Civil aircraft component reliability evaluation technology based on operation and maintenance data and migration method

Abstract

The invention provides a civil aircraft component reliability evaluation technology based on operation and maintenance data and a migration method, and belongs to the technical field of intelligent operation and maintenance and composite materials, and the method comprises the steps: constructing a benchmarking model component damage prediction database; environment classification is carried out on the part damage prediction data to obtain a damage characteristic parameter distribution correlation model, an engine age-damage characteristic parameter distribution function based on an environment classification result is constructed to obtain an engine age-damage characteristic parameter prediction model, and according to the damage characteristic parameter distribution correlation model and the engine age-damage characteristic parameter prediction model, part damage prediction is carried out. Deriving a reliability function to obtain a reliability evaluation strategy; a migration coefficient is obtained through an accelerated aging test, a preset civil aircraft equivalent damage prediction database is generated through data conversion, and a reliability curve of a preset aircraft type is generated by adopting a reliability evaluation strategy. The method solves the problem that the accuracy of civil aircraft component damage prediction is low due to the lack of operation and maintenance data in existing civil aircraft component reliability evaluation.

Description

Technical Field

[0001] The present invention belongs to the technical fields of intelligent operation and maintenance and composite materials, and particularly relates to a reliability evaluation technology and migration method for civil aircraft components based on operation and maintenance data. Background Art

[0002] To ensure that civil aircraft meet the requirements of continuous airworthiness during long-term operation, the industry has gradually developed an operation and maintenance technology system that includes various means such as monitoring, prediction, decision-making, and maintenance. After years of development, the operation and maintenance technology system for civil aircraft has gradually evolved from planned maintenance based on statistical optimization technology to condition-based maintenance based on prediction technology. That is, data analysis and prediction technology are used to optimize the maintenance strategies of aircraft systems / components. Its core idea is to monitor the status of aircraft systems / components in real time, predict potential faults / damages, and perform inspections and maintenance before the faults / damages develop to endanger flight safety, thereby reducing downtime and maintenance costs.

[0003] The prior art uses the Skywise platform to perform predictive maintenance on the environmental control system, power system, and landing gear system of Airbus benchmark models and A330 models, reducing the probability of unplanned maintenance by 10 - 50%; the prior art uses the Aviater platform to predict the damage of aircraft components, reducing the number of unplanned part replacements to 30 - 40%.

[0004] Most of the existing fault or damage prediction technologies directly select the visual data during operation as the basis, that is, use real-time monitoring data (such as QAR data) to perform state analysis and fault / damage prediction using methods such as backpropagation neural networks and grey prediction models, which are not applicable to civil aircraft components without sensors. The damage data of components can only be obtained from the operation and maintenance data of civil aircraft maintenance bases (MROs), and the data types are diverse and scattered; currently, there is little research on civil aircraft component damage prediction methods based on operation and maintenance data, and some existing civil aircraft have been in operation for a short time, and the operation and maintenance data still need to be accumulated for a long time. Summary of the Invention

[0005] Aiming at the above deficiencies in the prior art, the present invention provides a reliability evaluation technology and migration method for civil aircraft components based on operation and maintenance data, which solves the problem that the reliability evaluation of existing civil aircraft components lacks operation and maintenance data, resulting in low accuracy of civil aircraft component damage prediction.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is: a reliability evaluation technology and migration method for civil aircraft components based on operation and maintenance data, applied to the damage prediction of preset civil aircraft components, including the following steps: S1. Collect the historical operation and maintenance data of the components of the benchmark model, extract the typical damage characteristics and damage characteristic parameters of the components of the benchmark model, and collect the environmental data of the regions corresponding to the typical damage characteristics of the components of the benchmark model to construct a damage prediction database; S2. According to the damage prediction database and the damage characteristic parameters of the components of the benchmark model, classify the environmental data of the component damage prediction data, construct a distribution correlation model of damage characteristic parameters and an age-damage characteristic parameter prediction model, and use the cumulative distribution function for derivation. By obtaining the reliability curve of the benchmark model, obtain a reliability evaluation strategy; S3. Based on the environmental classification of the component damage prediction data, use the accelerated aging test to collect the damage characteristic parameters of the typical test pieces of the components of the preset model and the benchmark model to obtain the damage characteristic parameter migration coefficient; S4. According to the damage characteristic parameter migration coefficient, use data conversion and reliability evaluation strategy to generate the reliability curve of the preset model, and complete the reliability evaluation technology and migration of civil aircraft components.

[0007] The beneficial effects of the present invention are as follows: Through a multi-factor coupled damage, environment and age modeling mechanism, taking the damage characteristic parameters as the link, the present invention provides a reliability evaluation strategy and transfers the damage development trend of the benchmark model to the preset civil aircraft; It solves the practical problem that the reliability evaluation of existing civil aircraft components lacks operation and maintenance data, realizes the predictive operation and maintenance of preset civil aircraft components, and improves the accuracy of civil aircraft component damage prediction.

[0008] Further, the S1 includes the following steps: S101. Collect the historical operation and maintenance data of the components of the benchmark model, extract the maintenance data from the historical operation and maintenance data to obtain the typical damage characteristics and damage characteristic parameters of the components; S102. According to the typical damage characteristics and damage characteristic parameters of the components, collect the environmental data of the regions corresponding to the typical damage characteristics of the components of the benchmark model; S103. Combine the historical operation and maintenance data and the environmental data to construct a damage prediction database.

[0009] The beneficial effects of the above further solution are as follows: By obtaining the damage prediction database, extracting the typical damage characteristics and damage characteristic parameters of the components, and then collecting the environmental data in combination with the typical damage characteristics, the present invention improves the pertinence, accuracy and reliability of the damage prediction database.

[0010] Still further, the S2 includes the following steps: S201. According to the damage prediction database, use the clustering analysis method to classify the environmental data of the component damage prediction data, and use mathematical statistics theory to construct a distribution correlation model of damage characteristic parameters; S202. Fit the age-damage characteristic parameter distribution function based on environmental classification according to the component damage characteristic parameters of the benchmark model, solve the distribution parameters, and obtain the age damage characteristic parameter prediction model; S203. According to the damage characteristic parameter distribution correlation model and the age-damage characteristic parameter prediction model, use the cumulative distribution function to deduce the reliability function, and obtain the reliability evaluation strategy by obtaining the reliability curve of the benchmark model.

[0011] Furthermore, S201 includes the following steps: S2011. Perform dynamic cluster analysis on the aircraft environmental data including the value tube area in the damage prediction database according to the damage prediction database, and obtain the result of environmental classification of component damage prediction data; S2012. Based on the result of environmental classification of component damage prediction data, use the quantile-quantile plot to preliminarily identify and analyze the distribution characteristics of the damage characteristic parameter data, and judge to obtain the initial damage distribution type; S2013. Use Kolmogorov-Smirnov to test the initial damage distribution type, and construct the damage characteristic parameter distribution correlation model by verifying the goodness of fit between the damage characteristic parameter data and the initial damage distribution type.

[0012] The beneficial effect of the above further solution is that: through the cluster analysis method and mathematical statistics theory, the present invention classifies the environment of component damage prediction data and constructs the damage characteristic parameter distribution correlation model, improving the goodness of fit and accuracy of the distribution correlation model.

[0013] Furthermore, S202 includes the following steps: S2021. According to the component damage characteristic parameters of the benchmark model, use the curve estimation method to fit the age-damage characteristic parameter distribution function type based on environmental classification, and obtain the initial age-damage characteristic parameter prediction model with distribution parameters; S2022. Use the maximum likelihood estimation method to solve the distribution parameters, obtain the optimal distribution parameters, and update the initial age-damage characteristic parameter prediction model to obtain the age damage characteristic parameter prediction model.

[0014] The beneficial effect of the above further solution is that: by fully considering key influencing factors such as service environment differences and age changes, the present invention introduces environmental classification and age variables, making the age damage characteristic parameter prediction model more suitable for the real operation and maintenance scenario, and having stronger applicability and generalization ability compared with the traditional single characteristic modeling method, and being able to be applied to complex service environment conditions.

[0015] Furthermore, S203 is specifically: According to the damage characteristic parameter distribution correlation model and the aircraft age-damage characteristic parameter prediction model, using the cumulative distribution function, describe the statistical distribution characteristics of the damage characteristic parameters, derive the reliability function, and conduct reliability assessment on the damage evolution law under different service times. By obtaining the reliability curve of the benchmark model, obtain the reliability assessment strategy.

[0016] The beneficial effects of the above further solution are as follows: By constructing the aircraft age-damage characteristic parameter prediction model and deriving the reliability function, the present invention realizes the prediction of aircraft structure damage evolution and the reliability assessment of the benchmark model, establishes a reliability assessment system driven by operation and maintenance data, obtains the reliability assessment strategy, and at the same time provides a scientific basis for formulating maintenance decisions and aircraft health management strategies.

[0017] Furthermore, the specific content of S3 is as follows: Based on the environmental classification of component damage prediction data and the typical test pieces of components of the benchmark model, using the accelerated aging test, and combining with the preset mechanical property experiment, collect the damage characteristic parameters of the typical test pieces of components of the preset model and the benchmark model, fit the trend line, and obtain the damage characteristic parameter migration coefficient by calculating the ratio of the damage characteristic parameters of the benchmark model and the preset model in the preset mechanical property experiment.

[0018] Furthermore, the specific content of S4 is as follows: According to the damage characteristic parameter migration coefficient, use data conversion to convert the damage characteristic parameter data of the benchmark model into the estimated data of the preset model, obtain the equivalent damage characteristic parameter database of the preset model, and generate the reliability curve of the preset model based on the reliability assessment strategy to complete the reliability assessment technology and migration of civil aircraft components.

[0019] The beneficial effects of the above further solution are as follows: By obtaining the typical structure damage data of the benchmark model and the preset model under typical environmental conditions, extracting the key damage parameters and establishing the migration mapping relationship, the present invention proposes a quantitative calculation method for the "migration coefficient", realizes the effective migration of damage data between different models, and effectively solves the problem of data scarcity in the initial operation period of the preset model. Description of the Drawings

[0020] Figure 1 It is the flowchart of the method of the present invention.

[0021] Figure 2 It is the technical roadmap of this embodiment.

[0022] Figure 3 It is the fitting curve graph of the non-linear regression of the pit depth in high-temperature, high-humidity and high-precipitation areas in this embodiment.

[0023] Figure 4This is the reliability curve graph of the benchmark model in the high-temperature, high-humidity, and high-precipitation region in this embodiment.

[0024] Figure 5 This is the fitting curve graph of the damage characteristic parameters of the typical test piece of the inner flap lower wall panel in the high-temperature, high-humidity, and high-precipitation region in this embodiment.

[0025] Figure 6 This is the reliability curve graph of the domestic model's inner flap lower wall panel component in the high-temperature, high-humidity, and high-precipitation region in this embodiment. Specific implementation manners

[0026] The following describes the specific implementation manners of the present invention to facilitate those skilled in the art of this technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation manners. For those of ordinary skill in the art of this technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions made using the concept of the present invention are within the scope of protection.

[0027] Before describing this embodiment, the following terms are first explained: CDF: Cumulative distribution function; Q-Q plot: Quantile-quantile plot; Logistic distribution: Logistic regression distribution; K-S test: Kolmogorov-Smirnov test.

[0028] Embodiment As Figure 1 shown, the present invention provides a civil aircraft component reliability evaluation technology and migration method based on operation and maintenance data, and its implementation method is as follows: S1. Collect the historical operation and maintenance data of the benchmark model components, extract the typical damage characteristics and damage characteristic parameters of the benchmark model components, and collect the environmental data of the regions corresponding to the typical damage characteristics of the benchmark model components to construct a damage prediction database. The specific steps are as follows: S101. Collect the historical operation and maintenance data of the benchmark model components, extract the maintenance data from the historical operation and maintenance data, and obtain the typical damage characteristics and damage characteristic parameters of the components; S102. According to the typical damage characteristics and damage characteristic parameters of the components, collect the environmental data of the regions corresponding to the typical damage characteristics of the benchmark model components; S103. Combine the historical operation and maintenance data and the environmental data to construct a damage prediction database.

[0029] In this embodiment, the preset model uses a domestic model; As Figure 2As shown in the figure, collect the historical operation and maintenance data of the components of the benchmark model. The historical operation and maintenance data include flight information, maintenance data, and flight mission information. Flight information includes, but is not limited to, aircraft registration number, aircraft model, aircraft age, and value management area, etc. Maintenance data includes, but is not limited to, damage time, damage level, maintenance method, and maintenance time, etc. Flight mission information includes, but is not limited to, flight time, take-off and landing locations, overnight airport, and flight time, etc. Select a preset mechanical property experiment according to the actual use environment, operation and maintenance conditions, and stress conditions. Then, through the selected preset mechanical property experiment, extract the typical damage characteristics and damage characteristic parameters of the component. For example, when the flap structure is impacted by runway sand and gravel, use the depth of the pit after impact as the damage characteristic parameter. When the sliding door structure shows delamination under long-term humid and vibrating conditions, use prefabricated defects of different sizes as the damage characteristic parameter. According to the typical damage characteristics and damage characteristic parameters of the component, collect atmospheric environment data, and collect the monthly data of meteorological environment indicators in the value management area corresponding to the typical damage characteristics of the components of the benchmark model for more than ten years. Environmental factors include, but are not limited to, temperature, humidity, precipitation, and the number of days with high temperature and high humidity, etc. Combine the historical operation and maintenance data and environmental data to construct a damage prediction database. The damage prediction database contains four labels: source, classification index, index / variable, and type, as shown in Table 1 specifically.

[0030] Table 1

[0031] S2. According to the damage prediction database and the damage characteristic parameters of the components of the benchmark model, classify the component damage prediction data by environment, construct a distribution correlation model of damage characteristic parameters and an age-damage characteristic parameter prediction model, and use the cumulative distribution function for derivation. Through obtaining the reliability curve of the benchmark model, obtain the reliability evaluation strategy. The specific steps are as follows: S201. According to the damage prediction database, use the clustering analysis method to classify the component damage prediction data by environment, and use mathematical statistics theory to construct a distribution correlation model of damage characteristic parameters. The specific steps are as follows: S2011. According to the damage prediction database, perform dynamic clustering analysis on the aircraft environment data in the damage prediction database that contains the value management area to obtain the result of the environmental classification of the component damage prediction data. S2012. Based on the result of the environmental classification of the component damage prediction data, use the quantile-quantile plot to preliminarily identify and analyze the distribution characteristics of the damage characteristic parameter data, and judge to obtain the initial damage distribution type. S2013. Use the Kolmogorov-Smirnov test to check the type of the initial damage distribution. By verifying the goodness of fit between the damage characteristic parameter data and the type of the initial damage distribution, establish a distribution correlation model of the damage characteristic parameters.

[0032] In this embodiment, according to the damage prediction database, perform dynamic clustering analysis on the aircraft environment data in the value tube area included, use the data mining technology to establish the environmental classification of the component damage prediction data, and obtain the environmental classification result of the component damage prediction data; perform environmental division through the environmental factor clustering method; set the aircraft service environment impact factor set as , n , where ; where, represents the clustering objective function, which is used to measure the total sum of the within-class variances, represents the i th sample in the j th clustering cluster, represents the i th clustering cluster, k represents the total number of clustering clusters, represents the mean of the i th clustering cluster, represents the Euclidean distance; through iterative optimization, make each environmental factor have the minimum within-class variance in the area with similar damage patterns; divide the service environment into four areas with similar damage patterns to obtain the environmental classification result of the component damage prediction data; The environmental classification of the component damage prediction data is specifically: low temperature and low humidity area, relatively low temperature and low humidity area, medium temperature and medium humidity area, and high temperature, high humidity and high precipitation area; the specific division criteria and their corresponding characteristics are shown in Table 2: Table 2

[0033] In this embodiment, to systematically analyze the distribution characteristics of the aircraft structure damage characteristic parameters, use the mathematical statistics theory to establish a distribution correlation model of the damage characteristic parameters, specifically: First, based on the environmental classification of component damage prediction data, the distribution characteristics of damage feature parameter data are initially identified and analyzed using a quantile-quantile plot. Based on the complex service environment of the aircraft structure and past engineering experience, the exponential distribution, Weibull distribution, lognormal distribution, and Logistic distribution are selected as candidate models for the distribution of damage feature parameters. By separately plotting the Q-Q plots of the damage feature parameter data and each candidate distribution, the fitting situation between the sample data and the theoretical distribution is visually compared, and the possible distribution types are initially judged to obtain the initial damage distribution type. Furthermore, the Kolmogorov-Smirnov (K-S) test is used to quantitatively test the initial damage distribution type, verify the goodness of fit between the damage feature parameter data and the theoretical distribution, calculate the K-S statistic and its corresponding p-value, and make a judgment based on the p-value to obtain the distribution association model of the damage feature parameters.

[0034] S202. According to the component damage feature parameters of the benchmark model, fit the age-damage feature parameter distribution function based on environmental classification, and solve the distribution parameters to obtain the age damage feature parameter prediction model. The specific steps are as follows: S2021. According to the component damage feature parameters of the benchmark model, use the curve estimation method to fit the type of age-damage feature parameter distribution function based on environmental classification, and obtain the initial age-damage feature parameter prediction model with distribution parameters. S2022. Use the maximum likelihood estimation method to solve the distribution parameters to obtain the optimal distribution parameters, and update the initial age-damage feature parameter prediction model to obtain the age damage feature parameter prediction model.

[0035] In this embodiment, according to the component damage feature parameters of the benchmark model, the curve estimation method is used to perform a fitting analysis on the relationship between the age and damage feature parameters based on environmental classification. Based on the environmental classification results, in the medium temperature and medium humidity environment, an adaptability test is carried out to determine that the statistical distribution characteristics of the age and damage feature parameters in the high temperature, high humidity, and high precipitation environment conform to the exponential distribution, and the initial age-damage feature parameter prediction model with corresponding distribution parameters is obtained. The expression is as follows: ; where represents the damage feature parameter, represents the aircraft age, a and b both represent distribution parameters; Using the maximum likelihood estimation method, parameter estimation is carried out, and by constructing a likelihood function, the distribution parameter values are iteratively optimized to make them conform to the distribution characteristics of the observed data to the greatest extent, and the optimal distribution parameters are obtained. ; In addition, in a high-temperature, high-humidity, and high-precipitation environment, the damage characteristic parameters follow an exponential distribution. The nonlinear regression method is used for fitting, and the maximum likelihood estimation method is used to obtain the distribution parameters, resulting in Figure 3 the nonlinear regression fitting curve graph of the pit depth in high-temperature, high-humidity, and high-precipitation areas as shown. The expression of the relevant regression equation is as follows: ; In a low-temperature, low-humidity environment and a relatively low-temperature, low-humidity environment, the damage characteristic parameters follow a linear regression. The expression of the relevant regression equation is as follows: ; .

[0036] S203. According to the damage characteristic parameter distribution correlation model and the aircraft age-damage characteristic parameter prediction model, using the cumulative distribution function, the reliability function is deduced. By obtaining the reliability curve of the benchmark model, a reliability assessment strategy is obtained, specifically: According to the damage characteristic parameter distribution correlation model and the aircraft age-damage characteristic parameter prediction model, using the cumulative distribution function, the statistical distribution characteristics of the damage characteristic parameters are described, the reliability function is deduced, and the reliability assessment of the damage evolution law under different service times is carried out. By obtaining the reliability curve of the benchmark model, a reliability assessment strategy is obtained.

[0037] In this embodiment, based on the distribution correlation model of the damage characteristic parameters and the prediction model between the aircraft age and the damage characteristic parameters, the reliability assessment of the damage evolution law under different service times is carried out; the statistical distribution characteristics of the damage characteristic parameters are described by the CDF, and the reliability function is further deduced to quantitatively characterize the probability that the components of the benchmark model maintain a normal working state at different service stages; In high-temperature, high-humidity, and high-precipitation areas, the damage characteristic parameter data follows a logistic distribution. Therefore, the CDF expression is as follows: ; Among them, represents the cumulative distribution function of the damage area, represents the location parameter, which is the median of the damage area, represents the scale parameter, reflecting the dispersion degree of the distribution; Based on the distribution correlation model corresponding to the damage characteristic parameters and the prediction model between the aircraft age and the damage characteristic parameters, the corresponding probability distribution model and the relationship between the damage characteristic parameters and the aircraft age are obtained, and the relationship between the reliability and the aircraft age is deduced and solved. The reliability expression is as follows: ; Among them, denotes the reliability function, denotes the aircraft age, denotes the damage characteristic parameter predicted by the aircraft age which is obtained from the aircraft age prediction model; the reliability function is used to describe the probability that a component maintains its normal working state at different service times, thereby reflecting its fault propagation law, and obtaining the reliability curve of the benchmark model in high-temperature, high-humidity, and high-precipitation regions as shown in Figure 4 and obtaining the reliability evaluation strategy; According to the reliability evaluation strategy, the reliability functions of the aircraft structures in the other three types of service environments are derived; specifically, according to the environmental characteristics of different regions, such as temperature, humidity, and climate change laws, the reliability of the aircraft structures in each region is calculated respectively.

[0038] S3. Based on the environmental classification of component damage prediction data, using the accelerated aging test, the damage characteristic parameters of the typical test pieces of the components of the preset model and the benchmark model are collected to obtain the damage characteristic parameter migration coefficient, specifically: Based on the environmental classification of component damage prediction data and the typical test pieces of the components of the benchmark model, using the accelerated aging test and combining with the preset mechanical property experiment, the damage characteristic parameters of the typical test pieces of the components of the preset model and the benchmark model are collected, and the trend line is fitted. By calculating the ratio of the damage characteristic parameters of the benchmark model and the preset model in the preset mechanical property experiment, the damage characteristic parameter migration coefficient is obtained.

[0039] In this embodiment, based on the environmental classification of component damage prediction data and the typical test pieces of the components of the domestic model, under this environmental condition, the accelerated aging test is carried out, and combined with the preset mechanical property experiment. By accelerating the aging test, the process of damage accumulation is accelerated, the damage characteristic parameters of the typical test pieces of the components of the benchmark model and the domestic model are collected, and the obtained data can truly reflect the damage change trend of the two types of models under similar environmental conditions, and the migration relationship is obtained; Specifically, the damage characteristic parameters of the benchmark model and the domestic model are measured, and the trend line is fitted. By calculating the ratio of the damage characteristic parameters of the benchmark model and the domestic model under the corresponding impact, the damage characteristic parameter migration coefficient is obtained, and the expression is as follows: ; where denotes the damage characteristic parameter migration coefficient in the h th environment, denotes the damage characteristic parameter of the benchmark model in the h th environment, denotes the damage characteristic parameter of the domestic model in the h th environment, denotes the type of environment.

[0040] In this embodiment, the typical test piece of the domestic aircraft model component is selected as the flap component, and the preset mechanical property experiment corresponding to the flap component is the impact experiment; Taking the pit damage, which is the most common damage type in the flap rudder surface, as an example, different-energy impact tests are carried out on the typical test piece of the lower wall panel of the inner flap, the pit depth of the typical test piece after the test is measured, the damage characteristic parameters of the lower wall panel of the inner flap of the reference model and the domestic aircraft model are obtained, and based on the high-temperature, high-humidity and high-precipitation area, as Figure 5 shown, the damage characteristic parameters of the typical test piece of the lower wall panel of the inner flap in the high-temperature, high-humidity and high-precipitation area.

[0041] S4. According to the damage characteristic parameter migration coefficient, using data conversion and reliability evaluation strategies, generate the reliability curve of the preset aircraft model, and complete the reliability evaluation technology and migration of civil aircraft components. Specifically: According to the damage characteristic parameter migration coefficient, using data conversion, convert the damage characteristic parameter data of the reference model into the estimated data of the preset aircraft model, obtain the equivalent damage characteristic parameter database of the preset aircraft model, and generate the reliability curve of the preset aircraft model based on the reliability evaluation strategy, so as to complete the reliability evaluation technology and migration of civil aircraft components.

[0042] In this embodiment, using the damage characteristic parameter migration coefficient and data transfer, convert the damage characteristic parameter data of the existing reference model into the estimated data of the domestic aircraft model. The data transfer expression is as follows: ; Without directly obtaining the operation and maintenance data of the domestic aircraft model, based on the historical data of the reference model, speculate the damage data set of the domestic aircraft model, and adopt the reliability evaluation strategy obtained in S2 to perform reliability analysis on the converted domestic aircraft model data, establish the reliability curve of the domestic aircraft model in four types of environments, realize the reliability evaluation technology and migration of civil aircraft components, and then realize the predictive operation and maintenance of domestic civil aircraft components. Based on the high-temperature, high-humidity and high-precipitation area, as Figure 6 shown, the reliability curve of the lower wall panel component of the inner flap of the domestic aircraft model in the high-temperature, high-humidity and high-precipitation area.

[0043] In this embodiment, a multi-factor coupling damage-environment-aircraft age modeling mechanism is provided, which comprehensively considers the correlation between the service environment, aircraft age and typical damage types, and proposes a classification modeling strategy for different environments and a dynamic estimation method based on aircraft age. Realize the quantitative prediction of the damage evolution law under different service scenarios, and enhance the generalization ability and engineering adaptability of the model; A data-driven aircraft structure reliability assessment strategy is constructed. A damage feature distribution model is built using the historical operation and maintenance data of benchmark models, the cumulative distribution function is derived, and the reliability curve is generated. This strategy breaks away from the strong dependence of traditional physical modeling on failure mechanisms, improves the efficiency and scope of application of reliability assessment, and provides scientific support for condition monitoring and maintenance strategies; Damage feature parameter migration is also achieved. With the same damage source (the damage source of the flap component is foreign object impacts with different energies) as the equivalent point, damage feature parameter conversion is carried out; damage feature parameters are extracted, a migration mapping relationship is established, and a quantitative calculation method for the "migration coefficient" is proposed to achieve effective migration of damage data between different models. At the same time, this method can be combined with environmental aging experiments to improve the authenticity of damage migration and environmental simulation capabilities, effectively solving the problem of lack of data in the initial operation stage of domestic models.

Claims

1. A reliability evaluation technology and migration method for civil aircraft components based on operation and maintenance data, which are applied to the damage prediction of preset civil aircraft components, and are characterized in that Including the following steps: S1. Collect the historical operation and maintenance data of the components of the benchmark model, extract the typical damage characteristics and damage characteristic parameters of the components of the benchmark model, and collect the environmental data of the regions corresponding to the typical damage characteristics of the components of the benchmark model, and construct a damage prediction database; S2. According to the damage prediction database and the damage characteristic parameters of the components of the benchmark model, classify the environmental data of the component damage prediction data, construct a distribution correlation model of damage characteristic parameters and an age-damage characteristic parameter prediction model, and use the cumulative distribution function for derivation. By obtaining the reliability curve of the benchmark model, obtain a reliability evaluation strategy; S3. Based on the environmental classification of the component damage prediction data, use the accelerated aging test to collect the damage characteristic parameters of the typical test pieces of the components of the preset model and the benchmark model, and obtain the damage characteristic parameter migration coefficient; S4. According to the damage characteristic parameter migration coefficient, use data conversion and reliability evaluation strategy to generate the reliability curve of the preset model, and complete the reliability evaluation technology and migration of civil aircraft components.

2. The reliability evaluation technology and migration method of civil aircraft components based on operation and maintenance data according to claim 1, characterized in that The S1 includes the following steps: S101. Collect the historical operation and maintenance data of the components of the benchmark model, extract the maintenance data from the historical operation and maintenance data, and obtain the typical damage characteristics and damage characteristic parameters of the components; S102. According to the typical damage characteristics and damage characteristic parameters of the components, collect the environmental data of the regions corresponding to the typical damage characteristics of the components of the benchmark model; S103. Combine the historical operation and maintenance data and the environmental data to construct a damage prediction database.

3. The reliability evaluation technology and migration method of civil aircraft components based on operation and maintenance data according to claim 1, characterized in that, The S2 includes the following steps: S201. According to the damage prediction database, use the clustering analysis method to classify the environmental data of the component damage prediction data, and use the mathematical statistics theory to construct a distribution correlation model of damage characteristic parameters; S202. According to the damage characteristic parameters of the components of the benchmark model, fit the age-damage characteristic parameter distribution function based on environmental classification, solve the distribution parameters, and obtain an age damage characteristic parameter prediction model; S203. According to the distribution correlation model of damage characteristic parameters and the age-damage characteristic parameter prediction model, use the cumulative distribution function to derive the reliability function. By obtaining the reliability curve of the benchmark model, obtain a reliability evaluation strategy.

4. The reliability evaluation technology and migration method of civil aircraft components based on operation and maintenance data according to claim 3, characterized in that, The S201 includes the following steps: S2011. According to the damage prediction database, perform dynamic clustering analysis on the aircraft environmental data including the regions in the damage prediction database, and obtain the results of environmental classification of the component damage prediction data; S2012. Based on the results of the environmental classification of the component damage prediction data, use the quantile-quantile plot to preliminarily identify and analyze the distribution characteristics of the damage characteristic parameter data, and judge the initial damage distribution type; S2013. Use the Kolmogorov-Smirnov to test the initial damage distribution type, and verify the goodness of fit between the damage characteristic parameter data and the initial damage distribution type to construct a distribution correlation model of damage characteristic parameters.

5. The reliability evaluation technology and migration method of civil aircraft components based on operation and maintenance data according to claim 3, characterized in that, The S202 includes the following steps: S2021. Based on the component damage characteristic parameters of the benchmark model, using the curve estimation method, fit the distribution function type of the aircraft age - damage characteristic parameters classified by environment, and obtain the initial aircraft age - damage characteristic parameter prediction model with distribution parameters; S2022. Solve the distribution parameters using the maximum likelihood estimation method to obtain the optimal distribution parameters, and update the initial aircraft age - damage characteristic parameter prediction model to obtain the aircraft age damage characteristic parameter prediction model.

6. The reliability evaluation technology and migration method of civil aircraft components based on operation and maintenance data according to claim 3, characterized in that The specific content of S203 is as follows: According to the damage characteristic parameter distribution correlation model and the aircraft age - damage characteristic parameter prediction model, use the cumulative distribution function to describe the statistical distribution characteristics of the damage characteristic parameters, deduce the reliability function, and conduct reliability assessment on the damage evolution law under different service times. By obtaining the reliability curve of the benchmark model, obtain the reliability assessment strategy.

7. The reliability evaluation technology and migration method of civil aircraft components based on operation and maintenance data according to claim 1, characterized in that The specific content of S3 is as follows: Based on the environmental classification of the component damage prediction data and the typical test pieces of the benchmark model components, use the accelerated aging test, and combine with the preset mechanical property experiment to collect the damage characteristic parameters of the typical test pieces of the preset model and the benchmark model components, and fit the trend line. By calculating the ratio of the damage characteristic parameters of the benchmark model and the preset model in the preset mechanical property experiment, obtain the damage characteristic parameter migration coefficient.

8. The reliability evaluation technology and migration method of civil aircraft components based on operation and maintenance data according to claim 1, characterized in that, The specific content of S4 is as follows: According to the damage characteristic parameter migration coefficient, use data conversion to convert the damage characteristic parameter data of the benchmark model into the estimated data of the preset model, obtain the equivalent damage characteristic parameter database of the preset model, and generate the reliability curve of the preset model based on the reliability assessment strategy to complete the reliability assessment technology and migration of civil aircraft components.

Images