{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T15:18:22Z","timestamp":1772205502439,"version":"3.50.1"},"reference-count":65,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,1,17]],"date-time":"2025-01-17T00:00:00Z","timestamp":1737072000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Big Data"],"abstract":"<jats:p>Predicting the dynamics of chaotic systems is crucial across various practical domains, including the control of infectious diseases and responses to extreme weather events. Such predictions provide quantitative insights into the future behaviors of these complex systems, thereby guiding the decision-making and planning within the respective fields. Recently, data-driven approaches, renowned for their capacity to learn from empirical data, have been widely used to predict chaotic system dynamics. However, these methods rely solely on historical observations while ignoring the underlying mechanisms that govern the systems' behaviors. Consequently, they may perform well in short-term predictions by effectively fitting the data, but their ability to make accurate long-term predictions is limited. A critical challenge in modeling chaotic systems lies in their sensitivity to initial conditions; even a slight variation can lead to significant divergence in actual and predicted trajectories over a finite number of time steps. In this paper, we propose a novel Physics-Guided Learning (PGL) method, aiming at extending the scope of accurate forecasting as much as possible. The proposed method aims to synergize observational data with the governing physical laws of chaotic systems to predict the systems' future dynamics. Specifically, our method consists of three key elements: a data-driven component (DDC) that captures dynamic patterns and mapping functions from historical data; a physics-guided component (PGC) that leverages the governing principles of the system to inform and constrain the learning process; and a nonlinear learning component (NLC) that effectively synthesizes the outputs of both the data-driven and physics-guided components. Empirical validation on six dynamical systems, each exhibiting unique chaotic behaviors, demonstrates that PGL achieves lower prediction errors than existing benchmark predictive models. The results highlight the efficacy of our design of data-physics integration in improving the precision of chaotic system dynamics forecasts.<\/jats:p>","DOI":"10.3389\/fdata.2024.1506443","type":"journal-article","created":{"date-parts":[[2025,1,17]],"date-time":"2025-01-17T06:51:43Z","timestamp":1737096703000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Toward a physics-guided machine learning approach for predicting chaotic systems dynamics"],"prefix":"10.3389","volume":"7","author":[{"given":"Liu","family":"Feng","sequence":"first","affiliation":[]},{"given":"Yang","family":"Liu","sequence":"additional","affiliation":[]},{"given":"Benyun","family":"Shi","sequence":"additional","affiliation":[]},{"given":"Jiming","family":"Liu","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,1,17]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1051\/mmnp:2008070","article-title":"Epidemiology of dengue fever: a model with temporary cross-immunity and possible secondary infection shows bifurcations and chaotic behaviour in wide parameter regions","volume":"3","author":"Aguiar","year":"2008","journal-title":"Math. Model. Nat. Phenom"},{"key":"B2","doi-asserted-by":"publisher","first-page":"982","DOI":"10.1143\/PTP.67.982","article-title":"Topological aspects in chaos and in 2-k period doubling cascade","volume":"67","author":"Aizawa","year":"1982","journal-title":"Prog. Theoret. Phys"},{"key":"B3","doi-asserted-by":"publisher","first-page":"105609","DOI":"10.1016\/j.envsoft.2022.105609","article-title":"B-ama: A python-coded protocol to enhance the application of data-driven models in hydrology","volume":"160","author":"Amaranto","year":"2023","journal-title":"Environm. Model. Softw"},{"key":"B4","doi-asserted-by":"publisher","first-page":"111944","DOI":"10.1016\/j.jcp.2023.111944","article-title":"Adlgm: an efficient adaptive sampling deep learning galerkin method","volume":"477","author":"Aristotelous","year":"2023","journal-title":"J. Comput. Phys"},{"key":"B5","doi-asserted-by":"publisher","first-page":"20200083","DOI":"10.1098\/rsta.2020.0083","article-title":"Opportunities and challenges for machine learning in weather and climate modelling: hard, medium and soft ai","volume":"379","author":"Chantry","year":"2021","journal-title":"Philosoph. Trans. Royal Soc. A"},{"key":"B6","doi-asserted-by":"publisher","first-page":"373","DOI":"10.5194\/npg-27-373-2020","article-title":"Data-driven predictions of a multiscale lorenz 96 chaotic system using machine-learning methods: reservoir computing, artificial neural network, and long short-term memory network","volume":"27","author":"Chattopadhyay","year":"2020","journal-title":"Nonlinear Process. Geophys"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1465","DOI":"10.1142\/S0218127499001024","article-title":"Yet another chaotic attractor","volume":"9","author":"Chen","year":"1999","journal-title":"Int. J. Bifurcat. Chaos"},{"key":"B8","doi-asserted-by":"publisher","first-page":"111304","DOI":"10.1016\/j.chaos.2021.111304","article-title":"High-efficiency chaotic time series prediction based on time convolution neural network","volume":"152","author":"Cheng","year":"2021","journal-title":"Chaos, Solit. Fractals"},{"key":"B9","doi-asserted-by":"publisher","first-page":"507","DOI":"10.1109\/TCT.1971.1083337","article-title":"Memristor-the missing circuit element","volume":"18","author":"Chua","year":"1971","journal-title":"IEEE Trans. Circuit Theory"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1072","DOI":"10.1109\/TCS.1986.1085869","article-title":"The double scroll family","volume":"33","author":"Chua","year":"1986","journal-title":"IEEE Trans. Circuits Syst"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1488","DOI":"10.4249\/scholarpedia.1488","article-title":"Chua circuit","volume":"2","author":"Chua","year":"2007","journal-title":"Scholarpedia"},{"key":"B12","doi-asserted-by":"publisher","first-page":"6629","DOI":"10.1007\/s11071-024-09384-3","article-title":"Chaotic dynamical system of hopfield neural network influenced by neuron activation threshold and its image encryption","volume":"112","author":"Deng","year":"2024","journal-title":"Nonlinear Dyn"},{"key":"B13","doi-asserted-by":"publisher","first-page":"101237","DOI":"10.1016\/j.jocs.2020.101237","article-title":"Physics-informed echo state networks","volume":"47","author":"Doan","year":"2020","journal-title":"J. Comput. Sci"},{"key":"B14","first-page":"57","article-title":"\u201cLearning dynamical systems from data: an introduction to physics-guided deep learning,\u201d","volume-title":"Proceedings of the 4th French Regional Conference on Complex Systems","author":"Feng","year":"2024"},{"key":"B15","doi-asserted-by":"crossref","DOI":"10.1142\/1706","volume-title":"Chaos in Chemistry and Biochemistry","author":"Field","year":"1993"},{"key":"B16","doi-asserted-by":"publisher","first-page":"114425","DOI":"10.1016\/j.rse.2024.114425","article-title":"Deep learning solver unites sdgsat-1 observations and navier-stokes theory for oceanic vortex streets","volume":"315","author":"Gao","year":"2024","journal-title":"Remote Sens. Environ"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-021-25801-2","article-title":"Next generation reservoir computing","volume":"12","author":"Gauthier","year":"2021","journal-title":"Nat. Commun"},{"key":"B18","article-title":"\u201cChaos as an interpretable benchmark for forecasting and data-driven modelling,\u201d","volume-title":"NeurIPS 2021 Datasets and Benchmarks Track (Round 2)","author":"Gilpin","year":"2021"},{"key":"B19","doi-asserted-by":"publisher","first-page":"346","DOI":"10.1103\/PhysRevLett.50.346","article-title":"Characterization of strange attractors","volume":"50","author":"Grassberger","year":"1983","journal-title":"Phys. Rev. Lett"},{"key":"B20","doi-asserted-by":"publisher","first-page":"459","DOI":"10.1007\/BF00322848","article-title":"Order and chaos in chemistry and biology","volume":"337","author":"Hess","year":"1990","journal-title":"Fresenius' J. Analyt. Chem"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput"},{"key":"B22","doi-asserted-by":"publisher","first-page":"3088","DOI":"10.1073\/pnas.81.10.3088","article-title":"Neurons with graded response have collective computational properties like those of two-state neurons","volume":"81","author":"Hopfield","year":"1984","journal-title":"Proc. Nat. Acad. Sci"},{"key":"B23","doi-asserted-by":"publisher","first-page":"12485","DOI":"10.1007\/s11071-024-09696-4","article-title":"Design and analysis of a memristive hopfield switching neural network and application to privacy protection","volume":"112","author":"Hu","year":"2024","journal-title":"Nonlinear Dynam"},{"key":"B24","volume-title":"The \u201cEcho State\u201d Approach to Analysing and Training Recurrent Neural Networks-With an Erratum Note","author":"Jaeger","year":"2001"},{"key":"B25","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1063\/5.0065617","article-title":"Knowledge-based learning of nonlinear dynamics and chaos","volume":"31","author":"Jiahao","year":"2021","journal-title":"Chaos"},{"key":"B26","doi-asserted-by":"publisher","first-page":"106152","DOI":"10.1016\/j.neunet.2024.106152","article-title":"Data-driven learning of chaotic dynamical systems using discrete-temporal sobolev networks","volume":"173","author":"Kennedy","year":"2024","journal-title":"Neural Netw"},{"key":"B27","doi-asserted-by":"publisher","first-page":"85","DOI":"10.1016\/j.neunet.2023.12.008","article-title":"Memristor-induced hyperchaos, multiscroll and extreme multistability in fractional-order hnn: Image encryption and fpga implementation","volume":"171","author":"Kong","year":"2024","journal-title":"Neural Netw"},{"key":"B28","doi-asserted-by":"publisher","first-page":"987","DOI":"10.1109\/72.712178","article-title":"Artificial neural networks for solving ordinary and partial differential equations","volume":"9","author":"Lagaris","year":"1998","journal-title":"IEEE Trans. Neural Netw"},{"key":"B29","doi-asserted-by":"publisher","first-page":"655","DOI":"10.1016\/j.cnsns.2007.11.012","article-title":"Connecting the lorenz and chen systems via nonlinear control","volume":"14","author":"Li","year":"2009","journal-title":"Commun. Nonlinear Sci. Numer. Simulat"},{"key":"B30","doi-asserted-by":"publisher","first-page":"2250130","DOI":"10.1142\/S0218127422501309","article-title":"A novel multiscroll memristive hopfield neural network","volume":"32","author":"Li","year":"2022","journal-title":"Int. J. Bifurcat. Chaos"},{"key":"B31","doi-asserted-by":"publisher","first-page":"185796","DOI":"10.1109\/ACCESS.2019.2961164","article-title":"A simultaneous scrambling and diffusion color image encryption algorithm based on hopfield chaotic neural network","volume":"7","author":"Liu","year":"2019","journal-title":"IEEE Access"},{"key":"B32","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1175\/1520-0469(1963)020<0130:DNF>2.0.CO;2","article-title":"Deterministic nonperiodic flow","volume":"20","author":"Lorenz","year":"1963","journal-title":"J. Atmosph. Sci"},{"key":"B33","doi-asserted-by":"publisher","first-page":"113112","DOI":"10.1063\/1.4967730","article-title":"A chaotic model for the epidemic of ebola virus disease in west africa (2013-2016)","volume":"26","author":"Mangiarotti","year":"2016","journal-title":"Chaos"},{"key":"B34","doi-asserted-by":"publisher","first-page":"109953","DOI":"10.1016\/j.chaos.2020.109953","article-title":"A nonlinear epidemiological model considering asymptotic and quarantine classes for sars cov-2 virus","volume":"138","author":"Mishra","year":"2020","journal-title":"Chaos, Solitons Fractals"},{"key":"B35","first-page":"1","article-title":"\u201cPhysics-informed neural networks for power systems,\u201d","author":"Misyris","year":"2020","journal-title":"2020 IEEE Power"},{"key":"B36","doi-asserted-by":"publisher","first-page":"120155","DOI":"10.1016\/j.eswa.2023.120155","article-title":"Physics-informed hierarchical echo state network for predicting the dynamics of chaotic systems","volume":"228","author":"Na","year":"2023","journal-title":"Expert Syst. Appl"},{"key":"B37","doi-asserted-by":"publisher","first-page":"104229","DOI":"10.1016\/j.engappai.2021.104229","article-title":"Hierarchical delay-memory echo state network: a model designed for multi-step chaotic time series prediction","volume":"102","author":"Na","year":"2021","journal-title":"Eng. Appl. Artif. Intell"},{"key":"B38","doi-asserted-by":"publisher","first-page":"292","DOI":"10.1016\/j.neucom.2022.08.032","article-title":"MFRFNN: multi-functional recurrent fuzzy neural network for chaotic time series prediction","volume":"507","author":"Nasiri","year":"2022","journal-title":"Neurocomputing"},{"key":"B39","doi-asserted-by":"publisher","first-page":"13683","DOI":"10.1038\/s41598-023-39989-4","article-title":"Physics-informed neural networks for predicting gas flow dynamics and unknown parameters in diesel engines","volume":"13","author":"Nath","year":"2023","journal-title":"Sci. Rep"},{"key":"B40","doi-asserted-by":"publisher","first-page":"1749","DOI":"10.3390\/v15081749","article-title":"Physics-informed neural networks integrating compartmental model for analyzing COVID-19 transmission dynamics","volume":"15","author":"Ning","year":"2023","journal-title":"Viruses"},{"key":"B41","doi-asserted-by":"publisher","first-page":"10664","DOI":"10.1073\/pnas.1813901116","article-title":"Mapping solar system chaos with the geological orrery","volume":"116","author":"Olsen","year":"2019","journal-title":"Proc. Nat. Acad. Sci"},{"key":"B42","first-page":"382","article-title":"\u201cPhysics-informed long short-term memory for forecasting and reconstruction of chaos,\u201d","volume-title":"International Conference on Computational Science","author":"\u00d6zalp","year":"2023"},{"key":"B43","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1175\/1520-0477(1993)074<0049:ERAPAT>2.0.CO;2","article-title":"Extended-range atmospheric prediction and the lorenz model","volume":"74","author":"Palmer","year":"1993","journal-title":"Bull. Am. Meteorol. Soc"},{"key":"B44","article-title":"\u201cAutomatic differentiation in pytorch,\u201d","volume-title":"NeurIPS Autodiff Workshop","author":"Paszke","year":"2017"},{"key":"B45","doi-asserted-by":"publisher","first-page":"42598","DOI":"10.1063\/5.0042598","article-title":"Using machine learning to predict statistical properties of non-stationary dynamical processes: system climate, regime transitions, and the effect of stochasticity","volume":"31","author":"Patel","year":"2021","journal-title":"Chaos"},{"key":"B46","doi-asserted-by":"publisher","first-page":"024102","DOI":"10.1103\/PhysRevLett.120.024102","article-title":"Model-free prediction of large spatiotemporally chaotic systems from data: a reservoir computing approach","volume":"120","author":"Pathak","year":"2018","journal-title":"Phys. Rev. Lett"},{"key":"B47","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1063\/1.5010300","article-title":"Using machine learning to replicate chaotic attractors and calculate lyapunov exponents from data","volume":"27","author":"Pathak","year":"2017","journal-title":"Chaos"},{"key":"B48","doi-asserted-by":"publisher","first-page":"821","DOI":"10.1103\/PhysRevLett.64.821","article-title":"Synchronization in chaotic systems","volume":"64","author":"Pecora","year":"1990","journal-title":"Phys. Rev. Lett"},{"key":"B49","first-page":"323","article-title":"\u201cAutomatic-differentiated physics-informed echo state network (API-ESN),\u201d","volume-title":"International Conference on Computational Science","author":"Racca","year":"2021"},{"key":"B50","doi-asserted-by":"publisher","first-page":"686","DOI":"10.1016\/j.jcp.2018.10.045","article-title":"Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations","volume":"378","author":"Raissi","year":"2019","journal-title":"J. Comput. Phys"},{"key":"B51","doi-asserted-by":"publisher","first-page":"1381","DOI":"10.1109\/TNN.2005.857945","article-title":"Finite-element neural networks for solving differential equations","volume":"16","author":"Ramuhalli","year":"2005","journal-title":"IEEE Trans. Neural Netw"},{"key":"B52","doi-asserted-by":"publisher","first-page":"397","DOI":"10.1016\/0375-9601(76)90101-8","article-title":"An equation for continuous chaos","volume":"57","author":"R\u00f6ssler","year":"1976","journal-title":"Physics Letters A"},{"key":"B53","doi-asserted-by":"publisher","first-page":"275","DOI":"10.1016\/S0092-8240(77)80015-3","article-title":"Chaos in abstract kinetics: two prototypes","volume":"39","author":"R\u00f6ssler","year":"1977","journal-title":"Bull. Math. Biol"},{"key":"B54","doi-asserted-by":"publisher","first-page":"110045","DOI":"10.1016\/j.chaos.2020.110045","article-title":"Robustness of lstm neural networks for multi-step forecasting of chaotic time series","volume":"139","author":"Sangiorgio","year":"2020","journal-title":"Chaos, Solitons Fract"},{"key":"B55","doi-asserted-by":"publisher","first-page":"111570","DOI":"10.1016\/j.chaos.2021.111570","article-title":"Forecasting of noisy chaotic systems with deep neural networks","volume":"153","author":"Sangiorgio","year":"2021","journal-title":"Chaos, Solitons Fract"},{"key":"B56","volume-title":"Deep Learning in Multi-Step Prediction of Chaotic Dynamics: From Deterministic Models to Real-World Systems","author":"Sangiorgio","year":"2022"},{"key":"B57","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2024.3457039","article-title":"A physics-guided attention-based neural network for sea surface temperature prediction","volume":"62","author":"Shi","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sensing"},{"key":"B58","doi-asserted-by":"publisher","first-page":"1339","DOI":"10.1016\/j.jcp.2018.08.029","article-title":"Dgm: a deep learning algorithm for solving partial differential equations","volume":"375","author":"Sirignano","year":"2018","journal-title":"J. Comput. Phys"},{"key":"B59","doi-asserted-by":"crossref","DOI":"10.1142\/7183","volume-title":"Elegant Chaos: Algebraically Simple Chaotic Flows","author":"Sprott","year":"2010"},{"key":"B60","doi-asserted-by":"publisher","first-page":"4887","DOI":"10.1002\/grl.50940","article-title":"Projections of global changes in precipitation extremes from coupled model intercomparison project phase 5 models","volume":"40","author":"Toreti","year":"2013","journal-title":"Geophys. Res. Lett"},{"key":"B61","article-title":"\u201cAttention is all you need,\u201d","volume-title":"Proceedings of Advances in Neural Information Processing Systems","author":"Vaswani","year":"2017"},{"key":"B62","doi-asserted-by":"publisher","first-page":"191","DOI":"10.1016\/j.neunet.2020.02.016","article-title":"Backpropagation algorithms and reservoir computing in recurrent neural networks for the forecasting of complex spatiotemporal dynamics","volume":"126","author":"Vlachas","year":"2020","journal-title":"Neural Netw"},{"key":"B63","doi-asserted-by":"publisher","first-page":"229","DOI":"10.1038\/s41598-023-50690-4","article-title":"Synchronization of non-smooth chaotic systems via an improved reservoir computing","volume":"14","author":"Wu","year":"2024","journal-title":"Sci. Rep"},{"key":"B64","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2305.14452","article-title":"Fourier neural operators for arbitrary resolution climate data downscaling","author":"Yang","year":"2023","journal-title":"arXiv"},{"key":"B65","doi-asserted-by":"publisher","first-page":"11121","DOI":"10.1609\/aaai.v37i9.26317","article-title":"Are transformers effective for time series forecasting?","volume":"37","author":"Zeng","year":"2023","journal-title":"Proc. AAAI Conf. Artif. Intellig"}],"container-title":["Frontiers in Big Data"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdata.2024.1506443\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,17]],"date-time":"2025-01-17T06:51:49Z","timestamp":1737096709000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdata.2024.1506443\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,17]]},"references-count":65,"alternative-id":["10.3389\/fdata.2024.1506443"],"URL":"https:\/\/doi.org\/10.3389\/fdata.2024.1506443","relation":{},"ISSN":["2624-909X"],"issn-type":[{"value":"2624-909X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,17]]},"article-number":"1506443"}}