Aircraft structure state evaluation method and device, and electronic device

Abstract

The application discloses an aircraft structure state evaluation method and device and electronic equipment, and belongs to the technical field of aircraft health management, and the method comprises the steps of: screening a first type of monitoring index related to an aircraft to be evaluated according to a judgment matrix, wherein elements in the judgment matrix are used to represent whether there is an association between two monitoring indexes of the aircraft to be evaluated, and the judgment matrix is given by a domain expert; determining the confidence distribution of each monitoring index according to each type of monitoring index data of the aircraft to be evaluated; calculating the correlation coefficient of the first type of monitoring index according to the confidence distribution; and determining the structure state evaluation result of the aircraft to be evaluated according to the weight, reliability, correlation coefficient and confidence distribution of each monitoring index, wherein the correlation coefficient of a non-first type of monitoring index is a default value. The application has higher accuracy and better reliability.

Description

Technical Field

[0001] This invention belongs to the field of aircraft health management technology, specifically relating to an aircraft structural condition assessment method and device, and electronic equipment. Background Technology

[0002] Structural condition assessment of aircraft is a critical step in aircraft storage, essential for ensuring structural safety and mission reliability during launch, flight, and under extreme loads. Accurate understanding of the aircraft's structural condition is of significant engineering importance for ensuring mission success and improving flight system safety. Generally, continuous and reliable structural health information, independent of external visual inspection, can be provided to users by real-time monitoring of the mechanical properties of key aircraft components. However, traditional aircraft condition assessment suffers from poor assessment effectiveness due to the following three characteristics of aircraft systems.

[0003] Specifically, firstly, aircraft are composed of multiple highly coupled subsystems with complex interactions between them, making it difficult to establish accurate mathematical models through mechanistic analysis, thus affecting the accuracy of structural state assessment; secondly, the coupling relationships between subsystems lead to complex nonlinear correlations among monitoring indicators. If such correlations are not fully considered during information fusion, key information may be reused, causing overestimation problems and reducing the credibility of assessment results; thirdly, due to the inherent complexity and uncertainty of aircraft systems, correlation results obtained solely from statistical analysis of indicator data lack clear physical meaning, while relying solely on mechanistic derivation makes it difficult to obtain effective correlations in practice, resulting in insufficient persuasiveness of existing correlation calculation methods.

[0004] Therefore, current methods for assessing the condition of aircraft suffer from low accuracy and poor reliability. Summary of the Invention

[0005] This invention provides a method, apparatus, and electronic device for evaluating the structural condition of an aircraft, which can solve the above-mentioned technical problems.

[0006] In a first aspect, embodiments of the present invention provide a method for evaluating the structural condition of an aircraft, the method comprising: The first category of monitoring indicators that are relevant to the aircraft to be evaluated are selected based on the judgment matrix. The elements in the judgment matrix are used to characterize whether there is a correlation between two monitoring indicators of the aircraft to be evaluated. The judgment matrix is ​​provided by a domain expert. Based on the monitoring index data of the aircraft to be evaluated, determine the confidence distribution of the monitoring index; Calculate the correlation coefficient of the first type of monitoring indicators based on the confidence distribution; The structural state assessment result of the aircraft to be evaluated is determined based on the weight, reliability, correlation coefficient and confidence distribution of each monitoring indicator, wherein the correlation coefficient of non-Class I monitoring indicators is the default value.

[0007] Secondly, embodiments of the present invention provide an aircraft structural condition assessment device, comprising: The indicator screening unit is used to screen the first type of monitoring indicators that are related to the aircraft to be evaluated according to the judgment matrix. The elements in the judgment matrix are used to characterize whether there is a correlation between two monitoring indicators of the aircraft to be evaluated. The judgment matrix is ​​provided by a domain expert. The result-oriented correlation calculation unit is used to determine the confidence distribution of the monitoring indicators based on the monitoring indicator data of the aircraft to be evaluated; and to calculate the correlation coefficient of the first type of monitoring indicators based on the confidence distribution. An evaluation unit is used to determine the structural state evaluation result of the aircraft to be evaluated based on the weight, reliability, correlation coefficient and confidence distribution of each monitoring indicator, wherein the correlation coefficient of non-first-class monitoring indicators is a default value.

[0008] Thirdly, embodiments of the present invention provide an electronic device, including a processor and a memory, wherein the memory is used to store a computer program; the processor can be used to execute a calculator program (instructions) stored in the memory to implement the method of the first aspect described above.

[0009] The beneficial effects of this invention compared to existing technologies are as follows: By introducing a judgment matrix pre-constructed by domain experts based on aircraft mechanism knowledge, this invention can effectively distinguish between correlated monitoring indicators (i.e., the first type of monitoring indicators) and independent indicators, providing prior knowledge with clear physical meaning for subsequent fusion calculations. During the calculation process, this invention dynamically calculates the correlation coefficient based on the confidence distribution of the first type of monitoring indicators, quantifying the nonlinear interactions between indicators and avoiding overestimation problems caused by information duplication in subsequent evaluations. Simultaneously, assigning default correlation coefficients to independent indicators simplifies computational complexity. Finally, this invention performs a fusion evaluation of the aircraft by comprehensively considering indicator weights, reliability, correlation coefficients, and confidence distributions calculated based on real-time data. This fully utilizes the guidance of mechanism knowledge while leveraging the refined advantages of data-driven monitoring indicators, thereby improving the accuracy and reliability of state assessment results under conditions of strong coupling and complex models in aircraft systems. Attached Figure Description

[0010] Figure 1 A flowchart illustrating the implementation of an aircraft structural state assessment method provided by this invention; Figure 2A schematic diagram illustrating a fusion sequence provided in an embodiment of the present invention; Figure 3 A flowchart illustrating the implementation of a result-oriented relevant evidence reasoning method provided by this invention; Figure 4 A schematic diagram of a structural condition assessment device for an aircraft provided in an embodiment of the present invention; Figure 5 A schematic diagram of monitoring indicator data provided in an embodiment of the present invention; Figure 6 A schematic diagram of a correlation coefficient provided for an embodiment of the present invention; Figure 7 A comparative schematic diagram of structural state evaluation results provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0011] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0012] The method provided in this embodiment of the invention can be applied to electronic devices such as mobile terminals, personal laptops, and supercomputers. This embodiment of the invention does not impose any restrictions on the specific type of electronic device.

[0013] Example 1 Figure 1 The diagram shown illustrates an implementation flowchart of an aircraft structural state assessment method provided by the present invention. This method is illustrative and not limiting, and can be applied to the aforementioned electronic equipment. The method may include steps S101 to S104, which are described below.

[0014] S101, based on the judgment matrix, select the first type of monitoring indicators that are relevant to the aircraft to be evaluated.

[0015] In one example, when determining the correlation between aircraft monitoring indicators, although experts cannot provide precise numerical values ​​for the correlation coefficients, they can clearly identify which two monitoring indicators are related. Therefore, domain experts can combine their knowledge of aircraft mechanisms to generate a judgment matrix for the monitoring indicators. The elements of the judgment matrix can be used to characterize whether a correlation exists between two monitoring indicators.

[0016] For example, the aircraft can be an airplane, a drone, a launch vehicle, etc.

[0017] For example, monitoring indicators may include sensor data such as impact intensity, settlement value, humidity, and temperature.

[0018] For example, the judgment matrix can satisfy the following formula:

[0019] in, To determine the matrix, =1 indicates the first Monitoring indicators With the Monitoring indicators There is a connection. =0 means and There is no correlation. , All are less than or equal to positive integers and , This represents the total number of monitored indicators. The elements on the diagonal of the judgment matrix represent the correlation between each indicator and itself, and therefore their values ​​are all equal to 1.

[0020] By using a judgment matrix, expert knowledge can be embedded in subsequent correlation calculations, while also avoiding the negative impact of uncertain expert knowledge on the evaluation results.

[0021] In one example, if the first The elements of the row are divided If there is still an element with a value of 1 outside, then we can... It was selected as a primary monitoring indicator.

[0022] S102, Based on the monitoring index data of the aircraft to be evaluated, determine the confidence distribution of the monitoring index.

[0023] In one possible implementation, the correlations between aircraft structural condition monitoring indicators are complex and typically exist across multiple dimensions; however, regardless of the dimensions and degree of correlation, their impact is reflected in the results. Therefore, this invention converts all monitoring indicator data of the aircraft to be evaluated into a confidence distribution of structural condition assessment results, and calculates the correlations of the monitoring indicators based on the confidence distribution.

[0024] In one example, the first Monitoring indicators The confidence distribution can include A confidence vector for evaluation can satisfy the following formula: (1.1), in, For the first Monitoring indicators The confidence distribution, For the first One evaluation result, for point to The evaluation confidence vector, Less than or equal to positive integers, This represents the total number of evaluation results.

[0025] For example, ,Depend on Data at different times were evaluated as The confidence level is composed of [the following].

[0026] Specifically, each monitoring indicator includes multiple assessment thresholds that correspond one-to-one with the assessment results, and the assessment interval is divided by the assessment thresholds. In the Test time Monitoring data Assessed as confidence level It is based on Once the evaluation interval is determined, it can satisfy the following formula: (1.2), in, for The index of the evaluation interval. Less than or equal to positive integers, This represents the total number of tests. (Number) The evaluation intervals are ( , ], , The respective The, the +1 evaluation threshold, The center value.

[0027] For example, there are three evaluation results: "Good," "Medium," and "Poor." The center value of the temperature index is 25℃, and the three evaluation thresholds are 0, 2, and 4, respectively. A temperature monitoring data point of 25.6℃ was collected. According to the formula (1.2) above, =1, the probability (confidence level) of the temperature monitoring data being judged as "good", "medium" and "poor" is 0.7, 0.3 and 0 respectively.

[0028] By converting multi-source monitoring indicators involving different scales and magnitudes into their respective confidence distributions, subsequent correlation coefficients can be calculated using a unified data format.

[0029] S103, calculate the correlation coefficient of the first type of monitoring indicators based on the confidence distribution.

[0030] In one possible implementation, the distance correlation between any first-class monitoring indicator and its related indicators on each evaluation result can be calculated first. The mean of the distance correlation coefficients between the two indicators and all evaluation results can be taken as the overall correlation coefficient of the indicator. Then, the reciprocal of the sum of the overall correlation coefficients of the indicator and all its related indicators can be taken as the correlation coefficient of the indicator.

[0031] For example, distance correlation (dCor) is an indicator that measures the linear or nonlinear correlation between two sets of random variables. It can effectively monitor any type of nonlinear correlation and is simpler and more interpretable than mutual information methods.

[0032] In aircraft condition assessment, various monitoring indicators typically consist of one-dimensional, small sample data, with sample sizes ranging from 20 to 100. The dCor method does not suffer from the dimensionality explosion problem and is more suitable for correlation analysis between small sample variables compared to other methods.

[0033] In one example, if For a Class I monitoring indicator, its correlation coefficient can be calculated through the following steps S1031~S1035.

[0034] S1031, according to , Calculate the confidence distribution. and Distance matrix pointing to each evaluation result.

[0035] For example, point to The distance matrix can satisfy the following formula: (1.3), in, express point to In the distance matrix, the first OK List elements, representing and The distance between them , They are respectively In the Test time , No. Test time Assessed as confidence level Also less than or equal to positive integers, This represents the L2 norm.

[0036] For example, for The One relevant indicator, There is a correlation between it and its related indicators. It is a positive integer, and its maximum value is equal to the th element in the judgment matrix. The total number of elements with a row value of 1 is reduced by 1.

[0037] Specifically, The relevant indicators can be determined based on the first one in the judgment matrix. The elements of the row are determined.

[0038] For example, the fifth monitoring indicator For a Class I monitoring indicator, if the values ​​of the elements in the 5th row of the judgment matrix are 1, 0, 0, 1, 1, then... There are two relevant indicators, namely and .

[0039] Similarly, The distance matrix can also be calculated using formula (1.3).

[0040] For example, if For the first +1 monitoring indicator Then, the formula (1.3) above can be used to... Replace with +1, calculate The distance matrix.

[0041] S1032, for and The distance matrix pointing to each evaluation result is double-decentralized to obtain... and Euclidean matrices pointing to each evaluation result.

[0042] For example, The Euclidean matrix can satisfy the following formula: (1.4), in, for point to The Euclidean matrix.

[0043] Similarly, point to The Euclidean matrix can also be calculated using the formula (1.4) above.

[0044] S1033, Calculation and The distance correlation when pointing to the same evaluation result is obtained. and Between The distance correlation coefficient.

[0045] For example, and point to The distance correlation coefficient can be calculated using the following formula: (1.5), in, for and point to The distance correlation coefficient at time for and The distance covariance between them , They are respectively , With its own covariance, for point to The evaluation confidence vector at that time.

[0046] in: (1.6), (1.7), (1.8), in, for point to The Euclidean matrix.

[0047] S1034, and Between The mean of the distance correlation coefficients is used as and The overall correlation coefficient between them.

[0048] For example, and The overall correlation coefficient between them can satisfy the following formula: (1.9), in, for and The overall correlation coefficient between them.

[0049] S1035, Calculation The reciprocal of the sum of the overall correlation coefficients between it and all its related indicators is obtained. The correlation coefficient.

[0050] In one example, the overall correlation coefficients of all monitored indicators can be combined to obtain a correlation matrix, and then the first one in the correlation matrix can be... Add the values ​​of the row elements together to get The sum of the overall correlation coefficients .

[0051] For example, the correlation matrix can satisfy the following formula: (1.10), in, For the correlation matrix, For monitoring indicators and The overall correlation coefficient between them, if one of them is not a Class I monitoring indicator, or is not a Class II monitoring indicator. Relevant indicators (i.e.) If not 1), then =0.

[0052] For example, the correlation coefficient of the monitoring indicators can satisfy the following formula: (1.11), in, for The correlation coefficient, When the indicators are perfectly correlated When completely independent (e.g., non-Category I monitoring indicators). The default value is 1; .

[0053] S104. Based on the weight, reliability, correlation coefficient and confidence distribution of each monitoring indicator, determine the structural state assessment result of the aircraft to be evaluated.

[0054] In one possible implementation, the fusion order of all monitoring indicators can be determined based on the correlation coefficients of each indicator; then, according to the fusion order, all monitoring indicators are fused to obtain the joint probability quality of the indicator fusion result based on the weight, reliability, confidence distribution and correlation coefficient of each indicator; finally, the structural state assessment result of the aircraft to be evaluated is determined based on the joint probability quality of the indicator fusion result.

[0055] In one example, due to the correlation between the indicators, the order of indicator fusion no longer satisfies the commutative law. Therefore, the first type of monitoring indicators can be fused first in descending order of correlation coefficient, followed by the fusion of independent non-first type monitoring indicators. When fusion of non-first type monitoring indicators, the fusion order can be disregarded.

[0056] Therefore, the integration priority of the first type of monitoring indicators is higher than that of non-first type monitoring indicators, and the higher the correlation coefficient of the first type of monitoring indicators, the higher the integration priority.

[0057] For example, see Figure 2 If there are 5 monitoring indicators, they are respectively denoted as ~ ,in ~ There is a correlation between them, and ,index , If the indicators are independent of other indicators, then the fusion order of the monitoring indicators can be: First, fuse them in sequence. , , The result of the third-level fusion was obtained. Reintegration Indicators , Obtain the index fusion results .

[0058] This invention introduces a judgment matrix pre-constructed by domain experts based on aircraft mechanistic knowledge, effectively distinguishing correlated monitoring indicators (i.e., the first type of monitoring indicators) from independent indicators, providing physically meaningful prior knowledge for subsequent fusion calculations. During the calculation process, this invention dynamically calculates correlation coefficients based on the confidence distribution of the first type of monitoring indicators, quantifying the nonlinear interactions between indicators and avoiding overestimation problems caused by information duplication in subsequent evaluations. Simultaneously, assigning default correlation coefficients to independent indicators simplifies computational complexity. Finally, this invention performs a fusion evaluation of the aircraft by comprehensively considering indicator weights, reliability, correlation coefficients, and confidence distributions calculated based on real-time data. This fully leverages the guidance of mechanistic knowledge while utilizing the refined advantages of data-driven monitoring indicators, thereby improving the accuracy and reliability of state assessment results under conditions of strong coupling and complex models in aircraft systems.

[0059] Furthermore, when the correlation between indicators is high, a single indicator is more likely to provide consistent information. This invention, based on correlation coefficients, first fuses monitoring indicators with strong correlations before fusing independent indicators. Prioritizing the integration of highly correlated monitoring indicators allows for faster convergence to correct inference results, laying a more solid foundation for subsequent fusion. Simultaneously, computational and time resources are limited during evidence-based reasoning. Prioritizing highly correlated evidence ensures that resources are allocated primarily to evidence most likely to influence the fusion result, thereby optimizing resource allocation.

[0060] Example 2 Figure 3The diagram shown illustrates an implementation flowchart of a result-oriented, relevant evidence reasoning method provided by the present invention. As an example and not a limitation, this method can be one possible specific implementation of step S104 described above. The method may include steps S301 to S305, which are described below.

[0061] S301 calculates the weight and reliability of each monitoring indicator based on the monitoring indicator data of the aircraft to be evaluated.

[0062] In one example, the original weight and original reliability of each monitoring indicator can be calculated directly based on the monitoring indicator data of the aircraft, and then the original weight and reliability can be used to participate in the subsequent fusion calculation.

[0063] For example, the original weights of monitoring indicators can be calculated based on the coefficient of variation of the monitoring indicator data.

[0064] The coefficient of variation is a statistical indicator used to measure the degree of variability in data; it describes the dispersion of data relative to the mean. When the coefficient of variation of a certain monitoring indicator is greater than that of other indicators, it indicates that its relative volatility is greater and has a greater impact on the evaluation results, thus requiring a larger weighting value for that monitoring indicator.

[0065] Assuming the aircraft to be evaluated is at time point The same structural state monitoring index was used in each test, namely The monitoring dataset of the aircraft to be evaluated. It can be represented as: (1.12).

[0066] but The coefficient of variation can satisfy the following formula: (1.13), in, Indicates the operational time of the aircraft to be evaluated. back The coefficient of variation. , They represent Monitoring dataset The root mean square error and the mean value, .

[0067] in: (1.14), (1.15), in, express The first in Data (i.e.) ).

[0068] For example, the original weights of the monitoring indicators can be calculated using the following formula: (1.16), in, for The original weights.

[0069] For example, the raw reliability of a monitoring indicator can be calculated using the following formula: (1.17), in, for The original reliability, for The sufficiency coefficient can be represented by a fuzzy relation function. for Discount factor of the source.

[0070] In another example, in order to simulate the operating environment of the real aircraft structural condition assessment process to the greatest extent, the original weights and original reliability of each monitoring indicator can be optimized by combining historical monitoring indicator data and the historical real structural condition of the aircraft, and the optimized weights and reliability are used for subsequent fusion calculations.

[0071] For example, the original weights and original reliability of the monitoring indicators can be iteratively optimized in multiple rounds based on the optimization model, which can be expressed as the following formula: (1.18), in, , They are respectively The weights and reliability are substituted into the model during the first round of optimization based on the optimization model. , It can be the original weights and original reliability calculated based on historical monitoring indicator data. According to , The calculated index fusion result points to The confidence distribution, Indicates based on , , The mean square error between the obtained structural state assessment results and the historical actual structural state Indicates constraints. for The cohesion ratio, This indicates minimization.

[0072] Specifically: (1.19), in, The total number of historical data. For the first The structural status assessment results obtained from a set of historical monitoring indicator data, For the first A true historical structural state.

[0073] S302, determine the cohesion ratio of the monitoring indicators based on their weights and reliability.

[0074] In one example, the aggregation ratio of the monitoring indicator can satisfy the following formula: (1.20), in, for The cohesion ratio, , They are respectively The reliability and weight of.

[0075] S303, based on the cohesion ratio of the monitoring indicators and the assessment confidence vector, determines the basic probability quality of the monitoring indicators pointing to each assessment result.

[0076] In one example, the basic probabilistic quality of the monitoring indicator can be calculated using the following formula: (1.21), in, express point to The basic probability mass, The representation of the identification framework is the set of all possible outcomes. ; for The power set of represents the set consisting of all subsets of the recognition frame. This represents the empty set.

[0077] S304. Following the fusion order, the monitoring indicators are fused in multiple iterations based on the basic probability mass and cohesion ratio until all monitoring indicators are fused, thus obtaining the joint probability mass of the indicator fusion results.

[0078] In one example, the Evidence Reasoning Rule with Cohesion Ratio Factor (ERr-CRC) based on the results-oriented evidence reasoning method provided by this invention (see steps S3041-S3043 below) can be used to integrate monitoring indicators. When all monitoring indicators are non-Class I indicators, ERr-CRC will degenerate into an ER rule method that considers the cohesion coefficient; while when all monitoring indicators are uncorrelated and the evidence is absolutely reliable, ERr-CRC will degenerate into an evidence reasoning model.

[0079] S3041, determine the joint probability quality of the current fusion result based on the joint probability quality of the previous level fusion result and the basic probability quality and cohesion ratio of the current monitoring indicators to be fused.

[0080] For example, the monitoring indicators to be integrated are the monitoring indicators arranged in the order of integration.

[0081] For example, the ranking of the current monitoring indicator to be fused can be used to replace the values ​​in the following formulas (1.12) and (1.13). This is used to calculate the joint probability quality of the current fusion result.

[0082] S3042, Determine whether the rank of the current fusion result is less than... .

[0083] In one example, if the rank of the current fusion result is less than... If there are still monitoring indicators that have not been integrated, then the next integration can be carried out starting from step S3041.

[0084] In another example, if the rank of the current fusion result is equal to If there are no unintegrated monitoring indicators, then step S3042 can be performed.

[0085] S3043 outputs the joint probability quality of the current fusion result as the joint probability quality of the index fusion result.

[0086] For example, the joint probability quality of the index fusion results can satisfy the following formula: (1.22), in: (1.23), The result of the indicator fusion points to the first Evaluation results The joint probability quality, For the first The cohesion ratio of each monitoring indicator to be integrated For the first Individual monitoring indicators to be integrated point to Basic probability mass, , For both include composite set, For the first -1 level fusion result points to Basic probability mass, For the first Individual monitoring indicators to be integrated point to Basic probability mass, For identification framework a subset of Indicates the first -1 level fusion result points to The basic probability mass, For the first -1 monitoring indicator to be integrated The joint probability quality, For the first The aggregation ratio of each monitoring indicator to be integrated For the first -1 level fusion result points to The joint probability quality, , The first -1 level fusion result, the Level fusion results point to The joint probability quality, For the first Level fusion results point to The normalized basic probability mass.

[0087] S305 determines the evaluation result of the aircraft to be evaluated based on the joint probability quality of the index fusion results.

[0088] In one example, the confidence distribution of the index fusion results pointing to each evaluation result can be determined first based on the joint probability quality of the index fusion results, and then the structural state evaluation result of the aircraft to be evaluated can be determined based on the confidence distribution of the index fusion results.

[0089] For example, the confidence level of the final index fusion result pointing to each evaluation result can satisfy the following formula: (1.24), in, The result of indicator fusion points to The confidence distribution.

[0090] For example, the structural state assessment result of the aircraft to be evaluated can satisfy the following formula: (1.25), in, The structural condition assessment results for the aircraft under evaluation represent the average health status of the aircraft. express The utility value.

[0091] This invention employs an ER rule method based on discount factors, combined with an improved ER rule, to determine the final evaluation result. This method considers the influence of the agglomeration ratio to address the problem of unreasonable mixing weights, and can effectively improve the fusion accuracy.

[0092] Example 3 Figure 4 The diagram shown illustrates the structure of an aircraft structural condition assessment device according to an embodiment of the present invention. As an example and not a limitation, the device may include an index screening unit, a result-oriented correlation calculation unit, and an assessment unit.

[0093] For example, the indicator screening unit is used to screen the first type of monitoring indicators that are relevant to the aircraft to be evaluated according to the judgment matrix, wherein the elements in the judgment matrix are used to characterize whether there is a correlation between two monitoring indicators of the aircraft to be evaluated, and the judgment matrix is ​​provided by a domain expert; the correlation calculation unit is used to determine the confidence distribution of the monitoring indicators according to the monitoring indicator data of the aircraft to be evaluated; and calculate the correlation coefficient of the first type of monitoring indicators according to the confidence distribution; the evaluation unit is used to determine the structural state evaluation result of the aircraft to be evaluated according to the weight, reliability, correlation coefficient and confidence distribution of each monitoring indicator, wherein the correlation coefficient of non-first type monitoring indicators is a default value.

[0094] To better illustrate the beneficial effects of the present invention, the following simulation experiments were conducted: Simulation Experiment In the experiment, the structural state of a rocket-type spacecraft was evaluated using a rocket structural state perception simulation platform. This platform, a rocket structural state perception simulation platform, is centered on a wireless sensor network composed of monitoring sensor modules and a wireless gateway. By deploying monitoring sensor modules to collect various operational parameters of the monitored object, it achieves data acquisition and safety monitoring simulation for various large structures. Based on this, the invention selects data from four vibration sensors, six settlement sensors, one temperature sensor, and one humidity sensor as monitoring indicators. In the experiment, the rocket's structural state evaluation result has three possibilities: "good," "medium," and "poor." 200 sets of data are collected for each state, totaling 600 sets. From these, 150 sets are randomly selected as monitoring indicator data (50 sets for each possibility), and the remaining 450 sets are used as historical monitoring indicator data. Figure 5 The data displayed included some monitoring indicators. Since the output data from similar sensors are highly similar, Figure 5 The data only shows the vibration and settlement data on the right side.

[0095] The evaluation thresholds for each monitoring indicator when calculating correlation are shown in Table 1 below: Table 1. Sensor Reference Values ​​at Different Levels

[0096] Referring to the example in step S102 above, when the temperature data collected by the temperature sensor is 25.6℃, the probability of it being judged as "good," "medium," and "poor" is 70%, 30%, and 0%, respectively. As the deviation of the temperature data from the center value increases, the probability of the data being evaluated as "poor" also increases. When the distance from the center value is greater than... At that time, it must be evaluated as "poor".

[0097] The correlation coefficients between the various monitoring indicators were calculated based on Table 1 above, and the results are as follows: Figure 6 shown. Specifically, Figure 6 (a) in the figure represents the location simulation of the correlation coefficients of various monitoring indicators. Figure 6 (b) shows the specific values ​​of the correlation coefficients of each monitoring indicator calculated according to the formulas (1.1) to (1.11) above.

[0098] See Table 2 below. Table 2 shows the original weights and original reliability of each monitoring indicator calculated according to the above formulas (1.12)~(1.17) and (1.18)~(1.19), as well as the weights and reliability.

[0099] Table 2 Sensor Parameter Setting Table

[0100] Based on the correlation coefficients of each monitoring indicator in Table 1, the original weights and original reliability of each monitoring indicator in Table 2, and the weights and reliability, the following results are calculated based on the formulas (1.20) to (1.25) above: Figure 7 The structural condition assessment results of the rocket to be evaluated are shown.

[0101] See Figure 7 , Figure 7 The results of 150 sets of data are presented, with 50 sets of data for each possible evaluation result, displayed on blue-filled planes (representing "good"), orange-filled planes (representing "medium"), and green-filled planes (representing "poor") from front to back. Each plane has a boundary line of the same color as the plane, representing the standard utility value corresponding to each evaluation result. The closer the predicted structural state evaluation result is to the boundary line, the more accurate the evaluation. Specifically, each plane displays results obtained based on different methods, where the red line represents the present invention, the pink line represents the result obtained by fusing the original reliability and original weights based on the ERr-CRC method, and the blue line represents the result obtained by the ER rule model considering the cohesion ratio but not using the correlation coefficient.

[0102] Depend on Figure 7 It can be seen that the prediction accuracy of the present invention is high, while the other two methods deviate from the standard value by a large margin. This also illustrates the rationality of considering the correlation coefficient and optimizing the calculation method of monitoring indicators and reliability through cohesion ratio.

[0103] Therefore, this invention, by introducing a judgment matrix pre-constructed by domain experts based on aircraft mechanistic knowledge, can effectively distinguish between correlated monitoring indicators (i.e., the first type of monitoring indicators) and independent indicators, providing physically meaningful prior knowledge for subsequent fusion calculations. During the calculation process, this invention dynamically calculates correlation coefficients based on the confidence distribution of the first type of monitoring indicators, quantifying the nonlinear interactions between indicators and avoiding overestimation problems caused by information duplication in subsequent evaluations. Simultaneously, assigning default correlation coefficients to independent indicators simplifies computational complexity. Finally, this invention performs a fusion evaluation of the aircraft by comprehensively considering indicator weights, reliability, correlation coefficients, and confidence distributions calculated based on real-time data. This fully utilizes the guidance of mechanistic knowledge while leveraging the refined advantages of data-driven monitoring, thereby improving the accuracy and reliability of state assessment results under conditions of strong coupling and complex models in aircraft systems.

[0104] Example 4 Figure 8 The diagram shown is a structural schematic of an electronic device provided in an embodiment of the present invention. Figure 8The illustrated electronic device 800 may include: at least one processor 810 ( Figure 8 The diagram shows only one processor, a memory 820, and a computer program 830 stored in the memory 820 and executable on the at least one processor 810, wherein the processor 810 executes the computer program 830 to implement the steps in any of the above method embodiments.

[0105] The electronic device 800 may be a robot or other processing device capable of implementing the above methods. This embodiment of the invention does not impose any restrictions on the specific type of electronic device.

[0106] Those skilled in the art will understand that Figure 8 This is merely an example of electronic device 800 and does not constitute a limitation on the electronic device. It may include more or fewer components than shown, or combine certain components, or use different components. For example, the electronic device 800 may also include input / output interfaces.

[0107] The processor 810 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASTCs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0108] In some embodiments, the memory 820 may be an internal storage unit, such as a hard disk or RAM. In other embodiments, the memory 820 may be an external storage device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital card (SD), or a flash card. Furthermore, the memory 820 may include both internal and external storage units. The memory 820 is used to store the operating system, applications, a boot loader, data, and other programs, such as the program code of the computer program. The memory 820 can also be used to temporarily store data that has been output or will be output.

[0109] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0110] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the functions described above can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0111] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0112] This invention provides a computer program product that, when run on an electronic device, enables the electronic device to perform the steps described in the various method embodiments above.

[0113] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0114] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0115] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software 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 implementations should not be considered beyond the scope of this invention.

Claims

1. A method for assessing the structural condition of an aircraft, characterized in that, include: The first category of monitoring indicators that are relevant to the aircraft to be evaluated are selected based on the judgment matrix. The elements in the judgment matrix are used to characterize whether there is a correlation between two monitoring indicators of the aircraft to be evaluated. The judgment matrix is ​​provided by a domain expert. Based on the monitoring index data of the aircraft to be evaluated, determine the confidence distribution of the monitoring index; Calculate the correlation coefficient of the first type of monitoring indicators based on the confidence distribution; The structural state assessment result of the aircraft to be evaluated is determined based on the weight, reliability, correlation coefficient and confidence distribution of each monitoring indicator, wherein the correlation coefficient of non-Class I monitoring indicators is the default value.

2. The method for assessing the structural condition of an aircraft according to claim 1, characterized in that, The step of calculating the correlation coefficient of the first type of monitoring indicators based on the confidence distribution includes: Based on the confidence distribution, the distance correlation between the first type of monitoring indicator and its related indicators when they point to the same evaluation result is calculated, thus obtaining the distance correlation between the first type of monitoring indicator and its related indicators. There are several distance correlation coefficients, where the first type of monitoring indicator is correlated with its own related indicators. The total number of evaluation results; The The mean of the distance correlation coefficients is used as the overall correlation coefficient between the first type of monitoring indicator and its own related indicators; The correlation coefficient of the first type of monitoring indicator is obtained by calculating the reciprocal of the sum of the overall correlation coefficients between the first type of monitoring indicator and all its related indicators.

3. The method for assessing the structural condition of an aircraft according to claim 1, characterized in that, The weights and reliability of the monitoring indicators are obtained by optimizing the original weights and reliability using an optimization model. The optimization model satisfies the following formula: ， in, , The first Monitoring indicators Weights and reliability Less than or equal to positive integers, The total number of monitoring indicators, According to , The calculated index fusion result points to the first The confidence distribution of each evaluation result. Less than or equal to positive integers, For the total number of evaluation results, Indicates based on , , The mean square error between the obtained structural state assessment results and the historical actual structural state Indicates constraints. for The cohesion ratio, This indicates minimization.

4. The method for assessing the structural condition of an aircraft according to claim 1, characterized in that, The determination of the structural state assessment result of the aircraft to be evaluated based on the weight, reliability, correlation coefficient, and confidence distribution of each monitoring indicator includes: Based on the correlation coefficients, determine the fusion order of all monitoring indicators; According to the fusion order, the monitoring indicators are fused based on the weight, reliability, correlation coefficient and confidence distribution of each monitoring indicator to obtain the joint probability quality of the indicator fusion result; The structural state assessment result of the aircraft to be evaluated is based on the joint probability quality of the fusion results of the aforementioned indicators.

5. The method for assessing the structural condition of an aircraft according to claim 4, characterized in that, The first type of monitoring indicators have a higher fusion priority than the non-first type monitoring indicators, and the higher the correlation coefficient of the first type of monitoring indicators, the higher the fusion priority.

6. The method for assessing the structural condition of an aircraft according to claim 4, characterized in that, The process of fusing the monitoring indicators according to the fusion order, based on the weight, reliability, correlation coefficient, and confidence distribution of each indicator, to obtain the joint probability quality of the indicator fusion result includes: The aggregation ratio of the monitoring indicators is determined based on their weights and reliability. Based on the cohesion ratio, assessment confidence vector, and correlation coefficient of the monitoring indicators, the basic probability quality of the monitoring indicators pointing to each assessment result is determined. The confidence distribution of the monitoring indicators is composed of the assessment confidence vectors of the monitoring indicators pointing to each assessment result. The assessment confidence vector is evaluated by the monitoring index data as the first... Evaluation results The confidence level is composed of, Less than or equal to positive integers, The total number of evaluation results; According to the fusion order, based on the basic probability quality and the cohesion ratio, the monitoring indicators are iteratively fused in multiple rounds until all monitoring indicators are fused, thereby obtaining the joint probability quality of the indicator fusion result.

7. The method for assessing the structural condition of an aircraft according to claim 6, characterized in that, The joint probability quality of the index fusion results satisfies the following formula: ， in: ， in, The result of the index fusion The joint probability quality, For the first The cohesion ratio of each monitoring indicator to be integrated For the first Individual monitoring indicators to be integrated point to Basic probability mass, , For both include composite set, For the first -1 level fusion result points to Basic probability mass, For the first Individual monitoring indicators to be integrated point to Basic probability mass, For identification framework a subset of for power set, Indicates the first -1 level fusion result points to The basic probability mass, For the first -1 monitoring indicator to be integrated The joint probability quality, For the first The aggregation ratio of each monitoring indicator to be integrated Represents the empty set. For the first -1 level fusion result points to The joint probability quality, , The first -1 level fusion result, the Level fusion results point to The joint probability quality, For the first Level fusion results point to The normalized basic probability mass.

8. The method for assessing the structural condition of an aircraft according to claim 6, characterized in that, The basic probability mass satisfies the following formula: ， in, For the first Monitoring indicators point to The basic probability mass, , They are respectively The correlation coefficient and cohesion ratio Represents the recognition framework. for The Each evaluation confidence vector for The power set of .

9. An aircraft structural condition assessment device, characterized in that, include: The indicator screening unit is used to screen the first type of monitoring indicators that are related to the aircraft to be evaluated according to the judgment matrix. The elements in the judgment matrix are used to characterize whether there is a correlation between two monitoring indicators of the aircraft to be evaluated. The judgment matrix is ​​provided by a domain expert. The result-oriented correlation calculation unit is used to determine the confidence distribution of the monitoring indicators based on the monitoring indicator data of the aircraft to be evaluated; and to calculate the correlation coefficient of the first type of monitoring indicators based on the confidence distribution. An evaluation unit is used to determine the structural state evaluation result of the aircraft to be evaluated based on the weight, reliability, correlation coefficient and confidence distribution of each monitoring indicator, wherein the correlation coefficient of non-first-class monitoring indicators is a default value.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-8.

Images