CN114429082B - Circuit test point optimization method considering test uncertainty - Google Patents
Circuit test point optimization method considering test uncertainty Download PDFInfo
- Publication number
- CN114429082B CN114429082B CN202210021680.XA CN202210021680A CN114429082B CN 114429082 B CN114429082 B CN 114429082B CN 202210021680 A CN202210021680 A CN 202210021680A CN 114429082 B CN114429082 B CN 114429082B
- Authority
- CN
- China
- Prior art keywords
- particle
- iteration
- test
- value
- rate
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/28—Testing of electronic circuits, e.g. by signal tracer
- G01R31/2832—Specific tests of electronic circuits not provided for elsewhere
- G01R31/2836—Fault-finding or characterising
- G01R31/2846—Fault-finding or characterising using hard- or software simulation or using knowledge-based systems, e.g. expert systems, artificial intelligence or interactive algorithms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Geometry (AREA)
- Tests Of Electronic Circuits (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a circuit measuring point optimization method considering test uncertainty, which comprises the steps of constructing a multi-signal model of the test uncertainty, giving an optimization objective function and constraint conditions of isolation rate, detection rate, test cost, false alarm rate and missing detection rate under the model, then carrying out multi-objective optimization by using a multi-objective particle swarm algorithm and taking the test cost, the false alarm rate and the missing detection rate as optimization objectives and the detection rate and the isolation rate as constraint conditions, and finally obtaining a plurality of groups of optimized measuring point selection schemes.
Description
Technical Field
The invention belongs to the technical field of electronic system fault diagnosis, and particularly relates to a circuit test point optimization method considering test uncertainty.
Background
With the development of information technology, the complexity of an electronic system is higher and higher, and the complex test requirements are met by selecting an optimized test point, which is an important content in the field of current fault diagnosis. However, due to the fact that various external factors such as interference exist in an actual test environment, test data are not completely reliable, the actual test data have a lot of uncertainties, great difficulty is brought to testability design and analysis of a system, and how to optimally select a test point under the test uncertainty condition becomes one of the difficulties in the field of fault diagnosis.
Currently, many test optimization schemes for testing reliable conditions have been studied, but there is no in-depth study for testing point optimization considering uncertain conditions. Under the condition of meeting the isolation rate and the detection rate, how to optimize the measuring points as much as possible to reduce the testing cost, the false alarm rate and the missing detection rate so as to meet the testing and diagnosing requirements of an equipment system in the most reliable and economic mode is the problem to be solved by the method.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a circuit test point optimization method considering test uncertainty, and a plurality of groups of optimized test points are obtained under the condition of meeting the isolation rate and the detection rate.
In order to achieve the above object, the present invention provides a method for optimizing circuit test points considering test uncertainty, comprising the steps of:
(1) Constructing a multi-signal model H = { F, T, D, P, C, PF } with testing uncertainty;
wherein, F represents various fault sets of the circuit to be tested, and F = { F = { (F) 1 ,f 2 ,…,f i ,…,f m }，f i Representing the i-th fault, wherein m is the total number of faults; t represents all available measurement point sets of the circuit to be measured, T = { T = { (T) 1 ,t 2 ,…,t j ,…,t n }，t j Representing the jth available measuring point, wherein n is the total number of the available measuring points; p represents a prior probability set of a certain fault of the circuit to be tested, and P = { P = 1 ,p 2 ,…,p i ,…,p m }，p i Indicating the occurrence of a fault s i A priori probability of (a); c represents a test cost set corresponding to T, C = { C 1 ,c 2 ,…,c j ,…,c n }，c j Represents the test t j The cost of (d); d is a dependency matrix under the uncertainty of the measuring points, and is specifically expressed as follows:
wherein d is ij Indicates a fault f i At available measuring point t j Test information of d ij =0 or d ij (= 1) when d ij If =0, it indicates that a fault has occurred in the circuit under test f i Time can not pass through available measuring point t j Detecting; when d is ij If =1, it indicates that a fault f occurs in the circuit under test i Can pass through the available measuring point t j Detecting;
the PF is a false alarm rate matrix of the measuring point test, which is specifically expressed as:
wherein, pf ij Represents the measured point t j For a fault t j The test false alarm rate of (1);
(2) Carrying out measuring point optimization on the test uncertainty model by using a multi-target particle swarm algorithm based on crowded distance sorting;
(2.1) setting the position of the particle k in the multi-target particle swarm optimization as
Is taken to be 0 or 1, when>
Then, represent the measurement point t j Is selected when>
Then, represent the measurement point t j Is not selected, where j ∈ [1, n ]]；
(2.2) constructing an optimized objective function consisting of the test cost, the missed detection rate and the false alarm rate;
and (3) testing cost:
the omission rate is as follows:
false alarm rate:
(2.3) constructing a constraint condition consisting of the detection rate and the isolation rate;
detection rate:
isolation rate:
setting the lowest detection rate lambda 0 And isolation ratio lambda 1 Then the detection rate psi FD (x)≥λ 0 Isolation ratio psi FI (x)≥λ 1 ；
(2.4) carrying out measurement point optimization on the circuit to be measured based on the crowding distance sorting multi-target particle swarm algorithm;
(2.4.1) initializing a particle group P consisting of q particles, each having n dimensions and an initial velocity of 0, wherein the velocity of the kth particle is expressed as:
the initial position of the kth particle is expressed as
k∈[1,q]，j∈[1,n]；
Setting the maximum iteration time T of the multi-target particle swarm algorithm, wherein the current iteration time T belongs to [1, T ], and initializing T =1;
setting an external archive REP with the size of R, wherein R is more than q; initializing REP as an empty set;
setting an array of pbest of size qxn, pbestEach line is used for recording the corresponding position of the historical optimal value of each particle after each iteration, wherein the kth line after the t iteration is recorded as:
pbest stores the corresponding position of the historical optimal value of each particle;
setting an array gbest with the size of 1 multiplied by n, and recording the corresponding position of the global optimal value of the particle swarm after each iteration, wherein the global optimal value after the t iteration is recorded as
g is a particle which is optimally corresponding to the historical optimal value in the q particles;
(2.4.2) updating the speed and the position of each particle after the t iteration;
update the velocity of each particle:
V k (t)＝w×V k (t-1)+r 1 a 1 (P k (t-1)-X k (t-1))+a 2 r 2 (P g (t-1)-X k (t-1))
wherein w is an inertia factor, r 1 ，r 2 Is a random number varying in the range of (0, 1), a 1 ，a 2 Is an acceleration factor;
updating the position of each particle:
when X is k (t) in
Then, then->
Or else>
rand () generates a random number between (0, 1);
(2.4.3) calculating a fitness value and a constraint value after the t iteration;
substituting the position of each particle after the t iteration into objective functions C (X), L (X) and A (X) to obtain the fitness value of each particle, wherein the fitness value of the kth particle is recorded as
Respectively corresponding C (X), L (X) and A (X) to obtain calculated values;
substituting the position of each particle after the t-th iteration into the constraint condition psi FD (X)、ψ FI (X) obtaining a constraint value of each particle, wherein the constraint value of the kth particle is recorded as
Respectively corresponding to psi FD (X)、ψ FI (X) obtaining a calculated value;
(2.4.4) updating the historical optimal value position of each particle and the global optimal value position of the particle swarm after the t iteration;
and updating the historical optimal value position of each particle: the fitness value f of each particle after the t iteration k (t) comparing the historical best fitness value corresponding to the location stored in pbest if f k (t) if the historical optimal fitness value is greater than the historical optimal fitness value, replacing the corresponding historical optimal value position in the pbest, and otherwise, keeping the historical optimal value position unchanged;
updating the global optimal value position of the particle swarm: after the t iteration, the maximum fitness value is selected from the particles and is marked as f g (t) comparing with the global optimum corresponding to the location stored by gbest, if f g (t) if the global optimum value is greater than the global optimum value, replacing the global optimum value position in the gbest, otherwise, keeping the global optimum value position unchanged;
(2.4.5) calculating a crowding distance;
(2.4.5.1) extracting the fitness value of each particle after the t iteration
Then, carrying out descending order arrangement on the q groups of fitness values obtained under the same objective function;
(2.4.5.2) calculating the crowding distance dist of each particle after the t iteration k (t)；
Wherein the crowding distance obtained by the kth particle under the objective function C (X) is
And &>
The same can be obtained;
(2.4.6) archiving each particle;
extracting each particle stored in an external archive REP, and then arranging each particle after the t iteration in a descending order according to the size of the crowding distance; then, the memory size of the external archive REP is read, whether the remaining memory size of the external archive REP is larger than the total number q of the particles is judged, and if so, the sorted particles are stored in the external archive REP in sequence; otherwise, selecting the R particles with the largest crowding distance and storing the R particles in an external archive REP in sequence;
(2.4.7) outputting a preferred measuring point;
judging whether the current iteration time T reaches the maximum iteration time T, if T is less than T, adding 1 to the current iteration time T, returning to the step (2.4.2), and continuing the next iteration; otherwise, the loop is exited and R particles are output from the external archive REP, the position of each particle representing a preferred measuring point.
The invention aims to realize the following steps:
the invention relates to a circuit measuring point optimization method considering test uncertainty, which comprises the steps of constructing a multi-signal model of the test uncertainty, giving an optimization objective function and constraint conditions of isolation rate, detection rate, test cost, false alarm rate and missing detection rate under the model, then carrying out multi-objective optimization by using a multi-objective particle swarm algorithm and taking the test cost, the false alarm rate and the missing detection rate as optimization objectives and the detection rate and the isolation rate as constraint conditions, and finally obtaining a plurality of groups of optimized measuring point selection schemes.
Meanwhile, the circuit test point optimization method considering the test uncertainty further has the following beneficial effects:
(1) The optimization of the invention takes the test cost, the false alarm rate and the omission factor as the objective functions and takes the isolation rate and the detection rate as the constraint conditions, so that the conditions of high false alarm rate and omission factor can not occur in the optimized measuring points, and the invention is more suitable for practical application.
(2) According to the method, the test points under unreliable test conditions are optimized by using a binary multi-target particle swarm algorithm based on the crowding distance, and the crowding distance is adopted for archiving, so that the method is not easy to fall into local optimization, and a global optimal scheme can be found as much as possible;
(3) Compared with other optimization algorithms such as NSGA-II and simulated annealing particle swarm optimization, the method has the advantages that the optimization result is more excellent, and the optimal test scheme can be found more favorably.
Drawings
FIG. 1 is a flow chart of a preferred method of testing points for a circuit that accounts for test uncertainty in accordance with the present invention;
FIG. 2 is a flow chart of measuring point optimization of a circuit to be measured based on a multi-target particle swarm algorithm of crowded distance sorting;
FIG. 3 is a schematic representation of a preferred station.
Detailed Description
The following description of the embodiments of the present invention is provided in order to better understand the present invention for those skilled in the art with reference to the accompanying drawings. It is to be expressly noted that in the following description, a detailed description of known functions and designs will be omitted when it may obscure the main content of the present invention.
Examples
FIG. 1 is a flow chart of a preferred method of testing points for circuit in consideration of test uncertainty in accordance with the present invention.
In this embodiment, as shown in fig. 1, a preferred method for testing points of a circuit considering test uncertainty of the present invention includes the following steps:
s1, constructing a multi-signal model H = { F, T, D, P, C, PF } of testing uncertainty;
wherein, F represents various fault sets of the circuit to be tested, and F = { F = { (F) 1 ,f 2 ,…,f i ,…,f m }，f i Representing the i-th fault, wherein m is the total number of faults; t represents all available measurement point sets of the circuit to be measured, T = { T = { (T) 1 ,t 2 ,…,t j ,…,t n }，t j Representing the jth available measuring point, wherein n is the total number of the available measuring points; p represents a prior probability set of a certain fault of the circuit to be tested, and P = { P = 1 ,p 2 ,…,p i ,…,p m }，p i Indicating the occurrence of a fault s i A prior probability of (c); c represents a test cost set corresponding to T, C = { C 1 ,c 2 ,…,c j ,…,c n }，c j Represents the test t j The cost of (d); d is a dependency matrix under the uncertainty of the measuring points, and is specifically expressed as follows:
wherein, d ij Indicates a fault f i At available measuring point t j Test information of d ij =0 or d ij (= 1) when d ij If =0, it indicates that a fault has occurred in the circuit under test f i Can not pass through the available measuring point t j Detecting the result; when d is ij If =1, it indicates that a fault has occurred in the circuit under test f i Can pass through the available measuring point t j Detecting;
the PF is a false alarm rate matrix of the measuring point test, which is specifically expressed as:
wherein, pf ij Represents the measured point t j For a fault t j The test false alarm rate of (1);
in this example, the test data is derived from: yao zhai, smin jun, poplar commander and Qinyu; under unreliable test conditions, selecting [ J ] based on NSGA-II multi-objective test optimization, beijing university of aerospace, 2021,47 (04): 792-801) ]; a total of m =15 faults, n =20 test points, and the parameters are shown in tables 1-4;
table 1: prior failure probability of failure
Table 2: measuring point and cost
| Measuring point | t 1 | t 2 | t 3 | t 4 | t 5 | t 6 | t 7 | t 8 | t 9 | t 10 |
| Cost of | 60 | 66 | 120 | 60 | 52 | 90 | 50 | 60 | 20 | 36 |
| Measuring point | t 11 | t 12 | t 13 | t 14 | t 15 | t 16 | t 17 | t 18 | t 19 | t 20 |
| Cost of | 7 | 18 | 36 | 80 | 30 | 60 | 45 | 9 | 20 | 30 |
Table 3: probability of reliability of Circuit (%) under incomplete reliability of Point
Table 4: test unreliable point false alarm rate table (%)
S2, carrying out measuring point optimization on the test uncertainty model by using a multi-target particle swarm algorithm based on crowded distance sorting;
s2.1, setting the position of a particle k in the multi-target particle swarm optimization to be represented as
When takes on a value of 0 or 1, when>
Then, represent the measurement point t j Is selected when>
Then, represent the measurement point t j Is not selected, where j ∈ [1, n ]]；
S2.2, constructing an optimized objective function consisting of the test cost, the missing inspection rate and the false alarm rate;
and (3) testing cost:
the omission rate is as follows:
/>
false alarm rate:
s2.3, constructing a constraint condition consisting of the detection rate and the isolation rate;
detection rate:
isolation rate:
setting the lowest detection rate lambda 0 And isolation ratio lambda 1 Then the detection rate psi FD (x)≥λ 0 Isolation ratio psi FI (x)≥λ 1 ；
S2.4, carrying out measurement point optimization on the circuit to be measured based on the multi-target particle swarm algorithm of crowded distance sorting, wherein as shown in figure 2, the specific steps are as follows:
s2.4.1, initializing a particle group P consisting of q =500 particles, each having n dimensions and an initial velocity of 0, wherein the velocity of the kth particle is represented as:
the initial position of the kth particle is expressed as
k∈[1,q]，j∈[1,n]；
Setting the maximum iteration time T =100 of the multi-target particle swarm algorithm, setting the current iteration time T to be [1, T ], and initializing T =1;
setting an external archive REP with a size R =500, R > q; initializing REP as an empty set;
setting an array pbest with the size of q × n, wherein each line of pbest is used for recording the corresponding position of the historical optimal value of each particle after each iteration, and the kth line after the tth iteration is recorded as:
pbest stores the corresponding position of the historical optimal value of each particle;
setting an array gbest with the size of 1 multiplied by n, and recording the corresponding position of the global optimal value of the particle swarm after each iteration, wherein the global optimal value after the t iteration is recorded as
g is a particle which is optimally corresponding to the historical optimal value in the q particles;
s2.4.2, updating the speed and the position of each particle after the t iteration;
update the velocity of each particle:
V k (t)＝w×V k (t-1)+r 1 a 1 (P k (t-1)-X k (t-1))+a 2 r 2 (P g (t-1)-X k (t-1))
wherein, w is an inertia factor and takes the value of 0.4; r is a radical of hydrogen 1 、r 2 Random numbers varying in the range of (0, 1); a is a 1 、a 2 The values are all 2 for the acceleration coefficient;
updating the position of each particle:
when X is k (t) in
Then, then->
Or else>
rand () generates a random number between (0, 1);
s2.4.3, calculating a fitness value and a constraint value after the t iteration;
substituting the position of each particle after the t iteration into objective functions C (X), L (X) and A (X) to obtain the fitness value of each particle, wherein the fitness value of the kth particle is recorded as
Respectively corresponding C (X), L (X) and A (X) to obtain calculated values;
substituting the position of each particle after the t-th iteration into a constraint condition psi FD (X)、ψ FI (X) obtaining a constraint value of each particle, wherein the constraint value of the kth particle is recorded as
Respectively correspond to psi FD (X)、ψ FI (X) obtaining a calculated value;
s2.4.4, updating the historical optimal value position of each particle and the global optimal value position of the particle swarm after the t-th iteration;
and updating the historical optimal value position of each particle: fit each particle after the t-th iterationStress value f k (t) comparing the historical best fitness value corresponding to the location stored in pbest if f k (t) if the historical optimal fitness value is greater than the historical optimal fitness value, replacing the corresponding historical optimal value position in the pbest, and otherwise, keeping the historical optimal value position unchanged;
updating the global optimal value position of the particle swarm: after the t iteration, the maximum fitness value is selected from the particles and is marked as f g (t) comparing with the global optimum corresponding to the location stored by gbest, if f g (t) if the global optimum value is greater than the global optimum value, replacing the global optimum value position in the gbest, otherwise, keeping the global optimum value position unchanged;
s2.4.5, calculating a crowding distance;
s2.4.5.1, extracting the fitness value of each particle after the t iteration
Then, performing descending order arrangement on the q groups of fitness values obtained under the same objective function;
s2.4.5.2, calculating the crowding distance dist of each particle after the t iteration k (t)；
Wherein the crowding distance obtained by the kth particle under the objective function C (X) is
And &>
The same can be obtained;
s2.4.6, archiving each particle;
extracting each particle stored in an external archive REP, and then arranging each particle after the t iteration in a descending order according to the size of the crowding distance; then, the size of the internal memory of the external archive repository REP is read, whether the size of the residual internal memory of the external archive repository REP is larger than the total number q of the particles is judged, and if the size of the residual internal memory of the external archive repository REP is larger than the total number q of the particles, the sorted particles are stored in the external archive repository REP in sequence; otherwise, selecting the R particles with the largest crowding distance and storing the R particles in the external archive REP in sequence, and the benefit of using this archiving method is that the possible preferred results can be found as widely as possible, rather than concentrating all the preferred results at a certain place, i.e. without falling into local optimality, and finding as many preferred solutions as possible globally;
s2.4.7, outputting a preferred measuring point;
judging whether the current iteration time T reaches the maximum iteration time T, if T is less than T, adding 1 to the current iteration time T, returning to the step S2.4.2, and continuing the next iteration; otherwise, the loop is exited, R particles are output from the external archive REP, the position of each particle represents a preferred measuring point scheme, and the position of the corresponding particle with the largest crowding distance has higher priority as the preferred measuring point.
The final test result is shown in fig. 3, and the optimized combination of some selected test points is shown in table 5:
table 5 four optimal combinations in the optimization results
| Measuring point combination | Cost of | Rate of missing inspection | False alarm rate | |
| Cost optimization | {1,1,0,0,0,0,0,0,1,1,1,1,1,0,1,0,1,1,1,0} | 347 | 0.12％ | 1.07％ |
| The lowest missing rate | {1,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,0,1,1} | 442 | 0.016％ | 0.98％ |
| Lowest false alarm rate | {1,1,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,0,0,0} | 458 | 0.016％ | 0.78％ |
| Synthetic optimization | {1,1,0,0,0,0,0,0,1,1,1,1,1,0,1,1,1,1,0,0} | 387 | 0.016％ | 0.88％ |
As can be seen from Table 5, the method can reduce the test cost, the missing rate and the false alarm rate as much as possible on the premise of meeting the detection rate and the isolation rate, and optimize various schemes for people to select. We can select a satisfactory test scheme as required.
In summary, the test cost, the false alarm rate and the false alarm rate are used as objective functions, three aspects are integrated for optimization, so that the three target directions are all as small as possible, and the table 5 shows that the false alarm rate and the false alarm rate are not too high even if the lowest cost is selected.
It can be seen from fig. 3 that the optimization results of the present invention are not concentrated in one part, but rather are widely distributed, which is despite the benefit of using archiving based on the crowding distance, so that the optimization results do not fall into local optimality.
TABLE 6 comprehensive optimal comparison under three methods
As can be seen from Table 6, compared with the comprehensive optimization of the three methods, the method has obvious advantages, and particularly, compared with the NSGA-II and simulated annealing particle swarm optimization, the method has more obvious advantages in the optimization of the omission factor. Although the illustrative embodiments of the present invention have been described in order to facilitate those skilled in the art to understand the present invention, it is to be understood that the present invention is not limited to the scope of the embodiments, and that various changes may be made apparent to those skilled in the art as long as they are within the spirit and scope of the present invention as defined and defined in the appended claims, and all matters of the invention using the inventive concepts are protected.
Claims (1)
1. A method for optimizing circuit test points in view of test uncertainty, comprising the steps of:
(1) Constructing a multi-signal model H = { F, T, D, P, C, PF } with testing uncertainty;
wherein, F represents various fault sets of the circuit to be tested, and F = { F = { (F) 1 ,f 2 ,…,f i ,…,f m }，f i Representing the ith fault, wherein m is the total number of faults; t represents all available measurement point sets of the circuit to be measured, T = { T = { (T) 1 ,t 2 ,…,t j ,…,t n }，t j Representing the jth available measuring point, wherein n is the total number of the available measuring points; p represents a prior probability set of a certain fault of the circuit to be tested, and P = { P = 1 ,p 2 ,…,p i ,…,p m }，p i Indicating the occurrence of a fault s i A priori probability of (a); c represents a test cost set corresponding to T, C = { C 1 ,c 2 ,…,c j ,…,c n }，c j Represents the test t j The cost of (d); d is a dependency matrix under the uncertainty of the measuring points, and is specifically expressed as follows:
wherein d is ij Indicates a fault f i At available measuring point t j Test information of d ij =0 or d ij (= 1) when d ij If =0, it indicates that a fault has occurred in the circuit under test f i Time can not pass through available measuring point t j Detecting; when d is ij If =1, it indicates that a fault has occurred in the circuit under test f i Can pass through available measuring point t j Detecting;
the PF is a false alarm rate matrix of the measuring point test, which is specifically expressed as:
wherein, pf ij Represents the measured point t j For a fault t j The test false alarm rate of (1);
(2) Carrying out measuring point optimization on the test uncertainty model by using a multi-target particle swarm algorithm based on crowded distance sorting;
(2.1) setting the position of the particle k in the multi-target particle swarm optimization algorithm to be expressed as
Is taken to be 0 or 1, when>
Then, represent the measurement point t j Is selected when>
Then, represent the measurement point t j Is not selected, where j ∈ [1, n ]]；
(2.2) constructing an optimized objective function consisting of the test cost, the missed detection rate and the false alarm rate;
and (3) testing cost:
the omission rate is as follows:
false alarm rate:
(2.3) constructing a constraint condition consisting of the detection rate and the isolation rate;
detection rate:
/>
isolation rate:
setting the lowest detection rate lambda 0 And isolation ratio lambda 1 Then the detection rate psi FD (x)≥λ 0 Isolation ratio psi FI (x)≥λ 1 ；
(2.4) carrying out measurement point optimization on the circuit to be measured based on the crowding distance sorting multi-target particle swarm algorithm;
(2.4.1) initializing a particle group P consisting of q particles, each having n dimensions, and each having an initial velocity of 0, wherein the velocity of the kth particle is represented by:
the initial position of the kth particle is expressed as
Setting the maximum iteration times T of the multi-target particle swarm algorithm, setting the current iteration times T to be E [1, T ], and initializing T =1;
setting an external archive REP with the size of R, wherein R is more than q; initializing REP as an empty set;
setting an array pbest with the size of q multiplied by n, wherein each line of the pbest is used for recording the historical optimal value corresponding position of each particle after each iteration, and the kth line after the t iteration is recorded as:
pbest stores the corresponding position of the historical optimal value of each particle;
setting an array gbest with the size of 1 multiplied by n, and recording the corresponding position of the global optimal value of the particle swarm after each iteration, wherein the global optimal value after the t iteration is recorded as
g is a particle which is optimally corresponding to the historical optimal value in the q particles;
(2.4.2) updating the speed and the position of each particle after the t iteration;
update speed of each particle:
V k (t)＝w×V k (t-1)+r 1 a 1 (P k (t-1)-X k (t-1))+a 2 r 2 (P g (t-1)-X k (t-1))
wherein w is an inertia factor, r 1 、r 2 Is a random number varying in the range of (0, 1), a 1 、a 2 Is an acceleration factor;
updating the position of each particle:
when X is present k (t) in
Then, then->
Otherwise->
rand () generates a random number between (0, 1);
(2.4.3) calculating a fitness value and a constraint value after the t iteration;
substituting the position of each particle after the t iteration into objective functions C (X), L (X) and A (X) to obtain the fitness value of each particle, wherein the fitness value of the kth particle is recorded as
Respectively corresponding C (X), L (X) and A (X) to obtain calculated values;
substituting the position of each particle after the t-th iteration into the constraint condition psi FD (X)、ψ FI (X) obtaining a constraint value of each particle, wherein the constraint value of the kth particle is recorded as
Respectively correspond to psi FD (X)、ψ FI (X) obtaining a calculated value;
(2.4.4) updating the historical optimal value position of each particle and the global optimal value position of the particle swarm after the t-th iteration;
and updating the historical optimal value position of each particle: the fitness value f of each particle after the t iteration k (t) comparing with historical optimal fitness values corresponding to locations stored in pbest, if f k (t) if the historical optimal fitness value is greater than the historical optimal fitness value, replacing the corresponding historical optimal value position in the pbest, and otherwise, keeping the historical optimal value position unchanged;
updating the global optimal value position of the particle swarm: after the t-th iteration, the most significant of the particles is selectedLarge fitness value, noted as f g (t) comparing with the global optimum corresponding to the location stored by gbest, if f g (t) if the global optimum value is greater than the global optimum value, replacing the global optimum value position in the gbest, otherwise, keeping the global optimum value position unchanged;
(2.4.5), calculating the crowding distance;
(2.4.5.1) extracting the fitness value of each particle after the t iteration
Then, performing descending order arrangement on the q groups of fitness values obtained under the same objective function;
(2.4.5.2) calculating the crowding distance dist of each particle after the t iteration k (t)；
Wherein the crowding distance obtained by the kth particle under the objective function C (X) is
And &>
The same can be obtained;
(2.4.6) archiving each particle;
extracting each particle stored in an external archive REP, and then arranging each particle after the t iteration in a descending order according to the size of the crowding distance; then, the memory size of the external archive REP is read, whether the remaining memory size of the external archive REP is larger than the total number q of the particles is judged, and if so, the sorted particles are stored in the external archive REP in sequence; otherwise, selecting the R particles with the largest crowding distance and storing the R particles in an external archive REP in sequence;
(2.4.7) outputting a preferred measuring point;
judging whether the current iteration time T reaches the maximum iteration time T, if T is less than T, adding 1 to the current iteration time T, returning to the step (2.4.2), and continuing the next iteration; otherwise, the loop is exited and R particles are output from the external archive REP, the position of each particle representing a preferred measurement point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210021680.XA CN114429082B (en) | 2022-01-10 | 2022-01-10 | Circuit test point optimization method considering test uncertainty |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210021680.XA CN114429082B (en) | 2022-01-10 | 2022-01-10 | Circuit test point optimization method considering test uncertainty |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114429082A CN114429082A (en) | 2022-05-03 |
| CN114429082B true CN114429082B (en) | 2023-04-07 |
Family
ID=81311860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210021680.XA Active CN114429082B (en) | 2022-01-10 | 2022-01-10 | Circuit test point optimization method considering test uncertainty |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114429082B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101295008A (en) * | 2008-06-19 | 2008-10-29 | 电子科技大学 | Multi-target fault testing optimization method based on discrete particle swarm algorithm |
| US8793200B1 (en) * | 2009-09-22 | 2014-07-29 | Hrl Laboratories, Llc | Method for particle swarm optimization with random walk |
| CN105093095A (en) * | 2015-07-10 | 2015-11-25 | 四川航天系统工程研究所 | Improved particle swarm optimization diagnosis method and system of analog non-linear direct current circuit |
| CN105825063A (en) * | 2016-03-21 | 2016-08-03 | 北京航空航天大学 | Method for quantitatively selecting test point in design-for-test |
| CN106447025A (en) * | 2016-08-31 | 2017-02-22 | 电子科技大学 | Combined testability index distribution and test selection algorithm based on discrete particle swarm |
| CN107657311A (en) * | 2017-11-03 | 2018-02-02 | 电子科技大学 | Test method for optimizing based on multi-objective particle swarm algorithm |
| CN111412795A (en) * | 2020-03-16 | 2020-07-14 | 中国人民解放军海军航空大学 | Test point setting scheme generation method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9053426B2 (en) * | 2013-01-30 | 2015-06-09 | Harris Corporation | Distributed multi-layer particle swarm optimization based cognitive network |
-
2022
- 2022-01-10 CN CN202210021680.XA patent/CN114429082B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101295008A (en) * | 2008-06-19 | 2008-10-29 | 电子科技大学 | Multi-target fault testing optimization method based on discrete particle swarm algorithm |
| US8793200B1 (en) * | 2009-09-22 | 2014-07-29 | Hrl Laboratories, Llc | Method for particle swarm optimization with random walk |
| CN105093095A (en) * | 2015-07-10 | 2015-11-25 | 四川航天系统工程研究所 | Improved particle swarm optimization diagnosis method and system of analog non-linear direct current circuit |
| CN105825063A (en) * | 2016-03-21 | 2016-08-03 | 北京航空航天大学 | Method for quantitatively selecting test point in design-for-test |
| CN106447025A (en) * | 2016-08-31 | 2017-02-22 | 电子科技大学 | Combined testability index distribution and test selection algorithm based on discrete particle swarm |
| CN107657311A (en) * | 2017-11-03 | 2018-02-02 | 电子科技大学 | Test method for optimizing based on multi-objective particle swarm algorithm |
| CN111412795A (en) * | 2020-03-16 | 2020-07-14 | 中国人民解放军海军航空大学 | Test point setting scheme generation method and device |
Non-Patent Citations (2)
| Title |
|---|
| 王成刚 ; 应朝龙 ; 李建海 ; 刘志远 ; .基于DPSO的不确定系统测试性建模.电子器件.2010,(04),全文. * |
| 雷华军 ; 秦开宇 ; .测试不可靠条件下基于量子进化算法的测试优化选择.电子学报.2017,(10),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114429082A (en) | 2022-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111562108A (en) | Rolling bearing intelligent fault diagnosis method based on CNN and FCMC | |
| CN111695620B (en) | Method and system for detecting and correcting time sequence abnormal data of power system | |
| CN111814956B (en) | Multi-task learning air quality prediction method based on multi-dimensional secondary feature extraction | |
| CN112257963B (en) | Defect prediction method and device based on spaceflight software defect data distribution outlier | |
| CN114048688A (en) | Method for predicting service life of bearing of wind power generator | |
| CN111398798A (en) | Circuit breaker energy storage state identification method based on vibration signal interval characteristic extraction | |
| CN112149502A (en) | Unfavorable geology positioning forecasting method based on convolutional neural network | |
| CN109614512B (en) | Deep learning-based power equipment retrieval method | |
| CN114429082B (en) | Circuit test point optimization method considering test uncertainty | |
| CN117113066B (en) | Transmission line insulator defect detection method based on computer vision | |
| CN114357372A (en) | Aircraft fault diagnosis model generation method based on multi-sensor data driving | |
| CN112598666B (en) | Cable tunnel anomaly detection method based on convolutional neural network | |
| CN116842323A (en) | Abnormal detection method for operation data of water supply pipeline | |
| Ni et al. | Multi-meter intelligent detection and recognition method under complex background | |
| CN116127354A (en) | High-voltage cable partial discharge fault positioning method and system based on deep learning | |
| CN113221496B (en) | Fault diagnosis method based on three-dimensional testability analysis model | |
| CN113204894B (en) | Construction method and application of electric energy metering abnormality diagnosis model | |
| CN115471453A (en) | Power equipment fault diagnosis method based on image processing | |
| CN113887452A (en) | Fault diagnosis optimization method based on correlation matrix | |
| Zhao et al. | Industrial fault diagnosis based on few shot learning | |
| CN113971259B (en) | Error reliability parameter identification and correction method for power generation and transmission system element | |
| CN117494588B (en) | Method, equipment and medium for optimizing residual effective life of fan bearing | |
| CN115965823B (en) | Online difficult sample mining method and system based on Focal loss function | |
| CN114626195B (en) | Modeling method and system for solid oxide fuel cell system by using space-time data | |
| CN110942401B (en) | Intelligent communication method for electric power Internet of things |
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 |