Article
Thermodynamics
Laura Donato, Chiara Galletti, Alessandro Parente
Summary: Data assimilation is used to improve the predictive performance of a digital twin model of a semi-industrial combustion furnace. By accounting for underlying uncertainties, the study demonstrates the potential of data assimilation in building accurate and adaptive reduced-order models.
APPLIED THERMAL ENGINEERING
(2024)
Article
Environmental Sciences
Herve Petetin, Dene Bowdalo, Pierre-Antoine Bretonniere, Marc Guevara, Oriol Jorba, Jan Mateu Armengol, Margarida Samso Cabre, Kim Serradell, Albert Soret, Carlos Perez Garcia-Pando
Summary: This study investigates the improvement of air quality (AQ) forecasts using various model output statistics (MOS) methods. The results show that these methods can substantially improve the accuracy and correlations of the forecasts. The choice of MOS method depends on the forecast application and the desired predictive skill.
ATMOSPHERIC CHEMISTRY AND PHYSICS
(2022)
Article
Meteorology & Atmospheric Sciences
Jeffrey S. Whitaker, Anna Shlyaeva, Stephen G. Penny
Summary: This study compares two methods for incorporating a time-invariant, high-rank covariance estimate in an ensemble-based data assimilation system: the hybrid-covariance approach and the hybrid-gain approach. The results show that the simpler and less expensive hybrid-gain approach can achieve similar performance if the incremental normal-mode balance constraint applied to the ensemble-part of the hybrid-covariance update is turned off.
JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS
(2022)
Letter
Engineering, Aerospace
Jorge A. Ortega-Contreras, Yuriy S. Shmaliy, Jose A. Andrade-Lucio, Oscar G. Ibarra-Manzano
Summary: In this article, a novel approach is proposed to develop a robust H-2-OFIR filter for disturbed systems under initial and measurement errors. It is shown that the proposed filter outperforms other filters in terms of accuracy and robustness in GPS-based tracking of moving vehicles.
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS
(2022)
Article
Mathematics, Applied
Manoj K. Nambiar, Youmin Tang, Ziwang Deng
Summary: The study evaluates the performance of the Reduced-rank Sigma-Point Kalman filter (RSPKF) data assimilation method with the introduction of a localization scheme, which helps reduce the number of sigma points required and improve the minimum RMSE. The localization proves to be crucial in achieving optimal estimates independent of the state dimension of the model, highlighting its importance in oceanic or atmospheric General Circulation Models (GCMs) data assimilation.
PHYSICA D-NONLINEAR PHENOMENA
(2021)
Article
Meteorology & Atmospheric Sciences
Haohao Sun, Lili Lei, Zhengyu Liu, Liang Ning, Zhe-Min Tan
Summary: An analog offline ensemble Kalman filter (AOEnKF) is proposed, which constructs ensemble priors from a control climate simulation for each assimilation time based on an analog criterion using proxy observations. AOEnKF generates smaller posterior errors and requires much less computational cost compared to the online cycling EnKF (CEnKF). It has the advantages of having a more accurate prior ensemble mean and flow-dependent background error covariances compared to the commonly applied offline EnKF (OEnKF).
JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS
(2022)
Article
Meteorology & Atmospheric Sciences
Lars Nerger
Summary: The study introduces a hybrid filter combining LETKF and NETF with the performance improved by adjusting the hybrid weight. Results show that a hybrid variant applying NETF followed by LETKF yields the best results in complex nonlinear models. Calculating the hybrid weight based on skewness, kurtosis, and effective sample size reduces estimation errors and enhances stability of the hybrid filter.
QUARTERLY JOURNAL OF THE ROYAL METEOROLOGICAL SOCIETY
(2022)
Review
Environmental Sciences
Bowen Wang, Zhibin Sun, Xinyue Jiang, Jun Zeng, Runqing Liu
Summary: In 1960, R.E. Kalman published a paper introducing the Kalman filter, which provides a recursive solution to the discrete-data linear filtering problem. Over the years, the Kalman filter has become one of the most important tools in science and engineering due to the advancement of numerical computing and the increasing demand for various applications such as weather prediction and target tracking. This paper aims to collect and organize the fundamental principles of the Kalman filter and its derivative algorithms (including EKF, UKF, and EnKF), compare their advantages and limitations, and provide examples of their applications in data assimilation.
Article
Engineering, Ocean
Shintaro Gomi, Tsutomu Takagi, Katsuya Suzuki, Rika Shiraki, Ichiya Ogino, Shigeru Asaumi
Summary: A control method for changing the geometry of a fishing net was proposed, utilizing data assimilation to estimate unknown parameters and achieve the intended net geometry. The automatic control system was validated through numerical simulation experiments, demonstrating the successful control of net geometry using the extended Kalman filter.
APPLIED OCEAN RESEARCH
(2021)
Article
Engineering, Petroleum
Ricardo Vasconcellos Soares, Xiaodong Luo, Geir Evensen, Tuhin Bhakto
Summary: This study demonstrates the practical advantages of a new local analysis scheme in a 4D seismic history-matching problem. Compared to the Kalman gain localization scheme, the proposed local analysis scheme has improved capacity in handling big models and data sets, leading to faster convergence to the same level of data mismatch values.
Article
Automation & Control Systems
Yuriy S. Shmaliy, Oscar G. Ibarra Manzano, Jose A. Andrade Lucio
Summary: In this article, a robust H-2 optimal unbiased FIR predictor and filter are developed for uncertain and disturbed systems, demonstrating superior performance over existing filters under severe disturbances and large timing error.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2022)
Article
Mathematics, Interdisciplinary Applications
Dongrui Shao, Junyu Chu, Luonan Chen, Huanfei Ma
Summary: Data assimilation is crucial for both data driven and model driven research. The Kalman filter, a widely used data assimilation framework, has traditionally relied on theoretical models. However, recent efforts have aimed to develop model-free Kalman filters that solely rely on data. In this study, we propose a hybrid model framework that combines delay embedding theory and machine learning to bridge the gap between exact model-based and totally model-free methods. This hybrid approach is more flexible in application and has been validated using benchmark systems and real-world problems.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Automation & Control Systems
Juan J. Lopez Solorzano, Yuriy S. Shmaliy
Summary: This article proposes a robust RH H-2-FIR filter for uncertain system disturbances and demonstrates its superiority in quasi-periodic disturbed processes.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2022)
Article
Environmental Sciences
G. Piazzi, G. Thirel, C. Perrin, O. Delaigue
Summary: Skillful streamflow forecasts are crucial for water-related applications, with a growing emphasis on improving initial condition estimates through data assimilation. This study assesses the sensitivity of DA-based IC estimation to various uncertainties and model updates over 232 watersheds in France. The comparison of two ensemble-based techniques shows that accurate routing store estimates benefit the DA-based IC estimation, with the EnKF outperforming the PF in forecasting meteorological uncertainty.
WATER RESOURCES RESEARCH
(2021)
Article
Astronomy & Astrophysics
Zheqi Shen, Youmin Tang, Xiaojing Li, Yanqiu Gao
Summary: For coupled numerical models, there are two data assimilation strategies: strongly coupled data assimilation (SCDA) can update variables from different model components, potentially outperforming weakly coupled data assimilation (WCDA). The success of SCDA relies heavily on the accuracy of cross-component covariance, especially when the model components have different spatial scales. Newly proposed localization strategies can enhance the accuracy of SCDA analyses compared to WCDA in twin experiments.
EARTH AND SPACE SCIENCE
(2021)
Article
Energy & Fuels
T. Vezin, S. Meunier, L. Queval, J. A. Cherni, L. Vido, A. Darga, P. Dessante, P. K. Kitanidis, C. Marchand
Article
Water Resources
Hojat Ghorbanidehno, Jonghyun Lee, Matthew Farthing, Tyler Hesser, Eric F. Darve, Peter K. Kitanidis
Summary: Riverine bathymetry is important for shipping and flood management, and indirect measurements with sensor technology can be used to estimate river bed topography. Physics-based techniques are computationally expensive, while deep learning offers a data-driven approach with potential for efficient training using limited data. The proposed method combines DNN with PCA to image river bed topography using flow velocity observations, showing satisfactory performance in bathymetry estimation with low computational cost and small number of training samples.
ADVANCES IN WATER RESOURCES
(2021)
Article
Environmental Sciences
Xueyuan Kang, Amalia Kokkinaki, Peter K. Kitanidis, Xiaoqing Shi, Andre Revil, Jonghyun Lee, Abdellahi Soueid Ahmed, Jichun Wu
WATER RESOURCES RESEARCH
(2020)
Article
Engineering, Environmental
Mojtaba Forghani, Yizhou Qian, Jonghyun Lee, Matthew W. Farthing, Tyler Hesser, Peter K. Kitanidis, Eric F. Darve
Summary: This study proposes a two-stage process utilizing principal component geostatistical approach to estimate bathymetry probability density function and multiple machine learning algorithms to solve shallow water equations (SWEs) efficiently. The first stage predicts flow velocities without direct bathymetry measurement, while the second stage incorporates additional bathymetry information for improved accuracy and generalization. Fast solvers are capable of accurately predicting flow velocities with variable bathymetry and BCs at a significantly lower computational cost compared to traditional methods.
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT
(2021)
Article
Environmental Sciences
Xueyuan Kang, Amalia Kokkinaki, Peter K. Kitanidis, Xiaoqing Shi, Jonghyun Lee, Shaoxing Mo, Jichun Wu
Summary: Characterizing the architecture of dense nonaqueous phase liquid (DNAPL) source zones is crucial for designing efficient remediation strategies, but traditional drilling investigations provide limited information and affect the accuracy of geostatistical methods. By parameterizing the DNAPL saturation field using a physics-based approach, improved prior descriptions and better resolution can be achieved in characterizing the source zones. Additionally, incorporating hydrogeological and geophysical datasets in the inversion framework can further enhance the performance of the method.
WATER RESOURCES RESEARCH
(2021)
Article
Mathematics, Interdisciplinary Applications
Peter K. Kitanidis
Summary: This paper discusses the application of covariance and Fisher information matrix in inverse problems, as well as a reexamination within the Bayesian framework, proposing a lower bound for the covariance of the posterior probability density function.
GEM-INTERNATIONAL JOURNAL ON GEOMATHEMATICS
(2021)
Article
Engineering, Civil
Xueyuan Kang, Amalia Kokkinaki, Christopher Power, Peter K. Kitanidis, Xiaoqing Shi, Limin Duan, Tingxi Liu, Jichun Wu
Summary: The proposed geostatistical data assimilation method, using deep learning techniques, significantly improves the monitoring of DNAPL remediation. Experimental results show a reduction of 51% in the estimation error of DNAPL mass remediation compared to the standard EnKF method, indicating better real-time monitoring of DNAPL remediation.
JOURNAL OF HYDROLOGY
(2021)
Article
Multidisciplinary Sciences
Yi-Lin Tsai, Chetanya Rastogi, Peter K. Kitanidis, Christopher B. Field
Summary: The study discusses the importance of integrating social distancing with emergency evacuation operations and found that deep reinforcement learning can provide more efficient routing compared to other solutions. However, the time saved by deep reinforcement learning in evacuation does not compensate for the extra time required for social distancing as the emergency vehicle capacity approaches the number of people per household.
SCIENTIFIC REPORTS
(2021)
Article
Environmental Sciences
Lijing Wang, Peter K. Kitanidis, Jef Caers
Summary: Bayesian inversion is commonly used to quantify uncertainty of hydrological variables. This paper proposes a hierarchical Bayesian framework to quantify uncertainty of both global and spatial variables. The authors present a machine learning-based inversion method and a local dimension reduction method to efficiently estimate posterior probabilities and update spatial fields. Using three case studies, they demonstrate the importance of quantifying uncertainty of global variables for predictions and the acceleration effect of the local PCA approach.
WATER RESOURCES RESEARCH
(2022)
Article
Thermodynamics
Pragneshkumar Rajubhai Rana, Krithika Narayanaswamy, Sivaram Ambikasaran
Summary: Ignition delay time is a significant combustion property. This study proposes a data-driven approach to obtain the ignition delay time for new fuels. The proposed algorithm uses regression-based clustering and incorporates fuel structure to establish models. The accuracy of the algorithm is demonstrated using straight-chain alkane data.
COMBUSTION THEORY AND MODELLING
(2022)
Article
Water Resources
Mojtaba Forghani, Yizhou Qian, Jonghyun Lee, Matthew Farthing, Tyler Hesser, Peter K. Kitanidis, Eric F. Darve
Summary: This article presents a reduced-order model (ROM) based approach that utilizes a variational autoencoder (VAE) to compress bathymetry and flow velocity information, allowing for fast solving of bathymetry inverse problems. By constructing ROMs on a nonlinear manifold and employing a Hierarchical Bayesian setting, variational inference and efficient uncertainty quantification can be achieved using a small number of ROM runs.
ADVANCES IN WATER RESOURCES
(2022)
Article
Environmental Sciences
Xueyuan Kang, Amalia Kokkinaki, Xiaoqing Shi, Hongkyu Yoon, Jonghyun Lee, Peter K. Kitanidis, Jichun Wu
Summary: This study presents a framework that combines a deep-learning-based inversion method with a process-based upscaled model to estimate source zone architecture (SZA) metrics and mass discharge from sparse data. By improving the estimation method, the upscaled model accurately reproduces the concentrations and uncertainties of multistage effluents, providing valuable input for decision making in remediation applications.
WATER RESOURCES RESEARCH
(2022)
Article
Environmental Sciences
Simon Meunier, Peter K. Kitanidis, Amaury Cordier, Alan M. MacDonald
Summary: This study develops a numerical model to simulate the abstraction capacities of photovoltaic water pumping systems across Africa using openly available data. The model includes realistic geological constraints on pumping depth and sub-hourly irradiance time series. The simulation results show that for much of Africa, groundwater pumping using photovoltaic energy is limited by aquifer conditions rather than irradiance. These findings can help identify regions with high potential for photovoltaic pumping and guide large-scale investments.
COMMUNICATIONS EARTH & ENVIRONMENT
(2023)
Article
Physics, Mathematical
Vaishnavi Gujjula, Sivaram Ambikasaran
Summary: This article introduces a fast iterative solver for 2D scattering problems, utilizing the Green's function to represent the scattered field and discretizing the Lippmann-Schwinger equation with appropriate quadrature technique. The iterative solver accelerated by DAFMM and the new NCA method show significant advantages in numerical experiments.
COMMUNICATIONS IN COMPUTATIONAL PHYSICS
(2022)
Article
Engineering, Multidisciplinary
Simon Meunier, Loic Queval, Arouna Darga, Philippe Dessante, Claude Marchand, Matthias Heinrich, Judith A. Cherni, Elvire A. de la Fresnaye, Lionel Vido, Bernard Multon, Peter K. Kitanidis
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2020)
