CN116069646B - Test set determination method and system for multi-objective testability optimization - Google Patents
Test set determination method and system for multi-objective testability optimization Download PDFInfo
- Publication number
- CN116069646B CN116069646B CN202310096454.2A CN202310096454A CN116069646B CN 116069646 B CN116069646 B CN 116069646B CN 202310096454 A CN202310096454 A CN 202310096454A CN 116069646 B CN116069646 B CN 116069646B
- Authority
- CN
- China
- Prior art keywords
- test
- matrix
- determining
- fault
- row
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention aims to provide a test set determining method and a system for multi-objective testability optimization, which are used in the technical field of equipment or method for digital computing or data processing of specific applications, and are used for determining an optimized test set according to the test set and a fault-test correlation matrix so as to determine test resource types and test attribute types; constructing a three-dimensional matrix according to the test, the test resource type and the test attribute type in the optimized test set; respectively constructing matrix optimization functions corresponding to each test attribute to obtain a matrix optimization function set; constructing a decision matrix; solving a matrix optimization function group according to the three-dimensional matrix, and determining an optimal decision matrix; and carrying out testability optimization according to the optimal decision matrix. By using various test attributes as optimization targets to determine the test set, the rationality of testability optimization is improved. And the optimal decision matrix is determined by taking various test attributes as optimization targets for testability optimization, so that the rationality of testability optimization is improved.
Description
Technical Field
The invention relates to the technical field of equipment or methods specially applied to digital computing or data processing of specific applications, in particular to a test set determination method and a test set determination system for multi-objective testability optimization.
Background
The design for testability (Design for Testability, DFT) refers to a design method that comprehensively considers all test resources in the product design process, and under different test countermeasures (such as built-in test (BIT including software BIT and hardware BIT), automatic Test Equipment (ATE), manual Test (MTE), etc.), enables the product to be tested sufficiently through careful planning with minimum cost, and ensures that the test result has higher confidence. One of the important tasks of the design for testing is the problem of determining the test set.
Currently, the problem of determining the test set is considered under different test strategies, and usually, after the built-in test design is completed, the design of ATE and MTE is started at the shaping stage of the product. Even the built-in test and the external test equipment are separately designed for two departments, are independent of each other, lack of synthesis, are difficult to form scientific and reasonable grading test, and severely restrict the design level of the product. In addition, the cost is often considered when the testability optimization design is carried out, other key attributes (such as reliability, volume and the like) influencing the selection of the test set are not comprehensively considered, the optimization method cannot be suitable for the selection of the test set under different test strategies, and the test effect of the product cannot be optimized.
Disclosure of Invention
The invention aims to provide a test set determining method and a test set determining system for multi-objective testability optimization, which take various test attributes as optimization targets to determine the test set, so that the rationality of testability optimization is improved.
In order to achieve the above object, the present invention provides the following solutions:
a test set determination method for multi-objective testability optimization, comprising:
acquiring a test set and a fault-test correlation matrix of equipment; the fault-test correlation matrix is used for describing the correlation between the fault mode of the equipment and the test;
according to the fault-test correlation matrix, optimizing the test set to determine an optimized test set;
determining the types of the test resources and the types of the test attributes according to the tests in the fault-test correlation matrix;
taking the test in the optimized test set as an x coordinate, the test resource type as a y coordinate and the test attribute type as a z coordinate to construct a three-dimensional matrix;
respectively constructing matrix optimization functions corresponding to each test attribute to obtain a matrix optimization function set;
constructing a decision matrix; the decision matrix is used for describing the corresponding relation between the test and the test resource;
solving the matrix optimization function group according to the three-dimensional matrix, and determining an optimal decision matrix;
and performing testability optimization according to the optimal decision matrix.
Optionally, the test resource types include software built-in test, hardware built-in test, automatic test and manual test; the test attribute categories include cost, reliability, and volume.
Optionally, the optimizing processing is performed on the test set according to the test set and the fault-test correlation matrix, and the determining of the optimized test set specifically includes:
respectively determining a row vector array and a column vector array of the fault-test correlation matrix;
determining any proper subset of the array of column vectors as a set of pending optimized column vectors;
determining an identification vector of the pending optimized column vector group; the identification vector is a 0-1 matrix; when the j-th element in the identification vector is 1, the j-th element in the column vector group comprises the j-th column vector in the column vector array; when the j-th element in the identification vector is 0, the j-th element in the column vector array is not included in the column vector group;
determining the fault detection rate of the to-be-determined optimized column vector group according to the row vector array and the identification vector;
determining the fault isolation rate of the to-be-determined optimized column vector group according to the row vector array and the identification vector;
constructing a first test optimization function and a second test optimization function;
solving a first test optimization function and a second test optimization function according to the fault detection rate, the fault isolation rate and constraint conditions, and determining an optimal pending optimization column vector group;
and determining an optimization test set according to the optimal pending optimization column vector group.
Optionally, the determining, according to the row vector array and the identification vector, a failure detection rate of the pending optimized column vector group includes:
determining row vectors meeting detection conditions of the undetermined optimized column vector group in the row vector array as optimized row vectors; the detection condition of the undetermined optimized column vector group is that the sum of products of the row vectors and corresponding elements of the identification vectors is larger than 1;
obtaining failure rates of corresponding failures of all row vectors in the row vector array;
determining the sum of the fault rates of the faults corresponding to all the optimized row vectors as a first correlation quantity;
determining the sum of fault rates of all the row vectors in the row vector array corresponding to faults as a second correlation quantity;
and determining the ratio of the first correlation quantity to the second correlation quantity as a fault detection rate.
Optionally, the determining, according to the row vector array and the identification vector, a fault isolation rate of the to-be-determined optimized column vector group includes:
determining any row vector in the row vector array as a current row vector;
multiplying the current line vector by the corresponding element of the identification vector to obtain a current line identification matrix;
traversing all row vectors in the row vector array to obtain a plurality of row identification matrixes;
determining any row of identification matrix as a current first row of identification matrix;
determining any row identification matrix except the current first row identification matrix as a current second row identification matrix;
determining a norm value after exclusive OR operation is carried out on the current first row identification matrix and the current second row identification matrix;
updating the current second row identification matrix and returning to the step of determining the norm value after the exclusive or operation of the current first row identification matrix and the current second row identification matrix until all row identification matrices except the current first row identification matrix are traversed, so as to obtain a plurality of norm values corresponding to the current first row identification matrix as current Fan Shuzu;
performing continuous product operation on elements in the current range group to obtain isolation parameters of faults corresponding to the first row identification matrix;
updating a current first row identification matrix, and returning to the step of determining any row identification matrix except the current first row identification matrix as a current second row identification matrix until all row identification matrices are traversed to obtain isolation parameters of faults corresponding to each row identification matrix;
determining that the corresponding fault of the row identification matrix with the isolation parameter larger than 1 is an isolation fault;
determining the sum of the fault rates of all isolation faults as a third correlation quantity;
and determining the ratio of the third correlation quantity to the second correlation quantity as a fault isolation rate.
Alternatively to this, the method may comprise,
the first test optimization function is:
the second test optimization function is:
the constraint conditions are as follows:
wherein h (t) j ) In order to be a heuristic function,P(t j ) Coverage for the jth test in the test set; p (t) j )＝min{＜f i ＞，f i ∈F}；＜f i > is the i-th fault f that can be detected i Is a test number of (a); t j The I is the number of faults which can be detected by the j-th test; xj is the identifier corresponding to the j-th test, 1 represents selected, and 0 represents unselected; gamma ray FD And gamma FI The fault detection rate and the fault isolation rate are respectively; />And->The lower limit of the fault detection rate and the lower limit of the fault isolation rate are respectively defined.
Optionally, the determining an optimization test set according to the optimal pending optimization column vector group includes:
converting the optimal undetermined optimization column vector into a matrix form to obtain an optimal fault-test related undetermined matrix;
deleting the element sum of the row vector and the column vector which are 0 in the optimal fault-test related undetermined matrix to obtain an optimal fault-test related matrix;
determining a test corresponding to each column vector in the optimal fault-test correlation matrix to obtain an optimal test set;
determining faults corresponding to each row vector in the optimal fault-test correlation matrix to obtain an optimal fault set; the optimized test set comprises an optimal test set and an optimal fault set.
Optionally, the matrix optimization function set is:
wherein H is a decision matrix; determination matrix h= [ H ] ij ] n×q The method comprises the steps of carrying out a first treatment on the surface of the When h ij When=1, the ith test is performed using the jth test resource; when h ij When=0, it means that the ith test is not performed with the jth test resource; b. i represents a two-dimensional plane vector of the three-dimensional matrix when the z coordinate takes i; i=1, 2, 3.
A multi-objective, testability-optimized test set determination system, comprising:
the data acquisition module is used for acquiring a test set and a fault-test correlation matrix of the equipment; the fault-test correlation matrix is used for describing the correlation between the fault mode of the equipment and the test;
the optimization test set determining module is used for carrying out optimization processing on the test set according to the fault-test correlation matrix to determine an optimization test set;
the type determining module is used for determining the types of the test resources and the types of the test attributes according to the tests in the fault-test correlation matrix;
the three-dimensional matrix construction module is used for constructing a three-dimensional matrix by taking the test in the optimized test set as an x coordinate, the test resource type as a y coordinate and the test attribute type as a z coordinate;
the matrix optimization function group determining module is used for respectively constructing matrix optimization functions corresponding to each test attribute to obtain a matrix optimization function group;
the decision matrix construction module is used for constructing a decision matrix; the decision matrix is used for describing the corresponding relation between the test and the test resource;
the optimal decision matrix determining module is used for solving the matrix optimization function set according to the three-dimensional matrix to determine an optimal decision matrix;
and the testability optimization module is used for performing testability optimization according to the optimal decision matrix.
Optionally, the test resource types include software built-in test, hardware built-in test, automatic test and manual test; the test attribute categories include cost, reliability, and volume.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention aims to provide a test set determining method and a system for multi-objective testability optimization, which are used for optimizing a test set according to the test set and a fault-test correlation matrix to determine an optimized test set; determining the types of the test resources and the types of the test attributes according to the tests in the fault-test correlation matrix; taking the test in the optimized test set as an x coordinate, the test resource type as a y coordinate and the test attribute type as a z coordinate to construct a three-dimensional matrix; respectively constructing matrix optimization functions corresponding to each test attribute to obtain a matrix optimization function set; constructing a decision matrix; solving a matrix optimization function group according to the three-dimensional matrix, and determining an optimal decision matrix; and carrying out testability optimization according to the optimal decision matrix. By using various test attributes as optimization targets to determine the test set, the rationality of testability optimization is improved. And a plurality of test attributes are used as optimization targets to determine a test set (decision matrix), so that the rationality of testability optimization is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a test set determination method for multi-objective testability optimization in one embodiment of the invention;
FIG. 2 is a flow chart of a test set determination method for multi-objective testability optimization in accordance with the second embodiment of the present invention;
FIG. 3 is a schematic representation of a three-dimensional matrix in accordance with an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a test set determining method and a test set determining system for multi-objective testability optimization, which take various test attributes as optimization targets to determine the test set, so that the rationality of testability optimization is improved.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
As shown in fig. 1, the present embodiment provides a test set determining method for multi-objective testability optimization, including:
step 101: acquiring a test set and a fault-test correlation matrix of equipment; the failure-test correlation matrix is used to describe the correlation between the failure mode of the device and the test.
Step 102: and carrying out optimization processing on the test set according to the fault-test correlation matrix, and determining an optimized test set.
Step 103: determining the types of the test resources and the types of the test attributes according to the tests in the fault-test correlation matrix; the test resource types comprise software built-in tests, hardware built-in tests, automatic tests and manual tests; the test attribute categories include cost, reliability, and bulk.
Step 104: and constructing a three-dimensional matrix by taking the test in the optimized test set as an x coordinate, the test resource type as a y coordinate and the test attribute type as a z coordinate.
Step 105: and respectively constructing matrix optimization functions corresponding to each test attribute to obtain a matrix optimization function set.
Step 106: constructing a decision matrix; the decision matrix is used for describing the corresponding relation between the test and the test resource.
Step 107: and solving the matrix optimization function group according to the three-dimensional matrix, and determining an optimal decision matrix.
Step 108: and carrying out testability optimization according to the optimal decision matrix.
Step 102, specifically includes:
a row vector array and a column vector array of the fail-test correlation matrix are determined, respectively.
Any proper subset of the array of column vectors is determined to be the set of pending optimized column vectors.
Determining an identification vector of the pending optimized column vector set; the identification vector is a 0-1 matrix; when the j-th element in the identification vector is 1, the j-th element in the identification vector represents that the j-th column vector in the column vector array is included in the column vector group; the j-th element in the identification vector being 0 indicates that the j-th column vector in the column vector array is not included in the column vector group.
And determining the fault detection rate of the undetermined optimized column vector group according to the row vector array and the identification vector.
And determining the fault isolation rate of the undetermined optimized column vector group according to the row vector array and the identification vector.
And constructing a first test optimization function and a second test optimization function.
And solving the first test optimization function and the second test optimization function according to the fault detection rate, the fault isolation rate and constraint conditions, and determining the optimal pending optimization column vector group.
And determining an optimization test set according to the optimal pending optimization column vector group.
Specifically, determining the fault detection rate of the to-be-determined optimized column vector group according to the row vector array and the identification vector includes:
determining row vectors meeting detection conditions of a to-be-determined optimized column vector group in a row vector array as optimized row vectors; the detection condition of the undetermined optimized column vector group is that the sum of products of the row vector and corresponding elements of the identification vector is larger than 1.
And obtaining the fault rate of the corresponding faults of all the row vectors in the row vector array.
And determining the sum of the fault rates of the faults corresponding to all the optimized row vectors as a first correlation quantity.
And determining the sum of the fault rates of all the row vectors in the row vector array corresponding to the faults as a second correlation quantity.
And determining the ratio of the first correlation quantity to the second correlation quantity as a fault detection rate.
The method for determining the fault isolation rate of the undetermined optimized column vector group according to the row vector array and the identification vector comprises the following steps:
any one of the row vectors in the row vector array is determined to be the current row vector.
Multiplying the current line vector by the corresponding element of the identification vector to obtain the current line identification matrix.
Traversing all row vectors in the row vector array to obtain a plurality of row identification matrixes.
And determining any row of identification matrix as the current first row of identification matrix.
Any row identification matrix other than the current first row identification matrix is determined to be the current second row identification matrix.
And determining a norm value after the exclusive OR operation of the current first row identification matrix and the current second row identification matrix.
Updating the current second row identification matrix and returning to the step of determining the norm value after the exclusive or operation of the current first row identification matrix and the current second row identification matrix until all row identification matrices except the current first row identification matrix are traversed, so as to obtain a plurality of norm values corresponding to the current first row identification matrix as current Fan Shuzu.
And performing continuous product operation on the elements in the current Fan Shuzu to obtain isolation parameters of the faults corresponding to the first row identification matrix.
Updating the current first row identification matrix, and returning to the step of determining any row identification matrix except the current first row identification matrix as the current second row identification matrix until all row identification matrices are traversed, so as to obtain isolation parameters of faults corresponding to each row identification matrix.
And determining that the corresponding faults of the row identification matrix with the isolation parameters larger than 1 are isolation faults.
And determining the sum of the fault rates of all isolation faults as a third correlation quantity.
And determining the ratio of the third correlation quantity to the second correlation quantity as a fault isolation rate.
Wherein, the liquid crystal display device comprises a liquid crystal display device,
the first test point optimization function is as follows:
the second test optimization function is:
the constraint conditions are as follows:
wherein h (t) j ) In order to be a heuristic function,P(t j ) Coverage for the jth test in the test set; p (t) j )＝min{＜f i ＞，f i ∈F}；＜f i > is the i-th fault f that can be detected i Is a test number of (a); t j The I is the number of faults which can be detected by the j-th test; x is x j For the j-th test as a corresponding identifier, 1 indicates selected, and 0 indicates unselected; gamma ray FD And gamma FI The fault detection rate and the fault isolation rate are respectively; />And->The lower limit of the fault detection rate and the lower limit of the fault isolation rate are respectively defined.
Specifically, determining an optimization test set according to the optimal pending optimization column vector set includes:
and converting the optimal undetermined optimization column vector into a matrix form to obtain an optimal fault-test related undetermined matrix.
And deleting the element sum of the row vector and the column vector which are 0 in the optimal fault-test correlation undetermined matrix to obtain the optimal fault-test correlation matrix.
And determining a test corresponding to each column vector in the optimal fault-test correlation matrix to obtain an optimal test set.
Determining faults corresponding to each row vector in the optimal fault-test correlation matrix to obtain an optimal fault set; the optimized test set includes an optimized test set and an optimized fault set.
In addition, the matrix optimization function set is:
wherein H is a decision matrix; determination matrix h= [ H ] ij ] n×q The method comprises the steps of carrying out a first treatment on the surface of the When h ij When=1, the ith test is performed using the jth test resource; when h ij When=0, it means that the ith test is not performed with the jth test resource; b. i represents a two-dimensional plane vector of the three-dimensional matrix when the z coordinate takes i; i=1, 2, 3.
Example two
As in fig. 2, step1: and establishing a multi-signal flow diagram model through a system structure diagram.
The composition of the multi-signal flow graph model is that the modules Are connected by directional edges (Directed edges) and various fault modes of the composition units Are set in the corresponding modules, the propagation of faults is represented by signal flow directions, and the fault propagation relationship among the modules and the logic relationship of faults and test resources Are uniformly represented in a Directed graph. In practice, there is typically more than one failure mode in a constituent unit, and each failure mode is referred to as a multiple signal flow graph model because it affects a different functional signal. Constituent elements in a multiple signal flow graph model are generally represented by the following elements:
(1) A finite set of modules m= { M 1 ,m 2 ,…m l The modules here may be subsystems (Sub systems), line replaceable units (Line Replaceable Unit, LRU), shop replaceable units Shop Replaceable Unit, SRU), components or failure modes etc.;
(2) Limited test set t= { T 1 ,t 2 ,…t n And refers to all available test sets of the system.
(3) A finite set of test points p= { P 1 ,p 2 ,…p x And a test point, wherein at least one test is contained in the test point.
(4) A limited set of signals s= { S that can describe the function 1 ,s 2 ,…s r }。
(5) Each test point p i A set of test sets t (p j )，t(p j )∈T。
(6) Each test t j A set of detected signal sets st (t j )，st(t j )∈S。
(7) Each module m k A set of affected signal sets sm (m k )，sm(m k )∈S。
(8) Directed graph set dg= { M, P, E }, where E is a set of directed edges connecting system modules, representing the physical connection relationship and the functional correlation relationship of constituent units.
This step can result in all failure modes and all tested signal flow graph models. Generally, in order to facilitate the subsequent calculation, the multi-signal flow graph model obtained by simplifying the steps mainly comprises two aspects of contents:
1. the limited module set is used for simplifying and calculating that the modules are all in fault modes, so that the module set is the fault mode set: f= { F 1 ,F 2 ,…,F m }。
2. Limited test set t= { T 1 ,T 2 ,…T n And refers to all available test sets of the system.
Step2: and performing testability analysis on the multi-signal flow graph to generate a D matrix.
1. The method comprises the following steps: under ideal conditions, the fault mode and the available test have a certain correlation relationship without considering the uncertainty of fault propagation, false alarm, missing detection and other factors. If malfunction f m And test t n Associated, then: failure mode f m The occurrence of (c) will result in test t n The detection result is that the test does not pass; if test t n If the detection result is passing, the fault mode f can be determined m No occurrence occurs. Whether the detection result of the test passes or not and whether a relation between the two occurs or not to the failure mode which can be detected by the test can be determined, can be deduced from each other, and is called a correlation. By mathematical description of the graphic model, all fault modes F= { F in the graphic model are represented in the form of a Boolean matrix 1 ，f 2 ，…f m T= { T } and available test 1 ，t 2 ，…t n Correlation between. The mathematical model of a multiple signal flow graph is generally described by a fault-test correlation matrix (Dependency Matrix), also known as the D matrix, noted:
wherein: d matrix ith row:
f i ＝[d i1 d i2 … d in ]。
the j th column of the D matrix of detection information representing the i th failure mode that can be reacted by each test:
t j ＝[d 1j d 2j … d mj ] T 。
information representing each failure mode measurable by the jth test.
Ideally, the system's fault-test correlation matrix is a binary matrix, i.e., element d in the matrix ij Only 0 or 1, and the specific meaning is shown in the following formula:
2. outputting a result: according to the multi-signal flow diagram shown in the step1, all fault modes and all tested D matrixes can be obtained.
Step3: the testability is expected.
Let D matrix of the system under test be d= (D) ij ) p×q The system has p faults and q tests, and the fault set is F= { F 1 ，f 2 ，…，f p The failure rate of each failure is λ= { λ } 1 ，…，λ p Test set t= { T } 1 ，t 2 ，…t q }。x＝[x 1 ，x 2 ，…x q ]For the test set to be solved, if test t j Is selected, x j =1, otherwise x j ＝0。
1. Fault detection rate prediction model
Under the assumption of this, fault f i The conditions detected by the test set x are:
set F D A set of faults that are x detectable, namely:
the failure detection rate can be expressed as;
when considering failure rate data, it can be further rewritten as:
wherein lambda is i Is the failure rate of the ith failure.
2. Fault isolation rate prediction model
Suppose F ix ＝F i .*x＝(d ij x j ) 1×q Wherein F i And the row i of the D matrix is multiplied by corresponding elements of the two same-dimensional matrices, and the result is still the same-dimensional matrix. Failure f i And fault f j The conditions that can be isolated in test x are:
wherein the symbols areRepresenting an exclusive-or operation, I.I 1 Representing the vector 1 norm.
Set F I A set of x isolatable faults, namely:
the fault isolation rate can be expressed as:
considering the failure rate, it can be further rewritten as:
step4: testing for diagnosability assessment.
The test detection degree is test t j The number of faults that can be detected is recorded as |t j |。
The fault can be measured as any fault f in the fault set i The number of tests that can be detected is recorded as < f i ＞。
Test coverage t j Is one test in the test set T, let P (T j )＝min{＜f i ＞，f i E F }, called P (t) j ) At t j Is a coverage of (a).
Test t according to the definition above j Can use a heuristic function h (t j ) Description is made:
step5: and (5) establishing a test optimization model.
The test optimization is under the premise of meeting the fault detection rate and the fault isolation law. The set of tests that maximizes diagnosable performance of the equipment as a whole is selected while minimizing the number of tests. The optimized mathematical model is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,and->To meet the system requirementThe lower limit of the fault detection rate and the lower limit of the fault isolation rate are generally given according to actual needs.
Step6: and selecting an optimization model by the test resource.
Through step5, an optimized test set t= { T can be obtained 1 ，t 2 ，…，t n Test countermeasure set e= { E } 1 ，e 2 ，…，e q In this step, the test strategy, i.e., the form of the test, contains four types, i.e., e= { software BIT, hardware BIT, ATE, MTE }. Defining the Cartesian product of set T and set S: s=t×e= { S ij }＝{＜t i ，e j ＞|t i ∈T，e j E, to represent the consumed specific test resources of each test, and for each specific test occupying the specific test resources, define its set of attributes λ= { λ 1 ，λ 2 ，…，λ w In this step, w takes 3, and the tested attribute set is λ= { cost, reliability, volume }. Further definition ofTo test s ij K=1, 2, … …, w; y is then k Values under the kth attribute are evaluated for all tests.
By the expression, a three-dimensional matrix consisting of the test, the test strategy and the test attribute can be established, and is marked as B, and the specific meaning of the matrix is shown in figure 3. Definition of three-dimensional matrix consists of I x J x K real numbers a ijk The three-dimensional matrix of I row, J column and K longitudinal order is named as A I×J×K ＝[a ijk ] I×J×K 。
Plane vector in three-dimensional matrix: when A is I×J×K ＝[a ijk ] I×J×K Any one of the three variables i, j, k is taken as a constant, while the remaining two variables change from a minimum value to a maximum value, forming a flat dataform data table, called a flat vector (also called a two-dimensional matrix).
Assuming F, G as an m×n matrix, F, G can be determined at this pointMultiplying under the sense rule of multiplication, and the productIs an m x n matrix, denoted +.>Wherein h is ij ＝f ij ×g ij 。/>Representing a matrix multiplication operation.
Let the decision matrix h= [ H ] ij ] n×q Is a binary matrix, the matrix element is composed of 0 and 1, when h ij When=1, the ith test is performed using the jth test resource, otherwise, the test mode is not selected.
The final purpose of the test resource optimization is to make the whole manufacturing cost of the equipment lowest and the reliability of the equipment highest, and the established optimization model is as follows:
wherein H is i Row i of matrix D, b. i Representing the ith two-dimensional plane vector in the three-dimensional matrix B.
Step7: and (5) a crowd-target particle swarm optimization algorithm.
According to the models in the step5 and the step6, which belong to the multi-objective optimization problem, the step adopts a multi-objective particle swarm optimization method to solve the two models.
Pareto governance; in minimizing the multi-objective optimization problem, for n objective components f i (x) I=1, 2, …, n, any given two decision variables X a ，X b X is said to be true if the following two conditions are satisfied a Dominating X b 。
1) For the followingAll have f i (X a )≤f i (X b ) This is true.
2)So that f i (X a )＜f i (X b ) This is true.
A decision variable is said to be a non-dominant solution if, for that decision variable, there are no other decision variables that can dominate it.
The specific flow of the multi-target particle swarm optimization method is as follows:
step1: initialization of
The number M of particles in the population of particles and the number of iterations are determined. Initializing inertia weight omega and learning factor c 1 ，c 2 。
As described above, the test set to be solved x= [ x ] 1 ，x 2 ，…x q ]For a 1 xq matrix, the decision matrix h= [ H ] ij ] n×q Is an n x q matrix. Each particle in the particle group has a velocity matrix v i I=1, 2, …, M and one position matrix l i I=1, 2, …, M, the dimensions of the matrix are consistent with the problem x or H sought. Where the location matrix corresponds to what was sought in the original question, a variable ans is defined to hold the optimal answer.
Step2: calculating fitness
The position matrix l of each particle i Sequentially carrying out optimization targets of the problems to be optimized, taking the obtained result as an fitness vector of the particles, and measuring the quality degree of the positions of the particles according to the Pareto dominant relationship: for any two particles A and B, if the fitness vector of the particle A dominates the fitness vector of the particle B, the position of A is better than the position of B, and if the fitness vector of the particle A and the fitness vector of the particle B do not dominate each other, the position of A and the position of B cannot be judged whether the particle is good or bad.
ans is used for recording the optimal position and the position which cannot be judged, and setting a global optimal position G from ansExtracting a group of optimal positions to store in G, setting a particle optimal position P for each particle i The optimal position of each particle is recorded. ans the position of all the particles in the particle group is stored.
Step3: updating the velocity of particles
Velocity vector v for each particle i Updating, wherein the particle speed updating formula is as follows:
in the middle ofRepresenting the d-th dimensional velocity component of the ith particle in the t-th iteration, +.>Meaning similar thereto, ω being inertial weight, c 1 ，c 2 As a learning factor, rand () is a random number.
Step4: updating particle position
Updating the position vector of the particle according to the new velocity vector of the particle, wherein the updating formula is as follows:
in the method, in the process of the invention,representing the d-th dimensional position component of the i-th particle in the t-th iteration.
Step5: updating ans, G, P i
Position matrix l of new particle after updating i And sequentially carrying the new fitness vectors into an optimization target of the problem to be optimized, and obtaining the new fitness vectors.
1. Judging the quality degree of the particle position according to the Pareto dominant relationship, adding the position of the optimal position of the particle or the position which cannot be judged to be good or bad to an8, wherein ans comprises the position of the old particle and the position of the new particle, performing repeated item rejection on ans, and obtaining a new ans by using the dominant solution according to definition 6.
2. And randomly extracting a position vector corresponding to the non-dominant fitness vector from the new fitness to update G.
3. Comparing the new fitness of each particle with the old fitness, if the new fitness dominates the fitness, then P i For the position matrix corresponding to the new fitness, if the new fitness is dominated by the old fitness, P i Unchanged, if the two are not mutually dominant, randomly determining P i Whether or not to change.
Step6: repeating the iteration
Steps 3 to 5 are repeated until a prescribed number of iterations is reached.
Step7: conclusion is drawn
After the repeated iteration is completed, ans is the optimal solution.
Example III
The embodiment provides a test set determining system for multi-objective testability optimization, which comprises the following steps:
the data acquisition module is used for acquiring a test set and a fault-test correlation matrix of the equipment; the failure-test correlation matrix is used to describe the correlation between the failure mode of the device and the test.
And the optimization test set determining module is used for carrying out optimization processing on the test set according to the test set and the fault-test correlation matrix to determine the optimization test set.
And the category determining module is used for determining the category of the test resource and the category of the test attribute according to the test in the fault-test correlation matrix.
The three-dimensional matrix construction module is used for constructing a three-dimensional matrix by taking the test in the optimized test set as an x coordinate, the test resource type as a y coordinate and the test attribute type as a z coordinate.
And the matrix optimization function group determining module is used for respectively constructing matrix optimization functions corresponding to each test attribute to obtain a matrix optimization function group.
The decision matrix construction module is used for constructing a decision matrix; the decision matrix is used for describing the corresponding relation between the test and the test resource.
And the optimal decision matrix determining module is used for determining an optimal decision matrix according to the three-dimensional matrix solving matrix optimizing function set.
And the testability optimization module is used for performing testability optimization according to the optimal decision matrix.
The test resource types comprise software built-in tests, hardware built-in tests, automatic tests and manual tests; the test attribute categories include cost, reliability, and bulk.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The principles and embodiments of the present invention have been described in this specification with reference to specific examples, the description of which is only for the purpose of aiding in understanding the method of the present invention and its core ideas; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In summary, the present description should not be construed as limiting the invention.
Claims (8)
1. A test set determination method for multi-objective testability optimization, comprising:
acquiring a test set and a fault-test correlation matrix of equipment; the fault-test correlation matrix is used for describing the correlation between the fault mode of the equipment and the test;
according to the fault-test correlation matrix, optimizing the test set to determine an optimized test set;
determining the types of the test resources and the types of the test attributes according to the tests in the fault-test correlation matrix;
taking the test in the optimized test set as an x coordinate, the test resource type as a y coordinate and the test attribute type as a z coordinate to construct a three-dimensional matrix;
respectively constructing matrix optimization functions corresponding to each test attribute to obtain a matrix optimization function set;
constructing a decision matrix; the decision matrix is used for describing the corresponding relation between the test and the test resource;
solving the matrix optimization function group according to the three-dimensional matrix, and determining an optimal decision matrix;
performing testability optimization according to the optimal decision matrix;
according to the test set and the fault-test correlation matrix, optimizing the test set to determine an optimized test set, which comprises the following steps:
respectively determining a row vector array and a column vector array of the fault-test correlation matrix;
determining any proper subset of the array of column vectors as a set of pending optimized column vectors;
determining an identification vector of the pending optimized column vector group; the identification vector is a 0-1 matrix; when the j-th element in the identification vector is 1, the j-th element in the column vector group comprises the j-th column vector in the column vector array; when the j-th element in the identification vector is 0, the j-th element in the column vector array is not included in the column vector group;
determining the fault detection rate of the to-be-determined optimized column vector group according to the row vector array and the identification vector;
determining the fault isolation rate of the to-be-determined optimized column vector group according to the row vector array and the identification vector;
constructing a first test optimization function and a second test optimization function;
solving a first test optimization function and a second test optimization function according to the fault detection rate, the fault isolation rate and constraint conditions, and determining an optimal pending optimization column vector group;
determining an optimization test set according to the optimal pending optimization column vector group;
the matrix optimization function set is as follows:
wherein H is a decision matrix; determination matrix h= [ H ] ij ] n×q The method comprises the steps of carrying out a first treatment on the surface of the When h ij When=1, the ith test is performed using the jth test resource; when h ij When=0, it means that the ith test is not performed with the jth test resource; and b. i Representing a two-dimensional plane vector of the three-dimensional matrix when the z coordinate takes i; i=1, 2, 3.
2. The method for testing set determination of multi-objective testability optimization according to claim 1, wherein the test resource categories include software built-in tests, hardware built-in tests, automatic tests and manual tests; the test attribute categories include cost, reliability, and volume.
3. The method of claim 1, wherein determining a failure detection rate for a set of pending optimized column vectors based on the array of row vectors and the identification vector comprises:
determining row vectors meeting detection conditions of the undetermined optimized column vector group in the row vector array as optimized row vectors; the detection condition of the undetermined optimized column vector group is that the sum of products of the row vectors and corresponding elements of the identification vectors is larger than 1;
obtaining failure rates of corresponding failures of all row vectors in the row vector array;
determining the sum of the fault rates of the faults corresponding to all the optimized row vectors as a first correlation quantity;
determining the sum of fault rates of all the row vectors in the row vector array corresponding to faults as a second correlation quantity;
and determining the ratio of the first correlation quantity to the second correlation quantity as a fault detection rate.
4. The method of claim 3, wherein determining a fault isolation rate for a set of pending optimized column vectors based on the array of row vectors and the identification vector comprises:
determining any row vector in the row vector array as a current row vector;
multiplying the current line vector by the corresponding element of the identification vector to obtain a current line identification matrix;
traversing all row vectors in the row vector array to obtain a plurality of row identification matrixes;
determining any row of identification matrix as a current first row of identification matrix;
determining any row identification matrix except the current first row identification matrix as a current second row identification matrix;
determining a norm value after exclusive OR operation is carried out on the current first row identification matrix and the current second row identification matrix;
updating the current second row identification matrix and returning to the step of determining the norm value after the exclusive OR operation of the current first row identification matrix and the current second row identification matrix until the row identification matrixes except the current first row identification matrix in all row identification matrixes are traversed, so as to obtain a plurality of norm values corresponding to the current first row identification matrix as current Fan Shuzu;
performing continuous product operation on elements in the current range group to obtain isolation parameters of faults corresponding to the first row identification matrix;
updating a current first row identification matrix, and returning to the step of determining any row identification matrix except the current first row identification matrix as a current second row identification matrix until all row identification matrices are traversed to obtain isolation parameters of faults corresponding to each row identification matrix;
determining that the corresponding fault of the row identification matrix with the isolation parameter larger than 1 is an isolation fault;
determining the sum of the fault rates of all isolation faults as a third correlation quantity;
and determining the ratio of the third correlation quantity to the second correlation quantity as a fault isolation rate.
5. The method for testing set determination for multi-objective testability optimization according to claim 1,
the first test optimization function is:
the second test optimization function is:
the constraint conditions are as follows:
wherein h (t) j ) In order to be a heuristic function,P(t j ) Coverage for the jth test in the test set; p (t) j )＝min{<f i >，f i ∈F}；<f i >Is the ith fault f which can be detected i Is a test number of (a); t j The I is the number of faults which can be detected by the j-th test; x is x j For the identifier corresponding to the j-th test, 1 indicates selected, and 0 indicates unselected; gamma ray FD And gamma FI The fault detection rate and the fault isolation rate are respectively; />And->The lower limit of the fault detection rate and the lower limit of the fault isolation rate are respectively defined.
6. The method of claim 1, wherein determining an optimized test set from the optimal pending optimized column vector set comprises:
converting the optimal undetermined optimization column vector into a matrix form to obtain an optimal fault-test related undetermined matrix;
deleting the element sum of the row vector and the column vector which are 0 in the optimal fault-test related undetermined matrix to obtain an optimal fault-test related matrix;
determining a test corresponding to each column vector in the optimal fault-test correlation matrix to obtain an optimal test set;
determining faults corresponding to each row vector in the optimal fault-test correlation matrix to obtain an optimal fault set; the optimized test set comprises an optimal test set and an optimal fault set.
7. A multi-objective, testability-optimized test set determination system, comprising:
the data acquisition module is used for acquiring a test set and a fault-test correlation matrix of the equipment; the fault-test correlation matrix is used for describing the correlation between the fault mode of the equipment and the test;
the optimization test set determining module is used for carrying out optimization processing on the test set according to the fault-test correlation matrix to determine an optimization test set;
the type determining module is used for determining the types of the test resources and the types of the test attributes according to the tests in the fault-test correlation matrix;
the three-dimensional matrix construction module is used for constructing a three-dimensional matrix by taking the test in the optimized test set as an x coordinate, the test resource type as a y coordinate and the test attribute type as a z coordinate;
the matrix optimization function group determining module is used for respectively constructing matrix optimization functions corresponding to each test attribute to obtain a matrix optimization function group;
the decision matrix construction module is used for constructing a decision matrix; the decision matrix is used for describing the corresponding relation between the test and the test resource;
the optimal decision matrix determining module is used for solving the matrix optimization function set according to the three-dimensional matrix to determine an optimal decision matrix;
the testability optimization module is used for performing testability optimization according to the optimal decision matrix;
the optimization test set determining module is further used for determining a row vector array and a column vector array of the fault-test correlation matrix respectively; determining any proper subset of the array of column vectors as a set of pending optimized column vectors; determining an identification vector of the pending optimized column vector group; the identification vector is a 0-1 matrix; when the j-th element in the identification vector is 1, the j-th element in the column vector group comprises the j-th column vector in the column vector array; when the j-th element in the identification vector is 0, the j-th element in the column vector array is not included in the column vector group; determining the fault detection rate of the to-be-determined optimized column vector group according to the row vector array and the identification vector; determining the fault isolation rate of the to-be-determined optimized column vector group according to the row vector array and the identification vector; constructing a first test optimization function and a second test optimization function; solving a first test optimization function and a second test optimization function according to the fault detection rate, the fault isolation rate and constraint conditions, and determining an optimal pending optimization column vector group; determining an optimization test set according to the optimal pending optimization column vector group;
the matrix optimization function set is as follows:
wherein H is a decision matrix; determination matrix h= [ H ] ij ] n×q The method comprises the steps of carrying out a first treatment on the surface of the When h ij When=1, the ith test is performed using the jth test resource; when h ij When=0, it means that the ith test is not performed with the jth test resource; and b. i Representing a two-dimensional plane vector of the three-dimensional matrix when the z coordinate takes i; i=1, 2, 3.
8. The multi-objective, testability-optimized test set determination system according to claim 7, wherein the test resource categories comprise software built-in tests, hardware built-in tests, automatic tests and manual tests; the test attribute categories include cost, reliability, and volume.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310096454.2A CN116069646B (en) | 2023-02-10 | 2023-02-10 | Test set determination method and system for multi-objective testability optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310096454.2A CN116069646B (en) | 2023-02-10 | 2023-02-10 | Test set determination method and system for multi-objective testability optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116069646A CN116069646A (en) | 2023-05-05 |
| CN116069646B true CN116069646B (en) | 2023-08-11 |
Family
ID=86176589
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310096454.2A Active CN116069646B (en) | 2023-02-10 | 2023-02-10 | Test set determination method and system for multi-objective testability optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116069646B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPWO2022091228A1 (en) * | 2020-10-27 | 2022-05-05 | ||
| WO2022142009A1 (en) * | 2020-12-30 | 2022-07-07 | 平安科技（深圳）有限公司 | Blurred image correction method and apparatus, computer device, and storage medium |
| CN115165332A (en) * | 2022-06-29 | 2022-10-11 | 中国人民解放军海军航空大学 | Integrated design method and system for built-in test and comprehensive test of equipment |
| CN115184055A (en) * | 2022-06-28 | 2022-10-14 | 中国人民解放军海军航空大学 | Method and system for determining test set with optimized hierarchical testability |
-
2023
- 2023-02-10 CN CN202310096454.2A patent/CN116069646B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPWO2022091228A1 (en) * | 2020-10-27 | 2022-05-05 | ||
| WO2022142009A1 (en) * | 2020-12-30 | 2022-07-07 | 平安科技（深圳）有限公司 | Blurred image correction method and apparatus, computer device, and storage medium |
| CN115184055A (en) * | 2022-06-28 | 2022-10-14 | 中国人民解放军海军航空大学 | Method and system for determining test set with optimized hierarchical testability |
| CN115165332A (en) * | 2022-06-29 | 2022-10-11 | 中国人民解放军海军航空大学 | Integrated design method and system for built-in test and comprehensive test of equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116069646A (en) | 2023-05-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111130839B (en) | Flow demand matrix prediction method and system | |
| Bode | Neural networks for cost estimation: simulations and pilot application | |
| CN110009171B (en) | User behavior simulation method, device, equipment and computer readable storage medium | |
| Shahamiri et al. | Artificial neural networks as multi-networks automated test oracle | |
| CN107972070A (en) | Test method, test system and the computer-readable recording medium of robot performance | |
| CN106201871A (en) | Based on the Software Defects Predict Methods that cost-sensitive is semi-supervised | |
| CN111814342B (en) | Complex equipment reliability hybrid model and construction method thereof | |
| CN110007617B (en) | Uncertainty transmission analysis method of aircraft semi-physical simulation system | |
| CN113646714A (en) | Processing parameter setting method and device for production equipment and computer readable medium | |
| CN115184055B (en) | Test set determination method and system for hierarchical testability optimization | |
| CN116069646B (en) | Test set determination method and system for multi-objective testability optimization | |
| CN115165332B (en) | Method and system for integrally designing built-in test and comprehensive test of equipment | |
| Kleinke et al. | countimp 1.0–A multiple imputation package for incomplete count data | |
| KR102110316B1 (en) | Method and device for variational interference using neural network | |
| CN113484919B (en) | Magnetic anomaly inversion method, system, terminal and medium based on convolutional neural network | |
| CN113779116B (en) | Object ordering method, related equipment and medium | |
| CN114490412A (en) | Three-dimensional CAD software performance measurement method and device based on self-subtraction reverse cloud generator | |
| CN110045362B (en) | RNN-based radar target track identification method and device | |
| KR20130086083A (en) | Risk-profile generation device | |
| Yao et al. | Optimal selection of tests for fault diagnosis in multi-path system with time-delay | |
| CN114398291B (en) | Software regression testing influence domain analysis method and system | |
| CN114611059A (en) | Verification method and system of digital twin credibility evaluation method and storage medium | |
| Asghari Varzaneh et al. | Finding Reliable Identification Results for Nonlinear SDOF Systems, Based on Sensitivity Analyses. | |
| Chang | Dynamic robust design with missing data | |
| CN115423186A (en) | Cost prediction method, device, medium and equipment based on neural network model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |