0002] In many industrial applications, electronic components often face the problem of heat generation and temperature rise under harsh environments or long-term usage conditions, which may lead to performance degradation or device damage. Heat dissipation devices can solve this problem to a certain extent, but at the same time, the volume, mass, power consumption, and cost will also increase significantly. Therefore, it is necessary to consider the specific application objects and usage environments of the components and predict the heat generation process and characteristics accordingly. The prediction results can help improve the process during component design or assist in designing variable-frequency, low-power, and low-cost heat dissipation supporting devices.

[0003] However, in the prior art, the heat generation prediction for electronic components, especially aircraft-borne electronic components, mainly relies on environmental pressure tests. This test method usually only considers extreme conditions or a few typical working conditions, and its main purpose is to ensure that the components are not damaged. However, these tests are usually not sufficient to comprehensively reflect the temperature changes and thermal stresses of the components during actual use, and less attention is paid to the real-time temperature changes, heat dissipation power consumption, and long-term service life of the components.

[0004] Therefore, how to more comprehensively reflect the heat generation situation during the entire mission cycle is a research topic worthy of attention in this field. Summary of the Invention

[0005] In view of the above analysis, the embodiments of the present invention aim to provide a method, device, and storage medium for predicting the heat generation of aircraft-borne components to solve the problem that the prior art cannot reflect the heat generation situation during the entire mission cycle.

[0006] In the first aspect of this application, a method for predicting the heat generation of aircraft-borne components is provided, including:

[0007] Collect the original historical data set of aircraft-borne components;

[0008] Use the improved Latin hypercube sampling method to sample the original historical data set to obtain training data samples related to the heat generation prediction task;

[0009] Based on the training data samples, perform computational fluid dynamics simulations to obtain the thermal characteristic calculation results of aircraft-borne components under different working conditions;

[0010] Use the attribute information and flight condition information of aircraft-borne components as input data, and the thermal characteristic calculation results as output data, and train a convolutional neural network model based on the training data samples to generate a heat prediction neural network model;

[0011] The thermal prediction neural network model is used to perform thermal prediction based on new input data. This process includes:

[0012] Obtaining the attribute information of the aircraft's on-board components;

[0013] Obtaining in real time the flight condition information of the aircraft at the prediction moment;

[0014] Inputting the attribute information and the flight condition information into the thermal prediction neural network model to generate the temperature distribution data of the aircraft's on-board components at the prediction moment.

[0015] Optionally, the improved Latin hypercube sampling method is used to sample the original historical data set to obtain training data samples related to the heat prediction task, including:

[0016] Assuming that the sampling object is N-dimensional and the sampling times are M times, and uniform sampling from 0 to 1 is performed on each dimension, then the sampling process can be described as an N×M matrix, denoted as A; the improved Latin hypercube sampling method includes the following steps:

[0017] S1: Divide each dimension into M equal parts, that is, the length of each interval is 1 / M;

[0018] S2: Randomly draw N values from the first sub-interval [0, 1 / M] as the first column of matrix A;

[0019] S3: Randomly draw N values from the second sub-interval [1 / M, 2 / M] as the second column of matrix A;

[0020] S4: And so on until reaching the last sub-interval and forming the Nth column of matrix A;

[0021]

[0022] where P 11 , P 12 …, P 1M , P 21 …, P NM are the sampling values;

[0023] S5: Shuffle the order of each row of A to obtain matrix B. Finally, each column of matrix B represents a sample point, and there are M sample points;

[0024] S6: Use the parameter G max to evaluate whether the distribution of the sample points in matrix B is reasonable. If it is not reasonable, return for resampling; where G max is a parameter used to evaluate the uniform distribution degree of the samples in the multi-dimensional parameter space.

[0025] Optionally, the parameter G max is calculated according to the sampling dimension N and the number of sampling times M:

[0026] lnG max =(395.7N 0.0048 -394.1)-(0.24N -0.94 +1.03)lnM.

[0027] Optionally, the convolutional neural network includes a component attribute encoder, a flight condition encoder, and a thermal calculation decoder;

[0028] Among them, the component attribute encoder is used to obtain the attribute information of the aircraft's on-board components; the flight condition encoder is used to obtain the flight condition information of the aircraft; the thermal calculation decoder is used to sample the original historical data set by using an improved Latin hypercube sampling method to obtain training data samples related to the heat generation prediction task; based on the training data samples, hydrodynamic simulation calculations are performed to obtain the thermal characteristic calculation results of the on-board components under different working conditions.

[0029] Optionally, the attribute information of the aircraft's on-board components includes the component's material information, volume information, and shape information; the flight condition information includes flight altitude information, speed information, overload information, and air temperature information.

[0030] Optionally, the convolutional neural network adopts the form of a U-Net architecture, and the U-Net includes five channels:

[0031] The first channel: used to input the component attribute information encoded by the component attribute encoder;

[0032] The second channel: used to input the flight altitude information encoded by the component attribute encoder;

[0033] The third channel: used to input the speed information encoded by the component attribute encoder;

[0034] The fourth channel: used to input the overload information encoded by the component attribute encoder;

[0035] The fifth channel: used to input the air temperature information encoded by the component attribute encoder;

[0036] Taking the component attribute encoder and the flight condition encoder as inputs and the thermal calculation decoder as the output, the U-Net is trained to generate a thermal prediction neural network model.

[0037] Optionally, the hydrodynamic simulation calculations based on the training data samples to obtain the thermal characteristic calculation results of the on-board components under different working conditions include:

[0038] Collect the operating conditions of airborne components and determine the thermal characteristics to be studied;

[0039] Construct a geometric model of the airborne component so that the model reflects the shape and size of the actual component;

[0040] Generate a computational grid on the geometric model and set boundary conditions according to the actual working conditions and environment;

[0041] Select the corresponding hydrodynamic and heat transfer models for simulation calculation to obtain the calculation results of the thermal characteristics of the airborne component under different working conditions.

[0042] Optionally, the airborne component is an airborne radio altimeter.

[0043] In a second aspect of the present application, there is provided a device for predicting the heat generation of an aircraft airborne component, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it realizes the method for predicting the heat generation of an aircraft airborne component according to any one of the above.

[0044] In a third aspect of the present application, there is provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it realizes the method for predicting the heat generation of an aircraft airborne component according to any one of the above.

[0045] The method for predicting the heat generation of an aircraft airborne component provided by the present application collects the original historical data set of the airborne component; uses an improved Latin hypercube sampling method to sample the original historical data set to obtain training data samples related to the heat generation prediction task; performs hydrodynamic simulation calculations based on the training data samples to obtain the calculation results of the thermal characteristics of the airborne component under different working conditions; uses the attribute information and flight condition information of the aircraft airborne component as input data, and uses the calculation results of the thermal characteristics as output data, and trains a convolutional neural network model based on the training data samples to generate a thermal prediction neural network model; this thermal prediction neural network model is used to perform thermal prediction according to new input data and generate the temperature distribution data of the aircraft airborne component at the prediction moment. By using the improved Latin hypercube sampling method, the present application improves the uniformity and comprehensiveness of sampling, and then extracts the key factors affecting the heat generation of the component. Through the established thermal prediction neural network model, it is possible to realize real-time and full-process prediction of the thermal characteristics of aircraft airborne components.

[0046] In addition, the present application also provides an aircraft airborne component heat generation prediction device and a storage medium having the above technical effects. Description of the Drawings

[0047] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments described in the embodiments of this specification. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0048] Figure 1 It is a flowchart of a specific implementation manner of the aircraft airborne component heating prediction method provided by this application;

[0049] Figure 2 It is a schematic diagram of the convolutional neural network model structure of this application;

[0050] Figure 3 It is a schematic diagram of the U-Net model structure of this application;

[0051] Figure 4 It is a schematic diagram of the process of the thermal prediction neural network model of this application for thermal prediction based on new input data;

[0052] Figure 5 It is a block diagram of the structure of the aircraft airborne component heating prediction device provided by this application. Specific implementation manner

[0053] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. It should be noted that, without conflict, the implementation manners and the features in the implementation manners in this disclosure can be combined, separated, interchanged, and / or rearranged. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0054] The terms used herein are for the purpose of describing particular embodiments and are not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are also intended to include the plural forms. In addition, when the terms "comprising" and / or "including" and their variants are used in this specification, it is stated that there are the stated features, integers, steps, operations, components, assemblies, and / or groups thereof, but does not preclude the presence or addition of one or more other features, integers, steps, operations, components, assemblies, and / or groups thereof. It should also be noted that, as used herein, the terms "substantially", "about" and other similar terms are used as approximate terms and not as terms of degree, and thus they are used to explain the inherent deviations of measured, calculated and / or provided values that would be recognized by those of ordinary skill in the art.

[0055] A flowchart of a specific embodiment of the method for predicting the heat generation of aircraft airborne components provided by this application is as Figure 1 shown, and the method includes:

[0056] S101: Collect the original historical data set of airborne components.

[0057] Among them, the airborne component can be an airborne radio altimeter, and of course it can also be other aircraft airborne components, which are not limited herein. The original historical data set is various performance and environmental data under actual operating conditions. These data include but are not limited to the following types:

[0058] (1) Operating condition data:

[0059] Flight altitude: The operating data of the component at different altitudes;

[0060] Flight speed: The influence of different speeds on the performance of the component.

[0061] Ambient temperature: The operating data of the component under various external temperature conditions.

[0062] (2) Component performance data

[0063] Temperature record: The temperature data of the component during operation.

[0064] Power usage: Including records of current and voltage, and these data can help evaluate power consumption and heat generation.

[0065] Fault record: Record the situation where the component shows a decrease in performance or a fault under specific conditions.

[0066] (3) Physical and mechanical data:

[0067] Overload data: Record the physical stress or overload situation suffered by the component during flight.

[0068] Vibration data: The vibration level that the component experiences during flight, which may affect its heat generation and performance.

[0069] (4) Environmental data:

[0070] Air pressure data: The air pressure data at different flight altitudes, which affects the heat dissipation performance of the component.

[0071] Humidity data: The possible impact of environmental humidity on the performance of the component.

[0072] It can be understood that the above data can be collected through various types of airborne sensors with multiple dimensions. By collecting these data, the behavior patterns and performance of the component under various actual working conditions can be comprehensively understood, preparing for the next sampling step.

[0073] S102: Sample the original historical data set using an improved Latin hypercube sampling method to obtain training data samples related to the heat generation prediction task.

[0074] The heat generation process of the airborne altimeter is related to many factors such as flight altitude, speed, and even installation location. This step aims to select representative and useful samples for the heat generation prediction task from the extensive data set obtained in S101. Identify the most critical factors among them, ignore the secondary factors, and then establish a heat prediction neural network. A specific implementation method is to conduct sensitivity analysis, and the premise is sampling and experiments.

[0075] Latin Hypercube Sampling (LHS) is a low-cost sampling method and is the premise for analyzing the sensitivity of the model output to the input. Traditional LHS cannot evenly fill the entire problem space, so this application proposes an improved LHS method.

[0076] LHS is a stratified sampling technique that divides the sampling space into different layers according to a certain characteristic or rule, and then independently and randomly samples from different layers.

[0077] Assume that the sampling object is N-dimensional and the number of sampling times is M times. Uniform sampling from 0 to 1 is performed on each dimension. Then the sampling process can be described as an N×M matrix, denoted as A; the improved Latin hypercube sampling method includes the following steps:

[0078] S1: Divide each dimension into M equal parts, that is, the length of each interval is 1 / M.

[0079] S2: Randomly select N values from the first sub-interval [0, 1 / M] as the first column of matrix A.

[0080] S3: Randomly extract N values from the second sub - interval [1 / M, 2 / M] as the second column of matrix A.

[0081] S4: And so on until reaching the last sub - interval and forming the Nth column of matrix A;

[0082]

[0083] where P 11 , P 12 …, P 1M , P 21 …, P NM are the sampling values.

[0084] S5: Shuffle the order of each row of A to get matrix B. Finally, each column of matrix B represents a sample point, and there are M sample points in total.

[0085] S6: Use parameter G max to evaluate whether the distribution of sample points in matrix B is reasonable. If it is not reasonable, return for resampling.

[0086] where G max is a parameter used to evaluate the degree of uniform distribution of samples in the multi - dimensional parameter space. It is usually calculated based on the sample variance and sample standard deviation. Specifically, the sum of the sample variance and three times the sample standard deviation can be used as the calculation method of G max .

[0087] Define the function for evaluating the sampling effect as G:

[0088] where q represents a value in matrix B. Under a large number of experiments, the value of G approximately follows a normal distribution. The smaller G is, the more uniform the sampling is. Therefore, it is necessary to limit the upper limit of G during sampling, and thus limit the sampling effect. Denote the upper limit of G as G max .

[0089] Here, the sum of the sample variance and three times the sample standard deviation is used to represent G max . Assume N = 10, the LHS sampling is 1000 times. It is considered that usually the number of samples M should be greater than the object dimension N. Calculate the G max values for different M values as follows, where the results are reserved to two significant figures:

[0090]

[0091]

[0092] Fit the values of M and N: lnG max = q + plnM

[0093] Among them, p and q are fitting coefficients. Contrary to the above process, a set of G max and M are fixed here, and the variations of p and q with N are analyzed, where the results are retained to four significant figures:

[0094] N value p q N value p q 2 -1.154 2.863 12 -1.051 6.304 4 -1.093 4.202 14 -1.046 6.592 6 -1.069 4.954 16 -1.044 6.857 8 -1.062 5.523 18 -1.043 7.092 10 -1.058 5.971 20 -1.040 7.291

[0095] Similarly, the values of p and q are also fitted:

[0096] p = -0.24N -0.94 -1.03

[0097] q = 395.7N 0.0048 -394.1

[0098] Therefore, the parameter G max is calculated according to the sampling dimension N and the number of sampling times M:

[0099] lnG max = (395.7N 0.0048 - 394.1) - (0.24N -0.94 + 1.03)lnM.

[0100] Finally, G max is calculated according to the sampling dimension N and the number of sampling times M, and then whether the matrix B is reasonable is judged according to G max . If it is unreasonable, resampling is required.

[0101] Directly calculate G max from the sample data, which is a method for evaluating the sample quality based on the actual data, while the embodiment of the present application uses N and M to calculate G max which is a method for predicting the sample quality based on a theoretical or empirical model. Based on the statistical analysis of a large number of simulation or experimental data of LHS sampling, the relationship between N and M and G max is obtained. This method allows estimating the theoretical G max value before actually generating and evaluating the sample, which can be used to guide the sampling process and ensure that the generated sample set has sufficient uniformity and representativeness.

[0102] The improved sampling method provided by the present application generates representative samples from a large number of possible flight conditions and attribute information of airborne components. These samples represent various different flight situations and equipment states, aiming to comprehensively cover the possible operating environments. These samples not only cover a wide area of the feature space but also ensure the diversity of data, which is very important for the neural network to learn complex patterns.

[0103] The improved Latin hypercube sampling method of this application is proposed based on a large amount of previous analysis. It only needs to evaluate the sampling level according to the number of samplings and the problem dimension. In this way, the uniformity of the samples can be monitored in real time, allowing dynamic adjustment during the sampling process to improve the spatial distribution of the samples, which helps to improve the accuracy of model training and prediction.

[0104] The improved sampling method can ensure that the selected samples have good coverage and uniform distribution in each dimension (such as temperature, height, speed, etc.), so that the trained prediction model can more accurately reflect the thermal behavior of the components under various conditions. This is very important for designing more effective heat dissipation systems and predicting potential fault points in advance.

[0105] S103: Perform computational fluid dynamics simulation based on the training data samples to obtain the calculation results of the thermal characteristics of airborne components under different working conditions.

[0106] In this step, the specific process of performing computational fluid dynamics simulation based on the training data samples to obtain the calculation results of the thermal characteristics of airborne components under different working conditions includes:

[0107] Collect the operating conditions of the airborne components and determine the thermal characteristics to be studied. Collecting the operating conditions of the airborne components includes flight altitude, speed, ambient temperature, air pressure, etc. Clearly define the thermal characteristics to be studied, such as temperature distribution, heat flux, cooling efficiency, etc.

[0108] Construct a geometric model of the airborne component so that the model reflects the shape and size of the actual component. Optionally, the geometric model can also be simplified as necessary to remove details that have little impact on the flow field and thermal field to reduce the computational complexity.

[0109] Generate a computational grid on the geometric model and set boundary conditions according to the actual working conditions and environment.

[0110] Select the corresponding computational fluid dynamics and heat transfer models for simulation calculation to obtain the calculation results of the thermal characteristics of airborne components under different working conditions.

[0111] S104: Use the attribute information of the aircraft airborne components and the flight condition information as input data, and use the calculation results of the thermal characteristics as output data to train the convolutional neural network model based on the training data samples to generate a thermal prediction neural network model.

[0112] The structure of Convolutional Neural Networks (CNNs) includes convolutional layers, activation layers, pooling layers, and fully connected layers. Among them, the most important part is the convolutional layer, whose main function is to calculate matrices using convolutional kernels. For example, it can convert a 7×7 matrix convolutional layer into a 5×5 matrix, thereby accelerating the operation speed, but at the same time reducing the output dimension.

[0113] In order to maintain the size of the output matrix, this application designs a thermal prediction neural network model based on the encoder-decoder architecture of traditional CNNs and the variant U-Net model of fully convolutional neural networks.

[0114] As Figure 2 shown in the schematic diagram of the convolutional neural network model structure, the convolutional neural network includes a component attribute encoder, a flight condition encoder, and a thermal calculation decoder.

[0115] Among them, the component attribute encoder is used to obtain the attribute information of aircraft-borne components; the flight condition encoder is used to obtain the flight condition information of the aircraft; the thermal calculation decoder is used to sample the original historical data set using an improved Latin hypercube sampling method to obtain training data samples related to the heat prediction task; based on the training data samples, fluid dynamics simulation calculations are performed to obtain the thermal characteristic calculation results of aircraft-borne components under different working conditions.

[0116] The attribute information of aircraft-borne components includes component material information, volume information, and shape information; the flight condition information includes flight altitude information, speed information, overload information, and air temperature information.

[0117] Different from CNNs, U-Net replaces the fully connected layer with a convolutional layer, making the entire neural network completely composed of convolutional layers, pooling layers, and upsampling layers, so that the data or matrix only changes the resolution during the neural network operation and transmission process. Since U-Net belongs to a fully convolutional neural network, it cannot process the flight condition vector composed of flight altitude, speed, overload, air temperature, etc. like CNNs. It is considered to combine the row condition vectors into a five-channel matrix as the input of the entire neural network.

[0118] As Figure 3 shown in the U-Net model structure schematic diagram, U-Net includes five channels:

[0119] The first channel: the component attribute information encoded by the component attribute encoder;

[0120] The second channel: the flight altitude information encoded by the component attribute encoder;

[0121] The third channel: speed information encoded by the component attribute encoder;

[0122] The fourth channel: overload information encoded by the component attribute encoder;

[0123] The fifth channel: air temperature information encoded by the component attribute encoder;

[0124] Taking the component attribute encoder and the flight condition encoder as inputs and the thermal calculation decoder as the output, train a U-Net to generate a thermal prediction neural network model.

[0125] After generating the thermal prediction neural network model, the thermal prediction neural network model is used to perform thermal prediction based on new input data, as Figure 4 shown in the process diagram of the thermal prediction neural network model performing thermal prediction based on new input data. This process specifically includes:

[0126] S401: Obtain the attribute information of the aircraft's on-board components.

[0127] S402: Real-time obtain the flight condition information of the aircraft at the prediction moment.

[0128] Obtain flight condition information, such as flight altitude, speed, overload, and air temperature. Specifically, it can combine data from trajectory planning, airspace planning, and maneuver planning by integrating information from different sources, including the pre-determined flight plan, real-time flight data, and meteorological data.

[0129] Among them, trajectory planning involves determining the best path for the aircraft from the starting point to the destination. During the trajectory planning process, the ideal flight altitude and speed at different stages (such as climb, cruise, and descent) will be determined. These data are usually clearly specified in the flight plan and optimized according to flight performance, fuel efficiency, and air traffic control requirements.

[0130] Airspace planning involves managing and optimizing the flight activities of the aircraft within a specific spatial area to ensure safety and efficiency. Based on the traffic flow and restrictions in the airspace, the flight altitude and speed may need to be adjusted. Maneuvers such as sharp turns, climbs, or descents may result in instantaneous overloads. Congestion or avoidance measures in the airspace may require the aircraft to perform these maneuvers.

[0131] Maneuver planning is usually related to the performance of the aircraft and special situations that may be encountered during the flight. Specific maneuvers (such as obstacle avoidance or emergency descent) will directly affect the overload data. The pilot or the autopilot system performs these maneuvers as needed.

[0132] S403: Input the attribute information and the flight condition information into the thermal prediction neural network model to generate the temperature distribution data of the aircraft's on-board components at the prediction moment.

[0133] Specifically, according to the flight path planning, airspace planning and maneuver planning of the aircraft, as well as the material, volume and shape of the altimeter, the U-Net model can be used to predict in real time the heating temperature rise of the altimeter during flight, so as to provide a reference for the overall design of the altimeter and even the aircraft.

[0134] The convolutional neural network will change the matrix size and affect the data resolution. For this reason, the encoder-decoder model, the fully convolutional U-Net model, is proposed in this application, which can integrate component attributes, flight conditions and thermal simulation calculation results, and can be used to specifically improve the design of the airborne altimeter.

[0135] In addition, this application also provides an aircraft airborne component heating prediction device, such as Figure 5 The structural block diagram of the aircraft airborne component heating prediction device provided by this application. The device specifically includes a memory 51 and a processor 52. The memory 51 stores a computer program, and when the computer program is executed by the processor 52, it realizes the aircraft airborne component heating prediction method according to any one of the above.

[0136] In addition, this application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it realizes the aircraft airborne component heating prediction method according to any one of the above.

[0137] The computer-readable storage medium includes permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media do not include transitory media such as modulated data signals and carrier waves.

[0138] Those skilled in the art should also be able to further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0139] The steps of the methods or algorithms described in combination with the embodiments disclosed herein can be implemented by hardware, software modules executed by a processor, or a combination of both. The software modules can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0140] The specific implementation manners described above further elaborate on the purpose, technical solutions, and beneficial effects of this application. It should be understood that the above description is only the specific implementation manners of this application and is not used to limit the protection scope of this application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of this application should be included in the protection scope of this application.

Claims

1. A method for predicting the heat generation of aircraft on-board components, characterized in that, Including: Collecting the original historical data set of airborne components; Sampling the original historical data set by using an improved Latin hypercube sampling method to obtain training data samples related to the heating prediction task, including: assuming that the sampling object is N-dimensional and the sampling times are M times, and uniform sampling from 0 to 1 is performed on each dimension, then the sampling process can be described as an N×M matrix, denoted as A; the improved Latin hypercube sampling method includes the following steps: S1: Divide each dimension into M equal parts, that is, the length of each interval is 1 / M; S2: Randomly extract N values from the first sub-interval [0, 1 / M] as the first column of matrix A; S3: Randomly extract N values from the second sub-interval [1 / M, 2 / M] as the second column of matrix A; S4: And so on until reaching the last sub-interval and forming the Nth column of matrix A; Among them, P 11 , P 12 …, P 1M , P 21 …, P NM are the sampled values; S5: Shuffle the order of each row of A to obtain matrix B. Finally, each column of matrix B represents a sample point, and there are M sample points; S6: Adopt parameter G max Evaluate whether the sample point distribution of matrix B is reasonable. If it is not reasonable, return for resampling; where G max is a parameter used to evaluate the degree of uniform distribution of samples in the multi-dimensional parameter space; Performing hydrodynamic simulation calculations based on the training data samples to obtain the thermal characteristic calculation results of airborne components under different working conditions; Taking the attribute information of aircraft airborne components and the flight condition information as input data, and taking the thermal characteristic calculation results as output data, training a convolutional neural network model based on the training data samples to generate a thermal prediction neural network model; The thermal prediction neural network model is used to perform thermal prediction according to new input data, and this process includes: Obtaining the attribute information of aircraft airborne components; Obtaining the flight condition information of the aircraft in real time at the prediction moment; Inputting the attribute information and the flight condition information into the thermal prediction neural network model to generate the temperature distribution data of aircraft airborne components at the prediction moment.

2. The method for predicting the heat generation of aircraft on-board components according to claim 1, wherein The parameter G max Calculated based on the sampling dimension N and the number of sampling times M: lnG max = (395.7N 0.0048 - 394.1) - (0.24N -0.94 + 1.03) lnM。 3. The aircraft airborne component heating prediction method according to any one of claims 1 to 2, characterized in that The convolutional neural network includes a component attribute encoder, a flight condition encoder, and a thermal calculation decoder; Among them, the component attribute encoder is used to obtain the attribute information of aircraft airborne components; the flight condition encoder is used to obtain the flight condition information of the aircraft; the thermal calculation decoder is used to sample the original historical data set by using an improved Latin hypercube sampling method to obtain training data samples related to the heating prediction task; performing hydrodynamic simulation calculations based on the training data samples to obtain the thermal characteristic calculation results of airborne components under different working conditions.

4. The aircraft airborne component heating prediction method according to claim 3, wherein The attribute information of the aircraft airborne components includes the material information, volume information, and shape information of the components; the flight condition information includes the flight altitude information, speed information, overload information, and air temperature information.

5. The aircraft airborne component heating prediction method according to claim 4, wherein The convolutional neural network adopts the form of a U-Net architecture, and the U-Net includes five channels: The first channel: used to input the component attribute information encoded by the component attribute encoder; The second channel: used to input the flight altitude information encoded by the component attribute encoder; The third channel: used to input the speed information encoded by the component attribute encoder; The fourth channel: used to input the overload information encoded by the component attribute encoder; The fifth channel: for inputting the air temperature information encoded by the component attribute encoder; Taking the component attribute encoder and the flight condition encoder as inputs and the thermal calculation decoder as the output, training a U-Net to generate a thermal prediction neural network model.

6. The method for predicting the heat generation of aircraft airborne components according to any one of claims 1 to 2, characterized in that, Performing hydrodynamic simulation calculations based on the training data samples, and the obtained thermal characteristic calculation results of the airborne components under different working conditions include: Collecting the operating conditions of the airborne components and determining the thermal characteristics to be studied; Constructing a geometric model of the airborne component so that the model reflects the shape and size of the actual component; Generating a computational grid on the geometric model and setting boundary conditions according to the actual working conditions and environment; Selecting the corresponding hydrodynamic and heat transfer models for simulation calculations to obtain the thermal characteristic calculation results of the airborne components under different working conditions.

7. The method for predicting the heat generation of aircraft airborne components according to claim 3, wherein The airborne component is an airborne radio altimeter.

8. An aircraft airborne component heating prediction device, characterized in that, It includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it implements the method for predicting the heat generation of the aircraft airborne components according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, A computer program is stored thereon, and when the computer program is executed by a processor, it implements the method for predicting the heat generation of the aircraft airborne components according to any one of claims 1-7.

Images