Data assimilation(DA)and uncertainty quantification(UQ)are extensively used in analysing and reducing error propagation in high-dimensional spatial-temporal dynamics.Typical applications span from computational fluid ...Data assimilation(DA)and uncertainty quantification(UQ)are extensively used in analysing and reducing error propagation in high-dimensional spatial-temporal dynamics.Typical applications span from computational fluid dynamics(CFD)to geoscience and climate systems.Recently,much effort has been given in combining DA,UQ and machine learning(ML)techniques.These research efforts seek to address some critical challenges in high-dimensional dynamical systems,including but not limited to dynamical system identification,reduced order surrogate modelling,error covariance specification and model error correction.A large number of developed techniques and methodologies exhibit a broad applicability across numerous domains,resulting in the necessity for a comprehensive guide.This paper provides the first overview of state-of-the-art researches in this interdisciplinary field,covering a wide range of applications.This review is aimed at ML scientists who attempt to apply DA and UQ techniques to improve the accuracy and the interpretability of their models,but also at DA and UQ experts who intend to integrate cutting-edge ML approaches to their systems.Therefore,this article has a special focus on how ML methods can overcome the existing limits of DA and UQ,and vice versa.Some exciting perspectives of this rapidly developing research field are also discussed.Index Terms-Data assimilation(DA),deep learning,machine learning(ML),reduced-order-modelling,uncertainty quantification(UQ).展开更多
Molecular speciation of atmospheric organic matter was investigated during a short summer field campaign performed in a citrus fruit field in northern Corsica(June 2011).Aimedat assessing the performance on the field ...Molecular speciation of atmospheric organic matter was investigated during a short summer field campaign performed in a citrus fruit field in northern Corsica(June 2011).Aimedat assessing the performance on the field of newly developed analytical protocols,this work focuses on the molecular composition of both gas and particulate phases and provides an insight into partitioning behavior of the semi-volatile oxygenated fraction.Limonene ozonolysis tracers were specifically searched for,according to gas chromatography–mass spectrometry(GC–MS)data previously recorded for smog chamber experiments.A screening of other oxygenated species present in the field atmosphere was also performed.About sixty polar molecules were positively or tentatively identified in gas and/or particle phases.These molecules comprise a wide range of branched and linear,mono and di-carbonyls(C_3–C7),mono and di-carboxylic acids(C_3–C_18),and compounds bearing up to three functionalities.Among these compounds,some can be specifically attributed to limonene oxidation and others can be related toα-orβ-pinene oxidation.This provides an original snapshot of the organic matter composition at a Mediterranean site in summer.Furthermore,for compounds identified and quantified in both gaseous and particulate phases,an experimental gas/particle partitioning coefficient was determined.Several volatile products,which are not expected in the particulate phase assuming thermodynamic equilibrium,were nonetheless present in significant concentrations.Hypotheses are proposed to explain these observations,such as the possible aerosol viscosity that could hinder the theoretical equilibrium to be rapidly reached.展开更多
In this work,a combination of an acoustic emission (AE) technique and a machine learning (ML) algorithm (Random Forest (RF) and Gradient Boosting Regressor (GBR)) is developed to characterize the particle size distrib...In this work,a combination of an acoustic emission (AE) technique and a machine learning (ML) algorithm (Random Forest (RF) and Gradient Boosting Regressor (GBR)) is developed to characterize the particle size distribution in gas-solid fluidized bed reactors.A theoretical approach to explain the generation of acoustic emission signal in gas-solid flows is presented.An AE signal is generated in gas-solid fluidized beds due to the collision and friction between fluidized particles as well as between particles and the bed inner wall.The generated AE signal is in the form of an elastic wave with frequencies >100 KHz and it propagates through the gas-solid mixture.An inversion algorithm is used to extract the information about the particle size starting from the energy of the AE signal.The advantages of this AE technique are that it is a cheap,sensitive,non-intrusive,radiation-free,suitable for on-line measurements.Combining this AE technique with ML algorithms is beneficial for applications to industrial settings,reducing the cost of signal post-processing.Experiments were conducted in a pseudo-2D flat fluidized bed with four glass bead samples,with sizes ranging from 100 μm to 710 μm.AE signals were recorded with a sampling frequency of 5 MHz.The AE signal post-processing and data preparation for the ML process are explained.For the ML process,the AE frequency,AE energy and particle collision velocity data sets were divided into training (60%),cross-validation (20%) and test sets (20%).Two ensemble ML approaches,namely Random Forest and Gradient Boosting Regressor,are applied to predict particle sizes based on the AE signal features.The combination of these two models results in a coefficient of determination (R2) value greater than 0.9504.展开更多
基金the support of the Leverhulme Centre for Wildfires,Environment and Society through the Leverhulme Trust(RC-2018-023)Sibo Cheng,César Quilodran-Casas,and Rossella Arcucci acknowledge the support of the PREMIERE project(EP/T000414/1)+5 种基金the support of EPSRC grant:PURIFY(EP/V000756/1)the Fundamental Research Funds for the Central Universitiesthe support of the SASIP project(353)funded by Schmidt Futures–a philanthropic initiative that seeks to improve societal outcomes through the development of emerging science and technologiesDFG for the Heisenberg Programm Award(JA 1077/4-1)the National Natural Science Foundation of China(61976120)the Natural Science Key Foundat ion of Jiangsu Education Department(21KJA510004)。
文摘Data assimilation(DA)and uncertainty quantification(UQ)are extensively used in analysing and reducing error propagation in high-dimensional spatial-temporal dynamics.Typical applications span from computational fluid dynamics(CFD)to geoscience and climate systems.Recently,much effort has been given in combining DA,UQ and machine learning(ML)techniques.These research efforts seek to address some critical challenges in high-dimensional dynamical systems,including but not limited to dynamical system identification,reduced order surrogate modelling,error covariance specification and model error correction.A large number of developed techniques and methodologies exhibit a broad applicability across numerous domains,resulting in the necessity for a comprehensive guide.This paper provides the first overview of state-of-the-art researches in this interdisciplinary field,covering a wide range of applications.This review is aimed at ML scientists who attempt to apply DA and UQ techniques to improve the accuracy and the interpretability of their models,but also at DA and UQ experts who intend to integrate cutting-edge ML approaches to their systems.Therefore,this article has a special focus on how ML methods can overcome the existing limits of DA and UQ,and vice versa.Some exciting perspectives of this rapidly developing research field are also discussed.Index Terms-Data assimilation(DA),deep learning,machine learning(ML),reduced-order-modelling,uncertainty quantification(UQ).
文摘Molecular speciation of atmospheric organic matter was investigated during a short summer field campaign performed in a citrus fruit field in northern Corsica(June 2011).Aimedat assessing the performance on the field of newly developed analytical protocols,this work focuses on the molecular composition of both gas and particulate phases and provides an insight into partitioning behavior of the semi-volatile oxygenated fraction.Limonene ozonolysis tracers were specifically searched for,according to gas chromatography–mass spectrometry(GC–MS)data previously recorded for smog chamber experiments.A screening of other oxygenated species present in the field atmosphere was also performed.About sixty polar molecules were positively or tentatively identified in gas and/or particle phases.These molecules comprise a wide range of branched and linear,mono and di-carbonyls(C_3–C7),mono and di-carboxylic acids(C_3–C_18),and compounds bearing up to three functionalities.Among these compounds,some can be specifically attributed to limonene oxidation and others can be related toα-orβ-pinene oxidation.This provides an original snapshot of the organic matter composition at a Mediterranean site in summer.Furthermore,for compounds identified and quantified in both gaseous and particulate phases,an experimental gas/particle partitioning coefficient was determined.Several volatile products,which are not expected in the particulate phase assuming thermodynamic equilibrium,were nonetheless present in significant concentrations.Hypotheses are proposed to explain these observations,such as the possible aerosol viscosity that could hinder the theoretical equilibrium to be rapidly reached.
基金support from Engineering and Physical Sciences Research Council,UK,through the PREMIERE Programme Grant(EP/T000414/1).
文摘In this work,a combination of an acoustic emission (AE) technique and a machine learning (ML) algorithm (Random Forest (RF) and Gradient Boosting Regressor (GBR)) is developed to characterize the particle size distribution in gas-solid fluidized bed reactors.A theoretical approach to explain the generation of acoustic emission signal in gas-solid flows is presented.An AE signal is generated in gas-solid fluidized beds due to the collision and friction between fluidized particles as well as between particles and the bed inner wall.The generated AE signal is in the form of an elastic wave with frequencies >100 KHz and it propagates through the gas-solid mixture.An inversion algorithm is used to extract the information about the particle size starting from the energy of the AE signal.The advantages of this AE technique are that it is a cheap,sensitive,non-intrusive,radiation-free,suitable for on-line measurements.Combining this AE technique with ML algorithms is beneficial for applications to industrial settings,reducing the cost of signal post-processing.Experiments were conducted in a pseudo-2D flat fluidized bed with four glass bead samples,with sizes ranging from 100 μm to 710 μm.AE signals were recorded with a sampling frequency of 5 MHz.The AE signal post-processing and data preparation for the ML process are explained.For the ML process,the AE frequency,AE energy and particle collision velocity data sets were divided into training (60%),cross-validation (20%) and test sets (20%).Two ensemble ML approaches,namely Random Forest and Gradient Boosting Regressor,are applied to predict particle sizes based on the AE signal features.The combination of these two models results in a coefficient of determination (R2) value greater than 0.9504.
