CN113486742A - Fault identification method, device and system and computer readable storage medium - Google Patents
Fault identification method, device and system and computer readable storage medium Download PDFInfo
- Publication number
- CN113486742A CN113486742A CN202110697516.6A CN202110697516A CN113486742A CN 113486742 A CN113486742 A CN 113486742A CN 202110697516 A CN202110697516 A CN 202110697516A CN 113486742 A CN113486742 A CN 113486742A
- Authority
- CN
- China
- Prior art keywords
- fault
- calculating
- sample set
- weight
- mahalanobis distance
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 102
- 230000008569 process Effects 0.000 claims abstract description 23
- 230000005540 biological transmission Effects 0.000 claims abstract description 17
- 239000011159 matrix material Substances 0.000 claims description 37
- 238000004364 calculation method Methods 0.000 claims description 35
- 239000013598 vector Substances 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 10
- 239000002131 composite material Substances 0.000 claims description 3
- 150000001875 compounds Chemical class 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 abstract description 10
- 230000000694 effects Effects 0.000 abstract description 6
- 108010076504 Protein Sorting Signals Proteins 0.000 description 7
- 230000008901 benefit Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000035772 mutation Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012850 discrimination method Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Algebra (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides a fault identification method, a device, a system and a computer readable storage medium, comprising the following steps: acquiring a fault sample set; calculating the optimal value of the objective weight beta of the fault sample set based on an information entropy method; calculating the optimal value of the subjective weight value alpha of the fault sample set based on an analytic hierarchy process; calculating the combined weight of the fault sample set
(ii) a Calculating a minimum combined weighted mahalanobis distance value of the fault sample set based on a combined weighted mahalanobis distance method; and identifying the fault type of fault vibration generated by the gear teeth when the gearbox or the series-parallel gear transmission system runs by using the calculated minimum combined weighted Mahalanobis distance value according to the basic principle that the minimum combined weighted Mahalanobis distance value among the similar fault characteristics is minimum. The fault characteristic information identification algorithm in the invention can be used for aligning teethEarly failure of gears in a gearbox or a series-parallel gear transmission system generates an early warning effect to ensure safe operation of a gear system.
Description
Technical Field
The present invention relates to the field of fault feature identification and diagnosis technologies, and in particular, to a fault identification method, apparatus, system, and computer-readable storage medium.
Background
At present, at the beginning of a fault (early fault) of mechanical equipment, fault characteristic information is submerged by interference information of the surrounding environment and is easily ignored by equipment operation monitoring personnel. Early failures resulting in mechanical equipment are not effectively identified, further worsening the failure until a major failure occurs and the machine is shut down, or a major casualty event is caused, which has resulted in irreparable economic loss. Therefore, an effective method must be adopted to completely and accurately identify the early fault characteristic information of the equipment from various interference information, and remind the monitoring personnel of the equipment to adjust and maintain fault parts in the equipment in time, so that the occurrence probability of accidents is reduced.
At present, the most effective method for identifying early weak fault features is the mahalanobis distance method, and the characteristics of the fault are judged according to the vibration state. The method adopts the covariance distance between data to calculate the set similarity of two unknown samples and characteristic samples, and when the covariance distance between the unknown samples and the characteristic sample set is smaller, the similarity between the unknown samples and the characteristic samples is larger. Conversely, the smaller the degree of similarity. Further, the mahalanobis distance discrimination method has characteristics such as no linear distortion (regardless of the measurement unit) and no scale, and has a wide application space in actual fault identification and diagnosis using these characteristics. The method takes the mutual relation among the indexes of the characteristic elements into consideration, and can also eliminate the interference of the correlation among the characteristic variables. And moreover, the Mahalanobis distance calculation gets rid of the influence of various parameter dimensions, so that the Mahalanobis distance calculation is independent of the dimension of the characteristic measurement, and therefore the Mahalanobis distance calculation method has the advantages of being high in accuracy and the like of the fault characteristic diagnosis result. However, in the process of discrimination, the importance of samples with weak functions is exaggerated, so that the final calculation result loses practical guiding significance. The problem with this algorithm is determined by the nature of the algorithm itself. Therefore, to solve this problem, the mahalanobis distance algorithm needs to be improved accordingly.
In order to solve the above algorithm, the problem of importance of a relatively weak sample is exaggerated in the process of discrimination. The invention adopts an information entropy method and an analytic hierarchy process to carry out combined weighting on the importance degree of each sample according to a fault sample set so as to overcome the defect that the Markov distance method is insufficient in the importance division of the samples in the fault distinguishing process, and develop a set of complete methods capable of realizing the identification of the weak fault characteristics of the gear.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the existing mechanical equipment fault weak vibration feature identification technology, the invention provides a fault identification method, which is used for effectively identifying the mechanical equipment fault weak vibration feature from random vibration information and interference information of a dynamic system and ensuring the accuracy of feature information identification.
(II) technical scheme
The invention provides a fault identification method, which comprises the following steps:
acquiring a fault sample set; the fault sample set is a fault sample set of fault vibration generated by gear teeth when a gearbox or a parallel-serial gear transmission system operates; calculating the optimal value of the objective weight beta of the fault sample set based on an information entropy method; calculating the optimal value of the subjective weight value alpha of the fault sample set based on an analytic hierarchy process; calculating the combined weight of the fault sample set
Calculating a minimum combined weighted mahalanobis distance value of the fault sample set based on a combined weighted mahalanobis distance method; and identifying the fault type of fault vibration generated by the gear teeth when the gearbox or the series-parallel gear transmission system runs by using the calculated minimum combined weighted Mahalanobis distance value according to the basic principle that the minimum combined weighted Mahalanobis distance value among the similar fault characteristics is minimum.
Further, the method for calculating the optimal value of the subjective weight value alpha of the fault sample set based on the analytic hierarchy process comprises the following steps:
building a structural model:
establishing a structural model of a decision problem according to the characteristic information of the fault sample set;
obtaining weight set vectors of each layer: weight set A of decision target layer1＝[Au1,Au2,Au3]T(ii) a Weight set Au of decision criterion layer1＝[u11,u12,u13]T，Au2＝[u21,u22,u23]T，Au3＝[u31,u32,u33]T；
The subjective weight calculation process comprises the following steps:
and calculating a comparison matrix P of each layer in the following way:
the priority vectors W of the respective layers of the respective decision matrices are calculated and normalized,
in the formula, n is the dimension of the corresponding judgment matrix;
respectively for each judgment matrix PiThe consistency check is performed with the following formula CR:
in the formula, λmaxIs the maximum characteristic root of the corresponding judgment matrix, n is the dimension of the corresponding judgment matrix, and RI is the average random consistency index;
a composite priority vector for the structural model is calculated as shown in the following table:
calculating an optimal subjective weight value alpha:
further, the step of calculating the optimal objective weight value β of each indicator element based on an information entropy method includes the following steps:
assume n samples, an original matrix A with m decision indexes, and A ═ A1,A2,…,Ai,…,Am]TIf so, the attribute value corresponding to the jth index of the ith sample in the matrix A is aij; the original matrix A is calculated by the following method:
determining a forward indicator A of an original matrix A+And a negative index A-And normalizing the data, wherein the calculation method of the forward index set comprises the following steps:
the negative index set calculation method comprises the following steps:
calculating the specific gravity p occupied by the ith sample in the jth indexijThe calculation method is shown as the following formula:
calculating the information entropy e of the jth indexjThe calculation method is shown as the following formula:
calculating an objective weight value beta of the jth index of the sample according to the following formula:
further, the combined weight of the fault sample set is calculated
The calculation method is shown as the following formula:
in the formula, alpha is a subjective weight value, beta is an objective weight value, lambda is a preference factor, lambda is more than or equal to 0 and less than or equal to 1, when lambda is small, the combination weight is biased to alpha, and when lambda is large, the combination weight is biased to beta.
Further, the calculation process of calculating the minimum combined weighted mahalanobis distance value of each fault sample set includes the following steps:
calculating the Mahalanobis distance value D of the fault characteristic sampleMahaThe calculation method is shown as the following formula:
wherein, muYAs mean vector of fault features, CYA covariance matrix that is a fault feature;
calculating a combined weighted mahalanobis distance of fault feature samples
On the basis of the Mahalanobis distance, the fault feature samples are subjected to combined weighting to obtain a combined weighted Mahalanobis distance value
The calculation method is shown as the following formula:
in the formula (I), the compound is shown in the specification,
as a weight matrix, the weight matrix is,
is the combined weight.
The invention provides a fault identification method and a device thereof, comprising:
an acquisition module that acquires a fault sample set; the fault sample set is a fault sample set of fault vibration generated by gear teeth when a gearbox or a parallel-serial gear transmission system operates;
the first calculation module calculates the optimal value of the objective weight beta of the fault sample set based on an information entropy method;
the second calculation module is used for calculating the optimal subjective weight value alpha of the fault sample set based on an analytic hierarchy process;
a third calculation module for calculating a combined weight of the failure sample set
The fourth calculation module calculates a minimum combined weighted mahalanobis distance value of the fault sample set based on a combined weighted mahalanobis distance method;
and the identification module identifies the fault type of fault vibration generated by the gear teeth when the gearbox or the series-parallel gear transmission system operates by using the calculated minimum combined weighted Mahalanobis distance value according to the basic principle that the minimum combined weighted Mahalanobis distance value among the similar fault characteristics is minimum.
The invention provides a fault identification system, which comprises a memory and a processor; the memory for storing a computer program; when the processor executes the computer program, a fault identification method as described above is implemented.
The present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a fault identification method as described above.
(III) advantageous effects
The invention provides a fault identification method, which comprises the following steps: acquiring a fault sample set; the fault sample set is a fault sample set of fault vibration generated by gear teeth when a gearbox or a parallel-serial gear transmission system operates; calculating the optimal value of the objective weight beta of the fault sample set based on an information entropy method; calculating the optimal value of the subjective weight value alpha of the fault sample set based on an analytic hierarchy process; calculating the combined weight of the fault sample set
Calculating a minimum combined weighted mahalanobis distance value of the fault sample set based on a combined weighted mahalanobis distance method; and identifying the fault type of fault vibration generated by the gear teeth when the gearbox or the hybrid gear transmission system operates according to the minimum combined weighted Mahalanobis distance value. The whole set of fault characteristic information identification algorithm can generate an early warning effect on the early fault of the gear in the gear box or the series-parallel gear transmission system so as to ensure the safe operation of the gear system.
Compared with the prior art for identifying fault characteristic information by a weighted Mahalanobis distance method, the method has the beneficial effects that: the algorithm is most sensitive to weak mutation of mechanical dynamics behaviors, can effectively identify weak dynamic characteristic information from various interference information, has the minimum influence on effective characteristic information, and can ensure the accuracy and effectiveness of identification of various characteristic information. In addition, the gear in the gear box can effectively identify fault characteristic information at the initial stage of fault generation, and can give out corresponding early warning prompt, and the early warning performance of the gear can meet the requirements of practical application.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.
Fig. 1 is a schematic diagram of a fault sample set decision structure model establishment of a fault identification method in an embodiment of the present invention.
Fig. 2 is a flowchart illustrating a method for calculating subjective weight values of a fault sample set by using an analytic hierarchy process according to an embodiment of the present invention.
Fig. 3 is a flowchart of calculating objective weight values of a fault sample set by an information entropy method in a fault identification method according to an embodiment of the present invention.
Fig. 4 is a flowchart of calculating a minimum combined weighted mahalanobis distance value of a fault sample set by a combined weighted mahalanobis distance method according to a fault identification method in an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
Aiming at the defects of the current weak vibration feature identification technology of mechanical equipment faults, the invention provides a fault identification method, in particular to a fault identification method of combined weighted Mahalanobis distance based on information entropy and hierarchical analysis, which can effectively identify the weak vibration features of the mechanical equipment faults from random vibration information and interference information of a dynamic system and can ensure the accuracy of feature information identification.
The invention provides a fault identification method, which comprises the following steps:
1. determining subjective weight values by adopting an analytic hierarchy process:
the method is a decision-making method which decomposes a problem to be decided and related factors into a target layer, a criterion layer and a scheme layer, and performs qualitative analysis and quantitative analysis on the problem to be decided according to the three layers. On the basis of the analysis, the decision problem is mathematized by using less quantitative information, and a simple decision method can be provided for complex decision problems with multiple targets, multiple criteria or no structural characteristics.
In the invention, the subjective weight value of the characteristic sample set is determined by adopting an analytic hierarchy process, and the method comprises the following steps:
(1) building a structural model:
and establishing a structural model of the decision problem according to the characteristics of the fault sample set, as shown in FIG. 1.
The weight set vector of each level can be obtained from the graph: weight set a of decision target layer (decision target layer)1＝[Au1,Au2,Au3]T；
Weight set Au of decision rule layer (decision rule layer)1＝[u11,u12,u13]T，Au2＝[u21,u22,u23]T，Au3＝[u31,u32,u33]T。
The flow of calculating the subjective weight value is shown in fig. 2.
(2) Determining a comparison matrix P for each layer:
and (4) calculating a comparison matrix P of each layer by adopting an equation (1-1) according to the weight matrix in the step (1).
(3) Calculating each judgment matrix respectively, and normalizing the priority vector W of each layer, as shown in formula (1-2):
in the formula, n is the dimension of the corresponding judgment matrix.
(4) The consistency check is performed on each judgment matrix p (i) respectively, and the calculation formula CR is shown in the formula (1-3):
in the formula, λmaxIs the maximum characteristic root of the corresponding judgment matrix, n is the dimension of the corresponding judgment matrix, and RI is the average random consistency index.
(5) A composite priority vector for the structural model is calculated as shown in the following table:
(6) calculating an optimal subjective weight value alpha as shown in the formula (1-4):
2. determining objective weight values by using an information entropy method:
and calculating the objective weight value of the decision sample by adopting an information entropy method according to the distribution characteristic of the sample characteristic value.
When a certain attribute in a sample is stable and unchanged, the entropy of the attribute information is 1, and the weight (value) of the attribute is 0.
When the weight is 0, the influence degree of the sample attribute on the importance of the system decision result is negligible.
Conversely, when a certain attribute in a sample is more complicated to change, the smaller the value of the attribute information entropy is, the greater the weight of the attribute is. When the randomness of the set index is higher, the information entropy value is larger, and the importance degree of the index on the decision sample weight is higher. Conversely, the lower the importance level.
The calculation flow chart of the method is shown in FIG. 3.
In the invention, the method for determining the objective weight value of the characteristic sample set by adopting the information entropy method comprises the following steps:
(1) assume n samples, an original matrix A with m decision indexes, and A ═ A1,A2,…,Ai,…,Am]TAs shown in formula (2-1). The attribute value corresponding to the j index of the ith sample in the matrix A is aij。
(2) Determining a forward indicator A of a matrix A+And a negative index A-And normalizing the data to obtain a mathematical expression as shown in the formulas (2-2) and (2-3):
forward direction index set:
the negative index set calculation method comprises the following steps:
(3) calculating the specific gravity p occupied by the ith sample in the jth indexijAs shown in formulas (2-4):
(4) calculating the information entropy e of the jth indexjAs shown in formulas (2-5):
(5) calculating an objective weight value beta of the sample, as shown in formula (2-6):
3. calculating combined weight
Respectively adopting an information entropy method and an analytic hierarchy process to calculate a subjective weight value and an objective weight value of the fault sample set, and then adopting a method of formula (3-1) to calculate a combined weight value of the fault sample set
In the formula, alpha is a subjective weight value, beta is an objective weight value, lambda is a preference factor, lambda is more than or equal to 0 and less than or equal to 1, when lambda is small, the combination weight is biased to alpha, and when lambda is large, the combination weight is biased to beta.
4. Calculating the mahalanobis distance of the fault feature sample:
according to the Mahalanobis distance principle, the Mahalanobis distance value D of each sample set is calculated by adopting the formula (4-1)Maha：
In the formula, μ Y is a mean vector of the fault features, and CY is a covariance matrix of the fault features.
5. MeterCalculating combined weighted mahalanobis distance of fault feature samples
On the basis of the Mahalanobis distance, the fault feature samples are subjected to combined weighting to obtain a combined weighted Mahalanobis distance value
The calculation method is shown as the formula (5-1) and the formula (5-2):
in the formula (I), the compound is shown in the specification,
as a weight matrix, the weight matrix is,
is the combined weight.
The invention provides a detection device of a fault identification method, which writes the fault identification method into the detection device in a software form, and accurately identifies fault characteristic information from various noise fields when the detection device detects weak fault vibration generated by gear teeth in a gear box or a series-parallel gear transmission system.
The invention provides a fault identification device, which comprises a memory and a processor; the memory for storing a computer program; the processor is configured to implement a fault identification method as described above when executing the computer program.
The Memory mentioned in the above fault identification device may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.
The present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a fault identification method as described above. The computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk. In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
The invention relates to a fault identification method based on information entropy and hierarchical analysis and combined weighting Mahalanobis distance, and a system signal characteristic identification process is shown in figure 4.
(1) Resetting the system configuration:
before signal feature recognition, parameters in software and hardware are reset, parameter memory in the software and the hardware is eliminated, and influences of the software and the hardware parameters on feature information recognition effects are reduced.
(2) Calculating the subjective weight value alpha of the actual signal sequence:
according to the invention, an analytic hierarchy process is adopted, accurate calculation basis is provided for the subjective weight value alpha of the signal sequence, and the problem of low accuracy of calculating the subjective weight value alpha by adopting a traditional method is solved, so that the adverse effect of judging the importance degree of each element in the signal sequence by the weight value alpha is reduced.
(3) Calculating an objective weight value beta of the actual signal sequence:
the information entropy method used in the invention has the advantage of objectively reflecting the real condition of the importance degree of each sequence in the fault characteristic signal and providing accurate decision, and can provide a basis for calculating the optimal objective weight value beta for the fault signal sequence.
(4) Calculating the combination of the subjective weight value alpha and the objective weight value beta:
on the basis of the steps (2) and (3), the optimal weight values of alpha and beta of the actual signal sequence are calculated, and then the combined weight value of the fault sequence is calculated according to the formula (3-1)
(5) Calculating a combined weighted mahalanobis distance value for a fault signal sequence
On the basis of the above, the combined weighted Mahalanobis distance value of the fault signal is calculated by adopting the formula (5-1)
And according to
And judging the fault feature type.
The invention provides a fault identification method, which comprises the following steps: acquiring a fault sample set; the fault sample set is a fault sample set of fault vibration generated by gear teeth when a gearbox or a parallel-serial gear transmission system operates; calculating the optimal value of the objective weight beta of the fault sample set based on an information entropy method; calculating the optimal value of the subjective weight value alpha of the fault sample set based on an analytic hierarchy process; calculating the combined weight of the fault sample set
Calculating a minimum combined weighted mahalanobis distance value of the fault sample set based on a combined weighted mahalanobis distance method; and identifying the fault type of fault vibration generated by the gear teeth when the gearbox or the hybrid gear transmission system operates according to the minimum combined weighted Mahalanobis distance value. The whole set of fault characteristic information identification algorithm can generate an early warning effect on the early fault of the gear in the gear box or the series-parallel gear transmission system so as to ensure the safe operation of the gear system.
Compared with the prior art for identifying fault characteristic information by a weighted Mahalanobis distance method, the method has the beneficial effects that: the algorithm is most sensitive to weak mutation of mechanical dynamics behaviors, can effectively identify weak dynamic characteristic information from various interference information, has the minimum influence on effective characteristic information, and can ensure the accuracy and effectiveness of identification of various characteristic information. In addition, the gear in the gear box can effectively identify fault characteristic information at the initial stage of fault generation, and can give out corresponding early warning prompt, and the early warning performance of the gear can meet the requirements of practical application.
Compared with the prior art for identifying fault characteristic information by a weighted Mahalanobis distance method, the method has the beneficial effects that: the algorithm is most sensitive to weak mutation of mechanical dynamics behaviors, can effectively identify weak dynamic characteristic information from various interference information, has the minimum influence on effective characteristic information, and can ensure the accuracy and effectiveness of identification of various characteristic information. In addition, the gear in the gear box can effectively identify fault characteristic information at the initial stage of fault generation, and can give out corresponding early warning prompt, and the early warning performance of the gear can meet the requirements of practical application.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (8)
1. A method of fault identification, comprising:
acquiring a fault sample set; the fault sample set is a fault sample set of fault vibration generated by gear teeth when a gearbox or a parallel-serial gear transmission system operates;
calculating the optimal value of the objective weight beta of the fault sample set based on an information entropy method;
calculating the optimal value of the subjective weight value alpha of the fault sample set based on an analytic hierarchy process;
calculating the combined weight of the fault sample set
Calculating a minimum combined weighted mahalanobis distance value of the fault sample set based on a combined weighted mahalanobis distance method;
and identifying the fault type of fault vibration generated by the gear teeth when the gearbox or the series-parallel gear transmission system runs by using the calculated minimum combined weighted Mahalanobis distance value according to the basic principle that the minimum combined weighted Mahalanobis distance value among the similar fault characteristics is minimum.
2. The method for identifying the fault according to claim 1, wherein the step of calculating the optimal value of the subjective weight value α of the fault sample set based on the analytic hierarchy process comprises the following steps:
building a structural model:
establishing a structural model of a decision problem according to the characteristic information of the fault sample set;
obtaining weight set vectors of each layer: weight set A of decision target layer1＝[Au1,Au2,Au3]T(ii) a Weight set Au of decision criterion layer1＝[u11,u12,u13]T，Au2＝[u21,u22,u23]T，Au3＝[u31,u32,u33]T；
The subjective weight calculation process comprises the following steps:
and calculating a comparison matrix P of each layer in the following way:
the priority vectors W of the respective layers of the respective decision matrices are calculated and normalized,
in the formula, n is the dimension of the corresponding judgment matrix;
the consistency check is performed for each judgment matrix Pi, and the calculation formula CR is as follows:
in the formula, λmaxIs the maximum characteristic root of the corresponding judgment matrix, and n is the corresponding judgment momentDimension of the array, RI is the average random consistency index;
a composite priority vector for the structural model is calculated as shown in the following table:
calculating the optimal subjective weight value alpha,
3. the fault identification method according to claim 1, wherein the step of calculating the optimal objective weight value β of each indicator element based on the entropy method comprises the following steps:
assume n samples, an original matrix A with m decision indexes, and A ═ A1,A2,…,Ai,…,Am]TIf so, the attribute value corresponding to the jth index of the ith sample in the matrix A is aij; the original matrix A is calculated by the following method:
determining a forward indicator A of an original matrix A+And a negative index A-And normalizing the data, wherein the calculation method of the forward index set comprises the following steps:
the negative index set calculation method comprises the following steps:
calculating the specific gravity p occupied by the ith sample in the jth indexijThe calculation method is shown as the following formula:
calculating the information entropy e of the jth indexjThe calculation method is shown as the following formula:
calculating an objective weight value beta of the jth index of the sample according to the following formula:
4. the method according to claim 1, wherein the calculating of the combined weight of the failure sample set is performed by calculating the combined weight of the failure sample set
The calculation method is shown as the following formula:
in the formula, alpha is a subjective weight value, beta is an objective weight value, lambda is a preference factor, lambda is more than or equal to 0 and less than or equal to 1, when lambda is small, the combination weight is biased to alpha, and when lambda is large, the combination weight is biased to beta.
5. The method for identifying faults according to claim 1, wherein the step of calculating the minimum combined weighted mahalanobis distance value of each fault sample set comprises the following steps:
calculating the Mahalanobis distance value D of the fault characteristic sampleMahaThe calculation method is shown as the following formula:
wherein, μ Y is a mean vector of the fault characteristics, CY is a covariance matrix of the fault characteristics;
calculating a combined weighted mahalanobis distance of fault feature samples
On the basis of the Mahalanobis distance, the fault feature samples are subjected to combined weighting to obtain a combined weighted Mahalanobis distance value
The calculation method is shown as the following formula:
in the formula (I), the compound is shown in the specification,
as a weight matrix, the weight matrix is,
is the combined weight.
6. The apparatus of a fault identification method according to any one of claims 1-5, comprising:
an acquisition module that acquires a fault sample set; the fault sample set is a fault sample set of fault vibration generated by gear teeth when a gearbox or a parallel-serial gear transmission system operates;
the first calculation module calculates the optimal value of the objective weight beta of the fault sample set based on an information entropy method;
the second calculation module is used for calculating the optimal subjective weight value alpha of the fault sample set based on an analytic hierarchy process;
a third calculation module for calculating a combined weight of the failure sample set
The fourth calculation module calculates a minimum combined weighted mahalanobis distance value of the fault sample set based on a combined weighted mahalanobis distance method;
and the identification module identifies the fault type of fault vibration generated by the gear teeth when the gearbox or the series-parallel gear transmission system operates by using the calculated minimum combined weighted Mahalanobis distance value according to the basic principle that the minimum combined weighted Mahalanobis distance value among the similar fault characteristics is minimum.
7. A fault identification system comprising a memory and a processor; the memory for storing a computer program; a method of fault identification as claimed in any one of claims 1-5 when executed by said processor.
8. A computer-readable storage medium, characterized in that the storage medium has stored thereon a computer program which, when being executed by a processor, carries out a method of fault identification according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110697516.6A CN113486742B (en) | 2021-06-23 | 2021-06-23 | Fault identification method, device and system and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110697516.6A CN113486742B (en) | 2021-06-23 | 2021-06-23 | Fault identification method, device and system and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113486742A true CN113486742A (en) | 2021-10-08 |
| CN113486742B CN113486742B (en) | 2024-02-20 |
Family
ID=77935869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110697516.6A Active CN113486742B (en) | 2021-06-23 | 2021-06-23 | Fault identification method, device and system and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113486742B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130177065A1 (en) * | 2012-01-10 | 2013-07-11 | Qualcomm Incorporated | Method and Apparatus for a Robust Fractionally Spaced Equalizer |
| US20150199358A1 (en) * | 2014-01-15 | 2015-07-16 | Massachusetts Institute Of Technology | System and Method For Providing Unidimensional Scale Extraction from a Multidimensional Entity |
| WO2016028228A1 (en) * | 2014-08-21 | 2016-02-25 | Avennetz Technologies Pte Ltd | System, method and apparatus for determining driving risk |
| US20170147687A1 (en) * | 2015-11-24 | 2017-05-25 | International Business Machines Corporation | Dynamically linking information in a network |
| CN107702919A (en) * | 2017-06-30 | 2018-02-16 | 长沙理工大学 | Sliding bearing lubricating state monitoring method based on sound emission |
| US20190005400A1 (en) * | 2016-09-20 | 2019-01-03 | Southwest Petroleum University | A fuzzy evaluation and prediction method for running status of mechanical equipment with occurrence probability of failure modes |
| CN109558486A (en) * | 2018-10-30 | 2019-04-02 | 国家电网有限公司客户服务中心 | Electric power customer service client's demand intelligent identification Method |
| CN110135708A (en) * | 2019-04-25 | 2019-08-16 | 温州大学 | Ball valve quality multi-step Fuzzy Comprehensive Evaluation based on AHP and comentropy |
| CN110320894A (en) * | 2019-08-01 | 2019-10-11 | 陕西工业职业技术学院 | A kind of accurate Coal Pulverizing System of Thermal Power Plant fault detection method for dividing overlapping area data category |
| CN110346666A (en) * | 2019-07-15 | 2019-10-18 | 南京邮电大学盐城大数据研究院有限公司 | A kind of network transformer state analysis method differentiated based on weighted Mahalanobis distance method |
| KR20200058155A (en) * | 2018-11-19 | 2020-05-27 | 이은주 | System for evaluating technology of company |
-
2021
- 2021-06-23 CN CN202110697516.6A patent/CN113486742B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130177065A1 (en) * | 2012-01-10 | 2013-07-11 | Qualcomm Incorporated | Method and Apparatus for a Robust Fractionally Spaced Equalizer |
| US20150199358A1 (en) * | 2014-01-15 | 2015-07-16 | Massachusetts Institute Of Technology | System and Method For Providing Unidimensional Scale Extraction from a Multidimensional Entity |
| WO2016028228A1 (en) * | 2014-08-21 | 2016-02-25 | Avennetz Technologies Pte Ltd | System, method and apparatus for determining driving risk |
| US20170147687A1 (en) * | 2015-11-24 | 2017-05-25 | International Business Machines Corporation | Dynamically linking information in a network |
| US20190005400A1 (en) * | 2016-09-20 | 2019-01-03 | Southwest Petroleum University | A fuzzy evaluation and prediction method for running status of mechanical equipment with occurrence probability of failure modes |
| CN107702919A (en) * | 2017-06-30 | 2018-02-16 | 长沙理工大学 | Sliding bearing lubricating state monitoring method based on sound emission |
| CN109558486A (en) * | 2018-10-30 | 2019-04-02 | 国家电网有限公司客户服务中心 | Electric power customer service client's demand intelligent identification Method |
| KR20200058155A (en) * | 2018-11-19 | 2020-05-27 | 이은주 | System for evaluating technology of company |
| CN110135708A (en) * | 2019-04-25 | 2019-08-16 | 温州大学 | Ball valve quality multi-step Fuzzy Comprehensive Evaluation based on AHP and comentropy |
| CN110346666A (en) * | 2019-07-15 | 2019-10-18 | 南京邮电大学盐城大数据研究院有限公司 | A kind of network transformer state analysis method differentiated based on weighted Mahalanobis distance method |
| CN110320894A (en) * | 2019-08-01 | 2019-10-11 | 陕西工业职业技术学院 | A kind of accurate Coal Pulverizing System of Thermal Power Plant fault detection method for dividing overlapping area data category |
Non-Patent Citations (5)
| Title |
|---|
| N. NAJAT等: "Gene Clustering with Partition Around Mediods Algorithm Based on Weighted and Normalized Mahalanobis Distance", 《2017 INTERNATIONAL CONFERENCE ON INTELLIGENT INFORMATICS AND BIOMEDICAL SCIENCES (ICIIBMS)》, pages 140 - 145 * |
| R. AGUSTIANSYAH等: "Implementation of analytic network process (ANP) and analytic hierarchy process (AHP) method to determine priorities of roads to be repaired at Bogor City Department of Public Works", 《INTERNATIONAL CONFERENCE OF INFORMATION AND COMMUNICATION TECHNOLOGY (ICOICT)》, vol. 2013, 5 August 2013 (2013-08-05), pages 61 - 66 * |
| 李海洋等: "层次分析法在确定医疗设备维护优先级中的应用研究", 《中国医学装备》, vol. 18, no. 4, 15 April 2021 (2021-04-15), pages 159 - 163 * |
| 郝春燕: "基于神经网络的模拟电路故障诊断方法的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 5, pages 135 - 222 * |
| 颜子寒等: "基于加权马氏距离的改进深度嵌入聚类算法", 《计算机应用》, vol. 39, no. 2, pages 122 - 126 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113486742B (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109947086A (en) | Mechanical breakdown migration diagnostic method and system based on confrontation study | |
| CN113822421B (en) | Neural network-based anomaly locating method, system, equipment and storage medium | |
| CN108919059A (en) | A kind of electric network failure diagnosis method, apparatus, equipment and readable storage medium storing program for executing | |
| CN116467674B (en) | Intelligent fault processing fusion updating system and method for power distribution network | |
| CN112861459B (en) | Full-sensitivity significance-confrontation sampling yield optimization method and device | |
| Zhu et al. | Res-HSA: Residual hybrid network with self-attention mechanism for RUL prediction of rotating machinery | |
| CN112732690B (en) | Stabilizing system and method for chronic disease detection and risk assessment | |
| CN114971345A (en) | Quality measuring method, equipment and storage medium for built environment | |
| CN113487223B (en) | Risk assessment method and system based on information fusion | |
| CN115169534A (en) | Sample optimization training method of convolutional neural network and computer readable storage medium | |
| CN118211882B (en) | Product quality management system and method based on big data | |
| CN113486742B (en) | Fault identification method, device and system and computer readable storage medium | |
| CN116661954B (en) | Virtual machine abnormality prediction method, device, communication equipment and storage medium | |
| CN110543417A (en) | method and device for evaluating effectiveness of software test | |
| CN116862284A (en) | Marine diesel engine health state assessment method based on SOM network and rule fusion | |
| KR102212022B1 (en) | Method of automatically determining condition of hydro turbine in hydroelectric power plant and system for the same | |
| CN108763728B (en) | Mechanical fault diagnosis method using parallel deep neural network hierarchical feature extraction | |
| CN113869194A (en) | Variable parameter milling process signal marking method and system based on deep learning | |
| CN115208745A (en) | Fault positioning method and device for power system and fault positioning equipment | |
| CN113239626A (en) | Sensor layout optimization method based on system reliability | |
| CN116882333B (en) | Chip limiting frequency prediction method, device, equipment and medium | |
| CN112732549A (en) | Test program classification method based on cluster analysis | |
| CN112541554B (en) | Multi-mode process monitoring method and system based on time constraint and nuclear sparse representation | |
| CN118211171B (en) | Knowledge graph-based target path mining method | |
| Ilmiah | Application of the Fuzzy C-Means Method in Grouping Heart Abnormalities Based on Electrocardiogram Medical Records |
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 |