CN111078456B - Device fault diagnosis method, device, computer readable storage medium and electronic device - Google Patents
Device fault diagnosis method, device, computer readable storage medium and electronic device Download PDFInfo
- Publication number
- CN111078456B CN111078456B CN201911362088.0A CN201911362088A CN111078456B CN 111078456 B CN111078456 B CN 111078456B CN 201911362088 A CN201911362088 A CN 201911362088A CN 111078456 B CN111078456 B CN 111078456B
- Authority
- CN
- China
- Prior art keywords
- data set
- fault
- phenomenon
- diagnosed
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000003745 diagnosis Methods 0.000 title claims abstract description 45
- 230000006870 function Effects 0.000 claims description 15
- 238000013136 deep learning model Methods 0.000 claims description 11
- 241000820057 Ithone Species 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 10
- 208000037408 Device failure Diseases 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 7
- 210000002569 neuron Anatomy 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 239000013598 vector Substances 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 238000012417 linear regression Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 210000002364 input neuron Anatomy 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/079—Root cause analysis, i.e. error or fault diagnosis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0766—Error or fault reporting or storing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The invention discloses a device fault diagnosis method, a device, a computer readable storage medium and an electronic device, wherein the method comprises the following steps: determining occurrence probability of the acquired phenomenon data set to be diagnosed of the target equipment corresponding to the first fault data set according to a first model parameter between the first fault data set and the first phenomenon data set; determining a second fault data set corresponding to the phenomenon data set to be diagnosed according to the occurrence probability of the phenomenon data set to be diagnosed corresponding to each first fault data set; and determining a target fault reason corresponding to the target equipment according to the second fault data set corresponding to the phenomenon data set to be diagnosed. By the technical scheme of the invention, the fault cause of the equipment can be more accurately determined.
Description
Technical Field
The present invention relates to the field of energy technologies, and in particular, to a device fault diagnosis method, a device, a computer readable storage medium, and an electronic device.
Background
The equipment has the characteristics of large scale, high complexity and the like, and once the equipment fails, chain reaction can be caused to cause serious disastrous accidents, so that the fault diagnosis of the equipment is an important link for ensuring the safe and efficient operation of the equipment.
At present, a series of rules between fault phenomena and fault reasons are established mainly through comprehensive utilization of expert experience, so that a knowledge base is formed, when the fault phenomena are acquired, the fault phenomena are inferred through the knowledge base, the fault reasons corresponding to the fault phenomena are determined, and fault diagnosis is achieved.
However, the above-described fault diagnosis depends on expert experience, which depends on faults that have occurred, and these knowledge are limited, and thus, may result in relatively low accuracy of fault diagnosis.
Disclosure of Invention
The invention provides a device fault diagnosis method, a device, a computer readable storage medium and electronic equipment, which can improve the accuracy of fault diagnosis.
In a first aspect, the present invention provides an apparatus fault diagnosis method, including:
determining occurrence probability of the acquired phenomenon data set to be diagnosed of the target equipment corresponding to the first fault data set according to a first model parameter between the first fault data set and the first phenomenon data set;
determining a second fault data set corresponding to the phenomenon data set to be diagnosed according to the occurrence probability of the phenomenon data set to be diagnosed corresponding to each first fault data set;
And determining a target fault reason corresponding to the target equipment according to the second fault data set corresponding to the phenomenon data set to be diagnosed.
Preferably, the method further comprises:
inputting the phenomenon data set to be diagnosed into a deep learning model, and determining a third fault data set corresponding to the phenomenon data set to be diagnosed;
the determining, according to the second fault data set corresponding to the to-be-diagnosed phenomenon data set, a target fault cause corresponding to the target device includes:
and determining a target fault reason corresponding to the target equipment according to the second fault data set and the third fault data set corresponding to the to-be-diagnosed phenomenon data set.
Preferably, the deep learning model includes a neural network model, and the neural network model is obtained based on training of a second phenomenon data set corresponding to the target device and a fourth fault data set corresponding to the second phenomenon data set.
Preferably, the second phenomenon data set and the fourth failure data set carry device function labels of the target device.
Preferably, the method further comprises:
determining a fifth fault data set corresponding to the phenomenon data set to be diagnosed according to a preset fault knowledge base;
the determining, according to the second fault data set and the third fault data set corresponding to the to-be-diagnosed phenomenon data set, a target fault cause corresponding to the target device includes:
And determining a target fault reason corresponding to the target equipment according to the second fault data set, the third fault data set and the fifth fault data set corresponding to the to-be-diagnosed phenomenon data set.
Preferably, the determining, according to a preset fault knowledge base, a fifth fault data set corresponding to the to-be-diagnosed phenomenon data set includes:
calculating the similarity of the to-be-diagnosed phenomenon data set and each third phenomenon data set in a preset knowledge base;
and when the similarity corresponding to the third phenomenon data set meets a first preset condition, taking the fault data set corresponding to the third phenomenon data set in a preset knowledge base as a fifth fault data set corresponding to the phenomenon data set to be diagnosed.
Preferably, the number of occurrences and/or the weight value of the target fault cause corresponding to the second fault data set, the third fault data set and the fifth fault data set meet a second preset condition.
Preferably, the determining, according to the first model parameter between the first failure data set and the first phenomenon data set, the occurrence probability of the to-be-diagnosed phenomenon data set corresponding to the first failure data set includes:
determining second model parameters corresponding to the first fault data sets according to the first model parameters between the first fault data sets and each first phenomenon data set;
And determining occurrence probability of the phenomenon data set to be diagnosed corresponding to the first fault data set according to the first model parameter, the first phenomenon data set, the second model parameter and the phenomenon data set to be diagnosed.
Preferably, the determining, according to the first model parameter, the first phenomenon data, the second model parameter and the to-be-diagnosed phenomenon data set, the occurrence probability of the to-be-diagnosed phenomenon data set corresponding to the first fault data set includes:
calculating the occurrence probability of the first fault data set through the following formula; wherein the formula comprises:
wherein Pr characterizes the occurrence probability; y is k Characterizing a kth of said first failure data set; beta k Characterizing a second model parameter corresponding to the first failure dataset; x is x i Characterizing an ith one of said first phenomenon datasets; beta i Characterizing a first model parameter corresponding to an ith first phenomenon data set and a kth first fault data set; n represents a first number of failed phenomena in an ith one of said first phenomenon data sets; 2 n Characterizing a second number of the first phenomenon dataset; the first phenomenon data set includes data corresponding to n fault phenomena, respectively.
In a second aspect, the present invention provides an apparatus failure diagnosis device, comprising:
the probability determining module is used for determining the occurrence probability of the acquired to-be-diagnosed phenomenon data set of the target equipment corresponding to the first fault data set according to the first model parameters between the first fault data set and the first phenomenon data set;
the first diagnosis module is used for determining second fault data corresponding to the phenomenon data sets to be diagnosed according to the occurrence probability of the phenomenon data sets to be diagnosed corresponding to each first fault data set;
and the fault determining module is used for determining a target fault reason corresponding to the target equipment according to the second fault data set corresponding to the phenomenon data set to be diagnosed.
In a third aspect, the present invention provides a computer readable storage medium comprising execution instructions which, when executed by a processor of an electronic device, perform the method of any of the first aspects.
In a fourth aspect, the present invention provides an electronic device comprising a processor and a memory storing execution instructions, the processor performing the method according to any one of the first aspects when executing the execution instructions stored in the memory.
The invention provides a device fault diagnosis method, a device, a computer readable storage medium and an electronic device. In summary, through the technical scheme of the invention, the accuracy of fault diagnosis can be improved.
Further effects of the above-described non-conventional preferred embodiments will be described below in connection with the detailed description.
Drawings
In order to more clearly illustrate the embodiments of the invention or the prior art solutions, the drawings which are used in the description of the embodiments or the prior art will be briefly described below, it being obvious that the drawings in the description below are only some of the embodiments described in the present invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a schematic flow chart of an apparatus fault diagnosis method according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of step 102 in an apparatus fault diagnosis method according to an embodiment of the present invention;
FIG. 3 is a flow chart illustrating another method for diagnosing equipment failure according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart of a neural network in another device fault diagnosis method according to an embodiment of the present invention;
FIG. 5 is a schematic flow chart of another apparatus fault diagnosis method according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an apparatus for diagnosing a device failure according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram II of an apparatus fault diagnosis device according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of another apparatus for diagnosing a device failure according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of yet another device fault diagnosis apparatus according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram II of another device fault diagnosis apparatus according to an embodiment of the present invention;
fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to specific embodiments and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, an embodiment of the present invention provides an apparatus fault diagnosis method, including the following steps:
and step 103, determining a target fault reason corresponding to the target equipment according to the second fault data set corresponding to the phenomenon data set to be diagnosed.
In the embodiment shown in fig. 1, the method determines, through first model parameters between a first failure data set and a first phenomenon data set, occurrence probability of a to-be-diagnosed phenomenon data set of an acquired target device corresponding to the first failure data set, then determines one or more second failure data sets corresponding to the to-be-diagnosed phenomenon data set according to occurrence probability of the to-be-diagnosed phenomenon data set corresponding to each first failure data set, and then determines a target failure cause corresponding to the target device according to one or more second failure data sets corresponding to the to-be-diagnosed phenomenon data set. In summary, through the technical scheme of the invention, the accuracy of fault diagnosis can be improved.
In this implementation manner, according to the occurrence probability of the phenomenon data set to be diagnosed corresponding to each first fault data set, determining one or more second fault data sets corresponding to the phenomenon data set to be diagnosed may be implemented specifically by the following manner: and sequencing the occurrence probabilities according to the sequence from big to small, taking the first fault data sets corresponding to the first occurrence probabilities as second phenomenon data sets respectively, or determining the occurrence probabilities greater than a preset value, and taking the first fault data sets corresponding to the occurrence probabilities as second phenomenon data sets respectively.
In this implementation manner, determining, according to one or more second fault data sets corresponding to the to-be-diagnosed phenomenon data set, a target fault cause corresponding to the target device may be implemented specifically by: when the number of the second fault data sets is multiple, a plurality of fault reasons corresponding to the plurality of second fault data sets are combined, and the embodiment of the invention does not limit a combination mode, for example, the fault reason with the occurrence number greater than a preset value can be used as a target fault reason, or the fault reason with the highest occurrence number can be used as a target fault reason. When the number of the second fault data sets is one, a plurality of fault reasons corresponding to the second fault data sets can be used as target fault reasons.
Accordingly, according to the method provided by the embodiment of the invention, the time waste caused by the fact that the user maintains the target equipment by adopting the maintenance means corresponding to the failure cause of the target equipment can be avoided to a certain extent, so that the problem that the target equipment is further damaged due to the fact that the failed target equipment cannot be maintained timely and effectively is avoided.
In particular, target devices include, but are not limited to, gas boilers, gas pipelines, and other types of devices, and the methods provided by various embodiments of the present invention are particularly applicable to gas boilers. The target device comprises a plurality of fault reasons and a plurality of fault phenomena, the plurality of fault phenomena are formed into a plurality of first phenomenon data sets, and the plurality of fault reasons can form the plurality of first fault data sets. Wherein the different first phenomenon data sets correspond to different fault phenomena; the different first failure data sets correspond to different failure causes.
It should be noted that the to-be-diagnosed phenomenon data set indicates the operation condition of the target device, including a plurality of fault phenomena, and the data in the to-be-diagnosed phenomenon data set is usually obtained by analyzing the operation data of the target device collected by the sensor. The second set of fault data indicates a set of possible several fault causes for the target device when a fault phenomenon in the set of phenomenon data to be diagnosed occurs. The number of the target fault reasons is one or more, and the fault reasons which are possibly corresponding to the fault phenomena in the to-be-diagnosed phenomenon data set are indicated when the target equipment generates the fault phenomena.
When the first phenomenon data set is substituted as a feature into the model for the trained model, the model uses a parameter when the first phenomenon data set is classified as a first failure data set as a first model parameter. The first model parameters are self-learned by the model, for example, weights and biases in the neural network are parameters.
Referring to fig. 2, in an embodiment of the present invention, based on the embodiment of fig. 1, determining, in step 102, the occurrence probability of the to-be-diagnosed phenomenon dataset corresponding to the first failure dataset according to the first model parameter between the first failure dataset and the first phenomenon dataset includes:
Step 1021, determining second model parameters corresponding to the first fault data sets according to the first fault data sets and the first model parameters between each first phenomenon data set;
step 1022, determining occurrence probability of the phenomenon data set to be diagnosed corresponding to the first fault data set according to the first model parameter, the first phenomenon data set, the second model parameter and the phenomenon data set to be diagnosed.
Specifically, the first phenomenon data sets and the first fault data sets corresponding to the first phenomenon data sets are manually matched in advance, a multi-class regression model is trained based on the first phenomenon data sets which are manually matched in advance and correspond to the first phenomenon data sets, namely, each first phenomenon data set is taken as an input, the first fault data sets are taken as outputs, machine learning is carried out in the trained multi-class regression model, so that first model parameters between the first fault data sets and each first phenomenon data set are determined, and second model parameters corresponding to the first fault data sets are determined based on the first model parameters between the first fault data sets and each first phenomenon data set. The second model parameter is typically a first model parameter between the matched first phenomenon data set and a first fault data set corresponding to the first phenomenon data set.
Preferably, the determining, according to the first model parameter, the first phenomenon data, the second model parameter and the to-be-diagnosed phenomenon data set, the occurrence probability of the to-be-diagnosed phenomenon data set corresponding to the first fault data set includes:
calculating the occurrence probability of the first fault data set through the following formula; wherein the formula comprises:
wherein Pr characterizes the occurrence probability; y is k Characterizing a kth of said first failure data set; beta k Characterizing a second model parameter corresponding to the first failure dataset; x is x i Characterizing an ith one of said first phenomenon datasets; beta i Characterizing a first model parameter corresponding to an ith first phenomenon data set and a kth first fault data set; n represents a first number of failed phenomena in an ith one of said first phenomenon data sets; 2 n Characterizing a second number of the first phenomenon dataset; the first phenomenon data set includes data corresponding to n fault phenomena, respectively.
For example, the target device is a gas boiler, assuming that the gas boiler includes m phenomena, P i Represents the ith phenomenon, where each phenomenon P i There are two states of "0" and "1", and "0" and "1" are data. If this phenomenon is normal, P i =0; conversely, P i =1. These m phenomena can constitute 2 m A first phenomenon dataset:
at the same time, let us assume that we have n failure causes, q i Represents the ith failure cause, wherein each failure cause q i There are two states, 0 and 1. If the cause of the fault exists, q i =0; conversely, q i =1, then the n failure causes may generate 2 n A first failure data set:
the second failure data set is 2 n One of the first failure data sets, 2 m The details of the first phenomenon data set are shown in tables 1 and 2 n The details of the first failure data set are shown in table 2.
TABLE 1
TABLE 2
The "0" and "1" in tables 1 and 2 represent data corresponding to the failure phenomenon, the data being a carrier of information, and it is apparent that the first phenomenon data set is essentially a vector, and different positions reflect different failure phenomena.
For each first fault data set in table 2, machine learning is performed on the first fault data set and each first phenomenon data set in table 1 by using a preset linear regression model and a normalization factor (a logarithmic form of a distribution function), so as to determine a first model parameter between the first fault data set and each first phenomenon data set in table 1, and a second model parameter is determined from the first model parameters, wherein the linear regression model is not limited by the present invention. For example, assume that the regression equation in the linear regression model is a one-time equation, assume x in Table 1 1 And y in Table 2 1 Correspondingly, beta 1 Represents x 1 And y 1 First model parameters in between, let us assume that when x 1 Classified as y 1 The calculation formula is w 11 x 1 +b 11 Wherein x is 1 Refers to x in Table 1 1 Vector formed by row data of rows, w 11 And b 11 I.e. as first model parameter, i.e. beta 1 . Let x be 1 And y 1 There is a correspondence, i.e. x occurs 1 The cause of the failure phenomenon in (a) is y 1 In (a) the cause of the failure, at this time, w can be 11 And b 11 As a second model parameter.
It should be noted that the regression equations of the different first fault data sets in table 1 may be different or the same, which is not limited in the embodiment of the present invention.
As shown in fig. 3, based on step 101 and step 102 in the embodiment shown in fig. 1, another device fault diagnosis method provided in the embodiment of the present invention further includes the following steps:
In this embodiment, after the fault phenomenon in the to-be-diagnosed phenomenon data set is input into the deep learning model, the deep learning model may output a third fault data set corresponding to the to-be-diagnosed phenomenon data set. And then, combining the fault reasons in the third fault data set and the second fault data set, wherein the combination method is not limited, and any combination method in the prior art can be used for determining the target fault reason.
Preferably, the deep learning model includes a neural network, and the neural network model is obtained based on training of a second phenomenon data set corresponding to the target device and a fourth fault data set corresponding to the second phenomenon data set.
The neural network model has strong nonlinear mapping capability, can accurately simulate fault diagnosis of complex equipment, and has parallel processing capability, self-learning capability and memory capability. Each input layer neuron in the neural network corresponds to one failure phenomenon, and each output layer neuron corresponds to one or more failure causes.
The neural network is obtained by taking a second phenomenon data set as input and a fourth fault data set corresponding to the second phenomenon data set as output to train a neural network model. Obviously, the fault phenomenon in the phenomenon data set to be diagnosed is input to the input neuron in the corresponding neural network model, and finally the third fault data set corresponding to the phenomenon data set to be diagnosed is input.
The second phenomenon data set and the first phenomenon data set may be the same or different, and the fourth failure data set and the first failure data set may be the same or different. Typically, the second phenomenon data set is table 1 and the fourth failure data set is table 2.
Referring to fig. 4, based on the contents of the above tables 1 and 2, assuming that the second phenomenon data set is the first phenomenon data set and the fourth failure data set is the first failure data set, then P 1 To P m Representing data in a first phenomenon data set, y, in Table 1 1 To the point of
Respectively including the fault data sets, a, in table 2 above 1，1 To->
A) 2，1 To->
Each representing a neuron in a neural network, the neuron receiving an upper-layer input for each neuron and summing the upper-layer inputs after nonlinear linearization using an activation function to obtain an output of the neuron, for a 1，1 For illustration, a 1，1 The upper layer input of (2) is (P 1 *w 1 +b 1 )、(P 2 *w 2 +b 2 )、…、(P m *w m +b m )，a 1，1 The input of (2) is (P) 1 *w 1 +b 1 )*F(P 1 *w 1 +b 1 )+(P 2 *w 2 +b 2 )*F(P 2 *w 2 +b 2 )+…+(P m *w m +b m )*F(P m *w m +b m ) Where F represents an activation function, which may be an activation function in the prior art, for example, a ReLU function, a Sigmod function, etc., which is not limited in this embodiment of the present invention, and other neurons are similar and not described herein.
Preferably, the second phenomenon data set and the fourth failure data set carry device function labels of the target device.
Considering that the target device has more fault phenomena, the target device is usually located in a certain system, and the relationship between the fault phenomena and the fault reasons is complex, for example, the target device may cause a series of fault phenomena due to the occurrence of a problem of a certain part, so that for a complex system, the phenomenon data set is too large, and the required neural network model is too large, which is not beneficial to learning, so that the diagnosis efficiency and the accuracy of the diagnosis result may be reduced. The number of the device functions of the target device is smaller than that of the device phenomena, the chain reaction between the phenomena is ignored, and the number of the phenomenon data sets can be reduced only from the device functions, so that the diagnosis efficiency and the accuracy of the diagnosis result are improved to a certain extent.
Specifically, a plurality of device functions of the target device are determined, for each device function, a fault phenomenon and a fault reason corresponding to the device function are determined, and a relation between the fault phenomenon and the fault reason is established, so that a relation between the fault phenomenon corresponding to each fault reason and the fault phenomenon corresponding to the plurality of device functions is established, a phenomenon data set of a complex system is not excessively large, and therefore diagnosis efficiency and accuracy of diagnosis results are improved.
As shown in fig. 5, on the basis of step 101 and step 102 in the embodiment shown in fig. 1 and step 301 in the embodiment shown in fig. 3, the further device fault diagnosis method provided by the embodiment of the present invention further includes the following steps:
In this embodiment, according to a preset knowledge base, a fifth fault data set corresponding to the to-be-diagnosed phenomenon data set may be determined, and then, fault reasons in the fifth fault data sets of the second fault data set and the third fault data set may be combined, so as to determine a plurality of target fault reasons, where accuracy of the target fault reasons is relatively high.
In an embodiment of the present invention, on the basis of the embodiment shown in fig. 5, the determining, according to a preset fault knowledge base, a fifth fault data set corresponding to the to-be-diagnosed phenomenon data set includes:
calculating the similarity of the to-be-diagnosed phenomenon data set and each third phenomenon data set in a preset knowledge base;
and when the similarity corresponding to the third phenomenon data set meets a first preset condition, taking the fault data set corresponding to the third phenomenon data set in a preset knowledge base as a fifth fault data set corresponding to the phenomenon data set to be diagnosed.
In this embodiment, for each third phenomenon data set in the preset knowledge base, a similarity between the phenomenon data set to be diagnosed and the third phenomenon data set is determined, and when the similarity meets a first preset condition, a fault data set corresponding to the third phenomenon data set in the preset knowledge base is used as a fifth fault data set corresponding to the phenomenon data set to be diagnosed.
In this implementation, the similarity meeting the first preset condition includes, but is not limited to, the similarity being greater than a first preset value, or the similarity being maximum.
Specifically, the to-be-diagnosed phenomenon data set and the third phenomenon data set are vectors in nature, and the data items at the same position correspond to the same fault phenomenon, so that the similarity between the to-be-diagnosed phenomenon data set and the third phenomenon data set is calculated, where the similarity may be the sum of products of two data items corresponding to all the same fault phenomenon, and of course, in the prior art, the method for calculating the similarity between the two vectors may be any method, and the method for calculating the similarity is not limited in the embodiment of the present invention. For example, the processing steps may be performed,
Obviously, according to the working environment of the system where the target equipment is located, the system knowledge (reflecting the working mechanism and structural knowledge of the system) and the rule reflecting the cause and effect of the system, a preset knowledge base is established, wherein the preset knowledge base comprises, but is not limited to, a plurality of third phenomenon data sets and fault data sets corresponding to the third phenomenon data sets respectively. Wherein the third phenomenon data set and the first phenomenon data set may be equal, i.e. the third phenomenon data set may be table 1.
For example, the fifth failure data set may be calculated by the following formula.
Wherein A is i Characterizing an ith third phenomenon data set in a preset knowledge base; b, representing a phenomenon data set to be diagnosed; n (A) i B) representing the similarity of the ith third phenomenon data set and the phenomenon data set to be diagnosed in a preset knowledge base; a is that k And representing a fault data set corresponding to the third phenomenon data set corresponding to the maximum similarity.
Preferably, the determining, according to the second fault data set, the third fault data set and the fifth fault data set corresponding to the to-be-diagnosed phenomenon data set, the target fault cause corresponding to the target device includes:
and combining the fault reasons in the second fault data set, the third fault data set and the fifth fault data set corresponding to the to-be-diagnosed phenomenon data set according to a preset combination mode, and determining the target fault reason corresponding to the target equipment.
In this embodiment, according to a preset combination manner, the fault reasons in the second fault data set, the third fault data set and the fifth fault data set corresponding to the to-be-diagnosed phenomenon data set are combined, and the target fault reason corresponding to the target device is determined, so that the reference value of the target fault reason is ensured.
It should be noted that the second fault data set is essentially determined based on a shallow learning model, the third fault data set is determined based on a deep learning model, and the fifth fault data set is essentially determined based on rules, and different fault diagnosis methods are integrated to ensure the accuracy of the target fault cause. It should be noted that, the fault diagnosis method added into the bayesian network in the prior art may also be used to further improve the accuracy of the target fault cause.
Here, the target fault cause corresponds to the number of occurrences and/or the weight value of the second fault data set, the third fault data set and the fifth fault data set meeting a second preset condition. Wherein, the occurrence times and/or the weight value meet the second preset condition comprises: maximum number of occurrences, maximum weight value, maximum weighted average of number of occurrences and weight value, maximum weighted sum of number of occurrences and weight value, number of occurrences greater than a preset value, weight value greater than a preset value, weighted average of number of occurrences and weight value greater than a preset value, weighted sum of number of occurrences and weight value greater than a preset value, etc. The embodiment of the invention does not limit the content of the second preset condition, and specifically needs to be determined by combining the actual conditions.
The target fault cause corresponds to a number of occurrences in the second, third, and fifth fault data sets that indicates a total number of occurrences of the target fault cause in the second, third, and fifth fault data sets.
The target fault reasons correspond to the weight values of the second fault data set, the third fault data set and the fifth fault data set, the weight comprehensively considers the accuracy of different fault diagnosis methods, and the first weight values respectively corresponding to the second fault data set, the third fault data set and the fifth fault data set are considered. Specifically, for each fault reason, determining a plurality of fault data sets including the fault reason in the second fault data set, the third fault data set and the fourth fault data set, taking first weight values corresponding to the fault data sets as second weight values of the fault reason, and carrying out weighted average or weighted summation on the second weight values so as to determine a third weight of the fault reason, wherein the third weight values are weight values of the target fault reason corresponding to the second fault data set, the third fault data set and the fifth fault data set.
Based on the same concept as the method embodiment of the present invention, please refer to fig. 6, the embodiment of the present invention further provides an apparatus fault diagnosis device, which includes:
the probability determining module 601 is configured to determine, according to a first model parameter between a first failure data set and a first phenomenon data set, occurrence probability of the obtained to-be-diagnosed phenomenon data set of the target device corresponding to the first failure data set;
a first diagnostic module 602, configured to determine second fault data corresponding to the to-be-diagnosed phenomenon data set according to occurrence probability of the to-be-diagnosed phenomenon data set corresponding to each first fault data set;
the fault determining module 603 is configured to determine a target fault cause corresponding to the target device according to the second fault data set corresponding to the to-be-diagnosed phenomenon data set.
Referring to fig. 7, in one embodiment of the present invention, the first diagnostic module 602 includes: a parameter determination unit 6021 and a probability determination unit 6022; wherein,,
the parameter determining unit 6021 is configured to determine, according to a first model parameter between a first failure data set and each first phenomenon data set, a second model parameter corresponding to the first failure data set;
The probability determining unit 6022 is configured to determine, according to the first model parameter, the first phenomenon data set, the second model parameter, and the phenomenon data set to be diagnosed, an occurrence probability of the phenomenon data set to be diagnosed corresponding to the first failure data set.
In one embodiment of the present invention, the probability determining unit 6022 is configured to calculate the occurrence probability of the first failure data set by the following formula; wherein the formula comprises:
wherein Pr characterizes the occurrence probability; y is k Characterizing a kth of said first failure data set; beta k Characterizing a second model parameter corresponding to the first failure dataset; x is x i Characterizing an ith one of said first phenomenon datasets; beta i Characterizing a first model parameter corresponding to an ith first phenomenon data set and a kth first fault data set; n represents a first number of failed phenomena in an ith one of said first phenomenon data sets; 2 n Characterizing a second number of the first phenomenon dataset; the first phenomenon dataThe set includes data corresponding to each of the n failure phenomena.
Referring to fig. 8, in an embodiment of the present invention, on the basis of the probability determining module 601, the first diagnosing module 602, and the fault determining module 603 in the embodiment shown in fig. 6, the method further includes: a second diagnostic module 801; wherein,,
The prediction model 801 is configured to input the to-be-diagnosed phenomenon data set into a deep learning model, and determine a third fault data set corresponding to the to-be-diagnosed phenomenon data set;
the fault determining module 603 is configured to determine a target fault cause corresponding to the target device according to the second fault data set and the third fault data set corresponding to the to-be-diagnosed phenomenon data set.
In one embodiment of the present invention, the deep learning model includes a neural network model, and the neural network model is obtained based on training of a second phenomenon data set corresponding to the target device and a fourth fault data set corresponding to the second phenomenon data set.
In one embodiment of the present invention, the second phenomenon data set and the fourth failure data set carry device function labels of the target device.
Referring to fig. 9, based on the probability determining module 601, the first diagnosing module 602, the fault determining module 603 in the embodiment shown in fig. 6, and the prediction model 801 in the embodiment shown in fig. 8, in one embodiment of the present invention, the method further includes: a third diagnostic module 901; wherein,,
the third diagnostic module 901 is configured to determine a fifth fault data set corresponding to the to-be-diagnosed phenomenon data set according to a preset fault knowledge base;
The fault determining module 603 is configured to determine a target fault cause corresponding to the target device according to the second fault data set, the third fault data set, and the fifth fault data set corresponding to the to-be-diagnosed phenomenon data set.
Referring to fig. 10, in one embodiment of the present invention, the third diagnostic module 901 includes: a calculation unit 9011 and a failure diagnosis unit 9012; wherein,,
the calculating unit 9011 is configured to calculate a similarity between the to-be-diagnosed phenomenon data set and each third phenomenon data set in a preset knowledge base;
the fault diagnosis unit 9012 is configured to, when the similarity corresponding to the third phenomenon data set meets a first preset condition, use the fault data set corresponding to the third phenomenon data set in a preset knowledge base as a fifth fault data set corresponding to the to-be-diagnosed phenomenon data set.
In one embodiment of the present invention, the fault determining module 603 is configured to combine, according to a preset combination manner, fault reasons in the second fault data set, the third fault data set, and the fifth fault data set corresponding to the to-be-diagnosed phenomenon data set, and determine a target fault reason corresponding to the target device.
Fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. At the hardware level, the electronic device comprises a processor 111 and a memory 112 storing executable instructions, optionally together with an internal bus 113 and a network interface 114. The Memory 112 may include a Memory 1121, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory 1122 (non-volatile Memory), such as at least 1 disk Memory; the processor 111, the network interface 114, and the memory 112 may be connected to each other through an internal bus 113, and the internal bus 113 may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or an EISA (Extended Industry Standard Architecture ) bus, etc.; the internal bus 113 may be divided into an address bus, a data bus, a control bus, etc., and is represented by only one double-headed arrow in fig. 11 for convenience of representation, but does not represent only one bus or one type of bus. Of course, the electronic device may also include hardware required for other services. When the processor 111 executes the execution instructions stored in the memory 112, the processor 111 performs the method according to any of the embodiments of the present invention and is at least used to perform the method as shown in fig. 1, 2, 3 or 5.
In one possible implementation manner, the processor reads the corresponding execution instruction from the nonvolatile memory into the memory and then executes the execution instruction, and may also acquire the corresponding execution instruction from other devices, so as to form a device fault diagnosis device on a logic level. The processor executes the execution instructions stored in the memory to implement a device fault diagnosis method provided in any embodiment of the present invention by executing the execution instructions.
The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The embodiment of the invention also provides a computer readable storage medium, which comprises execution instructions, when the processor of the electronic device executes the execution instructions, the processor executes the method provided in any embodiment of the invention. The electronic device may specifically be an electronic device as shown in fig. 11; the execution instruction is a computer program corresponding to the device failure diagnosis apparatus.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method or a computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware aspects.
The embodiments of the present invention are described in a progressive manner, and the same and similar parts of the embodiments are all referred to each other, and each embodiment is mainly described in the differences from the other embodiments. In particular, for the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments in part.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or boiler that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or boiler. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of additional identical elements in a process, method, article or boiler comprising the element.
The foregoing is merely exemplary of the present invention and is not intended to limit the present invention. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are to be included in the scope of the claims of the present invention.
Claims (11)
1. A device failure diagnosis method, characterized by comprising:
determining occurrence probability of the acquired phenomenon data set to be diagnosed of the target equipment corresponding to the first fault data set according to a first model parameter between the first fault data set and the first phenomenon data set;
determining a second fault data set corresponding to the phenomenon data set to be diagnosed according to the occurrence probability of the phenomenon data set to be diagnosed corresponding to each first fault data set;
determining a target fault reason corresponding to the target equipment according to a second fault data set corresponding to the phenomenon data set to be diagnosed;
the determining, according to a first model parameter between a first failure data set and a first phenomenon data set, an occurrence probability of the to-be-diagnosed phenomenon data set corresponding to the first failure data set includes:
Determining second model parameters corresponding to the first fault data sets according to the first model parameters between the first fault data sets and each first phenomenon data set;
determining occurrence probability of the phenomenon data set to be diagnosed corresponding to the first fault data set according to the first model parameter, the first phenomenon data set, the second model parameter and the phenomenon data set to be diagnosed; the method specifically comprises the following steps:
calculating the occurrence probability of the first fault data set through the following formula; wherein the formula comprises:
wherein Pr characterizes the occurrence probability; y is k Characterizing a kth of said first failure data set; beta k Characterizing a second model parameter corresponding to the first failure dataset; x is x i Characterizing an ith one of said first phenomenon datasets; beta i Characterizing a first model parameter corresponding to an ith first phenomenon data set and a kth first fault data set; n represents a first number of failed phenomena in an ith one of said first phenomenon data sets; 2 n Characterizing a second number of the first phenomenon dataset; the first phenomenon data set includes data corresponding to n fault phenomena, respectively.
2. The method as recited in claim 1, further comprising:
Inputting the phenomenon data set to be diagnosed into a deep learning model, and determining a third fault data set corresponding to the phenomenon data set to be diagnosed;
the determining, according to the second fault data set corresponding to the to-be-diagnosed phenomenon data set, a target fault cause corresponding to the target device includes:
and determining a target fault reason corresponding to the target equipment according to the second fault data set and the third fault data set corresponding to the to-be-diagnosed phenomenon data set.
3. The method of claim 2, wherein the deep learning model comprises a neural network model that is trained based on a second phenomenon dataset corresponding to the target device and a fourth failure dataset corresponding to the second phenomenon dataset.
4. A method according to claim 3, wherein the second phenomenon data set and fourth failure data set carry device function labels of the target device.
5. The method as recited in claim 2, further comprising:
determining a fifth fault data set corresponding to the phenomenon data set to be diagnosed according to a preset fault knowledge base;
the determining, according to the second fault data set and the third fault data set corresponding to the to-be-diagnosed phenomenon data set, a target fault cause corresponding to the target device includes:
And determining a target fault reason corresponding to the target equipment according to the second fault data set, the third fault data set and the fifth fault data set corresponding to the to-be-diagnosed phenomenon data set.
6. The method according to claim 5, wherein determining a fifth fault data set corresponding to the to-be-diagnosed phenomenon data set according to a preset fault knowledge base includes:
calculating the similarity of the to-be-diagnosed phenomenon data set and each third phenomenon data set in a preset knowledge base;
and when the similarity corresponding to the third phenomenon data set meets a first preset condition, taking the fault data set corresponding to the third phenomenon data set in a preset knowledge base as a fifth fault data set corresponding to the phenomenon data set to be diagnosed.
7. The method according to claim 5, wherein determining the target fault cause corresponding to the target device according to the second, third and fifth fault data sets corresponding to the to-be-diagnosed phenomenon data set comprises:
and combining the fault reasons in the second fault data set, the third fault data set and the fifth fault data set corresponding to the to-be-diagnosed phenomenon data set according to a preset combination mode, and determining the target fault reason corresponding to the target equipment.
8. The method according to claim 7, wherein the target fault cause corresponds to the number of occurrences and/or weight values in the second, third and fifth fault data sets satisfying a second preset condition.
9. An apparatus failure diagnosis device, comprising:
the probability determining module is used for determining the occurrence probability of the acquired to-be-diagnosed phenomenon data set of the target equipment corresponding to the first fault data set according to the first model parameters between the first fault data set and the first phenomenon data set;
the first diagnosis module is used for determining second fault data corresponding to the phenomenon data sets to be diagnosed according to the occurrence probability of the phenomenon data sets to be diagnosed corresponding to each first fault data set;
the fault determining module is used for determining a target fault reason corresponding to the target equipment according to the second fault data set corresponding to the phenomenon data set to be diagnosed;
the first diagnostic module includes:
a parameter determining unit, configured to determine a second model parameter corresponding to a first failure data set according to the first model parameter between the first failure data set and each first phenomenon data set;
The probability determining unit is used for determining the occurrence probability of the phenomenon data set to be diagnosed corresponding to the first fault data set according to the first model parameter, the first phenomenon data set, the second model parameter and the phenomenon data set to be diagnosed; the probability determination unit calculates the occurrence probability of the first failure data set specifically by the following formula; wherein the formula comprises:
wherein Pr characterizes the occurrence probability; y is k Characterizing a kth of said first failure data set; beta k Characterizing a second model parameter corresponding to the first failure dataset; x is x i Characterizing an ith one of said first phenomenon datasets; beta i Characterizing a first model parameter corresponding to an ith first phenomenon data set and a kth first fault data set; n represents a first number of failed phenomena in an ith one of said first phenomenon data sets; 2 n Characterizing a second number of the first phenomenon dataset; the first phenomenon data set includes data corresponding to n fault phenomena, respectively.
10. A computer readable storage medium comprising execution instructions which, when executed by a processor of an electronic device, perform the method of any one of claims 1 to 8.
11. An electronic device comprising a processor and a memory storing execution instructions that, when executed by the processor, perform the method of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911362088.0A CN111078456B (en) | 2019-12-26 | 2019-12-26 | Device fault diagnosis method, device, computer readable storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911362088.0A CN111078456B (en) | 2019-12-26 | 2019-12-26 | Device fault diagnosis method, device, computer readable storage medium and electronic device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111078456A CN111078456A (en) | 2020-04-28 |
| CN111078456B true CN111078456B (en) | 2023-05-16 |
Family
ID=70317870
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911362088.0A Active CN111078456B (en) | 2019-12-26 | 2019-12-26 | Device fault diagnosis method, device, computer readable storage medium and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111078456B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113064779A (en) * | 2021-03-04 | 2021-07-02 | 山东英信计算机技术有限公司 | Machine delivery diagnosis method, system, equipment and storage medium |
| CN113689042B (en) * | 2021-08-25 | 2024-06-14 | 华自科技股份有限公司 | Fault source prediction method for monitoring node |
| CN116880442B (en) * | 2023-07-03 | 2024-06-11 | 广州汽车集团股份有限公司 | Fault diagnosis method, device, electronic equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101419267A (en) * | 2008-12-03 | 2009-04-29 | 中国电力科学研究院 | Electric network fault on-line rapid diagnosis and analysis method |
| CN102055604A (en) * | 2009-11-05 | 2011-05-11 | ***通信集团山东有限公司 | Fault location method and system thereof |
| CN102496028A (en) * | 2011-11-14 | 2012-06-13 | 华中科技大学 | Breakdown maintenance and fault analysis method for complicated equipment |
| CN105095675A (en) * | 2015-09-07 | 2015-11-25 | 浙江群力电气有限公司 | Switch cabinet fault feature selection method and apparatus |
| CN107168292A (en) * | 2017-06-21 | 2017-09-15 | 武汉理工大学 | Submarine navigation device circuit failure diagnosis method based on ELM algorithms |
| CN107222322A (en) * | 2016-03-22 | 2017-09-29 | ***通信集团陕西有限公司 | A kind of communication failure diagnostic method and device |
| CN108052974A (en) * | 2017-12-12 | 2018-05-18 | 苏州大学 | A kind of method for diagnosing faults, system, equipment and storage medium |
| EP3373041A1 (en) * | 2017-03-07 | 2018-09-12 | Honeywell International Inc. | False alarm distribution in advanced receiver autonomous integrity monitoring |
| CN108663600A (en) * | 2018-05-09 | 2018-10-16 | 广东工业大学 | A kind of method for diagnosing faults, device and storage medium based on power transmission network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106444703B (en) * | 2016-09-20 | 2018-12-07 | 西南石油大学 | Dynamic equipment running status fuzzy evaluation and prediction technique based on fault mode probability of happening |
-
2019
- 2019-12-26 CN CN201911362088.0A patent/CN111078456B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101419267A (en) * | 2008-12-03 | 2009-04-29 | 中国电力科学研究院 | Electric network fault on-line rapid diagnosis and analysis method |
| CN102055604A (en) * | 2009-11-05 | 2011-05-11 | ***通信集团山东有限公司 | Fault location method and system thereof |
| CN102496028A (en) * | 2011-11-14 | 2012-06-13 | 华中科技大学 | Breakdown maintenance and fault analysis method for complicated equipment |
| CN105095675A (en) * | 2015-09-07 | 2015-11-25 | 浙江群力电气有限公司 | Switch cabinet fault feature selection method and apparatus |
| CN107222322A (en) * | 2016-03-22 | 2017-09-29 | ***通信集团陕西有限公司 | A kind of communication failure diagnostic method and device |
| EP3373041A1 (en) * | 2017-03-07 | 2018-09-12 | Honeywell International Inc. | False alarm distribution in advanced receiver autonomous integrity monitoring |
| CN107168292A (en) * | 2017-06-21 | 2017-09-15 | 武汉理工大学 | Submarine navigation device circuit failure diagnosis method based on ELM algorithms |
| CN108052974A (en) * | 2017-12-12 | 2018-05-18 | 苏州大学 | A kind of method for diagnosing faults, system, equipment and storage medium |
| CN108663600A (en) * | 2018-05-09 | 2018-10-16 | 广东工业大学 | A kind of method for diagnosing faults, device and storage medium based on power transmission network |
Non-Patent Citations (2)
| Title |
|---|
| cui li 等."Diagnose of Sub-module Fault in Modular Multilevel Converters Based on Moving Average Method".《2019 CAA Symposium on Fault Detection，Supervision and Safety for Technical Process》.2019, * |
| 杨俊凯 ; 徐云杰 ; .故障现象矢量距离判别方法在收割机故障诊断中的应用.微型机与应用.2017,(第12期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111078456A (en) | 2020-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111078456B (en) | Device fault diagnosis method, device, computer readable storage medium and electronic device | |
| Lanubile et al. | Comparing models for identifying fault-prone software components. | |
| JP2008536219A (en) | Diagnosis and prediction method and system | |
| Jin et al. | Efficient board-level functional fault diagnosis with missing syndromes | |
| TWI727323B (en) | Repairable board detection device, method and storage medium | |
| WO2022077907A1 (en) | Adversarial attack detection method, system and device, and computer-readable storage medium | |
| CN113541985B (en) | Internet of things fault diagnosis method, model training method and related devices | |
| US20220221843A1 (en) | Anomaly detection in high-volume manufacturing lines | |
| Kumar et al. | A feature selection model for prediction of software defects | |
| US20220318982A1 (en) | Neural network architecture for automated part inspection | |
| CN116166967B (en) | Data processing method, equipment and storage medium based on meta learning and residual error network | |
| US11373285B2 (en) | Image generation device, image generation method, and image generation program | |
| US20180067831A1 (en) | Fine-Grained Causal Anomaly Inference for Complex System Fault Diagnosis | |
| CN112203311B (en) | Network element abnormity diagnosis method, device, equipment and computer storage medium | |
| CN106096635B (en) | The warning classification method of cost-sensitive neural network based on threshold operation | |
| GOYAL | Early prediction of software fault-prone module using artificial neural network | |
| CN113822441B (en) | Decision model training method, device, terminal equipment and storage medium | |
| CN115658515A (en) | Deep learning metamorphic test case sequencing method and computer readable medium | |
| CN106055483B (en) | The warning classification method of cost-sensitive neural network based on lack sampling operation | |
| CN114372560A (en) | Neural network training method, device, equipment and storage medium | |
| CN114116430A (en) | System reliability analysis method, device and equipment | |
| CN112733433A (en) | Equipment testability strategy optimization method and device | |
| US20240152805A1 (en) | Systems, methods, and non-transitory computer-readable storage devices for training deep learning and neural network models using overfitting detection and prevention | |
| CN113240000B (en) | Machine state monitoring method, readable storage medium and electronic device | |
| CN117112449B (en) | Maturity assessment method, device, equipment and medium of data management tool |
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 |