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
To assess whether a development strategy will be profitable enough, production forecasting is a crucial and difficult step in the process. The development history of other reservoirs in the same class tends to be studied to make predictions accurate. However, the permeability field, well patterns, and development regime must all be similar for two reservoirs to be considered in the same class. This results in very few available experiences from other reservoirs even though there is a lot of historical information on numerous reservoirs because it is difficult to find such similar reservoirs. This paper proposes a learn-to-learn method, which can better utilize a vast amount of historical data from various reservoirs. Intuitively, the proposed method first learns how to learn samples before directly learning rules in samples. Technically, by utilizing gradients from networks with independent parameters and copied structure in each class of reservoirs, the proposed network obtains the optimal shared initial parameters which are regarded as transferable information across different classes. Based on that, the network is able to predict future production indices for the target reservoir by only training with very limited samples collected from reservoirs in the same class. Two cases further demonstrate its superiority in accuracy to other widely-used network methods.
Caena, F., Stringher, C., 2022. Towards a new conceptualization of Learning to Learn. J. Aula Abierta 49 (3), 199–216. https://doi.org/10.17811/rifie.49.3.2020.199-216.
Dong, Y., Zhang, Y., Liu, F., et al., 2021. Reservoir production prediction model based on a stacked LSTM network and transfer learning. J. ACS omega 6 (50), 34700–34711. https://doi.org/10.1021/acsomega.1c05132.
Douglas, J., Furtado, F., Pereira, F., 1997. On the numerical simulation of waterflooding of heterogeneous petroleum reservoirs. J. Comput. Geosci. 1 (2), 155. https://doi.org/10.1023/A:1011565228179.
Du, E., Liu, Y., Cheng, Z., et al., 2022. Production forecasting with the interwell interference by integrating graph convolutional and long short-term memory neural network. J. SPE Reservoir Evaluat. Eng. 25 (2), 197–213. https://doi.org/10.2118/208596-PA.
Gupta, I., Samandarli, O., Burks, A., et al., 2021. Autoregressive and machine learning driven production forecasting-Midland Basin case study. In: SPE/AAPG/SEG Unconventional Resources Technology Conference. OnePetro. https://doi.org/10.15530/urtec-2021-5184.
Iwata, T., Kumagai, A., 2020. Meta-learning from tasks with heterogeneous attribute spaces. J. Advances in Neural Information Processing Systems 33, 6053–6063.
Jolliffe, I.T., Cadima, J.A.M., 2016. Principal component analysis: a review and recent developments. J. Phys. Sci. E. 374 (2065), 20150202. https://doi.org/10.1098/rsta.2015.0202.
Lemke, C., Gabrys, B.J.N., 2010. Meta-learning for time series forecasting and forecast combination. J. Neurocomput. 73 (10-12), 2006–2016. https://doi.org/10.1016/j.neucom.2009.09.020.
Li, Y., Guan, C., Guan, C., et al., 2019. An optimized neural network prediction model for gas assisted gravity drainage recovery based on dimensional analysis. Petroleum Science Bulletin 03, 288–299.
Li, M., Tang, Y., Ma, W., 2022. Few-shot traffic prediction with graph networks using locale as relational inductive biases. J. IEEE Transact. Intelligent Transport. Syst. https://doi.org/10.1109/TITS.2022.3219618.
Lu, L., Xiong, S., 2022. Few-shot driver identification via meta-learning. J. Expert Syst. Appl. 203, 117299. https://doi.org/10.1016/j.eswa.2022.117299.
Ma, Y., Zhao, S., Wang, W., et al., 2022. Multimodality in meta-learning: a comprehensive survey. J. Knowledge-Based Systems, 108976. https://doi.org/10.1016/j.knosys.2022.108976.
Mohd, R., Cornelio, J., Cho, Y., et al., 2022. Transfer learning with recurrent neural networks for long-term production forecasting in unconventional reservoirs. J. SPE 27 (4), 2425–2442. https://doi.org/10.2118/209594-PA.
Oreshkin, B.N., Dudek, G., Pełka, P., et al., 2021. N-BEATS neural network for mid-term electricity load forecasting. J. Appl. Energy 293, 116918. https://doi.org/10.1016/j.apenergy.2021.116918.
Song, S., Hong, B., Shi, B., et al., 2017. Research into calculation of natural gas well production based on an artificial neural network. Petroleum Science Bulletin 2 (3), 413–421.
Wang, Q., Yuan, C., Liu, Y., 2019. Learning deep conditional neural network for image segmentation. J. IEEE Transact. Multimedia 21 (7), 1839–1852. https://doi.org/10.1109/TMM.2018.2890360.
Yang, N., Zhang, B., Ding, G., et al., 2021. Specific emitter identification with limited samples: a model-agnostic meta-learning approach. J. IEEE Commun. Lett. 26 (2), 345–349. https://doi.org/10.1109/LCOMM.2021.3110775.
Zhong, Z., Wu, X., 2017. Pressure and saturation relationship in CBM reservoir and the semi-analytical production forecast model for the horizontal well. Petroleum Science Bulletin 2 (2), 251–257.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
京公网安备11010802044758号