CN113359682B - Equipment fault prediction method, device, equipment fault prediction platform and medium - Google Patents
Equipment fault prediction method, device, equipment fault prediction platform and medium Download PDFInfo
- Publication number
- CN113359682B CN113359682B CN202110738714.2A CN202110738714A CN113359682B CN 113359682 B CN113359682 B CN 113359682B CN 202110738714 A CN202110738714 A CN 202110738714A CN 113359682 B CN113359682 B CN 113359682B
- Authority
- CN
- China
- Prior art keywords
- state data
- dimension
- fault
- prediction model
- data set
- 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
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
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
-
- 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
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The application provides a device fault prediction method, a device fault prediction platform and a medium, and relates to the technical field of fault prediction. The method comprises the following steps: acquiring a state data set of a target device, wherein the state data set comprises state data of multiple dimensions of the target device; analyzing and processing the state data of each dimension in the state data set to construct a feature vector; and inputting the characteristic vector into a pre-trained fault prediction model to obtain the fault type of the target equipment output by the fault prediction model. By applying the embodiment of the application, the equipment fault type can be predicted in advance, and the phenomena of insufficient maintenance, excessive maintenance or maintainability fault are avoided.
Description
Technical Field
The present application relates to the field of failure prediction technologies, and in particular, to a method and an apparatus for predicting an equipment failure, an equipment failure prediction platform, and a medium.
Background
The equipment and test maintenance are currently an important topic and are problems to be solved urgently in the intelligent manufacturing industry, and the fault diagnosis of the equipment is a premise for realizing predictive maintenance.
At present, the time is mainly taken as a dimension, and equipment is regularly detected to ensure long-term stable operation of the equipment. However, this approach has the phenomena of "under-repair", "over-repair", or "serviceability failure".
Disclosure of Invention
An object of the present invention is to provide a method, an apparatus, a platform and a medium for predicting a device failure, which can predict a device failure type in advance, so as to avoid the phenomena of "insufficient maintenance", "excessive maintenance" or "maintainability failure".
In order to achieve the above purpose, the technical solutions adopted in the embodiments of the present application are as follows:
in a first aspect, an embodiment of the present application provides an apparatus failure prediction method, where the method includes:
acquiring a state data set of a target device, wherein the state data set comprises state data of multiple dimensions of the target device;
analyzing and processing the state data of each dimension in the state data set to construct a feature vector;
and inputting the characteristic vector into a pre-trained fault prediction model to obtain the fault type of the target equipment output by the fault prediction model.
Optionally, the analyzing and processing the state data of each dimension in the state data set to construct a feature vector includes:
determining a correlation coefficient between state data of each dimension in the state data set;
and constructing the feature vector according to the correlation coefficient.
Optionally, before determining a correlation coefficient between the status data of each dimension in the status data set, the method further includes:
preprocessing the state data of each dimension in the state data set to obtain a preprocessed state data set;
the determining a correlation coefficient between state data of each dimension in the state data set includes:
and determining a correlation coefficient between the state data of each dimension in the preprocessed state data set.
Optionally, the state data set includes frequency state data, and after obtaining the fault type of the target device output by the fault prediction model, the method further includes:
performing spectrum analysis on the frequency state data in the state data set to obtain a spectrum analysis result;
and determining the unit device of the fault type in the target equipment according to the spectrum analysis result.
Optionally, before the feature vector is input into a pre-trained fault prediction model and the fault type of the target device output by the fault prediction model is obtained, the method further includes:
constructing a training sample and a testing sample according to historical state data of a plurality of devices;
training an initial fault prediction model by using the training sample to obtain an intermediate fault prediction model;
and predicting the intermediate fault prediction model by using the test sample to obtain the fault prediction model.
Optionally, after the feature vector is input into a pre-trained fault prediction model and a fault type of the target device output by the fault prediction model is obtained, the method further includes:
and generating and outputting early warning information according to the fault type of the target equipment.
Optionally, the method further comprises:
generating a diagnosis report according to the fault type of the target equipment, wherein the diagnosis report comprises: the fault type and at least one fault reason corresponding to the fault type.
In a second aspect, an embodiment of the present application provides an apparatus for predicting a device failure, where the apparatus includes:
the device comprises an acquisition module, a storage module and a display module, wherein the acquisition module is used for acquiring a state data set of a target device, and the state data set comprises state data of multiple dimensions of the target device;
the building module is used for analyzing and processing the state data of each dimension in the state data set and building a feature vector;
and the determining module is used for inputting the feature vector into a pre-trained fault prediction model to obtain the fault type of the target equipment output by the fault prediction model.
Optionally, the building module is specifically configured to determine a correlation coefficient between state data of each dimension in the state data set; and constructing the feature vector according to the correlation coefficient.
Optionally, the apparatus further comprises: a processing module;
the processing module is used for preprocessing the state data of each dimension in the state data set, obtaining a preprocessed state data set;
correspondingly, the determining module is further configured to determine a correlation coefficient between the state data of each dimension in the preprocessed state data set.
Optionally, the status data set includes frequency status data, and the apparatus further includes: an analysis module;
the analysis module is used for carrying out spectrum analysis on the frequency state data in the state data set to obtain a spectrum analysis result;
correspondingly, the determining module is further configured to determine, according to the spectrum analysis result, a unit device of the type of the fault occurring in the target device.
Optionally, the apparatus further comprises: a training module and a prediction module;
the building module is further used for building a training sample and a testing sample according to the historical state data of the plurality of devices;
the training module is used for training an initial fault prediction model by using the training sample to obtain an intermediate fault prediction model;
and the prediction module is used for predicting the intermediate fault prediction model by using the test sample to obtain the fault prediction model.
Optionally, the apparatus further comprises: a generation module;
the generating module is further used for generating and outputting early warning information according to the fault type of the target equipment.
Optionally, the generating module is further configured to generate a diagnostic report according to the fault type of the target device, where the diagnostic report includes: the fault type and at least one fault reason corresponding to the fault type.
In a third aspect, an embodiment of the present application provides an apparatus failure prediction platform, including: the device fault prediction platform comprises a processor, a storage medium and a bus, wherein the storage medium stores machine readable instructions executable by the processor, when the device fault prediction platform runs, the processor and the storage medium communicate through the bus, and the processor executes the machine readable instructions to execute the steps of the device fault prediction method of the first aspect.
In a fourth aspect, the present application provides a storage medium, where a computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the device failure prediction method of the first aspect are performed.
The beneficial effect of this application is:
the embodiment of the application provides an equipment fault prediction method, an equipment fault prediction device, an equipment fault prediction platform and a medium, which can establish an association relationship between a fault type and state data of each dimension according to a relationship between a historical state data set and the fault type, analyze and process the state data of each dimension in the state data set corresponding to target equipment, determine the association relationship between the state data of each dimension and state data of other dimensions, construct a feature vector corresponding to the state data of each dimension based on the association relationship between the state data of each dimension and the state data of other dimensions, finally input the feature vector corresponding to the state data of each dimension into a pre-trained fault prediction model respectively, and predict the fault type of the target equipment in advance through the fault prediction model, so that the phenomena of 'insufficient maintenance', 'excessive maintenance' or 'maintainability fault' can be avoided.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
Fig. 1 is a schematic flowchart of an apparatus failure prediction method according to an embodiment of the present disclosure;
fig. 2 is a schematic flowchart of another method for predicting a failure of a device according to an embodiment of the present disclosure;
fig. 3 is a flowchart illustrating another method for predicting a device failure according to an embodiment of the present disclosure;
fig. 4 is a schematic flowchart of another method for predicting a device failure according to an embodiment of the present disclosure;
fig. 5 is a schematic diagram illustrating a display of warning information according to an embodiment of the present disclosure;
fig. 6 is a schematic diagram of a spectrum of state data according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of an apparatus failure prediction device according to an embodiment of the present disclosure;
fig. 8 is a schematic structural diagram of an equipment failure prediction platform according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
Before explaining the embodiments of the present application in detail, an application scenario of the present application will be described first. The application scenario may be specifically a scenario for predicting the health condition of the electromechanical device in the coal mine field, and may also be a scenario for predicting the health condition of the electromechanical device in other fields, which is not limited in the present application. The method includes the steps that electromechanical equipment in the coal mine field is taken as an example for explanation, the electromechanical equipment can comprise lifting equipment, pressure equipment, drainage equipment, driving equipment and the like, for one electromechanical equipment (target equipment), an equipment fault prediction platform obtains state data of the target equipment through acquisition equipment, fault types of the target equipment are predicted through the state data of the target equipment and a pre-trained fault prediction model, the equipment fault prediction platform can be installed on terminal equipment, and the specific form of the terminal equipment can be a tablet computer, a notebook computer, a desktop computer and the like.
The equipment fault prediction platform can obtain a fault prediction model through historical state data of a plurality of electromechanical equipment and a preset deep neural network framework, such as Long-Short Term Memory (LSTM) network framework training, the fault prediction model can predict whether the target equipment is in fault in advance, and the fault type of the fault is in fault, namely, the target equipment can be subjected to predictive maintenance, so that the automation level of coal mine electromechanical equipment management is improved, and the quality, the stability and the safety of the coal mine electromechanical equipment are ensured.
The device failure prediction method mentioned in the present application is exemplified as follows with reference to the accompanying drawings. Fig. 1 is a schematic flowchart of an apparatus failure prediction method provided in an embodiment of the present application, where the method is applicable to a background (e.g., a server) that supports the operation of the above-mentioned apparatus failure prediction platform. As shown in fig. 1, the method may include:
s101, acquiring a state data set of the target device, wherein the state data set comprises state data of multiple dimensions of the target device.
The multidimensional state data can indicate the types of the state parameters in a finger-like manner, wherein the state data are used for indicating specific values of the state parameters, and the dimensions are used for indicating the types of the state parameters. The type of the state parameter is related to the type of the monitoring device, the monitoring device related to the target equipment can be set in advance according to the type of the target equipment, state data of certain dimensionality of the target equipment can be obtained through the monitoring device, if the target equipment is pressure equipment, a pressure sensor can be set, and the pressure parameter of the pressure equipment can be obtained through the pressure sensor; if the target device is a driving device, a frequency monitoring device may be set, and the vibration information of the driving device is obtained through the frequency monitoring device, so as to obtain a frequency parameter, that is, the dimension of the state data of the target device has a certain correlation with the type of the target device.
It should be noted that the state parameters of the target device may further include a running time parameter, an installation location parameter, and the like, and the server may obtain specific values of the state parameters of multiple dimensions of the target device according to a preset period, form a state data set with the specific values of the state parameters of each type, and store the identifier of the target device in association with the state data set. The state data set may include, but is not limited to, state parameters such as current, voltage, frequency, temperature, displacement, speed, etc.
And S102, analyzing and processing the state data of each dimension in the state data set, and constructing a feature vector.
The incidence relation between the state parameters of all the dimensions can be established according to the relation between the historical state data set and the fault type, and the incidence analysis can be carried out on the state data of all the dimensions in the state data set of the target equipment according to the incidence relation.
Specifically, the state data of the abnormal dimension can be determined according to the historical state data sets corresponding to the fault types, for a plurality of historical state data sets corresponding to the same fault type, the association relationship between the state data of each abnormal dimension and the state data of other abnormal dimensions can be determined according to the state data of the abnormal dimension in each historical state data set, and by analogy, the association relationship between the dimensions corresponding to the state data can be finally established. And determining the association relationship between the state data of each dimension in the state data set of the target device according to the association relationship between the dimensions corresponding to the state data, and constructing a feature vector corresponding to the state data of each dimension according to the association relationship between the state data of each dimension and the state data of other dimensions.
And S103, inputting the feature vector into a pre-trained fault prediction model to obtain the fault type of the target equipment output by the fault prediction model.
The feature vectors corresponding to the state data of each dimension are respectively input into the fault prediction model, the fault prediction model can determine the matching degree between the target device and a plurality of fault types, and finally the fault type with the highest matching degree can be determined as the fault type of the target device. The fault types may include two types, one type is a mechanical fault type, such as a resonance fault, an imbalance fault, an error fault, a bearing damage fault, and the like, and the other type is an electrical fault, such as a stator magnetic field fault, a rotor fracture fault, and the like.
To sum up, in the device fault prediction method provided by the application, the incidence relation between the fault type and the state data of each dimension can be established according to the relation between the historical state data set and the fault type, the state data of each dimension in the state data set corresponding to the target device is analyzed and processed, the incidence relation between the state data of each dimension and the state data of other dimensions can be determined, the feature vector corresponding to the state data of each dimension can be established based on the incidence relation between the state data of each dimension and the state data of other dimensions, finally the feature vector corresponding to the state data of each dimension is respectively input into the pre-trained fault prediction model, and the fault type existing in the target device can be predicted in advance through the fault prediction model, so that the phenomena of 'insufficient maintenance', 'excessive maintenance' or 'maintainability fault' can be avoided.
Fig. 2 is a schematic flowchart of another device failure prediction method according to an embodiment of the present disclosure. Optionally, as shown in fig. 2, the analyzing and processing the state data of each dimension in the state data set to construct the feature vector includes:
s201, determining a correlation coefficient between state data of each dimension in the state data set.
And determining the incidence relation among the state data of each dimension in the state data set of the target equipment according to the incidence relation among the dimensions corresponding to the state data. Specifically, the association relationship between the dimensions corresponding to the respective pieces of state data, that is, the association relationship between the pieces of state data of various types, is determined in the following manner. For example, assuming that the fault types corresponding to three historical state data sets (e.g., state data set 1, state data set 2, and state data set 3) are all a-type faults, the state data set 1 may include a plurality of types of state data (e.g., state data a, state data b, state data c, and state data d … …), where the state data a and the state data c are both abnormal data; if the status data b in the status data set 2 is abnormal data and the status data b and the status data c in the status data set 3 are both abnormal data, it can be found that the status data a has a correlation with the status data b and the status data c, respectively.
That is, the correlation between the state data a and the state data b, and the correlation between the state data c are stronger than the correlation between other state data (e.g., the state data d), and the correlation between the state data a and the state data c is stronger than the correlation between the state data a and the state data b, and the correlation between the state data a and the state data b can be determined, and the strength of the correlation can be represented by the magnitude of the correlation coefficient, where the correlation coefficient between the state data a and the state data b can be represented by a1, the correlation coefficient between the state data a and the state data c can be represented by a2, and it can be known that a2 is greater than a1, and the correlation coefficient between each type of state data and other types of state data is stored in the database in the form of a list.
According to the obtained association relationship between the state data of each dimension (type), the association relationship between the state data of each dimension in the state data set of the target device can be determined, and the feature vector corresponding to the state data of each dimension can be constructed according to the association relationship between the state data of each dimension and the state data of other dimensions. Continuing the above example, assuming that the state data set corresponding to the target device includes the state data with the dimension a, the state data with the dimension b, and the state data with the dimension c, the correlation coefficient between the state data with the dimension a and the state data with the dimension b is a1, and the correlation coefficient between the state data with the dimension a and the state data with the dimension c is a2, and similarly, the correlation coefficient between the state data with the dimension a and the state data with other dimensions can be obtained by looking up the table, and finally, the correlation coefficient between the state data with each dimension in the state data set corresponding to the target device can be determined.
And S202, constructing a feature vector according to the correlation coefficient.
The state data of each dimension can be coded to obtain a coding result, and then the correlation coefficient between the state data of each dimension and the state data of other dimensions is combined to multiply the correlation coefficient with the corresponding coding result, so that the feature vector corresponding to the state data of each dimension can be finally obtained.
Therefore, the state data of multiple dimensions are connected, and the state data of each dimension has relevance of different degrees, so that the state data of each dimension is not an individual, the fault type of the target equipment can be more accurately predicted, an accurate and reliable basis can be provided for later maintenance management, the maintenance cost is saved, and accidental shutdown and major accidents of the target equipment are avoided. The scientization, standardization, systematization and informatization of equipment maintenance management are realized, the equipment maintenance management level is improved, and the safety and reliability of equipment operation are improved.
Optionally, before determining the correlation coefficient between the state data of each dimension in the state data set, the method further includes: and preprocessing the state data of each dimension in the state data set to obtain a preprocessed state data set.
Specifically, the dimension to which the state data that can cause the failure of the driving device belongs can be determined according to the corresponding relationship between the historical state data set and the failure type corresponding to the device (such as the driving device) of the same type as the target device, that is, the state data of some dimensions may not affect whether the driving device fails at all, the state data of the dimensions can be called as state data to be deleted, the state data matched with the dimension of the state data to be deleted can be determined from the state data set corresponding to the driving device and deleted, and then the correlation coefficient between the state data of the dimensions in the deleted state data set can be determined.
Fig. 3 is a flowchart illustrating another method for predicting a device failure according to an embodiment of the present disclosure. Optionally, as shown in fig. 3, the state data set includes frequency state data, and after obtaining the fault type of the target device output by the fault prediction model, the method may further include:
s301, performing spectrum analysis on the frequency state data in the state data set to obtain a spectrum analysis result.
S302, determining the unit device of the fault type in the target equipment according to the spectrum analysis result.
The target device has a corresponding working frequency, the frequency state data in the state data set corresponding to the target device is obtained by analyzing the acquired vibration signal by using a fast Fourier transform technology, the frequency state data is compared with the working frequency corresponding to the target device, the relationship between the frequency state data and the working frequency corresponding to the target device is determined, and if the frequency state data is a fraction multiple of the working frequency corresponding to the target device, the frequency spectrum analysis result is that the frequency state data of the target device is smaller than the working frequency.
Based on the above, the unit device of the type of the fault occurring in the target equipment can be determined according to the relationship between the fault frequency section corresponding to each unit device in the target equipment and the frequency state data of the target equipment. Assuming that the frequency state data of the target device just falls within the fault frequency interval corresponding to the unit apparatus 1, it can be predicted that the fault type of the unit apparatus 1 in the target device that is about to fail is the fault type corresponding to the target device.
Therefore, the frequency spectrum analysis is carried out on the frequency state data in the state data set corresponding to the target equipment, so that the fault of the fault type, which is about to occur in the unit device in the target equipment, can be obtained more accurately, and a more accurate and more reliable basis can be provided for the later dimension management.
Fig. 4 is a schematic flowchart of another method for predicting a device failure according to an embodiment of the present disclosure. Optionally, as shown in fig. 4, before the feature vector is input into a pre-trained fault prediction model to obtain a fault type of the target device output by the fault prediction model, the method further includes:
s401, constructing a training sample and a testing sample according to historical state data of a plurality of devices.
Wherein status data corresponding to different historical moments of a plurality of devices can be collected, the status data corresponding to each historical moment can comprise a plurality of dimensions, the state data of different dimensions for different historical moments of each device may constitute a state data set. For one device at a historical time, not only can a state data set corresponding to the device at the historical time be obtained, but also a real fault type corresponding to the device at the historical time can be obtained, the state data set and the real fault type can be combined into a group of samples, and finally, a plurality of groups of samples can be obtained, wherein the characteristics in each group of samples are state data of each dimension in the state data set, and the label is the real fault type. According to the training parameters, training samples for training the initial fault model and test samples for testing the trained model can be determined from the multiple groups of samples.
S402, training the initial fault prediction model by using the training samples to obtain an intermediate fault prediction model.
The neural network framework of the initial fault prediction model can be a Long-Short Term Memory (LSTM) network framework, characteristics in a training sample are used as input of the initial fault prediction model, a label in the training sample is used as output of the initial fault prediction model, and the initial fault prediction model is trained, that is, the initial fault prediction model can obtain a predicted fault type according to input state data, the predicted fault type is compared with a real fault type, learning parameters in the initial fault prediction model are adjusted according to errors between the predicted fault type and the real fault type until a training stop condition is met, and then an intermediate fault prediction model can be obtained.
And S403, predicting the intermediate fault prediction model by using the test sample to obtain a fault prediction model.
Inputting the characteristics in the test sample into the intermediate fault prediction model, wherein the intermediate fault prediction model can output the fault type corresponding to the characteristics in the test sample, comparing the fault type with the label in the test sample to obtain the accuracy of the intermediate fault prediction model, comparing the accuracy with a preset accuracy, when the accuracy is greater than or equal to the preset accuracy, representing that the intermediate fault prediction model is the fault prediction model, when the accuracy is less than the preset accuracy, continuously adjusting the learning parameters in the intermediate fault prediction model until the calculated accuracy is greater than or equal to the preset accuracy, and then taking the corresponding intermediate fault prediction model as the fault prediction model.
It can be seen that the health condition of each device can be quantitatively analyzed by using the trained fault prediction model.
Optionally, after the feature vector is input into the pre-trained fault prediction model and the fault type of the target device output by the fault prediction model is obtained, the method further includes: and generating and outputting early warning information according to the fault type of the target equipment.
On the premise that the fault type of the target device is known, the state data belonging to the abnormal dimension can be determined according to the relationship between the fault type and the state data, for example, if the fault type is resonance, the state data which may cause resonance includes frequency, speed and the like, that is, the warning information may include state data which needs to be focused, such as frequency data and speed data generated by the target device, and the frequency point and the speed point which exceed the threshold value may be identified by a reminding icon in a chart displaying the frequency data and the speed data. Fig. 5 is a schematic diagram illustrating display of warning information according to an embodiment of the present application, as shown in fig. 5, a worker may retrieve status data (e.g., speed data) related to a failure of a target device from a device failure prediction platform according to an actual demand, display the speed data in a manner of displaying a line graph, and display graph information of the speed data on an interface of the device failure prediction platform, where an abscissa in the graph represents time and an ordinate represents speed, and if the speed at a certain time is abnormal, the graph information may be identified by a reminder icon (e.g., an origin).
Therefore, the working personnel can know the health state of the target equipment more intuitively and react in time when the target equipment has potential faults.
Optionally, the method may further include: generating a diagnosis report according to the fault type of the target equipment, wherein the diagnosis report comprises: the fault type and at least one fault reason corresponding to the fault type.
Each fault type can correspond to at least one fault reason, the association relationship between the fault type and the fault reason can be stored in a database in advance, after the fault type of the target equipment is determined, the corresponding fault reason can be searched from the database, the fault type of the target equipment and the corresponding fault reason are recorded in a diagnosis report, the diagnosis report is stored in the database, and the equipment fault prediction platform can support workers to browse the diagnosis report and can download and print the diagnosis report in various file forms (such as PDF and XML).
Optionally, after the staff actually examines the target device, some information in the diagnosis report may be modified according to the examination result, for example, screening multiple initial fault causes, and the target device generates the fault cause of the fault type accurately.
Therefore, the fault reasons are added in the diagnosis report, so that reliable basis can be provided for later maintenance management, maintenance management personnel can efficiently maintain and manage the target equipment, and maintenance cost is saved. The diagnosis report is corrected through actual investigation, so that the fault prediction process of the target equipment can be optimized, and the fault prediction precision is improved.
Optionally, after acquiring the multi-dimensional state data of the target device, the device failure prediction platform may display the multi-dimensional data in a preset display manner, for example, the acceleration state data of the target device may be displayed by using a oscillogram; and respectively displaying the state data of speed, acceleration, displacement and the like by using the spectrogram. Fig. 6 is a schematic diagram of a spectrogram of state data according to an embodiment of the present application, where an abscissa on the spectrogram in fig. 6 represents frequency and an ordinate represents amplitude.
Therefore, the working state of the target equipment can be checked more visually from different angles by the staff, and the staff can know the key information of the target equipment in time conveniently.
Fig. 7 is a schematic structural diagram of an apparatus failure prediction device according to an embodiment of the present application. As shown in fig. 7, the apparatus may include:
an obtaining module 701, configured to obtain a state data set of a target device, where the state data set includes state data of multiple dimensions of the target device;
a building module 702, configured to analyze and process state data of each dimension in the state data set, and build a feature vector;
the determining module 703 is configured to input the feature vector into a pre-trained fault prediction model, and obtain a fault type of the target device output by the fault prediction model.
Optionally, the building module 702 is specifically configured to determine a correlation coefficient between state data of each dimension in the state data set; and constructing a feature vector according to the correlation coefficient.
Optionally, the apparatus further comprises: a processing module;
the processing module is used for preprocessing the state data of each dimensionality in the state data set to obtain a preprocessed state data set;
correspondingly, the determining module 703 is further configured to determine a correlation coefficient between the state data of each dimension in the preprocessed state data set.
Optionally, the status data set includes frequency status data, and the apparatus further includes: an analysis module;
the analysis module is used for carrying out spectrum analysis on the frequency state data in the state data set to obtain a spectrum analysis result;
correspondingly, the determining module 703 is further configured to determine, according to the result of the spectrum analysis, a unit device of the type of the fault occurring in the target device.
Optionally, the apparatus further comprises: a training module and a prediction module;
the building module 702 is further configured to build a training sample and a testing sample according to historical state data of the multiple devices;
the training module is used for training an initial fault prediction model by using a training sample to obtain an intermediate fault prediction model;
the prediction module is used for predicting the intermediate fault prediction model by using the test sample to obtain a fault prediction model.
Optionally, the apparatus further comprises: a generation module;
the generating module is further used for generating and outputting early warning information according to the fault type of the target equipment.
Optionally, the generating module is further configured to generate a diagnosis report according to the fault type of the target device, where the diagnosis report includes: the fault type and at least one fault reason corresponding to the fault type.
The above-mentioned apparatus is used for executing the method provided by the foregoing embodiment, and the implementation principle and technical effect are similar, which are not described herein again.
These above modules may be one or more integrated circuits configured to implement the above methods, such as: one or more Application Specific Integrated Circuits (ASICs), or one or more microprocessors, or one or more Field Programmable Gate Arrays (FPGAs), etc. For another example, when one of the above modules is implemented in the form of a Processing element scheduler code, the Processing element may be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).
Fig. 8 is a schematic structural diagram of an equipment failure prediction platform provided in an embodiment of the present application, and as shown in fig. 8, the equipment failure prediction platform may include: a processor 801, a storage medium 802, and a bus 803, wherein the storage medium 802 stores machine-readable instructions executable by the processor 801, and when the device failure prediction platform is operating, the processor 801 communicates with the storage medium 802 via the bus 803, and the processor 801 executes the machine-readable instructions to perform the steps of the above-described method embodiments. The specific implementation and technical effects are similar, and are not described herein again.
Optionally, the present application further provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the computer program performs the steps of the above method embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In another aspect, the indirect coupling or communication connection of devices or units may be electrical, mechanical or other.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute some steps of the methods according to the embodiments of the present application. And the aforementioned storage medium includes: a U disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (9)
1. A method for predicting a failure of a device, the method comprising:
acquiring a state data set of a target device, wherein the state data set comprises state data of multiple dimensions of the target device;
analyzing and processing the state data of each dimension in the state data set to construct a feature vector;
inputting the characteristic vector into a pre-trained fault prediction model to obtain the fault type of the target equipment output by the fault prediction model;
analyzing and processing the state data of each dimensionality in the state data set, and constructing a feature vector, wherein the analyzing and processing comprises the following steps:
determining a correlation coefficient between the state data of each dimension in the state data set, wherein the correlation coefficient between the state data of each dimension is used for indicating the correlation coefficient between the state data of each dimension and the state data of other dimensions;
constructing the feature vector according to the correlation coefficient;
the constructing the feature vector according to the correlation coefficient includes:
coding the state data of each dimension to obtain a coding result of the state data of each dimension;
and multiplying the correlation coefficient between the state data of each dimension and the state data of other dimensions by the encoding result of the state data of each dimension to construct the feature vector of the state data of each dimension.
2. The method of claim 1, wherein prior to determining the correlation coefficient between the state data for each dimension in the set of state data, the method further comprises:
preprocessing the state data of each dimension in the state data set to obtain a preprocessed state data set;
the determining a correlation coefficient between state data of each dimension in the state data set includes:
and determining a correlation coefficient between the state data of each dimension in the preprocessed state data set.
3. The method of claim 1 or 2, wherein the state data set includes frequency state data, and wherein after obtaining the fault type of the target device output by the fault prediction model, the method further comprises:
performing spectrum analysis on the frequency state data in the state data set to obtain a spectrum analysis result;
and determining the unit device with the fault type in the target equipment according to the spectrum analysis result.
4. The method of claim 1, wherein before inputting the feature vector into a pre-trained fault prediction model and obtaining the fault type of the target device output by the fault prediction model, the method further comprises:
constructing a training sample and a testing sample according to historical state data of a plurality of devices;
training an initial fault prediction model by using the training sample to obtain an intermediate fault prediction model;
and predicting the intermediate fault prediction model by using the test sample to obtain the fault prediction model.
5. The method of claim 1, wherein after inputting the feature vector into a pre-trained fault prediction model and obtaining the fault type of the target device output by the fault prediction model, the method further comprises:
and generating and outputting early warning information according to the fault type of the target equipment.
6. The method of claim 5, further comprising:
generating a diagnosis report according to the fault type of the target equipment, wherein the diagnosis report comprises: the fault type and at least one fault reason corresponding to the fault type.
7. An apparatus for predicting a failure of a device, the apparatus comprising:
the device comprises an acquisition module, a storage module and a display module, wherein the acquisition module is used for acquiring a state data set of a target device, and the state data set comprises state data of multiple dimensions of the target device;
the construction module is used for analyzing and processing the state data of each dimensionality in the state data set and constructing a feature vector;
the determining module is used for inputting the feature vectors into a pre-trained fault prediction model to obtain the fault type of the target equipment output by the fault prediction model;
the building module is specifically configured to determine a correlation coefficient between state data of each dimension in the state data set, where the correlation coefficient between state data of each dimension is used to indicate a correlation coefficient between state data of each dimension and state data of other dimensions; constructing the feature vector according to the correlation coefficient;
the building module is further specifically configured to encode the state data of each dimension to obtain an encoding result of the state data of each dimension; and multiplying the correlation coefficient between the state data of each dimension and the state data of other dimensions by the encoding result of the state data of each dimension to construct the feature vector of the state data of each dimension.
8. An equipment failure prediction platform, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the equipment failure prediction platform is running, the processor executing the machine-readable instructions to perform the steps of the equipment failure prediction method according to any one of claims 1-6.
9. A storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the device failure prediction method according to any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110738714.2A CN113359682B (en) | 2021-06-30 | 2021-06-30 | Equipment fault prediction method, device, equipment fault prediction platform and medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110738714.2A CN113359682B (en) | 2021-06-30 | 2021-06-30 | Equipment fault prediction method, device, equipment fault prediction platform and medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113359682A CN113359682A (en) | 2021-09-07 |
CN113359682B true CN113359682B (en) | 2022-12-02 |
Family
ID=77537490
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110738714.2A Active CN113359682B (en) | 2021-06-30 | 2021-06-30 | Equipment fault prediction method, device, equipment fault prediction platform and medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113359682B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114492869A (en) * | 2022-01-25 | 2022-05-13 | 王启明 | Power distribution system health diagnosis method based on Internet of things technology |
CN115338867A (en) * | 2022-08-31 | 2022-11-15 | 上海木蚁机器人科技有限公司 | Fault state monitoring method, device and equipment for mobile robot |
CN115805810A (en) * | 2022-10-31 | 2023-03-17 | 宁德时代新能源科技股份有限公司 | Battery failure prediction method, apparatus, device, storage medium, and program product |
CN116307259B (en) * | 2023-05-10 | 2023-10-31 | 天宇正清科技有限公司 | Maintainability prediction method, device, equipment and medium for mining equipment |
CN117454314B (en) * | 2023-12-19 | 2024-03-05 | 深圳航天科创泛在电气有限公司 | Wind turbine component running state prediction method, device, equipment and storage medium |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106324406A (en) * | 2016-09-21 | 2017-01-11 | 许继集团有限公司 | Transformer direct-current magnetic bias fault diagnosis method and device |
CN109612730A (en) * | 2018-12-29 | 2019-04-12 | 北京航天测控技术有限公司 | A kind of rolling bearing fault localization method based on prewhitening analysis |
CN110469461A (en) * | 2019-08-14 | 2019-11-19 | 树根互联技术有限公司 | A kind of fracture predictor method of blower tooth band, its device and readable storage medium storing program for executing |
CN110929934A (en) * | 2019-11-22 | 2020-03-27 | 深圳市通用互联科技有限责任公司 | Equipment failure prediction method and device, computer equipment and storage medium |
CN111060838A (en) * | 2019-12-31 | 2020-04-24 | 中国人民解放军陆军军医大学第二附属医院 | Medical electronic equipment switching power supply fault diagnosis method based on multi-dimensional feature fusion |
CN111105063A (en) * | 2018-10-26 | 2020-05-05 | 北京国双科技有限公司 | Fault prediction method, fault prediction device, model construction method, fault prediction device, processor and readable storage medium |
CN111914875A (en) * | 2020-06-05 | 2020-11-10 | 华南理工大学 | Fault early warning method of rotating machinery based on Bayesian LSTM model |
CN112785016A (en) * | 2021-02-20 | 2021-05-11 | 南京领行科技股份有限公司 | New energy automobile maintenance and fault monitoring and diagnosis method based on machine learning |
CN112884199A (en) * | 2021-01-15 | 2021-06-01 | 华自科技股份有限公司 | Method and device for predicting faults of hydropower station equipment, computer equipment and storage medium |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7035770B2 (en) * | 2003-12-18 | 2006-04-25 | Taiwan Semiconductor Manufacturing Company, Ltd. | Fuzzy reasoning model for semiconductor process fault detection using wafer acceptance test data |
CN103942568B (en) * | 2014-04-22 | 2017-04-05 | 浙江大学 | A kind of sorting technique based on unsupervised feature selection |
CN109034521B (en) * | 2018-06-07 | 2021-11-16 | 国电南瑞科技股份有限公司 | Intelligent operation and maintenance architecture design method of power grid dispatching control system |
CN111883228B (en) * | 2020-07-28 | 2023-07-07 | 平安科技(深圳)有限公司 | Knowledge graph-based health information recommendation method, device, equipment and medium |
CN111949886B (en) * | 2020-08-28 | 2023-11-24 | 腾讯科技(深圳)有限公司 | Sample data generation method and related device for information recommendation |
-
2021
- 2021-06-30 CN CN202110738714.2A patent/CN113359682B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106324406A (en) * | 2016-09-21 | 2017-01-11 | 许继集团有限公司 | Transformer direct-current magnetic bias fault diagnosis method and device |
CN111105063A (en) * | 2018-10-26 | 2020-05-05 | 北京国双科技有限公司 | Fault prediction method, fault prediction device, model construction method, fault prediction device, processor and readable storage medium |
CN109612730A (en) * | 2018-12-29 | 2019-04-12 | 北京航天测控技术有限公司 | A kind of rolling bearing fault localization method based on prewhitening analysis |
CN110469461A (en) * | 2019-08-14 | 2019-11-19 | 树根互联技术有限公司 | A kind of fracture predictor method of blower tooth band, its device and readable storage medium storing program for executing |
CN110929934A (en) * | 2019-11-22 | 2020-03-27 | 深圳市通用互联科技有限责任公司 | Equipment failure prediction method and device, computer equipment and storage medium |
CN111060838A (en) * | 2019-12-31 | 2020-04-24 | 中国人民解放军陆军军医大学第二附属医院 | Medical electronic equipment switching power supply fault diagnosis method based on multi-dimensional feature fusion |
CN111914875A (en) * | 2020-06-05 | 2020-11-10 | 华南理工大学 | Fault early warning method of rotating machinery based on Bayesian LSTM model |
CN112884199A (en) * | 2021-01-15 | 2021-06-01 | 华自科技股份有限公司 | Method and device for predicting faults of hydropower station equipment, computer equipment and storage medium |
CN112785016A (en) * | 2021-02-20 | 2021-05-11 | 南京领行科技股份有限公司 | New energy automobile maintenance and fault monitoring and diagnosis method based on machine learning |
Non-Patent Citations (1)
Title |
---|
唐明珠等.基于梯度提升决策树的风电机组齿轮箱故障检测.《湖南电力》.2019,(第06期), * |
Also Published As
Publication number | Publication date |
---|---|
CN113359682A (en) | 2021-09-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113359682B (en) | Equipment fault prediction method, device, equipment fault prediction platform and medium | |
EP3902992B1 (en) | Scalable system and engine for forecasting wind turbine failure | |
EP2905665B1 (en) | Information processing apparatus, diagnosis method, and program | |
EP3413154B1 (en) | Equipment diagnostic device, equipment diagnostic method, and equipment diagnostic program | |
EP2759938B1 (en) | Operations management device, operations management method, and program | |
US10852357B2 (en) | System and method for UPS battery monitoring and data analysis | |
EP3183622B1 (en) | Population-based learning with deep belief networks | |
RU2649542C1 (en) | Method and system of remote monitoring of objects | |
CN112083244B (en) | Integrated intelligent diagnosis system for faults of avionic equipment | |
US20210232104A1 (en) | Method and system for identifying and forecasting the development of faults in equipment | |
JP2000259223A (en) | Plant monitoring device | |
CN110852509A (en) | Fault prediction method and device of IGBT module and storage medium | |
CN112396250B (en) | Diesel engine fault prediction method, device, equipment and storage medium | |
Qiu et al. | Remaining useful life estimation for rolling bearing with SIOS-based indicator and particle filtering | |
CN111664083A (en) | Nuclear power main pump fault diagnosis method based on Bayesian network | |
da Silva Arantes et al. | A novel unsupervised method for anomaly detection in time series based on statistical features for industrial predictive maintenance | |
WO2023072724A1 (en) | System, apparatus and method for monitoring condition of an asset in technical installation | |
CN113948159B (en) | Fault detection method, device and equipment for transformer | |
CN115600695B (en) | Fault diagnosis method for metering equipment | |
JP7062505B2 (en) | Equipment management support system | |
US11339763B2 (en) | Method for windmill farm monitoring | |
de León Hijes et al. | Assessment of functional condition of equipment in industrial plants based on multiple measurements | |
Vicêncio et al. | An intelligent predictive maintenance approach based on end-of-line test logfiles in the automotive industry | |
Randrianandraina et al. | Wind turbine generator bearing anomaly detection and explanation using rrcf approach | |
Suzuki et al. | An anomaly detection system for advanced maintenance services |
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 |