Reliability index optimization allocation method based on operation data

Abstract

The application discloses a reliability index optimization distribution method based on operation data, and belongs to the technical field of airplane operation data optimization; the specific steps comprise the following steps that an operation parameter system evaluation model is constructed; a parameter influence weight coefficient of a product is calculated; operation data is configured to calculate operation parameters; steps 2 to 3 are repeated, operation parameters of all products in a system to be distributed are calculated, and the proportional importance of each parameter is calculated; and the reliability index optimization distribution result is generated in combination with the parameter influence weight coefficient and the parameter proportional importance. The operation data of the civil airplane is used to feedback the rationality of a design link, and effective support is provided for forming product performance promotion. Meanwhile, in the method, the analytic hierarchy process and proportional distribution are combined, the weighted coefficient is used to balance the influence of four types of properties, i.e., reliability, maintainability, usability and supportability, on the product performance, and finally, the obtained reliability distribution index considers comprehensive factors and is closer to the real operation state of each product in the system.

Description

Technical Field

[0001] This invention belongs to the field of aircraft operation data optimization technology, specifically relating to a method for optimizing the allocation of reliability indicators based on operation data. Background Technology

[0002] The development of operational support technologies and integrated support systems plays a crucial role in improving the continued airworthiness of civil aircraft. Conducting reliability analysis, fault diagnosis, and prediction based on operational data is an important way to continuously improve product performance. Currently, major foreign civil aircraft manufacturers (such as Airbus and Boeing) have established comprehensive operational data management systems and mature data acquisition, analysis, and feedback processes. For example, Boeing's "ISDP" program is a procedure in which Boeing collects reliability data, analyzes and provides feedback, develops measures, and supports users. Through relevant information from fleet operations and product reliability data, Boeing proactively supports users and continuously optimizes product design.

[0003] Airbus's Skywise intelligent aviation data platform implements the entire process of reliability data collection, analysis, and feedback. For example, when setting reliability targets for the A380, it extensively referenced aircraft projects such as the A340 and collected and analyzed historical fleet operating data to support the setting of specific reliability and direct maintenance cost targets for the A380. Domestic research has also incorporated several operational data points, such as maintenance frequency and importance, to conduct a series of optimization efforts regarding the allocation of reliability targets. Chen Mingsheng et al. introduced the number of landing gear system maintenance times in four quarters into the allocation of reliability indicators for the landing gear system, and found key elements based on grey relational analysis (Research on Reliability Relationship of Landing Gear [C]. Beijing: China Civil Aviation Magazine, 2015, 8: 309-311.); Wang Honghai et al. effectively combined the proportional combination method and expert scoring allocation method commonly used in actual engineering, providing a simple and easy-to-implement method for task reliability allocation in hybrid models in actual engineering (Task Reliability Allocation Method of Hybrid Models in Actual Engineering [J]. Progress in Aeronautical Engineering, 2010, 1(4): 101-105.); Qian et al. generated weight vectors to correct allocation influencing factors by combining grey relational analysis and expert scoring before linearly allocating each subsystem, thereby making up for the index allocation problem of insufficient design data events (Research of reliability allocation based on grey theory [J]. IOP Conference Series: Materials Science and Engineering, 2021, 1043(10): 1-9.); Similar to the previous paper, Wang et al. used grey relational analysis to generate weight functions, but used analytic hierarchy process in the construction of evaluation index system to further eliminate the influence of human factors in the algorithm (Research on comprehensive assessment method of emergency reliability evaluation of distribution network based on reformative grey clustering[J]. Electrical Measurement & Instrumentation, 2017, 54(18): 22-29.).

[0004] However, the above studies still have certain shortcomings. For example, the redistribution only considers two types of parameters among reliability, availability, maintainability, and supportability, and cannot cover all feedback links during operation. In addition, methods based on expert scoring rely too much on human experience and ignore the real results of operational data feedback. At present, the allocation of reliability indicators in China considers only one factor, making it difficult to evaluate the true reliability indicator requirements of products. The indicator allocation results cannot comprehensively reflect the reliability, maintainability, availability, and supportability requirements of civil aircraft throughout their entire life cycle. Summary of the Invention

[0005] The technical problem to be solved:

[0006] To avoid the shortcomings of existing technologies, this invention proposes a reliability index optimization allocation method based on operational data. It introduces the analytic hierarchy process to construct an evaluation model of the operational parameter system, generates parameter influence weight coefficients, and combines the proportional allocation method to optimize the reliability index allocation design. The allocation results reflect the reliability, maintainability, availability, and supportability index requirements of each product in the system under real operating conditions.

[0007] The technical solution of this invention is: a reliability index optimization allocation method based on operational data, characterized by the following specific steps:

[0008] Step 1: An evaluation model for the operating parameter system applicable to civil aircraft was established, consisting of two layers: the first layer is performance evaluation, and the second layer is the operating parameter system.

[0009] Step 2: Calculate the parameter impact weight index of the product;

[0010] a) Use a pairwise comparison method to evaluate the model layer by layer from the bottom layer, score the parameters or attributes, use the 1-9 scale method for scoring, and construct the judgment matrix G;

[0011] b) Perform a consistency check on the normalized judgment matrix G; after the check, solve for the weight vector w belonging to the eigenvalue λ using the following formula. max The eigenvectors are normalized.

[0012] Gw=λ max w

[0013] Where, λ max To determine the largest eigenvalue of matrix G;

[0014] Repeat steps a) and b) until both evaluation models have completed their scoring. Let the weight index of the first layer be w. i (i = 1, 2, 3, 4), the weight index of the second layer implying membership relationships is w. ij(j=1, 2, 3……9), then the final weight index of the parameter is calculated according to equation (4):

[0015] w iij =w i ·w ij ;

[0016] Step 3: The calculation methods for various operating parameters are as follows:

[0017]

[0018]

[0019]

[0020]

[0021] Sensitivity M is the result of numerical analysis calculation obtained from constructing the reliability model;

[0022] Among them, NUM rem The numbers represent the number of unplanned product replacements, h represents the average monthly flight hours, MTTR represents the average workshop repair time, and NUM represents the average monthly repair time. rep Indicates the number of product unit reports, NUM del-mech Indicates product mechanical delay events; MTBS indicates mean time to spare parts; NUM indicates... del-sup Indicates product warranty delay events, NUM install This indicates the number of products installed, MON indicates the cumulative number of months for data accumulation, and NFF indicates the aircraft's fault-free factor.

[0023] Step 4: Repeat steps 2 to 3 to calculate the operating parameters of all n products in the system to be assigned;

[0024] Step 5: Introduce parameter influence weighting coefficients to perform weighted calculations on the proportional importance of operating parameters. The product reliability index optimization allocation result is calculated according to the following formula; when the system as a whole meets the index requirements, the reallocation work is completed.

[0025]

[0026] Where, λ * k —Assign new reliability metrics to products or subsystems, namely failure rate; λ S * —The overall reliability index to be assigned to the system, i.e., the failure rate; λ s —Total system failure rate.

[0027] A further technical solution of the present invention is as follows: In step 1, the first-level performance evaluation of the operating parameter system evaluation model applicable to civil aircraft includes reliability, maintainability, availability, and supportability, wherein reliability describes the quality characteristics or importance of the product itself; maintainability describes the ease with which the product can be repaired; availability describes the ability of the product to complete its tasks during use; and supportability describes the ability of the product to obtain support support on the ground. The second-level operating parameter system includes failure rate and sensitivity decomposed by reliability, average repair rate and crew reporting rate decomposed by maintainability, unplanned replacement rate, daily utilization rate, and mechanical delay rate decomposed by availability, and average spare parts availability rate and support delay rate decomposed by supportability.

[0028] A further technical solution of the present invention is: in step 2, the method of performing a consistency check on the normalized judgment matrix G is as follows: if CR satisfies CR=CI / RI<0.1, then the judgment matrix G meets the consistency requirements; otherwise, it is considered that the judgment matrix G does not have sufficient consistency and needs to be adjusted again.

[0029] Where: CR is the random consistency ratio of the judgment matrix; RI is the average random consistency index of the judgment matrix; CI is the consistency index of the judgment matrix.

[0030]

[0031] Beneficial effects

[0032] The beneficial effects of this invention are as follows: This invention proposes a reliability index optimization allocation method based on operational data, which rationally utilizes effective operational data collected from civil aircraft to optimize the product reliability index allocation results. This invention collects operational data from a certain type of civil aircraft over five years of operation, and uses this method to optimize the reliability model of an asymmetric mission with left and right flap deflection angles for this type of civil aircraft (as attached). Figure 2 The reliability index redistribution work was carried out for the 13 products covered in the model shown in Table 2 to 3. First, the operational reliability parameters involved in the evaluation model were collected and organized. Second, the parameter influence weight index of the product was calculated by combining the analytic hierarchy process. Taking the P-ACE finished product (dual channel) as an example, the reliability index redistribution weight index is shown in Table 4. Finally, the proportional importance of the parameters was calculated, as shown in Table 5. The reliability index redistribution result was fed back according to Equation (7) combined with the weight index and proportional importance. The reliability redistribution index and the actual failure rate of the reliability model of the asymmetric mission reliability model of the left and right flaps of civil aircraft are compared as shown in the appendix. Figure 3 As shown, comparing the reliability redistribution index with the actual product failure rate, if the changes in the two tend to be consistent, it indicates that the redistribution result is reasonable and feasible.

[0033] Table 2 Operational Reliability Parameters (1)

[0034]

[0035] Table 3 Operational Reliability Parameters (2)

[0036]

[0037]

[0038] Table 4. Reassignment of Reliability Index Weighting Coefficients for P-ACE Finished Product (Dual-Channel)

[0039]

[0040] Table 5 Redistribution Results of Reliability Indicators

[0041]

[0042]

[0043] Feedback from civil aircraft operational data on the rationality of the design phase provides effective support for improving product performance within the operational support system. Methodologically, based on proportional allocation and combined with parameter influence weighting coefficients, a weighted balancing of the impact of four properties—reliability, maintainability, availability, and supportability—on product performance is employed. The resulting reliability allocation index comprehensively considers various factors and more closely approximates the actual operational status of products within the system. Attached Figure Description

[0044] Figure 1 A schematic diagram of the specific solution process of this invention;

[0045] Figure 2 A reliability model diagram of a certain type of civil aircraft with asymmetrical flap deflection angles;

[0046] Figure 3 A graph showing the comparison between the reliability index redistribution index and the actual failure rate. Detailed Implementation

[0047] The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention.

[0048] The specific implementation of the reliability index optimization allocation method based on operational data of this invention is as follows:

[0049] Effective operational data for civil aircraft is collected and acquired, including unplanned replacement frequency (NUMrem), average monthly flight hours (h), average workshop repair time (MTTR), number of crew reports (NUMrep), mechanical delay events (NUMdel-mech), mean time to spare parts up (MTBS), maintenance delay events (NUMdel-sup), number of aircraft installed (NUMinstall), cumulative months of data (MON), and aircraft fault-free factor (NFF). This data is then converted into operational parameters using a mathematical model. Based on the proportional importance of these operational parameters combined with their impact weighting coefficients, an optimized allocation of reliability indicators that comprehensively considers the reliability, maintainability, availability, and supportability of civil aircraft is generated. (See attached...) Figure 1 The process involves calculating reliability redistribution using the following steps:

[0050] (1) Construct an evaluation model for the operating parameter system;

[0051] (2) Calculate the parameter influence weighting coefficient of the product;

[0052] (3) Configure running data and calculate running parameters;

[0053] (4) Repeat steps 2 to 3 to calculate the operating parameters of all products in the system to be allocated, and calculate the proportional importance of each parameter;

[0054] (5) Combine the parameter influence weight coefficient and parameter proportion importance to generate the reliability index optimization allocation result.

[0055] Step 1: Based on the mission requirements of civil aircraft throughout their entire life cycle, their performance evaluation can be decomposed into reliability, maintainability, availability, and supportability. Reliability describes the quality characteristics or importance of the product itself; maintainability describes the ease with which the product can be repaired; availability describes the product's ability to complete its mission during use; and supportability describes the product's ability to obtain ground support. After initially decomposing the operational parameter system, some parameters are eliminated based on the adaptability of the methods and criteria. For example, the false alarm rate is eliminated because false alarms primarily indicate sensor quality issues rather than the product itself, and the false alarm rate is included in the operational parameters of electronic products; thus, to avoid overlapping evaluations, this type of data is eliminated. Similarly, the utilization rate of support equipment, which corresponds to the overall aircraft or system level, has little significance for evaluating the reliability of the product or subsystem and can also be eliminated. Finally, an operational parameter system evaluation model suitable for civil aircraft is established, as shown in the attached figure. Figure 1 As shown.

[0056] Step 2: After constructing the evaluation model of the civil aircraft's operating parameters, the analytic hierarchy process (AHP) is introduced to calculate the parameter influence weights of the product. First, a pairwise comparison method is used to judge from the bottom layer of the evaluation model layer by layer, scoring the parameters or attributes. The scoring standard is a 1-9 scale, and the principles are shown in Table 1.

[0057] Table 1. Judgment Principles of Scale 1-9

[0058]

[0059]

[0060] The scoring criteria are based on determining the importance of lower-level indicators relative to higher-level indicators with which they have a subordinate relationship, and constructing a judgment matrix G. Taking the three parameters belonging to availability as an example, the judgment matrix G is composed as shown in equation (1) by comparing the three factors pairwise.

[0061]

[0062] After normalizing the judgment matrix G, a consistency test is performed. When CR satisfies equation (2):

[0063] CR = CI / RI < 0.1 (2)

[0064] If the judgment matrix G is considered to meet the consistency requirements, then it is considered to lack sufficient consistency and needs to be readjusted using the analytic hierarchy process (AHP). Where: CR is the random consistency ratio of the judgment matrix; CI is the consistency index of the judgment matrix, defined as:

[0065]

[0066] Where, λ max Let G be the largest eigenvalue of the judgment matrix; RI is the average random consistency index of the judgment matrix. Definition:

[0067] Gw=λ max w (4)

[0068] Solving for w (representing the weight vector) reveals that it belongs to the eigenvalue λ. max The eigenvectors are normalized to obtain the desired w. Repeat the above steps until both layers of the evaluation model have been scored. Let w be the weight index for the first layer. i (i = 1, 2, 3, 4), the weight index of the second layer implying membership relationships is w. ij The final weight index of the parameter is calculated according to equation (5):

[0069] w iij =w i ·w ij (5)

[0070] Step 3: Relatively complete procedures exist both domestically and internationally for the collection and processing of operational data. This type of data can be converted into operational parameters used in the analysis through certain calculation methods. In this method, the main data reflecting the operational safety of civil aircraft include: Number of unplanned product replacements (NUM). rem ), Monthly average flight hours (h), Product mean workshop repair time (MTTR), Number of product crew reports (NUM) rep Product mechanical delay events (NUM) del-mech Mean Time Between Supplies (MTBS) and Number of Product Support Delay Events (NUM) del-sup ) and number of installed products (NUM) install The data includes the cumulative month (MON) and the aircraft's fault-free factor (NFF). To ensure the validity of the data, it should have been accumulated for at least 3 months. The calculation methods for various operating parameters are defined as follows: (6)

[0071]

[0072]

[0073]

[0074]

[0075] In addition, sensitivity M is the result of numerical analysis calculations obtained from constructing the reliability model.

[0076] Step 4: Repeat steps 2 and 3 to calculate the operating parameters of all n products in the system to be allocated. The traditional proportional allocation method uses the ratio of product failure rate to the total system failure rate in historical data as the basis for measuring the importance of the indicator. This method combines 10 types of operating data to calculate 9 types of operating parameters, including product reliability, maintainability, availability, and supportability, thus expanding the basis for the importance of the indicator to the proportional importance of 9 types of operating parameters.

[0077] Step 5: Introduce parameter influence weighting coefficients to perform weighted calculations on the proportional importance of operating parameters. The product reliability index optimization allocation result is calculated according to equation (7). When the system as a whole meets the index requirements, the redistribution work is completed.

[0078]

[0079] In the formula, —Assign new reliability metrics (failure rate) to products or subsystems; —The overall reliability metrics (failure rate) to be assigned to the system; λ S —Total system failure rate.

[0080] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A reliability index optimization allocation method based on operational data, characterized in that... The specific steps are as follows: Step 1: An evaluation model for the operating parameter system applicable to civil aircraft was established, consisting of two layers: the first layer is performance evaluation, and the second layer is the operating parameter system. The first level of performance evaluation includes reliability, maintainability, availability, and supportability. Reliability describes the quality characteristics or importance of the product itself; maintainability describes how easy or difficult it is to repair the product; availability describes the product's ability to perform its tasks during use; and supportability describes the product's ability to obtain support support on the ground. The second-level operating parameter system includes failure rate and sensitivity decomposed by reliability, average repair rate and unit reporting rate decomposed by maintainability, unplanned replacement rate, daily utilization rate and mechanical delay rate decomposed by availability, and average spare parts availability rate and maintenance delay rate decomposed by supportability. Step 2: Calculate the parameter impact weight index of the product; a) Use a pairwise comparison method to evaluate the model layer by layer from the bottom layer, score the parameters or attributes, use the 1-9 scale method for scoring, and construct the judgment matrix G; b) Perform a consistency check on the normalized judgment matrix G; after the check, solve for the weight vector using the following formula. w For belonging to eigenvalues λ max The eigenvectors are normalized. in, λ max To determine the largest eigenvalue of matrix G; Repeat steps a) and b) until both layers of the evaluation model have been scored. Let the weight index of the first layer be... w i (i=1, 2, 3, 4), the weight index of the second layer implying membership relationships is: w ij (j=1, 2, 3……9), then the final weight index of the parameter is calculated according to equation (4): ； A method for performing a consistency check on the normalized judgment matrix G, when CR satisfies If the consistency requirement is met, then the judgment matrix G is considered to be consistent; otherwise, the judgment matrix G is considered to be inconsistent and needs to be adjusted again. Where: CR is the random consistency ratio of the judgment matrix; RI is the average random consistency index of the judgment matrix; CI is the consistency index of the judgment matrix. ； Step 3: The calculation methods for various operating parameters are as follows: Sensitivity M is the result of numerical analysis calculation obtained from constructing the reliability model; in, NUM rem Indicates the number of unplanned product replacements. h The MTTR (Mean Time To Repair) indicates the average monthly flight hours. NUM rep Indicates the number of product unit reports, NUM del-mech This indicates product mechanical delay events; MTBS indicates the mean time to spare parts availability. NUM del-sup Indicates product warranty delay events, NUM install Indicates the number of products installed, MON This indicates the cumulative number of months for the data; NFF represents the aircraft's fault-free coefficient. Step 4: Repeat steps 2 and 3 to calculate all systems to be allocated. n Operating parameters of each product; Step 5: Introduce parameter influence weighting coefficients to perform weighted calculations on the proportional importance of operating parameters. The product reliability index optimization allocation result is calculated according to the following formula; when the system as a whole meets the index requirements, the reallocation work is completed. in, k =1,2,3...... n , λ * k —Assign new reliability metrics, such as failure rate, to products or subsystems; λ S * —The overall reliability index to be assigned to the system is the failure rate; λ s —Total system failure rate.

Images