A high-fidelity signal measurement and transmission system for full-aircraft lightning testing

Abstract

The present invention relates to the technical field of aircraft lightning protection test, and discloses a high-fidelity signal measurement and transmission system for a whole aircraft lightning test, comprising a high-fidelity data collection module, which is used for collecting observation data through a high-fidelity transmission system; a graph generation module, which is used for generating graph structure data; a sequence generation module, which is used for encoding measurement data of a key part of the aircraft represented by each node of the graph structure data and experimental environment data into a first sequence data; a lightning protection effect evaluation module, which inputs the graph structure data and the first sequence data into a lightning protection effect evaluation model, and outputs whether the lightning protection effect of the key part of the aircraft is qualified; the present invention evaluates whether the lightning protection effect of the key part of the aircraft is qualified by collecting high-fidelity data. Compared with evaluating the lightning protection effect through a single part, the system can simultaneously consider the observation sequence of each node and the dependency relationship between nodes, and comprehensively reflect the lightning protection effect of the entire aircraft.

Description

Technical Field

[0001] The invention relates to the technical field of aircraft lightning protection test, and more specifically, to a high-fidelity signal measurement and transmission system for aircraft whole-machine lightning test. Background Art

[0002] Aircraft may be struck by lightning during actual flight, so a full-aircraft lightning test is required to evaluate the lightning protection performance of the aircraft. In the lightning test, a high-fidelity signal measurement and transmission system is used to accurately measure and record the current, voltage, and electromagnetic field received by key parts of the aircraft to evaluate whether there are any defects in the lightning protection effect of the aircraft;

[0003] However, there are differences between the actual flight environment and the simulation test environment. Simulation tests are usually conducted in indoor laboratories where the temperature is controllable and relatively constant, but during actual flight, the aircraft will experience large temperature differences at different altitudes, and even encounter extreme high and low temperatures. At the same time, the laboratory environment usually maintains a low humidity level, but the aircraft will experience atmospheric environments with different humidity during flight, such as high humidity areas in thunderstorm clouds. Extreme temperature and high humidity can affect the insulation and electrical conductivity of the material, thereby affecting the lightning protection effect of key parts of the aircraft. Summary of the invention

[0004] The purpose of the present invention is to provide a high-fidelity signal measurement and transmission method, system and storage medium for aircraft full-machine lightning testing in order to solve the above-mentioned problems.

[0005] The present invention provides a high-fidelity signal measurement and transmission system for a whole-aircraft lightning test, comprising:

[0006] A high-fidelity data collection module, which is used to collect observation data through a high-fidelity transmission system. The observation data includes measurement data of key parts of the aircraft and lightning environment data of the aircraft;

[0007] A graph generation module is used to generate graph structure data, which includes nodes and edges. The nodes represent key parts of the aircraft or the lightning environment. The key parts of the aircraft include the fuselage, the left wing, the right wing, the landing gear, the right engine, and the left engine. There are edges between the nodes corresponding to the key parts of the aircraft connected on the aircraft, and there are edges between the node representing the lightning environment and all other nodes;

[0008] The node characteristics of the node representing the environment are equal to the environment characteristics, and the node characteristics of the node representing the key parts of the aircraft are equal to the part characteristics;

[0009] The key parts of the aircraft that are connected include the fuselage and left wing, the fuselage and right wing, the fuselage and landing gear, the left wing and right engine, and the right wing and left engine;

[0010] A sequence generation module is used to encode the measurement data of the key parts of the aircraft and the experimental environment data represented by each node of the graph structure data into a first sequence data; the first sequence data includes n sequence units, and the t-th sequence unit represents the node feature of a node at the t-th time sampling point;

[0011] A lightning protection effect evaluation module, which inputs the graph structure data and the first sequence data into a lightning protection effect evaluation model, the lightning protection effect evaluation model includes a first hidden layer, a second hidden layer and an output layer; inputs the first sequence data into the first hidden layer respectively, and outputs the first hidden state of each node respectively; inputs the graph structure data and the first hidden state of the node into the second hidden layer, and the second hidden layer outputs the second hidden state;

[0012] The output layer inputs the second hidden state and outputs the result of whether the lightning protection effect of the key parts of the aircraft is qualified.

[0013] Furthermore, the observation data is collected once at each time sampling point; the time sampling point refers to the moment when data is collected at a predetermined time interval, and the default value of the time interval is 1 millisecond.

[0014] Furthermore, environmental characteristics are obtained by processing lightning environmental data through feature engineering; and part characteristics are obtained by processing measurement data of key parts of the aircraft through feature engineering.

[0015] Furthermore, the formula for the first hidden layer is as follows:

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] in, , , represents the weight parameter, , , represents the bias parameter, represents the dot product, where represents the tth sequence unit of the first sequence data, , and Respectively represent the first, second, and third intermediate states, and They represent the t-th and t-1-th first hidden states respectively, n≥t≥1, n represents the total number of sequence units of the sequence data, and when t=1 , tanh is the hyperbolic tangent function, Represents a sigmoid function.

[0021] Furthermore, the formula for the second hidden layer is as follows:

[0022] ;

[0023] in Indicates The node second hidden state of the node, Indicates The aggregation coefficient of nodes, Indicates The set of nodes with which there are edges. represents the activation function, represents the state weight coefficient, Represents the nth first hidden state output when the sequence data of the jth node is input into the first hidden layer;

[0024] ;

[0025] ;

[0026] ;

[0027] in Indicates The nth first hidden state output when the sequence data of nodes is input into the first hidden layer, and Respectively represent the linear transformation features of the vth and jth nodes of the graph structure data, represents the aggregation weight coefficient, represents the splicing weight coefficient, T represents transposition, represents an exponential function with a natural constant as the base, and LeakyRelu represents the rectified linear unit function.

[0028] Furthermore, the calculation formula of the output layer is as follows:

[0029] ;

[0030] in Represents the output vector, and the two components of the output vector represent the The probability of qualified and unqualified lightning protection effects of key parts of the aircraft represented by each node, is the weight parameter, is the bias parameter, If the probability that the lightning protection effect of a key part of an aircraft is qualified is greater than the probability that it is unqualified, it means that the lightning protection effect of the key part is qualified; otherwise, the lightning protection effect of the key part is unqualified.

[0031] Furthermore, it also includes:

[0032] The transfer learning module is used to fine-tune the parameters of the lightning protection effect evaluation model. The fine-tuned lightning protection effect evaluation model is used in the actual flight environment.

[0033] Furthermore, the steps for fine-tuning the parameters of the lightning protection effect evaluation model are as follows:

[0034] Step 201: Loading lightning protection effect evaluation model and its parameters ;

[0035] Step 202: Take the collected observation data in the actual flight environment as input and calculate the target loss function of the model in the actual flight environment. :

[0036] ;

[0037] in, In actual flight conditions The actual lightning protection effect label of the key parts of the aircraft represented by each node, with a value of 0 or 1. 1 means that the lightning protection effect of the node is qualified in the actual flight environment, and 0 means that the lightning protection effect of the node is unqualified in the actual flight environment. represents the logarithmic function with the natural constant e as the base, is the first prediction of the model in the actual flight environment. The actual lightning protection effect labels of the key parts of the aircraft represented by each node;

[0038] Step 203: Update model parameters using gradient descent algorithm , get the fine-tuned migration model parameters ;

[0039] ;

[0040] in, is the regularized phase, is a regularization hyperparameter, the default value is 0.01, , represents the square of the 2-norm, Indicates Minimized The value of

[0041] Step 204: Output the new migration model after fine-tuning , as a lightning protection effectiveness evaluation model that will eventually be applied in actual flight environments.

[0042] The present invention provides a high-fidelity signal measurement and transmission method for a whole-aircraft lightning test of an aircraft. Based on the above-mentioned high-fidelity signal measurement and transmission system for a whole-aircraft lightning test of an aircraft, the following steps are performed:

[0043] Step 301, collecting observation data through a high-fidelity transmission system, the observation data including measurement data of key parts of the aircraft and lightning environment data of the aircraft;

[0044] Step 302, generating graph structure data, the graph structure data includes nodes and edges, the nodes represent key parts of the aircraft or the lightning environment, the key parts of the aircraft include the fuselage, the left wing, the right wing, the landing gear, the right engine, and the left engine, there are edges between the nodes corresponding to the key parts of the aircraft connected on the aircraft, and there are edges between the node representing the lightning environment and all other nodes;

[0045] Step 303, encoding the measurement data of the key parts of the aircraft and the experimental environment data represented by each node of the graph structure data into a first sequence data; the first sequence data includes n sequence units, and the t-th sequence unit represents the node feature of a node at the t-th time sampling point;

[0046] Step 304, inputting the graph structure data and the first sequence data into a lightning protection effect evaluation model, and the lightning protection effect evaluation model evaluates whether the lightning protection effect of the key parts of the aircraft is qualified;

[0047] Step 305, fine-tune the parameters of the lightning protection effect evaluation model, and the fine-tune lightning protection effect evaluation model is used in an actual flight environment.

[0048] The present invention provides a storage medium storing non-temporary computer-readable instructions. When the non-temporary computer-readable instructions are executed by a computer, the steps of the aforementioned high-fidelity signal measurement and transmission method for full-aircraft lightning testing of an aircraft can be executed.

[0049] The beneficial effect of the present invention is that the present invention evaluates whether the lightning protection effect of key parts of the aircraft is qualified by collecting high-fidelity data. Compared with evaluating the lightning protection effect through a single part, the system can simultaneously consider the observation sequence of each node and the dependency relationship between nodes, and comprehensively reflect the lightning protection effect of the entire aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a system module diagram of the present invention;

[0051] Figure 2The present invention is a flow chart of a high-fidelity signal measurement and transmission method for a whole-aircraft lightning test of an aircraft.

[0052] In the figure: high-fidelity data collection module 101, graph generation module 102, sequence generation module 103, lightning protection effect evaluation module 104, and transfer learning module 105. DETAILED DESCRIPTION

[0053] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0054] In a high-fidelity signal measurement and transmission system for aircraft whole-machine lightning test of the present invention, as Figure 1 As shown, including:

[0055] A high-fidelity data collection module 101, which is used to collect observation data through a high-fidelity transmission system, the observation data including measurement data of key parts of the aircraft and lightning environment data of the aircraft;

[0056] The observation data is collected once at each time sampling point; the time sampling point refers to the moment when data is collected at a predetermined time interval, and the default value of the time interval is 1 millisecond;

[0057] The key parts of the aircraft include the fuselage, left wing, right wing, landing gear, right engine, and left engine; the measured data include current, voltage, and electromagnetic pulse;

[0058] A graph generation module 102 is used to generate graph structure data, wherein the graph structure data includes nodes and edges, wherein the nodes represent key parts of the aircraft or the lightning environment, wherein the key parts of the aircraft include the fuselage, the left wing, the right wing, the landing gear, the right engine, and the left engine, wherein there are edges between the nodes corresponding to the key parts of the aircraft connected on the aircraft, and there are edges between the node representing the lightning environment and all other nodes;

[0059] The node characteristics of the node representing the environment are equal to the environment characteristics, and the node characteristics of the node representing the key parts of the aircraft are equal to the part characteristics;

[0060] The key parts of the aircraft that are connected include the fuselage and left wing, the fuselage and right wing, the fuselage and landing gear, the left wing and right engine, and the right wing and left engine;

[0061] In one embodiment of the present invention, the environmental features are obtained by processing lightning environmental data through feature engineering. The lightning environmental data is in text mode, and the Word Embedding algorithm is used as a feature engineering method.

[0062] The part features are obtained by feature engineering from the measurement data of the key parts of the aircraft. The measurement data is in text mode, and the Word Embedding algorithm is used as the feature engineering method.

[0063] Lightning environment data include discharge voltage amplitude, discharge voltage waveform (sine wave, pulse wave, etc.), discharge voltage polarity (positive or negative), discharge current amplitude, discharge current waveform, discharge current polarity, discharge duration, discharge repetition frequency, discharge position, ambient temperature, ambient pressure, and ambient humidity;

[0064] The measured data include the discharge voltage waveform applied to various parts of the aircraft and the current induced on the aircraft fuselage;

[0065] The sequence generation module 103 is used to encode the measurement data of the key parts of the aircraft and the experimental environment data represented by each node of the graph structure data into a first sequence data; the first sequence data includes n sequence units, and the t-th sequence unit represents the node feature of a node at the t-th time sampling point;

[0066] A lightning protection effect evaluation module 104, which inputs the graph structure data and the first sequence data into a lightning protection effect evaluation model, and the lightning protection effect evaluation model evaluates whether the lightning protection effect of the key parts of the aircraft is qualified;

[0067] The lightning protection effect evaluation model includes a first hidden layer, a second hidden layer and an output layer;

[0068] Input the first sequence data into the first hidden layer respectively, and output the first hidden state of each node respectively;

[0069] Input the graph structure data and the first hidden state of the node into the second hidden layer, and the second hidden layer outputs the second hidden state;

[0070] The output layer inputs the second hidden state and outputs the result of whether the lightning protection effect of the key parts of the aircraft is qualified;

[0071] In one embodiment of the present invention, the formula of the first hidden layer is as follows:

[0072]

[0073]

[0074]

[0075]

[0076] in, , , represents the weight parameter, , , represents the bias parameter, represents the dot product, where represents the tth sequence unit of the first sequence data, , and Respectively represent the first, second, and third intermediate states, and They represent the t-th and t-1-th first hidden states respectively, n≥t≥1, n represents the total number of sequence units of the sequence data, and when t=1 , tanh is the hyperbolic tangent function, Represents a sigmoid function.

[0077] In one embodiment of the present invention, the formula of the second hidden layer is as follows:

[0078]

[0079] in Indicates The node second hidden state of the node, Indicates The aggregation coefficient of nodes, Indicates The set of nodes with which there are edges. represents the activation function, represents the state weight coefficient, Represents the nth first hidden state output when the sequence data of the jth node is input into the first hidden layer;

[0080]

[0081]

[0082]

[0083] in Indicates The nth first hidden state output when the sequence data of nodes is input into the first hidden layer, and Respectively represent the linear transformation features of the vth and jth nodes of the graph structure data, represents the aggregation weight coefficient, represents the splicing weight coefficient, T represents transposition, represents an exponential function with a natural constant as the base, and LeakyRelu represents the rectified linear unit function.

[0084] In one embodiment of the present invention, the calculation formula of the output layer is as follows:

[0085]

[0086] in Represents the output vector, and the two components of the output vector represent the The probability of qualified and unqualified lightning protection effects of key parts of the aircraft represented by each node, is the weight parameter, is the bias parameter, If the probability that the lightning protection effect of a key part of an aircraft is qualified is greater than the probability that it is unqualified, it means that the lightning protection effect of the key part is qualified; otherwise, the lightning protection effect of the key part is unqualified.

[0087] In one embodiment of the present invention, the loss function of the lightning protection effect evaluation model is for:

[0088]

[0089] in, is the total number of nodes representing the key parts of the aircraft. The default value is 6. It is The actual lightning protection effect label of the key part of the aircraft represented by each node, with a value of 0 or 1. 1 means that the lightning protection effect of the node is qualified in the experimental environment, and 0 means that the lightning protection effect of the node is unqualified in the experimental environment. It is the first prediction of the lightning protection effect evaluation model. The lightning protection effect labels of the key parts of the aircraft represented by the nodes, Represents the logarithmic function with the natural constant e as the base.

[0090] The aforementioned lightning protection effect evaluation model uses observation data from simulation experiments during training; the experiment refers to observation data collected during the test field test;

[0091] In one embodiment of the present invention, the observation data collected by the test field test is collected by a high-fidelity transmission system, and the high-fidelity transmission system uses pulse current sensors, Rogowski coils, shunts, voltage probes, differential probes, magnetic field probes, electric field probes and other equipment to measure the current, voltage, electromagnetic field and other measurement data received by key parts of the aircraft, and the measurement data is transmitted through optical fibers;

[0092] The measured data is preprocessed through a high-frequency filter, amplifier, and attenuator, and the analog signal is sampled and quantized using an analog-to-digital converter to obtain a discrete digital signal. The conversion formula is as follows:

[0093]

[0094] in represents a discrete digital signal, is a continuous-time signal, is the sampling period, is the sampling point index.

[0095] However, there are differences in temperature and humidity between the actual flight environment and the simulated test environment. This difference will affect the results of the lightning protection effect evaluation. It is necessary to calibrate and optimize the original lightning protection effect evaluation model to improve the accuracy of the lightning protection effect evaluation results of the model in the actual flight environment.

[0096] In one embodiment of the present invention, a high-fidelity signal measurement and transmission system for aircraft whole-machine lightning testing further includes:

[0097] The transfer learning module 105 is used to fine-tune the parameters of the lightning protection effect evaluation model. The fine-tuned lightning protection effect evaluation model is used in the actual flight environment. The steps are as follows:

[0098] Step 201: Loading lightning protection effect evaluation model and its parameters ;

[0099] Step 202: Take the collected observation data in the actual flight environment as input and calculate the target loss function of the model in the actual flight environment. :

[0100]

[0101] in, In actual flight conditions The actual lightning protection effect label of the key parts of the aircraft represented by each node, with a value of 0 or 1. 1 means that the lightning protection effect of the node is qualified in the actual flight environment, and 0 means that the lightning protection effect of the node is unqualified in the actual flight environment. represents the logarithmic function with the natural constant e as the base, is the first prediction of the model in the actual flight environment. The actual lightning protection effect labels of the key parts of the aircraft represented by each node;

[0102] Step 203: Update model parameters using gradient descent algorithm , get the fine-tuned migration model parameters ;

[0103]

[0104] in, is the regularized phase, is a regularization hyperparameter, the default value is 0.01, , represents the square of the 2-norm, Indicates Minimized The value of

[0105] Step 204: Output the new migration model after fine-tuning , as a lightning protection effectiveness evaluation model that will eventually be applied in actual flight environments.

[0106] In one embodiment of the present invention, a high-fidelity signal measurement and transmission method for a full-aircraft lightning test is provided, such as Figure 2 As shown, the following steps are included:

[0107] Step 301, collecting observation data through a high-fidelity transmission system, the observation data including measurement data of key parts of the aircraft and lightning environment data of the aircraft;

[0108] Step 302, generating graph structure data, the graph structure data includes nodes and edges, the nodes represent key parts of the aircraft or the lightning environment, the key parts of the aircraft include the fuselage, the left wing, the right wing, the landing gear, the right engine, and the left engine, there are edges between the nodes corresponding to the key parts of the aircraft connected on the aircraft, and there are edges between the node representing the lightning environment and all other nodes;

[0109] Step 303, encoding the measurement data of the key parts of the aircraft and the experimental environment data represented by each node of the graph structure data into a first sequence data; the first sequence data includes n sequence units, and the t-th sequence unit represents the node feature of a node at the t-th time sampling point;

[0110] Step 304, inputting the graph structure data and the first sequence data into a lightning protection effect evaluation model, and the lightning protection effect evaluation model evaluates whether the lightning protection effect of the key parts of the aircraft is qualified;

[0111] Step 305, fine-tune the parameters of the lightning protection effect evaluation model, and the fine-tune lightning protection effect evaluation model is used in an actual flight environment.

[0112] At least one embodiment of the present invention provides a storage medium storing non-transitory computer-readable instructions. When the non-transitory computer-readable instructions are executed by a computer, the steps of the aforementioned high-fidelity signal measurement and transmission method for full-aircraft lightning testing of an aircraft can be executed.

[0113] The above describes an embodiment of the present invention, but this embodiment is not limited to the above-mentioned specific implementation mode. The above-mentioned specific implementation mode is merely illustrative and not restrictive. Under the guidance of this embodiment, ordinary technicians in this field can also make more forms of equivalent embodiments, all of which are within the protection of this embodiment.

Claims

1. A high-fidelity signal measurement and transmission system for aircraft lightning testing, characterized in that: include: A high-fidelity data collection module, which is used to collect observation data through a high-fidelity transmission system. The observation data includes measurement data of key parts of the aircraft and lightning environment data of the aircraft; A graph generation module is used to generate graph structure data, which includes nodes and edges. The nodes represent key parts of the aircraft or the lightning environment. The key parts of the aircraft include the fuselage, the left wing, the right wing, the landing gear, the right engine, and the left engine. There are edges between the nodes corresponding to the key parts of the aircraft connected on the aircraft, and there are edges between the node representing the lightning environment and all other nodes; The node characteristics of the node representing the environment are equal to the environment characteristics, and the node characteristics of the node representing the key parts of the aircraft are equal to the part characteristics; The key parts of the aircraft that are connected include the fuselage and left wing, the fuselage and right wing, the fuselage and landing gear, the left wing and right engine, and the right wing and left engine; A sequence generation module is used to encode the measurement data of the key parts of the aircraft and the experimental environment data represented by each node of the graph structure data into a first sequence data; the first sequence data includes n sequence units, and the t-th sequence unit represents the node feature of a node at the t-th time sampling point; A lightning protection effect evaluation module, which inputs the graph structure data and the first sequence data into a lightning protection effect evaluation model, the lightning protection effect evaluation model includes a first hidden layer, a second hidden layer and an output layer; inputs the first sequence data into the first hidden layer respectively, and outputs the first hidden state of each node respectively; inputs the graph structure data and the first hidden state of the node into the second hidden layer, and the second hidden layer outputs the second hidden state; The output layer inputs the second hidden state and outputs the result of whether the lightning protection effect of the key parts of the aircraft is qualified.

2. A high-fidelity signal measurement and transmission system for aircraft whole-machine lightning testing according to claim 1, characterized in that: Observation data is collected once at each time sampling point; a time sampling point refers to the moment when data is collected at a predetermined time interval, and the default value of the time interval is 1 millisecond.

3. The high-fidelity signal measurement and transmission system for aircraft lightning testing according to claim 1 is characterized in that: Environmental characteristics are obtained by processing lightning environment data through feature engineering; part characteristics are obtained by processing measurement data of key parts of the aircraft through feature engineering.

4. The high-fidelity signal measurement and transmission system for aircraft lightning testing according to claim 1 is characterized in that: The formula for the first hidden layer is as follows: ； ； ； ； in, , , represents the weight parameter, , , represents the bias parameter, represents the dot product, where represents the tth sequence unit of the first sequence data, , and Respectively represent the first, second, and third intermediate states, and They represent the t-th and t-1-th first hidden states respectively, n≥t≥1, n represents the total number of sequence units of the sequence data, and when t=1 , tanh is the hyperbolic tangent function, Represents a sigmoid function.

5. A high-fidelity signal measurement and transmission system for aircraft whole-machine lightning testing according to claim 4, characterized in that: The formula for the second hidden layer is as follows: ； in Indicates The node second hidden state of the node, Indicates The aggregation coefficient of nodes, Indicates The set of nodes with which there are edges. represents the S-type function, represents the state weight coefficient, Represents the nth first hidden state output when the sequence data of the jth node is input into the first hidden layer; ； ； ； in Indicates The nth first hidden state output when the sequence data of nodes is input into the first hidden layer, and Respectively represent the linear transformation features of the vth and jth nodes of the graph structure data, represents the aggregation weight coefficient, represents the splicing weight coefficient, T represents transposition, represents an exponential function with a natural constant as the base, and LeakyRelu represents the rectified linear unit function.

6. A high-fidelity signal measurement and transmission system for aircraft lightning testing according to claim 5, characterized in that: The calculation formula of the output layer is as follows: ； in Represents the output vector, and the two components of the output vector represent the The probability of qualified and unqualified lightning protection effects of key parts of the aircraft represented by each node, is the weight parameter, is the bias parameter, If the probability that the lightning protection effect of a key part of an aircraft is qualified is greater than the probability that it is unqualified, it means that the lightning protection effect of the key part is qualified; otherwise, the lightning protection effect of the key part is unqualified.

7. The high-fidelity signal measurement and transmission system for aircraft lightning testing according to claim 1 is characterized in that: Also includes: The transfer learning module is used to fine-tune the parameters of the lightning protection effect evaluation model. The fine-tuned lightning protection effect evaluation model is used in the actual flight environment.

8. The high-fidelity signal measurement and transmission system for aircraft lightning testing according to claim 7 is characterized in that: The steps to fine-tune the parameters of the lightning protection effectiveness evaluation model are as follows: Step 201: Loading lightning protection effect evaluation model and its parameters ; Step 202: Take the collected observation data in the actual flight environment as input and calculate the target loss function of the model in the actual flight environment. : ； in, is the total number of nodes representing the key parts of the aircraft, In actual flight conditions The actual lightning protection effect label of the key parts of the aircraft represented by each node, with a value of 0 or 1. 1 means that the lightning protection effect of the node is qualified in the actual flight environment, and 0 means that the lightning protection effect of the node is unqualified in the actual flight environment. represents the logarithmic function with the natural constant e as the base, is the first prediction of the model in the actual flight environment. The actual lightning protection effect labels of the key parts of the aircraft represented by each node; Step 203: Update model parameters using gradient descent algorithm , get the fine-tuned migration model parameters ; ； in, is the regularized phase, is a regularization hyperparameter, the default value is 0.01, , represents the square of the 2-norm, Indicates Minimized The value of Step 204: Output the new migration model after fine-tuning , as a lightning protection effectiveness evaluation model that will eventually be applied in actual flight environments.

9. A high-fidelity signal measurement and transmission method for aircraft full-machine lightning test, characterized in that: Based on the high-fidelity signal measurement and transmission system for aircraft whole-machine lightning test as described in claim 7 or 8, the following steps are performed: Step 301, collecting observation data through a high-fidelity transmission system, the observation data including measurement data of key parts of the aircraft and lightning environment data of the aircraft; Step 302, generating graph structure data, the graph structure data includes nodes and edges, the nodes represent key parts of the aircraft or the lightning environment, the key parts of the aircraft include the fuselage, the left wing, the right wing, the landing gear, the right engine, and the left engine, there are edges between the nodes corresponding to the key parts of the aircraft connected on the aircraft, and there are edges between the node representing the lightning environment and all other nodes; Step 303, encoding the measurement data of the key parts of the aircraft and the experimental environment data represented by each node of the graph structure data into a first sequence data; the first sequence data includes n sequence units, and the t-th sequence unit represents the node feature of a node at the t-th time sampling point; Step 304, inputting the graph structure data and the first sequence data into a lightning protection effect evaluation model, and the lightning protection effect evaluation model evaluates whether the lightning protection effect of the key parts of the aircraft is qualified; Step 305, fine-tune the parameters of the lightning protection effect evaluation model, and the fine-tune lightning protection effect evaluation model is used in an actual flight environment.

10. A storage medium, characterized in that: It stores non-transitory computer-readable instructions, and when the non-transitory computer-readable instructions are executed by a computer, the steps of a high-fidelity signal measurement and transmission method for a full-aircraft lightning test of an aircraft as described in claim 9 can be executed.

Images