{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/deep-convolutional-encoder-decoder-networks","title":"Deep convolutional encoder-decoder networks for uncertainty quantification of dynamic multiphase flow in heterogeneous media","arxiv_id":"1807.00882","date":"2018-07-02","proceeding":null,"authors":["Shaoxing Mo","Yinhao Zhu","Nicholas Zabaras","Xiaoqing Shi","Jichun Wu"],"abstract":"Surrogate strategies are used widely for uncertainty quantification of\ngroundwater models in order to improve computational efficiency. However, their\napplication to dynamic multiphase flow problems is hindered by the curse of\ndimensionality, the saturation discontinuity due to capillarity effects, and\nthe time-dependence of the multi-output responses. In this paper, we propose a\ndeep convolutional encoder-decoder neural network methodology to tackle these\nissues. The surrogate modeling task is transformed to an image-to-image\nregression strategy. This approach extracts high-level coarse features from the\nhigh-dimensional input permeability images using an encoder, and then refines\nthe coarse features to provide the output pressure/saturation images through a\ndecoder. A training strategy combining a regression loss and a segmentation\nloss is proposed in order to better approximate the discontinuous saturation\nfield. To characterize the high-dimensional time-dependent outputs of the\ndynamic system, time is treated as an additional input to the network that is\ntrained using pairs of input realizations and of the corresponding system\noutputs at a limited number of time instances. The proposed method is evaluated\nusing a geological carbon storage process-based multiphase flow model with a\n2500-dimensional stochastic permeability field. With a relatively small number\nof training data, the surrogate model is capable of accurately characterizing\nthe spatio-temporal evolution of the pressure and discontinuous CO2 saturation\nfields and can be used efficiently to compute the statistics of the system\nresponses.","url_abs":"http://arxiv.org/abs/1807.00882v1","url_pdf":"http://arxiv.org/pdf/1807.00882v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"deep-convolutional-encoder-decoder-networks","repo_url":"https://github.com/cics-nd/dcedn-gcs","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"image-to-image-regression","task_name":"Image-to-Image Regression"},{"task_slug":"uncertainty-quantification","task_name":"Uncertainty Quantification"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.00882","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}