Missile standby readiness rate evaluation method based on multi-source data

Abstract

The invention discloses a missile standby readiness rate evaluation method under multi-source data, belonging to the field of missile technology, comprising: classifying missile products into maintainable products and non-repairable products in the standby stage; performing availability calculation on the maintainable products to obtain the availability of the maintainable products; performing data fusion and type conversion on different environmental profile test data of the non-repairable products to obtain success-failure test data; based on the information equality principle, obtaining fused system success-failure data according to the availability and success-failure test data, wherein the system success-failure data includes the number of virtual tests and the number of virtual successes; obtaining final system success-failure test data based on the fused system success-failure test data and the own system success-failure data, and obtaining the missile standby readiness rate based on the final system success-failure data. Through the above technical scheme, the invention solves the problem of insufficient accuracy in the evaluation of the missile system standby readiness rate.

Description

Technical Field

[0001] The invention belongs to the technical field of missiles, and in particular relates to a missile standby readiness rate evaluation method under multi-source data. Background Art

[0002] Missiles are characterized by "long-term storage and one-time use" and are a type of high-reliability weapon equipment. The missile standby readiness rate usually refers to the possibility that a missile system that has completed standby preparation work can complete standby preparation work according to the established goals within the specified standby time. During the standby preparation stage, some products on the missile will be in working condition, and some products can be repaired and replaced, which can improve the final launch rate of the missile to a certain extent. Standby preparation is the duty launch state of the missile after the standby preparation stage. When evaluating the missile standby readiness rate, due to time and cost issues, it is usually impossible to obtain sufficient full missile test information. It is necessary to use the test information of each subsystem, component and different environment to expand the sample size to achieve the purpose of evaluation. Therefore, how to make full use of the test information of subsystems, components and different environments under different categories to construct a missile standby readiness rate evaluation model has become a very important issue.

[0003] At present, the requirements for product reliability in the military, aerospace and other fields are getting higher and higher. However, due to the high cost of destructive tests and the long product life, there are relatively few test samples available for reliability assessment. When evaluating the missile standby readiness rate, it is often impossible to obtain sufficient full missile test information. It is inevitable to use the test information of each subsystem, whole machine and component to evaluate the standby readiness rate of the whole missile. In the comprehensive evaluation of the missile system standby readiness rate, the use of subsystem-level and component-level test data under different categories can greatly improve the accuracy of the evaluation of the missile system standby readiness rate. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a missile standby readiness rate evaluation method based on multi-source data, so as to solve the problem that the above-mentioned prior art cannot effectively utilize subsystem-level and component-level test data under different categories, resulting in insufficient accuracy in the evaluation of the missile system standby readiness rate.

[0005] To achieve the above object, the present invention provides a missile standby readiness rate evaluation method under multi-source data, comprising:

[0006] Classify the missile products into repairable products and non-repairable products in the standby stage, among which the repairable products include those that can be repaired in continuous use and those that can be repaired in intermittent use;

[0007] Performing availability calculation on the repairable product to obtain the availability of the repairable product;

[0008] The different environmental profile test data of unrepairable products are fused and converted into different types to obtain success-failure test data. The fusion of different environmental profile data is based on the expert scoring method.

[0009] Based on the information equality principle, according to the availability and success / failure test data, the fusion system success / failure data is obtained, wherein the fusion system success / failure data includes the number of virtual tests and the number of virtual successes;

[0010] Based on the fusion system success / failure test data and the system's own success / failure data, the final system success / failure test data is obtained, and based on the final system success / failure data, the missile standby readiness rate is obtained.

[0011] Optionally, the process of calculating the availability of the repairable product includes:

[0012] Calculate the availability of the repairable product:

[0013] ;

[0014] Among them, A represents availability, Mean time between failures, represents the mean time to repair;

[0015] Availability calculation is performed on the intermittent use repairable product:

[0016]

[0017] in, Indicates the inverse of the usage rate during standby preparation.

[0018] Optionally, the process of fusing data from different environmental profiles includes:

[0019] Scoring different environments according to the expert scoring method, calculating the mean of the scoring results in the same environment, and obtaining the environmental factors of different environments;

[0020] According to the environmental factors of different environments, the different environmental profile test data of non-repairable products are converted into test data under the same environment.

[0021] Optionally, the process of obtaining the success or failure data of the fusion system includes:

[0022] For non-repairable products in the standby stage:

[0023]

[0024] According to the principle of information equality, the number of virtual tests of the system is calculated:

[0025] n ¯ = ∑ j = 1 a n j [ P j ln P j + ( 1 − P j )ln( 1 − P j )] ∑ j = 1 a [ P j ln P j + ( 1 − P j )ln( 1 − P j )]

[0026] Among them, s j , n j represents the success and failure test data of the jth non-repairable product, is the total number of unrepairable products in the standby phase, P j represents the reliability estimate of the jth unrepairable product, j represents the number of the unrepairable product, represents the number of virtual trials;

[0027] For repairable products in the standby stage:

[0028]

[0029] Among them, P i represents the estimated reliability value of the i-th repairable product, For the The availability of a product, i represents the repairable product number, n i represents the number of tests of the ith repairable product, is the total number of products, S i represents the number of successful attempts of the i-th repairable product;

[0030] According to the principle of information equality, the number of virtual successes of the system is calculated:

[0031]

[0032] is the number of virtual successes.

[0033] Optionally, the process of obtaining the final system success or failure test data includes:

[0034] Based on the success and failure test data of the fusion system and the success and failure data of the system itself, a chi-square distribution table is constructed, and the weight value is queried according to the chi-square distribution table, and the weight factor is calculated according to the weight value, where:

[0035]

[0036] in, is the weight factor corresponding to the i-th indicator, represents the i-th weight value, and n is the total number of weight values;

[0037] The fusion system success / failure test data and the own system success / failure test data are weighted and calculated according to the weight factor to obtain the final system success / failure test data.

[0038] Optionally, the type conversion process includes:

[0039] The exponential data under the same environment , converted into success or failure data , and obtain the success / failure test data:

[0040] n = [exp( r / η ) − 1 ] η 2 r

[0041]

[0042] in, represents the number of failures, represents the number of equivalent tasks, represents the total test time, represents the task time, n represents the total number of trials, and s represents the number of successes.

[0043] Optionally, among the repairable products, the continuously-use repairable product includes a computer, and the intermittently-use repairable product includes a servo mechanism and a controller.

[0044] Compared with the prior art, the present invention has the following advantages and technical effects:

[0045] a) This method takes into account the fact that there is little information in the missile full-range test and makes full use of the test information of the subsystem to complete the missile standby readiness rate assessment;

[0046] b) The use and maintenance of missile onboard products during the standby preparation phase were considered, the missile onboard products were classified, and the concept of availability was introduced to evaluate the onboard products;

[0047] c) Fully consider that each missile subsystem will obtain test data under different test environments, and make full use of it by converting it into test data under a single environment;

[0048] d) Considering that the missile has both success-failure type units and life-span type units, the test data are converted into success-failure type test data for evaluation through conversion methods.

[0049] e) When fusing the fused system-level test data with the own system-level test data, different methods for determining the fusion weight factor are given for two different situations: when the own system-level test data is missing or available. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0051] Figure 1 The figure is a flowchart of a method for evaluating missile standby readiness rate according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0053] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0054] The present invention proposes a missile standby readiness rate evaluation method based on multi-source data to solve the problem that the above-mentioned prior art cannot effectively utilize subsystem-level and component-level test data under different categories, resulting in insufficient accuracy in evaluating the missile system standby readiness rate. It should be noted that in addition to effectively utilizing the subsystem-level and component-level test data of missile products, the technical solution of the present invention also focuses on providing a method for comprehensively evaluating the missile standby readiness rate based on the classification of missile products, that is, based on the repairable data and the non-repairable data.

[0055] The present invention processes the missile subsystem-level test data and the system-level test data as the basis, and uses the data fusion method to evaluate the missile's standby readiness rate. First, the missile-mounted products are classified according to the use and maintenance conditions, and are divided into three categories: repairable products used continuously in the standby stage, repairable products used intermittently in the standby stage, and unrepairable products in the standby stage. Then, the availability of the two categories of products, namely, the repairable products used continuously in the standby stage and the repairable products used intermittently in the standby stage, is calculated. For the unrepairable products in the standby stage, starting from the test data of each subsystem level, the test data of different environmental profiles are fused into the test data under the same environmental profile, and then the life test data is converted into the success-failure test data. According to the converted subsystem success-failure test data, its reliability is obtained. Finally, the data fusion method is comprehensively applied to integrate with the system-level test data to obtain the final system-level success-failure test data, and the missile's standby readiness rate evaluation is completed.

[0056] The technical solution process of the present invention is described as follows:

[0057] The logic flow of the missile standby readiness rate evaluation method of the present invention is shown in the attached figure. Figure 1 , achieved through the following steps:

[0058] Step 1: Pop up product categories:

[0059] First, the on-board products (subsystems and components loaded on the missile) are classified according to their usage and maintenance conditions. They are divided into three categories: repairable products that are used continuously during the standby phase, repairable products that are used intermittently during the standby phase, and non-repairable products during the standby phase.

[0060] Step 2: Availability calculation:

[0061] The availability of the repairable products that are continuously used in the standby phase and the repairable products that are intermittently used in the standby phase obtained after classification in step 1 is calculated.

[0062] Step 3: Fusion of different environmental profile data at the subsystem level:

[0063] For non-repairable products in the standby stage, test data will be obtained in different environments such as warehouses, fields, and laboratories. The failure rates of components in different environments are different. The test data cannot be directly superimposed and used, and need to be converted.

[0064] Step 4: Experimental data type conversion:

[0065] For products that are not repairable in the standby phase, the types of units of various components of the missile are different, including both success-failure type units and life-span type units. This makes the type of reliability data obtained from the test complex and diverse, including both success-failure type and non-success-failure type test data. Therefore, for the convenience of evaluation, it is necessary to change the test data type from life-span type to success-failure type.

[0066] Step 5: Fusion of product data at different levels:

[0067] Based on the availability obtained in step 2 and the success / failure test data of the unrepairable product in step 4, the success / failure test data of the system level is calculated.

[0068] Step 6: Determine the system-level fusion weight factor:

[0069] Based on the system-level success and failure data obtained in step three and the system-level test data itself, determine the proportion of each when the two are fused, that is, the weight factor, and perform data fusion based on the weight factor.

[0070] Step 7: Calculate the standby readiness rate:

[0071] Using the data fusion results in step six as evaluation input, calculate the missile standby readiness rate.

[0072] The above technical solution of the present invention is described in detail:

[0073] Step 1: Pop up product categories:

[0074] The products on the missile are classified according to their usage and maintenance conditions. They are divided into three categories: repairable products that are used continuously in the standby phase, repairable products that are used intermittently in the standby phase, and non-repairable products in the standby phase.

[0075] Step 2: Availability calculation:

[0076] Calculate the availability of the repairable products (continuous use repairable products and intermittent use repairable products) among the products classified in step 1.

[0077] Availability of repairable products for continued use It can be calculated as follows:

[0078]

[0079] In the formula,

[0080] : Mean time between failures;

[0081] : Mean time to repair.

[0082] When calculating the availability of intermittently used repairable products, Make corrections, availability It can be calculated as follows:

[0083]

[0084] In the formula,

[0085] : Mean time between failures;

[0086] : Mean time to repair;

[0087] : The inverse of the usage rate during standby preparation.

[0088] in, , and All are known values.

[0089] If the mean time to repair and the average warranty interval If unknown, it can be calculated using the following method.

[0090] Mean time to repair and the average warranty interval Maintainability test data and supportability test data are used for evaluation and calculation respectively.

[0091] Mean time to repair The evaluation can be calculated using the following model:

[0092]

[0093]

[0094]

[0095] In the formula, represents the mean of the observed values ​​of maintenance time, X i represents the observed value of the i-th maintenance time, where i=1, 2, ..., M1, M1 represents the total number of observations of maintenance time, represents the confidence level of the repair time, represents the mean square error of the maintenance time observations.

[0096] The following model can be used to evaluate the average guarantee delay time:

[0097]

[0098]

[0099]

[0100] In the formula, represents the mean of the observed values ​​of the guarantee interval, represents the observation value of the i-th guarantee interval, where i=1, 2, ..., M2, M2 represents the total number of observation values ​​of the guarantee interval, represents the confidence level of the guaranteed interval time, represents the mean square error of the observations of the guarantee interval.

[0101] Step 3: Fusion of different environmental profile data at the subsystem level:

[0102] For products that cannot be repaired in the standby phase, each component of the missile will obtain test data in different environments such as launch sites and laboratories. The failure rate of components in different environments is different. The test data cannot be directly superimposed and used, and need to be converted.

[0103] The conversion method is as follows:

[0104] Combined with the expert scoring method, experts are invited to score the severity of each environment or environmental factors, and the test data are converted based on the scoring results.

[0105] The specific scoring rules are as follows: the basic score of the environmental information for obtaining the test samples is set to 1.0, and on this basis, the scores of the products in the historical sample environment (different from the environment for obtaining the test samples mentioned above) are given. Generally, there are three situations:

[0106] (1) K < 1, indicating that under the influence of this factor, the reliability of the product or functional component in the historical sample environment is higher than the current reliability.

[0107] (2) K = 1, indicating that under the influence of this factor, the reliability of the product or functional component in the historical sample environment is equivalent to the reliability of the current one.

[0108] (3) K>1, indicating that under the influence of this factor, the reliability of the product or functional component in the historical sample environment is worse than the current reliability.

[0109] Assuming that there are n experts scoring and there are m different environments, the environmental factors corresponding to each environment are:

[0110]

[0111] Among them, F pq Indicates different expert scoring results, F p It represents the environmental factor of the pth environment, p represents different environmental labels, and q represents different expert labels.

[0112] Finally, the data of different environmental profiles are converted into experimental data under a single environment according to the obtained environmental factors.

[0113] Step 4: Experimental data type conversion:

[0114] The unit types of missile components are different, including success-failure type units and life-span type units, which makes the reliability data type obtained from the test complex and diverse, including success-failure type and non-success-failure type test data such as exponential type. Therefore, for the convenience of evaluation, it is necessary to change the test data type from life-span type (exponential type) to success-failure type.

[0115] Exponential data ( is the number of failures, is the number of equivalent tasks, is the total test time, =Task time) is converted into success / failure data (n represents the total number of trials, s represents the number of successes) is calculated as follows:

[0116] n = [exp( r / η ) − 1 ] η 2 r

[0117] .

[0118] Step 5: Fusion of product data at different levels:

[0119] Based on the availability calculated in step 2 and the success / failure test data of the subsystem in step 4, the success / failure data of the system-level fusion system is calculated as follows:

[0120] First, for the unmaintainable standby phase, using the success / failure data in step 4, according to the principle of information equality, the average number of subsystems can be obtained to obtain the number of tests of the total system:

[0121]

[0122] n ¯ = ∑ j = 1 a n j [ P j ln P j + ( 1 − P j )ln( 1 − P j )] ∑ j = 1 a [ P j ln P j + ( 1 − P j )ln( 1 − P j )]

[0123] Where: s j , n j represents the success and failure test data of the jth non-repairable product, is the total number of unrepairable products in the standby phase, P j represents the reliability estimate of the jth unrepairable product, and j represents the label of the unrepairable product.

[0124] Then, for the continuously used repairable products and the intermittently used repairable products, using the availability obtained in step 2, we can get:

[0125]

[0126] in, For the The availability of a product, is the total number of products, P i represents the estimated reliability value of the i-th repairable product, n i represents the number of tests of the ith repairable product, S i Represents the number of successful attempts of the i-th repairable product.

[0127] Finally, using the "additivity" characteristic of information quantity in information theory and the general expression of information entropy, the above data are calculated to ensure that the information contained in the data is not lost. According to the principle of information equality, we can get:

[0128]

[0129] is the number of virtual trials, is the number of virtual successes. Both are not necessarily integers and can be replaced by the nearest integer.

[0130] Step 6: Determine the system-level fusion weight factor:

[0131] Based on the fusion system success / failure data obtained in step five and the system-level test data itself (data obtained from actual tests under the same environment), determine the proportion of each when the two are fused, that is, the weight factor.

[0132] When the system-level test data is missing, the fused system-level success and failure data can be directly used for evaluation. That is:

[0133]

[0134] is the number of virtual trials in step 5 ,

[0135] is the number of virtual successes in step 5 .

[0136] When there is system-level test data of its own, the system-level success and failure data and its own system-level test data can be evaluated by goodness-of-fit test.

[0137] If the system test information , Subsystem-level information distribution From the overall , construct the contingency table shown in Table 1:

[0138] Table 1

[0139]

[0140] make:

[0141] K i = [ | n b i − b i f − a i f | − n + a i + b i 2 ] 2 ( n + a i + b i ) n ⋅ ( n − f + a i ) ( f + b i ) ( a i + b i )

[0142] In the formula, is the Pearson chi-square statistic, which approximately follows the 1-degree-of-freedom Distribution, the system test information , Subsystem-level information distribution Substituting into the above formula, we can get .

[0143] Step 7: Calculate the standby readiness rate:

[0144] Using the system-level success and failure data fusion results in step 4 as data support, the missile standby readiness rate is calculated using the following formula:

[0145]

[0146] in: is the number of missiles that have successfully completed standby preparation, is the total number of missiles involved in standby preparation, and P represents the missile standby readiness rate.

[0147] Combined with relevant actual numerical data, the above technical solution of the present invention is described in detail:

[0148] This method is explained by taking a certain type of missile as an example.

[0149] This type of missile system consists of six parts connected in series: warhead, power system, body, computer, and servo mechanism. Among them, the warhead and body are success-failure systems, and the power system is an exponential system.

[0150] The test data of the warhead, body and missile system are shown in Table 2 below:

[0151] Table 2

[0152] Number of trials Number of successes Number of failures warhead 30 30 0 Projectile 30 29 1 Missile systems 5 4 1

[0153] The power system has test data at the launch site and the development laboratory. The test data at the launch site is , timed truncation The failure times are , the test data in the laboratory is , timed truncation The failure times are .

[0154] Step 1: Pop up product categories:

[0155] The products on the missile are classified according to their usage and maintenance conditions. Computers are repairable products that are used continuously during the standby preparation phase, servo mechanisms and controllers are repairable products that are used intermittently during the standby preparation phase, and warheads, power systems, and missile bodies are non-repairable products during the standby preparation phase.

[0156] Step 2: Availability calculation:

[0157] Computers use repairable products continuously in the standby preparation stage, while servo mechanisms and controllers use repairable products intermittently in the standby preparation stage. The availability of the three products is calculated. The mean time between failures, mean time to repair, and the inverse of the usage rate during the standby preparation process are shown in Table 3 below:

[0158] Table 3

[0159] MTBF(h) MTTR(h) K computer 15 0.5 1 Servo Mechanism 20 0.4 2 Controller 35 1 2.5

[0160] Computer availability is

[0161] The servo mechanism availability is

[0162] The controller availability is .

[0163] Step 3: Fusion of different environmental profile data at the subsystem level:

[0164] The control system has test data from the launch site and the development laboratory respectively. Since standby preparation is completed at the launch site, the test data in the development laboratory needs to be converted into test data at the launch site.

[0165] Invite three experts to score the two environments. Taking the launch site environment as the benchmark, we can get the relevant scores in Table 4:

[0166] Table 4

[0167] Expert 1 Expert 2 Expert 3 Launch Site 1 1 1 Research Laboratory 0.6 0.7 0.62

[0168] Get the environmental factor:

[0169]

[0170] The test data in the laboratory is converted into the number of test pieces put into the launch field environment. , timed truncation The failure times are .

[0171] Step 4: Experimental data type conversion:

[0172] Based on the exponential test data of the power system in step 1, it is converted into success-failure test data. The calculation results are as follows:

[0173] Number of equivalent tasks:

[0174]

[0175] Converted number of tests:

[0176] n 1 = [exp( r 1 / η 1 ) − 1 ] η 1 2 r 1 =19.69

[0177] n 2 = [exp( r 2 / η 2 ) − 1 ] η 2 2 r 2 =20.04

[0178] =39.73

[0179] Converted number of successful tests:

[0180] =16.68

[0181] =18.04

[0182] =36.72.

[0183] Step 5: Fusion of product data at different levels:

[0184] Based on the converted power system success / failure test data in step 4 and the warhead and body success / failure test data, the reliability is calculated. The calculation results are shown in Table 5 below:

[0185] Table 5

[0186] Number of trials Number of successes Number of failures warhead 30 30 0 Projectile 30 29 1 Powertrain 40 37 3

[0187]

[0188] n ¯ = ∑ j = 1 3 n j [ P j ln P j + ( 1 − P j )ln( 1 − P j )] ∑ j = 1 3 [ P j ln P j + ( 1 − P j )ln( 1 − P j )] = 36 . 48 .

[0189] Based on the availability of the three products, computer, servo mechanism and controller in step 2, the reliability calculation results are as follows:

[0190]

[0191]

[0192] Calculate the number of equivalent test successes:

[0193] S ¯ = ∑ j = 1 6 S j [ P j ln P j + ( 1 − P j )ln( 1 − P j )] ∑ j = 1 6 [ P j ln P j + ( 1 − P j )ln( 1 − P j )] = 31 . 38

[0194] The system-level success / failure test data is rounded to (36, 31).

[0195] Step 6: Determine the system-level fusion weight factor:

[0196] The final system success / failure test data is obtained by taking the system success / failure test data obtained by fusion in step 5 shown in Table 6 and the system success / failure test data itself as the data basis.

[0197] Table 6

[0198] Number of trials Number of successes Number of failures Fusion to test data 36 31 5 Own test data 5 4 1 and 41 35 6

[0199] Since there is system-level test data, the system-level success and failure data and the system-level test data can be evaluated by goodness of fit test. The calculation is as follows:

[0200] K i = [ | n b i − b i f − a i f | − n + a i + b i 2 ] 2 ( n + a i + b i ) n ⋅ ( n − f + a i ) ( f + b i ) ( a i + b i ) = 0 . 0 9 8

[0201] Look up the table:

[0202]

[0203] Indicates the number of trials in the system success / failure test data itself, Indicates the number of successes in the system success / failure test data.

[0204] Step 7: Calculate the missile standby readiness rate

[0205] Using the system-level success and failure data fusion results in step 4 as data support, the missile standby readiness rate is calculated, and the calculation results are as follows:

[0206] .

[0207] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A missile standby readiness rate assessment method based on multi-source data, characterized in that: include: Classify the missile products into repairable products and non-repairable products in the standby stage, among which the repairable products include those that can be repaired in continuous use and those that can be repaired in intermittent use; Performing availability calculation on the repairable product to obtain the availability of the repairable product; The different environmental profile test data of unrepairable products are fused and converted into different types to obtain success-failure test data. The fusion of different environmental profile data is based on the expert scoring method. Based on the information equality principle, according to the availability and success / failure test data, the fusion system success / failure data is obtained, wherein the fusion system success / failure data includes the number of virtual tests and the number of virtual successes; Based on the fusion system success / failure test data and the own system success / failure data, the final system success / failure test data is obtained, and based on the final system success / failure data, the missile standby readiness rate is obtained; The process of calculating the availability of the repairable product includes: Calculate the availability of the repairable product: ； Among them, A represents availability, Mean time between failures, represents the mean time to repair; Availability calculation for the intermittently used repairable product: in, Indicates the inverse of the utilization rate during standby preparation; The process of fusing different environmental profile data includes: Scoring different environments according to the expert scoring method, calculating the mean of the scoring results in the same environment, and obtaining the environmental factors of different environments; According to the environmental factors of different environments, the different environmental profile test data of non-repairable products are converted into test data under the same environment; The process of obtaining the success and failure data of the fusion system includes: For non-repairable products in the standby stage: According to the principle of information equality, the number of virtual tests of the system is calculated: Among them, s j , n j represents the success and failure test data of the jth non-repairable product, is the total number of unrepairable products in the standby phase, P j represents the reliability estimate of the jth unrepairable product, j represents the number of the unrepairable product, represents the number of virtual trials; For repairable products in the standby stage: Among them, P i represents the estimated reliability value of the i-th repairable product, For the The availability of a product, i represents the repairable product number, n i represents the number of tests of the ith repairable product, is the total number of products, S i represents the number of successful attempts of the i-th repairable product; According to the principle of information equality, the number of virtual successes of the system is calculated: is the number of virtual successes; The process of obtaining the final system success or failure test data includes: Based on the success and failure test data of the fusion system and the success and failure data of the system itself, a chi-square distribution table is constructed, and the weight value is queried according to the chi-square distribution table, and the weight factor is calculated according to the weight value, where: in, is the weight factor corresponding to the i-th indicator, represents the i-th weight value, and n is the total number of weight values; The fusion system success / failure test data and the own system success / failure test data are weighted and calculated according to the weight factor to obtain the final system success / failure test data; The type conversion process includes: The exponential data under the same environment , converted into success or failure data , and obtain the success / failure test data: in, represents the number of failures, represents the number of equivalent tasks, represents the total test time, represents the task time, n represents the total number of trials, and s represents the number of successes.

2. The method according to claim 1, characterized in that: Among the repairable products, the continuous use repairable products include computers, and the intermittent use repairable products include servo mechanisms and controllers.

Images