CN112381456A

CN112381456A - Equipment development phase testability evaluation and system based on conflict data fusion

Info

Publication number: CN112381456A
Application number: CN202011411320.8A
Authority: CN
Inventors: 狄鹏; 王旋; 胡涛; 陈童; 胡斌; 刘刚; 杨晶; 吕建伟
Original assignee: Naval University of Engineering PLA
Current assignee: Naval University of Engineering PLA
Priority date: 2020-12-03
Filing date: 2020-12-03
Publication date: 2021-02-19
Anticipated expiration: 2040-12-03
Also published as: CN112381456B

Abstract

The invention relates to the technical field of uncertain information decision, in particular to equipment development phase testability evaluation and a system based on conflict data fusion, wherein the method comprises the following steps: respectively converting the testability data of each development stage into a prior distribution form; constructing a corresponding quality function based on a prior distribution form of testability information of each development stage; synthesizing the quality function values of the testability data in each development stage to obtain the quality function value corresponding to the final data fusion result; and analyzing the testability evaluation result by using the quality function value of each final data fusion result. The invention can effectively reduce the adverse effect caused by the problem of test data conflict.

Description

Equipment development phase testability evaluation and system based on conflict data fusion

Technical Field

The invention relates to the technical field of uncertain information decision, in particular to equipment development phase testability evaluation and system based on conflict data fusion.

Background

Since the 21 st century, testability has been receiving increasing attention from equipment manufacturers and users as one of the common quality characteristics of equipment. With the development of equipment testability engineering, testability testing and evaluation methods are one of the hot spots of current research in the field of testability. Testability evaluation is an important link in equipment testability design, and is often used for testing and evaluating work performed to check whether equipment meets the testability development requirements. In the equipment development stage, defects of the testability design can be found through evaluating the testability level of the equipment, and the progress of the equipment design shaping process is concerned. However, in the process of equipment development, without a full-state prototype or a normal prototype, the collected fault samples and testability information cannot completely reflect the true level of equipment testability, and the equipment testability level is difficult to be evaluated comprehensively and accurately. Meanwhile, prior information such as expert experience information and the like in the equipment development stage has certain subjectivity and uncertainty, and the phenomenon of conflict in the data fusion process is easily caused. Therefore, how to scientifically and effectively carry out testability evaluation work in the equipment development stage becomes one of the hot spots of current testability field research.

Disclosure of Invention

In order to solve the technical problems, the equipment development phase testability evaluation and system based on the conflict data fusion can effectively reduce adverse effects caused by the conflict problem of test data.

On one hand, the equipment development phase testability evaluation method based on conflict data fusion provided by the invention comprises the following steps:

respectively converting the testability data of each development stage into a prior distribution form;

constructing a corresponding quality function based on the prior distribution form of the testability information in each development stage;

synthesizing the quality function values of the testability data in each development stage to obtain the quality function value corresponding to the final data fusion result;

and analyzing the testability evaluation result by using the quality function value of each final data fusion result.

Further, the development stage testability data includes: expert experience information, testability growth test data, and replaceable unit test data.

Further, the constructing a corresponding quality function specifically includes:

constructing an original quality function corresponding to testability data in a development stage;

respectively giving final fusion weights to the original quality functions of the testability data in each development stage to obtain corresponding modified quality functions;

and taking each corrected quality function as a quality function corresponding to the testability information in the development stage.

Further, the step of respectively assigning final fusion weights to the original quality functions of the testability data in each development stage further includes:

and constructing corresponding final fusion weight based on the credibility and uncertainty of the testability data in each development stage.

Furthermore, the reducing the testability data of each development stage into a prior distribution form includes:

and realizing the prior distribution form of the expert experience information by adopting a maximum entropy theory.

and (4) injecting the fault number, and solving the prior distribution form of the testability growth test data based on the Ongerbet model or the F distribution.

Further, the method specifically comprises the following steps:

and obtaining a prior distribution form of the test data of the replaceable unit based on a Bayesian formula.

In the above technical solution, the synthesizing of the quality function value of the testability data at each development stage specifically includes:

synthesizing the quality function values of the testability data of each development stage for N-1 times by applying a D-S evidence theory, wherein N is the number of the evidences;

and the obtained N-1 time fusion result is the quality function value of the final data fusion result.

Preferably, the constructing the corresponding final fusion weight based on the credibility and uncertainty of the testability data of each development stage further comprises:

calculating the credibility of the testability data in each development stage based on the Langmuir distance;

and determining the uncertainty of the testability data in each development stage by using a D-S evidence theory fusion method improved by the information entropy.

On the other hand, the equipment development stage testability evaluation system based on conflict data fusion provided by the invention comprises: a processor and a memory, the processor reading a computer program in the memory for performing the following operations:

The invention can effectively reduce the adverse effect caused by the problem of test data conflict and improve the accuracy of equipment testability index evaluation, thereby providing high-quality standard reference for equipment bearing and using parties.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a schematic flow chart of a method according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of an evaluation step according to an embodiment of the present invention;

FIG. 3 is a schematic flow chart illustrating the conversion of testability data into a priori distribution form at each development stage according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of a system according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1, the method for evaluating the testability of the equipment development stage based on the conflict data fusion provided by this embodiment includes:

101. respectively converting the testability data of each development stage into a prior distribution form;

the development stage testability data comprises: expert experience information, testability growth test data, and replaceable unit test data. Specifically, the method comprises the following steps:

1011. realizing a prior distribution form of expert experience information by adopting a maximum entropy theory;

1012. injecting the fault number, and solving a prior distribution form of the test growth experiment data based on a Gompertz (Gompertz) model or an F distribution;

1013. and obtaining a prior distribution form of the replaceable unit test data based on a Bayesian (Bayes) formula.

102. Constructing a corresponding quality function (mass) based on a prior distribution form of testability information at each development stage; specifically, the method comprises the following steps:

1021. constructing an original quality function corresponding to testability data in a development stage;

1022. respectively giving final fusion weights to the original quality functions of the testability data in each development stage to obtain corresponding modified quality functions;

1023. and taking each corrected quality function as a quality function corresponding to the testability information in the development stage.

103. Synthesizing the quality function values of the testability data in each development stage to obtain the quality function value corresponding to the final data fusion result; specifically, the method comprises the following steps:

1031. synthesizing the quality function values of the testability data of each development stage for N-1 times by applying a D-S evidence theory, wherein N is the number of the evidences;

104. And analyzing the testability evaluation result by using the quality function value of each final data fusion result.

The final fusion weight is respectively given to the original quality function of the testability data in each development stage, and the method also comprises the following steps:

105. and constructing corresponding final fusion weight based on the credibility and uncertainty of the testability data in each development stage.

Constructing corresponding final fusion weight based on the credibility and uncertainty of testability data in each development stage, wherein the method comprises the following steps:

106. calculating the credibility of the testability data in each development stage based on Lance (Lance) distance;

107. and determining the uncertainty of the testability data in each development stage by using a D-S evidence theory fusion method improved by the information entropy.

The method and the device can effectively solve the problem of test data conflict, improve the accuracy of equipment testability index evaluation, and provide high-quality standard reference for equipment bearing and using parties.

The above technical solutions of the embodiments of the present invention are described in detail below with reference to application examples:

as shown in fig. 2 and 3, the method for evaluating the testability of the equipment development stage based on the conflict data fusion provided by this embodiment includes:

step 1, collecting expert experience information, testability growth test data and replaceable unit test data existing in an equipment development stage;

in the equipment development stage, various testability information types exist, such as expert experience information, testability growth test data, replaceable unit test data and the like, and the testability information can be used as a priori information type for carrying out testability evaluation work in the development stage.

Step 2, converting the development stage testability information of each source into a prior distribution form;

step 2.1, aiming at expert experience information:

prior distribution of expert experience information pi_E(p) is usually expressed in terms of Beta distribution:

the determination of the prior distribution parameters is usually achieved by the maximum entropy method, so that the prior distribution pi_EThe information entropy of (p) is:

thus, the prior distributed parameter solution process can be transformed into a search such that the entropy function H [ π [ ]_E(p)]Planning the maximum time parameters a and b;

if the point estimate p is known₀Then it is a priori distributed pi_E(p) is:

the following equation (1) and equation (3) are obtained:

according to the Beta function property, the following steps are carried out:

then, combining formula (2) and formula (5) yields:

H[π_E(p)]＝ln(B(a,b))-a₁B₁-b₁B₂ (6)

wherein:

the prior distribution parameter pi_EThe optimal solution of (p) can be converted to solve the following planning problem:

step 2.2, aiming at 2) testing the experimental data of the increase:

the testability growth test is a special test which takes a testability index value agreed by a bearer and a user as a target, injects a fault into equipment through testability design, enables the equipment to operate in a specified environmental stress, observes the performance condition of the detection/isolation of statistical faults in a test system, finds the reason of failure detection/isolation failure or testability index not reaching the target, further improves the testability design and verifies the improvement measure. The testability growth test mainly comprises three processes of identifying design defects, feeding back problems and improving design, and can be summarized into a process of 'test-analysis-improvement-test'.

The Gompertz model is commonly used for describing a growth process, can be used for evaluating the reliability of a product and solving a testability growth curve, and has the following mathematical model:

where 0< u <1, 0< v <1, 0< w <1, i indicates that the equipment is in the ith test growth trial in the development phase.

By performing logarithmic transformation on equation (10), it is possible to obtain:

ln[p(i)]＝lnu+wⁱlnv (11)

suppose the equipment has performed m times of test growth tests in the development stage, wherein m is 3z, and z is a positive integer. Since the data of the test growth test is of a success-fail type, the data of the first test growth test can be recorded as (n)_i,c_i) Where i is 1,2, K, m, the estimated testability index point for the ith testability growth test is:

the test growth test data of m times was substituted into the formula (11), and it was found that:

the value u of the Gompertz parameter can be solved by the formula (13)^*、v^*、w^*Then substituting Gompertz model yields Gompertz's formula for the test growth experiment as:

the testability index point estimate for the final testability growth test obtained from the mth growth test data is:

the prior distribution of the testability growth test data may be represented by pi_I(p) and the entropy function of the prior distribution, as determined by the test growth test, is known to be:

therefore, the optimal solution of the a priori distribution parameters a, b determined by the testability growth test can be converted into solving the following programming problem:

step 2.3, for 3) replaceable unit test data:

in the equipment development stage, the difficulty of carrying out a system-level testability test is higher, and the collected testability test information is less. Replaceable units (including SRUs, LRUs, etc.) that are common in equipment are less difficult to perform testability tests and collect more data.

In engineering practice, a testability test with replaceable units as test objects follows a binomial distribution, which can be a Beta distribution. If a system has m replaceable units, let the testability index of the i (i ═ 1,2, K, m) th replaceable unit be p_iThen it is a priori distributed pi_R(p) can be represented as:

using a small amount of testability test data (n) collected from the replaceable unit during development_i,f_i) Wherein n is_iIs the ith oneNumber of fault samples injected by replacement unit, f_iFor its corresponding failure detection/isolation failure times, and substituting the data into Bayes' formula, p can be obtained_iPosterior distribution Beta (p)_i；a_i+n_i-f_i,b_i+f_i) Then p is_iThe posterior expectation and variance of (a) are:

by introducing the failure rate of the replaceable unit, the testability index value of the unit level can be reduced to the system level, and lambda is set_iRepresenting the failure rate of the ith replaceable unit, the system level testability index p is:

in engineering practice, the testability index p is influenced by the structural model, reliability, maintainability and the like of the equipment. For example, in terms of reliability, values obtained by different methods need to be selected according to model relationships such as series-parallel connection, k/n (g), and the like. In the embodiment, the influence of the coupling relationship of each unit on the testability index of the system level is not considered, and the testability index of the system level is obtained by weighting the testability index of the unit level by adopting a weighting method based on the unit failure rate.

The expectation and variance of the testability index p at the system level thus obtained are:

and the joint type (19) and (21) can solve the replaceable unit testability prior distribution parameter values a and b before data fusion.

In summary, the prior distribution of testability information at the development stage is shown in table 1.

TABLE 1 Prior distribution of testability information at development stage

Step 3, constructing an original mass function corresponding to the testability data in the development stage;

the method for evaluating the testability index of the equipment in the development stage is established on a D-S evidence theory, so that an identification frame of a testability index evaluation system can be regarded as consisting of three parts, namely expert experience information, testability growth test data and replaceable unit testability test data, and each part can be divided into 3 focal elements. Assuming that the target value of the equipment testability index given by the receiving party is P and the minimum acceptable value specified by the using party is P₁When the testability index P is reached>P₀Is the first focal element and is defined as H₁(ii) a When P is present₀>P₁Is the second focal unit and is defined as H₂(ii) a When P is present<P₁Is the third focal element and is defined as H₃。

For expert experience information (a)_E,b_E) Can be regarded as an evidence on the identification frame theta, and establishes a basic trust distribution function m of the evidence on the identification frame theta_E：

For test growth test data (a)_I,b_I) Establishing a basic trust distribution function m of the evidence on the identification frame theta_I：

Test data for replaceable units (a)_R,b_R) Establishing a basic trust distribution function m of the evidence on the identification frame theta_R：

The mass functions corresponding to the testability information at different development stages can be obtained by the equations (22) to (24), as shown in table 2:

TABLE 2 Mass function of testability information at development stage

Step 4, determining the reliability and uncertainty of testability data in the development stage by using a D-S evidence fusion method based on Lance distance and information entropy improvement, and constructing the final fusion weight of each data;

step 4.1, aiming at data credibility:

2) defining a system identification frame Θ ═ { a ═ a₁,A₂,L,A_M}, evidence E₁、E₂、K、E_nThe corresponding basic trust distribution function is m₁、m₂、K、m_nThen evidence E_i,E_jLancet distance d between_ijComprises the following steps:

wherein, i, j is 1,2, K, n, K is 1,2, K, M.

Distance d between evidences_ijCan reflect the difference between them, d_ijThe larger the value, the evidence E_iAnd E_jThe lower the confidence of (c). Therefore, the confidence level between the evidences can be used as the weight of the evidences, and the evidence E_iReliability Rel of_iIs defined as:

wherein:

step 4.2, aiming at data uncertainty:

setting A_i(i ═ 1,2, K, n) is a subset of the system recognition framework Θ, m (a)_i) For its corresponding belief function, | A_iI represents the subset A_iThe number of elements involved, then subset A_iThe confidence entropy of (d) is:

subset A_iThe more elements that are included, the greater the confidence entropy of the evidence, and the greater the uncertainty of the evidence. If the confidence entropy of an evidence body is smaller, the uncertainty of the evidence body is smaller, and the corresponding weight of the evidence body is larger in the final fusion process.

To avoid assigning zero weight to evidence in some cases, evidence weight is determined by computing an exponential form of confidence entropy:

after normalization, the weight determined according to the uncertainty of the evidence is as follows:

and 4.3, determining the final fusion weight based on the credibility and the uncertainty of the data:

determining fusion weight according to the consistency degree and the uncertainty degree of the evidence, and determining the fusion weight W of each evidence body_iComprises the following steps:

W_i＝Rel_i×Unc_i (30)

and (3) carrying out normalization processing on the fusion weight:

step 5, respectively endowing the mass functions of the testability information in the research stage with final fusion weights, and obtaining the corrected mass functions m_Avg(H_k) (ii) a Where k is 1,2,3, the solving process is shown in table 3:

TABLE 3 modified mass function m_Avg(H_k) Solution process of

Step 6, utilizing the corrected mass function m_Avg(H_k) Mass function m for replacing testability information in original development stage_E(H_k)、m_I(H_k)、m_R(H_k) (ii) a As shown in table 4:

TABLE 4 modified m_Avg(H_k) Alternative testability information mass function

Step 7, applying Dempster principle to synthesize the mass function values for N-1 times (N is the number of evidences, in this embodiment, N is 3), obtain the result of the 2 nd fusion,

following Dempster evidence combination rules;

the Dempster principle is applied, the mass function value is synthesized for 2 times, and the result of the 2 nd fusion is obtained

Following the Dempster evidence combination rule, the evidence fusion results are shown in Table 5:

TABLE 5 evidence fusion results

Step 8, according to the mass function value m 'of the final data fusion result'_EIR(H₁)、m'_EIR(H₂)、m'_EIR(H₃) And combining the equipment testability index values agreed by the bearing party and the using party to analyze the testability evaluation result.

In the embodiment, the reliability of the data is measured by calculating the Lance distance, the uncertainty of the data is measured by calculating the information entropy, the improved D-S evidence fusion method is designed to realize the testability index evaluation in the development stage of the equipment, the method is reliable and effective, and the equipment in the development stage can be accurately evaluated under the condition that the collected data has conflicts.

Furthermore, the invention also provides node equipment for the fog computing. As shown in fig. 4, each node device participating in the fog calculation at least includes a processor and a memory, and may further include a communication component, a sensor component, a power component, a multimedia component, and an input/output interface according to actual needs. The memory, the communication component, the sensor component, the power supply component, the multimedia component and the input/output interface are all connected with the processor. As mentioned above, the memory in the node device may be Static Random Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic memory, flash memory, etc., and the processor may be a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), a Digital Signal Processing (DSP) chip, etc. Other communication components, sensor components, power components, multimedia components, etc. may be implemented using common components found in existing smartphones and are not specifically described herein.

On the other hand, the equipment development stage testability evaluation system based on conflict data fusion comprises: a processor 42 and a memory 41, said processor 42 reading a computer program in said memory 41 for performing the following operations:

The equipment development stage testability evaluation system based on the conflict data fusion in this embodiment can implement all functions of the equipment development stage testability evaluation method based on the conflict data fusion in the above embodiments, and details are not repeated here.

It should be understood that the specific order or hierarchy of steps in the processes disclosed is an example of exemplary approaches. Based upon design preferences, it is understood that the specific order or hierarchy of steps in the processes may be rearranged without departing from the scope of the present disclosure. The accompanying method claims present elements of the various steps in a sample order, and are not intended to be limited to the specific order or hierarchy presented.

In the foregoing detailed description, various features are grouped together in a single embodiment for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the claimed embodiments of the subject matter require more features than are expressly recited in each claim. Rather, as the following claims reflect, invention lies in less than all features of a single disclosed embodiment. Thus, the following claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. To those skilled in the art; various modifications to these embodiments will be readily apparent, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

What has been described above includes examples of one or more embodiments. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the aforementioned embodiments, but one of ordinary skill in the art may recognize that many further combinations and permutations of various embodiments are possible. Accordingly, the embodiments described in this embodiment are intended to embrace all such alterations, modifications and variations that fall within the scope of the appended claims. Furthermore, to the extent that the term "includes" is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term "comprising" as "comprising" is interpreted when employed as a transitional word in a claim. Furthermore, any use of the term "or" in the specification of the claims is intended to mean a "non-exclusive or".

The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims

1. A device development phase testability evaluation method based on conflict data fusion is characterized by comprising the following steps:

2. The method of claim 1, wherein the development phase testability data comprises: expert experience information, testability growth test data, and replaceable unit test data.

3. The method for evaluating the testability of the equipment development stage based on the conflict data fusion as claimed in claim 1, wherein the constructing of the corresponding quality function specifically comprises:

4. The method for evaluating the testability of the equipment development stage based on the conflict data fusion as claimed in claim 1, wherein the final fusion weight is respectively assigned to the original quality function of the testability data of each development stage, and before the method further comprises:

5. The method for evaluating the testability of the equipment in the development stage based on the conflict data fusion as claimed in claim 2, wherein the step of respectively converting the testability data of the development stages into a priori distribution form specifically comprises the steps of:

6. The method for evaluating the testability of the equipment in the development stage based on the conflict data fusion as claimed in claim 2, wherein the step of respectively converting the testability data of the development stages into a priori distribution form specifically comprises the steps of:

7. The method for evaluating the testability of the equipment development stage based on the conflict data fusion as claimed in claim 2, specifically comprising:

8. The method for evaluating the testability of the equipment development stage based on the conflict data fusion as claimed in claim 1, wherein the step of synthesizing the quality function values of the testability data of each development stage specifically comprises:

9. The method of claim 4, wherein the step of constructing the final fusion weight based on the credibility and uncertainty of the testability data of each development phase further comprises:

10. An equipment development phase testability evaluation system based on conflict data fusion, comprising: a processor and a memory, the processor reading a computer program in the memory for performing the following operations: