Unlock instant, AI-driven research and patent intelligence for your innovation.

A dynamometer fault diagnosis method based on generative adversarial neural network

A neural network and dynamometer technology, applied in biological neural network models, neural learning methods, neural architectures, etc., can solve the problem of high fault false alarm/missing rate, improve specific recognition ability, reduce false alarms/missing The effect of reporting

Active Publication Date: 2022-07-22
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In recent years, based on the development of artificial intelligence technology and the accumulation of a large amount of on-site data, relevant scholars have begun to apply machine learning and deep learning technology to the diagnosis of dynamometer diagrams to realize automatic diagnosis. The reporting / missing rate is still high

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A dynamometer fault diagnosis method based on generative adversarial neural network
  • A dynamometer fault diagnosis method based on generative adversarial neural network
  • A dynamometer fault diagnosis method based on generative adversarial neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0171] This example is based on the field data of an oil field, with a total of 14,628 pieces of dynamometer working condition data.

[0172] First, according to the distribution of the number of samples, the adversarial neural network GAN is used to generate samples for faults such as continuous pumping and spraying, pump leakage, and other explanations, and 200 new samples are generated each. This embodiment does not involve optimization of hyperparameters, so the validation set data does not interfere with the model, so there is no need to divide the test set separately, and only the training set and the validation set need to be divided. Among them: 80% of the samples are used as the training set, and 20% of the samples are used as the validation set.

[0173] Then, establish an Xgboost classifier model according to the above step 5, and perform fault diagnosis on the sample.

[0174] Finally, the accuracy rate and recall rate of the validation set are calculated, and the...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a dynamometer diagram fault diagnosis method based on a generative confrontation neural network, belonging to the technical field of oil production fault diagnosis, comprising the following steps: data cleaning of dynamometer diagram sample database data; Figure characteristics, extract features from the dynamometer data points; use generative adversarial neural network to generate a small number of fault category samples, and conditionally constrain the output of the generator network during the generation process; based on the original samples and generated samples, Divide the data into training set, validation set, and test set; use the Xgboost classification algorithm to classify the samples; use the accuracy rate and recall rate to comprehensively evaluate the fault diagnosis results; use the classification model after training to conduct real-time monitoring and diagnosis of faults, Real-time judgment of fault types. The invention can significantly improve the specific identification ability of the classification model to the fault samples, and reduce the false alarm / missing alarm rate of the fault.

Description

technical field [0001] The invention belongs to the technical field of oil production fault diagnosis, and in particular relates to a dynamometer fault diagnosis method based on a generative confrontation neural network. Background technique [0002] The failure analysis of rod pump oil production usually relies on the dynamometer diagram as the judgment basis. The traditional diagnosis method is that technicians establish typical dynamometer diagrams under different faults based on oil production engineering knowledge, and judge the current oil well fault by comparing the actual dynamometer diagram with the typical dynamometer diagram. [0003] In recent years, based on the development of artificial intelligence technology and the accumulation of a large amount of field data, relevant scholars have begun to apply machine learning and deep learning technology to dynamometer diagram diagnosis to realize automatic diagnosis. The underreport / miss rate is still high. SUMMARY ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/2415G06F18/214
Inventor 张凯尹承哲曹晨张黎明张华清严侠刘丕养杨勇飞孙海姚军樊灵
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)