Method and system for evaluating correlation between performance design and reliability design of airborne product

Abstract

The invention belongs to the field of product reliability, and discloses an airborne product performance design and reliability design correlation evaluation method and system, and the method comprises the steps: obtaining basic evaluation data; obtaining a project fusion degree of reliability design; a key and important part control level is obtained; the reliability completion degree is obtained; obtaining a reliability growth coefficient; determining a correlation evaluation index weight of the performance design and the reliability design; obtaining a performance design and reliability design correlation evaluation result; and analyzing the correlation evaluation of the performance design and the reliability design. The system comprises a database, a project fusion degree module for reliability design, a key and important part control level module, a reliability completion degree module, a reliability growth coefficient module, a weight module, a correlation evaluation module and a correlation analysis module. According to the method, analysis is carried out based on reliability design and reliability test multi-source data and processes, so that the evaluation result of the correlation degree of the performance design and the reliability design is more real and accurate.

Description

Technical Field

[0001] The present invention belongs to the field of equipment reliability, and in particular relates to a method and system for evaluating the correlation between performance design and reliability design of an airborne product. Background Art

[0002] The development of domestic airborne products has gradually shifted from surveying and mapping to independent research and development. The traditional airborne product development method is driven by the realization of performance indicators and ignores reliability indicators, resulting in the risk of reliability indicators not being achieved, causing frequent early failures. Therefore, it is an inevitable trend to accurately evaluate the correlation between performance design and reliability design of airborne products. The correlation between performance design and reliability design of airborne products refers to the degree to which performance design work and reliability design work are organically combined. It is used to ensure that reliability work is implemented in the entire process of product development and to ensure that performance requirements and reliability requirements are achieved at the same time. Therefore, considering the correlation between performance design and reliability design can improve the drawbacks of the traditional design process, so that reliability requirements and the results obtained from design analysis work are embedded in the development process as equally important constraints to carry out related design work, and promote the development of independent research and development of airborne products in my country.

[0003] Evaluation is usually divided into qualitative evaluation and quantitative evaluation. If only one method is used, it may not objectively reflect the correlation or be too mechanical. Therefore, a comprehensive evaluation method is generally used to integrate different evaluation methods according to the characteristics of each method and apply them to solve a problem, so that the evaluation results are more accurate and effective. However, at present, there is a relative lack of evaluation on the correlation between performance design and reliability design of airborne products, and there is a lack of specific evaluation methods for the degree of integration of performance design and reliability design. Therefore, it is necessary to take the correlation between performance design and reliability design as the evaluation object, and establish a reasonable, comprehensive, systematic and engineering applicable evaluation method to correctly and objectively evaluate the degree of integration between the two. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention provides a method and system for evaluating the correlation between the performance design and reliability design of airborne products. The method and system take the correlation between the performance design and reliability design of airborne products as the evaluation object, evaluate the degree of integration of performance design and reliability design in engineering applications, and then put forward corresponding improvement suggestions. The method proposed in the present invention can work according to the airborne product design process to evaluate the degree to which airborne products combine reliability design in performance design; it can also be used to find the part of the performance design of airborne products that is currently least integrated with reliability design. In the case of limited resources, this part can be improved first, providing supervision and control support for the integration of reliability design in performance design of airborne products.

[0005] To achieve the above object, the present invention discloses a method for evaluating the correlation between performance design and reliability design of airborne products based on multi-source data fusion, which comprises the following steps:

[0006] Step 1: Obtain basic evaluation data;

[0007] The basic evaluation data consists of four parts of data obtained from two sources. The first source is obtained based on the reliability design of airborne products, including project integration and key component control level. The second source is obtained based on reliability experiments, including reliability completion and reliability growth coefficient.

[0008] Step 2: Obtain the project integration degree of reliability design;

[0009] When the number of reliability design projects is n, the calculation formula for the project integration degree Z of reliability design is:

[0010]

[0011] Among them, Z is the project integration degree of reliability design, n is the number of reliability design projects, a represents one of the reliability design projects, a=1,2…,n; n is a positive integer, Z a The integration degree of a single project;

[0012] Step 3: Obtain the control level of key parts;

[0013] The key component control level F∈(0,1) is:

[0014]

[0015] Among them, V is the average level of the control level of key items, F is the control level of key items, F∈(0,1), P ij represents the evaluation score of the i-th expert on the j-th indicator, i = 1, 2, ..., m, m is the number of experts, j = 1, 2, ..., 6;

[0016] Step 4: Get reliability completion degree;

[0017] The estimated MTBF for airborne products is:

[0018]

[0019] Where t is time, f(t) is the lifetime distribution density function after fusion;

[0020] The reliability completion degree Q is:

[0021]

[0022] Among them, Q is the reliability completion degree, MTBF is the estimated value of the reliability MTBF of the airborne product in the reliability experiment. 要求 is the target value of the reliability MTBF indicator;

[0023] Step 5: Obtain the reliability growth coefficient;

[0024] Taking 0.7 as the growth limit for normalization, the reliability growth coefficient U∈(0,1) of the performance design after reliability design is:

[0025]

[0026] Among them, U1 is the improvement degree of reliability growth test; U2 is the improvement degree of reliability development test;

[0027] Step 6: Determine the weights of the performance design and reliability design correlation evaluation indicators;

[0028] Let the project integration degree, key component control level, reliability growth coefficient and reliability completion degree of reliability design be evaluation indicators 1, 2, 3 and 4 respectively. The three-scale method is used to establish the judgment matrix A=(a ij ) 4×4 ,but:

[0029]

[0030] in,

[0031]

[0032] Normalize the matrix by column and establish the normalized matrix B = (b ij ) 4×4 , sum the rows of matrix B and perform normalization to establish the judgment matrix w = (w i ) 4×1 , the indicator weights of reliability design project integration, key component control level, reliability growth coefficient, and reliability completion are obtained as w1, w2, w3, and w4 respectively;

[0033] Step 7: Obtain the evaluation results of the correlation between performance design and reliability design;

[0034] The correlation evaluation value R between performance design and reliability design is:

[0035] R=Z×w1+S×w2+U×w3+Q×w4;

[0036] The larger the correlation evaluation value R between performance design and reliability design, the higher the support degree of reliability design for performance design.

[0037] Preferably, in step 1, the four parts of data include relevant data on project integration degree and key component control level, reliability completion degree and reliability growth coefficient, specifically:

[0038] (1) Obtain data related to the project integration of reliability design, including: reliability prediction, failure mode impact and criticality analysis, formulation of reliability design criteria, fault tree analysis, finite element analysis and durability analysis. The number of times each project should be carried out during the development process is m. a1 、The actual number of times m a2 , the number of iterations to improve the design m a3 , the number of times the design has not been improved by iteration m a4 , the number of times no improvement was provided to the design m a5 and the number of times the design is supported but not iterated a6 , where m a1 、m a2 、m a3 、m a4 、m a5 、m a6 a in the reliability design represents one of the items, a=1,2…,n; n is the number of reliability design items, n is a positive integer;

[0039] (2) Obtain relevant data on the control level of key items, including six indicators: key control level, improvement control level, distribution control level, quality control level, inspection control level, and procurement control level;

[0040] (3) Obtain relevant data on reliability completion, including: test time, number of responsible failures and confidence information during the reliability identification test, life test, and reliability acceptance test, as well as reliability index requirements;

[0041] (4) Obtain relevant data on the reliability growth factor, including the occurrence time of each failure observed during the reliability growth test, and the MTBF values ​​of the first and last tests of the reliability development test.

[0042] Preferably, in the step 2, the project integration degree of the reliability design is obtained:

[0043] The design support of item a in reliability design is:

[0044]

[0045] The completion degree of reliability design project a is:

[0046]

[0047] The integration degree Z of a single item of reliability design project a a , the formula is:

[0048]

[0049] Among them, w1 a Design support for item a designed for reliability, w1 a ∈(0,1),m a2 The number of times the project a designed for actual reliability is performed, m r5 The number of times that a project designed for reliability does not provide an improvement to the design; Y a The degree of completion of project a designed for reliability, Y a ∈(0,+∞),m a2 The number of times the project a designed for actual reliability is performed, m a1 The number of times a project designed for reliability should be carried out during the development process.

[0050] Preferably, in step 4, the specific steps for obtaining the fused life distribution density function f(t) are as follows:

[0051] First, for a single reliability experiment, the lower confidence limit at a given confidence level c is θ L ; Then the life distribution density function of airborne products is:

[0052] f(t)=λe -λt

[0053] in, t is time;

[0054] If the reliability experiment is of m types, then m life distribution density functions are fused, and the support vector of the mutual support degree between the reliability acceptance test and other tests is established with the reliability acceptance test as the center point, that is:

[0055] S=(S 11 S 12 … S 1m )

[0056] in, f1(t) represents the life distribution density function obtained from the reliability acceptance test; f i (t) represents the life distribution density function of the i-th reliability experiment, S 1i represents f1(t) and f i (t) the distance between them;

[0057] Determine the weight w21 of the reliability acceptance test as:

[0058]

[0059] Among them, L p and L q They represent the lengths of the confidence intervals of the life distribution density function obtained from the reliability acceptance test at confidence levels p and q, respectively, p>q;

[0060] After obtaining the weight of w21, the weight of the i-th reliability experiment w2 i for:

[0061]

[0062] Among them, S 1i represents the distance between the life distribution density function obtained from the reliability acceptance test and the life distribution density function of the i-th reliability experiment; w2 i is the weight of the i-th reliability experiment;

[0063] The final fusion model of life distribution density function is:

[0064]

[0065] Among them, f(t) is the lifetime distribution density function after fusion, w2 i is the weight of the reliability experiment of the i-th type; f i (t) represents the life distribution density function of the i-th reliability experiment, and m is the number of reliability experiment types.

[0066] Preferably, in step 5, the specific steps for obtaining the reliability growth test improvement degree U1 and the reliability development test improvement degree U2 are as follows:

[0067] The reliability growth curve is the AMSAA model, and its expression is:

[0068] E[N(t)]=at b

[0069] Where, N(t) is the cumulative number of failures that occur at cumulative test time t; a is the scale parameter; b is the growth shape parameter, and E[N(t)] is the mathematical expectation of N(t);

[0070] The reliability growth test improvement degree U1 is:

[0071]

[0072] Where N is the total number of observed faults; t i is the cumulative test time when the i-th fault occurs; T is the total cumulative test time; M is the first parameter, and its value satisfies

[0073] The reliability development test improvement degree Y2 is:

[0074]

[0075] Among them, t1 is the first MTBF obtained from the test, and t2 is the last MTBF obtained from the test.

[0076] Preferably, in step 6, after establishing the judgment matrix w, w also needs to pass a consistency test, specifically:

[0077] Judgment matrix w=(w i ) 4×1 The largest characteristic root of Among them, (Bw) i Represents the i-th row element after matrix B is multiplied by judgment matrix w;

[0078] Perform consistency test on the judgment matrix w, and the consistency index CI is expressed as:

[0079]

[0080] Random consistency index RI, when the number of indicators n is equal to 4, RI = 0.89, the consistency ratio CR is If CR<0t1, the eigenvector of the judgment matrix w is used to represent the weight vector;

[0081] If the consistency check fails, it means that the setting of matrix A is unreasonable, and the values ​​in matrix A need to be reset.

[0082] Preferably, the step seven further includes:

[0083] According to the value range of the performance design and reliability design correlation evaluation value R, segmentation is performed so that the performance design and reliability design correlation evaluation R corresponds to evaluation results of different levels.

[0084] Preferably, the method further includes step eight of analyzing the correlation evaluation between performance design and reliability design, specifically:

[0085] 1) Sort Z, F, Q, and U from low to high according to their scores to obtain a ranking table. The lowest score is the short board indicator item that combines performance design and reliability design;

[0086] 2) For the project integration degree Z, according to the reliability design analysis work project integration degree Z of item a i Sort by scores from low to high, and output a ranking table of the shortcomings of project integration;

[0087] 3) For the control level F of the key items, sort them from low to high according to the scoring results of the key item control level indicators to obtain a ranking table of the short board of the control level;

[0088] 4) For the reliability completion degree Q, if Q>0, it means that the specified reliability index requirements have been met; if Q<0, it means that the specified reliability index requirements have not been met;

[0089] 5) For the reliability growth coefficient U, based on the comparison of U1 and U2, a ranking table of the short boards according to the reliability growth degree is obtained;

[0090] In 1) to 5), the lower the score, the more improvement is needed.

[0091] The present invention also discloses a system using the performance design and reliability design correlation evaluation method of an airborne product, comprising:

[0092] Database, project integration degree module of reliability design, key component control level module, reliability completion degree module, reliability growth coefficient module, weight module and correlation evaluation module, wherein the project integration degree module of reliability design, key component control level module, reliability completion degree module and reliability growth coefficient module are connected to the database, weight module and correlation evaluation module respectively;

[0093] The database is used to store the data required for the performance reliability integrated design degree evaluation, including a reliability design project sub-database, a key component control level sub-database, a reliability completion sub-database and a reliability growth coefficient sub-database; wherein the reliability design project sub-database includes the number of times m that the reliability prediction, failure mode effect and criticality analysis, formulation of reliability design criteria, fault tree analysis, finite element analysis and durability analysis projects should be carried out in the development process. a1 , the number of iterations to improve the design m a3 , the number of times no improvement was provided to the design m a5 , the number of times the design is supported but not iterated again m a6 ; The key component control level sub-database includes six index values, namely, key control level, improvement control level, allocation control level, quality control level, inspection control level and procurement control level, obtained by expert scoring; the reliability completion sub-database includes the test time, number of responsible failures, confidence information and reliability index requirements of reliability identification test, life test and reliability acceptance test during the implementation process; the reliability growth coefficient sub-database includes the occurrence time of each failure observed during the reliability growth test, and the MTBF value of the first and last test of the reliability development test;

[0094] The project integration degree module of reliability design is used to obtain the project integration degree Z of reliability design, which is implemented in the way of step 2;

[0095] The critical component control level module is used to obtain the critical component control level F, which is implemented in the manner of step 3;

[0096] The reliability completion module is used to obtain the reliability completion Q, which is implemented in the way of step 4;

[0097] The reliability growth coefficient module is used to obtain the reliability growth coefficient U, which is implemented in the way of step 5;

[0098] The weight module is used to establish a judgment matrix using the three-scale method to obtain the weights of project integration degree Z, key component control level F, reliability completion degree Q and reliability growth coefficient U;

[0099] The correlation evaluation module obtains the correlation evaluation value of performance design and reliability design by using the method of step seven according to the weight output by the weight module, the project integration degree Z output by the project integration degree module of reliability design, the key component control level F output by the key component control level module, the reliability completion degree Q output by the reliability completion degree module, and the reliability growth coefficient U output by the reliability growth coefficient module.

[0100] Preferably, it also includes a correlation analysis module, which is respectively connected to the project integration module of reliability design, the key component control level module, the reliability completion module, the reliability growth coefficient module and the correlation evaluation module.

[0101] Compared with the prior art, the present invention has the following beneficial effects:

[0102] a) Analyze the multi-source data of reliability design and reliability test in the whole development stage to make the results of the evaluation of the degree of correlation between performance design and reliability design more real and accurate.

[0103] b) Comprehensively consider the relevant factors of performance design and reliability design, provide reliability improvement measures, provide design references, judge whether the specified performance requirements are met, and determine the evaluation indicators for the correlation between performance design and reliability design, so that the results of the correlation evaluation are more comprehensive.

[0104] c) This method establishes a performance design and reliability design correlation evaluation process, takes into account the support effect of reliability work for design work and the performance and reliability indicators required after the product design is finalized, and uses a multi-source data fusion method to extract four types of indicators to express the degree of correlation between performance design and reliability design, making the evaluation more scientific.

[0105] d) The results of this invention can quantitatively describe the degree of integration of performance design and reliability design in the development process, and output a ranking table of shortcomings of various indicators to identify problems, correct them in time, and provide direction for integrated control. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figure 1 Flow chart of correlation evaluation method for performance design and reliability design of airborne products;

[0107] Figure 2 A multi-source data system diagram for the performance design and reliability design correlation evaluation method of airborne products;

[0108] Figure 3 It is a data relationship diagram related to the integration degree of performance design and reliability design;

[0109] Figure 4 Schematic diagram of the structure of the correlation evaluation system for performance design and reliability design of airborne products. DETAILED DESCRIPTION

[0110] The exemplary embodiments, features and aspects of the present invention will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0111] The present invention provides a method and system for evaluating the correlation between performance design and reliability design of airborne products, which uses data obtained from reliability design and reliability test as data source and uses a multi-source data fusion method to evaluate the correlation between performance design and reliability design of airborne products. Figure 1 As shown, it includes the following steps:

[0112] Step 1: Obtain basic evaluation data.

[0113] Basic evaluation data sources such as Figure 2 As shown in the figure, it consists of four parts of data obtained from two sources. The first source is obtained based on the reliability design of airborne products, including project integration and key component control level. The second source is obtained based on reliability experiments, including reliability completion and reliability growth coefficient. The following is a detailed description of the four parts of data:

[0114] (1) Obtain data related to the project integration of reliability design. The projects of reliability design in this application include reliability prediction, failure mode effect and criticality analysis, formulation of reliability design criteria, fault tree analysis, finite element analysis and durability analysis. Therefore, the data related to the project integration of reliability design include: the number of times m that the projects such as reliability prediction, failure mode effect and criticality analysis, formulation of reliability design criteria, fault tree analysis, finite element analysis or durability analysis should be carried out in the development process. a1 , the number of iterations to improve the design m a3 , the number of times no improvement was provided to the design m a5, the number of times the design is supported but not iterated again m a6 , the number of iterations not reaching the design improvement is m a4 The number of times no improvement is provided to the design m a5 and the number of times the design is supported but not iterated a6 Additive composition, that is, m a4 =m a5 +m a6 , the actual reliability design project a is carried out m times a2 The number of iterations to design improvement m a3 and the number of times m that the design has not been improved a4 Additive composition, that is, m a2 =m a3 +m a4 , such as Figure 3 As shown. a1 、m a2 、m a3 、m a4 、m a5 、m a6 a in the above formula represents one of the reliability design items, a=1,2…,n; n is the number of reliability design items, n is a positive integer. For example, when a=1, the item is reliability prediction, then m 11 It indicates the number of times reliability is expected to be carried out in the development process, m 12 It means the number of times the actual reliability is expected to be carried out. The above times refer to the number of times the project is completed. In the fault tree analysis, the fault tree analysis of all failure modes at the same agreed level is counted as one time. The project integration data of reliability design can be obtained by looking up the reliability design report, performance design report or asking the designer.

[0115] (2) Obtain relevant data on the control level of key items. The control level of key items includes six indicators: key control level, improvement control level, allocation control level, quality control level, inspection control level and procurement control level. The control level of key items is determined by expert scoring. Key items refer to key items and important items.

[0116] Among them, the key control level, improvement control level, and allocation control level are evaluation indicators in design control, and the quality control level, inspection control level, and procurement control level are evaluation indicators in process control. The following is an overview of the six evaluation indicators. The overview of the evaluation indicators is shown in Table 1:

[0117] Table 1 Evaluation indicators and overview of the control level of key items

[0118]

[0119]

[0120] The evaluation indexes are divided into five levels. The higher the level, the better the reliability control of key components, the higher the reference degree of performance design to reliability design, and the higher the degree of correlation between performance design and reliability design. The explanation of each level of evaluation index is shown in Table 2.

[0121] Table 2 Evaluation criteria for key component control level evaluation indicators

[0122]

[0123]

[0124] (3) Obtain relevant data on reliability completion. Reliability experiments include reliability identification tests, life tests, and reliability acceptance tests. Therefore, relevant data on reliability completion include: the test time, number of responsible failures, confidence level, and other information during the reliability identification test, life test, and reliability acceptance test. Reliability index requirements also need to be clarified.

[0125] (4) Obtain relevant data on the reliability growth factor, including the occurrence time of each failure observed during the reliability growth test, and the MTBF (Mean Time Between Failure) values ​​of the first and last tests of the reliability development test.

[0126] Step 2: Get the project integration degree of reliability design.

[0127] Define the degree to which reliability design project a helps performance design as design support, denoted by w1 a ∈(0,1), the calculation formula is:

[0128]

[0129] Among them, w1 a Design support for item a designed for reliability, m a2 The number of times the project a designed for actual reliability is performed, m r5 The number of times a project designed for reliability provides no improvement to the design.

[0130] Define the degree of completion of reliability design project a as completion degree Y a ∈(0, +∞), the calculation formula is:

[0131]

[0132] Among them, Y a Completion of project a designed for reliability, ma2 The number of times the project a designed for actual reliability is performed, m a1 The number of times a project designed for reliability should be carried out during the development process.

[0133] Define the integration degree of reliability design project a and performance design as single project integration degree Z a , and its calculation formula is:

[0134]

[0135] When the number of reliability design projects that should be carried out during the development of airborne products is n, where n is a positive integer, the calculation formula for the project integration degree Z of reliability design is:

[0136]

[0137] Where Z is the project integration degree of reliability design, n is the number of reliability design projects, a represents one of the reliability design projects, a=1,2…,n; Z a The integration degree of a single project.

[0138] Step 3: Get the control level of key parts.

[0139] The key control level indicators include: key control level, improvement control level, allocation control level, quality control level, inspection control level and procurement control level, which are set as indicator 1, indicator 2, indicator 3, ..., indicator 6 respectively. Assume that the scoring results of m experts are as follows:

[0140]

[0141] Among them, P ij It represents the evaluation score of the i-th expert on the j-th indicator, i = 1, 2, ..., m, m is the number of experts, j = 1, 2, ..., 6, assuming that the weights of m experts are the same, the key control level F∈(0, 1), its calculation formula is:

[0142]

[0143] Among them, V is the average level of control of key parts, t is the control level of key parts, and the relative score between the average level of control of key parts and the highest level 5, F∈(0,1), represents the degree to which key parts are effectively controlled in the process of airborne product design.

[0144] Step 4: Get reliability completion.

[0145] Reliability experiments include reliability identification tests, life tests, and reliability acceptance tests. Based on the test time T and the number of responsible failures that occur during the reliability experiment, the MTBF (mean time between failures) of the system can be evaluated.

[0146] First, for a single reliability experiment, the lower confidence limit at a given confidence level c is:

[0147]

[0148] Among them, θ L is the lower confidence limit, T is the test duration, r is the number of responsible failures that occurred, It is a chi-square distribution with 2r+2 degrees of freedom, which is a statistic and can be obtained by looking up the table.

[0149] Therefore, the life distribution function and density function of airborne products are:

[0150] F(t)=1-e -λt

[0151] f(t)=λe -λt

[0152] in, t is time.

[0153] Secondly, to fuse the data of different reliability tests, it is necessary to determine the support weight of each type of test data for the results. Since each type of test has a certain correlation in the test conditions, it is assumed that the life distribution density functions obtained from different tests also have a certain correlation. The weight of the life distribution density function of each type of test is obtained according to the degree of mutual support between different tests. Assume that the life distribution density functions obtained from the two types of tests are f p (t) and f q (t), then the mutual support between the life distribution density functions can be expressed by D(f p ||f q ) to define:

[0154]

[0155] Among them, the definition If D(f p ||f q ) is larger, which means that the distance between the life distribution density functions obtained from the two types of experiments is smaller, and the degree of mutual support between the two is higher.

[0156] If the reliability experiment is of m types, then m life distribution density functions are fused. First, the mutual support degree of different reliability experiments is calculated. Since the reliability acceptance test in the reliability experiment is the test in the final stage of the development process, the reliability acceptance test is taken as the center point to establish the support vector of the mutual support degree between the reliability acceptance test and other tests, that is:

[0157] S=(S 11 S 12 … S 1m )

[0158] in, f1(t) represents the life distribution density function obtained from the reliability acceptance test; f i (t) represents the life distribution density function of the i-th type reliability experiment. In this embodiment, m=3, f2(t) and f3(t) are the life distribution density functions of the airborne product obtained by the reliability evaluation test and the life test, respectively. 1i represents f1(t) and f i (t) The distance between the two, if S 1i The larger the value, the smaller the distance, that is, the greater the support degree, S 1i Inversely proportional to the level of support.

[0159] Determine the weight of the reliability acceptance test. Since there is a certain gap between the life distribution obtained by the reliability acceptance test and the actual life distribution of the product, the credibility ρ of the life distribution density function f1(t) is used as its weight, that is, the weight w21 of the reliability acceptance test is:

[0160]

[0161] Among them, L p and L q They represent the length of the confidence interval of the life distribution density function obtained from the reliability acceptance test at the confidence levels of p and q, respectively, p>q. In engineering, the confidence lower limit with a confidence level of 50% is often used as the point estimate of the reliability parameter, while the confidence level of the reliability parameter estimate obtained based on the physical test data is 80%, so generally:

[0162]

[0163] Among them, L 0t8 and L 0t5 They represent the length of the confidence interval of the life distribution parameter obtained from the reliability acceptance test at 80% and 50% confidence levels. After obtaining the weight of w21, the weight w2 of the i-th reliability experiment i for:

[0164]

[0165] Among them, S 1i represents the distance between the life distribution density function obtained from the reliability acceptance test and the life distribution density function of the i-th reliability experiment; w2 i is the weight of the i-th reliability experiment;

[0166] The final fusion model of life distribution density function is:

[0167]

[0168] Among them, f(t) is the lifetime distribution density function after fusion, w2 i is the weight of the reliability experiment of the i-th type; f i (t) represents the life distribution density function of the i-th reliability experiment, and m is the number of reliability experiment types.

[0169] Finally, the estimated MTBF for the airborne product is:

[0170]

[0171] The reliability completion degree Q is defined as the ratio of the estimated value of the product reliability index of the reliability experiment to the required reliability index value. At this time, the required reliability index is MTBF. Using the target value in the parameter, the calculation formula is:

[0172]

[0173] Among them, Q is the reliability completion degree, MTBF is the estimated value of the reliability MTBF of the airborne product in the reliability experiment. 要求 It is the target value of the reliability MTBF indicator.

[0174] Step 5: Get the reliability growth coefficient.

[0175] First, collect reliability growth test data, continuously record the cumulative number of product failures N(t) and cumulative test time t, and assume that the reliability growth curve is the AMSAA (Army Material Systems Analysis Activity) model, which is expressed as:

[0176] E[N(t)]=at b

[0177] Where N(t) is the cumulative number of failures occurring at cumulative test time t; a is the scale parameter; b is the growth shape parameter, and E[N(t)] is the mathematical expectation of N(t).

[0178] The estimate for the growth shape parameter b is:

[0179]

[0180] Where N is the total number of observed faults; t i is the cumulative test time when the i-th fault occurs; T is the total cumulative test time; M is the first parameter, and its value satisfies

[0181] According to the above formula, the design improvement degree of the reliability growth test is derived, which is recorded as the reliability growth test improvement degree U1:

[0182]

[0183] Secondly, consider the design improvement effect brought by the reliability development test. Assuming that the first MTBF obtained by the test is t1, and the last MTBF obtained by the test is t2, the reliability improvement degree of the design improvement of the reliability development test is obtained, which is recorded as the reliability development test improvement degree U2:

[0184]

[0185] Assuming that the reliability growth test and the reliability development test have the same impact on reliability growth, and according to the relevant provisions in GJB1407, a reliability growth rate lower than 0.3 indicates that the corrective measures are not effective, and between 0.6-0.7 indicates that strong and effective fault analysis and corrective measures have been taken. Therefore, the overall reliability growth coefficient is processed and normalized with 0.7 as the growth limit. The reliability growth coefficient U∈(0,1) of the performance design after reliability design work is:

[0186]

[0187] Step 6: Determine the weights of the evaluation indicators related to performance design and reliability design.

[0188] According to the process of steps 2, 3, 4 and 5, a multi-source data system for performance design and reliability design correlation evaluation is constructed, as shown in the attached Figure 2 , and finally obtained four performance design and reliability design correlation evaluation indicators, their meanings and types are shown in Table 3:

[0189] Table 3 Classification and meaning of performance design and reliability design correlation evaluation indicators

[0190]

[0191] Let the project integration degree, key component control level, reliability growth coefficient, and reliability completion of reliability design be evaluation indicators 1, 2, 3, and 4 respectively. The evaluation indicators are established using the three-scale method to establish the judgment matrix A = (a ij ) 4×4 ,but:

[0192]

[0193] in,

[0194]

[0195] In order to obtain the maximum eigenvalue and corresponding eigenvector of the judgment matrix, it is necessary to normalize the matrix by column and establish a normalized matrix B = (b ij ) 4×4 ,but:

[0196]

[0197] in,

[0198] Sum the rows of matrix B and perform normalization to establish the judgment matrix w = (w i ) 4×1 ,but

[0199] w=(w1,w2,w3,w4) T

[0200] in,

[0201] Calculate the maximum characteristic root, that is

[0202] Among them, (Bw) i Represents the i-th row element after matrix B is multiplied by judgment matrix w.

[0203] Perform consistency test on the judgment matrix w, and the consistency index CI is expressed as:

[0204]

[0205] Random consistency index RI, when the number of indicators n is equal to 4, RI = 0.89, RI is obtained by looking up the average random consistency index RI table. The definition of consistency ratio CR is According to the regulations, if CR<0t1, the weight vector is represented by the eigenvector of the judgment matrix w.

[0206] If the consistency test fails, it means that the setting of matrix A is unreasonable and the value in matrix A needs to be reset: the consistency test is to determine whether there will be inconsistent errors in the matrix: for example, for indicators a, b and c, assuming that a>b, b>c are set in the matrix, but c>a exists, then the consistency test will fail.

[0207] Therefore, when the consistency test is passed, the indicator weights of the project integration degree, key parts control level, reliability growth coefficient, and reliability completion degree of reliability design are w1, w2, w3, and w4 respectively.

[0208] Step 7: Obtain the evaluation results of the correlation between performance design and reliability design.

[0209] Based on the index evaluation results obtained in steps 2, 3, 4 and 5 and the index weights obtained in step 6, the correlation evaluation of performance design and reliability design is obtained, and the output result is R:

[0210] R=Z×w1+S×w2+U×w3+Q×w4.

[0211] The larger the R value of the correlation evaluation between performance design and reliability design, the higher the support degree of reliability design for performance design.

[0212] Further, the value ranges of the four evaluation indicators are analyzed. The value ranges of key parts control level and reliability growth coefficient are between (0, 1); the reliability completion will be greater than 1 if the designed product exceeds the target value, and will be between (0, 1) if it does not exceed the target value but exceeds the threshold value, and will gradually decrease with the actual MTBF value; the project integration degree of reliability design is generally between (0, 1). If the number of iterations of design improvement is large, its range will be greater than 1, but a large number of iterations also indicates that the reliability design work has a high degree of support for the design work; According to the above analysis, the performance reliability integrated design degree R is graded into five levels, and the corresponding evaluation results are output according to the score range, as shown in Table 4.

[0213] Table 4 Analysis of the evaluation results of the correlation between performance design and reliability design

[0214]

[0215] Step 8: Analyze the correlation evaluation between performance design and reliability design.

[0216] Analyze the above output content to find the shortcomings in the combination of performance design and reliability design, and provide a reference for the combination of performance design and reliability design.

[0217] 1) First, Z, F, Q, and U should be sorted from low to high according to the scores to obtain a sorting table. The lowest score is the short board indicator item that combines performance design and reliability design.

[0218] 2) Further analyze the project integration degree Z, and the project integration degree Z of the reliability design analysis work of item a i Sort the scores from low to high, output a ranking table of the shortcomings of the project integration, and analyze the reasons for the low integration with performance design from the two aspects of reliability design support and completion.

[0219] 3) Further analyze the key-item control level F, sort the scoring results of the key-item control level indicators from low to high, obtain a ranking table of the shortcomings of the control level, and analyze the shortcomings of the work items in the key-item control process.

[0220] 4) Further analyze the reliability completion degree Q. If Q>0, it means that the specified reliability index requirements have been met. If Q<0, it means that the specified reliability index requirements have not been met, indicating that the combination of performance design and reliability design is very poor.

[0221] 5) Further analyze the reliability growth coefficient U, compare U1 and U2, and obtain a ranking table of shortcomings in terms of reliability growth degree.

[0222] In the above analysis, the lower the score, the more improvement is needed, and the degree of improvement of each indicator can be controlled according to the corresponding ranking table.

[0223] The present invention also provides a system using the performance design and reliability design correlation evaluation method of an airborne product, such as Figure 4 As shown, it includes a database 1, a project integration module 2 for reliability design, a key component control level module 3, a reliability completion module 4, a reliability growth coefficient module 5, a weight module 6, a correlation evaluation module 7 and a correlation analysis module 8, wherein the project integration module 2 for reliability design, the key component control level module 3, the reliability completion module 4 and the reliability growth coefficient module 5 are connected to the database 1, the weight module 6, the correlation evaluation module 7 and the correlation analysis module 8 respectively, and the correlation evaluation module 6 is also connected to the weight module 5 and the correlation analysis module 7 respectively.

[0224] Among them, database 1 is used to store the data required for the evaluation of the degree of integrated design of performance and reliability, including reliability design project sub-database 11, key component control level sub-database 12, reliability completion sub-database 13 and reliability growth coefficient sub-database 14. Among them, the reliability design project sub-database includes the number of times m that projects such as reliability prediction, failure mode effect and criticality analysis, formulation of reliability design criteria, fault tree analysis, finite element analysis and durability analysis should be carried out in the development process. a1 , the number of iterations to improve the design m a3 , the number of times no improvement was provided to the design m a5 , the number of times the design is supported but not iterated again m a6 . The key parts control level sub-database includes six indicator values, namely, key control level, improvement control level, allocation control level, quality control level, inspection control level and procurement control level obtained by expert scoring. The reliability completion sub-database includes the test time, number of responsible failures, confidence information, and reliability indicator requirements during the reliability identification test, life test and reliability acceptance test. The reliability growth coefficient sub-database includes the occurrence time of each failure observed during the reliability growth test, as well as the MTBF (Mean Time Between Failure) values ​​of the first and last tests of the reliability development test.

[0225] The project integration degree module 2 of reliability design is used to obtain the project integration degree Z of reliability design, which is implemented in the way of step 2.

[0226] The critical component control level module 3 is used to obtain the critical component control level F, which is implemented in the manner of step three.

[0227] The reliability completion module 4 is used to obtain the reliability completion Q, which is implemented in the manner of step 4.

[0228] The reliability growth coefficient module 5 is used to obtain the reliability growth coefficient U, which is implemented in the manner of step five.

[0229] The weight module 6 is used to establish a judgment matrix using the three-scale method to obtain the weights of the project integration degree Z, the key component control level F, the reliability completion degree Q and the reliability growth coefficient U.

[0230] The correlation evaluation module 7 obtains the performance design and reliability design correlation evaluation score according to the weight output by the weight module, the project integration degree Z output by the project integration degree module of the reliability design, the key component control level F output by the key component control level module, the reliability completion degree Q output by the reliability completion degree module, and the reliability growth coefficient U output by the reliability growth coefficient module, using the method of step seven.

[0231] The correlation analysis module 8 is used to determine the parts that need to be improved in the performance design to increase the relevance of reliability, specifically including: obtaining the evaluation results according to the performance design and reliability design relevance evaluation score R output by the correlation evaluation module, sorting the short board index items according to the project integration degree Z, the key component control level F, the reliability completion degree Q and the reliability growth coefficient U; and sorting the reliability design analysis work project integration degree Z for the ath item according to the reliability design project integration degree module. i According to the ranking from low to high scores, the reasons for the low integration with performance design can be further analyzed from the two aspects of reliability design support and completion; according to the distance S between the life distribution density function of the i-th reliability experiment and the reliability acceptance test in the key component control level module 1i Sorting is performed to obtain the shortcomings in the key parts control process; according to the reliability completion degree Q output by the reliability completion degree module, it is judged that if it is greater than 0, the specified reliability index requirements have been met, otherwise it has not been met; according to the comparison of the reliability growth test improvement degree U1 and the reliability development test improvement degree U2 in the reliability growth coefficient module, the shortcomings of the reliability growth degree are obtained.

[0232] Taking the development of a certain type of helicopter actuator as an example, the performance design and reliability design correlation evaluation method of the transmission system is implemented as follows:

[0233] Step 1: Obtain basic evaluation data.

[0234] Identify the required data and collect the data. The required data include the following:

[0235] (1) Obtain data related to the project integration of reliability design, including: reliability prediction, failure mode impact and criticality analysis, formulation of reliability design criteria, fault tree analysis, finite element analysis or durability analysis, and the number of times each project should be carried out during the development process. a1 、The actual number of times m a2 , the number of iterations to improve the design m a3 , the number of times the design has not been improved by iteration m a4 , the number of times no improvement was provided to the design m a5 , the number of times the design is supported but not iterated again m a6 , where a represents the corresponding reliability design item.

[0236] (2) Obtain relevant data on the control level of key items, including: experts' scoring data on the six indicators of key control level, improvement control level, allocation control level, quality control level, inspection control level and procurement control level.

[0237] (3) Obtain relevant data on reliability completion, including: the test time, number of responsible failures, confidence level, and other information during the reliability identification test, life test, and reliability acceptance test, as well as reliability index requirements.

[0238] (4) Obtain relevant data on the reliability growth factor, including: the occurrence time of each failure observed during the reliability growth test, and the MTBF value of the first and last tests of the reliability development test.

[0239] Step 2: Get the project integration degree of reliability design.

[0240] Based on the obtained project integration data of reliability design, the design support, completion, single project integration and project integration are calculated, and the results are shown in Table 5.

[0241] Table 5 Project integration degree of reliability design

[0242]

[0243] Step 3: Get the control level of key parts.

[0244] According to the scores given by the five invited experts on the six indicators of key control level, improvement control level, distribution control level, quality control level, inspection control level and procurement control level of the transmission system, the experts' scoring results were obtained, as shown in Table 6. Then, the key parts control level was obtained based on the experts' scoring results.

[0245] Table 6: Scoring table for key parts control level

[0246]

[0247]

[0248] It can be seen from Table 6 that the average level V of the control level of important items is 3.33. Compared with the highest level 5, the control level F of important items is 0.67.

[0249] Step 4: Get reliability completion.

[0250] According to the data of different reliability experiments, the life distribution density function and weight of each type of test are obtained. The fused life distribution density function is obtained, and then the estimated value of the transmission system reliability MTBF is obtained. It is then compared with the target value of the reliability MTBF indicator, and finally the reliability completion degree Q is obtained to be 1.02.

[0251] Step 5: Get the reliability growth coefficient.

[0252] According to the occurrence time of each failure during the reliability growth test and the MTBF values ​​of the first and last tests of the reliability development test, the comprehensive reliability growth coefficient U is 0.442.

[0253] Step 6: Determine the weights of the evaluation indicators related to performance design and reliability design.

[0254] Let the project integration degree, key component control level, reliability growth coefficient and reliability completion degree of reliability design be evaluation indicators 1, 2, 3 and 4 respectively. The evaluation indicators are established using the three-scale method to establish the judgment matrix:

[0255]

[0256] After normalization according to step 6, the project integration degree, key component control level, reliability completion degree and reliability growth coefficient weights of reliability design are obtained respectively.

[0257] Step 7: Obtain the evaluation results of the correlation between performance design and reliability design.

[0258] According to the formula of step 7 and the results of steps 2, 3, 4, 5 and 6, the output result of the correlation evaluation R between performance design and reliability design is 0.813. The evaluation result is: good. The degree of integrated design is high, and the reliability index basically meets the requirements. By sorting and analyzing the results of each index, it can be seen that the reliability growth coefficient is low, indicating that the effect of the corrective measures is average.

[0259] Integration degree Z of reliability design analysis work item a i According to the ranking from low to high scores, the ranking table of the shortcomings of the integration degree of reliability design project a is shown in Table 7. It is necessary to focus on the three work items of "formulating reliability design criteria", "failure mode effect and criticality analysis", and "fault tree analysis". For the reliability design criteria work item, its work has been carried out but has not been fed back to the design. The completion degree and design support degree of the failure mode effect and criticality analysis work item are both low, and the support for the design is less.

[0260] Table 7 Ranking of shortcomings of project integration

[0261] Project Name Project Integration Develop reliability design criteria 0.00 Failure Mode Effects and Criticality Analysis 0.5 Fault Tree Analysis 0.5 Reliability Prediction 0.67 Finite Element Analysis 1.00 Durability analysis 1.00

[0262] The embodiments described above are only descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for evaluating the correlation between performance design and reliability design of an airborne product, characterized by: It includes: Step 1: Obtain basic evaluation data; Basic evaluation data consists of four parts obtained from two sources. The first source is obtained based on the reliability design of airborne products, including project integration and key component control level. The second source is obtained based on reliability experiments, including reliability completion and reliability growth coefficient. Step 2: Obtain the project integration degree of reliability design; When the number of reliability design projects is n, the calculation formula for the project integration degree Z of reliability design is: Among them, Z is the project integration degree of reliability design, n is the number of reliability design projects, a represents one of the reliability design projects, a=1,2…,n; n is a positive integer, Z a The integration degree of a single project; Step 3: Obtain the control level of key parts; The key component control level F∈(0,1) is: Among them, V is the average level of control level of key items, F is the control level of key items, F∈(0,1), P ij represents the evaluation score of the i-th expert on the j-th indicator, i = 1, 2, ..., m, m is the number of experts, j = 1, 2, ..., 6; Step 4: Get reliability completion; The estimated MTBF for airborne products is: Where t is time, f(t) is the lifetime distribution density function after fusion; The reliability completion degree Q is: Among them, Q is the reliability completion degree, MTBF is the estimated value of the reliability of the airborne product in the reliability test. 要求 is the target value of the reliability MTBF indicator; Step 5: Obtain the reliability growth coefficient; Taking 0.7 as the growth limit for normalization, the reliability growth coefficient U∈(0,1) of the performance design after reliability design is: Among them, U1 is the improvement degree of reliability growth test; U2 is the improvement degree of reliability development test; Step 6: Determine the weight of the performance design and reliability design correlation evaluation indicators; Let the project integration degree, key component control level, reliability growth coefficient and reliability completion of reliability design be evaluation indicators 1, 2, 3 and 4 respectively. The three-scale method is used to establish the judgment matrix A=(a ij ) 4×4 ,but: in, Normalize the matrix by column and establish the normalized matrix B=(b ij ) 4×4 , sum the rows of matrix B and perform normalization to establish the judgment matrix w=(w i ) 4×1 , the indicator weights of reliability design project integration degree, key component control level, reliability growth coefficient, and reliability completion degree are obtained as w1, w2, w3, and w4 respectively; Step 7: Obtain the correlation evaluation results between performance design and reliability design; The correlation evaluation value R between performance design and reliability design is: R=Z×w1+S×w2+U×w3+Q×w4; The larger the correlation evaluation value R between performance design and reliability design, the higher the support degree of reliability design for performance design.

2. The method for evaluating the correlation between performance design and reliability design of an airborne product according to claim 1, characterized in that: In step 1, the four parts of data include project integration and key component control levels, reliability completion, and reliability growth coefficient, specifically: (1) Obtain data related to the project integration degree of reliability design, including: reliability prediction, failure mode impact and criticality analysis, formulation of reliability design criteria, fault tree analysis, finite element analysis and durability analysis, and the number of times each project should be carried out during the development process (m) a1 、The actual number of times m a2 , the number of iterations to design improvement m a3 , the number of times the design has not been improved a4 , the number of times no improvement was provided to the design m a5 and the number of times the design is supported but not iterated a6 , where m a1 、m a2 、m a3 、m a4 、m a5 、m a6 a represents one of the reliability design items, a=1,2…,n; n is the number of reliability design items, n is a positive integer; (2) Obtain relevant data on the control level of key items, including six indicators: key control level, improvement control level, distribution control level, quality control level, inspection control level, and procurement control level; (3) Obtain relevant data on reliability completion, including: test time, number of responsible failures and confidence information during the reliability identification test, life test, and reliability acceptance test, as well as reliability index requirements; (4) Obtain relevant data on the reliability growth factor, including the occurrence time of each failure observed during the reliability growth test, and the MTBF values ​​of the first and last tests of the reliability development test.

3. The method for evaluating the correlation between performance design and reliability design of an airborne product according to claim 1, characterized in that: In the step 2, the project integration degree of reliability design is obtained: The design support of item a in reliability design is: The completion degree of reliability design project a is: The integration degree Z of a single item of reliability design project a a , the formula is: Among them, w1 a Design support for item a designed for reliability, w1 a ∈(0,1),m a2 The number of times project a designed for actual reliability is performed, m r5 The number of times that a project designed for reliability does not provide an improvement to the design; Y a The degree of completion of project a designed for reliability, Y a ∈(0,+∞),m a2 The number of times project a designed for actual reliability is performed, m a1 The number of times that item a designed for reliability should be carried out during the development process.

4. The method for evaluating the correlation between performance design and reliability design of an airborne product according to claim 1, wherein: In step 4, the specific steps for obtaining the fused life distribution density function f(t) are as follows: First, for a single reliability experiment, the lower confidence limit at a given confidence level c is θ L ; Then the life distribution density function of airborne products is: f(t)=λe -λt in, t is time; If the reliability experiment is of m types, then m life distribution density functions are fused, and the support vector of the mutual support degree between the reliability acceptance test and other tests is established with the reliability acceptance test as the center point, that is: S=(S 11 S 12 … S 1m ) in, f1(t) represents the life distribution density function obtained from the reliability acceptance test; f i (t) represents the life distribution density function of the i-th type reliability experiment, S 1i represents f1(t) and f i (t) the distance between the two; Determine the weight w21 of the reliability acceptance test as: Among them, L p and L q They represent the lengths of the confidence intervals of the life distribution density function obtained from the reliability acceptance test at confidence levels p and q, respectively, p>q; After obtaining the weight of w21, the weight of the reliability experiment of type i, w2 i for: Among them, S 1i w2 represents the distance between the life distribution density function obtained from the reliability acceptance test and the life distribution density function of the i-th type reliability experiment; i is the weight of the reliability experiment of type i; The final fusion model of life distribution density function is: Among them, f(t) is the lifetime distribution density function after fusion, w2 i is the weight of the reliability experiment of type i; f i (t) represents the life distribution density function of the i-th type reliability experiment, and m is the number of reliability experiment types.

5. The method for evaluating the correlation between performance design and reliability design of an airborne product according to claim 1, characterized in that: In step 5, the specific steps for obtaining the reliability growth test improvement degree U1 and the reliability development test improvement degree U2 are as follows: The reliability growth curve is the AMSAA model, and its expression is: E[N(t)]=at b Where N(t) is the cumulative number of failures that occur during the cumulative test time t; a is the scale parameter; b is the growth shape parameter, and E[N(t)] is the mathematical expectation of N(t); The improvement degree U1 of the reliability growth test is: Where N is the total number of observed faults; t i is the cumulative test time when the i-th fault occurs; T is the total cumulative test time; M is the first parameter, and its value satisfies The reliability development test improvement degree U2 is: Where t1 is the first MTBF obtained from the test, and t2 is the last MTBF obtained from the test.

6. The method for evaluating the correlation between performance design and reliability design of an airborne product according to claim 1, characterized in that: In step 6, after establishing the judgment matrix w, w also needs to pass the consistency test, specifically: Judgment matrix w=(w i ) 4×1 The largest characteristic root of Among them, (Bw) i Represents the i-th row element after matrix B is multiplied by judgment matrix w; Perform consistency test on the judgment matrix w, and the consistency index CI is expressed as: Random consistency index RI, when the number of indicators n is equal to 4, RI = 0.89, the consistency ratio CR is If CR<0.1, the eigenvector of the judgment matrix w is used to represent the weight vector; If the consistency check fails, it means that the setting of matrix A is unreasonable, and the values ​​in matrix A need to be reset.

7. The method for evaluating the correlation between performance design and reliability design of an airborne product according to claim 1, characterized in that: The step seven further includes: According to the value range of the performance design and reliability design correlation evaluation value R, segmentation is performed so that the performance design and reliability design correlation evaluation R corresponds to evaluation results of different levels.

8. The method for evaluating the correlation between performance design and reliability design of an airborne product according to claim 1, characterized in that: Also includes: Step 8: Analyze the correlation between performance design and reliability design, specifically: 1) Sort Z, F, Q, and U by score from low to high to obtain a ranking table. The lowest-scoring indicator is the weak link in the combination of performance design and reliability design. 2) For the project integration degree Z, according to the reliability design analysis work in item a, the project integration degree Z i Sort by scores from low to high and output a ranking table of project integration shortcomings; 3) For the control level F of key items, sort them from low to high according to the scoring results of the key item control level indicators to obtain a ranking table of the short board of the control level; 4) For the reliability completion degree Q, if Q>0, it means that the specified reliability index requirements have been met; if Q<0, it means that the specified reliability index requirements have not been met; 5) For the reliability growth coefficient U, based on the comparison of U1 and U2, a ranking table of the short board in terms of reliability growth degree is obtained; From 1) to 5), the lower the score, the more improvement is needed.

9. A system using a method for evaluating the correlation between performance design and reliability design of an airborne product, characterized in that: It includes: Database, reliability design project integration module, key component control level module, reliability completion module, reliability growth coefficient module, weight module and correlation evaluation module, wherein the reliability design project integration module, key component control level module, reliability completion module and reliability growth coefficient module are connected to the database, weight module and correlation evaluation module respectively; The database is used to store the data required for the performance reliability integrated design degree evaluation, including a reliability design project sub-database, a key component control level sub-database, a reliability completion sub-database, and a reliability growth factor sub-database; wherein the reliability design project sub-database includes the number of times m that reliability prediction, failure mode effect and criticality analysis, formulation of reliability design criteria, fault tree analysis, finite element analysis, and durability analysis projects should be carried out during the development process. a1 , the number of iterations to design improvement m a3 , the number of times no improvement was provided to the design m a5 , the number of times the design is supported but not iterated again m a6 The key component control level sub-database includes six indicator values, namely, key control level, improvement control level, allocation control level, quality control level, inspection control level, and procurement control level, obtained by expert scoring. The reliability completion sub-database includes the test time, number of responsible failures, confidence information, and reliability indicator requirements during the reliability identification test, life test, and reliability acceptance test. The reliability growth factor sub-database includes the occurrence time of each failure observed during the reliability growth test, as well as the MTBF values ​​of the first and last tests of the reliability development test. The project integration degree module of reliability design is used to obtain the project integration degree Z of reliability design, which is achieved by the method in step 2; The critical component control level module is used to obtain the critical component control level F, which is implemented in the manner of step 3; The reliability completion module is used to obtain the reliability completion degree Q, which is implemented using the method in step 4; The reliability growth coefficient module is used to obtain the reliability growth coefficient U, which is implemented using the method in step 5; The weight module is used to establish a judgment matrix using the three-scale method to obtain the weights of project integration degree Z, key component control level F, reliability completion degree Q and reliability growth coefficient U; The correlation evaluation module obtains the correlation evaluation value of performance design and reliability design according to the weight output by the weight module, the project integration degree Z output by the project integration degree module of reliability design, the key component control level F output by the key component control level module, the reliability completion degree Q output by the reliability completion degree module, and the reliability growth coefficient U output by the reliability growth coefficient module, using the step seven method.

10. The system using the performance design and reliability design correlation evaluation method of an airborne product according to claim 9, characterized in that: Also includes: Correlation analysis module, The correlation analysis module is connected with the project integration module of reliability design, the key component control level module, the reliability completion module, the reliability growth coefficient module and the correlation evaluation module respectively.

Images