Physics-informed neural networks(PINNs)have shown remarkable prospects in solving the forward and inverse problems involving partial differential equations(PDEs).The method embeds PDEs into the neural network by calcu...Physics-informed neural networks(PINNs)have shown remarkable prospects in solving the forward and inverse problems involving partial differential equations(PDEs).The method embeds PDEs into the neural network by calculating the PDE loss at a set of collocation points,providing advantages such as meshfree and more convenient adaptive sampling.However,when solving PDEs using nonuniform collocation points,PINNs still face challenge regarding inefficient convergence of PDE residuals or even failure.In this work,we first analyze the ill-conditioning of the PDE loss in PINNs under nonuniform collocation points.To address the issue,we define volume weighting residual and propose volume weighting physics-informed neural networks(VW-PINNs).Through weighting the PDE residuals by the volume that the collocation points occupy within the computational domain,we embed explicitly the distribution characteristics of collocation points in the loss evaluation.The fast and sufficient convergence of the PDE residuals for the problems involving nonuniform collocation points is guaranteed.Considering the meshfree characteristics of VW-PINNs,we also develop a volume approximation algorithm based on kernel density estimation to calculate the volume of the collocation points.We validate the universality of VW-PINNs by solving the forward problems involving flow over a circular cylinder and flow over the NACA0012 airfoil under different inflow conditions,where conventional PINNs fail.By solving the Burgers’equation,we verify that VW-PINNs can enhance the efficiency of existing the adaptive sampling method in solving the forward problem by three times,and can reduce the relative L 2 error of conventional PINNs in solving the inverse problem by more than one order of magnitude.展开更多
Prognosis of bearing is critical to improve the safety,reliability,and availability of machinery systems,which provides the health condition assessment and determines how long the machine would work before failure occ...Prognosis of bearing is critical to improve the safety,reliability,and availability of machinery systems,which provides the health condition assessment and determines how long the machine would work before failure occurs by predicting the remaining useful life(RUL).In order to overcome the drawback of pure data-driven methods and predict RUL accurately,a novel physics-informed deep neural network,named degradation consistency recurrent neural network,is proposed for RUL prediction by integrating the natural degradation knowledge of mechanical components.The degradation is monotonic over the whole life of bearings,which is characterized by temperature signals.To incorporate the knowledge of monotonic degradation,a positive increment recurrence relationship is introduced to keep the monotonicity.Thus,the proposed model is relatively well understood and capable to keep the learning process consistent with physical degradation.The effectiveness and merit of the RUL prediction using the proposed method are demonstrated through vibration signals collected from a set of run-to-failure tests.展开更多
Physics-informed deep learning has drawn tremendous interest in recent years to solve computational physics problems,whose basic concept is to embed physical laws to constrain/inform neural networks,with the need of l...Physics-informed deep learning has drawn tremendous interest in recent years to solve computational physics problems,whose basic concept is to embed physical laws to constrain/inform neural networks,with the need of less data for training a reliable model.This can be achieved by incorporating the residual of physics equations into the loss function.Through minimizing the loss function,the network could approximate the solution.In this paper,we propose a mixed-variable scheme of physics-informed neural network(PINN)for fluid dynamics and apply it to simulate steady and transient laminar flows at low Reynolds numbers.A parametric study indicates that the mixed-variable scheme can improve the PINN trainability and the solution accuracy.The predicted velocity and pressure fields by the proposed PINN approach are also compared with the reference numerical solutions.Simulation results demonstrate great potential of the proposed PINN for fluid flow simulation with a high accuracy.展开更多
This paper proposes a physics-informed neural network(PINN)framework to analyze the nonlinear buckling behavior of a three-dimensional(3D)FG porous,slender beam resting on a Winkler-Pasternak foundation.PINNs need muc...This paper proposes a physics-informed neural network(PINN)framework to analyze the nonlinear buckling behavior of a three-dimensional(3D)FG porous,slender beam resting on a Winkler-Pasternak foundation.PINNs need much less training data to obtain high accuracy using a straightforward network.The powerful tool used in this work can handle any class of PDEs.We use the deep learning platform TensorFlow and DeepXDE library to design our network.In this study,the PINNs framework takes information from the governing differential equations of the beam system and the data from boundary conditions and outputs the critical nonlinear buckling load.The mathematical model is developed using Hamilton’s principle,considering geometry’s nonlinearity.The accuracy of the modeling framework is carefully examined by applying it to various boundary condition cases as well as the physical parameters such as 3D FG indexes on the nonlinear mechanical behaviors.Finally,the PINNs results are validated with those extracted from the generalized differential quadrature method(GDQM).It is found that the proposed PINN framework can characterize the nonlinear buckling behavior of 3D FG porous,slender beams with satisfactory accuracy.Furthermore,PINN is presented to accurately predict the nonlinear buckling behavior of the beam up to 71 times faster than the numerical method.展开更多
Recently,numerous studies have demonstrated that the physics-informed neural network(PINN)can effectively and accurately resolve hyperelastic finite deformation problems.In this paper,a PINN framework for tackling hyp...Recently,numerous studies have demonstrated that the physics-informed neural network(PINN)can effectively and accurately resolve hyperelastic finite deformation problems.In this paper,a PINN framework for tackling hyperelastic-magnetic coupling problems is proposed.Since the solution space consists of two-phase domains,two separate networks are constructed to independently predict the solution for each phase region.In addition,a conscious point allocation strategy is incorporated to enhance the prediction precision of the PINN in regions characterized by sharp gradients.With the developed framework,the magnetic fields and deformation fields of magnetorheological elastomers(MREs)are solved under the control of hyperelastic-magnetic coupling equations.Illustrative examples are provided and contrasted with the reference results to validate the predictive accuracy of the proposed framework.Moreover,the advantages of the proposed framework in solving hyperelastic-magnetic coupling problems are validated,particularly in handling small data sets,as well as its ability in swiftly and precisely forecasting magnetostrictive motion.展开更多
To reduce CO_(2) emissions in response to global climate change,shale reservoirs could be ideal candidates for long-term carbon geo-sequestration involving multi-scale transport processes.However,most current CO_(2) s...To reduce CO_(2) emissions in response to global climate change,shale reservoirs could be ideal candidates for long-term carbon geo-sequestration involving multi-scale transport processes.However,most current CO_(2) sequestration models do not adequately consider multiple transport mechanisms.Moreover,the evaluation of CO_(2) storage processes usually involves laborious and time-consuming numerical simulations unsuitable for practical prediction and decision-making.In this paper,an integrated model involving gas diffusion,adsorption,dissolution,slip flow,and Darcy flow is proposed to accurately characterize CO_(2) storage in depleted shale reservoirs,supporting the establishment of a training database.On this basis,a hybrid physics-informed data-driven neural network(HPDNN)is developed as a deep learning surrogate for prediction and inversion.By incorporating multiple sources of scientific knowledge,the HPDNN can be configured with limited simulation resources,significantly accelerating the forward and inversion processes.Furthermore,the HPDNN can more intelligently predict injection performance,precisely perform reservoir parameter inversion,and reasonably evaluate the CO_(2) storage capacity under complicated scenarios.The validation and test results demonstrate that the HPDNN can ensure high accuracy and strong robustness across an extensive applicability range when dealing with field data with multiple noise sources.This study has tremendous potential to replace traditional modeling tools for predicting and making decisions about CO_(2) storage projects in depleted shale reservoirs.展开更多
Extracting typical operational scenarios is essential for making flexible decisions in the dispatch of a new power system.A novel deep time series aggregation scheme(DTSAs)is proposed to generate typical operational s...Extracting typical operational scenarios is essential for making flexible decisions in the dispatch of a new power system.A novel deep time series aggregation scheme(DTSAs)is proposed to generate typical operational scenarios,considering the large amount of historical operational snapshot data.Specifically,DTSAs analyse the intrinsic mechanisms of different scheduling operational scenario switching to mathematically represent typical operational scenarios.A Gramian angular summation field-based operational scenario image encoder was designed to convert operational scenario sequences into highdimensional spaces.This enables DTSAs to fully capture the spatiotemporal characteristics of new power systems using deep feature iterative aggregation models.The encoder also facilitates the generation of typical operational scenarios that conform to historical data distributions while ensuring the integrity of grid operational snapshots.Case studies demonstrate that the proposed method extracted new fine-grained power system dispatch schemes and outperformed the latest high-dimensional feature-screening methods.In addition,experiments with different new energy access ratios were conducted to verify the robustness of the proposed method.DTSAs enable dispatchers to master the operation experience of the power system in advance,and actively respond to the dynamic changes of the operation scenarios under the high access rate of new energy.展开更多
Flash floods are one of the most dangerous natural disasters,especially in hilly terrain,causing loss of life,property,and infrastructures and sudden disruption of traffic.These types of floods are mostly associated w...Flash floods are one of the most dangerous natural disasters,especially in hilly terrain,causing loss of life,property,and infrastructures and sudden disruption of traffic.These types of floods are mostly associated with landslides and erosion of roads within a short time.Most of Vietnamis hilly and mountainous;thus,the problem due to flash flood is severe and requires systematic studies to correctly identify flood susceptible areas for proper landuse planning and traffic management.In this study,three Machine Learning(ML)methods namely Deep Learning Neural Network(DL),Correlation-based FeatureWeighted Naive Bayes(CFWNB),and Adaboost(AB-CFWNB)were used for the development of flash flood susceptibility maps for hilly road section(115 km length)of National Highway(NH)-6 inHoa Binh province,Vietnam.In the proposedmodels,88 past flash flood events were used together with 14 flash floods affecting topographical and geo-environmental factors.The performance of themodels was evaluated using standard statisticalmeasures including Receiver Operating Characteristic(ROC)Curve,Area Under Curve(AUC)and Root Mean Square Error(RMSE).The results revealed that all the models performed well(AUC>0.80)in predicting flash flood susceptibility zones,but the performance of the DL model is the best(AUC:0.972,RMSE:0.352).Therefore,the DL model can be applied to develop an accurate flash flood susceptibility map of hilly terrain which can be used for proper planning and designing of the highways and other infrastructure facilities besides landuse management of the area.展开更多
In recent years,physics-informed neural networks(PINNs) have become a representative method for solving partial differential equations(PDEs) with neural networks.PINNs provide a novel approach to solving PDEs through ...In recent years,physics-informed neural networks(PINNs) have become a representative method for solving partial differential equations(PDEs) with neural networks.PINNs provide a novel approach to solving PDEs through optimization algorithms,offering a unified framework for solving both forward and inverse problems.However,some limitations in terms of solution accuracy and generality have also been revealed.This paper systematically summarizes the limitations of PINNs and identifies three root causes for their failure in solving PDEs:(1) poor multiscale approximation ability and ill-conditioning caused by PDE losses;(2) insufficient exploration of convergence and error analysis,resulting in weak mathematical rigor;(3) inadequate integration of physical information,causing mismatch between residuals and iteration errors.By focusing on addressing these limitations in PINNs,we outline the future directions and prospects for the intelligent computing of PDEs:(1) analysis of illconditioning in PINNs and mitigation strategies;(2) improvements to PINNs by enforcing temporal causality;(3) empowering PINNs with classical numerical methods.展开更多
Visual tracking is a classical computer vision problem with many applications.Efficient convolution operators(ECO)is one of the most outstanding visual tracking algorithms in recent years,it has shown great performanc...Visual tracking is a classical computer vision problem with many applications.Efficient convolution operators(ECO)is one of the most outstanding visual tracking algorithms in recent years,it has shown great performance using discriminative correlation filter(DCF)together with HOG,color maps and VGGNet features.Inspired by new deep learning models,this paper propose a hybrid efficient convolution operators integrating fully convolution network(FCN)and residual network(ResNet)for visual tracking,where FCN and ResNet are introduced in our proposed method to segment the objects from backgrounds and extract hierarchical feature maps of objects,respectively.Compared with the traditional VGGNet,our approach has higher accuracy for dealing with the issues of segmentation and image size.The experiments show that our approach would obtain better performance than ECO in terms of precision plot and success rate plot on OTB-2013 and UAV123 datasets.展开更多
Power flow adjustment is a sequential decision problem.The operator makes decisions to ensure that the power flow meets the system's operational constraints,thereby obtaining a typical operating mode power flow.Ho...Power flow adjustment is a sequential decision problem.The operator makes decisions to ensure that the power flow meets the system's operational constraints,thereby obtaining a typical operating mode power flow.However,this decision-making method relies heavily on human experience,which is inefficient when the system is complex.In addition,the results given by the current evaluation system are difficult to directly guide the intelligent power flow adjustment.In order to improve the efficiency and intelligence of power flow adjustment,this paper proposes a power flow adjustment method based on deep reinforcement learning.Combining deep reinforcement learning theory with traditional power system operation mode analysis,the concept of region mapping is proposed to describe the adjustment process,so as to analyze the process of power flow calculation and manual adjustment.Considering the characteristics of power flow adjustment,a Markov decision process model suitable for power flow adjustment is constructed.On this basis,a double Q network learning method suitable for power flow adjustment is proposed.This method can adjust the power flow according to the set adjustment route,thus improving the intelligent level of power flow adjustment.The method in this paper is tested on China Electric Power Research Institute(CEPRI)test system.展开更多
Lithium-ion batteries are key drivers of the renewable energy revolution,bolstered by progress in battery design,modelling,and management.Yet,achieving high-performance battery health prognostics is a significant chal...Lithium-ion batteries are key drivers of the renewable energy revolution,bolstered by progress in battery design,modelling,and management.Yet,achieving high-performance battery health prognostics is a significant challenge.With the availability of open data and software,coupled with automated simulations,deep learning has become an integral component of battery health prognostics.We offer a comprehensive overview of potential deep learning techniques specifically designed for modeling and forecasting the dynamics of multiphysics and multiscale battery systems.Following this,we provide a concise summary of publicly available lithium-ion battery test and cycle datasets.By providing illustrative examples,we emphasize the efficacy of five techniques capable of enhancing deep learning for accurate battery state prediction and health-focused management.Each of these techniques offers unique benefits.(1)Transformer models address challenges using self-attention mechanisms and positional encoding methods.(2) Transfer learning improves learning tasks within a target domain by leveraging knowledge from a source domain.(3) Physics-informed learning uses prior knowledge to enhance learning algorithms.(4)Generative adversarial networks(GANs) earn praise for their ability to generate diverse and high-quality outputs,exhibiting outstanding performance with complex datasets.(5) Deep reinforcement learning enables an agent to make optimal decisions through continuous interactions with its environment,thus maximizing cumulative rewards.In this Review,we highlight examples that employ these techniques for battery health prognostics,summarizing both their challenges and opportunities.These methodologies offer promising prospects for researchers and industry professionals,enabling the creation of specialized network architectures that autonomously extract features,especially for long-range spatial-temporal connections across extended timescales.The outcomes could include improved accuracy,faster training,and enhanced generalization.展开更多
为在稀疏测点超孔隙水压力数据条件下预测饱和软土的固结行为,引入物理信息深度算子网络(physics-informed deep operator network,PI-DeepONet)方法,并利用稀疏孔隙水压力测点数据对饱和土体全域内超孔隙水压力分布进行实时预测。通过...为在稀疏测点超孔隙水压力数据条件下预测饱和软土的固结行为,引入物理信息深度算子网络(physics-informed deep operator network,PI-DeepONet)方法,并利用稀疏孔隙水压力测点数据对饱和土体全域内超孔隙水压力分布进行实时预测。通过分析常规黏土变形固结及软弱黏土大变形固结2个实例进行预测,引入相对L2误差和R2这2个评估指标,验证了PI-DeepONet算法在预测全域超孔隙水压力演化方面的性能,并与纯数据驱动的DeepONet算法的计算结果进行了对比。预测结果表明:在相同的测点数目和各测点拥有相同超孔隙水压力数据量的条件下,DeepONet算法对全域超孔隙水压力的预测绝对误差在10^(-2)~10^(-1)左右,而PI-DeepONet算法的绝对误差范围则在10^(−3)~10^(-2)左右,表现出更好的预测效果。其次,在常规黏土变形固结行为研究中,通过对超孔隙水压力数据添加3种不同噪声水平来模拟现场监测环境,观察到即使噪声水平达到5%,PI-DeepONet算法仍能在水压力数据稀疏且带噪声的条件下提供高质量的全域超孔隙水压力实时预测。最后,在软弱黏土大变形固结行为研究中,将PI-DeepONet算法运用于上下边界排水速率不同的固结问题中,发现训练好的一维模型在单一测点条件下,能对其他界面参数条件下饱和土体全域内超孔隙水压力分布规律进行准确预测,表明PIDeepONet算法能为岩土工程中相关问题提供新的解决办法。展开更多
Delivering energy flexibility at the district scale entails coordinating control actions across many buildings to shape aggregate demand;this coordination depends on training and deploying control policies and optimiz...Delivering energy flexibility at the district scale entails coordinating control actions across many buildings to shape aggregate demand;this coordination depends on training and deploying control policies and optimization routines,which in turn require predictive models that can be queried efficiently over large building clusters.However,conventional physics-based simulators are computationally prohibitive for large-scale control training,and simple data-driven surrogates often lack the generalization needed for heterogeneous clusters.This paper introduces ScaleONet,a deep operator network framework designed for scalable,control-oriented modeling of building-cluster thermal dynamics.ScaleONet leverages the DeepONet paradigm to decouple and share learning across buildings:an LSTM-based branch network encodes outdoor climate and individual HVAC control signals,while a multilayer perceptron(MLP)-based trunk network embeds prediction timestamps,enabling fast predictions for growing clusters with negligible extra cost for each additional building or timestep.To the authors’knowledge,this is the first operator-learning method tailored to indoor air temperature forecasting in heterogeneous building clusters.Validation on thirty Belgian buildings(GenkNet)simulated in Dymola shows that,although a non-operator-learning LSTM baseline slightly outperforms ScaleONet for single-building cases,its error grows monotonically with cluster size.In contrast,ScaleONet’s median per-building-per-day RMSE decreases from 0.59°C at three buildings to 0.53°C at ten and 0.47°C at thirty,compared to 0.95°C for the LSTM at thirty buildings-a 51%reduction in prediction error.Error analysis across envelope heat-loss coefficients(UAbuilding)further reveals that while the LSTM’s RMSE increases for high-𝑈𝐴structures,ScaleONet maintains uniformly low error.With millisecond-scale inference(approximately 4 ms per sample for thirty buildings),ScaleONet is well suited for large-scale reinforcement learning,receding-horizon optimization,and real-time model predictive control.展开更多
Optical solitons play an important role in long-distance,high-capacity communications.To enhance the precision of soliton dynamics modeling,the authors combine incremental learning techniques with physics-informed neu...Optical solitons play an important role in long-distance,high-capacity communications.To enhance the precision of soliton dynamics modeling,the authors combine incremental learning techniques with physics-informed neural network.The novel model employs a process of knowledge distillation and fine-tuning to continually integrate fresh physical information into the neural network.This iterative approach leads to a constant improvement in the network's ability to extract features.The authors conduct experiments on three solitons,and the new method significantly reduces the error compared to the general physics-informed neural network.The modeling approach put forward in this research is anticipated to contribute to the advancement of all-optical computing research and facilitate the development of novel fiber optic communication systems.展开更多
Surface wave tomography based on dispersion is an important approach for imaging the velocity structure of the Earth's crust and upper mantle.Traditional surface wave tomography methods based on dispersion data ty...Surface wave tomography based on dispersion is an important approach for imaging the velocity structure of the Earth's crust and upper mantle.Traditional surface wave tomography methods based on dispersion data typically involve a multistep process:initial construction of 2D phase/group velocity maps,a point-wise inversion of dispersion data to derive 1D profiles of shear wave velocity as a function of depth at each grid point,and final construction of the 3D velocity model.However,conventional 2D tomography methods have certain limitations.For instance,linearized inversion methods are highly sensitive to the choice of the initial velocity model and regularization parameters,while eikonal tomography method requires dense observations.Here,we propose a surface wave tomography method based on physics-informed neural networks,which can construct the phase/group velocity maps of multiple measurement periods simultaneously,eliminating the need for repeated individual inversions for each period.The network comprises two branches,one taking in the coordinates of the virtual source and station as well as period as input to fit the observed surface wave travel times,and another one taking in the station coordinates and period to predict the phase/group velocity.The two branches are constrained by the eikonal equation.After the training is completed,the velocity of each grid point in each period can be queried using the neural network and form the phase/group velocity maps.We test the new method using data from the Feidong and the Weifang dense seismic arrays deployed around the Tanlu Fault Zone in eastern China,and compare the tomography results with those of the traditional method.We demonstrate that the new method is a meshless tomography approach with data adaptive resolution.In addition,it does not require an initial velocity model or explicit regularizations.This method is highly automated,simple,and user-friendly,and it has great potential for integration with existing automatic dispersion curve extraction techniques to achieve automated surface wave tomography without human intervention.展开更多
基金supported by the National Natural Science Foundation of China(Grant No.92152301)the National Key Research and Development Program of China(Grant No.2022YFB4300200)the Shaanxi Provincial Key Research and Development Program(Grant No.2023-ZDLGY-27).
文摘Physics-informed neural networks(PINNs)have shown remarkable prospects in solving the forward and inverse problems involving partial differential equations(PDEs).The method embeds PDEs into the neural network by calculating the PDE loss at a set of collocation points,providing advantages such as meshfree and more convenient adaptive sampling.However,when solving PDEs using nonuniform collocation points,PINNs still face challenge regarding inefficient convergence of PDE residuals or even failure.In this work,we first analyze the ill-conditioning of the PDE loss in PINNs under nonuniform collocation points.To address the issue,we define volume weighting residual and propose volume weighting physics-informed neural networks(VW-PINNs).Through weighting the PDE residuals by the volume that the collocation points occupy within the computational domain,we embed explicitly the distribution characteristics of collocation points in the loss evaluation.The fast and sufficient convergence of the PDE residuals for the problems involving nonuniform collocation points is guaranteed.Considering the meshfree characteristics of VW-PINNs,we also develop a volume approximation algorithm based on kernel density estimation to calculate the volume of the collocation points.We validate the universality of VW-PINNs by solving the forward problems involving flow over a circular cylinder and flow over the NACA0012 airfoil under different inflow conditions,where conventional PINNs fail.By solving the Burgers’equation,we verify that VW-PINNs can enhance the efficiency of existing the adaptive sampling method in solving the forward problem by three times,and can reduce the relative L 2 error of conventional PINNs in solving the inverse problem by more than one order of magnitude.
基金support in part by China Postdoctoral Science Foundation (No.2021M702634)National Science Foundation of China (No.52175116).
文摘Prognosis of bearing is critical to improve the safety,reliability,and availability of machinery systems,which provides the health condition assessment and determines how long the machine would work before failure occurs by predicting the remaining useful life(RUL).In order to overcome the drawback of pure data-driven methods and predict RUL accurately,a novel physics-informed deep neural network,named degradation consistency recurrent neural network,is proposed for RUL prediction by integrating the natural degradation knowledge of mechanical components.The degradation is monotonic over the whole life of bearings,which is characterized by temperature signals.To incorporate the knowledge of monotonic degradation,a positive increment recurrence relationship is introduced to keep the monotonicity.Thus,the proposed model is relatively well understood and capable to keep the learning process consistent with physical degradation.The effectiveness and merit of the RUL prediction using the proposed method are demonstrated through vibration signals collected from a set of run-to-failure tests.
文摘Physics-informed deep learning has drawn tremendous interest in recent years to solve computational physics problems,whose basic concept is to embed physical laws to constrain/inform neural networks,with the need of less data for training a reliable model.This can be achieved by incorporating the residual of physics equations into the loss function.Through minimizing the loss function,the network could approximate the solution.In this paper,we propose a mixed-variable scheme of physics-informed neural network(PINN)for fluid dynamics and apply it to simulate steady and transient laminar flows at low Reynolds numbers.A parametric study indicates that the mixed-variable scheme can improve the PINN trainability and the solution accuracy.The predicted velocity and pressure fields by the proposed PINN approach are also compared with the reference numerical solutions.Simulation results demonstrate great potential of the proposed PINN for fluid flow simulation with a high accuracy.
文摘This paper proposes a physics-informed neural network(PINN)framework to analyze the nonlinear buckling behavior of a three-dimensional(3D)FG porous,slender beam resting on a Winkler-Pasternak foundation.PINNs need much less training data to obtain high accuracy using a straightforward network.The powerful tool used in this work can handle any class of PDEs.We use the deep learning platform TensorFlow and DeepXDE library to design our network.In this study,the PINNs framework takes information from the governing differential equations of the beam system and the data from boundary conditions and outputs the critical nonlinear buckling load.The mathematical model is developed using Hamilton’s principle,considering geometry’s nonlinearity.The accuracy of the modeling framework is carefully examined by applying it to various boundary condition cases as well as the physical parameters such as 3D FG indexes on the nonlinear mechanical behaviors.Finally,the PINNs results are validated with those extracted from the generalized differential quadrature method(GDQM).It is found that the proposed PINN framework can characterize the nonlinear buckling behavior of 3D FG porous,slender beams with satisfactory accuracy.Furthermore,PINN is presented to accurately predict the nonlinear buckling behavior of the beam up to 71 times faster than the numerical method.
基金supported by the National Natural Science Foundation of China(Nos.12072105 and 11932006)。
文摘Recently,numerous studies have demonstrated that the physics-informed neural network(PINN)can effectively and accurately resolve hyperelastic finite deformation problems.In this paper,a PINN framework for tackling hyperelastic-magnetic coupling problems is proposed.Since the solution space consists of two-phase domains,two separate networks are constructed to independently predict the solution for each phase region.In addition,a conscious point allocation strategy is incorporated to enhance the prediction precision of the PINN in regions characterized by sharp gradients.With the developed framework,the magnetic fields and deformation fields of magnetorheological elastomers(MREs)are solved under the control of hyperelastic-magnetic coupling equations.Illustrative examples are provided and contrasted with the reference results to validate the predictive accuracy of the proposed framework.Moreover,the advantages of the proposed framework in solving hyperelastic-magnetic coupling problems are validated,particularly in handling small data sets,as well as its ability in swiftly and precisely forecasting magnetostrictive motion.
基金This work is funded by National Natural Science Foundation of China(Nos.42202292,42141011)the Program for Jilin University(JLU)Science and Technology Innovative Research Team(No.2019TD-35).The authors would also like to thank the reviewers and editors whose critical comments are very helpful in preparing this article.
文摘To reduce CO_(2) emissions in response to global climate change,shale reservoirs could be ideal candidates for long-term carbon geo-sequestration involving multi-scale transport processes.However,most current CO_(2) sequestration models do not adequately consider multiple transport mechanisms.Moreover,the evaluation of CO_(2) storage processes usually involves laborious and time-consuming numerical simulations unsuitable for practical prediction and decision-making.In this paper,an integrated model involving gas diffusion,adsorption,dissolution,slip flow,and Darcy flow is proposed to accurately characterize CO_(2) storage in depleted shale reservoirs,supporting the establishment of a training database.On this basis,a hybrid physics-informed data-driven neural network(HPDNN)is developed as a deep learning surrogate for prediction and inversion.By incorporating multiple sources of scientific knowledge,the HPDNN can be configured with limited simulation resources,significantly accelerating the forward and inversion processes.Furthermore,the HPDNN can more intelligently predict injection performance,precisely perform reservoir parameter inversion,and reasonably evaluate the CO_(2) storage capacity under complicated scenarios.The validation and test results demonstrate that the HPDNN can ensure high accuracy and strong robustness across an extensive applicability range when dealing with field data with multiple noise sources.This study has tremendous potential to replace traditional modeling tools for predicting and making decisions about CO_(2) storage projects in depleted shale reservoirs.
基金The Key R&D Project of Jilin Province,Grant/Award Number:20230201067GX。
文摘Extracting typical operational scenarios is essential for making flexible decisions in the dispatch of a new power system.A novel deep time series aggregation scheme(DTSAs)is proposed to generate typical operational scenarios,considering the large amount of historical operational snapshot data.Specifically,DTSAs analyse the intrinsic mechanisms of different scheduling operational scenario switching to mathematically represent typical operational scenarios.A Gramian angular summation field-based operational scenario image encoder was designed to convert operational scenario sequences into highdimensional spaces.This enables DTSAs to fully capture the spatiotemporal characteristics of new power systems using deep feature iterative aggregation models.The encoder also facilitates the generation of typical operational scenarios that conform to historical data distributions while ensuring the integrity of grid operational snapshots.Case studies demonstrate that the proposed method extracted new fine-grained power system dispatch schemes and outperformed the latest high-dimensional feature-screening methods.In addition,experiments with different new energy access ratios were conducted to verify the robustness of the proposed method.DTSAs enable dispatchers to master the operation experience of the power system in advance,and actively respond to the dynamic changes of the operation scenarios under the high access rate of new energy.
基金funded by Vietnam National Foundation for Science and Technology Development (NAFOSTED)under Grant No.105.08-2019.03.
文摘Flash floods are one of the most dangerous natural disasters,especially in hilly terrain,causing loss of life,property,and infrastructures and sudden disruption of traffic.These types of floods are mostly associated with landslides and erosion of roads within a short time.Most of Vietnamis hilly and mountainous;thus,the problem due to flash flood is severe and requires systematic studies to correctly identify flood susceptible areas for proper landuse planning and traffic management.In this study,three Machine Learning(ML)methods namely Deep Learning Neural Network(DL),Correlation-based FeatureWeighted Naive Bayes(CFWNB),and Adaboost(AB-CFWNB)were used for the development of flash flood susceptibility maps for hilly road section(115 km length)of National Highway(NH)-6 inHoa Binh province,Vietnam.In the proposedmodels,88 past flash flood events were used together with 14 flash floods affecting topographical and geo-environmental factors.The performance of themodels was evaluated using standard statisticalmeasures including Receiver Operating Characteristic(ROC)Curve,Area Under Curve(AUC)and Root Mean Square Error(RMSE).The results revealed that all the models performed well(AUC>0.80)in predicting flash flood susceptibility zones,but the performance of the DL model is the best(AUC:0.972,RMSE:0.352).Therefore,the DL model can be applied to develop an accurate flash flood susceptibility map of hilly terrain which can be used for proper planning and designing of the highways and other infrastructure facilities besides landuse management of the area.
基金supported by the National Natural Science Foundation of China (Grant No.92152301)。
文摘In recent years,physics-informed neural networks(PINNs) have become a representative method for solving partial differential equations(PDEs) with neural networks.PINNs provide a novel approach to solving PDEs through optimization algorithms,offering a unified framework for solving both forward and inverse problems.However,some limitations in terms of solution accuracy and generality have also been revealed.This paper systematically summarizes the limitations of PINNs and identifies three root causes for their failure in solving PDEs:(1) poor multiscale approximation ability and ill-conditioning caused by PDE losses;(2) insufficient exploration of convergence and error analysis,resulting in weak mathematical rigor;(3) inadequate integration of physical information,causing mismatch between residuals and iteration errors.By focusing on addressing these limitations in PINNs,we outline the future directions and prospects for the intelligent computing of PDEs:(1) analysis of illconditioning in PINNs and mitigation strategies;(2) improvements to PINNs by enforcing temporal causality;(3) empowering PINNs with classical numerical methods.
文摘Visual tracking is a classical computer vision problem with many applications.Efficient convolution operators(ECO)is one of the most outstanding visual tracking algorithms in recent years,it has shown great performance using discriminative correlation filter(DCF)together with HOG,color maps and VGGNet features.Inspired by new deep learning models,this paper propose a hybrid efficient convolution operators integrating fully convolution network(FCN)and residual network(ResNet)for visual tracking,where FCN and ResNet are introduced in our proposed method to segment the objects from backgrounds and extract hierarchical feature maps of objects,respectively.Compared with the traditional VGGNet,our approach has higher accuracy for dealing with the issues of segmentation and image size.The experiments show that our approach would obtain better performance than ECO in terms of precision plot and success rate plot on OTB-2013 and UAV123 datasets.
文摘Power flow adjustment is a sequential decision problem.The operator makes decisions to ensure that the power flow meets the system's operational constraints,thereby obtaining a typical operating mode power flow.However,this decision-making method relies heavily on human experience,which is inefficient when the system is complex.In addition,the results given by the current evaluation system are difficult to directly guide the intelligent power flow adjustment.In order to improve the efficiency and intelligence of power flow adjustment,this paper proposes a power flow adjustment method based on deep reinforcement learning.Combining deep reinforcement learning theory with traditional power system operation mode analysis,the concept of region mapping is proposed to describe the adjustment process,so as to analyze the process of power flow calculation and manual adjustment.Considering the characteristics of power flow adjustment,a Markov decision process model suitable for power flow adjustment is constructed.On this basis,a double Q network learning method suitable for power flow adjustment is proposed.This method can adjust the power flow according to the set adjustment route,thus improving the intelligent level of power flow adjustment.The method in this paper is tested on China Electric Power Research Institute(CEPRI)test system.
文摘Lithium-ion batteries are key drivers of the renewable energy revolution,bolstered by progress in battery design,modelling,and management.Yet,achieving high-performance battery health prognostics is a significant challenge.With the availability of open data and software,coupled with automated simulations,deep learning has become an integral component of battery health prognostics.We offer a comprehensive overview of potential deep learning techniques specifically designed for modeling and forecasting the dynamics of multiphysics and multiscale battery systems.Following this,we provide a concise summary of publicly available lithium-ion battery test and cycle datasets.By providing illustrative examples,we emphasize the efficacy of five techniques capable of enhancing deep learning for accurate battery state prediction and health-focused management.Each of these techniques offers unique benefits.(1)Transformer models address challenges using self-attention mechanisms and positional encoding methods.(2) Transfer learning improves learning tasks within a target domain by leveraging knowledge from a source domain.(3) Physics-informed learning uses prior knowledge to enhance learning algorithms.(4)Generative adversarial networks(GANs) earn praise for their ability to generate diverse and high-quality outputs,exhibiting outstanding performance with complex datasets.(5) Deep reinforcement learning enables an agent to make optimal decisions through continuous interactions with its environment,thus maximizing cumulative rewards.In this Review,we highlight examples that employ these techniques for battery health prognostics,summarizing both their challenges and opportunities.These methodologies offer promising prospects for researchers and industry professionals,enabling the creation of specialized network architectures that autonomously extract features,especially for long-range spatial-temporal connections across extended timescales.The outcomes could include improved accuracy,faster training,and enhanced generalization.
文摘为在稀疏测点超孔隙水压力数据条件下预测饱和软土的固结行为,引入物理信息深度算子网络(physics-informed deep operator network,PI-DeepONet)方法,并利用稀疏孔隙水压力测点数据对饱和土体全域内超孔隙水压力分布进行实时预测。通过分析常规黏土变形固结及软弱黏土大变形固结2个实例进行预测,引入相对L2误差和R2这2个评估指标,验证了PI-DeepONet算法在预测全域超孔隙水压力演化方面的性能,并与纯数据驱动的DeepONet算法的计算结果进行了对比。预测结果表明:在相同的测点数目和各测点拥有相同超孔隙水压力数据量的条件下,DeepONet算法对全域超孔隙水压力的预测绝对误差在10^(-2)~10^(-1)左右,而PI-DeepONet算法的绝对误差范围则在10^(−3)~10^(-2)左右,表现出更好的预测效果。其次,在常规黏土变形固结行为研究中,通过对超孔隙水压力数据添加3种不同噪声水平来模拟现场监测环境,观察到即使噪声水平达到5%,PI-DeepONet算法仍能在水压力数据稀疏且带噪声的条件下提供高质量的全域超孔隙水压力实时预测。最后,在软弱黏土大变形固结行为研究中,将PI-DeepONet算法运用于上下边界排水速率不同的固结问题中,发现训练好的一维模型在单一测点条件下,能对其他界面参数条件下饱和土体全域内超孔隙水压力分布规律进行准确预测,表明PIDeepONet算法能为岩土工程中相关问题提供新的解决办法。
基金supported by KU Leuven,Belgium through the TECHPED-C2 project(C24M/21/021)which investigates tech-nically feasible and effective solutions for Positive Energy Districts.Additional support was provided by the National University of Singa-pore,Singapore through the Start-Up Grant(A-0009876-00-00)the Ministry of Education,Singapore,under the Academic Research Fund Tier 1(A-8003235-00-00).
文摘Delivering energy flexibility at the district scale entails coordinating control actions across many buildings to shape aggregate demand;this coordination depends on training and deploying control policies and optimization routines,which in turn require predictive models that can be queried efficiently over large building clusters.However,conventional physics-based simulators are computationally prohibitive for large-scale control training,and simple data-driven surrogates often lack the generalization needed for heterogeneous clusters.This paper introduces ScaleONet,a deep operator network framework designed for scalable,control-oriented modeling of building-cluster thermal dynamics.ScaleONet leverages the DeepONet paradigm to decouple and share learning across buildings:an LSTM-based branch network encodes outdoor climate and individual HVAC control signals,while a multilayer perceptron(MLP)-based trunk network embeds prediction timestamps,enabling fast predictions for growing clusters with negligible extra cost for each additional building or timestep.To the authors’knowledge,this is the first operator-learning method tailored to indoor air temperature forecasting in heterogeneous building clusters.Validation on thirty Belgian buildings(GenkNet)simulated in Dymola shows that,although a non-operator-learning LSTM baseline slightly outperforms ScaleONet for single-building cases,its error grows monotonically with cluster size.In contrast,ScaleONet’s median per-building-per-day RMSE decreases from 0.59°C at three buildings to 0.53°C at ten and 0.47°C at thirty,compared to 0.95°C for the LSTM at thirty buildings-a 51%reduction in prediction error.Error analysis across envelope heat-loss coefficients(UAbuilding)further reveals that while the LSTM’s RMSE increases for high-𝑈𝐴structures,ScaleONet maintains uniformly low error.With millisecond-scale inference(approximately 4 ms per sample for thirty buildings),ScaleONet is well suited for large-scale reinforcement learning,receding-horizon optimization,and real-time model predictive control.
基金supported by the National Key R&D Program of China under Grant No.2022YFA1604200the National Natural Science Foundation of China under Grant Nos.11975001 and 12075034+1 种基金the Beijing Natural Science Foundation under Grant No.JQ21019the Hebei Key Laboratory of Physics and Energy Technology under Grant No.HBKLPET202303。
文摘Optical solitons play an important role in long-distance,high-capacity communications.To enhance the precision of soliton dynamics modeling,the authors combine incremental learning techniques with physics-informed neural network.The novel model employs a process of knowledge distillation and fine-tuning to continually integrate fresh physical information into the neural network.This iterative approach leads to a constant improvement in the network's ability to extract features.The authors conduct experiments on three solitons,and the new method significantly reduces the error compared to the general physics-informed neural network.The modeling approach put forward in this research is anticipated to contribute to the advancement of all-optical computing research and facilitate the development of novel fiber optic communication systems.
基金supported by the Scientific Research Foundation for High-level Talents of Anhui University of Science and Technology(Grant No.2024yjrc64)the National Natural Science Foundation of China(Grant No.42230101)the Deep Earth Probe and Mineral Resources Exploration-National Science and Technology Major Project(Grant No.2025ZD1004903)。
文摘Surface wave tomography based on dispersion is an important approach for imaging the velocity structure of the Earth's crust and upper mantle.Traditional surface wave tomography methods based on dispersion data typically involve a multistep process:initial construction of 2D phase/group velocity maps,a point-wise inversion of dispersion data to derive 1D profiles of shear wave velocity as a function of depth at each grid point,and final construction of the 3D velocity model.However,conventional 2D tomography methods have certain limitations.For instance,linearized inversion methods are highly sensitive to the choice of the initial velocity model and regularization parameters,while eikonal tomography method requires dense observations.Here,we propose a surface wave tomography method based on physics-informed neural networks,which can construct the phase/group velocity maps of multiple measurement periods simultaneously,eliminating the need for repeated individual inversions for each period.The network comprises two branches,one taking in the coordinates of the virtual source and station as well as period as input to fit the observed surface wave travel times,and another one taking in the station coordinates and period to predict the phase/group velocity.The two branches are constrained by the eikonal equation.After the training is completed,the velocity of each grid point in each period can be queried using the neural network and form the phase/group velocity maps.We test the new method using data from the Feidong and the Weifang dense seismic arrays deployed around the Tanlu Fault Zone in eastern China,and compare the tomography results with those of the traditional method.We demonstrate that the new method is a meshless tomography approach with data adaptive resolution.In addition,it does not require an initial velocity model or explicit regularizations.This method is highly automated,simple,and user-friendly,and it has great potential for integration with existing automatic dispersion curve extraction techniques to achieve automated surface wave tomography without human intervention.
