CN113326889A - Method and apparatus for training a model - Google Patents
Method and apparatus for training a model Download PDFInfo
- Publication number
- CN113326889A CN113326889A CN202110670871.4A CN202110670871A CN113326889A CN 113326889 A CN113326889 A CN 113326889A CN 202110670871 A CN202110670871 A CN 202110670871A CN 113326889 A CN113326889 A CN 113326889A
- Authority
- CN
- China
- Prior art keywords
- sample data
- data
- train
- target
- target sample
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The application discloses a method and a device for training a model, and relates to the technical field of machine learning, intelligent detection and industrial big data. The method comprises the following steps: acquiring a target sample data set and a label of target sample data in the target sample data set, wherein the target sample data comprises continuous sample data; extracting data characteristics of target sample data, including: performing statistical calculation on continuous sample data in a first preset time period, and determining a statistical calculation result as the data characteristics of the continuous sample data; and training an initial fault detection model by adopting the data characteristics of the target sample data and the label of the target sample data, and obtaining a target fault detection model. The efficiency of detecting equipment faults can be improved by adopting the target fault detection model.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to the field of machine learning, intelligent detection, and industrial big data technologies, and more particularly, to a method and apparatus for training a model.
Background
The rail transit is an indispensable traffic mode in people's production and life, and the trouble of detecting rail train can avoid the train to take place the accident when moving. Train components are regularly inspected, usually manually, to troubleshoot detected train faults.
However, the conventional method for detecting the train fault has the problem of low detection efficiency.
Disclosure of Invention
The present disclosure provides a method, an apparatus, an electronic device, and a computer-readable storage medium for training a model.
According to a first aspect, there is provided a method for training a model, the method comprising: acquiring a target sample data set and a label of target sample data in the target sample data set, wherein the target sample data comprises continuous sample data; extracting data characteristics of target sample data, including: performing statistical calculation on continuous sample data in a first preset time period, and determining a statistical calculation result as the data characteristics of the continuous sample data; and training an initial fault detection model by adopting the data characteristics of the target sample data and the label of the target sample data, and obtaining a target fault detection model.
According to a second aspect, there is provided a method for detecting a fault, the method comprising: acquiring real-time data of a detection object in the operation process; determining whether a fault exists in the running state of a detected object by adopting the data characteristics of real-time data and a target fault detection model, wherein the target fault detection model is obtained by adopting the data characteristics of target sample data and the label training of the target sample data, the target sample data comprises continuous sample data, and the data characteristics of the continuous sample data comprise the result of statistical calculation of the continuous sample data in a first preset time period; and sending out alarm information in response to determining that the operation state of the detection object has a fault.
According to a third aspect, there is provided an apparatus for training a model, the apparatus comprising: the device comprises an acquisition unit, a storage unit and a processing unit, wherein the acquisition unit is configured to acquire a target sample data set and a label of target sample data in the target sample data set, and the target sample data comprises continuous sample data; an extraction unit configured to extract data features of target sample data, including: performing statistical calculation on continuous sample data in a first preset time period, and determining a statistical calculation result as the data characteristics of the continuous sample data; and the training unit is configured to train the initial fault detection model by adopting the data characteristics of the target sample data and the label of the target sample data, and obtain the target fault detection model.
According to a fourth aspect, there is provided an apparatus for detecting a fault, the apparatus comprising: the second acquisition unit is configured to acquire real-time data in the running process of the detection object; the prediction unit is configured to determine whether a fault exists in the operation state of the detection object by adopting the data characteristics of real-time data and a target fault detection model, wherein the target fault detection model is obtained by adopting the data characteristics of target sample data and the label training of the target sample data, the target sample data comprises continuous sample data, and the data characteristics of the continuous sample data comprise a result of statistical calculation on the continuous sample data in a first preset time period; and the early warning unit is configured to send out warning information in response to the fact that the operation state of the detection object has a fault.
According to a fifth aspect, embodiments of the present disclosure provide an electronic device, comprising: one or more processors: a storage device for storing one or more programs which, when executed by one or more processors, cause the one or more processors to implement the method for training a model as provided in the first aspect, or the method for training a model as provided in the second aspect.
According to a sixth aspect, embodiments of the present disclosure provide a computer readable storage medium having stored thereon a computer program, wherein the program, when executed by a processor, implements the method for training a model as provided by the first aspect, or implements the method for training a model as provided by the second aspect.
The method and the device for training the model, provided by the disclosure, are used for acquiring a target sample data set and a label of target sample data in the target sample data set, wherein the target sample data comprises continuous sample data; extracting data characteristics of target sample data, including: performing statistical calculation on continuous sample data in a first preset time period, and determining a statistical calculation result as the data characteristics of the continuous sample data; the data characteristics of the target sample data and the label of the target sample data are adopted to train the initial fault detection model, the target fault detection model is obtained, and the obtained target fault detection model is adopted to detect equipment faults, so that the fault detection efficiency and accuracy can be improved.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The drawings are included to provide a better understanding of the present solution and are not intended to limit the present application. Wherein:
FIG. 1 is an exemplary system architecture diagram in which embodiments of the present application may be applied;
FIG. 2 is a flow diagram of one embodiment of a method for training a model according to the present application;
FIG. 3 is a flow diagram of another embodiment of a method for training a model according to the present application;
FIG. 4 is a flow diagram of one embodiment of a method for detecting faults in accordance with the present application;
FIG. 5 is a flow chart of one example of a method for detecting a fault according to the present application;
FIG. 6 is a schematic block diagram of one embodiment of an apparatus for training models according to the present application;
FIG. 7 is a schematic block diagram of one embodiment of an apparatus for detecting faults in accordance with the present application;
FIG. 8 is a block diagram of an electronic device for implementing a method for training a model according to an embodiment of the present application.
Detailed Description
The following description of the exemplary embodiments of the present application, taken in conjunction with the accompanying drawings, includes various details of the embodiments of the application for the understanding of the same, which are to be considered exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
Fig. 1 illustrates an exemplary system architecture 100 to which embodiments of the present method for authenticating a system or apparatus for authenticating a system may be applied.
As shown in fig. 1, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. Various emulation class processes or processes for testing the system can be installed on the terminal devices 101, 102, 103. The terminal devices 101, 102, 103 may also have various client applications installed thereon, such as an information input application, a video application, a play application, an audio application, a search application, a shopping application, a financial application, and the like.
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting receiving server messages, including but not limited to smartphones, tablets, e-book readers, electronic players, laptop portable computers, desktop computers, and the like.
The terminal apparatuses 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various hardware modules to be verified or electronic devices, and when the terminal devices 101, 102, 103 are software, they may be installed in the electronic devices listed above. It may be implemented as multiple pieces of software or software modules (e.g., multiple software modules to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.
The server 105 may obtain, through the terminal devices 101, 102, and 103, a target sample data set and a tag of target sample data in the target sample data set, where the target sample data includes continuous sample data, and extract data features of the target sample data, including: and performing statistical calculation on the continuous sample data, determining the result of the statistical calculation as the data characteristics of the continuous sample data, training an initial fault detection model by adopting the data characteristics of the target sample data and the label of the target sample data, and obtaining a target fault detection model.
It should be noted that the method for authenticating the system provided by the embodiment of the present disclosure is generally performed by the server 105, and accordingly, the apparatus for authenticating the system is generally disposed in the server 105.
It should be understood that the number of devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of devices, networks, and servers, as desired for an implementation.
With continued reference to FIG. 2, a flow 200 of one embodiment of a method for training a model according to the present disclosure is shown. A method for training a model, comprising the steps of:
In this embodiment, an executing subject (e.g., a server shown in fig. 1) of the method for training a model may acquire a target sample data set and a tag of the target sample data in the target sample data set from a terminal device or a cloud storage in a wired or wireless manner. The target sample data in the target sample data set may be ambient environment data or operation parameter data of the device or any vehicle during normal operation/working, and correspondingly, the tag of the target sample data may be that the device or the vehicle has no fault; the target sample data can be ambient environment data or operation parameter data of the device or any vehicle when the device or the vehicle fails, and accordingly, the label of the target sample data can be that the device or the vehicle has a fault.
The target sample data includes continuously changing sample data, such as power consumption data of the equipment, environmental temperature data of the environment where the equipment is located, traveling speed data of the vehicle, main air pipe pressure data of the train, and the like.
In this embodiment, the data feature of the target sample data may be extracted, specifically, for the continuous sample data, the statistical data such as a maximum value, a minimum value, a mean value, or a median of the continuous sample data in the first preset time period may be calculated, and a result of the statistical calculation may be determined as the data feature of the continuous sample data. For example, the maximum value of the train main air pipe pressure in a preset time period is used as the data characteristic of the target sample data (specifically, the train main air pipe pressure). The first preset period may be an arbitrary period.
And 203, training an initial fault detection model by using the data characteristics of the target sample data and the label of the target sample data, and obtaining a target fault detection model.
In this embodiment, the data characteristics of the target sample data and the label of the target sample data are used to train the initial fault detection model, and the trained target fault detection model is obtained, so that the target fault detection model includes the mapping relationship between the data characteristics and the data label, that is, the target fault detection model can obtain the data label based on the data characteristics. The initial fault detection model may be a model stored locally by the server, or may be any initial model acquired by the server based on the terminal device or the cloud storage space.
According to the method for training the model, a target sample data set and a label of target sample data in the target sample data set are obtained, wherein the target sample data comprises continuous sample data; extracting data characteristics of target sample data, including: carrying out statistical calculation on the continuous sample data, and determining the result of the statistical calculation as the data characteristics of the continuous sample data; the data characteristics of the target sample data and the label of the target sample data are adopted to train the initial fault detection model, the target fault detection model is obtained, and the obtained target fault detection model is adopted to detect equipment faults, so that the fault detection efficiency and accuracy can be improved.
Optionally, the target sample data further comprises discrete sample data; extracting data characteristics of target sample data, including: and counting the frequency of the occurrence of the preset characteristics of the discrete sample data in the second preset time period, and determining the frequency counting result of the preset characteristics as the data characteristics of the discrete sample data.
In this embodiment, the target sample data further includes discrete sample data, for example, the discrete sample data may be discretized sample data such as operating status data of the device in a certain period and/or at a certain time point, environment data of an environment where the device is located, geographic location data where the vehicle is located, period data (e.g., a day period or a night period) when the device/vehicle operates, and the like.
The data characteristic extraction of the target sample data comprises the following steps: and counting the frequency of the discrete sample data with the preset characteristics in a second preset time period, and determining the frequency counting result of the preset characteristics as the data characteristics of the discrete sample data. The preset feature is a feature associated with the discrete sample data, for example, if the discrete sample data is operating state data of the device within a certain period of time, the preset feature may be stable in operation or abnormal in operating state; if the discrete sample data is the environmental data of the environment where the equipment is located, the preset characteristic can be that the equipment is in an extreme working environment or the equipment is in a normal temperature working environment; if the discrete sample data is the geographic position data of the vehicle, the preset characteristics can be that the vehicle is positioned on an overhead bridge, in a tunnel, in a mountain region and the like; if the discrete sample data is time period data when the vehicle runs, the preset characteristic may be that the vehicle running time is day, night or a traffic peak time period, and the like.
Specifically, the frequency of occurrence of the preset characteristic of the discrete sample data in the preset time period before the device or the vehicle has a fault may be determined as the data characteristic of the discrete sample data. For example, if the discrete sample data is the output voltage state data of the device, the number of times of fluctuation of the output voltage of the device in the preset time period may be determined as the data characteristic of the discrete sample data; if the discrete sample data is the geographic position where the vehicle is located, determining the number of times that the vehicle is located on the viaduct (or the number of times that the vehicle is located in the tunnel and the number of times that the vehicle is located in the mountain land section) in the preset time period as the data characteristic of the discrete sample data; if the time data of the vehicle running when the sample data is dispersed, the number of times that the vehicle runs in the night time period in the preset time period can be determined as the data characteristic of the discrete sample data.
In this embodiment, the target sample data further includes discrete sample data, and when the data features of the discrete sample data are extracted, the frequency statistical result of the preset features appearing in the discrete sample data is determined as the data features of the discrete sample data, so that the efficiency and accuracy of determining the data features of the discrete sample data can be improved.
With further reference to FIG. 3, a flow 300 of another embodiment of a method for training a model is shown. The process 300 of the method for training a model includes the steps of:
In this embodiment, an executing subject (for example, a server shown in fig. 1) of the method for training the model may acquire a sample data set from a terminal device or a cloud storage in a wired or wireless manner, where the sample data set includes positive sample data and negative sample data, the positive sample data used for training the model refers to sample data of a tag representing "fault exists", and the negative sample data used for training the model refers to sample data of a tag representing "no fault" or no tag.
It can be understood that, in the working/operating process of the device or the vehicle, the event that a fault occurs is a small probability event compared with the event that the device or the vehicle is in a normal working state, so that in the acquired sample data set, the data volume of positive sample data is far smaller than that of negative sample data, the positive sample data in the sample data set belongs to sample data of a rare class, and the negative sample data in the sample data set belongs to sample data of a rich class.
In this embodiment, the sample data set may be subjected to sample equalization processing, so as to increase the data size of the rare positive sample data or decrease the data size of the rich negative sample data, so that the data size of the positive sample data and the data size of the negative sample data in the sample data set after the sample equalization processing are in a uniform state, and the sample data set after the sample equalization processing is determined as the target sample data set. Specifically, the positive sample data may be subjected to an oversampling process, or the negative sample data may be subjected to an undersampling process to increase the positive sample data, or decrease the negative sample data.
in this embodiment, the tag of the target sample data in the target sample data set may be acquired from a terminal device, a cloud storage, or the sample data set. The target sample data includes continuous sample data, for example, continuously changing sample data such as power consumption data of equipment, environment temperature data of equipment, driving speed data of a vehicle, and main air duct pressure data of a train.
And 305, training an initial fault detection model by using the data characteristics of the target sample data and the label of the target sample data, and obtaining a target fault detection model.
In this embodiment, the descriptions of step 304 and step 305 are the same as the descriptions of step 202 and step 203, and are not repeated here.
Compared with the embodiment described in fig. 2, the method for training the model provided by this embodiment adds the steps of obtaining the sample data set and performing sample equalization on the positive sample data and the negative sample data in the sample data set, so as to train the initial fault detection model by using the sample data set after sample equalization, thereby improving the detection accuracy of the trained target fault detection model and enabling the trained target fault detection model to have better fitting capability and generalization capability.
Optionally, performing sample equalization processing on the sample data set, and determining the sample data set after the sample equalization processing as a target sample data set, including: carrying out down-sampling processing on the negative sample data; and determining a set constructed by the positive sample data and the negative sample data after the down-sampling processing as a target sample data set.
In this embodiment, downsampling (or sampling) processing may be performed on the negative sample data of the rich category, and a set constructed by all positive sample data in the sample data set and a part of negative sample data extracted from all negative sample data in the sample data set may be determined as the target sample data set.
In the embodiment, the data volume of the negative sample data in the sample data set and in the rich categories is reduced by downsampling all the negative sample data, so that the efficiency of sample equalization on the sample data set can be improved.
Optionally, performing sample equalization processing on the sample data set, and determining the sample data set after the sample equalization processing as a target sample data set, including: synthesizing new positive sample data based on the positive sample data; and determining a set constructed by the positive sample data, the newly synthesized positive sample data and the negative sample data as a target sample data set.
In this embodiment, data synthesis processing may be performed on positive sample data of a rare category, and a set constructed by all positive sample data in the sample data set, newly synthesized positive sample data, and all negative sample data in the sample data set may be determined as a target sample data set. For example, a manual minority over-Sampling (SMOTE) method may be used to generate new positive sample data.
In this embodiment, the data synthesis of all positive sample data is performed to increase the data size of the positive sample data in the sample data set and in the scarce category, so that the efficiency of performing sample equalization on the sample data set can be improved.
In some optional implementations of the embodiments described above in connection with fig. 2 and 3, the target sample data comprises at least one of: the pressure of a train air pipe, the pressure of a train air spring, the running speed of a train, the passenger capacity of the train, the temperature of fresh air in a main air pipe of the train, the running state of a train air compressor, the running time of the train and the position of the train; the tag of the target sample data includes: the air pipe of the train has faults and is not faulted; and the target fault detection model is used for detecting the faults of the train air pipes.
In this embodiment, the target sample data may include at least one of the following: train air pipe pressure, train air spring pressure, train running speed, train passenger capacity, temperature of fresh air in a train main air pipe, running state of a train air compressor, train running time and train position. Wherein, the pressure of the train air duct, the pressure of the train air spring, the running speed of the train, the passenger capacity of the train and the temperature of the fresh air in the main air duct of the train belong to continuous sample data; the operation state of the air compressor of the train (the preset characteristic can be a normal working state or an abnormal working state), the running time of the train (the preset characteristic can be a daytime period or a night period), the position of the train (the preset characteristic can be on an overhead bridge, on a mountain road section or on a ramp road section) belong to discrete sample data.
The tag of the target sample data may include: the air pipe of the train has a fault or is not faulted. For example, the target sample data acquired at a certain time includes: when the train runs at 350 km/h, the train is positioned on an overhead bridge, the air spring pressure of the train is 3bar (pressure intensity unit), and when the train generates the set of sample data or within a preset time period of generating the set of sample data, the air pipe of the train breaks down, the set of sample data can be marked as a 'fault-existing' label.
The model for completing the target fault detection based on the target sample data training can be a model for detecting the air pipe fault of the train.
In some optional implementation manners of the embodiments described above with reference to fig. 2 and fig. 3, the obtained target sample data may be unlabeled data, and the classification model may be trained by using the unlabeled target sample data and based on an unsupervised model training method, so that the classification model may classify the unlabeled target sample data, and label the classified unlabeled target sample data based on manual experience, so as to expand the sample data set used for training the initial model.
With continued reference to FIG. 4, a flow 400 of one embodiment of a method for training a model according to the present disclosure is shown. Method for detecting a fault, comprising the steps of:
In this embodiment, an executing body (for example, a server shown in fig. 1) of the method for detecting the fault may obtain, in a wired or wireless manner, real-time data of the detection object in the operation process, for example, data such as voltage and current output by the device in the operation process, environmental parameters such as temperature and humidity of an environment in which the device is currently located, a real-time running speed of the train in the operation process, or a current running position of the train, a number of passengers carried by the train in the current running process, and the like from the terminal device, the service server, or the cloud storage.
In this embodiment, the data characteristics of the real-time data and the target fault detection model may be used to determine whether a fault exists in the operating state of the detection object. Specifically, the data characteristics of the real-time data may be input into the target fault detection model, and the label output by the target fault detection model may be used as the prediction result of the operation state of the detection object.
The training method of the target fault detection model comprises the following steps: acquiring a target sample data set and a label of target sample data in the target sample data set, wherein the target sample data comprises continuous sample data; extracting data characteristics of target sample data, including: carrying out statistical calculation on the continuous sample data, and determining the result of the statistical calculation as the data characteristics of the continuous sample data; and training an initial fault detection model by adopting the data characteristics of the target sample data and the label of the target sample data, and obtaining a target fault detection model. The target fault detection model may also be trained using the method in the embodiment as described in fig. 2 or fig. 3.
In response to determining that the operating state of the detection object has a fault, an alarm message is issued, step 403.
In this embodiment, if the operation state of the detection target predicted by the target failure detection model is that there is a failure, alarm information is issued.
The method for detecting the fault provided by the embodiment obtains the real-time data of the detected object in the operation process, determines whether the operation state of the detected object has the fault by adopting the data characteristics of the real-time data and the target fault detection model, and sends out the alarm information after determining that the operation state of the detected object has the fault, so that whether the detected object has the fault can be automatically detected, the detection efficiency is improved, and the manual operation and maintenance cost is reduced. In addition, a target fault detection model can be utilized, whether the detected object is about to have a fault or not is pre-warned based on real-time data in the current operation process of the detected object, accidents can be avoided, and the operation efficiency of the system can be improved.
Optionally, the real-time data comprises continuous data; the method for detecting a fault includes: and carrying out statistical calculation on the continuous data, carrying out statistical calculation on the continuous data in a third preset time period, and determining the result of the statistical calculation as the data characteristic of the continuous data.
In this embodiment, statistical data such as a maximum value, a minimum value, a mean value, or a median of the continuous data in the third preset time period may be calculated, and a result of the statistical calculation may be determined as a data feature of the continuous data. For example, the maximum value of the main air pipe pressure of the train in a preset time period is used as the data characteristic of continuous data. The third preset period may be an arbitrary period.
In this embodiment, the real-time data includes continuous data, and the statistical calculation result of the continuous data is determined as the data characteristic of the continuous data, so that the efficiency and accuracy of determining the data characteristic of the continuous data can be improved.
Optionally, the target sample data comprises discrete sample data, and the data characteristics of the discrete sample data comprise: counting the frequency of the occurrence of the preset characteristics of the discrete sample data in the second preset time period; the real-time data comprises discrete data; the method for detecting a fault includes: and determining the statistical result of the frequency of the occurrence of the preset features of the discrete data in the fourth preset time period as the data features of the discrete data.
In this embodiment, the target sample data includes discrete sample data, and the data characteristics of the discrete sample data include a statistical result of a frequency of occurrence of a preset characteristic of the discrete sample data in a fourth preset time period, which is determined as the data characteristics of the discrete sample data; the real-time data comprises discrete data, and the method for detecting the fault comprises the following steps: and counting the frequency of the discrete data with the preset characteristics, and determining the frequency counting result of the preset characteristics as the data characteristics of the discrete data.
Specifically, the number of times that the discrete data has the preset feature within the preset time period of the detection object may be determined as the data feature of the discrete data. For example, if the discrete data is output voltage state data of the device, the number of times of fluctuation of the output voltage of the device in the preset time period may be determined as the data characteristic of the discrete data; if the discrete data is the geographic position where the vehicle is located, determining the number of times that the vehicle is located on the viaduct (or the number of times that the vehicle is located in the tunnel and the number of times that the vehicle is located in the mountain land section) in the preset time period as the data characteristic of the discrete data; if the discrete data is time data when the vehicle runs, the number of times that the vehicle runs in the night time period in the preset time period can be determined as the data characteristic of the discrete data.
In this embodiment, the real-time data includes discrete data, and the frequency statistical result of the occurrence of the preset feature in the discrete data is determined as the data feature of the discrete data, so that the efficiency and accuracy of determining the data feature of the discrete data can be improved.
Optionally, the detection object comprises a train, and the real-time data comprises at least one of: the method comprises the following steps of (1) air pipe pressure of a train, air spring pressure of the train, current running speed of the train, current passenger capacity of the train, temperature of fresh air in a main air pipe of the train, running state of an air compressor of the train, current time and current position of the train; detecting the operating state of the object includes: the air pipe state of the train.
In this embodiment, the detection object includes a train, and the real-time data of the train includes at least one of the following: the air pipe pressure of the train, the air spring pressure of the train, the current running speed of the train, the current passenger capacity of the train, the temperature of fresh air in a main air pipe of the train, the running state of an air compressor of the train, the current time and the current position of the train. Wherein, the air pipe pressure of the train, the air spring pressure of the train, the current running speed of the train, the current passenger capacity of the train and the temperature of fresh air in a main air pipe of the train belong to continuous data; the running state of an air compressor of the train, the current time and the current position of the train belong to discrete data. Determining whether the operating state of the detection object has a fault based on the target fault detection model comprises: whether the air pipe of the train has a fault or not.
For example, as shown in fig. 5, the target fault detection model may be deployed in an operation and maintenance system of a train (detection model deployment), and the server collects data generated during train operation in real time by using various sensors installed on the train, including: the air pipe pressure of the train, the air spring pressure of the train, the current running speed of the train, the current passenger capacity of the train, the temperature of fresh air in a main air pipe of the train, the running state of an air compressor of the train, the current time and the current position of the train, and inputting the data characteristics of the real-time data acquired at the current time point or the real-time data acquired in a preset time period before the current time point into a target fault detection model, so that the target fault detection model predicts whether a fault exists in a main air duct of the train based on the data characteristics (fault real-time prediction), and if the output/prediction result of the target fault detection model is 'fault existence' (based on sample data and a label adopted in the process of training the model, the output of the target fault detection model can also be a more specific fault label, such as 'air leakage exists'), alarming information is sent (fault alarm).
The first preset time period, the second preset time period, the third preset time period and the fourth preset time period in the disclosure may be the same preset time period or different preset time periods.
With further reference to fig. 6, as an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an apparatus for training a model, which corresponds to the method embodiments shown in fig. 2 and 3, and which can be applied in various electronic devices.
As shown in fig. 6, the apparatus 600 for training a model of the present embodiment includes: acquisition section 601, extraction section 602, and training section 603. The device comprises an acquisition unit, a storage unit and a processing unit, wherein the acquisition unit is configured to acquire a target sample data set and a label of target sample data in the target sample data set, wherein the target sample data comprises continuous sample data; an extraction unit configured to extract data features of target sample data, including: performing statistical calculation on continuous sample data in a first preset time period, and determining a statistical calculation result as the data characteristics of the continuous sample data; and the training unit is configured to train the initial fault detection model by adopting the data characteristics of the target sample data and the label of the target sample data, and obtain the target fault detection model.
In some embodiments, the target sample data further comprises discrete sample data; an extraction unit comprising: and the extraction module is configured to count the frequency of the occurrence of the preset features of the discrete sample data in the second preset time period, and determine the frequency counting result of the preset features as the data features of the discrete sample data.
In some embodiments, the obtaining unit comprises: the device comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is configured to acquire a sample data set, the sample data set comprises positive sample data and negative sample data, and the data volume of the positive sample data is smaller than that of the negative sample data; and the determining module is configured to perform sample equalization processing on the sample data set and determine the sample data set after the sample equalization processing as a target sample data set.
In some embodiments, the determining module comprises: the down-sampling module is configured to perform down-sampling processing on the negative sample data; and the first determining sub-module is configured to determine a set constructed by the positive sample data and the downsampled negative sample data as a target sample data set.
In some embodiments, the determining module comprises: a data synthesis module configured to synthesize new positive sample data based on the positive sample data; and the second determination submodule is configured to determine a set constructed by the positive sample data, the newly synthesized positive sample data and the negative sample data as a target sample data set.
In some embodiments, the target sample data comprises at least one of: the pressure of a train air pipe, the pressure of a train air spring, the running speed of a train, the passenger capacity of the train, the temperature of fresh air in a main air pipe of the train, the running state of a train air compressor, the running time of the train and the position of the train; the tag of the target sample data includes: the air pipe of the train has faults and is not faulted; and the target fault detection model is used for detecting the faults of the train air pipes.
The units in the apparatus 600 described above correspond to the steps in the method described with reference to fig. 2 and 3. Thus, the operations, features and technical effects that can be achieved by the methods for training a model described above are also applicable to the apparatus 600 and the units included therein, and are not described herein again.
With further reference to fig. 7, as an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an apparatus for training a model, which corresponds to the method embodiment shown in fig. 4, and which can be applied in various electronic devices.
As shown in fig. 7, the apparatus 700 for detecting a failure of the present embodiment includes: a second acquisition unit 701, a prediction unit 702, and an early warning unit 703. The second acquisition unit is configured to acquire real-time data in the running process of the detection object; the prediction unit is configured to determine whether a fault exists in the operation state of the detection object by adopting the data characteristics of real-time data and a target fault detection model, wherein the target fault detection model is obtained by adopting the data characteristics of target sample data and the label training of the target sample data, the target sample data comprises continuous sample data, and the data characteristics of the continuous sample data comprise a result of statistical calculation on the continuous sample data in a first preset time period; and the early warning unit is configured to send out warning information in response to the fact that the operation state of the detection object has a fault.
In some embodiments, the real-time data comprises continuous data; the device comprises: and the first feature extraction unit is configured to perform statistical calculation on the continuous data in the third preset time period and determine the result of the statistical calculation as the data feature of the continuous data.
In some embodiments, the target sample data comprises discrete sample data, the data characteristics of the discrete sample data comprising: counting the frequency of the occurrence of the preset characteristics of the discrete sample data in the second preset time period; the real-time data comprises discrete data; the device comprises: and the second feature extraction unit is configured to determine a statistical result of the frequency of occurrence of the preset features of the discrete data in the fourth preset time period as the data features of the discrete data.
In some embodiments, the detection object comprises a train, and the real-time data comprises at least one of: the method comprises the following steps of (1) air pipe pressure of a train, air spring pressure of the train, current running speed of the train, current passenger capacity of the train, temperature of fresh air in a main air pipe of the train, running state of an air compressor of the train, current time and current position of the train; detecting the operating state of the object includes: the air pipe state of the train.
The units in the apparatus 700 described above correspond to the steps in the method described with reference to fig. 4. Thus, the operations, features and technical effects that can be achieved by the methods for training a model described above are also applicable to the apparatus 700 and the units included therein, and are not described herein again.
There is also provided, in accordance with an embodiment of the present application, an electronic device, a readable storage medium, and a computer program product.
FIG. 8 shows a schematic block diagram of an example electronic device 800 that may be used to implement embodiments of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application that are described and/or claimed herein.
As shown in fig. 8, the apparatus 800 includes a computing unit 801 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)802 or a computer program loaded from a storage unit 808 into a Random Access Memory (RAM) 803. In the RAM803, various programs and data required for the operation of the device 800 can also be stored. The calculation unit 801, the ROM 802, and the RAM803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.
A number of components in the device 800 are connected to the I/O interface 805, including: an input unit 806, such as a keyboard, a mouse, or the like; an output unit 807 such as various types of displays, speakers, and the like; a storage unit 808, such as a magnetic disk, optical disk, or the like; and a communication unit 805 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 805 allows the device 800 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for implementing the methods of the present application may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this application, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present application may be executed in parallel, may be executed sequentially, or may be executed in different orders, as long as the desired data of the technical solution disclosed in the present application can be realized, and the present disclosure is not limited thereto.
The above-described embodiments should not be construed as limiting the scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (23)
1. A method for training a model, comprising:
acquiring a target sample data set and a label of target sample data in the target sample data set, wherein the target sample data comprises continuous sample data;
extracting data characteristics of the target sample data, including: performing statistical calculation on the continuous sample data in a first preset time period, and determining the result of the statistical calculation as the data characteristic of the continuous sample data;
and training an initial fault detection model by adopting the data characteristics of the target sample data and the label of the target sample data, and obtaining a target fault detection model.
2. The method of claim 1, wherein the target sample data further comprises discrete sample data;
the extracting the data characteristics of the target sample data comprises the following steps:
and counting the frequency of the occurrence of the preset characteristics of the discrete sample data in a second preset time period, and determining the frequency counting result of the preset characteristics as the data characteristics of the discrete sample data.
3. The method of claim 1, wherein said obtaining a target sample data set comprises:
acquiring a sample data set, wherein the sample data set comprises positive sample data and negative sample data, and the data volume of the positive sample data is smaller than that of the negative sample data;
and carrying out sample equalization processing on the sample data set, and determining the sample data set after the sample equalization processing as the target sample data set.
4. The method according to claim 3, wherein said performing sample equalization processing on the sample data set and determining the sample data set after sample equalization processing as the target sample data set comprises:
carrying out downsampling processing on the negative sample data;
and determining a set constructed by the positive sample data and the downsampled negative sample data as the target sample data set.
5. The method according to claim 3, wherein said performing sample equalization processing on the sample data set and determining the sample data set after sample equalization processing as the target sample data set comprises:
synthesizing new positive sample data based on the positive sample data;
and determining the set constructed by the positive sample data, the newly synthesized positive sample data and the negative sample data as the target sample data set.
6. The method of claim 1, wherein the target sample data comprises at least one of: the pressure of a train air pipe, the pressure of a train air spring, the running speed of a train, the passenger capacity of the train, the temperature of fresh air in a main air pipe of the train, the running state of a train air compressor, the running time of the train and the position of the train; the tag of the target sample data includes: the air pipe of the train has faults and is not faulted; and the target fault detection model is used for detecting the faults of the train air pipes.
7. A method for detecting a fault, comprising:
acquiring real-time data of a detection object in the operation process;
determining whether a fault exists in the running state of the detection object by adopting the data characteristics of the real-time data and a target fault detection model, wherein the target fault detection model is obtained by adopting the data characteristics of target sample data and the label training of the target sample data, the target sample data comprises continuous sample data, and the data characteristics of the continuous sample data comprise the result of statistical calculation on the continuous sample data in a first preset time period;
and responding to the determined operation state of the detection object to generate a fault, and sending out alarm information.
8. The method of claim 7, wherein the real-time data comprises continuous data;
the method comprises the following steps:
and carrying out statistical calculation on the continuous data in a third preset time period, and determining the result of the statistical calculation as the data characteristic of the continuous data.
9. The method of claim 7, wherein the target sample data comprises discrete sample data, data characteristics of which comprise: counting the frequency of the occurrence of preset characteristics of the discrete sample data in a second preset time period; the real-time data comprises discrete data;
the method comprises the following steps:
and determining the statistical result of the frequency of the occurrence of the preset features of the discrete data in a fourth preset time period as the data features of the discrete data.
10. The method of claim 7, wherein the test object comprises a train, and the real-time data comprises at least one of: the air pipe pressure of the train, the air spring pressure of the train, the current running speed of the train, the current passenger capacity of the train, the temperature of fresh air in a main air pipe of the train, the running state of an air compressor of the train, the current time and the current position of the train; the detecting the operating state of the object includes: the air pipe state of the train.
11. An apparatus for training a model, comprising:
the device comprises an acquisition unit, a storage unit and a processing unit, wherein the acquisition unit is configured to acquire a target sample data set and a label of target sample data in the target sample data set, and the target sample data comprises continuous sample data;
an extracting unit configured to extract data features of the target sample data, including: performing statistical calculation on the continuous sample data in a first preset time period, and determining the result of the statistical calculation as the data characteristic of the continuous sample data;
and the training unit is configured to train an initial fault detection model by adopting the data characteristics of the target sample data and the label of the target sample data, and obtain a target fault detection model.
12. The apparatus of claim 11, in which the target sample data further comprises discrete sample data;
the extraction unit includes:
the extraction module is configured to count the frequency of the occurrence of the preset features of the discrete sample data in a second preset time period, and determine the frequency counting result of the preset features as the data features of the discrete sample data.
13. The apparatus of claim 11, wherein the obtaining unit comprises:
the device comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is configured to acquire a sample data set, the sample data set comprises positive sample data and negative sample data, and the data volume of the positive sample data is smaller than that of the negative sample data;
and the determining module is configured to perform sample equalization processing on the sample data set, and determine the sample data set after sample equalization processing as the target sample data set.
14. The apparatus of claim 13, wherein the means for determining comprises:
a down-sampling module configured to perform down-sampling processing on the negative sample data;
a first determining sub-module configured to determine the set constructed by the positive sample data and the downsampled negative sample data as the target sample data set.
15. The apparatus of claim 13, wherein the means for determining comprises:
a data synthesis module configured to synthesize new positive sample data based on the positive sample data;
a second determining submodule configured to determine the set constructed from the positive sample data, the newly synthesized positive sample data, and the negative sample data as the target sample data set.
16. The apparatus of claim 11, wherein the target sample data comprises at least one of: the pressure of a train air pipe, the pressure of a train air spring, the running speed of a train, the passenger capacity of the train, the temperature of fresh air in a main air pipe of the train, the running state of a train air compressor, the running time of the train and the position of the train; the tag of the target sample data includes: the air pipe of the train has faults and is not faulted; and the target fault detection model is used for detecting the faults of the train air pipes.
17. An apparatus for detecting a fault, comprising:
the second acquisition unit is configured to acquire real-time data in the running process of the detection object;
the prediction unit is configured to determine whether a fault exists in the operation state of the detection object by using the data characteristics of the real-time data and a target fault detection model, wherein the target fault detection model is obtained by using the data characteristics of target sample data and the label training of the target sample data, the target sample data comprises continuous sample data, and the data characteristics of the continuous sample data comprise a result of statistical calculation on the continuous sample data in a first preset time period;
and the early warning unit is configured to send out warning information in response to the fact that the operation state of the detection object has a fault.
18. The apparatus of claim 17, wherein the real-time data comprises continuous data;
the device comprises:
a first feature extraction unit configured to perform statistical calculation on the continuous data within a third preset period and determine a result of the statistical calculation as a data feature of the continuous data.
19. The apparatus of claim 17, wherein the target sample data comprises discrete sample data, data characteristics of which comprise: counting the frequency of the occurrence of preset characteristics of the discrete sample data in a second preset time period; the real-time data comprises discrete data;
the device comprises:
and the second feature extraction unit is configured to determine a statistical result of the frequency of occurrence of the preset features of the discrete data in a fourth preset time period as the data features of the discrete data.
20. The apparatus of claim 17, wherein the detection object comprises a train, and the real-time data comprises at least one of: the air pipe pressure of the train, the air spring pressure of the train, the current running speed of the train, the current passenger capacity of the train, the temperature of fresh air in a main air pipe of the train, the running state of an air compressor of the train, the current time and the current position of the train; the detecting the operating state of the object includes: the air pipe state of the train.
21. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6 or claims 7-10.
22. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any of claims 1-6 or claims 7-10.
23. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-6 or claims 7-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110670871.4A CN113326889A (en) | 2021-06-16 | 2021-06-16 | Method and apparatus for training a model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110670871.4A CN113326889A (en) | 2021-06-16 | 2021-06-16 | Method and apparatus for training a model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113326889A true CN113326889A (en) | 2021-08-31 |
Family
ID=77423612
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110670871.4A Pending CN113326889A (en) | 2021-06-16 | 2021-06-16 | Method and apparatus for training a model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113326889A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117873839A (en) * | 2024-03-12 | 2024-04-12 | 苏州元脑智能科技有限公司 | Fault detection method, device, equipment and storage medium of complex computing system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109591793A (en) * | 2019-02-01 | 2019-04-09 | 王家元 | A kind of fault detection system of rail wagon brake system air hose |
| CN109947080A (en) * | 2019-03-21 | 2019-06-28 | 北京明略软件系统有限公司 | A kind of method, apparatus of fault diagnosis, computer storage medium and terminal |
| CN110705592A (en) * | 2019-09-03 | 2020-01-17 | 平安科技（深圳）有限公司 | Classification model training method, device, equipment and computer readable storage medium |
| WO2020173228A1 (en) * | 2019-02-26 | 2020-09-03 | 京东数字科技控股有限公司 | Joint training method and apparatus for machine learning model, device, and storage medium |
| CN112232405A (en) * | 2020-10-13 | 2021-01-15 | 中车青岛四方机车车辆股份有限公司 | Fault prediction, monitoring and diagnosis method of gearbox and corresponding device |
| CN112395684A (en) * | 2020-10-30 | 2021-02-23 | 长春工业大学 | Intelligent fault diagnosis method for high-speed train running part system |
| CN112633407A (en) * | 2020-12-31 | 2021-04-09 | 深圳云天励飞技术股份有限公司 | Method and device for training classification model, electronic equipment and storage medium |
| CN112785420A (en) * | 2021-01-26 | 2021-05-11 | 上海明略人工智能(集团)有限公司 | Credit scoring model training method and device, electronic equipment and storage medium |
-
2021
- 2021-06-16 CN CN202110670871.4A patent/CN113326889A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109591793A (en) * | 2019-02-01 | 2019-04-09 | 王家元 | A kind of fault detection system of rail wagon brake system air hose |
| WO2020173228A1 (en) * | 2019-02-26 | 2020-09-03 | 京东数字科技控股有限公司 | Joint training method and apparatus for machine learning model, device, and storage medium |
| CN109947080A (en) * | 2019-03-21 | 2019-06-28 | 北京明略软件系统有限公司 | A kind of method, apparatus of fault diagnosis, computer storage medium and terminal |
| CN110705592A (en) * | 2019-09-03 | 2020-01-17 | 平安科技（深圳）有限公司 | Classification model training method, device, equipment and computer readable storage medium |
| CN112232405A (en) * | 2020-10-13 | 2021-01-15 | 中车青岛四方机车车辆股份有限公司 | Fault prediction, monitoring and diagnosis method of gearbox and corresponding device |
| CN112395684A (en) * | 2020-10-30 | 2021-02-23 | 长春工业大学 | Intelligent fault diagnosis method for high-speed train running part system |
| CN112633407A (en) * | 2020-12-31 | 2021-04-09 | 深圳云天励飞技术股份有限公司 | Method and device for training classification model, electronic equipment and storage medium |
| CN112785420A (en) * | 2021-01-26 | 2021-05-11 | 上海明略人工智能(集团)有限公司 | Credit scoring model training method and device, electronic equipment and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| 王秋鹏 等: "《车辆运用与管理》", 30 April 2018, 成都：西南交通大学出版社 * |
| 荆学娜: "高速列车制动系统气制动泄漏故障快速检测方法", 《内燃机与配件》 * |
| 边莉 等: "《交叉熵算法在电子工程领域中的应用》", 31 August 2016, 西安：西安电子科技大学出版社 * |
| 陈宗海 主编: "《系统仿真技术及其应用 第17卷》", 30 September 2016, 合肥：中国科学技术大学出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117873839A (en) * | 2024-03-12 | 2024-04-12 | 苏州元脑智能科技有限公司 | Fault detection method, device, equipment and storage medium of complex computing system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112669867B (en) | Debugging method and device of noise elimination algorithm and electronic equipment | |
| CN110347694B (en) | Equipment monitoring method, device and system based on Internet of things | |
| CN113313962A (en) | Signal lamp fault monitoring method and device, electronic equipment and storage medium | |
| CN113326889A (en) | Method and apparatus for training a model | |
| CN115359471A (en) | Image processing and joint detection model training method, device, equipment and storage medium | |
| CN113723607A (en) | Training method, device and equipment of space-time data processing model and storage medium | |
| US20200342769A1 (en) | Systems and methods for predicting flight safety events using a learning-based model | |
| CN114697247B (en) | Fault detection method, device, equipment and storage medium of streaming media system | |
| CN116155541A (en) | Automatic machine learning platform and method for network security application | |
| CN115641360A (en) | Battery detection method and device based on artificial intelligence and electronic equipment | |
| CN114677848A (en) | Perception early warning system, method, device and computer program product | |
| CN114298772A (en) | Information display method, device, equipment and storage medium | |
| CN114419551A (en) | Channel monitoring method, device, equipment and storage medium | |
| EP3723067A1 (en) | Systems and methods for predicting flight safety events using a learning-based model | |
| CN114399660A (en) | Fault type determination method and device, electronic equipment and storage medium | |
| CN113887414A (en) | Target detection method, target detection device, electronic equipment and storage medium | |
| CN113887101A (en) | Visualization method and device of network model, electronic equipment and storage medium | |
| CN113989720A (en) | Target detection method, training method, device, electronic equipment and storage medium | |
| CN112836626B (en) | Accident determining method and device, model training method and device and electronic equipment | |
| CN114093170B (en) | Generation method, system and device of annunciator control scheme and electronic equipment | |
| CN114445711B (en) | Image detection method, image detection device, electronic equipment and storage medium | |
| CN115345247B (en) | Model training and highway axle load determining method, device, equipment and medium | |
| CN113963322B (en) | Detection model training method and device and electronic equipment | |
| CN113138903B (en) | Method and apparatus for tracking performance of a storage system | |
| CN112700657A (en) | Method and device for generating detection information, road side equipment and cloud control platform |
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 |