CN116522998A - Information processing method, device, equipment and medium - Google Patents
Information processing method, device, equipment and medium Download PDFInfo
- Publication number
- CN116522998A CN116522998A CN202310720465.3A CN202310720465A CN116522998A CN 116522998 A CN116522998 A CN 116522998A CN 202310720465 A CN202310720465 A CN 202310720465A CN 116522998 A CN116522998 A CN 116522998A
- Authority
- CN
- China
- Prior art keywords
- water injection
- injection pump
- samples
- fault diagnosis
- operation information
- 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
- 230000010365 information processing Effects 0.000 title claims abstract description 61
- 238000003672 processing method Methods 0.000 title claims abstract description 27
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 186
- 238000002347 injection Methods 0.000 claims abstract description 184
- 239000007924 injection Substances 0.000 claims abstract description 184
- 238000012549 training Methods 0.000 claims abstract description 181
- 238000003745 diagnosis Methods 0.000 claims abstract description 111
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000005070 sampling Methods 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 71
- 238000000034 method Methods 0.000 claims description 48
- 238000004590 computer program Methods 0.000 claims description 17
- 238000013527 convolutional neural network Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 description 17
- 238000010586 diagram Methods 0.000 description 12
- 238000012360 testing method Methods 0.000 description 11
- 238000004891 communication Methods 0.000 description 6
- 238000011176 pooling Methods 0.000 description 6
- 238000002372 labelling Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000001066 destructive effect Effects 0.000 description 2
- 235000019580 granularity Nutrition 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002332 oil field water Substances 0.000 description 2
- 239000003208 petroleum Substances 0.000 description 2
- 239000010865 sewage Substances 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 210000002364 input neuron Anatomy 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 210000004205 output neuron Anatomy 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B51/00—Testing machines, pumps, or pumping installations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Mechanical Engineering (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Control Of Positive-Displacement Pumps (AREA)
Abstract
The embodiment of the application provides an information processing method, device, equipment and medium, comprising the following steps: acquiring water injection pump operation information of a water injection pump in a first preset time, wherein the water injection pump operation information comprises a plurality of sampling signals; classifying and calculating the operation information of the water injection pump based on the fault diagnosis model to obtain the operation state of the water injection pump corresponding to the operation information of the water injection pump; the fault diagnosis model comprises a plurality of model parameters, the model parameters are obtained after a loss function is trained on the basis of training samples to a preset fault diagnosis model, and the loss function is a function obtained after weighting processing is performed on the basis of the number of samples of the training samples corresponding to the running state of the water injection pump. According to the embodiment of the application, whether the water injection pump fails or not can be accurately determined.
Description
Technical Field
The application belongs to the technical field of information processing, and particularly relates to an information processing method, an information processing device, information processing equipment and an information processing medium.
Background
The water injection pump is the main equipment in oilfield water injection production and has an important influence on petroleum exploitation operation. However, in the operation process of the water injection pump, the water injection pump may fail due to the conditions of high water injection pressure of the water injection well, too high water temperature of the reinjection sewage, poor water quality, strong corrosiveness, large load, severe working condition and the like. In order to avoid unnecessary losses due to failure of the water injection pump, it is necessary to monitor the operating state of the water injection pump in real time in order to determine in time whether the water injection pump has failed.
However, in the prior art, a worker generally determines the operation state of the water injection pump according to related experience so as to further determine whether the water injection pump fails or what kind of failure occurs, or monitors the operation information of the water injection pump based on the failure threshold divided by the expert system so as to determine whether the water injection pump fails. However, the method cannot accurately judge whether the water injection pump fails.
Disclosure of Invention
The embodiment of the application provides an information processing method, an information processing device, information processing equipment and an information processing medium, which can accurately determine whether a water injection pump fails.
In a first aspect, an embodiment of the present application provides an information processing method, including:
acquiring water injection pump operation information of a water injection pump in a first preset time, wherein the water injection pump operation information comprises a plurality of sampling signals;
classifying and calculating the operation information of the water injection pump based on the fault diagnosis model to obtain the operation state of the water injection pump corresponding to the operation information of the water injection pump;
the fault diagnosis model comprises a plurality of model parameters, the model parameters are obtained after a loss function is trained on the basis of training samples to a preset fault diagnosis model, and the loss function is a function obtained after weighting processing is performed on the basis of the number of samples of the training samples corresponding to the running states of the water injection pumps.
In an alternative implementation of the first aspect, the loss function satisfies the following formula:
wherein,,is the probability of model estimation samples, +.>Indicating that the training sample belongs to->The probability distribution of a class is determined,，/>indicating that the training sample belongs to->Probability distribution of class->Is->The actual number of samples of the class training samples, +.>Is->The effective number of samples of the class training sample and is an exponential function of the actual number of samples, +.>Is a superparameter that controls the weights.
In an alternative embodiment of the first aspect, the fault diagnosis model comprises a convolutional neural network comprising at least a first convolutional layer, a pooling layer and a SoftMax layer, the first convolutional layer comprising a plurality of second convolutional layers juxtaposed to each other, the plurality of second convolutional layers being different in size from each other.
In an optional implementation manner of the first aspect, before performing a classification calculation on the water injection pump operation information based on the fault diagnosis model to obtain a water injection pump operation state corresponding to the water injection pump operation information, the method further includes:
acquiring a training sample set, wherein the training sample set comprises a plurality of training samples and a label running state corresponding to each training sample, and the training samples comprise water injection pump running information samples;
For each training sample, the following is performed:
classifying and calculating the training samples based on a preset fault diagnosis model to obtain reference running states corresponding to the training samples;
determining a loss function value of a preset fault diagnosis model according to a reference running state of a target training sample and a label running state of the target training sample, wherein the target training sample is any one of a plurality of training samples;
and training the preset fault diagnosis model by using a training sample based on the loss function value of the preset fault diagnosis model to obtain a trained fault diagnosis model.
In an optional implementation manner of the first aspect, before acquiring the training sample set, the method further includes:
obtaining original water injection pump operation information samples of the water injection pump in a plurality of second preset times respectively, wherein the second preset time is longer than the first preset time;
and dividing the original water injection pump operation information samples according to a sliding time window algorithm aiming at any one of the original water injection pump operation information samples in a plurality of second preset time to obtain a plurality of water injection pump operation information samples corresponding to each original water injection pump operation information sample.
In a second aspect, an embodiment of the present application provides an information processing apparatus, including:
The water injection pump operation information comprises a plurality of sampling signals;
the classification calculation module is used for carrying out classification calculation on the operation information of the water injection pump based on the fault diagnosis model so as to obtain the operation state of the water injection pump corresponding to the operation information of the water injection pump;
the fault diagnosis model comprises a plurality of model parameters, the model parameters are obtained after a loss function is trained on the basis of training samples to a preset fault diagnosis model, and the loss function is a function obtained after weighting processing is performed on the basis of the number of samples of the training samples corresponding to the running states of the water injection pumps.
In an alternative embodiment of the second aspect, the fault diagnosis model comprises a convolutional neural network comprising at least a first convolutional layer, a pooling layer and a SoftMax layer, the first convolutional layer comprising a plurality of second convolutional layers juxtaposed to each other, the plurality of second convolutional layers being different in size from each other.
In a third aspect, there is provided an electronic device comprising: a memory for storing computer program instructions; a processor for reading and executing computer program instructions stored in a memory to perform the information processing method provided in any optional implementation manner of the first aspect.
In a fourth aspect, there is provided a computer storage medium having stored thereon computer program instructions which, when executed by a processor, implement the information processing method provided by any of the alternative embodiments of the first aspect.
In a fifth aspect, a computer program product is provided, instructions in the computer program product, when executed by a processor of an electronic device, cause the electronic device to perform an information processing method implementing any of the alternative embodiments of the first aspect.
In the embodiment of the application, the operation information of the water injection pump in the first preset time can be obtained, and then the obtained operation information of the water injection pump can be classified and calculated based on the fault diagnosis model, so that the operation state of the water injection pump corresponding to the operation information of the water injection pump can be obtained. The fault diagnosis model may include a plurality of model parameters, where the model parameters are obtained by training a preset fault diagnosis model based on training samples by using a loss function, and the loss function is obtained by weighting the number of samples of the training samples corresponding to the operation states of the water injection pumps. In this way, in the process of training the preset fault diagnosis model, the situation that the number of various training samples is unbalanced is considered, so that a more accurate fault diagnosis model can be obtained, and further, whether the water injection pump breaks down and the fault type can be accurately determined based on the running state of the water injection pump output by the fault diagnosis model.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments of the present application will be briefly described, and it is possible for a person skilled in the art to obtain other drawings according to these drawings without inventive effort.
Fig. 1 is a schematic diagram of a training flow of a fault diagnosis model in an information processing method according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a preset fault diagnosis model before improvement according to an embodiment of the present application;
FIG. 3 is a schematic structural diagram of an improved preset fault diagnosis model according to an embodiment of the present application;
fig. 4 is a schematic flow chart of an information processing method according to an embodiment of the present application;
FIG. 5 is a graph showing loss curves of different neural networks according to embodiments of the present application;
fig. 6 is a schematic structural view of an information processing apparatus according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
Features and exemplary embodiments of various aspects of the present application are described in detail below to make the objects, technical solutions and advantages of the present application more apparent, and to further describe the present application in conjunction with the accompanying drawings and the detailed embodiments. It should be understood that the specific embodiments described herein are intended to be illustrative of the application and are not intended to be limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present application by showing examples of the present application.
It is noted that relational terms such as first and second, and the like are 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. Moreover, 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 … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.
The term "and/or" is herein merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone.
The water injection pump is the main equipment in oilfield water injection production and has an important influence on petroleum exploitation operation. However, in the operation process of the water injection pump, the water injection pump may fail due to the conditions of high water injection pressure of the water injection well, too high water temperature of the reinjection sewage, poor water quality, strong corrosiveness, large load, severe working condition and the like. In order to avoid unnecessary losses due to failure of the water injection pump, it is necessary to monitor the operating state of the water injection pump in real time in order to determine in time whether the water injection pump has failed.
However, in the prior art, a worker generally determines the operation state of the water injection pump according to related experience so as to further determine whether the water injection pump fails or what kind of failure occurs, or monitors the operation information of the water injection pump based on the failure threshold divided by the expert system so as to determine whether the water injection pump fails. However, the method cannot accurately judge whether the water injection pump fails.
Based on the problems in real life, with the continuous expansion of applications in various fields and the improvement of enterprise data consciousness in deep learning in recent years, many students propose a method for diagnosing faults of industrial equipment by combining a deep learning network. However, because of the large amount of data support required for training the deep learning model, the ideal balance data set generally used in the prior art, i.e. the number of positive training samples is substantially identical to the number of negative training samples. However, in actual production, after the water injection pump fails, the field engineers can decide to stop the operation, so that the data of the water injection pump in a normal state is far more than the data of the water injection pump in a failure state, and the accuracy of the output result of the model obtained by training is lower, and further whether the water injection pump fails or not cannot be accurately determined based on the output result of the model.
In order to solve the above problems, embodiments of the present application provide an information processing method, an apparatus, a device, and a medium, where the method may obtain operation information of a water injection pump within a first preset time, and further may perform classification calculation on the obtained operation information of the water injection pump based on a fault diagnosis model, so as to obtain an operation state of the water injection pump corresponding to the operation information of the water injection pump. The fault diagnosis model may include a plurality of model parameters, where the model parameters are obtained by training a preset fault diagnosis model based on training samples by using a loss function, and the loss function is obtained by weighting the number of samples of the training samples corresponding to the operation states of the water injection pumps. In this way, in the process of training the preset fault diagnosis model, the situation that the number of various training samples is unbalanced is considered, so that a more accurate fault diagnosis model can be obtained, and further, whether the water injection pump breaks down and the fault type can be accurately determined based on the running state of the water injection pump output by the fault diagnosis model.
The execution subject of the information processing method provided in the embodiment of the present application may be an information processing apparatus, or a control module for executing the information processing method in the information processing apparatus. In the embodiment of the present application, an information processing method provided in the embodiment of the present application will be described by taking an example in which an information processing apparatus executes an information processing method.
In addition, in the information processing method provided in the embodiment of the present application, the operation information of the water injection pump needs to be processed by using a pre-trained fault diagnosis model, so that the fault diagnosis model needs to be trained before the information processing is performed by using the fault diagnosis model. Accordingly, a specific implementation of the training method for a fault diagnosis model provided in the embodiments of the present application is described below with reference to the accompanying drawings.
As shown in fig. 1, the embodiment of the present application provides a training method for a fault diagnosis model, where an execution body of the method is an information processing apparatus, and the method may be specifically implemented by the following steps:
s110, acquiring a training sample set.
The training sample set may include a plurality of training samples and a label running state corresponding to each training sample. Each training sample may include a water injection pump operation information sample of the water injection pump within a second preset time, and each water injection pump operation information sample may include a plurality of sampling signals. The second preset time may be a time preset based on actual experience or situation, for example, the second preset time may be ten minutes or one hour, which is not limited herein.
It should be noted that the above-mentioned related label operation state may include states such as normal operation, bearing bush wear failure, pump head spring taking, pump head loosening, motor eccentricity, destructive wear of a ball outer ring of a double bearing, bearing bracket damage, cross head wear, double bearing bracket damage, wear in a plunger, plunger loosening, western bearing ball outer ring wear, bearing bush wear, pump head spring taking mixing failure, bearing bush wear, anchor bolt loosening mixing failure, bearing bush wear, motor eccentricity mixing failure, bearing bush wear, motor bolt loosening mixing failure, and the like.
In order to obtain a more accurate training sample set and further train the fault diagnosis model better, in a specific embodiment, obtaining the training sample set may specifically include the following steps:
step 1, a plurality of training samples are obtained.
Specifically, by installing the vibration sensor on the water injection pump, the information processing device may obtain the water injection pump operation information samples of the water injection pump respectively in a plurality of second preset times through the vibration sensor, and the sampling frequency of the vibration sensor may be 3200 hz, based on which the information processing device may collect 8192 sampling signals in each second preset time.
In addition, because of the influence of factors such as measurement noise, instrument degradation, and working condition change, the plurality of training samples included in the obtained training sample set need to be preprocessed, wherein the preprocessing may include outlier cleaning processing, invalid data deleting processing, and the like, so that the situation that the accuracy of the model to be trained subsequently is low due to the abnormality or invalidation of the training sample can be avoided.
And 2, labeling the running states of the labels corresponding to the training samples one by one.
Specifically, the label running state of each training sample may be labeled by a manual labeling manner, or the label running state of each training sample may be directly labeled by the information processing device, and the specific labeling manner is not limited herein.
In the labeling process, 80% of labeled sample data can be used as a training sample, 20% of labeled sample data can be used as a test sample, and the distribution ratio of the specific training sample and the test sample is not excessively limited here. It should be noted that, in the process of dividing the training samples and the test samples, the division ratio of the majority class and the minority class is the same, and for example, it is assumed that, in the training samples, the division ratio of the training samples corresponding to the state 1 and the training samples corresponding to the state 2 is 2:1:2, and correspondingly, in the test samples, the division ratio of the test samples corresponding to the state 1 and the test samples corresponding to the state 2 and the test samples corresponding to the state 3 should also be 2:1:2. In addition, the intersection of the training sample and the test sample is an empty set.
It should be noted that, since the fault diagnosis model needs to be subjected to multiple iterative processes to adjust the loss function value thereof until the loss function value satisfies the training stop condition, the trained fault diagnosis model is obtained. However, in each iterative training, if only one training sample is input, too little sample size is unfavorable for training adjustment of the fault diagnosis model. Therefore, the training sample set needs to be divided into a plurality of training samples, and thus, the fault diagnosis model can be iteratively processed by using the training samples in the training sample set.
Therefore, the obtained training samples can be marked to obtain the label running states corresponding to the training samples one by one, and a training sample set containing the training samples and the label running states corresponding to each training sample can be obtained. Thus, the training of the subsequent model is facilitated.
S120, for each training sample, classifying and calculating the training sample based on a preset fault diagnosis model to obtain a reference running state corresponding to the training sample.
Specifically, after the information processing device acquires the plurality of training samples, for each training sample in the plurality of training samples, the information processing device may perform classification calculation on the training sample based on a preset fault diagnosis model, so as to obtain a reference running state corresponding to the training sample.
S130, determining a loss function value of a preset fault diagnosis model according to the reference running state of the target training sample and the label running state of the target training sample.
Wherein the target training sample is any one of a plurality of training samples.
Specifically, the information processing device may further determine a loss function value of the preset fault diagnosis model based on a reference running state corresponding to any one of the plurality of training samples and a label running state corresponding to the training sample, so as to perform iterative training on the preset fault diagnosis model based on the loss function value, and further obtain a more accurate fault diagnosis model.
And S140, training the preset fault diagnosis model by using a training sample based on the loss function value of the preset fault diagnosis model to obtain a trained fault diagnosis model.
Specifically, in order to obtain a better trained fault diagnosis model, under the condition that the loss function value does not meet the training stop condition, model parameters of a preset fault diagnosis model are adjusted, and the fault diagnosis model with the parameters adjusted is continuously trained by using a training sample until the loss function value meets the training stop condition, so as to obtain the trained fault diagnosis model.
It should be noted that, after the trained fault diagnosis model is obtained, the trained fault diagnosis model needs to be tested by using the test sample set obtained by the division, where the maximum iteration number may be set based on the actual requirement or the actual situation, and for example, the maximum iteration number may be 500. And in the test process, when the fault diagnosis model is continuously preset times, for example, 50 times, and the model precision is not improved, the water injection pump operation information sample can be processed based on the preset fault diagnosis model so as to determine the corresponding water injection pump operation state, otherwise, the model is continuously trained. In addition, as shown in fig. 5, the built fault diagnosis model is trained and tested, and the optimizer uses an Adam optimizer, so that the final training accuracy can reach 74.3%.
Because the prior art has the problem that the sample number of various training samples is unbalanced, based on the problem, the related loss function (SoftMax cross entropy loss function) is weighted by using a class balance function, and the obtained loss function is:
wherein,,is the probability of model estimation samples, +.>Indicating that the training sample belongs to->The probability distribution of a class is determined, ，/>Indicating that the training sample belongs to->Probability distribution of class->Is->The actual number of samples of the class training samples, +.>Is->The effective number of samples of the class training sample and is an exponential function of the actual number of samples, +.>Is a superparameter that controls the weights.
In addition, it should be noted that the above-mentioned preset fault diagnosis model may include network parameters including: the number of input neurons is 2048, and the number of output neurons is 16.
It should be further noted that, considering that the components of the obtained water injection pump operation information sample are complex, the signal characteristics included in the single-scale convolutional neural network are difficult to be fully extracted, and based on the signal characteristics, the model result of the preset fault diagnosis model can be improved so as to better extract the signal characteristics of the water injection pump operation information sample. In an example, the network structure of the above-mentioned preset fault diagnosis model before improvement may be shown in fig. 2, where the network structure includes 5 convolution layers (Conv 1-Conv 5), 3 maximum pooling layers (MaxPooling 1-MaxPooling 3), 2 full connection layers (FullyConnected 1-FullyConnected 2), and one SoftMax layer, the convolution kernels of the convolution layers are all 5, the number of the convolution kernels is 16, 32, and 64, the activation functions of the convolution layers are ReLu functions, a BN layer is connected after each convolution layer, the size of the maximum pooling layer is 5, and a ReLu function layer is added between the two full connection layers.
The improved network structure of the preset fault diagnosis model is shown in fig. 3, and two parallel convolution layers are respectively added into two convolution layers Conv2 and Conv4 and are marked as convolution layers Conv21, conv22, conv41 and Conv42. Wherein the convolution kernels of Conv21 and Conv41 are 3, the number of convolution kernels is 16, the convolution kernels of Conv22 and Conv42 are 7, the number of convolution kernels is 32, the newly added convolution layers also use ReLu as an activation function, and one BN layer normalizes the output, the characteristics of each convolution layer in the parallel convolution layers are spliced in series, and if the input into the parallel convolution layers is exemplifiedThe output is +.>。
In this embodiment, the information processing apparatus obtains a training sample set, performs classification calculation on the training samples based on a preset fault diagnosis model for each training sample in the training sample set, so as to obtain a reference running state corresponding to the training sample, further, may determine a loss function value of the preset fault diagnosis model according to the reference running state of the target training sample and the tag running state of the target training sample, and further, may train the preset fault diagnosis model by using the training samples based on the loss function value until the loss function value satisfies a training stop condition, so as to ensure that a more accurate fault diagnosis model may be obtained.
In one embodiment, before acquiring the training sample set, the above-mentioned information processing method further includes:
obtaining original water injection pump operation information samples of the water injection pump in a plurality of third preset times respectively, wherein the third preset time is longer than the second preset time;
and dividing the original water injection pump operation information samples according to a sliding time window algorithm aiming at any one of the original water injection pump operation information samples in a plurality of third preset time to obtain a plurality of water injection pump operation information samples corresponding to each original water injection pump operation information sample.
In some embodiments, the third preset time is greater than the second preset time, where both the third preset time and the second preset time may be preset based on actual experience or situation, and are not excessively limited herein.
It should be noted that, the sliding time window algorithm refers to selecting a time window with a fixed length, dividing a piece of data at a certain position, and then sliding to another position of the data for dividing again, so that the number of data can be increased without losing the original information.
In one example, after the original water injection pump operation information samples are obtained, the original water injection pump operation information samples are cut in a sliding time window mode, and each original water injection pump operation information sample can be cut into 7 water injection pump operation information samples, and each water injection pump operation information sample comprises 2048 acquisition signals, assuming that the window size is 2048 and the sliding step size is 1024.
In this embodiment, the original water injection pump operation information sample may be processed by a sliding time window algorithm to obtain a plurality of water injection pump operation information samples corresponding to the original water injection pump operation information sample, so that the number of training samples may be increased to facilitate training of a subsequent model.
Based on this, the information processing method provided in the embodiment of the present application is described in detail below with reference to fig. 4.
Fig. 4 is a schematic flow chart of an information processing method according to an embodiment of the present application.
As shown in fig. 4, the execution subject of the method may be an information processing apparatus, and the method may specifically include the steps of:
s410, water injection pump operation information of the water injection pump in a first preset time is obtained.
The above-mentioned water injection pump operation information may include a plurality of sampling signals. The water injection pump may be a plunger type water injection pump or other type of water injection pump. The first preset time may be a preset period based on actual experience or situation, for example, the preset period may be one day or one hour, which is not limited herein. The above-mentioned sample signals may include one or more of an acceleration signal, a velocity signal, a displacement signal, a torque signal, and a sound pressure signal.
In one example, it is assumed that the water injection pump is a plunger type water injection pump on which a vibration sensor is mounted, and based on this, the information processing apparatus can acquire water injection pump operation information of the above-described plunger type water injection pump within a first preset time through the vibration sensor.
S420, classifying and calculating the operation information of the water injection pump based on the fault diagnosis model to obtain the operation state of the water injection pump corresponding to the operation information of the water injection pump.
The fault diagnosis model comprises a plurality of model parameters, the model parameters are obtained after a loss function is adjusted on the basis of training samples to preset model parameters in the fault diagnosis model, and the loss function is a function obtained after weighting processing is carried out on the basis of the number of samples of the training samples corresponding to the running states of the water injection pumps.
The information processing device may input the water injection pump operation information to a failure diagnosis model after acquiring the water injection pump operation information, and perform classification calculation on the water injection pump operation information through the failure diagnosis model to obtain a water injection pump operation state corresponding to the water injection pump operation information.
It should be noted that the above-mentioned operation state of the water injection pump may include states such as normal operation, bearing bush wear failure, pump head spring taking, pump head loosening, motor eccentricity, destructive wear of the outer ring of the double bearing ball, bearing bracket damage, cross head wear, double bearing bracket damage, wear in the plunger, plunger loosening, western bearing ball outer ring wear, bearing bush wear, pump head spring taking mixing failure, bearing bush wear, anchor bolt loosening mixing failure, bearing bush wear, motor eccentricity mixing failure, bearing bush wear, motor bolt loosening mixing failure, and the like.
In the embodiment of the application, the operation information of the water injection pump in the first preset time can be obtained, and then the obtained operation information of the water injection pump can be classified and calculated based on the fault diagnosis model, so that the operation state of the water injection pump corresponding to the operation information of the water injection pump can be obtained. The fault diagnosis model may include a plurality of model parameters, where the model parameters are obtained by training a preset fault diagnosis model based on training samples by using a loss function, and the loss function is obtained by weighting the number of samples of the training samples corresponding to the operation states of the water injection pumps. Therefore, in the process of training the preset fault diagnosis model, the situation that the number of training samples is unbalanced is considered, a more accurate fault diagnosis model can be obtained, and whether the water injection pump breaks down and the fault type can be accurately determined based on the running state of the water injection pump output by the fault diagnosis model.
Because of considering the situation that the number of various training samples is unbalanced in the model training process in real life, the embodiment of the application can use the class balance function to improve the SoftMax cross entropy function so as to solve the problem of the influence of the unbalanced number of various samples on the accuracy of model output results in the prior art, and based on the influence, in one embodiment, the above-mentioned related loss function can satisfy the following formula:
Wherein,,is the probability of model estimation samples, +.>Indicating that the training sample belongs to->The probability distribution of a class is determined,，/>indicating that the training sample belongs to->Probability distribution of class->Is->The actual number of samples of the class training samples, +.>Is->The effective number of samples of the class training sample and is an exponential function of the actual number of samples, +.>Is a superparameter that controls the weights.
In this embodiment, the loss function value is a function obtained by weighting the number of training samples corresponding to the operation state of the water injection pump with a class balance function. Therefore, after the class balance function is weighted, the magnitude of the loss function and the effective sample number of the samples are inversely related, and the misclassification cost of a few classes is improved, so that the overall classification accuracy of the model can be improved.
In addition, because the types of parts and the number of parts of the water injection pump and the types of movement involved in the operation of the water injection pump are various, the cost of the obtained operation information of the water injection pump to be processed is complex, the signal characteristics included in the single-scale convolutional neural network are difficult to fully extract, and the accuracy of the operation state of the water injection pump which is determined later is low easily.
To solve the above problem, in some embodiments, the fault diagnosis model according to the embodiments of the present application may include a convolutional neural network, where the convolutional neural network includes at least one first convolutional layer, a pooling layer, and a SoftMax layer, and the first convolutional layer includes a plurality of second convolutional layers that are juxtaposed with each other, and the sizes of the plurality of second convolutional layers are different from each other.
In one example, the first convolution layer includes three mutually independent second convolution layers that differ in size from each other if the input into the first convolution layer is characterized byThe input feature->Three second convolution layers with different scales are needed to pass through simultaneously, and the respective outputs of the three second convolution layers are respectively +.>、/>、Thus, the characteristics with different granularity in the operation information of the water injection pump can be respectively extracted through the three second convolution layers, and then the output characteristics of the three convolution layers with different scales are connected in series to obtain the output characteristics of the first convolution layerAnd then input it into the next layer.
In this embodiment, features with different granularities in the operation information of the water injection pump may be extracted through a plurality of second convolution layers, so that the accuracy of the operation state of the water injection pump may be more accurately determined.
In addition, in order to demonstrate the advantages of the method of the present invention over the prior art, as shown in fig. 5, curves 11, 12, 13 are loss curves of the convolutional neural network, the quasi-balanced convolutional neural network, and the fault diagnosis model during training, and curves 21, 22, 23 are loss curves of the convolutional neural network, the quasi-balanced convolutional neural network, and the fault diagnosis model during testing, respectivelyA curve. In this way, by comparing the classification precision of the same type model and the fault diagnosis model on the same data set, the same type model is the built convolutional neural network, and the loss function is a cross entropy loss functionThe class-balanced parallel convolutional neural network model provided by the invention has higher convergence speed and higher classification precision, and the advantages of the optimization method provided by the invention are reflected.
Based on the same inventive concept, the embodiment of the application also provides an information processing device. An information processing apparatus provided in an embodiment of the present application will be described in detail with reference to fig. 6.
Fig. 6 is a schematic structural diagram of an information processing apparatus according to an embodiment of the present application.
As shown in fig. 6, the information processing apparatus 600 may include: an acquisition module 610 and a classification calculation module 620.
The acquiring module 610 is configured to acquire water injection pump operation information of the water injection pump within a first preset time, where the water injection pump operation information includes a plurality of sampling signals;
the classification calculation module 620 is configured to perform classification calculation on the operation information of the water injection pump based on the fault diagnosis model, so as to obtain an operation state of the water injection pump corresponding to the operation information of the water injection pump;
the fault diagnosis model comprises a plurality of model parameters, the model parameters are obtained after a loss function is trained on the basis of training samples to a preset fault diagnosis model, and the loss function is a function obtained after weighting processing is performed on the basis of the number of samples of the training samples corresponding to the running state of the water injection pump.
In one embodiment, the loss function satisfies the following formula:
wherein,,is a model estimation sampleProbability of book (I)>Indicating that the training sample belongs to->The probability distribution of a class is determined,，/>indicating that the training sample belongs to->Probability distribution of class->Is->The actual number of samples of the class training samples, +.>Is->The effective number of samples of the class training sample and is an exponential function of the actual number of samples, +.>Is a superparameter that controls the weights.
In one embodiment, the fault diagnosis model comprises a convolutional neural network comprising at least a first convolutional layer, a pooling layer and a SoftMax layer, the first convolutional layer comprising a plurality of second convolutional layers juxtaposed to each other, the plurality of second convolutional layers being different in size from each other.
In one embodiment, the information processing apparatus related to the foregoing further includes a training module, where the training module is specifically configured to:
before classifying and calculating the operation information of the water injection pump based on the fault diagnosis model to obtain the operation state of the water injection pump corresponding to the operation information of the water injection pump, acquiring a training sample set, wherein the training sample set comprises a plurality of training samples and the label operation state corresponding to each training sample, and the training samples comprise the operation information sample of the water injection pump in a second preset time;
for each training sample, the following is performed:
classifying and calculating the training samples based on a preset fault diagnosis model to obtain reference running states corresponding to the training samples;
determining a loss function value of a preset fault diagnosis model according to a reference running state of a target training sample and a label running state of the target training sample, wherein the target training sample is any one of a plurality of training samples;
and training the preset fault diagnosis model by using a training sample based on the loss function value of the preset fault diagnosis model to obtain a trained fault diagnosis model.
In one embodiment, the information processing module further includes a sample dividing module, where the sample dividing module is specifically configured to:
Before a training sample set is obtained, obtaining original water injection pump operation information samples of the water injection pump in a plurality of third preset times respectively, wherein the third preset time is longer than the second preset time;
and dividing the original water injection pump operation information samples according to a sliding time window algorithm aiming at any one of the original water injection pump operation information samples in a plurality of third preset time to obtain a plurality of water injection pump operation information samples corresponding to each original water injection pump operation information sample.
In the embodiment of the application, the operation information of the water injection pump in the first preset time can be obtained, and then the obtained operation information of the water injection pump can be classified and calculated based on the fault diagnosis model, so that the operation state of the water injection pump corresponding to the operation information of the water injection pump can be obtained. The fault diagnosis model may include a plurality of model parameters, where the model parameters are obtained by training a preset fault diagnosis model based on training samples by using a loss function, and the loss function is obtained by weighting the number of samples of the training samples corresponding to the operation states of the water injection pumps. Therefore, in the process of training the preset fault diagnosis model, the situation that the number of training samples is unbalanced is considered, a more accurate fault diagnosis model can be obtained, and whether the water injection pump breaks down and the fault type can be accurately determined based on the running state of the water injection pump output by the fault diagnosis model.
Each module in the information processing apparatus provided in the embodiment of the present application may implement the method steps in the embodiment shown in fig. 4, and may achieve the technical effects corresponding to the steps, which are not described herein for brevity.
Fig. 7 shows a schematic hardware structure of an electronic device according to an embodiment of the present application.
A processor 701 may be included in an electronic device, as well as a memory 702 in which computer program instructions are stored.
In particular, the processor 701 described above may include a Central Processing Unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured to implement one or more integrated circuits of embodiments of the present application.
Memory 702 may include mass storage for data or instructions. By way of example, and not limitation, memory 702 may comprise a Hard Disk Drive (HDD), floppy Disk Drive, flash memory, optical Disk, magneto-optical Disk, magnetic tape, or universal serial bus (Universal Serial Bus, USB) Drive, or a combination of two or more of the foregoing. The memory 702 may include removable or non-removable (or fixed) media, where appropriate. Memory 702 may be internal or external to the integrated gateway disaster recovery device, where appropriate. In a particular embodiment, the memory 702 is a non-volatile solid state memory.
The memory may include Read Only Memory (ROM), random Access Memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical/tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors) it is operable to perform the operations described with reference to methods in accordance with aspects of the present disclosure.
The processor 701 implements any one of the information processing methods of the above embodiments by reading and executing computer program instructions stored in the memory 702.
In one example, the electronic device may also include a communication interface 703 and a bus 710. As shown in fig. 7, the processor 701, the memory 702, and the communication interface 703 are connected by a bus 710 and perform communication with each other.
The communication interface 703 is mainly used for implementing communication between each module, device, unit and/or apparatus in the embodiments of the present application.
Bus 710 includes hardware, software, or both that couple the components of the online data flow billing device to each other. By way of example, and not limitation, the buses may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a micro channel architecture (MCa) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus, or a combination of two or more of the above. Bus 710 may include one or more buses, where appropriate. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.
In addition, in combination with the information processing method in the above embodiment, the embodiment of the application may be implemented by providing a computer storage medium. The computer storage medium has stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement the information processing method provided in the embodiments of the present application.
Embodiments of the present application also provide a computer program product, where instructions in the computer program product, when executed by a processor of an electronic device, cause the electronic device to perform an information processing method as provided in the embodiments of the present application.
It should be clear that the present application is not limited to the particular arrangements and processes described above and illustrated in the drawings. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present application are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions, or change the order between steps, after appreciating the spirit of the present application.
The functional blocks shown in the above block diagrams may be implemented in hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, a plug-in, a function card, or the like. When implemented in software, the elements of the present application are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine readable medium or transmitted over transmission media or communication links by a data signal carried in a carrier wave. A "machine-readable medium" may include any medium that can store or transfer information. Examples of machine-readable media include electronic circuitry, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio Frequency (RF) links, and the like. The code segments may be downloaded via computer networks such as the internet, intranets, etc.
It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be different from the order in the embodiments, or several steps may be performed simultaneously.
Aspects of the present disclosure are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable information processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable information processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to being, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware which performs the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In the foregoing, only the specific embodiments of the present application are described, and it will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the systems, modules and units described above may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein. It should be understood that the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present application, which are intended to be included in the scope of the present application.
Claims (9)
1. An information processing method, characterized in that the method comprises:
acquiring water injection pump operation information of a water injection pump in a first preset time, wherein the water injection pump operation information comprises a plurality of sampling signals;
classifying and calculating the water injection pump operation information based on a fault diagnosis model to obtain a water injection pump operation state corresponding to the water injection pump operation information;
the fault diagnosis model comprises a plurality of model parameters, the model parameters are obtained after a loss function is trained on the basis of training samples, and the loss function is a function obtained after weighting processing is performed on the basis of the number of samples of the training samples corresponding to the running states of the water injection pumps.
2. The method of claim 1, wherein the loss function satisfies the following formula:
wherein (1)>Is the probability of model estimation samples, +.>Indicating that the training sample belongs to->Probability distribution of class->，/>Indicating that the training sample belongs to->Probability distribution of class->Is->The actual number of samples of the class training samples, +.>Is the firstThe effective number of samples of the class training sample and is an exponential function of the actual number of samples, +.>Is a superparameter that controls the weights.
3. The method of claim 1, wherein the fault diagnosis model comprises a convolutional neural network comprising at least a first convolutional layer, a pooled layer, and a SoftMax layer, the first convolutional layer comprising a plurality of second convolutional layers juxtaposed to one another, the plurality of second convolutional layers being different in size from one another.
4. The method of claim 1, wherein prior to classifying the water injection pump operation information based on the fault diagnosis model to obtain a water injection pump operation state corresponding to the water injection pump operation information, the method further comprises:
acquiring a training sample set, wherein the training sample set comprises a plurality of training samples and a label running state corresponding to each training sample, and the training samples comprise water injection pump running information samples of a water injection pump in a second preset time;
For each training sample, the following is performed:
classifying and calculating the training samples based on a preset fault diagnosis model to obtain reference running states corresponding to the training samples;
determining a loss function value of a preset fault diagnosis model according to a reference running state of a target training sample and a label running state of the target training sample, wherein the target training sample is any one of the plurality of training samples;
and training the preset fault diagnosis model by using a training sample based on the loss function value of the preset fault diagnosis model to obtain a trained fault diagnosis model.
5. The method of claim 4, wherein prior to acquiring the training sample set, the method further comprises:
obtaining original water injection pump operation information samples of the water injection pump in a plurality of third preset times respectively, wherein the third preset time is longer than the second preset time;
and dividing the original water injection pump operation information samples within the third preset time according to a sliding time window algorithm aiming at any one of the original water injection pump operation information samples to obtain a plurality of water injection pump operation information samples corresponding to each original water injection pump operation information sample.
6. An information processing apparatus, characterized in that the apparatus comprises:
the water injection pump operation information comprises a plurality of sampling signals;
the classification calculation module is used for carrying out classification calculation on the water injection pump operation information based on a fault diagnosis model so as to obtain a water injection pump operation state corresponding to the water injection pump operation information;
the fault diagnosis model comprises a plurality of model parameters, the model parameters are obtained after a loss function is trained on the basis of training samples, and the loss function is a function obtained after weighting processing is performed on the basis of the number of samples of the training samples corresponding to the running states of the water injection pumps.
7. The apparatus of claim 6, wherein the fault diagnosis model comprises a convolutional neural network comprising at least a first convolutional layer, a pooled layer, and a SoftMax layer, the first convolutional layer comprising a plurality of second convolutional layers juxtaposed to one another, the plurality of second convolutional layers being different in size from one another.
8. An electronic device, the device comprising: a processor and a memory storing computer program instructions;
The processor reads and executes the computer program instructions to implement the information processing method according to any one of claims 1 to 5.
9. A computer storage medium, characterized in that the computer storage medium has stored thereon computer program instructions which, when executed by a processor, implement the information processing method according to any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310720465.3A CN116522998A (en) | 2023-06-16 | 2023-06-16 | Information processing method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310720465.3A CN116522998A (en) | 2023-06-16 | 2023-06-16 | Information processing method, device, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116522998A true CN116522998A (en) | 2023-08-01 |
Family
ID=87396110
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310720465.3A Pending CN116522998A (en) | 2023-06-16 | 2023-06-16 | Information processing method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116522998A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112699597A (en) * | 2020-12-09 | 2021-04-23 | 西安交通大学 | Nuclear power starting water pump rolling bearing fault detection method and system |
| CN113033309A (en) * | 2021-02-25 | 2021-06-25 | 北京化工大学 | Fault diagnosis method based on signal downsampling and one-dimensional convolution neural network |
| WO2021243838A1 (en) * | 2020-06-03 | 2021-12-09 | 苏州大学 | Fault diagnosis method for intra-class self-adaptive bearing under variable working conditions |
| CN115879026A (en) * | 2022-12-07 | 2023-03-31 | 吉林大学 | Rotary drilling rig hydraulic system fault diagnosis method based on convolutional neural network-integrated classifier |
-
2023
- 2023-06-16 CN CN202310720465.3A patent/CN116522998A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021243838A1 (en) * | 2020-06-03 | 2021-12-09 | 苏州大学 | Fault diagnosis method for intra-class self-adaptive bearing under variable working conditions |
| CN112699597A (en) * | 2020-12-09 | 2021-04-23 | 西安交通大学 | Nuclear power starting water pump rolling bearing fault detection method and system |
| CN113033309A (en) * | 2021-02-25 | 2021-06-25 | 北京化工大学 | Fault diagnosis method based on signal downsampling and one-dimensional convolution neural network |
| CN115879026A (en) * | 2022-12-07 | 2023-03-31 | 吉林大学 | Rotary drilling rig hydraulic system fault diagnosis method based on convolutional neural network-integrated classifier |
Non-Patent Citations (1)
| Title |
|---|
| 郑宇: "基于深度学习的样本不平衡问题研究", 中国优秀硕士学位论文全文数据库信息科技辑, pages 2 - 4 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lou et al. | Machinery fault diagnosis based on domain adaptation to bridge the gap between simulation and measured signals | |
| CN107228766B (en) | Based on the Fault Diagnosis of Roller Bearings for improving multiple dimensioned fuzzy entropy | |
| Tsui et al. | Prognostics and health management: A review on data driven approaches | |
| Tong et al. | Bearing fault diagnosis based on domain adaptation using transferable features under different working conditions | |
| CN113887136B (en) | Electric automobile motor bearing fault diagnosis method based on improved GAN and ResNet | |
| CN108181105B (en) | Rolling bearing fault pre-diagnosis method and system based on logistic regression and J divergence | |
| Liang et al. | A deep learning method for motor fault diagnosis based on a capsule network with gate-structure dilated convolutions | |
| CN112633339A (en) | Bearing fault intelligent diagnosis method, bearing fault intelligent diagnosis system, computer equipment and medium | |
| CN113532829A (en) | Reciprocating compressor fault diagnosis method based on improved RCMDE | |
| Wang et al. | Remaining useful life prediction considering joint dependency of degradation rate and variation on time-varying operating conditions | |
| CN116992226A (en) | Water pump motor fault detection method and system | |
| CN116577677A (en) | Discharging test system and method for retired power battery | |
| CN115081673B (en) | Abnormality prediction method and device for oil and gas pipeline, electronic equipment and medium | |
| CN115791174B (en) | Rolling bearing abnormality diagnosis method, system, electronic equipment and storage medium | |
| CN116089870A (en) | Industrial equipment fault prediction method and device based on meta-learning under small sample condition | |
| CN115876467A (en) | Pseudo label transfer type two-stage field self-adaptive rolling bearing fault diagnosis method | |
| CN114491828A (en) | Compound fault diagnosis method based on multi-gating hybrid expert multi-task learning | |
| CN116482524A (en) | Power transmission and distribution switch state detection method and system | |
| CN116012681A (en) | Method and system for diagnosing motor faults of pipeline robot based on sound vibration signal fusion | |
| CN115470717A (en) | Method, device, equipment and storage medium for predicting remaining life of robot | |
| CN116295948A (en) | Abnormality detection method, system and storage medium of industrial temperature sensor in large temperature difference environment | |
| CN112990329B (en) | System abnormality diagnosis method and device | |
| CN114861349A (en) | Rolling bearing RUL prediction method based on model migration and wiener process | |
| CN116502131B (en) | Bearing fault diagnosis model training method and device based on transfer learning | |
| Lee et al. | Remaining useful life estimation for ball bearings using feature engineering and extreme learning machine |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230801 |