CN113158140A - Aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion - Google Patents

Aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion Download PDF

Info

Publication number
CN113158140A
CN113158140A CN202110273838.8A CN202110273838A CN113158140A CN 113158140 A CN113158140 A CN 113158140A CN 202110273838 A CN202110273838 A CN 202110273838A CN 113158140 A CN113158140 A CN 113158140A
Authority
CN
China
Prior art keywords
damage
maintenance
analysis
structural
source information
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110273838.8A
Other languages
Chinese (zh)
Inventor
刘君强
关小玲
左洪福
陆晓华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Aeronautics and Astronautics
Original Assignee
Nanjing University of Aeronautics and Astronautics
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Aeronautics and Astronautics filed Critical Nanjing University of Aeronautics and Astronautics
Priority to CN202110273838.8A priority Critical patent/CN113158140A/en
Publication of CN113158140A publication Critical patent/CN113158140A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/20—Administration of product repair or maintenance

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Operations Research (AREA)
  • Algebra (AREA)
  • Human Resources & Organizations (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Evolutionary Biology (AREA)
  • Strategic Management (AREA)
  • Quality & Reliability (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Evolutionary Computation (AREA)
  • Economics (AREA)
  • Computing Systems (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)

Abstract

The invention discloses an aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion, which comprises the following steps: comprehensively considering four input sources of special event analysis, damage tolerance data, structure MSG-3 analysis and field repair experience, and selecting maintenance projects; establishing a maintenance interval model of multi-source information fusion; a repair interval for the aircraft structure is calculated using the multi-source information-fused repair interval model. The method solves the problem of single information source in the traditional aircraft structure fault diagnosis method, realizes effective fusion of multi-source information, and supports more accurate structure maintenance interval determination.

Description

Aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion
Technical Field
The invention belongs to the field of aircraft structure maintenance, and particularly relates to an aircraft structure maintenance project selection and maintenance interval analysis method.
Background
The core content of the civil aircraft maintenance outline is to determine maintenance tasks and maintenance intervals, and whether the maintenance tasks and the intervals are formulated reasonably or not directly influences the airworthiness safety and the maintenance economy of the aircraft.
The structural repair manual is an instructional document that provides the user with identification of the aircraft structure, instructional information about permissible damage and repair techniques, and instructional guidance on the implementation of a structural damage repair in order to assist the aircraft user in achieving and maintaining maximum utilization of the aircraft at a minimum cost and with an optimal repair implementation.
The establishment of regular maintenance plans for modern commercial aircraft is mainly based on the MSG-3 method. The method comprises the steps of firstly determining some evaluation indexes and rating important structural items of an airplane, secondly determining a comprehensive grade by adopting matrix conversion, then determining the length of a detectable crack according to the corresponding relation between the comprehensive grade and the length of the detectable crack, and finally determining the detection grade and the interval according to the length of the detectable crack and by combining a crack propagation curve.
At present, research on preventive maintenance decision of airplane structure fatigue damage can be divided into two parts, namely a theoretical method and an engineering method: the engineering method mainly includes methods of boeing and air passenger, and research on theory mainly focuses on time delay models, impact models and the like.
The invention is suitable for the problem that the structural analysis and interval determination in China at present are lack of methods in practice, and the determination only according to the traditional method can cause inaccurate results.
Disclosure of Invention
In order to solve the technical problems mentioned in the background technology, the invention provides an aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion.
In order to achieve the technical purpose, the technical scheme of the invention is as follows:
the method for selecting the aircraft structure maintenance project and analyzing the maintenance interval based on the multi-source information fusion comprises the following steps:
(1) comprehensively considering four input sources of special event analysis, damage tolerance data, structure MSG-3 analysis and field repair experience, and selecting maintenance projects;
(2) establishing a maintenance interval model of multi-source information fusion, wherein the model comprises a quintuple (W, FD, ST, I, R), wherein W represents a wiener process model, FD represents a fatigue damage-based method, ST represents a statistical conversion-based method, I represents an integration method based on W, FD and ST, and R represents an obtained result;
(3) a repair interval for the aircraft structure is calculated using the multi-source information-fused repair interval model.
Further, the method for selecting the maintenance items comprises a method for selecting MSG-3 maintenance items and a method for determining newly added maintenance items.
Further, the MSG-3 maintenance project selection method comprises failure consequence determination, failure possibility determination and rating system analysis;
the failure consequence determination means that structural items or structural details which have significant influence on the structural integrity of the aircraft flying, landing, supercharging or control load after failure are determined as important positions SSI, and damage tolerance or safe life characteristics of the structural items or the structural details are required to be determined;
the failure possibility determination means that according to effective evaluation of the load condition and the operation environment, a structural project or a region which most possibly shows structural degradation signs firstly is determined, and structural degradation reasons are three basic structural damages including accidental damage, environmental deterioration and fatigue damage;
the rating system analyses as follows:
for each important location SSI, an environmental deterioration level EDR and an accidental damage level ADR are determined, which take into account the detectability and sensitivity to damage and at the same time determine basic structural examination requirements for this important location SSI;
all important positions SSI sensitive to fatigue damage also have to determine the fatigue damage level FDR;
if the detection of fatigue damage to a certain important position SSI before the position SSI reaches the critical crack size cannot be realized, the position SSI is classified into a 'safe life' class; for each "safe life" item, its life limit is determined from analysis and/or experimentation; the important positions SSI of the class "safe life" are limited by environmental degradation and accidental damage assessment, and FDR reflects the detectable probability of fatigue damage before reaching a critical state;
critical damage is calculated using a residual strength analysis program, allowable loads, and typical material properties; crack propagation speed is predicted according to typical service load, bearing range and typical material characteristics; to ensure that damage is detected before critical dimensions are reached, inspection work should be selected based on fatigue and damage tolerance analysis and experience with similar structural designs; when the fatigue and damage tolerance analysis report is approved, the recommended intervals in the analysis report need to be re-evaluated according to the MSG-3 analysis, and the intervals can be further modified along with the fatigue test result of the complete machine or the part.
For the non-metal structure, FDR analysis is not carried out, and EDR and ADR are adopted to formulate an inspection outline;
for metal structures, a fatigue damage analysis procedure was employed.
Further, the method for determining the newly added maintenance items comprises the following steps:
selecting rules of maintenance items considering the structural design characteristics;
the special event analysis is to determine important special events occurring in the airplane structure according to the occurrence probability in each use stage;
the damage analysis is to determine important damage according to the maturity of the technology adopted for the structural design of the airplane, the damage type corresponding to the technology and the sensitivity to the damage, and the special events and the damage are finally implemented in specific physical objects, namely potential damage units;
determining the analysis object is to determine the dangerous crisis degree of the potential damage unit exposed to the danger according to the installation position, and continuously analyzing the high-risk unit;
the maintenance task is determined according to the analysis of the harmfulness of the analysis object;
maintenance item selection rules considering field maintenance experience;
a repair item selection rule based on the damage tolerance design analysis;
the source merging principle of each repair task is as follows: judging whether sources of the repair tasks overlap or not by considering the damage occurrence part, the damage mode and the damage influence, and determining a merging principle according to the side key points of the sources;
formal repair item selection principle: and comprehensively considering the damage sensitivity and the damage occurrence probability, and determining a formal repair item selection principle by taking the determined damage sensitivity grade and the calculated or counted damage probability as main indexes.
Further, the specific process of step (3) is as follows:
(3-1) determining a maintenance interval based on the wiener process:
(3-1-1) establishing a log-likelihood function of a degradation model based on historical performance degradation data, and estimating prior distribution of model parameters by using a maximum likelihood and one-dimensional search method;
(3-1-2) acquiring real-time performance degradation monitoring data of the structure, and updating the structure performance degradation model parameters in real time by using a Bayes method;
(3-1-3) obtaining an expected estimated value of posterior distribution of parameters in the structural performance degradation model;
(3-1-4) obtaining a lifetime satisfying the reliability limit, the time interval of which is a maintenance interval;
(3-2) determining a maintenance interval based on the fatigue damage:
(3-2-1) initializing setting;
(3-2-2) judging whether to perform initial check threshold value calculation and repeated check threshold value calculation;
(3-2-3) calculating an initial threshold value;
(3-2-4) calculating a duplicate checking threshold value;
(3-3) determining a maintenance interval based on a statistical conversion method:
(3-3-1) selecting a statistical distribution for analysis;
(3-3-2) analyzing whether the statistical distribution meets the hypothesis test, if so, turning to the step (3-3-3), and if not, returning to the step (3-3-1);
(3-3-3) performing statistical conversion using the distribution satisfying the hypothesis test;
(3-3-4) obtaining a maintenance interval based on the converted statistical result;
and (3-4) obtaining a final maintenance interval by using a Bayesian fusion method.
Adopt the beneficial effect that above-mentioned technical scheme brought:
the aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion is more suitable for engineering practice than the traditional method, and the maintenance interval obtained by the method is more accurate than that obtained by a single method.
Drawings
FIG. 1 is an overall flow chart of the present invention;
FIG. 2 is a flow chart of a maintenance interval algorithm of the present invention;
FIG. 3 is an SSI illustration and an SSI analysis diagram in an embodiment;
FIG. 4 is a diagram showing the analysis results of the maintenance items B and C in the embodiment.
Detailed Description
The technical scheme of the invention is explained in detail in the following with the accompanying drawings.
The invention designs an aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion, as shown in figure 1, the steps are as follows:
step 1, comprehensively considering four input sources of special event analysis, damage tolerance data, structure MSG-3 analysis and field repair experience, and selecting maintenance projects.
And 2, establishing a maintenance interval model of multi-source information fusion on the basis of the step 1, wherein the model comprises a quintuple (W, FD, ST, I, R), W represents a wiener process, FD represents a method based on fatigue damage FD, ST represents a method based on statistical conversion, I represents an integration method based on the three methods, and R represents an obtained result.
And 3, calculating the maintenance interval of the aircraft structure by using an algorithm based on the multi-source information fusion model.
In the embodiment, the aircraft structure maintenance item selection includes the selection rule of the maintenance item of MSG-3 and the newly added maintenance item determination rule.
The selection rules for MSG-3 repair projects include fault consequences (significant locations), fault probabilities (possible location analysis), and rating system analysis rules.
1. Consequences of failure (important position)
Structural items or details that have a significant impact on the structural integrity of the aircraft in safe flight, landing, pressurization and control loads after failure are identified as SSI and damage tolerance or safe life characteristics of these items need to be determined.
2. Probability of failure (possible location)
Based on an effective assessment of the loading conditions and operating environment, structural items or regions are identified that are most likely to first exhibit signs of structural degradation from three basic structural damage sources (referred to as accidental damage, environmental degradation, and fatigue damage). Typical item/structural details that are generally classified as SSI are:
1) connectors between the main elements;
2) static connections that require lubrication to prevent wear;
3) fatigue sensitive areas, such as:
-stress concentration
- - -non-continuity
Preload connections (especially those subjected to cyclic tension/compression forces)
-a splice
- -Primary junction
-skin opening
-door and window surround construction
Structures with the possibility of multiple cracks
4) Corrosion sensitive areas such as under-kitchen and toilet structures, under-fuselage substructure and items subject to stress corrosion.
5) Items/areas susceptible to accidental injury from external causes and maintenance activities, such as:
-in the vicinity of the exit or carrying door;
-items/areas close to areas where maintenance is frequent or where corrosive liquids leak.
6) A safe life item.
3. Rating system
For each SSI an environmental degradation level (EDR) and an accidental injury level (ADR) are determined. These classes take into account the detectability and sensitivity to damage, while also determining the examination requirements for the underlying structure for this SSI.
All SSI sensitive to fatigue damage also determine the Fatigue Damage Rating (FDR).
If it is not possible to detect fatigue damage for a certain SSI before it reaches the critical crack size, it is classified as "safe life".
For each "safe life" item, its life limit is determined from analysis and/or experimentation. Furthermore, SSI of the "safe life" class is limited by Environmental Degradation (ED) and occasional damage (AD) assessments. The FDR can be obtained by applying a special inspection schema, which reflects the detectable probability of fatigue damage before reaching a critical state.
The critical damage is calculated using the residual strength analysis program, the allowable load and typical material properties. Crack propagation speed is predicted from typical service loads, load ranges and typical material properties. To ensure that damage is detected before the critical dimension is reached, the inspection effort should be selected by analyzing the design experience of fatigue and damage tolerance and similar structures. When the fatigue and damage tolerance analysis report is approved, the recommended intervals in the report need to be re-evaluated according to the MSG-3 analysis. These intervals may be further modified as a result of fatigue testing of the machine or component.
For non-metal structures, because the structural design avoids the propagation of damage cracks, the structure is not subjected to FDR analysis, but an inspection outline should be formulated by EDR and ADR.
For metal structures, the Fatigue Damage (FD) analysis procedure was used. Most SSI are suitable candidates for fatigue-related sampling procedures. Sampling procedures related to fatigue focus on FD inspections of long-lived (high-frequency) aircraft in a fleet. The sampling procedure is a statistical method based on the following parameters:
predicted average life
- -sample size
- -Damage Tolerance Analysis (DTA) report
- -examination results
The details of the sampling procedure will be finalized before a maintenance item reaches its fatigue damage threshold.
The selection rule of the newly added maintenance items comprises the following steps:
1) and selecting rules of maintenance items considering the structural design characteristics.
2) The special event analysis is to determine important special events occurring in the aircraft structure according to the occurrence probability in each use stage.
3) The damage analysis is to determine important damage according to the maturity of the technology adopted by the aircraft structure design, the damage type corresponding to the technology and the sensitivity to the damage. Both the specific event and the damage eventually arrive at a specific physical object, i.e., a potentially damaged unit.
4) The analysis object is determined according to the installation position to determine the dangerous danger degree of the potential damage unit, so that the high-risk unit is continuously analyzed. This is a necessary measure to further reduce the amount of analysis tasks.
5) The maintenance task is determined according to the analysis of the harmfulness of the analysis object, so that the required maintenance task such as detection, positioning, measurement, recovery, replacement and repair is determined.
6) Maintenance item selection rules that take into account field maintenance experience.
7) The analyzed repair item selection rules are designed based on damage tolerance.
8) The source merging principle of each repair task is as follows: and (4) judging whether the sources of the repair tasks overlap or not by considering factors such as the damage occurrence part, the damage mode, the damage influence and the like, and determining a merging principle according to the side emphasis of each source.
9) Formal repair item selection principle: and comprehensively considering the damage sensitivity and the damage occurrence probability, and determining a formal repair item selection principle by taking the determined damage sensitivity grade and the calculated or counted damage probability as main indexes.
In this embodiment, a maintenance interval model based on multi-source information fusion is proposed based on the above contents, and includes a quintuple < W, FD, ST, I, R >, where W represents a wiener process, FD represents a method based on fatigue damage FD, ST represents a method based on statistical conversion, I represents an integration method based on the above three methods, and R represents an obtained result.
1. Wiener process
Wiener Process, the amount of degradation X (t) can be expressed as
X(t)＝X(0)+θt+σB(t) (1)
Where X (0) represents the amount of initial performance degradation, θ represents the drift coefficient in the Wiener process, σ represents the diffusion coefficient in the Wiener process, and b (t) obeys standard brownian motion.
Let t be (Δ t)1,...,ΔtM)TAnd Yi＝(Δyi,1,...,Δyi,M)TWhere Δ t isj＝tj-tj-1,Δyij＝Yi(tj)-Yi(tj-1)，tjThe runtime of the aircraft structure at time j. From the above equation and the independence assumption, Y can be derivediObeying multivariate normal distribution, the mean and variance are respectively:
Yi～N(μ,Σ)
μ＝μθt
Figure BDA0002975687360000091
Ω＝σ2Q+γ2IM
Q＝[min{ti,tj}]1＜i,j≤M
wherein μ and γ represent vernapassesThe unknown parameters in the process are obtained by a logarithm natural estimation method. M is the order, IMRepresenting an M-order identity matrix.
According to the assumption of independence between the performance monitoring data, for unknown parameter Θ ═ μθ,σθThe log-likelihood function of σ, γ } can be expressed as
Figure BDA0002975687360000101
Wherein, N is the number of sample data.
To easily solve the maximum likelihood estimation about the unknown parameters, the unknown parameters are redefined as
Figure BDA0002975687360000102
At this time, with respect to the unknown parameter
Figure BDA0002975687360000103
Can be rewritten as
Figure BDA0002975687360000104
For the above formula, the parameter mu is obtainedθAnd
Figure BDA0002975687360000105
the first partial derivative of (A) can be obtained
Figure BDA0002975687360000106
Figure BDA0002975687360000107
Given the partial parameters σ, γ, the values of equations (5), (6) are made 0, and the parameter μ can be obtainedθAnd
Figure BDA0002975687360000108
is a limited maximum likelihood estimate of
Figure BDA0002975687360000109
Figure BDA00029756873600001010
When the parameter mu is obtainedθAnd
Figure BDA00029756873600001011
after the maximum likelihood estimation expression is substituted into the formula (4), the parameters can be obtained
Figure BDA00029756873600001012
And
Figure BDA00029756873600001013
the cross-sectional log likelihood function of (1). At this time, with respect to the unknown parameter
Figure BDA00029756873600001014
And
Figure BDA00029756873600001015
the maximum similarity estimation value can be determined in a two-dimensional search mode to obtain parameters
Figure BDA00029756873600001016
And
Figure BDA00029756873600001017
after the maximum likelihood estimated value is substituted into the formulas (7) and (8) again, the parameter mu to be solved can be determinedθAnd
Figure BDA0002975687360000111
the maximum likelihood estimate of (a).
2. FD analysis method
The FD maintenance interval system is composed of an inspection threshold value, a repeated inspection interval and an interval time unit. The fatigue damage check threshold, the repeat check interval are determined from the results of the fatigue and damage tolerance analysis and revised based on usage experience, supplemental experimentation or engineering analysis. The time unit of the fatigue damage check task is represented by the flight cycle.
1. Initial inspection threshold
The inspection threshold is defined as the number of flight cycles, i.e. the inspection period, corresponding to the moment at which the first inspection should be performed. There are two methods for threshold determination:
(1) determined by dividing the time interval between the initial defect size propagation to the detectable crack length by the dispersion factor. The maximum allowable value is half of the target time length for designing and using the airplane, and the formula is represented as follows:
Figure BDA0002975687360000112
in the above formula, NTHIndicating the number of flights to the inspection threshold,
Figure BDA0002975687360000113
representing the number of flights from initial defect to detectable size, NDSORepresenting the target of the design and use of the airplane and calculating the number of the flying times. K1 represents the dispersion coefficient of the crack data source. K3 represents the dispersion factor of the environmental impact. K4 represents the dispersion coefficient taking into account uncertainty in the analysis.
(2) Determined from the time interval between the propagation of the initial defect size to the critical crack length divided by the dispersion factor.
Figure BDA0002975687360000114
In the above formula, the first and second carbon atoms are,
Figure BDA0002975687360000115
representing the number of flights from the initial defect to the critical crack.
In the case where the detectable crack length is difficult to determine, it is recommended to determine the inspection threshold using the 2 nd method.
2. Repeated checking of threshold value
The inspection interval is repeated, i.e. the interval between one inspection and the next. Often repeated multiple times. The repeat inspection begins when the crack propagates to a size that is truly monitored. The inspection period was taken as the time between the detectable crack length and the crack length divided by the dispersion factor.
Figure BDA0002975687360000121
In the above formula, NrepIndicating the number of flights for which the inspection interval is repeated,
Figure BDA0002975687360000122
the table is the number of flights from detectable crack to critical crack. K2 represents the dispersion coefficient of the path. K3 represents the dispersion factor of the environmental impact. K4 represents the dispersion coefficient taking into account uncertainty in the analysis.
The inspection threshold values for all FD-generated tasks are determined from design data or usage and maintenance experience of similar models, or conservatively taken as a repeat inspection interval value. This check threshold value can be revised based on the experience of the fleet at a later time.
The fatigue damage check threshold is determined by the host manufacturer based on the results of the fatigue and damage tolerance analysis and revised based on usage experience, supplemental experimentation or engineering analysis.
3. Method for converting based on statistical analysis
The method is based on a large number of data samples, and the main theoretical basis is probability theory and mathematical statistics. There are many standard statistical distributions that can be used as models for the reliability parameters, and the type of statistical distribution used in different cases depends on the nature of the data. In structural life data analysis, the most common types of distributions are: exponential distribution, normal distribution, lognormal distribution, gamma distribution, and the like.
The average ratio conversion method is a general calculation method of the environmental factors, and the calculation method of the environmental factors is an expected ratio of the distribution in the two environments. Namely:
Figure BDA0002975687360000123
the minimum variance unbiased estimation of environmental factors (MVUE) calculation method and results are different for different distributions.
For a normal distribution of x to N (μ, σ)2) The coefficient of variation of X is
Figure BDA0002975687360000124
Let X, Y obey normal distribution, are independent of each other, and have the same coefficient of variation:
Figure BDA0002975687360000125
then the environmental factor
Figure BDA0002975687360000131
Taking a sample X with the capacity of n in X1,x2,…,xi,…,xnTaking a sample Y with capacity m in Y1,y2,…,yi,…,ynThe mean values of the samples are respectively
Figure BDA0002975687360000132
The sample variances are respectively
Figure BDA0002975687360000133
Then the environmental factor
Figure BDA0002975687360000134
Is a Minimum Variance Unbiased Estimation (MVUE) of
Figure BDA0002975687360000135
For a lognormal distribution of x to LN (mu, sigma)2) The coefficient of variation of X is
Figure BDA0002975687360000136
Let X, Y obey the log-normal distribution, are independent of each other, and have the same coefficient of variation:
Figure BDA0002975687360000137
then the environmental factor
Figure BDA0002975687360000138
Taking a sample X with the capacity of n in X1,x2,…,xi,…,xnTaking a sample Y with capacity m in Y1,y2,…,yi,…,ynThe mean values of the samples are respectively
Figure BDA0002975687360000139
The sample variances are respectively
Figure BDA00029756873600001311
Then the environmental factor
Figure BDA00029756873600001312
Is a Minimum Variance Unbiased Estimation (MVUE) of
Figure BDA00029756873600001313
Wherein T isx,TyAre the statistics of X and Y.
For an exponential distribution X to E (λ), the coefficient of variation of X is
Figure BDA00029756873600001314
Let X, Y obey an exponential distribution X-E (lambda)x),Y～E(λy) Independent of each other and having the same coefficient of variation, or environmental factor
Figure BDA00029756873600001315
Taking a sample X with the capacity of n in X1,x2,…,xi,…,xnTaking a sample Y with capacity m in Y1,y2,...,yi,...,ymRecording the mean value of the sampleIs otherwise provided with
Figure BDA00029756873600001316
The sample variances are respectively
Figure BDA00029756873600001317
Then the environmental factor
Figure BDA00029756873600001318
Is a Minimum Variance Unbiased Estimation (MVUE) of
Figure BDA00029756873600001319
For a gamma distribution X to gamma (α, β), the coefficient of variation of X is
Figure BDA00029756873600001320
Let X, Y obey the gamma distribution X-gamma (alpha)x,βx),Y～Γ(αy,βy) Independently of each other, the environmental factor alpha must be such that they have the same coefficient of variationx＝αy. Taking a sample X with the capacity of n in X1,x2,…,xi,…,xnTaking a sample Y with capacity m in Y1,y2,...,yi,...,ymThe mean values of the samples are respectively
Figure BDA0002975687360000141
The sample variances are respectively
Figure BDA0002975687360000142
Then the environmental factor
Figure BDA0002975687360000143
Is a Minimum Variance Unbiased Estimation (MVUE) of
Figure BDA0002975687360000144
Based on the three methods, a maintenance interval algorithm based on multi-source information fusion is provided, and the flow chart of the algorithm is shown in FIG. 2:
step 1: determining a maintenance interval based on a maintenance process
Step 1.1: establishing a log-likelihood function of a degradation model based on historical performance degradation data, and estimating prior distribution of model parameters by using a maximum likelihood and one-dimensional search method;
step 1.2: after real-time performance degradation monitoring data of the structure are obtained, updating the structure performance degradation model parameters in real time by using a Bayes method;
step 1.3: solving the parameters u, sigma in the structural performance degradation model2An expected estimate of the posterior distribution of (a);
step 1.4: a life satisfying a reliability of 0.9 or more is obtained. The time interval during which the lifetime is present is the maintenance interval.
Step 2: determining maintenance intervals based on FD
Step 2.1: initializing and setting;
step 2.2: judging whether to perform initial check threshold value calculation and repeated check threshold value calculation;
step 2.3: calculating an initial threshold value by using an initial inspection threshold formula;
step 2.4: the duplicate inspection threshold value is calculated using a duplicate inspection threshold formula.
And step 3: statistical conversion-based method
Step 3.1: selecting a statistical distribution (e.g., normal distribution, exponential distribution, etc.) for analysis;
step 3.2: analyzing whether the statistical distribution satisfies a hypothesis test. If yes, the step 3.3 is carried out, and if not, the step 3.1 is returned to;
step 3.3: performing statistical conversion using distributions satisfying hypothesis testing;
step 3.4: and obtaining the maintenance interval based on the converted statistical result.
And 4, step 4: a bayesian fusion method is used to obtain the final maintenance interval. Information fusion is the comprehensive application of a plurality of models collected in one frame. In multi-model prediction, when the effect of a model is excellent, a larger weight is given to the model, and conversely, a smaller weight is given to the model. The calculation effect is that in comparison with the actual value error, the smallest actual value error should be given a larger weight.
After the Bayesian model fusion, the expected value and variance of the posterior can be expressed as:
Figure BDA0002975687360000151
Figure BDA0002975687360000152
wherein the content of the first and second substances,
Figure BDA0002975687360000153
is a model MjCalculated variance, ρ, based on dataset DjAre the corresponding probabilities of discrete random variables. As can be seen from equation (13), the variance of bayesian fusion includes two terms: firstly, evaluating the dispersion degree in a set; the second is the variance of the model itself.
Take a model SSI 55-10-03 (wing rib, horizontal stabilizer) as an example. The horizontal stabilizer rib is positioned in the wing tip cover, between the front beam and the rear beam and the upper and lower side wall plates, and in the main bearing wing box, and is mainly used for maintaining the pneumatic appearance and mainly bearing the flight load. The main structure comprises wing ribs, rib angle pieces and connecting angle pieces.
The end sealing ribs (No. 12 ribs) at two ends of the overhanging section and the No. 1 ribs at the root part are coordinated and connected with surrounding parts more, the shapes of the parts are relatively complex, an aluminum alloy thick plate machine made of 7050-T7451 is selected for adding, No. 2-10 ribs are selected for mainly maintaining the pneumatic appearance and not participating in the integral force transmission of a vertical tail, a CFRP honeycomb sandwich structure is selected for paving, and then the materials are subjected to high-temperature curing molding by a hot pressing tank; the rib angle sheet for connecting the wing ribs and the stringer adopts a CFRP honeycomb sandwich structure; the connecting angle piece for connecting the No. 1 rib with the front and rear beams is made of 7050-T7451 aluminum alloy material, as shown in figure 3.
Based on the results of the previous analysis, based on the rules of MSG-3 herein and the selection rules of newly added maintenance items. SSI 55-10-03 generates the following MSG-3 tasks:
TABLE 1 SSI maintenance item selection
Figure BDA0002975687360000161
To test the effectiveness of the invention, analysis was performed using a wiener process model, FD method, statistical conversion method and the method proposed by the present invention. Table 2 shows the experimental results of maintenance item 1.
Table 2 maintenance item 1 experimental results
Accuracy of measurement The invention Wiener process Statistical conversion FD
Airline
1 data 0.917 0.820 0.70 0.706
Airline 2 data 0.916 0.832 0.70 0.704
Airline 3 data 0.916 0.830 0.72 0.695
Airline 4 data 0.915 0.811 0.74 0.715
Airline 5 data 0.916 0.812 0.74 0.705
Mean value of 0.916 0.821 0.72 0.705
To better verify the effectiveness of the present invention, further experimental results are given for maintenance project B and maintenance project C, as shown in fig. 4.
Therefore, as can be seen from the results of the experiments in Table 2 and FIG. 4, the method of the present invention has better accuracy than the conventional method.
The embodiments are only for illustrating the technical idea of the present invention, and the technical idea of the present invention is not limited thereto, and any modifications made on the basis of the technical scheme according to the technical idea of the present invention fall within the scope of the present invention.

Claims (5)

1. The method for selecting the aircraft structure maintenance project and analyzing the maintenance interval based on multi-source information fusion is characterized by comprising the following steps of:
(1) comprehensively considering four input sources of special event analysis, damage tolerance data, structure MSG-3 analysis and field repair experience, and selecting maintenance projects;
(2) establishing a maintenance interval model of multi-source information fusion, wherein the model comprises a quintuple (W, FD, ST, I, R), wherein W represents a wiener process model, FD represents a fatigue damage-based method, ST represents a statistical conversion-based method, I represents an integration method based on W, FD and ST, and R represents an obtained result;
(3) a repair interval for the aircraft structure is calculated using the multi-source information-fused repair interval model.
2. The method for selecting aircraft structure maintenance items and analyzing maintenance intervals based on multi-source information fusion of claim 1, wherein the method for selecting maintenance items comprises a method for selecting MSG-3 maintenance items and a method for determining newly added maintenance items.
3. The method for selecting the aircraft structure maintenance project and analyzing the maintenance interval based on the multi-source information fusion of claim 2, wherein the method for selecting the MSG-3 maintenance project comprises fault consequence determination, fault possibility determination and rating system analysis;
the failure consequence determination means that structural items or structural details which have significant influence on the structural integrity of the aircraft flying, landing, supercharging or control load after failure are determined as important positions SSI, and damage tolerance or safe life characteristics of the structural items or the structural details are required to be determined;
the failure possibility determination means that according to effective evaluation of the load condition and the operation environment, a structural project or a region which most possibly shows structural degradation signs firstly is determined, and structural degradation reasons are three basic structural damages including accidental damage, environmental deterioration and fatigue damage;
the rating system analyses as follows:
for each important location SSI, an environmental deterioration level EDR and an accidental damage level ADR are determined, which take into account the detectability and sensitivity to damage and at the same time determine basic structural examination requirements for this important location SSI;
all important positions SSI sensitive to fatigue damage also have to determine the fatigue damage level FDR;
if the detection of fatigue damage to a certain important position SSI before the position SSI reaches the critical crack size cannot be realized, the position SSI is classified into a 'safe life' class; for each "safe life" item, its life limit is determined from analysis and/or experimentation; the important positions SSI of the class "safe life" are limited by environmental degradation and accidental damage assessment, and FDR reflects the detectable probability of fatigue damage before reaching a critical state;
critical damage is calculated using a residual strength analysis program, allowable loads, and typical material properties; crack propagation speed is predicted according to typical service load, bearing range and typical material characteristics; to ensure that damage is detected before critical dimensions are reached, inspection work should be selected based on fatigue and damage tolerance analysis and experience with similar structural designs; when the fatigue and damage tolerance analysis report is approved, the recommended intervals in the analysis report need to be re-evaluated according to the MSG-3 analysis, and the intervals can be further modified along with the fatigue test result of the complete machine or the part.
For the non-metal structure, FDR analysis is not carried out, and EDR and ADR are adopted to formulate an inspection outline;
for metal structures, a fatigue damage analysis procedure was employed.
4. The method for selecting aircraft structure maintenance items and analyzing maintenance intervals based on multi-source information fusion according to claim 2, wherein the method for determining the newly added maintenance items comprises:
selecting rules of maintenance items considering the structural design characteristics;
the special event analysis is to determine important special events occurring in the airplane structure according to the occurrence probability in each use stage;
the damage analysis is to determine important damage according to the maturity of the technology adopted for the structural design of the airplane, the damage type corresponding to the technology and the sensitivity to the damage, and the special events and the damage are finally implemented in specific physical objects, namely potential damage units;
determining the analysis object is to determine the dangerous crisis degree of the potential damage unit exposed to the danger according to the installation position, and continuously analyzing the high-risk unit;
the maintenance task is determined according to the analysis of the harmfulness of the analysis object;
maintenance item selection rules considering field maintenance experience;
a repair item selection rule based on the damage tolerance design analysis;
the source merging principle of each repair task is as follows: judging whether sources of the repair tasks overlap or not by considering the damage occurrence part, the damage mode and the damage influence, and determining a merging principle according to the side key points of the sources;
formal repair item selection principle: and comprehensively considering the damage sensitivity and the damage occurrence probability, and determining a formal repair item selection principle by taking the determined damage sensitivity grade and the calculated or counted damage probability as main indexes.
5. The method for selecting the aircraft structure maintenance project and analyzing the maintenance interval based on the multi-source information fusion as claimed in claim 1, wherein the specific process of the step (3) is as follows:
(3-1) determining a maintenance interval based on the wiener process:
(3-1-1) establishing a log-likelihood function of a degradation model based on historical performance degradation data, and estimating prior distribution of model parameters by using a maximum likelihood and one-dimensional search method;
(3-1-2) acquiring real-time performance degradation monitoring data of the structure, and updating the structure performance degradation model parameters in real time by using a Bayes method;
(3-1-3) obtaining an expected estimated value of posterior distribution of parameters in the structural performance degradation model;
(3-1-4) obtaining a lifetime satisfying the reliability limit, the time interval of which is a maintenance interval;
(3-2) determining a maintenance interval based on the fatigue damage:
(3-2-1) initializing setting;
(3-2-2) judging whether to perform initial check threshold value calculation and repeated check threshold value calculation;
(3-2-3) calculating an initial threshold value;
(3-2-4) calculating a duplicate checking threshold value;
(3-3) determining a maintenance interval based on a statistical conversion method:
(3-3-1) selecting a statistical distribution for analysis;
(3-3-2) analyzing whether the statistical distribution meets the hypothesis test, if so, turning to the step (3-3-3), and if not, returning to the step (3-3-1);
(3-3-3) performing statistical conversion using the distribution satisfying the hypothesis test;
(3-3-4) obtaining a maintenance interval based on the converted statistical result;
and (3-4) obtaining a final maintenance interval by using a Bayesian fusion method.
CN202110273838.8A 2021-03-15 2021-03-15 Aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion Pending CN113158140A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110273838.8A CN113158140A (en) 2021-03-15 2021-03-15 Aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110273838.8A CN113158140A (en) 2021-03-15 2021-03-15 Aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion

Publications (1)

Publication Number Publication Date
CN113158140A true CN113158140A (en) 2021-07-23

Family

ID=76887349

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110273838.8A Pending CN113158140A (en) 2021-03-15 2021-03-15 Aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion

Country Status (1)

Country Link
CN (1) CN113158140A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115221640A (en) * 2022-09-16 2022-10-21 深圳市金精博科技有限公司 Intelligent full-automatic aviation rudder wing inspection system based on infrared induction principle
US11958789B1 (en) * 2022-12-15 2024-04-16 Kunming University Of Science And Technology Method for determining consistency coefficient of power-law cement grout

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115221640A (en) * 2022-09-16 2022-10-21 深圳市金精博科技有限公司 Intelligent full-automatic aviation rudder wing inspection system based on infrared induction principle
CN115221640B (en) * 2022-09-16 2022-12-27 深圳市金精博科技有限公司 Intelligent full-automatic aviation rudder wing inspection system based on infrared induction principle
US11958789B1 (en) * 2022-12-15 2024-04-16 Kunming University Of Science And Technology Method for determining consistency coefficient of power-law cement grout

Similar Documents

Publication Publication Date Title
Goranson Fatigue issues in aircraft maintenance and repairs
Iyyer et al. Aircraft life management using crack initiation and crack growth models–P-3C Aircraft experience
CN113158140A (en) Aircraft structure maintenance project selection and maintenance interval analysis method based on multi-source information fusion
Farrar et al. An introduction to damage prognosis
CN104392122A (en) Probabilistic life evaluation method based on crack detection probability model
Berens et al. Risk analysis for aging aircraft fleets
Lin et al. Probabilistic approach to damage tolerance design of aircraft composite structures
Wanhill Fatigue requirements for aircraft structures
EP2957883A2 (en) Method for prognostics of an aircraft structure based on structural testing
Wanhill Fatigue requirements for aircraft structures
Kordonsky et al. Choice of the best time scale for system reliability analysis
Rusk et al. Bayesian updating of damage size probabilities for aircraft structural life-cycle management
Millwater et al. Probabilistic sensitivity-based ranking of damage tolerance analysis elements
Huang et al. A method for reliability assessment of aircraft structures subject to accidental damage
Wu et al. Probabilistic function evaluation system (ProFES) for reliability-based design
Chen et al. Integration of structural health monitoring with scheduled maintenance of aircraft composite structures
Kim et al. Bayesian approach for fatigue life prediction from field data
Rudd et al. Probabilistic fracture mechanics analysis methods for structural durability
CN114492074A (en) Probabilistic damage tolerance assessment analysis method
Shekhter et al. Assessment of the effect of pitting corrosion on the safe life prediction of the P3-C.
Jing et al. System health management for unmanned aerial vehicle: conception, state-of-art, framework and challenge
Wanhill et al. Fracture mechanics in aircraft failure analysis: Uses and limitations
White Review of methods and approaches for the structural risk assessment of aircraft
Xiong et al. A durability model incorporating safe life methodology and damage tolerance approach to assess first inspection and maintenance period for structures
Molent et al. Verification of an airframe fatigue life monitoring system using ex-service structure

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination