Aircraft storage health state assessment method

Abstract

The invention relates to the technical field of aircraft storage and maintenance support, and particularly discloses an aircraft storage health state assessment method, which comprises the following steps of: firstly, acquiring health state parameters of an aircraft through an acquisition board card; acquiring health state parameter data from the cloud, and inputting the health state parameter data into the embedded platform for processing and analysis; then, according to the health state parameters, health degrees of key single-machine sub-devices, subsystems and aircrafts are generated and visually presented; and finally, according to the health state of each subsystem, carrying out health state evaluation on the aircraft, and giving an equipment combination scheme and evaluation result details. According to the method for evaluating the storage health state of the aircraft, by referring to the actual storage condition of the aircraft, evaluating and analyzing the health state of the batch of aircrafts and reasonably matching resources, the overall health state of the aircrafts of the same batch is optimized, reference is provided for maintenance personnel, the guarantee ability of the aircrafts is improved, and the maintenance efficiency is improved. The waste of maintenance resources is reduced, and the maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft storage maintenance and support, and in particular to a method for evaluating the storage health status of an aircraft. Background Art

[0002] According to the development trend of national defense high-tech, the types and numbers of aircraft are increasing, the complexity of system composition is getting higher and higher, the scale is getting larger and larger, and it is gradually developing towards high integration, automation and high efficiency. Traditional fault detection and maintenance support technology can no longer meet the precise maintenance and support needs of modern aircraft and modern warfare. Therefore, higher requirements are also placed on the reliability, maintainability and comprehensive support system of aircraft.

[0003] Aircraft have the characteristics of "long-term storage, small amount of testing, and one-time use". The health status assessment and prediction method of long-term storage of aircraft will help to realize the transition of equipment health management technology from theoretical research to engineering practice application in the aerospace field, effectively improve the support capability of aerospace system aircraft, reduce the waste of maintenance resources, and reduce maintenance costs. At the same time, it can also be promoted and applied to other large and complex equipment fields, which has certain social value and economic benefits. Summary of the invention

[0004] The purpose of the present invention is to provide a method for evaluating the storage health status of an aircraft, which is suitable for autonomous health management and maintenance support of aircraft equipment. By scientifically evaluating the health status of the aircraft and predicting its health status development trend based on test parameters, a maintenance plan can be made in advance to eliminate potential faults in the bud as much as possible, and the performance of the aircraft can be maximized in terms of comprehensive support and on-site combat.

[0005] To achieve the above object, the present invention provides a method for evaluating the storage health status of an aircraft, comprising the following steps:

[0006] S1. Collect the health status parameters of the aircraft through the acquisition board;

[0007] S2, obtain health status parameter data from the cloud and input it into the embedded platform for processing and analysis;

[0008] S3. Generate the health status of key stand-alone sub-equipment, subsystems, and aircraft based on health status parameters, and present them visually;

[0009] S4. Evaluate the health status of the aircraft based on the health status of each subsystem, and provide an equipment combination plan and detailed evaluation results.

[0010] Preferably, in step S1, the principles to be followed when collecting the health status parameters are: testability, representativeness, systematicity, and a combination of qualitative and quantitative indicators.

[0011] Preferably, the specific process of step S2 is: using various sensors and single-machine testing technologies to collect full life cycle data of aircraft equipment, including test parameters and environmental information of each key single-machine sub-equipment, and using normalization and outlier processing to process the original test data as input parameters of the health status assessment and prediction algorithm.

[0012] Preferably, the specific process of step S3 is: select parameter indicators of key stand-alone sub-equipment in each subsystem as health status assessment parameters, first normalize the test data, determine the weight of each indicator, then make a comprehensive decision based on the health status assessment algorithm, and finally obtain the health status level and health degree of each subsystem.

[0013] Preferably, in step S3, the test data x is compared with the standard value x s For comparison, the normalized parameters will be mapped to [0,1]. Assuming that the threshold range of the health status test parameter is min max[x min ,x max ], the specific calculation process of the normalized value is as follows:

[0014]

[0015] When evaluating health status, it is necessary to highlight the parameters with poor status and give them a greater weight, that is, the larger the normalized value, the greater the weight:

[0016]

[0017] Among them, λ i is the normalized value;

[0018] Assume that the number of health assessment indicators of aircraft is n, the number of aircraft in the same batch is m, and x ij represents the test value of the jth health assessment index of the i-th aircraft, and the information entropy of the j-th health status assessment index is recorded as:

[0019]

[0020] Among them, p ij is the normalized value of the indicator, then the weight of the jth indicator calculated using the entropy weight method is:

[0021]

[0022] The aircraft has n health status assessment indicators, and the final weight of the i-th indicator is:

[0023]

[0024] Among them, ω i1 ,ω i2 are the weights of the i-th indicator calculated using the test parameter normalization method and the entropy weight method, respectively.

[0025] Preferably, in step S3, assuming there are n pieces of evidence, the final weight vector is ω=(ω1,ω1,…,ω n ), then the discount rate of evidence is expressed as:

[0026] α i =ω i / ω max , where ω max =max(ω1,ω1,…,ω n ), and use the evidence discount rate to modify the evidence source. The basic probability after modification is:

[0027] m′ i (A k )=α i m i (A k ) (6)

[0028] Among them, ω i represents the weight of the i-th source of evidence;

[0029] The additional basic probability assignment function added after evidence discount is:

[0030]

[0031] Among them, m′ i (Θ) is the supplementary basic probability assignment function added after evidence discounting;

[0032] The evidence sources corrected by the discount rate are fused n-1 times using the Dempster synthesis rule. The corresponding synthesis rule is recorded as:

[0033]

[0034] Among them, α i is the discount rate mentioned above, m i (A k ) is the basic probability assignment in the above text;

[0035] If there is a conflict in the evidence, the Jousselme distance function is introduced to calculate the credibility of each piece of evidence. Assume that m i and m jis the evidence in the aircraft health status assessment framework and the corresponding basic probability distribution, then the Jousselme distance is:

[0036]

[0037] The similarity between the two is calculated based on the Jousselme distance between them:

[0038] S(m i ,m j )=1-d BPA (m i ,m j ) (10)

[0039] After weighted averaging, we get the average similarity between this evidence and other evidence:

[0040]

[0041] The credibility of the i-th evidence is recorded as:

[0042]

[0043] Preferably, the health status levels are divided into: excellent, good, fair, observation and overhaul.

[0044] Preferably, in step S4, the health status of the aircraft is evaluated, including evaluating and predicting the health status of a single aircraft and evaluating and predicting the health status of a batch of aircraft.

[0045] Preferably, in step S4, a maintenance plan is formulated based on the aircraft health status assessment prediction result, including the maintenance / repair task type, start and end time, required maintenance / repair resources and maintenance personnel information.

[0046] Therefore, the present invention adopts the above-mentioned aircraft storage health status assessment method, and the beneficial effects are as follows:

[0047] (1) The health assessment system of the present invention modifies the evidence source according to the "discount rate". In the process of information fusion, the evidence source is first pre-processed using the evidence "discount rate". When there is conflicting evidence, it is not completely denied, but replaced by average evidence, thus avoiding the loss of original evidence source information.

[0048] (2) Based on the validity and relevance of the evidence itself, the present invention introduces the evidence "discount rate" to correct the evidence source, uses the Josselme distance function to calculate the credibility of the evidence, replaces the conflicting evidence with the weighted average evidence, and uses the improved DS evidence theory to fuse the test data information. This can solve the problem of evidence conflict, realize the health status assessment of key single-machine sub-equipment, and further quantify the health status level into health degree, which is convenient for quantitative analysis and comparison.

[0049] (3) The present invention proposes an equipment combination optimization method based on a genetic algorithm, which can maintain the overall health status of the equipment at a relatively stable level and make the overall health status of the batch of equipment reach the optimal level.

[0050] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flow chart of a health status assessment module of an embodiment of an aircraft storage health status assessment method of the present invention;

[0052] Figure 2 It is a flow chart of establishing HAFPN in an embodiment of an aircraft storage health status assessment method of the present invention;

[0053] Figure 3 A Petri net model for evaluating the health status of a single aircraft in an embodiment of an aircraft storage health status evaluation method of the present invention;

[0054] Figure 4 The present invention is a flowchart of a genetic algorithm searching for an optimal equipment combination in an embodiment of a method for evaluating the storage health status of an aircraft. DETAILED DESCRIPTION

[0055] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0056] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0057] like Figure 1 As shown in the figure, a method for evaluating the storage health status of an aircraft is mainly used for the health management and maintenance of an aircraft during long-term storage. The health status of key stand-alone sub-equipment, aircraft subsystems and system level is evaluated through a multi-level health assessment system.

[0058] First, a three-level health status assessment sequence of "key stand-alone sub-equipment-subsystem-aircraft" is established, including data cloud storage and PC. The data cloud storage is for the aircraft manager to upload the monitored aircraft health status parameters to the cloud, and the PC input is the health status parameters in the cloud. The method includes the following steps:

[0059] S1. The health status parameters of the aircraft are collected through the acquisition board. The principles to be followed when collecting health status parameters are: testability, representativeness, systematicity, and a combination of qualitative and quantitative indicators.

[0060] Testability: When selecting health assessment indicators, the first consideration is the testability of the indicator to ensure that the test data can be directly obtained through testing;

[0061] Representativeness: There are many indicators that affect the health status of aircraft. Usually, only the main factors are selected for analysis. However, it is necessary to ensure that the established indicator system fully reflects the current health status of the aircraft, that is, the indicators selected must be typical and representative;

[0062] Systematicity: Since the system composition of the aircraft has obvious hierarchical characteristics, the relationship between indicators should be fully considered when selecting indicators. The selected indicators should have clear connotations and be able to meet the hierarchical relationship between layers to avoid overlap and intersection;

[0063] Combining qualitative and quantitative indicators When evaluating the health status of an aircraft, a comprehensive evaluation should be conducted from both qualitative and quantitative perspectives.

[0064] S2. Obtain health status parameter data from the cloud and input it into the embedded platform for processing and analysis. The specific process is: use various sensors and single-machine testing technologies to collect the full life cycle data of aircraft equipment, including the test parameters and environmental information of each key single-machine sub-equipment, and use normalization, outlier and other data processing technologies to process the original test data as input parameters for the health status assessment and prediction algorithm.

[0065] S3. Generate the health of key stand-alone sub-equipment, subsystem, and aircraft based on the health status parameters and present them visually. The specific process of health status assessment of key stand-alone sub-equipment and subsystem level is as follows: select parameter indicators of key stand-alone sub-equipment in each subsystem as health status assessment parameters, first normalize the test data, determine the weight of each indicator, and then make a comprehensive decision based on the health status assessment algorithm, and finally obtain the health status level and health of each subsystem.

[0066] The method for determining the health index weight in this embodiment is as follows:

[0067] (1) Test parameter normalization

[0068] Since the dimensions of the various test parameters of the aircraft are different and cannot be directly compared, the test parameters need to be normalized first:

[0069] Compare the test data x with the standard value x s For comparison, the normalized parameters will be mapped to [0,1]. Assuming that the threshold range of the health status test parameter is min max[x min ,x max ], the specific calculation process of the normalized value is as follows:

[0070]

[0071] Among them, λ i is the normalized value;

[0072] When evaluating health status, it is necessary to highlight the parameters with poor status and give them a greater weight, that is, the larger the normalized value, the greater the weight:

[0073]

[0074] (2) Assume that the number of health assessment indicators of aircraft is n, the number of aircraft in the same batch is m, and x ij represents the test value of the jth health assessment index of the i-th aircraft, and the information entropy of the j-th health status assessment index is recorded as:

[0075]

[0076] Where: p ij is the normalized value of the indicator, then the weight of the jth indicator calculated using the entropy weight method is:

[0077]

[0078] Combining the above two methods, the aircraft has n health status assessment indicators, and the final weight of the i-th indicator is:

[0079]

[0080] Among them, ω i1 ,ω i2 are the weights of the i-th indicator calculated using the test parameter normalization method and the entropy weight method, respectively.

[0081] like Figure 1 As shown, an improved information fusion method is used to perform aircraft health assessment:

[0082] Assume there are n pieces of evidence, and the final weight vector is ω=(ω1,ω1,…,ω n ), then the “discount rate” of the evidence is expressed as:

[0083] α i =ω i / ω max , where ω max =max(ω1,ω1,…,ω n ), and use the evidence “discount rate” to modify the evidence source. The basic probability assignment after modification is:

[0084] m′ i (A k )=α i m i (A k ) (6)

[0085] Among them, ω i represents the weight of the i-th source of evidence;

[0086] The additional basic probability assignment function added after the evidence "discount" is:

[0087]

[0088] Among them, m′ i (Θ) is the supplementary basic probability assignment function added after evidence discounting;

[0089] The evidence sources modified by the “discount rate” are fused n-1 times using the Dempster synthesis rule. The corresponding synthesis rule is recorded as:

[0090]

[0091] Among them, α i is the discount rate mentioned above, m i (A k ) is the basic probability assignment in the above text;

[0092] If there is a conflict in the evidence, the Jousselme distance function is introduced to calculate the credibility of each piece of evidence. Assume that m i and m j is the evidence in the aircraft health status assessment framework and the corresponding basic probability distribution, then the Jousselme distance is:

[0093]

[0094] The similarity between the two is calculated based on the Jousselme distance between them:

[0095] S(m i ,m j )=1-d BPA (m i ,mj ) (10)

[0096] After weighted averaging, we can get the average similarity between this evidence and other evidence:

[0097]

[0098] The credibility of the i-th evidence is recorded as:

[0099]

[0100] The health status level of an aircraft can be differentiated according to the degree of deviation of the test parameters from the standard value or threshold range. It can be divided into five levels: excellent, good, general, observation and maintenance.

[0101] The "excellent", "good" and "average" conditions of the aircraft are all healthy, and no maintenance measures are needed for the time being; the "observation" state is a sub-healthy state. Since the health of the aircraft tends to deteriorate over time, aircraft in a sub-healthy state need to attract the attention of maintenance and support personnel. The test cycle can be appropriately shortened, and corresponding maintenance plans can be formulated based on the health status assessment and prediction results.

[0102] The health degree indicates the degree to which the current health state of the aircraft deviates from the standard health state. It is represented by HD (Health Degree) and has a value range of [0,1]. When HD = 0, it means that the aircraft is faulty and will be scrapped; when HD = 1, it means that the aircraft is in good health and can perform tasks normally. The health state level assessment results and health degree scoring results can further quantify the health level and make the fuzzy quantity precise. Assume that the membership degree at a certain health state level is m i , then the calculation formula of health HD is:

[0103] HD=0.1×m1+0.3×m2+0.5×m3+0.7×m4+0.9×m5 (13)

[0104] The m in this formula i is the membership degree of a certain health status level. The membership degree uses the ridge distribution membership function to describe the uncertainty relationship between the various health states of the aircraft. The formula of the ridge distribution membership function is:

[0105]

[0106] In formula (14), a and b represent the upper limit and lower limit, respectively. Taking the normalized value of the evaluation index test parameter as the variable, the ridge membership function corresponding to the excellent, good, general, and observation states can be recorded as:

[0107]

[0108] This membership degree can be used as the basic probability assignment BPA in the DS evidence theory. After being fused using the Dempster synthesis rule, the proposition corresponding to the largest basic probability assignment is the health status level of the aircraft. The health status level is further quantified according to formula (13) to obtain the health status of the aircraft.

[0109] S4. Evaluate the health status of the aircraft based on the health status of each subsystem, and provide an equipment combination plan and detailed evaluation results.

[0110] Conduct health status assessment on aircraft, including health status assessment and prediction of single aircraft and health status assessment and prediction of batches of aircraft.

[0111] Single aircraft health status assessment and prediction: Based on the health status of each key single-machine sub-equipment and subsystem, a system-level health status assessment of a single aircraft is performed. Based on the assessment results, the current health status of the single aircraft can be obtained.

[0112] Batch aircraft health status assessment prediction: In the actual storage process, since each subsystem and key single-machine sub-equipment are stored separately, the assessment of the health status of batch aircraft in the same batch is essentially a problem of optimal combination. How to randomly combine all current subsystems and key single-machine sub-equipment to make the health status of the entire batch of aircraft reach the optimal state to meet the needs of large-scale operations is the optimal solution to this problem.

[0113] In view of the hierarchical composition of aircraft systems, the concept of fuzzy Petri net for health status assessment is introduced. The health status of key single-machine sub-equipment is taken as the input place, and the hierarchical analysis method is used to analyze the weights of key single-machine sub-equipment and subsystems in evaluating high-level health status. The weights are used as the weights of input places and transition places. Through the place transition rules, the health status of each subsystem and the aircraft system level is gradually inferred.

[0114] The structure of the aircraft health status assessment fuzzy Petri net is established. When evaluating the health status of a certain aircraft (subsystem), the system will first automatically select the health status of the target aircraft (subsystem)'s front library, solve the health status of the target library according to the library transition rules, and visualize it in the interface.

[0115] The forward reasoning of the aircraft health status assessment fuzzy Petri net (HAFPN) can reflect the dynamic change process of the health status of each hierarchical structure of the entire aircraft system. The process of establishing HAFPN is as follows: Figure 2 As shown, first determine the key stand-alone sub-equipment and subsystem composition, then establish the input library, transition library and termination library, scan and update the status of the input library to determine the health status of the key stand-alone sub-equipment, determine whether the change is triggered, if so, proceed to the next step, otherwise repeat the scan and update, then determine whether the termination library is terminated, if so, proceed to the next step, otherwise repeat the scan and update, then update the transition library status, determine the subsystem health status, update the termination library status, and determine the system health status.

[0116] The Petri net model for evaluating the health status of a single aircraft is as follows: Figure 3 As shown, most of the low-level key stand-alone sub-devices or subsystems directly affect the health status of the high-level subsystem or system level, and the input library represents the health status of each key stand-alone sub-device.

[0117] The aircraft system is a temporary combination based on battlefield conditions, mission types and combat objectives. The optimal solution for the equipment combination is found by using a combination optimization method based on genetic algorithms. The basic steps of the solution are as follows: Figure 4 As shown:

[0118] 1. Coding: Coding is a mapping process from genotype to expression form. Due to the complex combination of aircraft, integer coding is selected. The gene is consistent with the phenotype. The coding length is 50, which respectively represents the permutations and combinations of the five subsystems.

[0119] 2. Selection: The process of selection is to select individuals with strong adaptability and enter the next generation to participate in crossover and mutation.

[0120] 3. Crossover and mutation: Use the selected excellent individuals to perform crossover and mutation to produce new species. Mapping it in the algorithm is the process of jumping out of the local optimal solution.

[0121] 4. Verification: Use the fitness function to verify the new offspring. If the fitness function is met, the offspring will replace the chromosomes with insufficient fitness in the population.

[0122] Repeat steps 2, 3, and 4 until the optimal solution is found and decoded, which is the optimal aircraft combination solution.

[0123] Based on the aircraft health status assessment and prediction results, a maintenance plan is formulated, including the type of maintenance / repair task, start and end time, required maintenance / repair resources and maintenance personnel information.

[0124] Therefore, the present invention adopts the above-mentioned aircraft storage health status assessment method, which is suitable for autonomous health management and maintenance support of aircraft equipment. By scientifically assessing the health status of the aircraft and predicting its health status development trend based on test parameters, a maintenance plan can be made in advance to eliminate potential faults in the bud as much as possible, and the performance of the aircraft can be maximized in terms of comprehensive support and on-the-spot combat.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for evaluating the storage health status of an aircraft, characterized in that: The following steps are involved: S1. Collect the health status parameters of the aircraft through the acquisition board; S2, obtain health status parameter data from the cloud and input it into the embedded platform for processing and analysis; S3. Generate the health status of key stand-alone sub-devices, subsystems, and aircraft based on health status parameters and present them visually; S4. Evaluate the health status of the aircraft based on the health status of each subsystem, and provide an equipment combination plan and detailed evaluation results.

2. The method for evaluating the storage health status of an aircraft according to claim 1, wherein: In step S1, the principles to be followed when collecting the health status parameters are: testability, representativeness, systematicity, and a combination of qualitative and quantitative indicators.

3. The method for evaluating the storage health status of an aircraft according to claim 2, wherein: The specific process of step S2 is: using various sensors and single-machine testing technologies to collect data on the entire life cycle of aircraft equipment, including test parameters and environmental information of each key single-machine sub-equipment, and using normalization and outlier processing to process the original test data as input parameters for the health status assessment and prediction algorithm.

4. The method for evaluating the storage health status of an aircraft according to claim 3, wherein: The specific process of step S3 is: select the parameter indicators of key single-machine sub-equipment in each subsystem as health status assessment parameters, first normalize the test data, determine the weight of each indicator, then make a comprehensive decision based on the health status assessment algorithm, and finally obtain the health status level and health degree of each subsystem.

5. The method for evaluating the storage health status of an aircraft according to claim 4, wherein: In step S3, the test data x is compared with the standard value x s For comparison, the normalized parameters will be mapped to [0,1]. Assuming that the threshold range of the health status test parameter is min max [x min ,x max ], the specific calculation process of the normalized value is as follows: When evaluating health status, it is necessary to highlight parameters with poorer status and give them a greater weight. That is, the larger the normalized value, the greater the weight: Among them, λ i is the normalized value; Assume that the number of aircraft health assessment indicators is n, the number of aircraft in the same batch is m, and x ij represents the test value of the jth health assessment index of the i-th aircraft, and the information entropy of the j-th health status assessment index is recorded as: Among them, p ij is the normalized value of the indicator, then the weight of the j-th indicator calculated using the entropy weight method is: The aircraft has n health status evaluation indicators, and the final weight of the i-th indicator is: Among them, ω i1 、ω i2 are the weights of the i-th indicator calculated using the test parameter normalization method and entropy weight method, respectively.

6. The method for evaluating the storage health status of an aircraft according to claim 5, wherein: In step S3, assuming there are n pieces of evidence, the final weight vector is ω=(ω1,ω1,…,ω n ), then the discount rate of evidence is expressed as: α i =ω i / ω max , where ω max =max(ω1,ω1,…,ω n ), and use the evidence discount rate to modify the evidence source. The basic probability after modification is: m' i (A k )=a i m i (A k ) (6) Among them, ω i represents the weight of the i-th evidence source; The supplementary basic probability assignment function added after evidence discount is: Among them, m' i (Θ) is the supplementary basic probability assignment function added after evidence discounting; The evidence sources corrected by the discount rate are fused n-1 times using Dempster's synthesis rule. The corresponding synthesis rule is recorded as: Among them, α i is the discount rate mentioned above, m i (A k ) is assigned the basic probability above; If there is a conflict in the evidence, the Josselme distance function is introduced to calculate the credibility of each piece of evidence. Assume that m i and m j is the evidence and the corresponding basic probability distribution in the aircraft health status assessment framework, then the Jouselme distance is: The similarity between the two is calculated based on the Jousellme distance between them: S(m i ,m j )=1-d BPA (m i ,m j ) (10) After weighted averaging, we get the average similarity between this evidence and other evidence: The credibility of the i-th evidence is recorded as:

7. The method for evaluating the storage health status of an aircraft according to claim 6, wherein: The health status levels are divided into: excellent, good, fair, observation and maintenance.

8. The method for evaluating the storage health status of an aircraft according to claim 7, wherein: In step S4, the health status of the aircraft is evaluated, including the health status evaluation and prediction of a single aircraft and the health status evaluation and prediction of a batch of aircraft.

9. The method for evaluating the storage health status of an aircraft according to claim 8, characterized in that: In step S4, a maintenance plan is formulated based on the aircraft health status assessment prediction results, including the maintenance / repair task type, start and end time, required maintenance / repair resources and maintenance personnel information.

Images