CN111522308A - Fault diagnosis method and device, storage medium and computer equipment - Google Patents
Fault diagnosis method and device, storage medium and computer equipment Download PDFInfo
- Publication number
- CN111522308A CN111522308A CN202010307946.8A CN202010307946A CN111522308A CN 111522308 A CN111522308 A CN 111522308A CN 202010307946 A CN202010307946 A CN 202010307946A CN 111522308 A CN111522308 A CN 111522308A
- Authority
- CN
- China
- Prior art keywords
- fault
- matrix
- parameter
- parameter matrix
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
- G05B19/41885—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by modeling, simulation of the manufacturing system
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32339—Object oriented modeling, design, analysis, implementation, simulation language
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The invention discloses a fault diagnosis method, a fault diagnosis device, a storage medium and computer equipment, wherein the method comprises the following steps: when a fault occurs, acquiring operation data in a set time period before a fault time point; determining an initial parameter matrix according to the operation data in the set time period, and performing iteration through orthogonal transformation based on the initial parameter matrix to obtain an updated parameter matrix; determining main parameters which cause that the probability of generating the current fault is higher than a set value in parameter vectors corresponding to the faults based on the updated parameter matrix, and determining associated parameters according to the positions of the main parameters in the updated parameter matrix to form a target parameter set; and determining the reason and the corresponding probability of the fault according to the target parameter set and the fault tree.
Description
Technical Field
The invention relates to the technical field of big data, in particular to a fault diagnosis method, a fault diagnosis device, a storage medium and computer equipment.
Background
At present, with the progress of industrial automation, the degree of automated production in various industries is increasing. When equipment has faults, the method for troubleshooting the faults mainly comprises the following modes:
firstly, diagnosing faults according to equipment operation data by means of expert experience;
secondly, a database is established according to expert experience, abnormal parameters are searched for after the fault occurs, and the fault corresponding to the abnormal parameters is judged based on the database.
However, the first method is highly dependent on human factors and is inefficient. For the second method, a database needs to be established by combing huge data, the difficulty is high, the accuracy is low, and along with the increasingly complex functions of an automation system, the reasons for generating the same fault are often many, the real reason of the fault may not be directly related to abnormal parameters, and many branches may occur in the process of determining the fault based on the abnormal parameters, which may result in the fact that the real reason of the fault cannot be judged.
Disclosure of Invention
In order to solve the existing technical problems, embodiments of the present invention provide a fault diagnosis method, apparatus, storage medium, and computer device with high diagnosis efficiency and more accurate diagnosis result.
The technical scheme of the embodiment of the invention is realized as follows:
a fault diagnosis method comprising:
when a fault occurs, acquiring operation data in a set time period before a fault time point;
determining an initial parameter matrix according to the operation data in the set time period, and performing iteration through orthogonal transformation based on the initial parameter matrix to obtain an updated parameter matrix;
determining main parameters which cause that the probability of generating the current fault is higher than a set value in parameter vectors corresponding to the faults based on the updated parameter matrix, and determining associated parameters according to the positions of the main parameters in the updated parameter matrix to form a target parameter set;
and determining the reason and the corresponding probability of the fault according to the target parameter set and the fault tree.
Before obtaining the updated parameter matrix, the method further includes:
determining whether the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than a threshold value;
if the maximum value in the non-diagonal elements in the parameter matrix after iteration is not smaller than the threshold value, taking the parameter matrix after iteration as an updated initial parameter matrix, and returning to the step of performing iteration through orthogonal transformation based on the initial parameter matrix;
and if the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than the threshold value, obtaining the updated parameter matrix.
Wherein the iterating through orthogonal transformation based on the initial parameter matrix comprises:
constructing an initialization identity matrix;
calculating a vector rotation angle based on a maximum value among non-diagonal elements in the initial parameter matrix;
and constructing an orthogonal matrix according to the vector rotation angle, performing rotation transformation on the column vectors of the initial parameter matrix and the initialization unit matrix based on the orthogonal matrix, and iterating the initial parameter matrix.
Wherein the fault diagnosis method further comprises:
and iterating the unit matrix according to the orthogonal matrix.
Before determining the main parameters, which cause the probability of generating the current fault to be higher than the set value, in the parameter vector corresponding to the fault based on the updated parameter matrix, the method further includes:
and arranging the column vectors in the updated parameter matrix according to the magnitude of the modulus, and synchronously transforming the unit matrix according to the updated parameter matrix.
Wherein, the determining an initial parameter matrix according to the operation data in the set time period comprises:
determining a parameter vector corresponding to each parameter according to the operation data of each time point in the set time period;
subtracting the mean value of the corresponding row from the data in the matrix formed by the parameter vectors, and dividing the mean value by the standard deviation of the corresponding row to obtain a centralized matrix;
and obtaining the initial parameter matrix according to the covariance matrix of the centralized matrix.
After determining the cause and the corresponding probability of the fault according to the target parameter set and the fault tree, the method comprises the following steps:
generating a report form of the fault analysis result and storing the report form; and/or
And generating a report form of the fault analysis result and sending the report form to the terminal.
A fault diagnosis apparatus comprising:
the data acquisition module is used for acquiring operation data in a set time period before a fault time point when the fault occurs;
the iteration module is used for determining an initial parameter matrix according to the operating data in the set time period, and performing iteration through orthogonal transformation based on the initial parameter matrix to obtain an updated parameter matrix;
the parameter determining module is used for determining main parameters which cause that the probability of generating the current fault is higher than a set value in parameter vectors corresponding to the faults based on the updated parameter matrix, determining associated parameters according to the positions of the main parameters in the updated parameter matrix and forming a target parameter set;
and the fault diagnosis module is used for determining the reason and the corresponding probability of the fault according to the target parameter set and the fault tree.
A storage medium stores a computer program that, when executed by a processor, causes the processor to perform the steps of the fault diagnosis method provided by the embodiments of the present application.
A computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of the fault diagnosis method provided by an embodiment of the present application.
The fault diagnosis method, apparatus, storage medium and computer device provided in the above embodiments, wherein when a fault occurs, an initial parameter matrix is formed by obtaining operating data within a set time period before the fault occurs, an updated parameter matrix is obtained by performing an iteration through orthogonal transformation, a main parameter with a relatively largest contribution to the current fault is found from a parameter vector corresponding to the fault in the updated parameter matrix, a correlation parameter is determined according to a position of the main parameter in the updated parameter matrix to form a target parameter set, a parameter range of a most probable cause of the fault occurrence can be locked by forming the target parameter set, a fault tree is called again, a cause of the fault and a probability corresponding to the cause of the fault are analyzed according to parameters extracted from the target parameter set, so that a targeted fault diagnosis can be performed, a fault cause range is narrowed, and a fault diagnosis efficiency is improved, the purpose of rapid fault diagnosis and accuracy improvement is greatly achieved.
Drawings
Fig. 1 is an application scenario diagram of a fault diagnosis method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a fault diagnosis method according to an embodiment of the present invention;
FIG. 3 is a flow chart of a fault diagnosis method in an alternative embodiment of the present invention;
FIG. 4 is a schematic diagram of a fault diagnosis method according to another embodiment of the present invention;
FIG. 5 is a schematic diagram of a fault tree in an embodiment of the present invention;
fig. 6 is a schematic diagram of a fault diagnosis apparatus according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further elaborated by combining the drawings and the specific embodiments in the specification.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
In the following description, reference is made to the expression "some embodiments" which describes a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict.
Referring to fig. 1, an optional application scenario diagram of the fault diagnosis method provided in the embodiment of the present application includes a server 10 and a terminal, where the server 10 may be a cloud terminal composed of one or more physical servers, and the terminal may include at least one of the following: the system comprises a device to be monitored 22 directly or indirectly in communication connection with the server, an electronic device terminal 21 such as a personal computer, a mobile phone and a tablet for managing the device to be monitored 22, a data server 23 for storing the operation data of the device to be monitored 22 and the like. The device 22 to be monitored can be various production devices and living devices. When the device 22 to be monitored has a fault, the server 10 extracts the operation data in the set time period before the fault of the device 22 to be monitored occurs, and by operating the computer program of the fault diagnosis method provided by the embodiment of the present application, the rapid diagnosis of the cause of the fault of the device to be monitored is realized. The operation data acquired by the server 10 may be directly acquired from the device to be monitored 22, or the operation data in the set time period before the failure occurs is sent to the server 10 through the electronic device terminal 21 or the data server 23.
It should be noted that the terminal may also only include the electronic device terminal 21 communicatively connected to the server 10, a client program for implementing the fault diagnosis method is installed in the electronic device terminal 21, a server program for implementing the fault diagnosis method is installed in the server 10, and a user may perform communication with the server 10 by operating the electronic device terminal 21, including but not limited to sending a fault diagnosis instruction to the server 10 and/or obtaining a fault diagnosis result sent by the server 10.
Referring to fig. 2, a flowchart of a fault diagnosis method provided in an embodiment of the present application can be applied to the server shown in fig. 1, and the fault diagnosis method includes the following steps.
And step S101, when the fault is determined, acquiring the operation data in a set time period before the fault time point.
Determining that a fault has occurred may include: setting threshold value ranges for various operation parameters of the equipment respectively, and determining that the equipment fails when one or more operation parameters are determined to exceed the threshold value ranges. The failure time point refers to the time when the device failed. The set time period may be a preset time period or a time period input by the user through the terminal. As an optional implementation manner, when it is determined that a fault occurs, acquiring operation data in a set time period before a fault time point includes: the method comprises the steps that a server side monitors equipment to be monitored, when the server side determines that operation parameters of the equipment to be monitored exceed a threshold range, a fault time point of the equipment to be monitored is recorded, operation data of the equipment to be monitored in a set time period before the fault time point are acquired from the equipment to be monitored, and/or operation data of the equipment to be monitored in the set time period before the fault time point are acquired from an electronic equipment terminal for managing the equipment to be monitored, and/or the operation data of the equipment to be monitored in the set time period before the fault time point are acquired from a data server for storing the operation data of the equipment to be monitored.
As another optional implementation, when it is determined that a fault occurs, acquiring operation data in a set time period before a fault time point includes: the equipment or the electronic equipment terminal monitors the running state of the equipment, and sends a fault notification to the server when the equipment determines that the fault occurs; the method comprises the steps that an electronic equipment terminal obtains time period information input by a user and sends the time period information to a server side; and after receiving the fault notification, the server records the fault time point of the equipment to be monitored, and acquires the operation data of the equipment to be monitored in the set time period before the fault time point from the equipment to be monitored, and/or the electronic equipment terminal for managing the equipment to be monitored, and/or the data server for storing the operation data of the equipment to be monitored according to the time period information received from the electronic equipment terminal.
Step S103, an initial parameter matrix is determined according to the operation data in the set time period, and iteration is performed through orthogonal transformation based on the initial parameter matrix to obtain an updated parameter matrix.
The operation data of the equipment can be divided into a plurality of categories, and a plurality of groups of data of the operation parameters of the different categories are formed according to the data of the operation parameters of the different categories at a plurality of set time points, so that an initial parameter matrix is formed. The operation data of the equipment can be classified according to different parameters, and a plurality of groups of data of the parameters are formed according to the data of the different parameters at a plurality of set time points respectively, so that an initial parameter matrix is formed. Taking the example that the operation data of the equipment can be divided into n categories, or the operation data of the equipment can be divided into n parameters, an initial parameter matrix with n rows and x m columns is formed according to the data of the n categories of operation parameters at the set m time points respectively. Iteration is carried out through orthogonal transformation based on the initial parameter matrix, dimension reduction is carried out on the initial parameter matrix through the orthogonal transformation, and the initial parameter matrix is converted into linear uncorrelated principal components in initial parameter variables to form an updated parameter matrix.
Step S105, based on the updated parameter matrix, determining a main parameter which causes the probability of generating the current fault to be higher than a set value in a parameter vector corresponding to the fault, and determining a related parameter according to the position of the main parameter in the updated parameter matrix to form a target parameter set.
The parameter vector corresponding to the fault may refer to a vector of parameters that determine that the device is out of the corresponding threshold range when it fails. The main parameter of the parameter vector, which causes the probability of the current fault being higher than the set value, may be a parameter in the corresponding parameter vector, which is sorted in the set range according to the size of the data value, as the main parameter, or a parameter of the maximum value in the corresponding parameter vector as the main parameter.
Taking the initial parameter matrix of n rows and x m columns for dimensionality reduction to obtain an updated parameter matrix of n rows and x n columns as an example, the parameter corresponding to the fault is a2, and the maximum value or the parameter ranked in the first s bits is selected from the parameter vector corresponding to the parameter a2 as the main parameter. And determining a correlation parameter according to the position of the main parameter in the updated parameter matrix, wherein the correlation parameter can be a parameter which is larger than a threshold value in the principal component row vector where the main parameter corresponds, and the main parameter and the correlation parameter together form a target parameter set.
And step S107, determining the reason and the corresponding probability of the fault according to the target parameter set and the fault tree.
The fault tree may refer to a correspondence between parameter anomalies and fault types established from expert experience or from historical fault data. The server searches the fault tree in the range of the target parameter set by calling the fault tree to determine the cause and the corresponding probability of the fault, so that the efficiency of determining fault diagnosis is improved. The target parameter set is a combination of parameters which are determined most possibly to cause the current fault after dimension reduction analysis is carried out on the operation data in a set time period before the fault time point, and the data are analyzed to form the target parameter set, so that the range of subsequently calling a fault tree for diagnosis can be greatly reduced, the accuracy is ensured, and the diagnosis efficiency is improved.
In the fault diagnosis method provided in the above embodiment, when a fault occurs, the server side forms an initial parameter matrix by obtaining operation data within a set time period before the fault occurs, performs iteration through orthogonal transformation to obtain an updated parameter matrix, searches for a main parameter having a relatively largest contribution to the current fault from a parameter vector corresponding to the fault in the updated parameter matrix, determines a relevant parameter according to a position of the main parameter in the updated parameter matrix to form a target parameter set, can lock a parameter range of a most probable cause of the fault occurrence by forming the target parameter set, invokes a fault tree, analyzes the cause of the fault and a probability corresponding to the cause of the fault according to parameters extracted from the target parameter set, and thus, can perform targeted fault diagnosis, narrow a range of troubleshooting for determining the cause of the fault when the fault occurs, and improve fault diagnosis efficiency, the purpose of rapid fault diagnosis and accuracy improvement is greatly achieved.
In some embodiments, referring to fig. 3, in step S103, determining an initial parameter matrix according to the operation data in the set time period, and performing an iteration through an orthogonal transformation based on the initial parameter matrix to obtain an updated parameter matrix includes:
step S1031, determining an initial parameter matrix according to the operation data in the set time period, and performing iteration through orthogonal transformation based on the initial parameter matrix;
step S1032, determining whether the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than a threshold value;
if the maximum value in the non-diagonal elements in the parameter matrix after iteration is not less than the threshold value, returning to execute the step S1031, taking the parameter matrix after iteration as an updated initial parameter matrix, and returning to the step of performing iteration through orthogonal transformation based on the initial parameter matrix;
and if the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than the threshold value, executing step S1033 to obtain the updated parameter matrix.
In the process of performing iteration through orthogonal transformation based on the initial parameter matrix, the iteration frequency can be multiple times, the judgment condition of iteration termination is set to be that the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than the threshold, and the iteration frequency and the length of the parameter vector in the updated parameter matrix obtained after iteration can be adjusted through setting the threshold.
Wherein the iterating through orthogonal transformation based on the initial parameter matrix comprises:
constructing an initialization identity matrix;
calculating a vector rotation angle based on a maximum value among non-diagonal elements in the initial parameter matrix;
and constructing an orthogonal matrix according to the vector rotation angle, performing rotation transformation on the column vectors of the initial parameter matrix and the initialization unit matrix based on the orthogonal matrix, and iterating the initial parameter matrix.
The dimension of the initialized identity matrix is the same as the dimension of the initialized parameter matrix. Assuming that an initialized parameter matrix is Dx, the maximum value of the non-diagonal elements in the initialized parameter matrix is dpq, and a calculation formula for calculating the vector rotation angle theta according to the maximum value dpq of the non-diagonal elements in the initialized parameter matrix is as follows:
the calculation formula for constructing the orthogonal matrix U according to the vector rotation angle theta is as follows:
initializing unit matrix U-En, let Upp Uqq cos θ, Uqp-Upq sin θ (equation 2)
Performing rotation transformation on the column vectors of the initial parameter matrix and the initial unit matrix based on the orthogonal matrix to obtain the initial parameter matrix DxThe calculation formula for iteration is as follows:
Dx＝UTDXu (formula 3)
T ═ TU (equation 4)
Will initialize the parameter matrix DxObtaining an updated parameter matrix D after iterationxAnd performing rotation transformation on the column vectors of the initial parameter matrix and the initialized unit matrix based on the orthogonal matrix, performing iteration on the unit matrix to obtain an updated unit matrix T, judging whether the maximum value in the off-diagonal elements in the parameter matrix obtained after the iteration is smaller than a threshold value, if not, taking the parameter matrix after the iteration as the updated initial parameter matrix, and returning to the step of performing the iteration through the orthogonal transformation based on the initial parameter matrix. Optionally, the step of returning to the step of performing iteration through orthogonal transformation based on the initial parameter matrix specifically includes: updating and calculating a vector rotation angle theta based on a formula (1) according to the maximum value of the non-diagonal elements in the parameter matrix after iteration, constructing an updated orthogonal matrix U according to the vector rotation angle theta, performing rotation transformation on the column vectors of the initial parameter matrix after updating and the initialized unit matrix based on the updated orthogonal matrix, and completing iteration once again to obtain the parameter matrix after iteration. After the returning step is executed and iteration is executed, whether the maximum value of the off-diagonal elements in the parameter matrix obtained after iteration is smaller than the threshold value or not is judged again, if not, the iteration is executed again, if yes, the iteration is terminated, and the parameter matrix after iteration is used as the parameter matrixTo update the parameter matrix.
In some embodiments, before determining, based on the updated parameter matrix, the main parameter of the parameter vector corresponding to the fault, which causes the probability of generating the current fault to be higher than the set value, the step S105 further includes:
and step S104, arranging the column vectors in the updated parameter matrix according to the magnitude of the modulus, and synchronously transforming the unit matrix according to the updated parameter matrix.
The arranging according to the magnitude of the modulus may mean that the column vectors in the updated parameter matrix are sequentially arranged from left to right according to the magnitude of the modulus. The initial parameter matrix is a diagonal matrix through an updated parameter matrix after iteration, the larger the modulus value is, the more the corresponding element components in the matrix are, the more the column vectors in the updated parameter matrix are arranged from left to right according to the modulus value, and the identity matrix is synchronously transformed according to the updated parameter matrix, so that the elements in the matrix can be sequentially arranged from left to right according to the importance degree of the components. The column vectors in the updated parameter matrix are sequentially arranged according to the magnitude of a modulus, the unit matrix is synchronously transformed according to the updated parameter matrix, and in the subsequent step of determining the main parameter which causes the probability of generating the current fault in the parameter vector corresponding to the fault to be higher than the set value based on the updated parameter matrix, the unit matrix can be transposed, and the sequence of the elements in the unit matrix is corresponding to the row and column positions of each parameter in the updated parameter matrix, so that the elements in the parameter vector corresponding to each parameter are sequentially arranged according to the important program of the components, and the elements corresponding to the largest principal component can be more conveniently and rapidly found from the updated parameter matrix, or the elements corresponding to the principal components with the sequencing in the previously set range can be found.
In some embodiments, the determining an initial parameter matrix according to the operating data within the set time period includes:
determining a parameter vector corresponding to each parameter according to the operation data of each time point in the set time period;
subtracting the mean value of the corresponding row from the data in the matrix formed by the parameter vectors, and dividing the mean value by the standard deviation of the corresponding row to obtain a centralized matrix;
and obtaining the initial parameter matrix according to the covariance matrix of the centralized matrix.
The set time period may be a preset time period with a fixed duration, or may be a time period set by the terminal to send the long information to the server after the user inputs the long information through the terminal after determining the device failure. The time points are located at different moments in the time period, the set time period can be divided according to the length of the time period and the quantity of the operating data according to time intervals to determine the quantity of the time points, and then the operating data values of the parameters corresponding to the time points in the time period are determined to determine the parameter vectors corresponding to the parameters. Taking the operating DATA corresponding to m time points in the set time period as an example, the operating DATA values of the parameters a1, a2 to an at the time points 1 and 2 to m are respectively selected, the parameter vectors of the parameter a1 are a11 and a12 … a1m, the parameter vector of the parameter a2 is a21 and a22 … a2m, and the parameter vector of the parameter a … is an1 and an2 … anm, optionally, according to the operating DATA of each time point in the set time period, the matrix DATA formed by the parameter vectors corresponding to each parameter may be determined as follows:
centralizing DATA in a matrix DATA formed by the parameter vectors, specifically comprising: subtracting the average value of the row where the DATA in the matrix DATA formed by the parameter vectors is located from the DATA in the matrix DATA, and dividing the average value by the standard deviation of the corresponding row to obtain a centralized matrix X, wherein the calculation formula is as follows:
wherein the content of the first and second substances,
obtaining a calculation formula of the initial parameter matrix according to the covariance matrix of the centralized matrix, wherein the calculation formula comprises the following steps:
the covariance matrix of the centralized matrix is calculated, i.e. the correlation between the dimensions is calculated, and the larger the value of the element, the higher the correlation between the features corresponding to the table below. The covariance matrix of the centralized matrix is a symmetric matrix, the diagonal elements of the covariance matrix represent the variance of each dimension, and the off-diagonal elements represent the correlation between different dimensions. And obtaining the initial parameter matrix according to the covariance matrix of the centralized matrix, so that iteration is conveniently carried out on the initial parameter matrix subsequently to determine the eigenvalue and the eigenvector of the covariance, and the obtained eigenvector matrix is used as an updated parameter matrix.
After determining the cause and the corresponding probability of the fault according to the target parameter set and the fault tree, the step S107 includes:
step S108, generating a report form of the fault analysis result and storing the report form; and/or
And step S109, generating a report form of the fault analysis result and sending the report form to the terminal.
And the server determines the fault reason and the corresponding probability to form a fault analysis result, and generates a report of the fault analysis result for storage so that a user can call and check the fault analysis result when needed. Optionally, the server may also directly send the fault analysis result generation report to the terminal that manages the device to be monitored, so that the user can know the fault diagnosis result in time, and adopt corresponding measures or repair the fault according to the fault diagnosis result.
The server side can also acquire evaluation information of a fault analysis result, and update the fault tree according to the evaluation information and the corresponding fault analysis result. The evaluation information may be information that a user verifies a fault analysis result and determines that adjustment is required after receiving the fault analysis result, and when the user determines that the fault analysis result needs to be adjusted, the evaluation information may send modification data of the current fault analysis result to the server, and the server receives the modification data and then synchronously stores the modification data and the corresponding fault analysis result, and updates the fault tree according to the modification data. The evaluation information may also be information that a user verifies a fault analysis result after receiving the fault analysis result, and the server side may store the verified evaluation result as positive sample data.
In order to further specifically understand the fault diagnosis method provided in the embodiment of the present application, please refer to fig. 4, a service end is taken as a cloud server, a user communicates with the cloud server through a mobile phone terminal to monitor a device to be monitored, and the device to be monitored includes n pieces of operating data of parameters. Assuming that n parameters of the equipment to be monitored are a1 and a2 … an respectively, the fault diagnosis method comprises the following steps:
step S11, the cloud server collects data; the cloud server is in direct communication with the equipment to be detected, or is in communication with an electronic equipment terminal for managing the equipment to be detected or a storage server for storing operation data of the equipment to be detected, so that real-time operation data of the equipment to be detected are obtained. The data of the equipment can be uploaded to the cloud server in real time, a static or dynamic threshold value can be set for each parameter by combining the industry and the equipment characteristics, and when a certain parameter exceeds the threshold value, some reasons related to the parameter in the characterization system may have problems.
Step S12, determining whether the device is faulty; if not, that is, if the maximum value of the non-diagonal elements in the parameter matrix after the iteration is not less than the threshold value, returning to execute the step S12, and if so, that is, if the maximum value of the non-diagonal elements in the parameter matrix after the iteration is less than the threshold value, executing the step S13;
step S13, extracting data of all parameters in a period of time before the fault;
the period of time before the fault may be a preset period of time, or may be a period of time determined by the mobile phone terminal sending the start time information to the cloud server after the user inputs the start time information through the mobile phone terminal. The length of the time period will affect how much data is extracted. Optionally, the length of the setting time period may also correspond to different faults, for example, a mapping relation table between the fault and the length of the time period may be established in advance, and when it is determined that the fault of the parameter 1 occurs, the operation data of all the parameters in the setting time period 1 before the fault of the parameter 1 occurs are correspondingly extracted; and correspondingly extracting the operation data of all the parameters in the set time period 2 before the parameter 2 fault occurs when the parameter 2 fault is determined to occur, and so on, correspondingly extracting the operation data of all the parameters in the set time period n before the parameter n fault occurs when the parameter n fault occurs, or extracting the operation data of all the parameters in the default set time period when the parameter x fault occurs.
Step S14, data cleaning; cleaning the operating data to form a matrix of parameter vectors corresponding to the parameters, assuming that the parameter a2 fails, extracting all the operating data in a set time period before the parameter a2 fails, obtaining operating data values corresponding to a plurality of time points of the parameters in the set time period, and obtaining m groups of data of n parameters; the m sets of data for the n parameters are represented in the form of a column vector as follows:
step S15, principal component analysis; the principal component analysis mainly comprises the steps of centralizing data of the matrix, then solving a covariance matrix to obtain an initial parameter matrix, and iterating the initial parameter matrix based on orthogonal transformation to obtain an updated parameter matrix, and the method comprises the following steps:
step A1: the DATA of the matrix is centered, the mean is subtracted from each row in the DATA, and divided by the standard deviation of that row as follows:
step A2: solving the covariance matrix of the centralized matrix to obtain an initial parameter matrix as follows:
iterating the initial parameter matrix, including:
step A3: establishing an initialized identity matrix T-En;
step A4: determining the maximum value in the non-diagonal elements in the initial parameter matrix, and assuming the maximum value as dpq;
step A5: calculating a vector rotation angle of the initial parameter matrix according to the maximum value of the non-diagonal elements in the initial parameter matrix
Step A6: establishing an initialized intermediate variable orthogonal matrix U ═ En, and enabling U to be equal to Enpp＝Uqq＝cosθ，Uqp＝-Upq＝-sinθ；
Step A7: updating an initial parameter matrix based on an intermediate variable orthogonal matrix, performing rotation transformation on column vectors of the initial parameter matrix and an initialization unit matrix, and performing iteration D on the initial parameter matrix and the initialization unit matrixX＝UTDXU，T＝TU；
Step A7: steps a4 to a7 are repeated until dpq < (say ═ 0.0001).
Step S16, determining a principal component Tp with the largest contribution rate to the fault parameter in the updated parameter matrix after iteration; the column vectors in the updated parameter matrix after iteration can be arranged from left to right according to the magnitude of the modulus, wherein the modulus of the leftmost column vector is the largest, and meanwhile, the unit matrix T is synchronously transformed according to the updated parameter matrix after iteration, and then the unit matrix is transformed and corresponds to the parameters a1 and a2 … an, as shown in the following table:
watch 1
| a1 | a2 | … | an | |
| Principal component 1 | t11 | t12 | … | t1n |
| Principal component 2 | t21 | t22 | … | t2n |
| … | … | … | … | … |
| Principal component n | tn1 | tn2 | … | tnn |
And finding out a column vector corresponding to the fault parameter a2 from the table, finding out a maximum value from the column vector, and taking a principal component corresponding to a row corresponding to the maximum value as a principal component Tp with the maximum contribution rate to the fault parameter.
Step S17, extracting a parameter set Z with the contribution rate larger than a threshold value alpha in the principal component Tp; extracting all parameters larger than a threshold value alpha (for example, alpha is 0.3) from a line where the principal component Tp is located to form a parameter set Z; still assuming that the parameter a2 is an example of a fault parameter, and as shown in the above table i, assuming that the maximum value is determined to be t22 according to the parameter vector corresponding to the parameter a2, t22 is a main parameter with the highest probability of generating a current fault in the parameter vector corresponding to the fault, the principal component 2 corresponding to the row in which the main parameter t22 is located is the principal component with the highest contribution rate to the fault parameter, and a parameter with a contribution rate larger than the threshold value α is extracted from the row vector of the principal component 2, so as to form a parameter set Z ═ Z1, Z2, Z3 … zk }.
Step S18, calling a fault tree, and deducing possible fault reasons and probability thereof in the range of the parameter set Z; referring to fig. 5, the fault tree may be pre-established according to expert experience or historical fault diagnosis data, a binary tree is established according to parameter abnormal conditions and fault results, and leaf nodes are finally found by layer to obtain fault types. Still taking the above-mentioned formed parameter set Z ═ { Z1, Z2, Z3 … zk } as an example, it is assumed that there are three kinds Z1, Z2, and Z5 which are possible causes of failure by calling the failure tree and judging in the parameter set Z range, where all binary tree nodes where the failure cause Z1 is obtained include a1 and a2, all binary tree nodes where the failure cause Z2 is obtained include a3, and all binary tree nodes where the failure cause Z3 is obtained have a2, a6, a9, τ 1 ═ a1+ a2 is used as the corresponding probability index of the failure cause Z2, τ 2 ═ a 56 is used as the corresponding probability index of the failure cause Z2, τ 3 ═ a2+ a6+ a9 is used as the corresponding probability index of the failure cause Z3, and thus τ 1/(1 + 2+ 3), τ 2+ 863 + a6+ a 9) and τ 3/(τ 72 + Z3 + Z1) are used as the corresponding probability indexes of the failure cause Z3, τ 3/(τ 3 + 823 + 3 + Z/(τ 3), τ 3 + 3, z2, z 5.
And step S19, generating a report, storing and sending the report to the terminal equipment.
The fault diagnosis method provided by the above embodiment of the application combines principal component analysis with fault tree calling analysis, when a device fails, a cloud server acquires operation data within a set time period before the failure occurs, performs data cleaning and forms an initial parameter matrix, iterates the initial parameter matrix based on orthogonal transformation based on the principal component analysis, determines principal components with a probability of generating a current failure higher than a set value according to the iterated matrix, forms a target parameter set, can lock a range of parameters most likely to cause the failure occurrence by forming the target parameter set, calls a fault tree, analyzes the cause of the failure and the corresponding probability according to the parameters extracted from the target parameter set, and thus before calling the fault tree, performs principal component analysis on the data from the perspective of the data itself to find partial parameters most likely to cause the failure occurrence, the method can carry out targeted fault diagnosis, greatly reduce the troubleshooting range for determining the fault reason when the fault occurs, improve the fault diagnosis efficiency, and realize the purposes of quickly diagnosing the fault and improving the accuracy.
Referring to fig. 6, in another aspect of the present embodiment, a fault diagnosis apparatus is further provided, including a data obtaining module, configured to obtain, when a fault occurs, operation data in a set time period before a fault time point; the iteration module is used for determining an initial parameter matrix according to the operating data in the set time period, and performing iteration through orthogonal transformation based on the initial parameter matrix to obtain an updated parameter matrix; the parameter determining module is used for determining main parameters which cause that the probability of generating the current fault is higher than a set value in parameter vectors corresponding to the faults based on the updated parameter matrix, determining associated parameters according to the positions of the main parameters in the updated parameter matrix and forming a target parameter set; and the fault diagnosis module 23 is configured to determine a cause and a corresponding probability of a fault according to the target parameter set and the fault tree. Wherein, optionally, the data acquisition module can be data receiving module 21 partly, data receiving module 21 can include thing networking module, and cloud server passes through thing networking module and waits to examine equipment communication connection, follows examine equipment operation data and the storage of examining equipment direct or indirectly receiving equipment, when confirming when breaking down, the data acquisition module selects the operation data in the time quantum before the fault time point from the operation data that thing networking module received. Or, the data acquisition module may also extract the operation data in the set time period before the failure time point from other servers storing the operation data of the device to be detected through the internet of things module when the failure is determined.
Before the updated parameter matrix is obtained, determining whether the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than a threshold value; if the maximum value in the non-diagonal elements in the parameter matrix after iteration is not smaller than the threshold value, taking the parameter matrix after iteration as an updated initial parameter matrix, and returning to the step of performing iteration through orthogonal transformation based on the initial parameter matrix; and if the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than the threshold value, obtaining the updated parameter matrix.
The iteration module is specifically used for constructing an initialization identity matrix; calculating a vector rotation angle based on a maximum value among non-diagonal elements in the initial parameter matrix; and constructing an orthogonal matrix according to the vector rotation angle, performing rotation transformation on the column vectors of the initial parameter matrix and the initialization unit matrix based on the orthogonal matrix, and iterating the initial parameter matrix.
The iteration module is further configured to iterate the identity matrix according to the orthogonal matrix.
The parameter determining module is further configured to arrange column vectors in the updated parameter matrix according to a magnitude of a modulus value and synchronously transform the identity matrix according to the updated parameter matrix before determining a main parameter, which causes a probability of generating a current fault to be higher than a set value, in a parameter vector corresponding to the fault based on the updated parameter matrix.
The iteration module is specifically configured to determine a parameter vector corresponding to each parameter according to the operating data of each time point in the set time period; subtracting the mean value of the corresponding row from the data in the matrix formed by the parameter vectors, and dividing the mean value by the standard deviation of the corresponding row to obtain a centralized matrix; and obtaining the initial parameter matrix according to the covariance matrix of the centralized matrix.
The fault diagnosis device further comprises an analysis result report storage module 24, which is used for generating a report from the fault analysis result and storing the report; and/or generating a report form of the fault analysis result and sending the report form to the terminal.
It will be understood by those skilled in the art that the structure of the failure diagnosis apparatus shown in fig. 6 does not constitute a limitation of the failure diagnosis apparatus, and the respective modules may be entirely or partially implemented by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a controller in the computer device, or can be stored in a memory in the computer device in a software form, so that the controller can call and execute operations corresponding to the modules. In other embodiments, more or fewer modules than those shown in the figures may be included in the fault diagnosis apparatus, for example, the iteration module and the parameter determination module may incorporate the principal component analysis module 22, and the fault diagnosis module invokes the fault tree to determine the cause and the corresponding probability of the fault based on the analysis result of the principal component analysis module 22, so as to obtain the fault diagnosis result.
In another aspect of the embodiments of the present application, a storage medium is further provided, in which a computer program is stored, and when the computer program is executed by a processor, the processor is caused to execute the steps of the fault diagnosis method provided in any one of the embodiments of the present application.
In another aspect of the embodiments of the present application, a computer device is further provided, which includes a memory and a processor, where the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to execute the steps of the fault diagnosis method provided in any one of the above embodiments of the present application.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a non-volatile computer-readable storage medium, and can include the processes of the embodiments of the methods described above when the program is executed. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. The scope of the invention is to be determined by the scope of the appended claims.
Claims (10)
1. A fault diagnosis method, comprising:
when a fault occurs, acquiring operation data in a set time period before a fault time point;
determining an initial parameter matrix according to the operation data in the set time period, and performing iteration through orthogonal transformation based on the initial parameter matrix to obtain an updated parameter matrix;
determining main parameters which cause that the probability of generating the current fault is higher than a set value in parameter vectors corresponding to the faults based on the updated parameter matrix, and determining associated parameters according to the positions of the main parameters in the updated parameter matrix to form a target parameter set;
and determining the reason and the corresponding probability of the fault according to the target parameter set and the fault tree.
2. The fault diagnosis method according to claim 1, wherein before obtaining the updated parameter matrix, further comprising:
determining whether the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than a threshold value;
if the maximum value in the non-diagonal elements in the parameter matrix after iteration is not smaller than the threshold value, taking the parameter matrix after iteration as an updated initial parameter matrix, and returning to the step of performing iteration through orthogonal transformation based on the initial parameter matrix;
and if the maximum value of the non-diagonal elements in the parameter matrix after iteration is smaller than the threshold value, obtaining the updated parameter matrix.
3. The fault diagnosis method according to claim 1, wherein said iterating through an orthogonal transformation based on said initial parameter matrix comprises:
constructing an initialization identity matrix;
calculating a vector rotation angle based on a maximum value among non-diagonal elements in the initial parameter matrix;
and constructing an orthogonal matrix according to the vector rotation angle, performing rotation transformation on the column vectors of the initial parameter matrix and the initialization unit matrix based on the orthogonal matrix, and iterating the initial parameter matrix.
4. The fault diagnosis method according to claim 3, further comprising:
and iterating the unit matrix according to the orthogonal matrix.
5. The fault diagnosis method according to claim 3, wherein before determining the main parameters of the parameter vector corresponding to the fault that cause the probability of generating the current fault to be higher than the set value based on the updated parameter matrix, further comprising:
and arranging the column vectors in the updated parameter matrix according to the magnitude of the modulus, and synchronously transforming the unit matrix according to the updated parameter matrix.
6. The fault diagnosis method according to claim 1, wherein the determining an initial parameter matrix from the operational data over the set time period comprises:
determining a parameter vector corresponding to each parameter according to the operation data of each time point in the set time period;
subtracting the mean value of the corresponding row from the data in the matrix formed by the parameter vectors, and dividing the mean value by the standard deviation of the corresponding row to obtain a centralized matrix;
and obtaining the initial parameter matrix according to the covariance matrix of the centralized matrix.
7. The method according to any one of claims 1 to 6, wherein the determining the cause and the corresponding probability of the fault according to the target parameter set and the fault tree comprises:
generating a report form of the fault analysis result and storing the report form; and/or
And generating a report form of the fault analysis result and sending the report form to the terminal.
8. A failure diagnosis device characterized by comprising:
the data acquisition module is used for acquiring operation data in a set time period before a fault time point when the fault occurs;
the iteration module is used for determining an initial parameter matrix according to the operating data in the set time period, and performing iteration through orthogonal transformation based on the initial parameter matrix to obtain an updated parameter matrix;
the parameter determining module is used for determining main parameters which cause that the probability of generating the current fault is higher than a set value in parameter vectors corresponding to the faults based on the updated parameter matrix, determining associated parameters according to the positions of the main parameters in the updated parameter matrix and forming a target parameter set;
and the fault diagnosis module is used for determining the reason and the corresponding probability of the fault according to the target parameter set and the fault tree.
9. A storage medium storing a computer program which, when executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 7.
10. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010307946.8A CN111522308B (en) | 2020-04-17 | 2020-04-17 | Fault diagnosis method and device, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010307946.8A CN111522308B (en) | 2020-04-17 | 2020-04-17 | Fault diagnosis method and device, storage medium and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111522308A true CN111522308A (en) | 2020-08-11 |
| CN111522308B CN111522308B (en) | 2021-10-29 |
Family
ID=71904408
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010307946.8A Active CN111522308B (en) | 2020-04-17 | 2020-04-17 | Fault diagnosis method and device, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111522308B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112303811A (en) * | 2020-10-27 | 2021-02-02 | 珠海格力电器股份有限公司 | Processing method, device and system for air conditioner operation data, air conditioner and storage medium |
| CN112986734A (en) * | 2021-02-19 | 2021-06-18 | 北京首钢自动化信息技术有限公司 | Fault detection device of voice communication system |
| CN113868236A (en) * | 2021-09-29 | 2021-12-31 | 东风汽车有限公司东风日产乘用车公司 | Electric vehicle fault monitoring method, device and computer readable storage medium |
| CN115712840A (en) * | 2022-11-15 | 2023-02-24 | 中国人民解放军陆军工程大学 | Multi-fault diagnosis method and system for electronic information system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6498993B1 (en) * | 2000-05-30 | 2002-12-24 | Gen Electric | Paper web breakage prediction using bootstrap aggregation of classification and regression trees |
| CN201017225Y (en) * | 2006-12-22 | 2008-02-06 | 浙江大学 | Polymerization of propylene production data detecting and failure diagnosis device |
| US20180060448A1 (en) * | 2015-03-27 | 2018-03-01 | Hitachi, Ltd. | Computer and method of creating graph data |
| CN108020781A (en) * | 2017-12-19 | 2018-05-11 | 上海电机学院 | A kind of circuit breaker failure diagnostic method |
| CN108227653A (en) * | 2017-12-28 | 2018-06-29 | 湖州师范学院 | A kind of large-scale nonlinear course monitoring method based on randomization core pivot element analysis |
| CN108255783A (en) * | 2018-01-12 | 2018-07-06 | 上海海事大学 | A kind of marine vessel power drive system method for diagnosing faults based on wavelet analysis |
| CN109459993A (en) * | 2018-12-06 | 2019-03-12 | 湖南师范大学 | A kind of process flow industry process online adaptive Fault monitoring and diagnosis method |
| CN109919446A (en) * | 2019-02-01 | 2019-06-21 | 北京航空航天大学 | A kind of quality accident root primordium recognition methods |
| CN110288022A (en) * | 2019-06-26 | 2019-09-27 | 华北水利水电大学 | A kind of vibration fault diagnosis of hydro-turbine generating unit algorithm of extracting parameter non-linear relation |
-
2020
- 2020-04-17 CN CN202010307946.8A patent/CN111522308B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6498993B1 (en) * | 2000-05-30 | 2002-12-24 | Gen Electric | Paper web breakage prediction using bootstrap aggregation of classification and regression trees |
| CN201017225Y (en) * | 2006-12-22 | 2008-02-06 | 浙江大学 | Polymerization of propylene production data detecting and failure diagnosis device |
| US20180060448A1 (en) * | 2015-03-27 | 2018-03-01 | Hitachi, Ltd. | Computer and method of creating graph data |
| CN108020781A (en) * | 2017-12-19 | 2018-05-11 | 上海电机学院 | A kind of circuit breaker failure diagnostic method |
| CN108227653A (en) * | 2017-12-28 | 2018-06-29 | 湖州师范学院 | A kind of large-scale nonlinear course monitoring method based on randomization core pivot element analysis |
| CN108255783A (en) * | 2018-01-12 | 2018-07-06 | 上海海事大学 | A kind of marine vessel power drive system method for diagnosing faults based on wavelet analysis |
| CN109459993A (en) * | 2018-12-06 | 2019-03-12 | 湖南师范大学 | A kind of process flow industry process online adaptive Fault monitoring and diagnosis method |
| CN109919446A (en) * | 2019-02-01 | 2019-06-21 | 北京航空航天大学 | A kind of quality accident root primordium recognition methods |
| CN110288022A (en) * | 2019-06-26 | 2019-09-27 | 华北水利水电大学 | A kind of vibration fault diagnosis of hydro-turbine generating unit algorithm of extracting parameter non-linear relation |
Non-Patent Citations (2)
| Title |
|---|
| DING S ETAL: "On the application of PCA_technique to fault diagnosis", 《TSINGHUA SCIENCE AND TECHNOLOGY》 * |
| 马智明 等: "主成分分析法在故障诊断中的应用", 《数据统计与管理》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112303811A (en) * | 2020-10-27 | 2021-02-02 | 珠海格力电器股份有限公司 | Processing method, device and system for air conditioner operation data, air conditioner and storage medium |
| CN112303811B (en) * | 2020-10-27 | 2021-12-14 | 珠海格力电器股份有限公司 | Processing method, device and system for air conditioner operation data, air conditioner and storage medium |
| CN112986734A (en) * | 2021-02-19 | 2021-06-18 | 北京首钢自动化信息技术有限公司 | Fault detection device of voice communication system |
| CN113868236A (en) * | 2021-09-29 | 2021-12-31 | 东风汽车有限公司东风日产乘用车公司 | Electric vehicle fault monitoring method, device and computer readable storage medium |
| CN113868236B (en) * | 2021-09-29 | 2023-06-30 | 东风汽车有限公司东风日产乘用车公司 | Electric vehicle fault monitoring method, equipment and computer readable storage medium |
| CN115712840A (en) * | 2022-11-15 | 2023-02-24 | 中国人民解放军陆军工程大学 | Multi-fault diagnosis method and system for electronic information system |
| CN115712840B (en) * | 2022-11-15 | 2023-06-13 | 中国人民解放军陆军工程大学 | Multi-fault diagnosis method and system for electronic information system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111522308B (en) | 2021-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111522308B (en) | Fault diagnosis method and device, storage medium and computer equipment | |
| CN111897673B (en) | Operation and maintenance fault root cause identification method and device, computer equipment and storage medium | |
| CN111108362B (en) | Abnormal sound detection device, abnormal model learning device, abnormal sound detection method, and recording medium | |
| CN110378052B (en) | Equipment residual life prediction method considering future working conditions based on cyclic neural network | |
| Benkwitz et al. | Problems related to confidence intervals for impulse responses of autoregressive processes | |
| US20070061144A1 (en) | Batch statistics process model method and system | |
| CN111338836A (en) | Method, device, computer equipment and storage medium for processing fault data | |
| CN111314173B (en) | Monitoring information abnormity positioning method and device, computer equipment and storage medium | |
| CN110232678B (en) | Image uncertainty prediction method, device, equipment and storage medium | |
| CN110912908B (en) | Network protocol anomaly detection method and device, computer equipment and storage medium | |
| CN112527604A (en) | Deep learning-based operation and maintenance detection method and system, electronic equipment and medium | |
| CN111897695A (en) | Method and device for acquiring KPI abnormal data sample and computer equipment | |
| WO2014179873A1 (en) | Quantitative analysis of signal related measurements for trending and pattern recognition | |
| CN111259952A (en) | Abnormal user identification method and device, computer equipment and storage medium | |
| CN111207926A (en) | Fault diagnosis method based on rolling bearing, electronic device and storage medium | |
| CN114997744A (en) | Equipment health assessment method and device, computer equipment and medium | |
| CN111124898B (en) | Question-answering system testing method and device, computer equipment and storage medium | |
| CN110570487B (en) | Undersampling model generation method, image reconstruction method, device and computer equipment | |
| CN113110961B (en) | Equipment abnormality detection method and device, computer equipment and readable storage medium | |
| CN113672782A (en) | Internet of things equipment importance matching method based on data query | |
| US20230161653A1 (en) | Method of managing system health | |
| CN113593581B (en) | Voiceprint discrimination method, voiceprint discrimination device, computer device and storage medium | |
| CN111160464B (en) | Industrial high-order dynamic process soft measurement method based on multi-hidden-layer weighted dynamic model | |
| CN109842689B (en) | Method and device for selecting server access | |
| CN112307086A (en) | Automatic data verification method and device in fire service |
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 |