A digital twin monitoring method for the lifecycle health status of in-service aircraft structures

Abstract

A digital twin monitoring method for the full life-cycle health status of in-service aircraft structures, belonging to the technical field of digital twins, is disclosed. First, a health status data acquisition scheme is designed to obtain health status data under actual flight conditions. Second, the health status data is filtered, extracted, and optimized to construct a health status dataset of the in-service aircraft structure based on flight data and mechanical response data. Third, a digital twin model of flight data, mechanical response data, and life-cycle monitoring is established to achieve high-precision prediction of the remaining fatigue life of the in-service aircraft structure. Finally, an intelligent planning scheme for inspection and maintenance tasks is formulated, classifying inspection and maintenance levels based on the remaining fatigue life and fatigue reliability coefficient. This invention enables the prediction of the remaining fatigue life of in-service aircraft structures and the intelligent planning of inspection and maintenance tasks, forming a digital twin monitoring process for the full life-cycle health status of in-service aircraft structures, facilitating full life-cycle health status monitoring and management for in-service aircraft.

Description

Technical Field

[0001] This invention belongs to the field of digital twin technology and relates to a digital twin monitoring method for the whole life health status of an aircraft structure in service. Background Technology

[0002] Aircraft structural health monitoring technology records health status data during aircraft flight, calculates fatigue damage, and estimates the remaining structural life, enabling the monitoring and assessment of aircraft structural health status. This plays a crucial role in structural design and maintenance planning. Traditional methods involve deploying sensors on test or ground-based aircraft to monitor structural mechanical response data and determine structural health status. For example, Chinese invention patent CN117326090A discloses an online health monitoring device and method for light aircraft landing gear structures. This method, targeting light aircraft landing gear structures, establishes a relationship between different crack types, ultrasonic crack types, and the remaining life of landing gear wheel forks by statistically analyzing different crack forms and load cycles under ground testing, thus achieving health monitoring. However, this method is limited by factors such as spatial layout; sensors cannot completely cover critical areas of the aircraft structure, making it difficult to achieve full-field monitoring of structural mechanical response. Furthermore, it cannot consider the differences in single-aircraft service under actual flight conditions, only reflecting the mechanical response of the monitored aircraft structure under specific flight conditions, resulting in inaccurate health status monitoring and analysis, and high costs.

[0003] Digital twin technology, by integrating real data to simulate actual motion processes, establishes a data connection between physical and digital entities, enabling real-time monitoring and analysis of the physical entity's true state, and thus achieving simulation, monitoring, and prediction of the physical entity. Therefore, utilizing digital twin technology to achieve high-precision simulation of realistic mechanical responses, and subsequently to assess and predict remaining lifespan, provides a solution for intelligent monitoring of the full lifespan health status of in-service aircraft structures. For example, Chinese invention patent CN117077327A discloses a bearing life prediction method and system based on digital twins, which uses real-time collected bearing state parameters to establish a digital twin model and obtain lifespan data under multiple operating conditions. However, the stringent requirements of in-service aircraft and the massive volume of health status data necessitate further research into how to obtain health status data under actual flight conditions of in-service aircraft and construct a high-precision lifespan monitoring digital twin model to achieve accurate remaining lifespan assessment.

[0004] Therefore, based on the above issues, it is necessary to design a health status data acquisition scheme for in-service aircraft structures to obtain health status data under actual flight conditions. The health status data includes mechanical response data and flight data. It is also necessary to filter, extract, and optimize the flight data and mechanical response data to construct a health status dataset of flight data-mechanical response data for in-service aircraft structures. Furthermore, it is necessary to establish a digital twin model of flight data-mechanical response data-life monitoring to achieve high-precision prediction of the remaining fatigue life of in-service aircraft structures. Finally, it is necessary to develop an intelligent planning scheme for inspection and maintenance tasks, classifying inspection and maintenance levels based on the remaining fatigue life and fatigue reliability coefficient. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a digital twin monitoring method for the full life-cycle health status of in-service aircraft structures. This digital twin monitoring method collects health status data of in-service aircraft structures under actual flight conditions, filters, extracts, and optimizes the health status data, constructs a digital twin model of flight data, mechanical response data, and life-cycle monitoring, predicts the remaining fatigue life of the in-service aircraft structure, and then formulates intelligent planning schemes for inspection and maintenance tasks. This enables full life-cycle health status monitoring of in-service aircraft structures, facilitating full life-cycle health monitoring and management for in-service aircraft.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A digital twin monitoring method for the full life-cycle health status of an aircraft structure includes the following steps:

[0008] The first step is to design a health status data acquisition scheme for in-service aircraft structures to obtain health status data under actual flight conditions. This health status data includes mechanical response data and flight data. Specifically:

[0009] Step 1.1: Select a mechanical response data acquisition device that meets airworthiness requirements. On the one hand, this ensures the continuity and reliability of mechanical response data acquisition during flight, and the mechanical response data acquisition device can withstand flight conditions such as low temperature, large temperature difference, turbulence, and overload impact. On the other hand, it avoids affecting the normal operation of the aircraft during flight, so a mechanical response data acquisition device that is lightweight, small in size, low in energy consumption, and resistant to electromagnetic interference is selected.

[0010] Furthermore, the mechanical response data includes strain, stress, displacement, temperature, and acceleration. The mechanical response data acquisition equipment includes an acquisition instrument, sensors, acquisition lines, power lines, and fixed supports. The sensors include strain sensors, displacement sensors, temperature sensors, and acceleration sensors.

[0011] Step 1.2: Design the sensor layout scheme and determine the monitoring locations for mechanical response data. Based on mechanical mechanisms, simulation results, and maintenance data, ensure coverage of stress concentration areas and areas of significant strain in the aircraft structure, providing accurate input for the subsequent construction of a digital twin model of flight data and mechanical response data.

[0012] Furthermore, the stress concentration areas of the aircraft structure include holes, bends, and gaps, and the areas with significant strain include areas of varying thickness and connection areas.

[0013] Step 1.3: Based on the mechanical response data acquisition equipment selected in Step 1.1 and the sensor layout scheme determined in Step 1.2, design the installation scheme for the mechanical response data acquisition equipment. The acquisition instrument is installed close to the mechanical response data monitoring location to reduce redundant acquisition lines, and is secured using a fixed support to ensure stability during flight. During sensor installation, the sensor installation locations are positioned and labeled. After grinding and cleaning, flame-retardant adhesive is used to install the sensors, ensuring no damage and secure installation. Flame-retardant protective adhesive is applied to the sensor surface. Flame-retardant tape is wrapped around the connectors of the sensor leads, acquisition lines, and power cords for insulation and fire protection. The wiring arrangement is determined based on the mechanical response data monitoring location. After installation, the data acquisition equipment is debugged to ensure stable readings and good equipment operation.

[0014] Step 1.4: Develop a health status data acquisition plan and begin health status data acquisition. Use the mechanical response data acquisition equipment installed in Step 1.3 to collect mechanical response data, and use the flight parameter recording system onboard the aircraft to collect flight data. Divide the health status data acquisition into three phases: In the first phase, collect health status data m times to check and ensure data stability and accuracy; the second phase is the period when the sensors are operating stably, during which data is collected t times per week to ensure sufficient data for subsequent health status monitoring before the end of the second phase; in the third phase, collect data p times per month until a large number of sensor data anomalies are detected.

[0015] Furthermore, the flight data includes weight, altitude, speed, acceleration, attitude, control, engine, and environmental data; the first phase is within two weeks of starting health status data collection, with 3 ≤ m ≤ 14 collection times; the second phase is within three months of starting health status data collection, with 2 ≤ t ≤ 7 collection times per week; the third phase is within ten months of starting health status data collection, with 2 ≤ p ≤ 30 collection times per month.

[0016] The second step involves filtering, extracting, and optimizing flight data and mechanical response data to construct a health status dataset of flight data and mechanical response data for in-service aircraft structures. Specifically:

[0017] Step 2.1: For the flight data collected in Step 1.4, check and process abnormal values ​​of flight data caused by sensor failure and erroneous flight data during the data acquisition process, and fill in missing values; use correlation analysis to screen and reduce the correlation of flight data, set the correlation threshold of flight data as θ, calculate the correlation between each pair of flight data, and delete the flight data with higher correlation to other flight data in the two pairs of flight data with correlation > θ, so as to reduce redundancy.

[0018] Furthermore, the correlation analysis method may be Pearson correlation analysis or Spearman correlation analysis; the correlation threshold θ ranges from 0 to 1.

[0019] Step 2.2: For the mechanical response data collected in Step 1.4, the drift correction method is used to correct the drift, the noise reduction method is used to reduce the noise, and the resampling method is used to align the collection time and unify the collection frequency of the processed flight data and mechanical response data, so as to obtain flight data and mechanical response data with consistent collection time and collection frequency, thus obtaining the flight data-mechanical response data health status dataset, and dividing it into training set and test set.

[0020] Furthermore, the drift correction method includes linear regression, polynomial fitting, or piecewise function method; the noise reduction method includes wavelet noise reduction or wavelet packet decomposition; and the resampling method includes interpolation, averaging, decimation, Fourier transform, or wavelet transform.

[0021] The third step is to establish a digital twin model of flight data, mechanical response data, and life monitoring to achieve high-precision prediction of the remaining fatigue life of in-service aircraft structures. Specifically:

[0022] Step 3.1: Based on the flight data-mechanical response data health status data obtained in Step 2.2, establish a digital twin model of flight data-mechanical response data. Import the flight data and mechanical response data from the training set of the flight data-mechanical response data health status dataset into the machine learning model. Use the flight data from the training set as the input and the mechanical response data from the training set as the output of the machine learning model. Train the mapping relationship between flight data and mechanical response data using the machine learning model to obtain the digital twin model of flight data-mechanical response data. By inputting flight data into the digital twin model of flight data-mechanical response data, mechanical response data prediction is achieved. To evaluate the accuracy of the digital twin model of flight data-mechanical response data, use the mechanical response data from the test set of the flight data-mechanical response data health status dataset as the true value. Input the flight data from the test set into the digital twin model of flight data-mechanical response data to obtain the predicted value of mechanical response data for the test set. Use the error between the predicted value and the true value of mechanical response data for the test set as the evaluation index to assess the accuracy of the digital twin model of flight data-mechanical response data.

[0023] Furthermore, the machine learning models mentioned include Gaussian process regression, random forest, support vector machine, deep neural network, and extreme gradient boosting tree, all of which are known methods in the art. The error range between the predicted values ​​of the mechanical response data of the test set and the actual values ​​of the mechanical response data of the test set is 0~1. When the error range is between [0~0.1) (i.e., 0 ≤ error range < 0.1), the accuracy of the flight data-mechanical response data digital twin model meets the requirements and can be used for subsequent analysis. When the error range is between [0.1~1] (i.e., 0.1 ≤ error range ≤ 1), the accuracy of the flight data-mechanical response data digital twin model does not meet the requirements, and the machine learning model needs to be adjusted to reduce the error. Step 3.2: Establish a digital twin model of flight data-mechanical response data-life monitoring. Input the flight data of the test set into the digital twin model of flight data-mechanical response data established in step 3.1 to obtain the predicted values ​​of the mechanical response data of the test set. The peak and valley values ​​of the predicted mechanical response data are obtained by extracting peak-valley pairs. These peak and valley values ​​are used as inputs to the fatigue remaining life prediction method. The fatigue remaining life is calculated using the fatigue remaining life prediction method, resulting in a digital twin model of flight data-mechanical response data-life monitoring. By inputting flight data into the digital twin model of flight data-mechanical response data-life monitoring, mechanical response data prediction is achieved, and the fatigue remaining life of N flights is calculated.

[0024] Furthermore, the methods for extracting peak-valley pairs include rainflow counting, time-domain extraction, principal component analysis, or wavelet transform; the methods for predicting fatigue remaining life include nominal stress, local stress-strain, continuous damage mechanics, or detailed fatigue rating (DFR) method.

[0025] The fourth step is to develop an intelligent planning scheme for inspection and maintenance tasks, classifying inspection and maintenance levels based on remaining fatigue life and fatigue reliability coefficients. Specifically:

[0026] Step 4.1: Based on the maintenance party's damage tolerance requirements for the in-service aircraft structure, set the safety threshold as R flights. Based on the fatigue remaining life N obtained in step 3.2, when N > R, the in-service aircraft structure is in a healthy and safe period; when N ≤ R, the in-service aircraft structure is in a healthy and dangerous period.

[0027] Step 4.2, based on the health status determined in Step 4.1, and assuming the aircraft is within its safe health period, primarily conduct inspection tasks, including light-duty, medium-duty, and heavy-duty inspections. Based on the fatigue reliability coefficient of the aircraft structure, the inspection reliability level is classified into levels 1 to n.

[0028] When the fatigue reliability coefficient of the aircraft structure in service is the maximum value FRF max If the reliability level is n, which is the highest reliability level, then the actual number of flights of the aircraft structure in service is N. R =N / FRF max At that time, carry out lightweight inspection tasks;

[0029] When the fatigue reliability coefficient of the aircraft structure in service is the median value FRF, then Level 1 < Inspection Reliability Level < Level n. Therefore, when the actual number of flights N of the aircraft structure in service is... R When =N / FRF, perform a medium-scale check task;

[0030] When the fatigue reliability coefficient of the aircraft structure in service is at its minimum value FRF min If the reliability level is 1, which is the lowest reliability level, then the actual number of flights N of the aircraft structure in service is... R =N / FRF min At that time, a major inspection task was carried out.

[0031] Furthermore, the lightweight inspection tasks include visual inspection and surface penetration testing; the medium-weight inspection tasks include magnetic particle testing and eddy current testing; and the heavyweight inspection tasks include ultrasonic testing, radiographic testing, and infrared thermal imaging testing. The fatigue reliability coefficient is a coefficient introduced in this field to meet the reliability target requirements of fatigue design, and its value ranges from 1.0 to FRF. min <FRF<FRFmax ≤4.0; the reliability level n of the inspection is in the range of 3 to 10.

[0032] Step 4.3: Based on the health status determined in Step 4.1, for aircraft in a critical health condition, the main maintenance tasks are carried out, and the aircraft is scheduled for inspection and maintenance as soon as possible. First, inspections are conducted progressively from light to heavy-duty to confirm whether the aircraft structure is damaged, its location, and the degree of damage, thus obtaining an understanding of the structural damage situation. Then, based on the structural damage situation, maintenance levels are determined, and maintenance tasks are carried out, including surface maintenance, local repairs, component-level repairs, area-level repairs, and structural-level repairs.

[0033] The beneficial effects of this invention are as follows:

[0034] (1) The present invention designs a health status data acquisition scheme for the structure of service aircraft, which can meet the airworthiness requirements of service aircraft and obtain health status data under real flight conditions.

[0035] (2) The present invention establishes a method for screening, extracting and optimizing flight data and mechanical response data, which can reduce the redundancy of flight data and mechanical response data and obtain a high-quality flight data-mechanical response data health status dataset.

[0036] (3) The present invention establishes a digital twin model of flight data-mechanical response data-life monitoring, which can predict mechanical response data by inputting flight data into the digital twin model of flight data-mechanical response data-life monitoring, and then calculate the fatigue remaining life, forming a set of digital twin monitoring process for the full life health status of the aircraft structure in service.

[0037] (4) The present invention has developed an intelligent planning scheme for inspection and maintenance tasks, which can classify inspection and maintenance levels based on fatigue remaining life and fatigue reliability coefficient, making it convenient to carry out full life health monitoring and management for aircraft in service. Attached Figure Description

[0038] Figure 1 A flowchart illustrating the implementation of a digital twin monitoring method for the structural health status throughout the entire service life of an aircraft.

[0039] Figure 2 This is a schematic diagram of a digital twin model for flight data, mechanical response data, and life monitoring.

[0040] Figure 3 This is a schematic diagram of an intelligent planning scheme for inspection and maintenance tasks. Detailed Implementation

[0041] This specification will describe the embodiments of the present invention in detail with reference to the accompanying drawings. It should be particularly noted that the specific embodiments described below are only for the purpose of helping to understand the technical solutions of the present invention and do not constitute a limitation on the scope of protection. To clearly illustrate the key technical points, the accompanying drawings only present technical features directly related to the invention, omitting other non-essential details. Based on the technical solutions of the present invention, any alternative solutions that can be obtained by those skilled in the art through conventional technical means should be considered within the scope of protection of the claims of the present invention.

[0042] Figure 1 A flowchart illustrating the implementation of a digital twin monitoring method for the full life-cycle health status of an aircraft structure, provided for the implementation of this invention. (See attached flowchart.) Figure 1 As shown, the present invention provides a digital twin monitoring method for the full life-cycle health status of an aircraft structure, comprising:

[0043] The first step is to design a health status data acquisition scheme for in-service aircraft structures to obtain health status data under actual flight conditions. This health status data includes mechanical response data and flight data. Specifically:

[0044] Step 1.1: Select airworthiness-compliant mechanical response data acquisition equipment for the fuselage panel structure, including a data acquisition unit, sensors, acquisition cables, power cables, and mounting brackets. This ensures the continuity and reliability of mechanical response data acquisition during flight, requiring the equipment to withstand low temperatures, large temperature differences, turbulence, and overload impacts. It also avoids impacting normal aircraft operation during flight, selecting lightweight, compact, low-power, and electromagnetically interference-resistant mechanical response data acquisition equipment. The data acquisition unit selected is the small, lightweight DH5916 rugged miniature dynamic data acquisition and analysis system, capable of withstanding low temperatures, large temperature differences, turbulence, and overload impacts, and can accurately test mechanical response data such as force, pressure, displacement, velocity, and acceleration. The sensor uses the BA120 series sensor (single-chip) for monitoring, with an operating temperature range of -80℃ to +150℃, and can provide temperature self-compensation for aluminum alloy materials. The acquisition cables and power cables are made of flame-retardant materials, and the mounting brackets are made of composite honeycomb sandwich panels.

[0045] Step 1.2: Based on the sensors selected in Step 1.1, design the sensor layout scheme and determine the monitoring locations for mechanical response data. Based on mechanical mechanisms, simulation results, and maintenance data, ensure coverage of stress concentration areas and areas of significant strain in the aircraft structure, providing accurate input for the subsequent construction of a digital twin model of flight data-mechanical response data. Based on mechanical mechanisms, strain gauges are placed away from bolt holes and the influence of connection points, such as in the center of the milled skin area, and in stress-critical areas, such as around holes. Based on simulation results, they are placed in areas where strain is significant under different operating conditions, such as the bulkhead flange, bulkhead web, and the center of the milled skin area. Based on maintenance data, they are placed in areas prone to fatigue crack initiation, such as around bolt holes. A total of 32 sensors are deployed on the bulkhead, connectors, and fuselage skin, with 7 sensors on the bulkhead flange, 7 on the bulkhead web, 8 on connectors, 9 in the center of the milled fuselage skin area, and 1 on the floor.

[0046] Step 1.3: Based on the mechanical response data acquisition equipment selected in Step 1.1 and the sensor layout scheme determined in Step 1.2, design the installation scheme for the mechanical response data acquisition equipment. Two data acquisition units were installed in the overhead bins near the fuselage panel of the monitoring aircraft. Unit 1 was connected to sensors 1001-1016, and Unit 2 was connected to sensors 2017-2032. The data acquisition units were secured with high-strength bolts and fixed with brackets to ensure stability during flight. During sensor installation, the sensor positions were located and labeled. After grinding and cleaning, the sensors were installed using flame-retardant adhesive, ensuring no damage and secure installation. Flame-retardant protective adhesive was also applied to the sensor surface. The data acquisition cables were bundled and fixed along the outer fabric of the aircraft cabin. Holes were made in the outer fabric for cable routing, and an opening was made at the rear of the overhead bins for connecting the data acquisition cables and power cables. The sensor leads were soldered to the data acquisition cables using aviation-grade tools. Flame-retardant protective adhesive was applied to the sensors, and the leads were secured with tape to prevent damage to the sensors due to aircraft vibrations. After installation, the data acquisition equipment was tested to ensure stable readings and good operating condition.

[0047] Step 1.4: Develop a health status data acquisition plan and begin health status data acquisition. Use the mechanical response data acquisition equipment installed in Step 1.3 to collect mechanical response data, and use the flight parameter recording system built into the aircraft to collect flight data. Health status data acquisition is divided into three phases: In the first phase, within the first two weeks, health status data is collected five times to check and ensure acquisition stability and data accuracy; in the second phase, from the third week to the end of the third month, during the period of stable sensor operation, data is collected twice a week, totaling 60 health status data collections before the end of the second phase, meeting the data requirements for subsequent health status monitoring; in the third phase, from the fourth month onwards, data is collected twice a month until the end of the sixth month when a large number of sensor data anomalies are detected.

[0048] The second step involves filtering, extracting, and optimizing flight data and mechanical response data to construct a health status dataset of flight data and mechanical response data for in-service aircraft structures. Specifically:

[0049] Step 2.1: Based on the flight data collected in Step 1.4, 120 categories of flight data related to the mechanical response of the fuselage panels were initially screened from 2000 flight data types for analysis. Abnormal values ​​in the flight data caused by sensor malfunctions and erroneous flight data during data acquisition were checked and processed, and missing values ​​were filled. Specifically, 26 categories of flight data were removed due to sensor malfunction or data collected directly from non-sensor sources, and 20 categories required preprocessing such as deleting abrupt changes and filling in missing values. Spearman correlation analysis was used to filter the flight data based on correlation. A correlation threshold of θ=0.90 was set, and the correlation between each pair of flight data was calculated. Flight data with a correlation greater than θ that had a higher correlation with other flight data were removed, resulting in the removal of 46 categories of flight data. Finally, 48 categories of flight data were retained for subsequent analysis and training, reducing data redundancy and improving subsequent training efficiency.

[0050] Step 2.2: For the mechanical response data collected in Step 1.4, the drift is corrected using linear regression, and noise is reduced using wavelet packet decomposition. For the health status data of 60 real flights, the collection time of the processed flight data and mechanical response data is aligned and the collection frequency is unified to 8Hz using interpolation resampling, resulting in flight data and mechanical response data with consistent collection time and frequency. These are then divided into training and test sets, resulting in 60 sets of flight data-mechanical response data health status datasets, of which 56 sets are used as the training set and 4 sets are used as the test set. Each set contains 48 columns of flight data and 32 columns of mechanical response data.

[0051] The third step is to establish a digital twin model of flight data, mechanical response data, and life monitoring, such as... Figure 2 As shown, this enables high-precision prediction of the remaining fatigue life of in-service aircraft structures. Specifically:

[0052] Step 3.1: Based on the flight data-mechanical response data health status data obtained in Step 2.2, establish a digital twin model of flight data-mechanical response data. Using the flight data and mechanical response data from the training set of the flight data-mechanical response data health status dataset, select the extreme gradient boosting tree model as the machine learning model for this example. Use 48 columns of flight data from 56 training sets as input to the machine learning model, and use 32 columns of mechanical response data from 56 training sets as output to the extreme gradient boosting tree model. Train the mapping relationship between flight data and mechanical response data using the extreme gradient boosting tree model to obtain the flight data-mechanical response data digital twin model. This allows the prediction of mechanical response data by inputting flight data into the flight data-mechanical response data digital twin model. To evaluate the accuracy of the flight data-mechanical response data digital twin model, three columns of mechanical responses (1001, 1011, and 2007) from four test sets in the flight data-mechanical response data health status dataset were used as the true values. The flight data from the four test sets were input into the flight data-mechanical response data digital twin model. The prediction error for the mechanical response of test set 1001 was 0.049, and the training time was 797.38s; the prediction error for the mechanical response of test set 1011 was 0.114, and the training time was 669.07s; and the prediction error for the mechanical response of test set 2007 was 0.057, and the training time was 547.90s.

[0053] Step 3.2: Establish a digital twin model of flight data, mechanical response data, and life monitoring. Input 48 columns of flight data from 4 test sets into the digital twin model of flight data and mechanical response data established in Step 3.1 to obtain 32 columns of predicted mechanical response data values ​​from the 4 test sets. Use the rainflow counting method to obtain the peak and valley values ​​of the predicted mechanical response data values. Use the peak and valley values ​​as inputs to the fatigue remaining life prediction method to obtain the first principal stress peak value as 44.295 MPa and the valley value as -6.898 MPa. Calculate the fatigue remaining life using the detailed fatigue rating (DFR) method to obtain the digital twin model of flight data, mechanical response data, and life monitoring. This allows the prediction of mechanical response data by inputting flight data into the digital twin model of flight data, mechanical response data, and life monitoring, and then calculates the fatigue remaining life N = 32216 flights.

[0054] The fourth step is to develop an intelligent planning scheme for inspection and maintenance tasks, classifying inspection and maintenance levels based on remaining fatigue life and fatigue reliability coefficients, such as... Figure 3 As shown. Specifically:

[0055] Step 4.1: Based on the maintenance party's damage tolerance requirements for the structure of the aircraft in service, the safety threshold is set to R = 10,000 flights. Based on the fatigue remaining life N = 32216 obtained in step 3.2, N > R, and the structure is in a healthy and safe period.

[0056] Step 4.2, based on the health status determined in Step 4.1, and assuming the aircraft is within its safe health period, primarily conduct inspection tasks, including light-duty, medium-duty, and heavy-duty inspections. Based on the fatigue reliability coefficient of the aircraft structure, the inspection reliability level is classified into 1-3 levels.

[0057] Let the fatigue reliability coefficient of the aircraft structure in service be FRF. max =3.0, then the reliability level is checked to be level 3, which is the highest reliability level. Therefore, when the actual number of flights N of the aircraft structure in service is... R =N / FRF max When the value is 10738, a light inspection task is carried out, including visual inspection.

[0058] Let the fatigue reliability coefficient of the in-service aircraft structure be FRF=2.0, then the reliability level is checked as level 2. Then, when the actual number of flights N of the in-service aircraft structure... R When N / FRF=16108, a medium-scale inspection task is carried out to perform eddy current testing;

[0059] Let the fatigue reliability coefficient of the aircraft structure in service be FRF. min =1.5, then the reliability level is checked as level 1, which is the lowest reliability level. Therefore, when the actual number of flights N of the aircraft structure in service is... R =N / FRF min At 21477, a major inspection task was carried out, including ultrasonic testing.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit its scope of protection. Those skilled in the art, based on a full understanding of the foregoing embodiments, can make reasonable adjustments to the technical solutions or equivalent substitutions to some technical features, and these modified or substituted solutions should still fall within the scope of protection defined by the claims of the present invention.

Claims

1. A digital twin monitoring method for the full life-cycle health status of an aircraft structure, characterized in that, The digital twin monitoring method for the full life-cycle health status of the in-service aircraft structure includes the following steps: The first step is to design a health status data acquisition scheme for the structure of the aircraft in service to obtain health status data under actual flight conditions. The health status data includes mechanical response data and flight data. The second step involves filtering, extracting, optimizing, and managing flight data and mechanical response data to construct a health status dataset of flight data and mechanical response data for the structure of in-service aircraft, which is then divided into training and testing sets. The third step is to establish a digital twin model of flight data, mechanical response data, and life monitoring to achieve high-precision prediction of the remaining fatigue life of the aircraft structure in service; specifically: Step 3.1: Based on the flight data and mechanical response data in the training set, train the mapping relationship from flight data to mechanical response data, establish a digital twin model of flight data-mechanical response data, and realize mechanical response data prediction; Step 3.2: Establish a digital twin model of flight data, mechanical response data, and life monitoring; The flight data of the test set is input into the flight data-mechanical response data digital twin model to obtain the predicted value of the mechanical response data of the test set; the peak and valley values ​​of the predicted value of the mechanical response data are used as input to obtain the flight data-mechanical response data-life monitoring digital twin model, and the fatigue remaining life is calculated as N flights. The fourth step is to develop an intelligent planning scheme for inspection and maintenance tasks, classifying inspection and maintenance levels based on remaining fatigue life and fatigue reliability coefficients; specifically: Step 4.1: Based on the maintenance party's damage tolerance requirements for the in-service aircraft structure, set the safety threshold to R flights, and compare it with the fatigue remaining life N obtained in Step 3.2 to determine the health status of the in-service aircraft structure, including the safe health status period and the dangerous health status period. Step 4.2: For aircraft in a healthy and safe period, conduct inspection tasks; classify the inspection reliability level as 1 to n based on the fatigue reliability coefficient of the aircraft structure. Step 4.3: For situations where the aircraft is in a critical health condition, carry out maintenance tasks and arrange for the aircraft to be parked for inspection and maintenance.

2. The method for digital twin monitoring of the structural health status of an aircraft throughout its service life, as described in claim 1, is characterized in that... The first step is specifically as follows: Step 1.1: Select a mechanical response data acquisition device; the mechanical response data acquisition device includes an acquisition instrument, sensors, acquisition lines, power lines, and a fixed support, wherein the sensors include strain sensors, displacement sensors, temperature sensors, and acceleration sensors; Step 1.2: Design the sensor layout scheme and determine the monitoring locations for mechanical response data, which includes strain, stress, displacement, temperature, and acceleration. Based on mechanical mechanisms, simulation results, and maintenance data, ensure coverage of stress concentration areas and areas with significant strain in the aircraft structure to provide accurate input for the subsequent construction of a digital twin model of flight data and mechanical response data. Step 1.3: Based on the mechanical response data acquisition equipment selected in Step 1.1 and the sensor layout scheme determined in Step 1.2, design the installation scheme for the mechanical response data acquisition equipment; Step 1.4: Develop a health status data collection plan and begin health status data collection; collect mechanical response data using the mechanical response data collection equipment installed in Step 1.3, and collect flight data using the flight parameter recording system built into the aircraft in service; The health status data collection is divided into three stages: In the first stage, health status data is collected m times to check and ensure the stability and accuracy of the data; the second stage is the period when the sensor is working stably, during which data is collected t times per week to ensure that the amount of data collected before the end of the second stage meets the needs of subsequent health status monitoring; in the third stage, data is collected p times per month until a large number of sensor data anomalies are found.

3. The method for digital twin monitoring of the structural health status of an aircraft throughout its service life, as described in claim 2, is characterized in that... In the first step: In step 1.2, the stress concentration areas of the aircraft structure include holes, bends, and notches, and the areas with significant strain include areas of varying thickness and connection areas. In step 1.4, the flight data includes weight, altitude, speed, acceleration, attitude, control, engine, and environmental data; the first stage is within two weeks of starting health status data collection, with 3 ≤ m ≤ 14 collection times; the second stage is within three months of starting health status data collection, with 2 ≤ t ≤ 7 collection times per week; the third stage is within ten months of starting health status data collection, with 2 ≤ p ≤ 30 collection times per month.

4. The method for digital twin monitoring of the structural health status of an aircraft throughout its service life, as described in claim 2, is characterized in that... The second step is as follows: Step 2.1: For the flight data collected in Step 1.4, check and process abnormal values ​​of flight data caused by sensor failure and erroneous flight data during the data acquisition process, and fill in missing values; use correlation analysis to screen and reduce the correlation of flight data, set the correlation threshold of flight data as θ, calculate the correlation between each pair of flight data, and delete the flight data with higher correlation to other flight data in the two pairs of flight data with correlation > θ. Step 2.2: For the mechanical response data collected in Step 1.4, perform drift correction and noise reduction processing on it, and use a resampling method to align the collection time and unify the collection frequency of the processed flight data and mechanical response data to obtain flight data and mechanical response data with consistent collection time and collection frequency, thus obtaining the flight data-mechanical response data health status dataset, and dividing it into training set and test set.

5. The digital twin monitoring method for the full life-cycle health status of an aircraft structure according to claim 4, characterized in that, In the second step: In step 2.1, the correlation analysis method used is Pearson correlation analysis or Spearman correlation analysis; the correlation threshold θ ranges from 0 to 1. In step 2.2, a drift correction method is used to correct the drift, which includes linear regression, polynomial fitting, or piecewise function. A noise reduction method is used to reduce noise, which includes wavelet noise reduction or wavelet packet decomposition. The resampling method includes interpolation, averaging, decimation, Fourier transform, or wavelet transform.

6. The digital twin monitoring method for the full life-cycle health status of an aircraft structure according to claim 4, characterized in that, The third step specifically involves: The specific process for predicting mechanical response data in step 3.1 is as follows: Based on the health status data of flight data and mechanical response data obtained in step 2.2, a digital twin model of flight data and mechanical response data is established; the flight data and mechanical response data of the training set are imported into the machine learning model, with the flight data in the training set as the input of the machine learning model and the mechanical response data in the training set as the output of the machine learning model. The mapping relationship between flight data and mechanical response data is trained using the machine learning model to obtain the digital twin model of flight data and mechanical response data. By inputting flight data into the digital twin model of flight data and mechanical response data, mechanical response data prediction is achieved. In step 3.2, the specific process of establishing the digital twin model of flight data-mechanical response data-life monitoring is as follows: The peak and valley values ​​of the predicted mechanical response data are obtained by extracting peak-valley pairs. These peak and valley values ​​are used as inputs to the fatigue remaining life prediction method. The fatigue remaining life is calculated using the fatigue remaining life prediction method, resulting in a digital twin model of flight data-mechanical response data-life monitoring. By inputting flight data into the digital twin model of flight data-mechanical response data-life monitoring, mechanical response data prediction is achieved, and the fatigue remaining life of N flights is calculated.

7. The digital twin monitoring method for the full life-cycle health status of an aircraft structure according to claim 6, characterized in that, The third step specifically involves: In step 3.1, in order to evaluate the accuracy of the flight data-mechanical response data digital twin model, the mechanical response data in the test set is used as the true value. The flight data of the test set is input into the flight data-mechanical response data digital twin model to obtain the predicted value of the mechanical response data of the test set. The error between the predicted value of the mechanical response data of the test set and the true value of the mechanical response data of the test set is used as the evaluation index to assess the accuracy of the flight data-mechanical response data digital twin model. In step 3.1, the machine learning model includes Gaussian process regression, random forest, support vector machine, deep neural network or extreme gradient boosting tree; In step 3.2, the peak-valley pair extraction method includes rainflow counting, time-domain extraction, principal component analysis, or wavelet transform; the fatigue remaining life prediction method includes nominal stress method, local stress-strain method, continuous damage mechanics method, or detailed fatigue rating method.

8. The method for digital twin monitoring of the structural health status of an aircraft throughout its service life, as described in claim 6, is characterized in that... In step 3.1, the error range between the predicted value of the mechanical response data of the test set and the actual value of the mechanical response data of the test set is 0~1. When 0≤error range<0.1, the accuracy of the digital twin model of flight data-mechanical response data meets the requirements. When 0.1≤error range≤1, the accuracy of the digital twin model of flight data-mechanical response data does not meet the requirements, and the machine learning model needs to be adjusted to reduce the error.

9. The digital twin monitoring method for the full life-cycle health status of an aircraft structure according to claim 6, characterized in that, The fourth step is specifically as follows: The specific method for judging the health status of the aircraft structure in step 4.1 is as follows: when N > R, the aircraft structure is in a safe period of health; when N ≤ R, the aircraft structure is in a dangerous period of health. In step 4.2, based on the health status determined in step 4.1, inspection tasks are carried out for situations where the health status is within a safe period. These tasks include light-weight, medium-weight, and heavy-weight inspection tasks. The inspection reliability level is divided into levels 1 to n, specifically: When the fatigue reliability coefficient of the aircraft structure in service is the maximum value FRF max If the reliability level is n, which is the highest reliability level, then the actual number of flights of the aircraft structure in service is N. R =N / FRF max At that time, carry out lightweight inspection tasks; When the fatigue reliability coefficient of the aircraft structure in service is the median value FRF, then Level 1 < Inspection Reliability Level < Level n. Therefore, when the actual number of flights N of the aircraft structure in service is... R When =N / FRF, perform a medium-scale check task; When the fatigue reliability coefficient of the aircraft structure in service is at its minimum value FRF min If the reliability level is 1, which is the lowest reliability level, then the actual number of flights N of the aircraft structure in service is... R =N / FRF min At that time, a major inspection task was carried out.

10. The method for digital twin monitoring of the structural health status of an aircraft throughout its service life, as described in claim 9, is characterized in that... In the fourth step mentioned above: In step 4.2, the lightweight inspection tasks include visual inspection and surface penetration testing; the medium-weight inspection tasks include magnetic particle testing and eddy current testing; and the heavyweight inspection tasks include ultrasonic testing, radiographic testing, and infrared thermal imaging testing. The fatigue reliability coefficient ranges from 1.0 to FRF. min <FRF<FRF max The reliability level n described in ≤4.0 ranges from 3 to 10; Step 4.3 specifically involves: firstly, performing inspection tasks from lightweight to heavyweight levels to confirm whether the aircraft structure is damaged, the location of the damage, and the level of damage, thereby obtaining the damage status of the aircraft structure; then, classifying the maintenance level according to the damage status of the aircraft structure and carrying out maintenance tasks, including surface maintenance, local repair, component-level repair, area-level repair, and structural-level repair.

Images