CN114708675A - DAE-based electric submersible pump fault diagnosis method, system, terminal and medium - Google Patents
DAE-based electric submersible pump fault diagnosis method, system, terminal and medium Download PDFInfo
- Publication number
- CN114708675A CN114708675A CN202210260063.5A CN202210260063A CN114708675A CN 114708675 A CN114708675 A CN 114708675A CN 202210260063 A CN202210260063 A CN 202210260063A CN 114708675 A CN114708675 A CN 114708675A
- Authority
- CN
- China
- Prior art keywords
- data
- submersible pump
- fault
- electric submersible
- dae
- 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
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
- G07C3/08—Registering or indicating the production of the machine either with or without registering working or idle time
-
- 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
- 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
- G06N3/045—Combinations of networks
-
- 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
Abstract
The invention discloses a DAE-based electric submersible pump fault diagnosis method, a DAE-based electric submersible pump fault diagnosis system, a DAE-based electric submersible pump fault diagnosis terminal and a DAE-based electric submersible pump fault diagnosis medium, and relates to the technical field of electric submersible pump fault diagnosis, wherein the technical scheme is as follows: collecting operation data of the production process of the electric submersible pump in real time; preprocessing the operation data and extracting relevant attribute features; analyzing the related attribute characteristics through a noise reduction self-coding model to obtain fault characteristics in the related attribute characteristics; and classifying the fault characteristics by adopting a characteristic classification model to obtain the fault type of the electric submersible pump corresponding to the fault characteristics. The method combines the noise reduction self-coding model and the characteristic classification model, can effectively estimate the running state of the electric submersible pump, judges the type of the electric submersible pump fault to be generated in advance, and effectively improves the timeliness of fault diagnosis.
Description
Technical Field
The invention relates to the technical field of electric submersible pump fault diagnosis, in particular to a DAE-based electric submersible pump fault diagnosis method, system, terminal and medium.
Background
The Electric Submersible Pump (ESP) lifting technology is widely applied to non-flowing high-yield wells and high-water-content wells. The ESP system has strong relevance and multi-unit structure, and the ESP can lift crude oil from the ground to the surface under the conditions of higher temperature and greater depth, which easily causes the ESP to break down in the production process, thereby causing production interruption and serious economic loss.
In recent years, the development of sensors makes the real-time production condition monitoring of the ESP easier, the sensors can record the production data of the ESP in real time according to a certain time interval, and the effective processing and analysis of the production data of the ESP well can provide a new idea for diagnosing the faults of the ESP. The current fault diagnosis technology for pump equipment is mainly based on the data of the operating voltage and current of a pump; on one hand, the timeliness of fault diagnosis is poor only according to the operating voltage and current data of the pump; on the other hand, the influence of ESP production data and complex environment on the failure of the pump equipment is not considered, and the accuracy of the failure diagnosis result of the pump equipment is to be further improved.
Therefore, how to design a method, a system, a terminal and a medium for diagnosing the fault of the DAE-based electric submersible pump, which can overcome the defects, is a problem which needs to be solved urgently at present.
Disclosure of Invention
In order to solve the defects in the prior art, the invention aims to provide a DAE-based electric submersible pump fault diagnosis method, system, terminal and medium, and combines a noise reduction self-coding model and a feature classification model, so that the running state of the electric submersible pump can be effectively estimated, the type of the electric submersible pump fault to be generated can be judged in advance, and the timeliness of fault diagnosis is effectively improved.
The technical purpose of the invention is realized by the following technical scheme:
in a first aspect, a DAE-based electrical submersible pump fault diagnosis method is provided, comprising the steps of:
collecting operation data of the production process of the electric submersible pump in real time;
preprocessing the operation data and extracting relevant attribute features;
analyzing the related attribute characteristics through a noise reduction self-coding model to obtain fault characteristics in the related attribute characteristics;
and classifying the fault characteristics by adopting a characteristic classification model to obtain the fault type of the electric submersible pump corresponding to the fault characteristics.
Further, the preprocessing of the running data comprises data cleaning and data padding;
the data cleaning comprises the following steps: deleting the attribute of the corresponding column in the corresponding data when the vacancy value of the attribute in the data exceeds the first vacancy rate; deleting the corresponding data samples when the vacancy rate of the data samples exceeds a second vacancy rate; when the characteristic data is a single value, deleting the corresponding characteristic data; when the production time of the electric submersible pump is less than 24 hours, deleting corresponding sample data;
and the data filling adopts a linear interpolation method to fill the vacant data.
Further, the process of extracting the relevant attribute features specifically includes:
analyzing the maximum information coefficient between the characteristic data in the operation data and each target variable by taking four variables of wellhead temperature, bottom hole flowing pressure, pump current and converted daily liquid production as target variables;
and screening out the related attribute characteristics by combining the Pearson correlation coefficient and the maximum information coefficient.
Further, the related attribute characteristics include bottom hole flowing pressure, converted daily liquid yield, test liquid amount, daily liquid yield _ distribution, test water amount, converted daily liquid yield, daily liquid yield _ distribution, daily oil yield _ distribution, test oil amount, converted daily oil yield, pump inlet temperature, water content _ distribution, water-gas ratio _ distribution, well head temperature, oil pressure, pump current and pump voltage.
Further, the analysis process of the noise reduction self-coding model on the relevant attribute features specifically comprises:
dividing the related attribute characteristics into data to be measured and long-term normal data;
inputting long-term normal data into a noise reduction self-coding model, pre-training and adjusting, and calculating to obtain a standard reconstruction value and a reconstruction error range;
inputting the data to be detected into a noise reduction self-coding model, and calculating to obtain a prediction reconstruction value and a reconstruction error range corresponding to the data to be detected;
and judging the data to be measured corresponding to the predicted reconstruction value and the reconstruction error range exceeding the standard reconstruction value and the reconstruction error range as fault characteristics.
Further, the noise reduction self-coding model establishes a seven-layer neural network after pre-training and tuning:
the first layer is configured with 20 coding nodes;
the second layer is configured with 10 coding nodes;
the third layer is configured with 5 coding nodes;
the fourth layer is configured with 3 coding nodes;
the fifth layer is configured with 5 decoding nodes;
the sixth layer is configured with 10 decoding nodes;
the seventh layer is configured with 20 decoding nodes.
Furthermore, the characteristic classification model is provided with an SVM classifier and a GAS optimizer, and the electric submersible pump fault types comprise four types of voltage unbalance, pipe column leakage, overload pump stopping and underload pump stopping.
In a second aspect, a DAE-based electrical submersible pump fault diagnostic system is provided, comprising:
the data acquisition module is used for acquiring the operation data of the electric submersible pump in the production process in real time;
the characteristic extraction module is used for extracting relevant attribute characteristics after preprocessing the operation data;
the fault analysis module is used for analyzing the related attribute characteristics through the noise reduction self-coding model to obtain fault characteristics in the related attribute characteristics;
and the fault classification module is used for classifying and processing the fault characteristics by adopting the characteristic classification model to obtain the fault type of the electric submersible pump corresponding to the fault characteristics.
In a third aspect, there is provided a computer terminal comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the method of diagnosing a fault in a DAE-based electrical submersible pump according to any one of the first aspect when executing the program.
In a fourth aspect, there is provided a computer readable medium having a computer program stored thereon, the computer program being executable by a processor to implement the method of fault diagnosis for a DAE-based electric submersible pump as defined in any one of the first aspects.
Compared with the prior art, the invention has the following beneficial effects:
1. the DAE-based electric submersible pump fault diagnosis method provided by the invention combines the noise reduction self-coding model and the feature classification model, can effectively estimate the running state of the electric submersible pump, judges the type of the electric submersible pump fault to be generated in advance, and effectively improves the timeliness of fault diagnosis;
2. according to the method, data cleaning and data filling processing are carried out on the operation data of the electric submersible pump, and the detailed processing mode is optimally designed, so that the interference on subsequent processing analysis can be effectively reduced, and the reasonability and reliability of extraction of relevant attribute features are ensured;
3. according to the invention, the influence of ESP production data and a complex environment on the fault of the pump equipment is considered, and the pump current and the pump voltage are combined, so that the accuracy of the fault diagnosis result of the pump equipment can be effectively improved, and the timeliness of fault diagnosis is further improved;
4. according to the invention, the neural network structure of the noise reduction self-coding model is optimally designed according to the category and the number of the related attribute characteristics, so that abnormal data can be accurately and reliably identified;
5. according to the invention, the SVM classifier and the GAs optimizer are combined to establish the feature classification model, so that the accuracy of the whole model can be obviously improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
FIG. 1 is a flow chart of fault analysis in an embodiment of the invention;
FIG. 2 is a graph of four interpolation effects in an embodiment of the present invention;
FIG. 3 is a graph illustrating the calculated maximum information coefficient for an embodiment of the present invention, where a is the correlation coefficient between the reduced daily fluid production and the MIC for each variable, b is the correlation coefficient between the wellhead temperature and the MIC for each variable, c is the correlation coefficient between the bottom hole flow pressure and the MIC for each variable, and d is the correlation coefficient between the pump current and the MIC for each variable;
FIG. 4 is a diagram illustrating the training results of a noise reduction self-coding model according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of the results of fault diagnosis in an embodiment of the present invention;
FIG. 6 is a functional diagram of a feature classification model according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of the fitness of GAs in an embodiment of the present invention;
fig. 8 is a system block diagram in an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to examples and accompanying drawings, and the exemplary embodiments and descriptions thereof are only used for explaining the present invention and are not meant to limit the present invention.
Example 1: the DAE-based electrical submersible pump fault diagnosis method is specifically realized by the following steps as shown in FIG. 1.
First, data acquisition
The running data of the production process of the electric submersible pump is collected in real time through the sensor, and the detection data can be also retrieved from a data collection and monitoring control System (SCADA), a database and the like, so that data support is provided for the subsequent steps.
Second, data cleaning
Due to some non-human factors, the detection system and the sensor have some missing or abnormal values of collected data, and data cleaning is needed to be carried out on the data.
In this embodiment, when the vacancy value of the attribute in the data exceeds 90%, the attribute of the corresponding column in the corresponding data is deleted, and the first vacancy rate may also be set as another percentage according to the requirement. And when the vacancy rate of the data samples exceeds 90%, deleting the corresponding data samples, and setting the second vacancy rate as other percentage ratios according to requirements. When the feature data is a single value, deleting the corresponding feature data, such as surface casing pressure, technical casing pressure and other variables, which have no influence on the data and can increase the experiment running time. When the production time of the electric submersible pump is less than 24h, deleting corresponding sample data as shown in table 1.
TABLE 1 production time less than 24 hours
Data padding
The interpolation method generally comprises an inverse distance weighted interpolation method, a Newton interpolation method, a linear interpolation method and a random forest filling method. The experiment was conducted using the "831352627" well as an example, and the four interpolation effects are shown in FIG. 2.
The distance between the inserted data and the variable of various interpolation algorithms is calculated through Euclidean distance for comparison, as shown in Table 2, the interpolation method taking linear interpolation as data has better effect on the operation data considering the influence of ESP production data and complex environment on the pump equipment fault in the application. The data padding uses a linear interpolation method to pad the vacant data.
TABLE 2 Euclidean distances of the four interpolation distance results
Feature extraction
The data attribute characteristics of the submersible pump are more, some characteristics are not meaningful for researching the fault of the submersible pump and can negatively influence the subsequent data research, so that the characteristics need to be selected by a method with different contrasts, and the method is shown in table 3. The electric submersible pump has the characteristics of multiple samples and complex attributes, and a method with low computational complexity and high robustness is selected.
Selecting four variables of wellhead temperature, bottom hole flowing pressure, pump current and reduced daily liquid production as target variables, wherein the four different indexes need higher robustness; the electric submersible pump has large data sample amount, needs simpler calculation and saves the algorithm of the running time. Maximum information coefficients between the other features and each are analyzed. There were the following 4 results, as shown in FIG. 3.
Therefore, in this embodiment, the process of extracting the relevant attribute features specifically includes: analyzing the maximum information coefficient between the characteristic data in the operation data and each target variable by taking four variables of wellhead temperature, bottom hole flowing pressure, pump current and converted daily liquid production as target variables; and screening out the related attribute characteristics by combining the Pearson correlation coefficient and the maximum information coefficient.
The relevant attribute characteristics comprise bottom hole flowing pressure, converted daily liquid yield, test liquid amount, daily liquid yield _ distribution, test water amount, converted daily liquid yield, daily liquid yield _ distribution, daily oil yield _ distribution, test oil amount, converted daily oil yield, pump inlet temperature, water content _ distribution, water-gas ratio _ distribution, wellhead temperature, oil pressure, pump current and pump voltage.
Fifth, failure analysis
It can be seen from the data shown in table 4 that the value of the "pump voltage" is generally large, and is larger than the "oil pressure" by more than one hundred times, which requires normalization and standardization processes to provide data support for model training.
Table 4 electric submersible pump section data
An autoencoder is a type of artificial neural network that learns efficient data value encoding in an unsupervised manner. The purpose of an auto-encoder is to learn the representation (encoding) of a set of data, typically for dimensionality reduction. Both the decoder and encoder learn so that the auto-encoder attempts to generate an expression from the reduced-dimension code that is as close as possible to its original input. A noise reduction self-encoder (DAE) has the core idea that original data is corroded, the DAE overcomes the damage, encoding and decoding operations are carried out, the original data is reconstructed, and characteristics with high stability can be extracted.
Aiming at the fact that the types of the selected data are more, a noise reduction self-coding model constructed by an automatic encoder is selected for data analysis, and as shown in fig. 1, the analysis process of the noise reduction self-coding model on the relevant attribute features specifically comprises the following steps: dividing the related attribute characteristics into data to be measured and long-term normal data; inputting long-term normal data into a noise reduction self-coding model, pre-training and adjusting, and calculating to obtain a standard reconstruction value and a reconstruction error range; inputting the data to be detected into a noise reduction self-coding model, and calculating to obtain a prediction reconstruction value and a reconstruction error range corresponding to the data to be detected; and judging the data to be measured corresponding to the predicted reconstruction value and the reconstruction error range exceeding the standard reconstruction value and the reconstruction error range as fault characteristics.
In this embodiment, the noise reduction self-coding model establishes a seven-layer neural network after pre-training and tuning: the first layer is configured with 20 coding nodes; the second layer is configured with 10 coding nodes; the third layer is configured with 5 coding nodes; the fourth layer is configured with 3 coding nodes; the fifth layer is configured with 5 decoding nodes; the sixth layer is configured with 10 decoding nodes; the seventh layer is configured with 20 decoding nodes.
As shown in fig. 4, from the training results analyzed by the above-mentioned noise-reduced self-coding model, it can be seen that the training loss and the verification loss tend to be stable when the epoch is about 250, and the training set does not have "bounce", i.e. no overfitting phenomenon.
As shown in fig. 5, it can be seen that the fault analysis described above clearly distinguishes between normal data and fault data, and plays a role in monitoring faults. The health data below the horizontal line and the abnormal data above the horizontal line indicate that the machine is abnormal, but the abnormal data is not necessarily recognized by a human.
And sixthly, when the machine is monitored to be abnormal, fault identification needs to be carried out on the fault, so that the fault maintenance of technicians is facilitated, the maintenance cost is reduced, and the maintenance risk is reduced.
In the embodiment, the electric submersible pump fault types comprise four types of voltage unbalance, pipe column leakage, overload pump stop and underload pump stop. By comparing various classification models, "SVM", "decision tree", "Softmax", "Xgboost", and "bayesian classifier", see table 5.
TABLE 5 comparison of models
The improved multi-classification SVM model has higher model precision through comparison, and meanwhile, the Genetic Algorithm (GA) is used for further adjusting parameters of the model, so that the model has more accurate accuracy.
As shown in FIG. 6, the processing flow of the SVM classifier and the GAs optimizer comprises: determining an SVM model, initializing parameters in the SVM, coding an initial value by Gas, combining a data preprocessing result of a Gas optimizer and an error obtained by SVM training to obtain a fitness value, performing selection, intersection and mutation operations, calculating the fitness value to obtain an optimal parameter, and obtaining fault classification.
As shown in fig. 7, in this embodiment, the best parameter value and the optimal fitness function value of the population are [8.069773385747773,0.12224313967837981], and the final tuning result reaches 97.35%.
Example 2: the DAE-based electric submersible pump fault diagnosis system is used for realizing the method described in the embodiment 1 and comprises a data acquisition module, a feature extraction module, a fault analysis module and a fault classification module as shown in FIG. 8.
The data acquisition module is used for acquiring the operation data of the production process of the electric submersible pump in real time; the characteristic extraction module is used for extracting relevant attribute characteristics after preprocessing the operation data; the fault analysis module is used for analyzing the related attribute characteristics through the noise reduction self-coding model to obtain fault characteristics in the related attribute characteristics; and the fault classification module is used for classifying the fault characteristics by adopting the characteristic classification model to obtain the fault type of the electric submersible pump corresponding to the fault characteristics.
The working principle is as follows: the method combines the noise reduction self-coding model and the characteristic classification model, can effectively estimate the running state of the electric submersible pump, judges the type of the electric submersible pump fault to be generated in advance, and effectively improves the timeliness of fault diagnosis; data cleaning and data filling processing are carried out on the operation data of the electric submersible pump, and a detailed processing mode is optimally designed, so that the interference on subsequent processing analysis can be effectively reduced, and the reasonability and reliability of extraction of relevant attribute features are ensured; in addition, the influence of ESP production data and a complex environment on the faults of the pump equipment is considered, and the pump current and the pump voltage are combined, so that the accuracy of the fault diagnosis result of the pump equipment can be effectively improved, and the timeliness of fault diagnosis is further improved; in addition, the neural network structure of the noise reduction self-coding model is optimally designed according to the category and the number of the related attribute characteristics, so that abnormal data can be accurately and reliably identified; in addition, the accuracy of the whole model can be obviously improved by combining the SVM classifier and the GAS optimizer to establish a feature classification model.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above embodiments are only examples of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. The DAE-based electric submersible pump fault diagnosis method is characterized by comprising the following steps of:
collecting operation data of the production process of the electric submersible pump in real time;
preprocessing the operation data and extracting relevant attribute features;
analyzing the related attribute characteristics through a noise reduction self-coding model to obtain fault characteristics in the related attribute characteristics;
and classifying the fault characteristics by adopting a characteristic classification model to obtain the fault type of the electric submersible pump corresponding to the fault characteristics.
2. The DAE-based electrical submersible pump fault diagnostic method of claim 1, wherein the pre-processing of operational data includes data cleaning and data padding;
the data cleaning comprises the following steps: deleting the attribute of the corresponding column in the corresponding data when the vacancy value of the attribute in the data exceeds the first vacancy rate; when the vacancy rate of the data samples exceeds a second vacancy rate, deleting the corresponding data samples; when the characteristic data is a single value, deleting the corresponding characteristic data; when the production time of the electric submersible pump is less than 24 hours, deleting corresponding sample data;
and the data filling adopts a linear interpolation method to fill the vacant data.
3. The method as claimed in claim 1, wherein the extracting process of the related attribute features is specifically as follows:
analyzing the maximum information coefficient between the characteristic data in the operation data and each target variable by taking four variables of wellhead temperature, bottom hole flowing pressure, pump current and converted daily liquid production as target variables;
and screening out the related attribute characteristics by combining the Pearson correlation coefficient and the maximum information coefficient.
4. The DAE-based electrical submersible pump fault diagnosis method of claim 1, wherein the associated attribute characteristics include bottom hole flow pressure, reduced daily liquid production, test liquid quantity, daily liquid production _ allocated, test water quantity, reduced daily water production, daily water production _ allocated, daily gas production _ allocated, daily oil production _ allocated, test oil quantity, reduced daily oil production, pump inlet temperature, water cut _ allocated, water to gas ratio _ allocated, wellhead temperature, oil pressure, pump current, and pump voltage.
5. The DAE-based electrical submersible pump fault diagnosis method of claim 1, wherein the analysis of the relevant attribute features by the noise reduction self-coding model specifically comprises:
dividing the related attribute characteristics into data to be measured and long-term normal data;
inputting long-term normal data into a noise reduction self-coding model, pre-training and adjusting, and calculating to obtain a standard reconstruction value and a reconstruction error range;
inputting the data to be detected into a noise reduction self-coding model, and calculating to obtain a prediction reconstruction value and a reconstruction error range corresponding to the data to be detected;
and judging the data to be measured corresponding to the predicted reconstruction value and the reconstruction error range exceeding the standard reconstruction value and the reconstruction error range as fault characteristics.
6. The DAE-based electrical submersible pump fault diagnosis method of claim 5, wherein the noise reduction self-coding model establishes a seven-layer neural network after pre-training and tuning:
the first layer is configured with 20 coding nodes;
the second layer is configured with 10 coding nodes;
the third layer is configured with 5 coding nodes;
the fourth layer is configured with 3 coding nodes;
the fifth layer is configured with 5 decoding nodes;
the sixth layer is configured with 10 decoding nodes;
the seventh layer is configured with 20 decoding nodes.
7. The DAE-based electrical submersible pump fault diagnosis method of claim 1, wherein the feature classification model is configured with SVM classifier and GAS optimizer, and the electrical submersible pump fault types include three types of voltage imbalance, pipe string loss, overload pump shutdown, and underload pump shutdown.
8. DAE-based electric submersible pump fault diagnosis system is characterized by comprising:
the data acquisition module is used for acquiring the operation data of the electric submersible pump in the production process in real time;
the characteristic extraction module is used for extracting relevant attribute characteristics after preprocessing the operation data;
the fault analysis module is used for analyzing the related attribute characteristics through the noise reduction self-coding model to obtain fault characteristics in the related attribute characteristics;
and the fault classification module is used for classifying and processing the fault characteristics by adopting the characteristic classification model to obtain the fault type of the electric submersible pump corresponding to the fault characteristics.
9. A computer terminal comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor when executing the program implements the DAE based electrical submersible pump fault diagnosis method according to any one of claims 1-7.
10. A computer-readable medium having a computer program stored thereon, the computer program being executable by a processor to implement the method of fault diagnosis for a DAE-based electric submersible pump as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210260063.5A CN114708675A (en) | 2022-03-16 | 2022-03-16 | DAE-based electric submersible pump fault diagnosis method, system, terminal and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210260063.5A CN114708675A (en) | 2022-03-16 | 2022-03-16 | DAE-based electric submersible pump fault diagnosis method, system, terminal and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114708675A true CN114708675A (en) | 2022-07-05 |
Family
ID=82168232
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210260063.5A Pending CN114708675A (en) | 2022-03-16 | 2022-03-16 | DAE-based electric submersible pump fault diagnosis method, system, terminal and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114708675A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115614292A (en) * | 2022-11-02 | 2023-01-17 | 昆明理工大学 | Vibration monitoring device and method for vertical water pump unit |
| CN117539155A (en) * | 2024-01-09 | 2024-02-09 | 深圳市威诺达工业技术有限公司 | Optimized control method of electric submersible pump |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180094512A1 (en) * | 2016-09-30 | 2018-04-05 | General Electric Company | Systems and methods for optimizing an efficiency of a variable frequency drive |
| US20180223830A1 (en) * | 2017-02-03 | 2018-08-09 | Zilift Holdings, Limited | Inline monitoring package for an electric submersible pump system |
| US20180375456A1 (en) * | 2017-06-27 | 2018-12-27 | General Electric Company | Systems and methods for active damping of a motor |
| US20210335020A1 (en) * | 2020-04-23 | 2021-10-28 | Saudi Arabian Oil Company | Generation and Implementation of Multi-Layer Encoded Data Images |
-
2022
- 2022-03-16 CN CN202210260063.5A patent/CN114708675A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180094512A1 (en) * | 2016-09-30 | 2018-04-05 | General Electric Company | Systems and methods for optimizing an efficiency of a variable frequency drive |
| US20180223830A1 (en) * | 2017-02-03 | 2018-08-09 | Zilift Holdings, Limited | Inline monitoring package for an electric submersible pump system |
| US20180375456A1 (en) * | 2017-06-27 | 2018-12-27 | General Electric Company | Systems and methods for active damping of a motor |
| US20210335020A1 (en) * | 2020-04-23 | 2021-10-28 | Saudi Arabian Oil Company | Generation and Implementation of Multi-Layer Encoded Data Images |
Non-Patent Citations (1)
| Title |
|---|
| 杨佳丽: "海上油气水井电潜泵的数据分析及故障诊断研究", 工程科技Ⅰ辑, no. 2022, pages 1 - 71 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115614292A (en) * | 2022-11-02 | 2023-01-17 | 昆明理工大学 | Vibration monitoring device and method for vertical water pump unit |
| CN115614292B (en) * | 2022-11-02 | 2023-10-27 | 昆明理工大学 | Vibration monitoring device and method for vertical water pump unit |
| CN117539155A (en) * | 2024-01-09 | 2024-02-09 | 深圳市威诺达工业技术有限公司 | Optimized control method of electric submersible pump |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114708675A (en) | DAE-based electric submersible pump fault diagnosis method, system, terminal and medium | |
| CN104170244B (en) | Submersible electric pump monitors and failure predication | |
| CN109738939B (en) | Earthquake precursor data anomaly detection method | |
| CN111914883B (en) | Spindle bearing state evaluation method and device based on deep fusion network | |
| CN108647786B (en) | Rotary machine on-line fault monitoring method based on deep convolution countermeasure neural network | |
| CN112414694B (en) | Equipment multistage abnormal state identification method and device based on multivariate state estimation technology | |
| CN112179691B (en) | Mechanical equipment running state abnormity detection system and method based on counterstudy strategy | |
| CN111861272A (en) | Multi-source data-based complex electromechanical system abnormal state detection method | |
| CN110738255A (en) | device state monitoring method based on clustering algorithm | |
| CN113177537B (en) | Fault diagnosis method and system for rotary mechanical equipment | |
| CN112305388B (en) | On-line monitoring and diagnosing method for insulation partial discharge faults of generator stator winding | |
| KR20210017651A (en) | Method for Fault Detection and Fault Diagnosis in Semiconductor Manufacturing Process | |
| CN111881594B (en) | Non-stationary signal state monitoring method and system for nuclear power equipment | |
| CN114021620B (en) | BP neural network feature extraction-based electric submersible pump fault diagnosis method | |
| CN116578940A (en) | Bearing fault new type identification and diagnosis method based on mixed depth self-coding | |
| CN113721000B (en) | Method and system for detecting abnormity of dissolved gas in transformer oil | |
| Yang et al. | A fault identification method for electric submersible pumps based on dae-svm | |
| CN109255201A (en) | A kind of ball screw assembly, health evaluating method based on SOM-MQE | |
| CN113670611A (en) | Bearing early degradation evaluation method, system, medium and electronic equipment | |
| CN108386182B (en) | Oil well production abnormity early warning method based on SPC control chart and weighted decision tree | |
| CN114112390B (en) | Nonlinear complex system early fault diagnosis method | |
| CN115392109A (en) | LSTM multivariable time series anomaly detection method based on generative model | |
| CN114580472A (en) | Large-scale equipment fault prediction method with repeated cause and effect and attention in industrial internet | |
| CN114997485A (en) | Overflow condition prediction model training method and device and overflow condition prediction method | |
| KR20230075150A (en) | Method for managing system health |
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 |