CN105760672B - A kind of Trouble Diagnostic Method of Machinery Equipment - Google Patents
A kind of Trouble Diagnostic Method of Machinery Equipment Download PDFInfo
- Publication number
- CN105760672B CN105760672B CN201610095084.0A CN201610095084A CN105760672B CN 105760672 B CN105760672 B CN 105760672B CN 201610095084 A CN201610095084 A CN 201610095084A CN 105760672 B CN105760672 B CN 105760672B
- Authority
- CN
- China
- Prior art keywords
- fault
- vector
- failure
- symptom
- determining
- 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.)
- Expired - Fee Related
Links
- 238000002405 diagnostic procedure Methods 0.000 title abstract 3
- 208000024891 symptom Diseases 0.000 claims abstract description 115
- 238000000034 method Methods 0.000 claims abstract description 40
- 238000003745 diagnosis Methods 0.000 claims abstract description 22
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 abstract description 2
- 230000015556 catabolic process Effects 0.000 abstract 2
- 206010039740 Screaming Diseases 0.000 description 5
- 230000007547 defect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 2
- 238000007789 sealing Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000036760 body temperature Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
Abstract
The invention discloses a kind of Trouble Diagnostic Method of Machinery Equipment, belongs to mechanical fault diagnosis technical field.It is characterized in that this method comprises the following steps：The probability vector B of parts failure is determined according to failure symptom1；According to the Parts Breakdown historical data of statistics, the failure rate vector B of parts inherently is determined2；The probability vector B to be broken down according to parts1And the failure rate vector B of parts inherently2, it is weighted, determines the combined chance vector B that parts break down；The size of the combined chance of parts failure is pressed, determines failure cause.The diagnostic method has considered Parts Breakdown rate, failure mechanism, failure symptom obvious degree, failure symptom and has obtained Multiple factors, the correctness of its diagnostic result such as complexity and increase considerably.
Description
Technical Field
The invention relates to a fault diagnosis method based on fault rate and fault symptoms, and belongs to the technical field of fault diagnosis of mechanical equipment.
Background
The fault diagnosis technology is a technology for judging whether a mechanical device is normal or not and finding a fault in time according to the running state of the mechanical device, and is a powerful guarantee for safe production and efficient running of the mechanical device, but the implementation of the technology faces many challenges: because the mechanical equipment has a plurality of parts, the movement process of the mechanical equipment is the sum of mechanical movements of partial parts of the mechanical equipment or all the parts, so that the faults of the mechanical equipment usually have the characteristics of multiple sources, complexity and concealment; the cause of the fault occurrence is complex, generally, the same fault can show various characteristics, and the same fault characteristic can be caused by different faults; in addition, most of the existing fault diagnosis methods make judgment based on simple fault symptoms, do not consider the fault mechanism of equipment and the difficulty degree of obtaining the fault symptoms, and have poor objectivity of diagnosis results.
Disclosure of Invention
The invention aims to overcome the defects of a fault diagnosis method under the condition of a single factor, and provides an equipment fault diagnosis method based on the fault rate and the fault symptoms of a part, so that the diagnosis result is more scientific and reasonable.
In order to achieve the above purpose, the present invention provides an apparatus fault diagnosis method based on a part fault rate and fault symptoms, namely a mechanical apparatus fault diagnosis method, which comprises the following steps:
1) Determining probability vector B1 of fault of part according to fault symptoms
(1) Establishing a fault symptom set U and a fault reason set V;
establishing a fault symptom set, U = { U = 1 ，u 2 ，…，u m In which u 1 ，u 2 ，…，u m The fault symptoms of the equipment are shown, and m is the number of the fault symptoms; set of causes of failure V = { V = { (V) } 1 ,v 2 ,…,v n }，v 1 ,v 2 ,…,v n Indicates the cause of the failure, and n is the number of causes of the failure.
(2) Determining a fault symptom significance vector A
According to the significance degree of the fault symptoms, scoring each fault symptom to form a fault symptom significance degree vector A = (a) 1 ，a 2 ，…，a m )，a 1 ，a 2 ，…，a m Indicating the significance of each symptom of the failure, the scores were according to the following table.
TABLE 1 Scale of severity of symptom of failure
(3) Determining a difficulty vector F obtained in the field of symptoms of a fault
TABLE 2 assessment of the difficulty of on-site acquisition of symptoms of a failure
According to the difficulty degree obtained by the fault symptom, determining a difficulty degree vector F obtained on the site of the fault symptom according to the table
F＝(f 1 ,f 2 ,…,f m )
Wherein f is 1 ，f 2 ，…，f m Indicating a fault symptom u of the device 1 ，u 2 ，…，u m The difficulty of obtaining, m is the number of symptoms of the failure
(4) Determining a judgment matrix R
Establishing a judgment matrix R and determining fault symptoms u i For fault v j Degree of membership r ij Then the membership of m fault symptoms to n fault causes forms an m × n evaluation matrix R.
In the formula r ij Is the symptom of failure u i For the fault reason v j The degree of membership of (c) satisfies 0. Ltoreq. R ij ≤1。
In the present invention, the symptom u is specific to each failure i The fault causes which may cause the fault symptoms are compared pairwise and expressed by the calibration numbers shown in the following tableThe magnitude relation of the fault symptom to the degree of membership of the fault reason. For each fault symptom u i Constructing a pairwise comparison matrix C i 。
TABLE 3 pairwise comparison matrix element scaling method
In the invention, pairwise comparison matrixes are respectively constructed by a plurality of experts, and after the pairwise comparison matrixes are constructed, the sum method, the root method or the power method is used for solving the membership degree of the fault symptom to the fault reason.
(5) Determining probability vector B of fault of related parts 1
Probability vector B of fault of part 1 ＝(b 11 ,b 12 ,…,b 1n )，b 11 ,b 12 ,…,b 1n Indicating a fault v 1 ,v 2 ,…,v n The size of the probability of existence, n is the number of faults.
On the basis of comprehensively considering the severity of the fault symptoms and the difficulty of obtaining the fault symptoms on site, the probability vector B of the fault of the part is obtained by the following formula 1
B 1 ＝(k 1 A+k 2 F)○R
Above formula k 1 And k 2 For vector A and vector F, the weighting coefficients, k in the present invention 1 Take 0.8,k 2 Taking 0.2, O is an operation rule, and the invention adopts a weighted average method.
2) Determining the fault rate vector B inherent to the relevant part 2
(1) And according to the statistical historical data, sequencing the fault interval time from small to large, and establishing a fault interval time table.
(2) According to the fault interval time table, estimating the reliability by a median rank formula
R(t i )＝1-(i-0.3)/(n+0.4)
Wherein t is 1 ,t 2 ,…,t n Is a fault interval time, and t 1 ≤t 2 ≤…≤t n N is the total number of acquisition faults
(3) Drawing a Weibull probability chart according to the estimated reliability value
Let x = lnt i ，y＝ln[-lnR(t i )]Thus (t) 1 ,R(t 1 )),(t 2 ,R(t 2 )),…,(t n ,R(t n ) The data set is transformed into ((x) 1 ,y 1 ),(x 2 ,y 2 ),…,(x n ,y n ) Data points are plotted on a Weibull probability map to obtain a Weibull probability map.
(4) On Weibull frequency paper, the left part of the graph is fitted with a straight line L 1 To obtain a straight line L 1 Has a slope and an intercept on the y-axis of k 1 And b 1 Fitting an asymptote L at the right end of the graph 2 To obtain a straight line L 2 Has a slope and an intercept on the y-axis of k 2 And b 2 。
(5) And determining the reliability function according to the double segmentation Weibull function.
Wherein
α 1 ＝exp(-b 1 /k 1 )
β 1 ＝k 1
α 2 ＝exp(-b 2 /k 2 )
β 2 ＝k 2
(6) Determining the failure rate inherent to the component itself
V = { V) for fault set 1 ,v 2 ,…,v n Determining the failure rate b of each unit in the failure set 2i ，i＝1,2,3,…,n。
b 2i ＝1-R(t)
(7) By using the method, the inherent fault rate vector B of the related part is determined 2
B 2 ＝(b 21 ,b 22 ,..b 2n )
3) And (2) carrying out weighting operation according to the fault symptom to determine the fault probability of the part in the step 1) and the inherent fault rate of the part in the step 2), and determining the comprehensive fault probability of the part.
(1) Normalizing the probability vector B of a failure of a relevant component determined from the symptoms of the failure 1
(2) Normalizing fault rate vector B of related parts 2
(3) Probability vector B of failure of normalized related parts 1 And the normalized fault rate vector B of the related parts 2 Performing weighted average calculation to obtain the comprehensive probability vector B of the fault of the part, namely
B＝k 3 B 1 +k 4 B 2
In the above formula k 3 And k 4 Are respectively vector B 1 Sum vector B 1 In the present invention, k is 3 Take 0.5,k 4 Take 0.5.
4) And determining the fault reason according to the probability of the fault of the part.
Compared with the existing fault diagnosis method for mechanical equipment, the fault diagnosis method for mechanical equipment has the following advantages:
1. in the process of diagnosing the fault of the mechanical equipment, the fault symptom of the equipment is considered, the inherent fault rate of the equipment parts is also considered, the phenomenon and the source are comprehensively considered, and the defect that the fault diagnosis accuracy rate of the mechanical equipment is low because only a single factor of the fault symptom of the equipment is considered in the conventional fault diagnosis method is overcome;
2. in the process of judging the fault according to the fault symptom, the invention not only considers the obvious degree of the fault symptom, but also considers the difficulty degree of obtaining the fault symptom, and avoids the defect that the fault diagnosis accuracy of the mechanical equipment is low because the current fault diagnosis method only considers the single factor of the obvious degree of the equipment fault symptom.
Drawings
Figure 1 is a flow chart of the fuzzy fault diagnosis steps of the present invention,
FIG. 2 is a Weibull probability plot of the present invention.
Detailed Description
The invention provides an equipment fault diagnosis method based on part fault rate and fault symptoms, a flow chart of which is shown in figure 1, and the following concrete implementation steps are described by taking fault diagnosis of a hydraulic system of a certain numerical control machine as an example:
step 1, determining the probability vector B of the fault of the part according to the fault symptom 1
(1) Establishing a fault symptom set U and a fault reason set V;
establishing a fault symptom set, U = { U = 1 ，u 2 ，…，u m In which u 1 ，u 2 ，…，u m The fault symptoms of the equipment are shown, and m is the number of the fault symptoms; set of causes of failure V = { V = { (V) } 1 ,v 2 ,…,v n }，v 1 ,v 2 ,…,v n Indicates the cause of the failure, and n is the number of causes of the failure.
For the numerical control machine hydraulic system, when a fault occurs, the display value of a pressure gauge is very low for the fault symptom; the hydraulic cylinder is weak to work; screaming sound and noise exist in the system; the temperature of the pump body is higher. I.e. the set of fault symptoms U = { U = { (U) } 1 ,u 2 ,u 3 ,u 4 And = (the display value of a pressure gauge is low, a hydraulic cylinder cannot work well, screaming sound and noise exist in a system, and the temperature of an oil pump is higher).
The hydraulic system of the numerical control machine tool has a fault part, namely, an internal part of an overflow valve is clamped, damaged or abraded; the internal parts of the oil pump are seriously worn; the oil filter is blocked by dirt; the hydraulic cylinder sealing element is damaged or not tightly sealed; the oil pump driving device slips. So the set of fault causes V = { V = { V } 1 ,v 2 ,v 3 ,v 4 ,v 5 And = (the internal parts of the overflow valve are clamped, damaged or worn, the internal parts of the oil pump are seriously worn, the oil filter is blocked by dirt, the sealing element of the hydraulic cylinder is damaged or not tightly sealed, and the oil pump driving device slips).
(2) Determining a fault symptom significance vector A
According to the significance degree of the fault symptoms, scoring each fault symptom to form a fault symptom significance degree vector A = (a) 1 ，a 2 ，…，a m )，a 1 ，a 2 ，…，a m Indicating the significance of each symptom of the failure, the scores were according to the following table.
TABLE 1 Scale of severity of symptom of failure
Observing the hydraulic system of the numerical control machine tool, and finding that the display value of a pressure gauge is very low; the force output by the hydraulic cylinder is smaller; the screaming sound and noise of the system are large; the pump temperature is slightly higher. From table 1, a fault symptom significance vector a = {0.7,0.3,0.7,0.3} is determined.
(3) Determining a difficulty vector F obtained in the field of symptoms of a fault
TABLE 2 assessment of the difficulty of on-site acquisition of symptoms of a failure
According to the difficulty degree obtained by the fault symptom, determining a difficulty degree vector F obtained on the site of the fault symptom according to the table
F＝(f 1 ,f 2 ,…,f m )
Wherein f is 1 ，f 2 ，…，f m Indicating a fault symptom u of the device 1 ，u 2 ，…，u m The difficulty of obtaining the symptom is m, and the m is the number of the symptom of the failure.
For the embodiment of the invention, the fault symptom pressure gauge shows a low u value 1 Easy observation and weak operation of hydraulic cylinder 2 Difficult observation, screaming sound and noise u in the system 3 Easy observation, high temperature of oil pump u 4 Easy to observe, so the failure symptom field obtained difficulty vector F = (0.3,0.7,0.5,0.5).
(4) Determining a judgment matrix R
Establishing a judgment matrix R and determining fault symptoms u i For the fault reason v j Degree of membership r of ij Then the membership of m fault symptoms to n fault causes forms an m × n evaluation matrix R.
In the formula r ij Is the symptom of failure u i For the fault reason v j The degree of membership of (c) satisfies 0. Ltoreq. R ij ≤1。
In the present invention, each failure symptom u i Comparing the fault causes possibly causing the fault symptoms pairwise, and using the calibration number table shown in the following tableShowing the magnitude relation of the fault symptom to the degree of membership of the fault reason. For each fault symptom u i Constructing a pairwise comparison matrix C i 。
TABLE 3 pairwise comparison matrix element scaling method
In the invention, pairwise comparison matrixes are respectively constructed by a plurality of experts, and after the pairwise comparison matrixes are constructed, the sum method, the root method or the power method is used for solving the membership degree of the fault symptom to the fault reason.
In the present example, the fault symptom u showing a low value is indicated for the pressure gauge 1 Constructing a two-by-two comparison matrix thereof,
calculated fault symptoms u with low pressure gauge display value 1 Membership degree vector r for various fault reasons 1
r 1 ＝(0.03,0.43,0.11,0.32,0.11)
Similarly, the fault symptom u of the smaller force output by the hydraulic cylinder is calculated 2 Membership vector r for various fault causes 2
r 2 ＝(0.02,0.41,0.26,0.28,0.03)
Systematic screaming and loud fault symptoms u 3 For various reasonsMembership vector r of barrier cause 3
r 3 ＝(0.39,0.21,0.11,0.09,0.20)
Fault symptom u of slightly high pump body temperature 4 Membership vector r for various fault causes 4
r 4 ＝(0.31,0.38,0.14,0.04,0.13)
(5) Determining a probability vector B of a failure of an associated component 1
Probability vector B of fault of part 1 ＝(b 11 ,b 12 ,…,b 1n )，b 11 ,b 12 ,…,b 1n Indicating a fault v 1 ,v 2 ,…,v n The size of the probability of existence, n is the number of faults.
On the basis of comprehensively considering the severity of the fault symptoms and the difficulty of obtaining the fault symptoms on site, the probability vector B of the fault of the related parts is obtained by the following formula 1 。
B 1 ＝(k 1 A+k 2 F)○R
Above formula k 1 And k 2 For vector A and vector F, the weighting coefficients, k in the present invention 1 Take 0.8,k 2 Taking 0.2, O is an operation rule, and the invention adopts a weighted average method.
In this example, a = {0.7,0.3,0.7,0.3}, F = (0.3,0.7,0.5,0.5)
Step 2, determining inherent fault rate vector B of related parts 2
(1) According to historical data, sorting the fault interval time from small to large, and establishing a fault interval time table
(2) According to the fault interval time table, estimating the reliability by a median rank formula
R(t i )＝1-(i-0.3)/(n+0.4)
Wherein t is 1 ,t 2 ,…,t n Is time to failure, and t 1 ≤t 2 ≤…≤t n N is the total number of acquisition faults
Take the overflow valve as an example to illustrate how to estimate the failure rate. Counting the operation data of a certain type of machine tool from 1 month 1 day in 2012 to 30 months in 2013, and the following table 4 shows the fault records of the overflow valve, wherein x and y are the coordinate values of discrete data points (because the sample is larger, only partial processing results are given here)
TABLE 4 Fault data and its Weibull conversion Table
(3) Drawing a Weibull probability chart according to the estimated reliability value
Let x = lnt i ，y＝ln[-lnR(t i )]Thus (t) 1 ,R(t 1 )),(t 2 ,R(t 2 )),…,(t n ,R(t n ) The data set is transformed into ((x) 1 ,y 1 ),(x 2 ,y 2 ),…,(x n ,y n ) Data points are plotted on a Weibull probability map to obtain a Weibull probability map.
The coordinate values of all the data scatter points in Table 4 are plotted on Weibull Probability Paper (WPP), as shown in FIG. 2.
(4) On Weibull frequency paper, the left part of the graph is fitted with a straight line L 1 To obtain a straight line L 1 Has a slope and an intercept on the y-axis of k 1 And b 1 Fitting an asymptote L at the right end of the graph 2 To obtain a straight line L 2 Has a slope and an intercept on the y-axis of k 2 And b 2 。
As shown in FIG. 2, a straight line L is fitted to the left part of the graph 1 The slope and the intercept on the y-axis are k 1 =0.763 and b 1 ＝-5.564
At the right end of the graph, an asymptote L is fitted 2 The slope and the intercept on the y-axis are k 2 =1.303 and b 2 ＝-9.029
(5) And determining the reliability function according to the double segmentation Weibull function.
Wherein
α 1 ＝exp(-b 1 /k 1 )
β 1 ＝k 1
α 2 ＝exp(-b 2 /k 2 )
β 2 ＝k 2
α 1 ＝exp(-b 1 /k 1 )＝exp((5.564/0.763)＝1468.9
β 1 ＝k 1 ＝0.763
α 2 ＝exp(-b 2 /k 2 )＝exp(9.029/1.303)＝1021.9
β 2 ＝k 2 ＝1.303
Finally, the reliability function of the double Weibull segmental model of the overflow valve is obtained as
In this example, the relief valve operates for 305 hours with a reliability of 0.7145 according to the above formula
The reliability functions of the oil pump, oil filter, seal and oil pump drive can be determined in the same way.
(6) Determining failure rates inherent to the components themselves
For the set of failure causes V = { V = { V = } 1 ,v 2 ,…,v n Determining the failure rate b of each unit in the failure cause set 2i ，i＝1,2,3,…,n。
b 21 ＝1-R(t)＝1-0.7145＝0.2855
(7) By using the method, the inherent fault rate vector B of the related part is determined 2
B 2 ＝(b 21 ,b 22 ,..b 2n )
B 2 ＝(0.2855,0.6791,0.2674,0.31,0.302)
And 3, performing weighted operation according to the probability of the fault of the part determined based on the fault symptom in the step 1 and the inherent fault rate of the part determined in the step 2, and determining the comprehensive probability of the fault of the part.
(1) Normalizing the probability vector B of a failure of a relevant component determined from the symptoms of the failure 1
B 1 = (0.3890,0.6902,0.2872,0.3778,0.2558) after normalization processing,
B 1 ＝(0.1945,0.3451,0.1436,0.1889,0.1279)
(2) Normalizing fault rate vector B of related parts 2
B 2 ＝(0.2855,0.6791,0.2674,0.31,0.302) are normalized,
B 2 ＝(0.1548,0.3683,0.1450,0.1681,0.1638)
(3) Probability vector B of fault of related parts determined according to fault symptoms 1 And fault rate vector B of related parts 2 Performing weighted average calculation to obtain the comprehensive fault probability vector B of the parts, i.e.
B＝k 3 B 1 +k 4 B 2
In the above formula k 3 And k 4 Are respectively vector B 1 Sum vector B 2 In the present invention, k is 3 Take 0.5,k 4 Take 0.5.
B＝0.5×(0.1548,0.3683,0.1450,0.1681,0.1638)+0.5×(0.1945,0.3451,0.1436,0.1889,0.1279)
＝(0.1747,0.3567,0.1442,0.1785,0.1459)。
And 4, determining the fault reason according to the comprehensive fault probability of the parts.
Sequencing elements in the comprehensive fault probability vector B from big to small: b2> b4> b1> b5> b3, it can be seen that the most likely cause of system pressure deficiency is severe wear of the internal parts of the oil pump.
Claims (2)
1. A mechanical equipment fault diagnosis method comprises the following steps:
1) Determining the probability vector B of the fault of the part according to the fault symptoms 1 ；
2) Determining the inherent failure rate vector B of the part according to the statistical part failure historical data 2 ；
3) Determining the probability vector B of the fault of the part according to the step 1) 1 And step 2) determining the fault rate vector B inherent to the part per se 2 Carrying out weighting operation to determine a comprehensive probability vector B of the fault of the part;
4) Determining the fault reason according to the magnitude of the comprehensive probability of the fault of the part;
characterized in that the probability vector B for determining the fault of the part according to the fault symptoms in the step 1) is 1 The method comprises the following steps:
(1) Respectively establishing a fault symptom set U and a fault reason set V according to the following formula;
fault symptom set U = { U = 1 ，u 2 ，…，u m In which u 1 ，u 2 ，…，u m The fault symptoms of the equipment are shown, and m is the number of the fault symptoms;
set of causes of failure V = { V = { (V) } 1 ,v 2 ,…,v n In which v is 1 ,v 2 ,…,v n Indicating fault reasons, wherein n is the number of the fault reasons;
(2) Determining a fault symptom significance vector A
According to the significance degree of the fault symptoms, scoring each fault symptom to form a fault symptom significance degree vector A = (a) 1 ，a 2 ，…，a m ) Wherein a is 1 ，a 2 ，…，a m The obvious degree of each fault symptom is shown, and the value of the score is taken according to the following table 1;
TABLE 1 Scale of severity of symptom of failure
(3) Determining a difficulty vector F for field acquisition of symptoms of a fault
TABLE 2 assessment of the difficulty of on-site acquisition of symptoms of a failure
According to the difficulty degree obtained by the fault symptom, determining a difficulty degree vector F obtained on the site of the fault symptom according to the table
F＝(f 1 ,f 2 ,…,f m )
Wherein f is 1 ，f 2 ，…，f m Indicating a fault symptom u of the device 1 ，u 2 ，…，u m The difficulty of obtaining, m is the number of symptoms of the failure;
(4) Determining a judgment matrix R
Establishing a judgment matrix R and determining fault symptoms u i For the fault reason v j Degree of membership r ij Then the membership degrees of the m fault symptoms to the n fault reasons form an m multiplied by n order judgment matrix R;
in the above formula r ij Is the symptom of failure u i For the fault reason v j The degree of membership of (c) satisfies 0. Ltoreq. R ij ≤1；
For each fault symptom u i Comparing every two fault causes which may cause the fault symptom, and adopting a standard number shown in the following table 3 to represent the magnitude relation of the fault symptom to the membership degree of the fault causes; for each fault symptom u i Constructing a pairwise comparison matrix C i ；
TABLE 3 pairwise comparison matrix element scaling method
Constructing pairwise comparison matrixes by multiple experts respectively, and solving the membership degree of fault symptoms to fault reasons by using a root method after constructing pairwise comparison matrixes;
(5) Determining probability vector B of fault of related parts 1
Probability vector B of fault of part 1 ＝(b 11 ,b 12 ,…,b 1n ) Wherein b is 11 ,b 12 ,…,b 1n Indicates the cause of the failure v 1 ,v 2 ,…,v n The existence probability is large, and n is the number of fault reasons;
on the basis of comprehensively considering the severity of the fault symptoms and the difficulty of obtaining the fault symptoms on site, the probability vector B of the fault of the related parts is obtained by using the following formula 1
B 1 ＝(k 1 A+k 2 F)○R
In the above formula, k 1 And k 2 Is vector A and vector F weight coefficients, where k 1 Take 0.8,k 2 Taking 0.2, and O represents an operation rule, namely a weighted average method;
step 2) determining the inherent failure rate vector B of the part according to the statistical part failure historical data 2 The method comprises the following steps:
(1) According to historical data, sorting the fault interval time from small to large, establishing a fault interval time table,
(2) Calculating the reliability R (t) of the fault according to the fault interval time table by using a median rank formula i )：
R(t i )＝1-(i-0.3)/(n+0.4)
Wherein t is 1 ,t 2 ,…,t n Is a fault interval time, and t 1 ≤t 2 ≤…≤t n N is the total number of acquisition faults,
(3) Drawing a Weibull probability chart according to the calculated reliability value
Let x = lnt i ，y＝ln[-lnR(t i )]Will (t) 1 ,R(t 1 )),(t 2 ,R(t 2 )),…,(t n ,R(t n ) Conversion of data set to ((x) 1 ,y 1 ),(x 2 ,y 2 ),…,(x n ,y n ) Data points of the data set are plotted on a Weibull probability map to obtain a Weibull probability map;
(4) Fitting the left part of the figure of the Deweibull probability chart obtained in the step (3) to a straight line L on Weibull frequency paper 1 To obtain a straight line L 1 Has a slope and an intercept on the y-axis of k 1 And b 1 Fitting an asymptote L at the right end of the graph 2 To obtain a straight line L 2 Has a slope and an intercept on the y-axis of k 2 And b 2 ；
(5) Determining a reliability function R (t) according to the double segmentation Weibull function;
wherein
α 1 ＝exp(-b 1 /k 1 )
β 1 ＝k 1
α 2 ＝exp(-b 2 /k 2 )
β 2 ＝k 2
(6) Determining the failure rate inherent to the component itself
V = { V) for set of faults 1 ,v 2 ,…,v n Determining the failure rate b of each unit in the failure set 2i ，i＝1,2,3,…,n；
b 2i ＝1-R(t)
(7) By using the method, the inherent fault rate vector B of the related part is determined 2
B 2 ＝(b 21 ,b 22 ,..b 2n )；
The method for determining the comprehensive probability vector B of the fault of the part in the step 3) comprises the following steps:
(1) Normalizing the probability vector B of a failure of a relevant component determined from the symptoms of the failure 1 ，
(2) Normalizing fault rate vector B of related parts 2 ，
(3) The normalized probability vector B of the fault of the related parts determined according to the fault symptoms 1 And normalized fault rate vector B of related parts 2 Performing weighted average calculation to obtain the comprehensive probability vector B of the fault of the part, namely
B＝k 3 B 1 +k 4 B 2
In the above formula k 3 And k 4 Are respectively vector B 1 Sum vector B 2 Wherein k is 3 Take 0.5,k 4 Take 0.5.
2. The method for diagnosing the faults of the mechanical equipment according to claim 1, wherein the method for solving the membership degree of the fault symptoms to the fault reasons further comprises a sum method or a power method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610095084.0A CN105760672B (en) | 2016-02-22 | 2016-02-22 | A kind of Trouble Diagnostic Method of Machinery Equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610095084.0A CN105760672B (en) | 2016-02-22 | 2016-02-22 | A kind of Trouble Diagnostic Method of Machinery Equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105760672A CN105760672A (en) | 2016-07-13 |
| CN105760672B true CN105760672B (en) | 2018-04-24 |
Family
ID=56330900
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610095084.0A Expired - Fee Related CN105760672B (en) | 2016-02-22 | 2016-02-22 | A kind of Trouble Diagnostic Method of Machinery Equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105760672B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10635094B2 (en) * | 2016-09-16 | 2020-04-28 | Siemens Aktiengesellschaft | Failure models for embedded analytics and diagnostic/prognostic reasoning |
| CN108858194B (en) * | 2018-07-10 | 2020-10-27 | 华北水利水电大学 | Control method and device of Boolean network robot |
| CN112641129A (en) * | 2020-12-22 | 2021-04-13 | 红云红河烟草(集团)有限责任公司 | Method for pre-judging cigarette machine fault by using cigarette automatic monitoring data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462846A (en) * | 2014-12-22 | 2015-03-25 | 山东鲁能软件技术有限公司 | Intelligent device failure diagnosis method based on support vector machine |
| CN105158598A (en) * | 2015-08-15 | 2015-12-16 | 国家电网公司 | Fault prediction method suitable for power equipment |
| CN105224782A (en) * | 2014-10-16 | 2016-01-06 | 华北电力大学 | A kind of converting equipment probability of malfunction computing method based on fault mode |
-
2016
- 2016-02-22 CN CN201610095084.0A patent/CN105760672B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105224782A (en) * | 2014-10-16 | 2016-01-06 | 华北电力大学 | A kind of converting equipment probability of malfunction computing method based on fault mode |
| CN104462846A (en) * | 2014-12-22 | 2015-03-25 | 山东鲁能软件技术有限公司 | Intelligent device failure diagnosis method based on support vector machine |
| CN105158598A (en) * | 2015-08-15 | 2015-12-16 | 国家电网公司 | Fault prediction method suitable for power equipment |
Non-Patent Citations (4)
| Title |
|---|
| 基于多源信息的延时约束加权模糊Petri网故障诊断模型;吴文可 等;《电力系统自动化》;20131225;第37卷(第24期);第43-53页 * |
| 基于模糊随机Petri网的可重构制造系统可靠性分析;齐继阳 等;《组合机床与自动化加工技术》;20150430(第4期);第156-160页 * |
| 电网故障诊断的智能方法综述;边莉 等;《电力系统保护与控制》;20140201;第42卷(第3期);第146-153页 * |
| 融合集对分析和关联规则的变压器故障诊断方法;谢龙君 等;《中国电机工程学报》;20150120;第35卷(第2期);第279-280页第2.1-2.3节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105760672A (en) | 2016-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102011576B (en) | Method for hierarchically diagnosing fault of rod-comprising oil pumping system based on indicator diagram | |
| DE102022201761A1 (en) | Method, system and storage medium for automatically diagnosing devices | |
| CN110738274A (en) | nuclear power device fault diagnosis method based on data driving | |
| CN109489977B (en) | KNN-AdaBoost-based bearing fault diagnosis method | |
| CN101689051B (en) | abnormal factor specifying method and system | |
| CN105760672B (en) | A kind of Trouble Diagnostic Method of Machinery Equipment | |
| Zhou et al. | Latent degradation indicators estimation and prediction: A Monte Carlo approach | |
| CN102252843B (en) | Assessment method for rolling bearing performance variation | |
| CN116226646B (en) | Method, system, equipment and medium for predicting health state and residual life of bearing | |
| CN113834657A (en) | Bearing fault early warning and diagnosis method based on improved MSET and frequency spectrum characteristics | |
| CN113095402B (en) | Code input-based generation countermeasure network fault detection method and system | |
| CN104008294B (en) | System and method for detecting abnormality of bearing | |
| CN110990788A (en) | Bearing residual life prediction method based on ternary wiener process | |
| CN112904810B (en) | Process industry nonlinear process monitoring method based on effective feature selection | |
| EP3948438B1 (en) | Method and system for anomaly detection and diagnosis in industrial processes and equipment | |
| WO2022147853A1 (en) | Complex equipment power pack fault prediction method based on hybrid prediction model | |
| CN110008898A (en) | Industrial equipment data edges processing method based on symbol and convolutional neural networks | |
| WO2019043600A1 (en) | Remaining useful life estimator | |
| WO2021114320A1 (en) | Wastewater treatment process fault monitoring method using oica-rnn fusion model | |
| CN112000081B (en) | Fault monitoring method and system based on multi-block information extraction and Mahalanobis distance | |
| CN110555049A (en) | fault reason suspected degree analysis method based on measurement abnormality correlation degree model | |
| CN112598144A (en) | CNN-LSTM burst fault early warning method based on correlation analysis | |
| CN110333694A (en) | Numerical control device method for diagnosing faults based on Fuzzy Petri Net | |
| CN117032102A (en) | Equipment fault prediction method and system based on big data | |
| Yacout | Logical analysis of maintenance and performance data of physical assets, ID34 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20191203 Address after: 315040 Zhejiang city of Ningbo province high tech Zone Juxian 587 Lane Road No. 15 Building 2 building 033 17-1 Patentee after: CHINA E-TECH (NINGBO) MARITIME ELECTRONICS RESEARCH INSTITUTE Co.,Ltd. Address before: Meng Xi Road 212003 Zhenjiang city of Jiangsu province Jingkou District No. 2 Patentee before: Jiangsu University of Science and Technology |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180424 |