CN115994657A

CN115994657A - Unit division optimization method and system adopting universal detection tool for detection

Info

Publication number: CN115994657A
Application number: CN202211591573.7A
Authority: CN
Inventors: 张宁; 孙强; 李佳宽; 饶喆; 陈文须; 李华
Original assignee: Naval University of Engineering PLA
Current assignee: Naval University of Engineering PLA
Priority date: 2022-12-12
Filing date: 2022-12-12
Publication date: 2023-04-21

Abstract

The invention discloses a unit division optimizing method and system adopting a universal detection tool for detection, and belongs to the field of electronic equipment fault investigation. The invention constructs a candidate division set by traversing group leader units and preferentially selecting group member units, wherein the group leader units are units with longest unit inspection time consumption, when the group leader units are preferentially selected, a plurality of units with highest fault probability, which are not more than the group leader unit state inspection time consumption, are selected from all current non-inspected units, are used as the group member units, a feasible unit division for the jth detection is jointly formed with the feasible group leader units, the division with the smallest overall equivalent detection time is selected from all candidate feasible unit division modes as the current division, the optimal division of all units is finally realized, the division optimization method realizes the globally optimal neighborhood solution, and the average inspection time is greatly reduced.

Description

Unit division optimization method and system adopting universal detection tool for detection

Technical Field

The invention belongs to the field of electronic equipment fault detection, and particularly relates to a unit division optimizing method and system for detecting by using a universal detection tool.

Background

The interfaces of electronic devices are easy to standardize, and detection tools capable of detecting the same general type but different specific specification and models exist, and are called universal detection tools. For example, a device has a plurality of hard disks, and each hard disk is loaded with a respective program and data file, when a part of a hard disk is damaged, an occasional fault is caused, and the model of each hard disk is not identical, but is a standard interface, so that a universal detection tool can be adopted at the moment, and meanwhile, a disk damage check or a random data reading test can be performed on a plurality of hard disks. As equipment/systems become more powerful and more advanced in performance, the equipment/systems become more complex. When a certain fault phenomenon occurs in the complex equipment/system, the possible reasons behind the fault phenomenon are numerous, and the workload of searching the fault unit is huge.

The inspection tool is an important maintenance resource and the number of units that can be inspected at most at one time by the universal inspection tool is called the capacity of the inspection tool. Generally, the larger the capacity of the inspection tool, the more quickly the failure-causing unit can be inspected, but at the same time the more costly the inspection tool is to be used, such as for cost, space occupation, and the like.

The decision process of selecting together check elements for use with a common detection tool when formulating an investigation scheme is referred to as element partitioning. When a fault investigation scheme is formulated, the basic problems faced are: how to preferentially select a plurality of units to be tested and find a faulty unit as soon as possible each time the detection tool is used? This problem has not been solved well in the industry at present. Especially for complex equipment or systems, the number of units to be examined is few tens or more than hundred. The unit division can be regarded as an arrangement and combination problem of the units to be checked, for example, when the number of the units is 16 and the capacity of the detection tool is 4, the division number exceeds 6300 ten thousand, and the traversing mode can not be adopted to effectively optimize the checking scheme. At present, units are divided mainly by virtue of self experience of first-line maintenance personnel, the divided quality is uneven, and a more accurate time equivalent result for completing fault detection is difficult to give, so that a scientific fault detection scheme is not facilitated to be formulated.

Disclosure of Invention

Aiming at the defects of the prior art, the invention aims to provide a unit division optimizing method and system detected by a universal detection tool, and aims to solve the problems that the existing unit division method depends on experience and the division quality is uneven.

To achieve the above object, in a first aspect, the present invention provides a unit division optimizing method using a general detection tool, the method comprising:

s1, acquiring capacity m of a general detection tool, the number of electronic class units forming electronic equipment, the state detection time of each electronic class unit and the probability of failure of each electronic class unit in task time;

s2, the ratio of the number of the electronic class units to the capacity is rounded upwards, the maximum detection times nm is used, the detection sequence j=1 is initialized, and an undetected unit set is initialized to form all electronic class units of the electronic equipment;

s3, finding out all feasible group leader units from the current undetected unit set, wherein the group leader units are units with the longest unit inspection time consumption; for each feasible group length unit, selecting m-1 units which are not more than the unit state detection time consuming and have highest fault probability, and forming a feasible unit division for the jth detection together with the feasible group length units, wherein all the feasible unit divisions jointly form a feasible unit division candidate set for the jth detection;

s4, selecting a division mode with minimum overall equivalent detection time from a feasible unit division candidate set detected at the j-th time as a unit division result detected at the j-th time, and updating an undetected unit set into a difference set between a current undetected unit set and the unit division result detected at the j-th time;

s5, updating j=j+1, if j is less than nm, entering S3, otherwise, taking the current undetected unit set as a unit division result of the last 1 times of detection.

Preferably, step S3 includes:

s31. initializing the grouping number s=0, i=1, and the number na=n- (j-1) m of the current undetected unit set elements, where n is the number of electronic class units forming the electronic device;

s32. initialize group leader unit number z=a _i The number of the unit of the candidate member, which is the rest elements except z, in the undetected unit set A is marked as At;

s33, finding out that all unit checking time is not more than tc in the At _z Is placed in an array idy, if the length of the array idy is not less than m-1, the step S34 is carried out, and if not, the step S36 is directly carried out;

s34, pairing array pf _idy After descending order, the unit numbers corresponding to the largest first m-1 elements are arranged in an array, zy is the group member unit number, pf _idy The unit failure probabilities corresponding to the units in the array idy are numbered for all units;

s35, updating s=s+1, placing [ z zy ] on the s line of the matrix zall, and calculating the whole equivalent detection time;

s36, updating i=i+1, if i is less than or equal to _na S32, otherwise, S4.

Preferably, the calculation formula of the overall equivalent detection time is as follows:

wherein tc _z Pf for checking time of group leader unit _z For the failure probability of the group leader unit, pg is the accumulated result of the failure probabilities of all units except the group leader unit in the possible unit division results.

Preferably, the method further comprises:

s6, calculating average fault detection time Tc detected by using a general detection tool:

therein, zPf _j For checking the sum of the failure probabilities of all units for the j-th time, tx _j For the accumulated checking time of the previous j times of detection, pf _r Is the probability of a failure of unit r.

Preferably, the calculation formula of the accumulated check time of the previous j detections is as follows:

therein, ztc _r The maximum value of the inspection time in all the units is inspected for the r-th time.

Preferably, the method further comprises:

s6, calculating fault detection probability distribution detected by a general detection tool:

wherein px is _j The probability of finding the fault part for the first j times is that n is the number of electronic class units constituting the electronic equipment.

To achieve the above object, in a second aspect, the present invention provides a unit division optimizing system using a general detection tool, comprising: a processor and a memory; the memory is used for storing computer execution instructions; the processor is configured to execute the computer-executable instructions such that the method of the first aspect is performed.

In general, the above technical solutions conceived by the present invention have the following beneficial effects compared with the prior art:

the invention provides a unit division optimizing method and a system adopting a general detection tool to detect, which are characterized in that a candidate division set is formed by traversing group leader units and preferentially selecting group member units, wherein the group leader units are units with the longest time consumption for unit inspection, when the group member units are preferentially selected, m-1 units with the highest failure probability and time consumption for unit inspection are selected from all current undetected units, and the m-1 units are taken as the group member units and are combined with the feasible group leader units to form a feasible unit division for jth detection, the division with the smallest overall equivalent detection time is selected from all candidate feasible unit division modes to be taken as the current division, and finally, the optimized division of all units is realized.

Drawings

FIG. 1 is a flow chart of a method for optimizing unit division using a universal detection tool according to the present invention.

Fig. 2 is a schematic diagram of simulation verification results of probability distribution of troubleshooting time provided by the embodiment of the invention.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

The invention provides a method for optimizing and dividing units by adopting a universal detection tool aiming at electronic equipment, which can correlate the troubleshooting workload, the troubleshooting time and the capacity, effectively reduce the average troubleshooting time and improve the timeliness of maintenance work.

FIG. 1 is a flow chart of a method for optimizing unit division using a universal detection tool according to the present invention. As shown in fig. 1, the method includes:

s1, acquiring the capacity m of a general detection tool, the number of electronic units forming the electronic equipment, the state detection time of each electronic unit and the probability of failure of each electronic unit in task time.

In engineering, electronic components that are normally used all belong to the exponential lifetime piece, such as: printed circuit board packages, electronic components, resistors, capacitors, integrated circuits, etc. The exponential type unit refers to a unit whose lifetime obeys an exponential distribution Exp (u), and the physical meaning of the parameter u is a lifetime average. Density function of exponential distributionThe number is

The invention has the following conventions: (1) Some equipment is composed of a plurality of electronic type units, and for convenience of description, the life of each unit is described in terms of time. (2) at any time, at most 1 unit fails. When a unit fails, normal operation of the equipment is affected, and certain failure phenomena occur in the equipment, and repair work is required. (3) The order in which the units are status checked in finding the cause of the failure is independent, i.e.: there are no cases where a specific requirement for the inspection order is made, such as "unit a must be inspected first and then unit B". (4) The life distribution law of each unit, the time taken for performing a status check (normal or not) for each unit, and the time at which a task is to be executed are known. (5) After the detection tool starts to detect a plurality of units at the same time, only if the units are detected, the detection tool gives the detection result of each unit.

The relevant variable conventions of the present invention are as follows: the capacity of the detection tool is recorded as m; the number of units is denoted n; lifetime compliance exponential distribution Exp (u) of unit i _i ) The method comprises the steps of carrying out a first treatment on the surface of the The time consumed for the status check of cell i is denoted as tc _i The method comprises the steps of carrying out a first treatment on the surface of the The task time is noted Tw. These variables are known amounts.

The invention regards all units to be inspected in one inspection using the inspection tool as a group of units, wherein the unit with the longest inspection time is called group leader unit and the other units are called group member units.

The failure probability Pf of each cell is calculated by traversal.

1) Let i=1;

2) Integral calculation Pf _i ：

When k=i, the number of times,

when k+.i, +.>

3) Let i=i+1, if i is less than or equal to n, then execute 2), otherwise execute S2.

And S2, the ratio of the number of the electronic class units to the capacity is rounded up, the maximum detection times nm is used, the detection sequence j=1 is initialized, and the undetected unit set is initialized to be all the electronic class units forming the electronic equipment.

Let the maximum number of detections nm=n/m and take the integer upwards. The unit numbers 1 to n are placed in the set a, and the number of unit numbers in the set a is denoted as na.

S3, finding out all feasible group leader units from the current undetected unit set, wherein the group leader units are units with the longest time consumption for unit inspection; for each feasible group leader unit, selecting m-1 units which are not more than the unit group leader unit and have the time consumption in state inspection and highest fault probability, and forming one feasible unit division for the jth detection together with the feasible group leader unit, wherein all the feasible unit divisions jointly form a feasible unit division candidate set for the jth detection.

Preferably, step S3 includes:

s36, updating i=i+1, if i is less than or equal to _na S32, otherwise, S4.

S4, selecting a division mode with minimum overall equivalent detection time from the feasible unit division candidate set detected at the j-th time as a unit division result detected at the j-th time, and updating the undetected unit set into a difference set between the current undetected unit set and the unit division result detected at the j-th time.

And S5, updating j=j+1, if j is less than nm, entering S3, otherwise, taking the current undetected unit set as a unit division result of the last 1 times of detection.

Preferably, the method further comprises: s6, calculating average fault detection time Tc detected by using a general detection tool:

Preferably, the method further comprises: s6, calculating fault detection probability distribution detected by a general detection tool:

The evaluation unit divides the time consuming result of the result Mdy.

1) The initialization detection sequence number j=1, and the average trouble shooting time tc=0.

2) The j-th row of non-zero elements in the matrix Mdy are placed in an array zj, and the unit number of the j-th detection is stored in zj.

3) Finding the maximum value of the corresponding unit check time from all the zj units, and storing the maximum value in ztc _j . Accumulating fault probability of all units of zj and storing the fault probability in zPf _j 。

4) Updating

Wherein pxj is at time tx _j The probability of finding a faulty piece is found.

5) Updating j=j+1, if j is less than or equal to nm, executing 2), otherwise, terminating calculation, and outputting unit division results Mdy of each detection, average troubleshooting time Tc, troubleshooting time probability distribution tx and px when using the general detection tool.

Examples

It is known that a certain part is composed of 20 electronic class units, the task time is 100 hours, the capacity of the detection tool is 5, and the related information is shown in table 1. By adopting the method, the average detection time for finding the fault at the moment is calculated.

TABLE 1

Unit numbering	Distribution parameter u	Status checking time/min
			1	865	7
2	1170	55
			3	3705	48
4	670	8
			5	2350	12
6	2940	25
			7	2480	14
8	3935	28
			9	1400	35
10	2325	44
			11	1700	52
12	3990	22
			13	2240	59
14	2330	47
			15	2490	35
16	1105	43
			17	3165	44
18	4025	45
			19	810	61
20	1855	54

1) The probability Pf of each unit failure was calculated by traversal, and the result is shown in table 2.

TABLE 2

/>

2) Let the maximum detection number nm=4, put the unit numbers 1 to 20 in the set a, the number na=20 of unit numbers in the set a, and the detection number j=1.

3) The feasible unit division zall of the jth detection and the equivalent detection time tpf are optimally calculated. When j=1, the unit division zall and the equivalent detection time tpf thereof are shown in table 3.

TABLE 3 Table 3

4) According to tpf, the optimal cell partition is selected from zall and the result is saved to Mdy.

The minimum value tpm=100.12 in the array tpf, the corresponding sequence number is denoted as im=8, and the element in row im of the matrix zall is selected as the unit sequence number zj= [12,4,1,5,7] detected 1 st time, and stored in row 1 of Mdy in the matrix.

5) Update j=j+1, delete zj from a, na=na-m, i=1, if j < nm performs 3), otherwise, directly place the element in a on row nm of matrix Mdy.

Executing 3) and 4) for detecting j=2, 3 and 4 in sequence, and Mdy calculating results are shown in table 4.

TABLE 4 Table 4

6) The evaluation unit divides the time consumption of the result Mdy, tc=87.2, and the probability distribution result px of the troubleshooting completion time tx is shown in table 5.

TABLE 5

7) The calculation is terminated, and the unit division result Mdy, the average troubleshooting time Tc, and the probability distribution px of the troubleshooting time tx are output. The optimized fault checking scheme is as follows: the unit numbers of the first detection are 12,4,1,5 and 7, which consume 22 minutes, and the probability of finding a fault part is 0.32; if the first detection unit is normal, sequentially performing second, third and fourth detection, wherein the second detection unit numbers are 19, 16, 2, 9 and 11, the third detection unit numbers are 14, 10, 15, 6 and 17, and the fourth detection unit numbers are 3, 8, 13, 18 and 20; the probability of finding a faulty part in 83, 130, 189 minutes is 0.69, 0.85, 1.00 in order. The average troubleshooting time was 87.2 minutes.

A simulation model can be established to verify the correctness of the method, and the simulation model is briefly described as follows:

(1) Generating n random numbers simT _i ，1≤i≤n，simT _i Obeying the life distribution law of the unit i.

(2) At all simT _i The minimum number is found, and the corresponding serial number is marked as g, namely: simT _g ≤simT _i ，1≤i≤n。

(3) If simT _g If Tw is less than Tw, the simulation is valid, and the simulation is carried out according to the detectionThe detection tool is used for detecting at most m units at a time until the unit g is detected, so that the detection time for finding the fault cause at the time can be obtained.

After a large number of simulation times, the average fault investigation time and the probability distribution condition thereof can be obtained through statistics.

Fig. 1 shows probability distribution results of the fault detection time of the above examples obtained by using the simulation method and the method of the present invention, respectively. From this figure, the results of the two were very consistent. 10000 fault investigation schemes are randomly generated according to the above calculation example, the simulation results of the minimum value, the maximum value and the average value of the average fault investigation time are respectively 101.2 minutes, 168.8 minutes and 137.8 minutes, and compared with the result 87.2 minutes of the method, the optimization effect of the method is obvious.

Table 6 shows the simulation results of the above-mentioned example average trouble shooting time and the evaluation results of the method of the present invention when the detection tool capacities were 1 to 10, respectively. As can be seen from Table 6, the two are very consistent.

TABLE 6

By using the method of the invention, the average detection time of each detection tool capacity can be calculated in a certain range by traversing, the result similar to the result shown in the table 6 can be obtained, and the detection tool capacity can be selected in an auxiliary way according to the result: for example, as can be seen from table 6, when the capacity exceeds 4, the average troubleshooting time shortened is limited, and when the capacity is not more than 4 from the viewpoint of cost performance; if the average troubleshooting time is required to be not more than 100 minutes, the capacity is at least 4.

It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims

1. The unit division optimizing method adopting the universal detection tool for detection is characterized by comprising the following steps:

2. The method of, wherein step S3 comprises:

s33, finding out that all unit checking time is not more than tc in the At _z Is placed in an array idy, if the length of the array idy is not less than m-1, the process proceeds to S34, otherwise, the process is straightS36 is accessed;

s36. updating i=i+1, if i is not greater than na, entering S32, otherwise, entering S4.

3. The method of claim 1, wherein the overall equivalent detection time is calculated as:

wherein tc _z Pf for checking time of group leader unit _z For the failure probability of the group leader unit, pg is the accumulated result of the failure probabilities of all units except the group leader unit in the feasible unit division results.

4. The method of claim 1, wherein the method further comprises:

5. The method of claim 4 wherein the accumulated inspection time for the first j detections is calculated as follows:

6. The method of claim 1, wherein the method further comprises:

7. A unit division optimization system employing universal detection tools for detection, comprising: a processor and a memory; the memory is used for storing computer execution instructions; the processor configured to execute the computer-executable instructions such that the method of any one of claims 1 to 6 is performed.