Print Email Facebook Twitter Machine Learning-Based Surrogate Modeling for Urban Water Networks Title Machine Learning-Based Surrogate Modeling for Urban Water Networks: Review and Future Research Directions Author Garzón Díaz, J.A. (TU Delft Sanitary Engineering) Kapelan, Z. (TU Delft Sanitary Engineering) Langeveld, J.G. (TU Delft Sanitary Engineering; Partners4UrbanWater) Taormina, R. (TU Delft Sanitary Engineering) Date 2022 Abstract Surrogate models replace computationally expensive simulations of physically-based models to obtain accurate results at a fraction of the time. These surrogate models, also known as metamodels, have been employed for analysis, control, and optimization of water distribution and urban drainage systems. With the advent of machine learning (ML), water engineers have increasingly resorted to these data-driven techniques to develop metamodels of urban water networks (UWNs). In this article, we review 31 recent articles on ML-based metamodeling of UWNs to outline the state-of-the-art of the field, identify outstanding gaps, and propose future research directions. For each article, we critically examined the purpose of the metamodel, the metamodel characteristics, and the applied case study. The review shows that current metamodels suffer several drawbacks, including (a) the curse of dimensionality, hindering implementation for large case studies; (b) black-box deterministic nature, limiting explainability and applicability; and (c) rigid architecture, preventing generalization across multiple case studies. We argue that researchers should tackle these issues by resorting to recent advancements in ML concerning inductive biases, robustness, and transferability. Recently developed neural network architectures, which extend deep learning methods to graph data structures, are preferred candidates for advancing surrogate modeling in UWNs. Furthermore, we foresee increasing efforts for complex applications where metamodels may play a fundamental role, such as uncertainty analysis and multi-objective optimization. Lastly, the development and comparison of ML-based metamodels can benefit from the availability of new benchmark datasets for urban drainage systems and realistic complex networks. Subject artificial neural networksmachine learningsurrogate modelingurban drainage systemswater distribution systemswater networks To reference this document use: http://resolver.tudelft.nl/uuid:e6310127-02c8-410d-bb2d-ae5a04bed9b7 DOI https://doi.org/10.1029/2021WR031808 ISSN 0043-1397 Source Water Resources Research, 58 (5) Part of collection Institutional Repository Document type review Rights © 2022 J.A. Garzón Díaz, Z. Kapelan, J.G. Langeveld, R. Taormina Files PDF Water_Resources_Research_ ... ew_and.pdf 746.83 KB Close viewer /islandora/object/uuid:e6310127-02c8-410d-bb2d-ae5a04bed9b7/datastream/OBJ/view