AI Chat Paper
Discover the SciOpen Platform and Achieve Your Research Goals with Ease.
Search articles, authors, keywords, DOl and etc.
{{expandStatus?'Exit ':''}}Advanced Search
In the production of the sucker rod well, the dynamic liquid level is important for the production efficiency and safety in the lifting process. It is influenced by multi-source data which need to be combined for the dynamic liquid level real-time calculation. In this paper, the multi-source data are regarded as the different views including the load of the sucker rod and liquid in the wellbore, the image of the dynamometer card and production dynamics parameters. These views can be fused by the multi-branch neural network with special fusion layer. With this method, the features of different views can be extracted by considering the difference of the modality and physical meaning between them. Then, the extraction results which are selected by multinomial sampling can be the input of the fusion layer. During the fusion process, the availability under different views determines whether the views are fused in the fusion layer or not. In this way, not only the correlation between the views can be considered, but also the missing data can be processed automatically. The results have shown that the load and production features fusion (the method proposed in this paper) performs best with the lowest mean absolute error (MAE) 39.63 m, followed by the features concatenation with MAE 42.47 m. They both performed better than only a single view and the lower MAE of the features fusion indicates that its generalization ability is stronger. In contrast, the image feature as a single view contributes little to the accuracy improvement after fused with other views with the highest MAE. When there is data missing in some view, compared with the features concatenation, the multi-view features fusion will not result in the unavailability of a large number of samples. When the missing rate is 10%, 30%, 50% and 80%, the method proposed in this paper can reduce MAE by 5.8, 7, 9.3 and 20.3 m respectively. In general, the multi-view features fusion method proposed in this paper can improve the accuracy obviously and process the missing data effectively, which helps provide technical support for real-time monitoring of the dynamic liquid level in oil fields.
Baltrusaitis, T., Ahuja, C., Morency, L.P., 2018. Multimodal machine learning: a survey and taxonomy. IEEE. T. Pattern. Anal. 41 (2), 423-443. https://doi.org/10.1109/TPAMI.2018.2798607.
Chavarriaga, R., Sagha, H., Calatroni, A., et al., 2013. The opportunity challenge: a benchmark database for on-body sensor-based activity recognition. Pattern Recogn. Lett. 34 (15), 2033-2042.
Chen, D.C., Zhang, R.C., Meng, H.X., et al., 2015. The study and application of dynamic liquid level calculation model based on dynamometer card of oil wells. Sci. Technol. Eng. 15 (32), 32-35. https://doi.org/10.3969/j.issn.1671-1815.2015.32.006 (in Chinese).
Chen, D.F., Han, X.L., Yang, J., 2008. Discuss on survey method for liquid level of oil well. Well Test. 17 (2), 60-61. https://doi.org/10.3969/j.issn.1004-4388.2008.02.024 (in Chinese).
Choi, J.H., Lee, J.S., 2019. EmbraceNet: a robust deep learning architecture for multimodal classification. Inf. Fusion 51, 259-270. https://doi.org/10.1016/j.inffus.2019.02.010.
Gibbs, S.G., 1963. Predicting the behavior of sucker-rod pumping systems. JPT 15 (7), 769-778. https://doi.org/10.2118/588-PA.
Han, Y., Song, X.P., Li, K., et al., 2022. Hybrid modeling for submergence depth of the pumping well using stochastic configuration networks with random sampling. J. Pet. Sci. Eng. 208, 109423. https://doi.org/10.1016/j.petrol.2021.109423.
Hou, J.L., Jiang, H.C., Wan, C.F., et al., 2022. Deep learning and data augmentation based data imputation for structural health monitoring system in multi-sensor damaged state. Measurement 196, 111206. https://doi.org/10.1016/j.measurement.2022.111206.
Hou, Y.B., Gao, X.W., Li, X.Y., 2019. Prediction for dynamic liquid level of sucker rod pumping using generation of multi-scale state characteristics in oil field production. CIESC J. 70 (S2), 311-321. https://doi.org/10.11949/0438-1157.20190352 (in Chinese).
LeCun, Y., Bottou, L., Bengio, Y., et al., 1998. Gradient-based learning applied to document recognition. Proc. IEEE 86 (11), 2278-2324.
Li, X.Y., Gao, X.W., Hou, Y.B., 2015. Online dynamic Gaussian process regression for dynamic liquid level soft sensing of sucker-rod pumping well. CIESC J. 66 (6), 2150-2158. https://doi.org/10.11949/j.issn.0438-1157.20141791 (in Chinese).
Li, X.Y., Gao, X.W., Li, K., et al., 2016. Ensemble soft sensor modeling for dynamic liquid level of oil well based on multi-source information feature fusion. CIESC J. 67 (6), 2469-2479. https://doi.org/10.11949/j.issn.0438-1157.20151673 (in Chinese).
Li, Z.C., Tian, L., Jiang, Q.C., et al., 2020b. Fault diagnostic method based on deep learning and multimodel feature fusion for complex industrial processes. Ind. Eng. Chem. Res. 59 (40), 18061-18069. https://doi.org/10.1021/acs.iecr.0c03082.
Li, S., Wan, H.Q., Song, L.Y., 2020a. An adaptive data fusion strategy for fault diagnosis based on the convolutional neural network. Measurement 165, 108122. https://doi.org/10.1016/j.measurement.2020.108122.
Liang, X., Zhang, Z.H., 2021. Research on depth of oil well moving liquid surface based on short-term energy and LSTM. Comput. Mod. 308, 15-19+26. https://doi.org/10.3969/j.issn.1006-2475.2021.04.003 (in Chinese).
Liguori, A., Markovic, R., Ferrando, M., et al., 2023. Augmenting energy time-series for data-efficient imputation of missing values. Appl. Energy 334, 120701. https://doi.org/10.1016/j.apenergy.2023.120701.
Mohammadpoor, M., Torabi, F., 2018. Big data analytics in oil and gas industry: An emerging trend. Petroleum 7 (2), 241-242. https://doi.org/10.1016/j.petlm.2018.11.001.
Nguyen, T., Gosine, R.G., Warrian, P., 2020. A systematic review of big data analytics for oil and gas industry 4.0. IEEE Access 8, 61183-61201. https://doi.org/10.1109/ACCESS.2020.2979678.
Ordóñez, F.J., Roggen, D., 2016. Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors 16 (1), 1-25. https://doi.org/10.3390/s16010115.
Tian, Z.D., 2021. Kernel principal component analysis-based least squares support vector machine optimized by improved grey wolf optimization algorithm and application in dynamic liquid level forecasting of beam pump. Trans. Inst. Meas. Control 42 (6), 1135-1150. https://doi.org/10.1177/0142331219885273.
Vergara, A., Fonollosa, J., Mahiques, J., et al., 2013. On the performance of gas sensor arrays in open sampling systems using Inhibitory Support Vector Machines. Sensor. Actuator. B Chem. 185, 462-477.
Wang, T., Duan, Z.W., Li, K., 2017. Adaptive ensemble modeling for dynamic liquid level of oil well based on improved AdaBoost method. J. Electron. Meas. Instrum. 31 (8), 1342-1348. https://doi.org/10.13382/j.jemi.2017.08.025 (in Chinese).
Xu, X.W., Tao, Z.R., Ming, W.W., 2020. Intelligent monitoring and diagnostics using a novel integrated model based on deep learning and multi-sensor feature fusion. Measurement 165, 108086. https://doi.org/10.1016/j.measurement.2020.108086.
Yang, L.P., 2010. The dynamic liquid level calculation of the sucker rod well by dynamometer cards. Petrol. Geol. Eng. 24 (5), 101-103. https://doi.org/10.3969/j.issn.1000-7393.2011.06.030 (in Chinese).
Yu, D.L., Qi, W.G., Ding, B., et al., 2018. Study of forecasting producing fluid level of submersible reciprocating pump on the basis of chaotic time series. Comput. Sci. Appl. 8 (6), 1034-1044. https://doi.org/10.12677/CSA.2018.86115 (in Chinese).
Zhang, H., 2003. Discussion of detecting fluid level by pressure gauge. Well Test. 12 (5), 49-50. https://doi.org/10.3969/j.issn.1004-4388.2003.05.018 (in Chinese).
Zhang, H.L., Li, P., Xie, Q., et al., 2007. Preliminary study and application of the method for calculating dynamic liquid level by dynamometer cards. Qinghai Shiyou 25 (2), 31-35 (in Chinese).
Zhang, Q., Wu, X.D., 1984. Pumping well diagnostic technique and its application. Journal of east China Petroleum Institute 2, 145-159 (in Chinese).
Zhang, S.L., Luo, Y., Wu, Z.M., et al., 2011. Corrected algorithm for calculating dynamic fluid level with indicator diagram for rob-pumped well. Oil Drill. Prod. Technol. 33 (6), 122-124. https://doi.org/10.3969/j.issn.1000-7393.2011.06.030 (in Chinese).
Zhou, W., Liu, J., Gan, L.Q., 2018. Dynamic liquid level detection method based on resonant frequency difference for oil wells. Turk. J. Electr. Eng. Co. 26 (6), 2968-2976. https://doi.org/10.3906/elk-1805-68.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
京公网安备11010802044758号