[1]LI Kewen,DU Congcong,HUANG Zongchao,et al.Early warning model for abnormal workingconditions of CBiA-PSL pumping wells[J].CAAI Transactions on Intelligent Systems,2022,17(2):295-302.[doi:10.11992/tis.202106007]
Copy
Early warning model for abnormal workingconditions of CBiA-PSL pumping wells
CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume:
17
Number of periods:
2022 2
Page number:
295-302
Column:
学术论文—机器感知与模式识别
Public date:
2022-03-05
- Title:
- Early warning model for abnormal workingconditions of CBiA-PSL pumping wells
- Keywords:
- convolutional neural network; bidirectional gated recurrent unit; attention; positive sharing loss; loss function; identification of abnormal working conditions; working condition diagnosis and early warning; data imbalance
- CLC:
- TP18
- DOI:
- 10.11992/tis.202106007
- Abstract:
- In the process of oilfield production, which is affected by various factors, oil wells are prone to abnormal working conditions, such as pump and pipe leakages, which will reduce the output of oil wells and even lead to lying in wells. Early warning of abnormal working conditions is an important task of intelligent oilfield management. Based on the convolutional neural network–bidirectional gated recurrent unit (CNN–BiGRU) joint network, an improved network structure CNN–BiGRU attention-positive sharing loss (CBiA-PSL) model is proposed for early warning of abnormal oil well conditions. The model uses the CNN to learn the depth features of the gray image of the sample, BiGRU to effectively prevent information loss and strengthen the connection between the features of the CNN pool layer, and attention mechanism to weigh the hidden state to complete effective feature screening. To address the imbalance of the working condition dataset, a PSL function, which divides the abnormal data (positive class) into subclasses, is proposed. Each subclass shares the loss of the entire positive class and provides a high weight to the positive class with only a few samples. The experimental results show that the CBiA-PSL model has a better prediction effect and a higher accuracy for anomaly and overall prediction than other models.
- References:
-
[1] ZHANG Bin, GAO Xianwen, LI Xiangyu. Complete simulation and fault diagnosis of sucker-rod pumping (includes associated comment)[J]. SPE production & operations, 2021, 36(2): 277–290.
[2] DUAN Jimiao, LIU Huishu, JIANG Junzhe, et al. Numerical prediction of wax deposition in oil-gas stratified pipe flow[J]. International journal of heat and mass transfer, 2017, 105: 279–289.
[3] ALSAIHATI A, ELKATATNY S, MAHMOUD A A, et al. Use of machine learning and data analytics to detect downhole abnormalities while drilling horizontal wells, with real case study[J]. Journal of energy resources technology, 2021, 143(4): 043201.
[4] ?ZMEN ?, SINANO?LU C, CALISKAN A, et al. Prediction of leakage from an axial piston pump slipper with circular dimples using deep neural networks[J]. Chinese journal of mechanical engineering, 2020, 33(1): 28.
[5] RAHMANIFARD H, PLAKSINA T. Application of artificial intelligence techniques in the petroleum industry: a review[J]. Artificial intelligence review, 2019, 52(4): 2295–2318.
[6] ZHAO Minghang, KANG M, TANG Baoping, et al. Deep residual networks with dynamically weighted wavelet coefficients for fault diagnosis of planetary gearboxes[J]. IEEE transactions on industrial electronics, 2018, 65(5): 4290–4300.
[7] ZHANG Yiyi, LI Jiaxi, FAN Xianhao, et al. Moisture prediction of transformer oil-immersed polymer insulation by applying a support vector machine combined with a genetic algorithm[J]. Polymers, 2020, 12(7): 1579.
[8] 周斌, 王延江, 刘伟锋, 等. 基于Hessian正则化支持向量机的多视角协同识别抽油机井工况方法[J]. 石油学报, 2018, 39(12): 1429–1436
ZHOU Bin, WANG Yanjiang, LIU Weifeng, et al. A working condition recognition method of sucker-rod pumping wells based on Hessian-regularized SVM and multi-view co-training algorithm[J]. Acta petrolei sinica, 2018, 39(12): 1429–1436
[9] 王利君, 支志英, 贾鹿, 等. 基于SCRF的抽油井结蜡预测方法优化研究[J]. 计算机科学, 2019, 46(S2): 599–603
WANG Lijun, ZHI Zhiying, JIA Lu, et al. Study on optimized method for predicting paraffin deposition of pumping wells based on SCRF[J]. Computer science, 2019, 46(S2): 599–603
[10] LEI Yaguo, YANG Bin, JIANG Xinwei, et al. Applications of machine learning to machine fault diagnosis: a review and roadmap[J]. Mechanical systems and signal processing, 2020, 138: 106587.
[11] WEI Jingliang, GAO Xianwen. Fault diagnosis of sucker rod pump based on deep-broad learning using motor data[J]. IEEE access, 2020, 8: 222562–222571.
[12] CHEN Xiaohan, ZHANG Beike, GAO Dong. Bearing fault diagnosis base on multi-scale CNN and LSTM model[J]. Journal of intelligent manufacturing, 2021, 32(4): 971–987.
[13] CABRERA D, GUAMáN A, ZHANG Shaohui, et al. Bayesian approach and time series dimensionality reduction to LSTM-based model-building for fault diagnosis of a reciprocating compressor[J]. Neurocomputing, 2020, 380: 51–66.
[14] 魏晓良, 潮群, 陶建峰, 等. 基于LSTM和CNN的高速柱塞泵故障诊断[J]. 航空学报, 2021, 42(3): 423876
WEI Xiaoliang, CHAO Qun, TAO Jianfeng, et al. Cavitation fault diagnosis method for high-speed plunger pumps based on LSTM and CNN[J]. Acta aeronautica et astronautica sinica, 2021, 42(3): 423876
[15] LIANG Rong, CHANG Xintan, JIA Pengtao, et al. Mine gas concentration forecasting model based on an optimized BiGRU network[J]. ACS omega, 2020, 5(44): 28579–28586.
[16] SHEN Wei, WANG Xinggang, WANG Yan, et al. DeepContour: a deep convolutional feature learned by positive-sharing loss for contour detection[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, USA, 2015: 3982-3991.
[17] CHEN Yushi, JIANG Hanlu, LI Chunyang, et al. Deep feature extraction and classification of hyperspectral images based on convolutional neural networks[J]. IEEE transactions on geoscience and remote sensing, 2016, 54(10): 6232–6251.
[18] LU Wenjie, LI Jiazheng, WANG Jingyang, et al. A CNN-BiLSTM-AM method for stock price prediction[J]. Neural computing and applications, 2021, 33(10): 4741–4753.
[19] DIAZ M, MOETESUM M, SIDDIQI I, et al. Sequence-based dynamic handwriting analysis for Parkinson’s disease detection with one-dimensional convolutions and BiGRUs[J]. Expert systems with applications, 2021, 168: 114405.
[20] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[EB/OL]. (2017-05-12)[2021-05-01].https://arxiv.org/abs/1706.03762.
[21] JAGVARAL B, LEE W K, ROH J S, et al. Path-based reasoning approach for knowledge graph completion using CNN-BiLSTM with attention mechanism[J]. Expert systems with applications, 2020, 142: 112960.
- Similar References:
Memo
-
Last Update:
1900-01-01