Physics-informed neural networks(PINNs)have emerged as a promising class of scientific machine learning techniques that integrate governing physical laws into neural network training.Their ability to enforce different...Physics-informed neural networks(PINNs)have emerged as a promising class of scientific machine learning techniques that integrate governing physical laws into neural network training.Their ability to enforce differential equations,constitutive relations,and boundary conditions within the loss function provides a physically grounded alternative to traditional data-driven models,particularly for solid and structural mechanics,where data are often limited or noisy.This review offers a comprehensive assessment of recent developments in PINNs,combining bibliometric analysis,theoretical foundations,application-oriented insights,and methodological innovations.A biblio-metric survey indicates a rapid increase in publications on PINNs since 2018,with prominent research clusters focused on numerical methods,structural analysis,and forecasting.Building upon this trend,the review consolidates advance-ments across five principal application domains,including forward structural analysis,inverse modeling and parameter identification,structural and topology optimization,assessment of structural integrity,and manufacturing processes.These applications are propelled by substantial methodological advancements,encompassing rigorous enforcement of boundary conditions,modified loss functions,adaptive training,domain decomposition strategies,multi-fidelity and transfer learning approaches,as well as hybrid finite element–PINN integration.These advances address recurring challenges in solid mechanics,such as high-order governing equations,material heterogeneity,complex geometries,localized phenomena,and limited experimental data.Despite remaining challenges in computational cost,scalability,and experimental validation,PINNs are increasingly evolving into specialized,physics-aware tools for practical solid and structural mechanics applications.展开更多
The S38C railway axle undergoes induction hardening,resulting in a gradient-distributed microstructure and mechanical properties.The accurate identification of gradient-distributed plastic parameters for the S38C axle...The S38C railway axle undergoes induction hardening,resulting in a gradient-distributed microstructure and mechanical properties.The accurate identification of gradient-distributed plastic parameters for the S38C axle remains a challenging task.To tackle this challenge,the present study proposes a novel approach for identifying the gradient-distributed plastic parameters for the S38C axle by integrating nano-indentation techniques with the machine learning method.Firstly,nano-indentation tests are conducted along the radial direction of the S38C axle to obtain the gradient-distributed load-displacement curves,nano-hardness,and elastic modulus.Subsequently,the dimensionless analysis is performed to obtain the representative stress,strain,and yield stress from load-displacement curves.These parameters are then incorporated into the machine learning method as physical information to identify the gradient-distributed plastic parameters of the S38C axle.The results indicate that the proposed method based on the physics-informed neural network and multi-fidelity neural network successfully identifies the gradient-distributed plastic parameters of the S38C axles and demonstrates superior prediction accuracy and generalization compared with the purely data-driven machine learning method.展开更多
The growing use of lithium-ion batteries in electric transportation and grid-scale storage systems has intensified the need for accurate and highly generalizable state-of-health(SOH)estimation.Conventional approaches ...The growing use of lithium-ion batteries in electric transportation and grid-scale storage systems has intensified the need for accurate and highly generalizable state-of-health(SOH)estimation.Conventional approaches often suffer from reduced accuracy under dynamically uncertain state-of-charge(SOC)operating ranges and heterogeneous aging stresses.This study presents a unified SOH estimation framework that integrates physics-informed modeling,subspace identification,and Transformer-based learning.A reduced-order model is derived from simplified electrochemical dynamics,providing an interpretable and computationally efficient representation of battery behavior.Subspace identification across a wide SOC and SOH range yields degradation-sensitive features,which the Transformer uses to capture long-range aging dynamics via multi-head self-attention.Experiments on LiFePO4 cells under joint-cell training show consistently accurate SOH estimation,with a maximum error of 1.39%,demonstrating the framework’s effectiveness in decoupling SOC and SOH effects.In cross-cell validation,where training and validation are performed on different cells,the model maintains a maximum error of 2.06%,confirming strong generalization to unseen aging trajectories.Comparative experiments on LiFePO_(4)and public LiCoO_(2)datasets confirm the framework’s cross-chemistry applicability.By extracting low-dimensional,physically interpretable features via subspace identification,the framework significantly reduces training cost while maintaining high SOH estimation accuracy,outperforming conventional data-driven models lacking physical guidance.展开更多
Accurate state of health(SOH)estimation is essential for the safe and reliable operation of lithium-ion batteries.However,existing methods face significant challenges,primarily because they rely on complete charge–di...Accurate state of health(SOH)estimation is essential for the safe and reliable operation of lithium-ion batteries.However,existing methods face significant challenges,primarily because they rely on complete charge–discharge cycles and fixed-form physical constraints,which limit adaptability to different chemistries and real-world conditions.To address these issues,this study proposes an approach that extracts features from segmented state of charge(SOC)intervals and integrates them into an enhanced physics-informed neural network(PINN).Specifically,voltage data within the 25%–75%SOC range during charging are used to derive statistical,time–frequency,and mechanism-based features that capture degradation trends.A hybrid PINN-Lasso-Transformer-BiLSTM architecture is developed,where Lasso regression enables sparse feature selection,and a nonlinear empirical degradation model is embedded as a learnable physical term within a dynamically scaled composite loss.This design adaptively balances data-driven accuracy with physical consistency,thereby enhancing estimation precision,robustness,and generalization.The results show that the proposed method outperforms conventional neural networks across four battery chemistries,achieving root mean square error and mean absolute error below 1%.Notably,features from partial charging segments exhibit higher robustness than those from full cycles.Furthermore,the model maintains strong performance under high temperatures and demonstrates excellent generalization capacity in transfer learning across chemistries,temperatures,and C-rates.This work establishes a scalable and interpretable solution for accurate SOH estimation under diverse practical operating conditions.展开更多
Physics-informed neural networks(PINNs)have been shown as powerful tools for solving partial differential equations(PDEs)by embedding physical laws into the network training.Despite their remarkable results,complicate...Physics-informed neural networks(PINNs)have been shown as powerful tools for solving partial differential equations(PDEs)by embedding physical laws into the network training.Despite their remarkable results,complicated problems such as irregular boundary conditions(BCs)and discontinuous or high-frequency behaviors remain persistent challenges for PINNs.For these reasons,we propose a novel two-phase framework,where a neural network is first trained to represent shape functions that can capture the irregularity of BCs in the first phase,and then these neural network-based shape functions are used to construct boundary shape functions(BSFs)that exactly satisfy both essential and natural BCs in PINNs in the second phase.This scheme is integrated into both the strong-form and energy PINN approaches,thereby improving the quality of solution prediction in the cases of irregular BCs.In addition,this study examines the benefits and limitations of these approaches in handling discontinuous and high-frequency problems.Overall,our method offers a unified and flexible solution framework that addresses key limitations of existing PINN methods with higher accuracy and stability for general PDE problems in solid mechanics.展开更多
Accelerated aging tests are widely used to rapidly evaluate the durability of materials,of which thermal-oxidative aging is the most common approach.To quantitatively predict the effects of multiple coupled factors,th...Accelerated aging tests are widely used to rapidly evaluate the durability of materials,of which thermal-oxidative aging is the most common approach.To quantitatively predict the effects of multiple coupled factors,this study takes polyamide66 reinforced with glass fiber(PA66-GF)as a model system and proposed a high-precision paradigm for coupled thermal-oxidative aging.By integrating Arrhenius-type reaction kinetics with oxygen diffusion,a predictive formula that holistically captures the nonlinear synergistic effects of multiple factors was developed,thereby overcoming the limitations of traditional single-variable models.A systematic evaluation of the stepwise improved formulas through nonlinear fitting showed that the coefficient of determination(R^(2))increased from 0.223 to 0.803,elucidating the fundamental reason why conventional approaches fail in quantitative prediction.These formulae were further embedded as physical constraints into a physics-informed neural network(PINN),which further enhanced the predictive performance,with the proposed formula achieving a peak R^(2)of 0.946.The results highlight that robust data fitting alone is insufficient;the decisive factor for the success of PINN lies in whether the embedded formula faithfully reflects the underlying physical mechanisms.When applied to polyamide 6 reinforced with glass fiber(PA6-GF),the Formula-constrained PINN maintained a high level of accuracy(R^(2)=0.916),demonstrating its strong cross-system generalizability.In summary,this work establishes a robust hybrid physics-machine learning framework that combines high accuracy with transferability for predicting the thermal-oxidative aging behavior of composite material systems.展开更多
Physics-informed neural networks(PINNs)are vital for machine learning and exhibit significant advantages when handling complex physical problems.The PINN method can rapidly predict ^(220)Rn progeny concentration and i...Physics-informed neural networks(PINNs)are vital for machine learning and exhibit significant advantages when handling complex physical problems.The PINN method can rapidly predict ^(220)Rn progeny concentration and is very important for regulating and measuring this property.To construct a PINN model,training data are typically preprocessed;however,this approach changes the physical characteristics of the data,with the preprocessed data potentially no longer directly conforming to the original physical equations.As a result,the original physical equations cannot be directly employed in the PINN.Consequently,an effective method for transforming physical equations is crucial for accurately constraining PINNs to model the ^(220)Rn progeny concentration prediction.This study presents an equation adaptation approach for neural networks,which is designed to improve prediction of ^(220)Rn progeny concentration.Five neural network models based on three architectures are established:a classical network,a physics-informed network without equation adaptation,and a physics-informed network with equation adaptation.The transport equation of the ^(220)Rn progeny concentration is transformed via equation adaption and integrated with the PINN model.The compatibility and robustness of the model with equation adaption is then analyzed.The results show that PINNs with equation adaption converge consistently with classical neural networks in terms of the training and validation loss and achieve the same level of prediction accuracy.This outcome indicates that the proposed method can be integrated into the neural network architecture.Moreover,the prediction performance of classical neural networks declines significantly when interference data are encountered,whereas the PINNs with equation adaption exhibit stable prediction accuracy.This performance demonstrates that the proposed method successfully harnesses the constraining power of physical equations,significantly enhancing the robustness of the resultant PINN models.Thus,the use of a physics-informed network with equation adaption can guarantee accurate prediction of ^(220)Rn progeny concentration.展开更多
The neutron diffusion equation plays a pivotal role in nuclear reactor analysis.Nevertheless,employing the physics-informed neural network(PINN)method for its solution entails certain limitations.Conventional PINN app...The neutron diffusion equation plays a pivotal role in nuclear reactor analysis.Nevertheless,employing the physics-informed neural network(PINN)method for its solution entails certain limitations.Conventional PINN approaches generally utilize a fully connected network(FCN)architecture that is susceptible to overfitting,training instability,and gradient vanishing as the network depth increases.These challenges result in accuracy bottlenecks in the solution.In response to these issues,the residual-based resample physics-informed neural network(R2-PINN)is proposed.It is an improved PINN architecture that replaces the FCN with a convolutional neural network with a shortcut(S-CNN).It incorporates skip connections to facilitate gradient propagation between network layers.Additionally,the incorporation of the residual adaptive resampling(RAR)mechanism dynamically increases the number of sampling points.This,in turn,enhances the spatial representation capabilities and overall predictive accuracy of the model.The experimental results illustrate that our approach significantly improves the convergence capability of the model and achieves high-precision predictions of the physical fields.Compared with conventional FCN-based PINN methods,R 2-PINN effectively overcomes the limitations inherent in current methods.Thus,it provides more accurate and robust solutions for neutron diffusion equations.展开更多
Physics-informed neural networks(PINNs),as a novel artificial intelligence method for solving partial differential equations,are applicable to solve both forward and inverse problems.This study evaluates the performan...Physics-informed neural networks(PINNs),as a novel artificial intelligence method for solving partial differential equations,are applicable to solve both forward and inverse problems.This study evaluates the performance of PINNs in solving the temperature diffusion equation of the seawater across six scenarios,including forward and inverse problems under three different boundary conditions.Results demonstrate that PINNs achieved consistently higher accuracy with the Dirichlet and Neumann boundary conditions compared to the Robin boundary condition for both forward and inverse problems.Inaccurate weighting of terms in the loss function can reduce model accuracy.Additionally,the sensitivity of model performance to the positioning of sampling points varied between different boundary conditions.In particular,the model under the Dirichlet boundary condition exhibited superior robustness to variations in point positions during the solutions of inverse problems.In contrast,for the Neumann and Robin boundary conditions,accuracy declines when points were sampled from identical positions or at the same time.Subsequently,the Argo observations were used to reconstruct the vertical diffusion of seawater temperature in the north-central Pacific for the applicability of PINNs in the real ocean.The PINNs successfully captured the vertical diffusion characteristics of seawater temperature,reflected the seasonal changes of vertical temperature under different topographic conditions,and revealed the influence of topography on the temperature diffusion coefficient.The PINNs were proved effective in solving the temperature diffusion equation of seawater with limited data,providing a promising technique for simulating or predicting ocean phenomena using sparse observations.展开更多
For the diagnostics and health management of lithium-ion batteries,numerous models have been developed to understand their degradation characteristics.These models typically fall into two categories:data-driven models...For the diagnostics and health management of lithium-ion batteries,numerous models have been developed to understand their degradation characteristics.These models typically fall into two categories:data-driven models and physical models,each offering unique advantages but also facing limitations.Physics-informed neural networks(PINNs)provide a robust framework to integrate data-driven models with physical principles,ensuring consistency with underlying physics while enabling generalization across diverse operational conditions.This study introduces a PINN-based approach to reconstruct open circuit voltage(OCV)curves and estimate key ageing parameters at both the cell and electrode levels.These parameters include available capacity,electrode capacities,and lithium inventory capacity.The proposed method integrates OCV reconstruction models as functional components into convolutional neural networks(CNNs)and is validated using a public dataset.The results reveal that the estimated ageing parameters closely align with those obtained through offline OCV tests,with errors in reconstructed OCV curves remaining within 15 mV.This demonstrates the ability of the method to deliver fast and accurate degradation diagnostics at the electrode level,advancing the potential for precise and efficient battery health management.展开更多
The Internet of Things(IoT)ecosystem faces growing security challenges because it is projected to have 76.88 billion devices by 2025 and $1.4 trillion market value by 2027,operating in distributed networks with resour...The Internet of Things(IoT)ecosystem faces growing security challenges because it is projected to have 76.88 billion devices by 2025 and $1.4 trillion market value by 2027,operating in distributed networks with resource limitations and diverse system architectures.The current conventional intrusion detection systems(IDS)face scalability problems and trust-related issues,but blockchain-based solutions face limitations because of their low transaction throughput(Bitcoin:7 TPS(Transactions Per Second),Ethereum:15-30 TPS)and high latency.The research introduces MBID(Multi-Tier Blockchain Intrusion Detection)as a groundbreaking Multi-Tier Blockchain Intrusion Detection System with AI-Enhanced Detection,which solves the problems in huge IoT networks.The MBID system uses a four-tier architecture that includes device,edge,fog,and cloud layers with blockchain implementations and Physics-Informed Neural Networks(PINNs)for edge-based anomaly detection and a dual consensus mechanism that uses Honesty-based Distributed Proof-of-Authority(HDPoA)and Delegated Proof of Stake(DPoS).The system achieves scalability and efficiency through the combination of dynamic sharding and Interplanetary File System(IPFS)integration.Experimental evaluations demonstrate exceptional performance,achieving a detection accuracy of 99.84%,an ultra-low false positive rate of 0.01% with a False Negative Rate of 0.15%,and a near-instantaneous edge detection latency of 0.40 ms.The system demonstrated an aggregate throughput of 214.57 TPS in a 3-shard configuration,providing a clear,evidence-based path for horizontally scaling to support overmillions of devices with exceeding throughput.The proposed architecture represents a significant advancement in blockchain-based security for IoT networks,effectively balancing the trade-offs between scalability,security,and decentralization.展开更多
The packaging quality of coaxial laser diodes(CLDs)plays a pivotal role in determining their optical performance and long-term reliability.As the core packaging process,high-precision laser welding requires precise co...The packaging quality of coaxial laser diodes(CLDs)plays a pivotal role in determining their optical performance and long-term reliability.As the core packaging process,high-precision laser welding requires precise control of process parameters to suppress optical power loss.However,the complex nonlinear relationship between welding parameters and optical power loss renders traditional trial-and-error methods inefficient and imprecise.To address this challenge,a physics-informed(PI)and data-driven collaboration approach for welding parameter optimization is proposed.First,thermal-fluid-solid coupling finite element method(FEM)was employed to quantify the sensitivity of welding parameters to physical characteristics,including residual stress.This analysis facilitated the identification of critical factors contributing to optical power loss.Subsequently,a Gaussian process regression(GPR)model incorporating finite element simulation prior knowledge was constructed based on the selected features.By introducing physics-informed kernel(PIK)functions,stress distribution patterns were embedded into the prediction model,achieving high-precision optical power loss prediction.Finally,a Bayesian optimization(BO)algorithm with an adaptive sampling strategy was implemented for efficient parameter space exploration.Experimental results demonstrate that the proposedmethod effectively establishes explicit physical correlations between welding parameters and optical power loss.The optimized welding parameters reduced optical power loss by 34.1%,providing theoretical guidance and technical support for reliable CLD packaging.展开更多
Energy efficiency stands as an essential factor when implementing deep reinforcement learning(DRL)policies for robotic control systems.Standard algorithms,including Deep Deterministic Policy Gradient(DDPG),primarily o...Energy efficiency stands as an essential factor when implementing deep reinforcement learning(DRL)policies for robotic control systems.Standard algorithms,including Deep Deterministic Policy Gradient(DDPG),primarily optimize task rewards but at the cost of excessively high energy consumption,making them impractical for real-world robotic systems.To address this limitation,we propose Physics-Informed DDPG(PI-DDPG),which integrates physics-based energy penalties to develop energy-efficient yet high-performing control policies.The proposed method introduces adaptive physics-informed constraints through a dynamic weighting factor(λ),enabling policies that balance reward maximization with energy savings.Our motivation is to overcome the impracticality of rewardonly optimization by designing controllers that achieve competitive performance while substantially reducing energy consumption.PI-DDPG was evaluated in nine MuJoCo continuous control environments,where it demonstrated significant improvements in energy efficiency without compromising stability or performance.Experimental results confirm that PI-DDPG substantially reduces energy consumption compared to standard DDPG,while maintaining competitive task performance.For instance,energy costs decreased from 5542.98 to 3119.02 in HalfCheetah-v4 and from1909.13 to 1586.75 in Ant-v4,with stable performance in Hopper-v4(205.95 vs.130.82)and InvertedPendulum-v4(322.97 vs.311.29).Although DDPG sometimes yields higher rewards,such as in HalfCheetah-v4(5695.37 vs.4894.59),it requires significantly greater energy expenditure.These results highlight PI-DDPG as a promising energy-conscious alternative for robotic control.展开更多
Accurate real-time monitoring of internal temperature in lithium-ion batteries remains critical for preventing thermal runaway,as conventional approaches sacrifice either computational efficiency or cross-scenario rob...Accurate real-time monitoring of internal temperature in lithium-ion batteries remains critical for preventing thermal runaway,as conventional approaches sacrifice either computational efficiency or cross-scenario robustness.We present a generalized fuzzy physics-informed framework that distills thermally sensitive electrochemical processes while circumventing redundant physical constraints,thereby establishing an explicit mechanism-constrained mapping between frequency-domain signals and internal temperature.This framework facilitates online thermal estimation,with dynamic validations in LiFePO_4/graphite 18650-type cells confirming real-time capability with near-instantaneous acquisition(~6 s per measurement),exceptional accuracy(±0.5℃) within the operational temperature range(30-50℃),and operational resilience across 20 %-80 % state-of-charge.The framework maintains predictive fidelity(±1.0℃ at 30℃ and ±4.0℃ at 60℃,95 % prediction intervals) across 80 %-100 % state-of-health while demonstrating adaptability to cathode materials and structural architectures.This strategy resolves the competing imperatives of physical interpretability,computational efficiency,and crossscenario generalizability,offering a universal paradigm for embedded thermal management in safetycritical applications.展开更多
In this study,we propose a constraint learning strategy based on interpretability analysis to improve the convergence and accuracy of the enriched physics-informed neural network(EPINN),which is applied to simulate tw...In this study,we propose a constraint learning strategy based on interpretability analysis to improve the convergence and accuracy of the enriched physics-informed neural network(EPINN),which is applied to simulate two-phase flow in heterogeneous porous media.Specifically,we first analyze the layerwise outputs of EPINN,and identify the distinct functions across layers,including dimensionality adjustment,pointwise construction of non-equilibrium potential,extraction of high-level features,and the establishment of long-range dependencies.Then,inspired by these distinct modules,we propose a novel constraint learning strategy based on regularization approaches,which improves neural network(NN)learning through layer-specific differentiated updates to enhance cross-timestep generalization.Since different neu ral network layers exhibit varying sensitivities to global generalization and local regression,we decrease the update frequency of layers more sensitive to local learning under this constraint learning strategy.In other words,the entire neural network is encouraged to extract more generalized features.The superior performance of the proposed learning strategy is validated through evaluations on numerical examples with varying computational complexities.Post hoc analysis reveals that gradie nt propagation exhibits more pronounced staged characte ristics,and the partial differential equation(PDE)residuals are more uniformly distributed under the constraint guidance.Interpretability analysis of the adaptive constraint process suggests that maintaining a stable information compression mode facilitates progressive convergence acceleration.展开更多
In the electroslag remelting(ESR)process,it mainly relies on thermal experiments or analysis via mechanistic models to realize the physical fields simulation of the electromagnetic field and temperature field coupled ...In the electroslag remelting(ESR)process,it mainly relies on thermal experiments or analysis via mechanistic models to realize the physical fields simulation of the electromagnetic field and temperature field coupled transfer,which has the limitations of high cost,a large amount of calculating data and high computing power requirements.A novel network based on physics-informed neural network(PINN)was designed to realize the fast and high-fidelity prediction of the distribution of electromagnetic field and temperature field in ESR process.The physical laws were combined with the deep learning network through PINN,and physical constraints were embedded to achieve effective solution of partial differential equations(PDEs).PINN was used to minimize the loss function consisting of data error,physical information error and boundary condition error.The physical laws and boundary condition constraints in the ESR process were considered to maintain high PDE solution accuracy under different spatial and temporal resolutions.Automatic differentiation(Autodiff)technique and gradient descent algorithm were used to optimize the network parameters.The experimental results show that compared with the mechanistic models,PINN can effectively replace thermal experiments to realize the physical field simulation of ESR process with only a few experimental data,which can avoid the disadvantages of pure data-driven network simulation that requires a large amount of training data.Moreover,the solution of PINN has good physical interpretability and reliability of simulation results.For simulating electromagnetic field and temperature field distribution,the training time of the network is only 140 and 203 s,and the regression indicators of root mean square error can reach 12.65 and 13.76,respectively.展开更多
Accurately forecasting peak particle velocity(PPV)during blasting operations plays a crucial role in mitigating vibration-related hazards and preventing economic losses.This research introduces an approach to PPV pred...Accurately forecasting peak particle velocity(PPV)during blasting operations plays a crucial role in mitigating vibration-related hazards and preventing economic losses.This research introduces an approach to PPV prediction by combining conventional empirical equations with physics-informed neural networks(PINN)and optimizing the model parameters via the Particle Swarm Optimization(PSO)algorithm.The proposed PSO-PINN framework was rigorously benchmarked against seven established machine learning approaches:Multilayer Perceptron(MLP),Extreme Gradient Boosting(XGBoost),Random Forest(RF),Support Vector Regression(SVR),Gradient Boosting Decision Tree(GBDT),Adaptive Boosting(Adaboost),and Gene Expression Programming(GEP).Comparative analysis showed that PSO-PINN outperformed these models,achieving RMSE reductions of 17.82-37.63%,MSE reductions of 32.47-61.10%,AR improvements of 2.97-21.19%,and R^(2)enhancements of 7.43-29.21%,demonstrating superior accuracy and generalization.Furthermore,the study determines the impact of incorporating empirical formulas as physical constraints in neural networks and examines the effects of different empirical equations,particle swarm size,iteration count in PSO,regularization coefficient,and learning rate in PINN on model performance.Lastly,a predictive system for blast vibration PPV is designed and implemented.The research outcomes offer theoretical references and practical recommendations for blast vibration forecasting in similar engineering applications.展开更多
Enforcing initial and boundary conditions(I/BCs)poses challenges in physics-informed neural networks(PINNs).Several PINN studies have gained significant achievements in developing techniques for imposing BCs in static...Enforcing initial and boundary conditions(I/BCs)poses challenges in physics-informed neural networks(PINNs).Several PINN studies have gained significant achievements in developing techniques for imposing BCs in static problems;however,the simultaneous enforcement of I/BCs in dynamic problems remains challenging.To overcome this limitation,a novel approach called decoupled physics-informed neural network(d PINN)is proposed in this work.The d PINN operates based on the core idea of converting a partial differential equation(PDE)to a system of ordinary differential equations(ODEs)via the space-time decoupled formulation.To this end,the latent solution is expressed in the form of a linear combination of approximation functions and coefficients,where approximation functions are admissible and coefficients are unknowns of time that must be solved.Subsequently,the system of ODEs is obtained by implementing the weighted-residual form of the original PDE over the spatial domain.A multi-network structure is used to parameterize the set of coefficient functions,and the loss function of d PINN is established based on minimizing the residuals of the gained ODEs.In this scheme,the decoupled formulation leads to the independent handling of I/BCs.Accordingly,the BCs are automatically satisfied based on suitable selections of admissible functions.Meanwhile,the original ICs are replaced by the Galerkin form of the ICs concerning unknown coefficients,and the neural network(NN)outputs are modified to satisfy the gained ICs.Several benchmark problems involving different types of PDEs and I/BCs are used to demonstrate the superior performance of d PINN compared with regular PINN in terms of solution accuracy and computational cost.展开更多
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
Using in-situ microstructure observations from 2010 to 2018,this study investigates the performance and generalization of machine learning models in parameterizing turbulent mixing in the northwestern South China Sea....Using in-situ microstructure observations from 2010 to 2018,this study investigates the performance and generalization of machine learning models in parameterizing turbulent mixing in the northwestern South China Sea.The results show that the data-driven extreme gradient boosting(XGBoost)performs better than the other four models,i.e.,random forest,neural network,linear regression and support vector machine regression.In order to further improve the generalization of machine learning-based parameterization method,we propose a physics-informed machine learning(PIML)that couples the MacKinnon-Gregg model(known as the MG model)and Osborn’s formula to the XGBoost model.The correlation coefficient(r)and root mean square error(RMSE)between the estimated and observed 1g(ε)(whereεdenotes the turbulent kinetic energy dissipation rate)from the PIML are improved by 14%and 16%,respectively.The results also show that PIML effectively improves the generalization of the XGBoost-based parameterization method,enhancing r and RMSE by 35%and 75%,respectively.展开更多
基金funded by National Research Council of Thailand(contract No.N42A671047).
文摘Physics-informed neural networks(PINNs)have emerged as a promising class of scientific machine learning techniques that integrate governing physical laws into neural network training.Their ability to enforce differential equations,constitutive relations,and boundary conditions within the loss function provides a physically grounded alternative to traditional data-driven models,particularly for solid and structural mechanics,where data are often limited or noisy.This review offers a comprehensive assessment of recent developments in PINNs,combining bibliometric analysis,theoretical foundations,application-oriented insights,and methodological innovations.A biblio-metric survey indicates a rapid increase in publications on PINNs since 2018,with prominent research clusters focused on numerical methods,structural analysis,and forecasting.Building upon this trend,the review consolidates advance-ments across five principal application domains,including forward structural analysis,inverse modeling and parameter identification,structural and topology optimization,assessment of structural integrity,and manufacturing processes.These applications are propelled by substantial methodological advancements,encompassing rigorous enforcement of boundary conditions,modified loss functions,adaptive training,domain decomposition strategies,multi-fidelity and transfer learning approaches,as well as hybrid finite element–PINN integration.These advances address recurring challenges in solid mechanics,such as high-order governing equations,material heterogeneity,complex geometries,localized phenomena,and limited experimental data.Despite remaining challenges in computational cost,scalability,and experimental validation,PINNs are increasingly evolving into specialized,physics-aware tools for practical solid and structural mechanics applications.
基金supported by the National Key Research and Development Plan(Grant No.2022YFB3401901)the National Natural Science Foundation of China(Grant Nos.12192210,12192214,12072295,and 12222209)+1 种基金Independent Project of State Key Laboratory of Rail Transit Vehicle System(Grant No.2023TPL-T03)Fundamental Research Funds for the Central Universities(Grant No.2682023CG004).
文摘The S38C railway axle undergoes induction hardening,resulting in a gradient-distributed microstructure and mechanical properties.The accurate identification of gradient-distributed plastic parameters for the S38C axle remains a challenging task.To tackle this challenge,the present study proposes a novel approach for identifying the gradient-distributed plastic parameters for the S38C axle by integrating nano-indentation techniques with the machine learning method.Firstly,nano-indentation tests are conducted along the radial direction of the S38C axle to obtain the gradient-distributed load-displacement curves,nano-hardness,and elastic modulus.Subsequently,the dimensionless analysis is performed to obtain the representative stress,strain,and yield stress from load-displacement curves.These parameters are then incorporated into the machine learning method as physical information to identify the gradient-distributed plastic parameters of the S38C axle.The results indicate that the proposed method based on the physics-informed neural network and multi-fidelity neural network successfully identifies the gradient-distributed plastic parameters of the S38C axles and demonstrates superior prediction accuracy and generalization compared with the purely data-driven machine learning method.
基金supported by the National Natural Science Foundation of China(No.52207228)the Beijing Natural Science Foundation,China(No.3224070)the National Natural Science Foundation of China(No.52077208).
文摘The growing use of lithium-ion batteries in electric transportation and grid-scale storage systems has intensified the need for accurate and highly generalizable state-of-health(SOH)estimation.Conventional approaches often suffer from reduced accuracy under dynamically uncertain state-of-charge(SOC)operating ranges and heterogeneous aging stresses.This study presents a unified SOH estimation framework that integrates physics-informed modeling,subspace identification,and Transformer-based learning.A reduced-order model is derived from simplified electrochemical dynamics,providing an interpretable and computationally efficient representation of battery behavior.Subspace identification across a wide SOC and SOH range yields degradation-sensitive features,which the Transformer uses to capture long-range aging dynamics via multi-head self-attention.Experiments on LiFePO4 cells under joint-cell training show consistently accurate SOH estimation,with a maximum error of 1.39%,demonstrating the framework’s effectiveness in decoupling SOC and SOH effects.In cross-cell validation,where training and validation are performed on different cells,the model maintains a maximum error of 2.06%,confirming strong generalization to unseen aging trajectories.Comparative experiments on LiFePO_(4)and public LiCoO_(2)datasets confirm the framework’s cross-chemistry applicability.By extracting low-dimensional,physically interpretable features via subspace identification,the framework significantly reduces training cost while maintaining high SOH estimation accuracy,outperforming conventional data-driven models lacking physical guidance.
基金supported by the Shanghai Pilot Program for Basic Research(22T01400100-18)the National Natural Science Foundation of China(22278127 and 12447149)+1 种基金the Fundamental Research Funds for the Central Universities(2022ZFJH004)the Postdoctoral Fellowship Program of CPSF(GZB20250159).
文摘Accurate state of health(SOH)estimation is essential for the safe and reliable operation of lithium-ion batteries.However,existing methods face significant challenges,primarily because they rely on complete charge–discharge cycles and fixed-form physical constraints,which limit adaptability to different chemistries and real-world conditions.To address these issues,this study proposes an approach that extracts features from segmented state of charge(SOC)intervals and integrates them into an enhanced physics-informed neural network(PINN).Specifically,voltage data within the 25%–75%SOC range during charging are used to derive statistical,time–frequency,and mechanism-based features that capture degradation trends.A hybrid PINN-Lasso-Transformer-BiLSTM architecture is developed,where Lasso regression enables sparse feature selection,and a nonlinear empirical degradation model is embedded as a learnable physical term within a dynamically scaled composite loss.This design adaptively balances data-driven accuracy with physical consistency,thereby enhancing estimation precision,robustness,and generalization.The results show that the proposed method outperforms conventional neural networks across four battery chemistries,achieving root mean square error and mean absolute error below 1%.Notably,features from partial charging segments exhibit higher robustness than those from full cycles.Furthermore,the model maintains strong performance under high temperatures and demonstrates excellent generalization capacity in transfer learning across chemistries,temperatures,and C-rates.This work establishes a scalable and interpretable solution for accurate SOH estimation under diverse practical operating conditions.
基金Project supported by the Basic Science Research Program through the National Research Foundation(NRF)of Korea funded by the Ministry of Science and ICT(No.RS-2024-00337001)。
文摘Physics-informed neural networks(PINNs)have been shown as powerful tools for solving partial differential equations(PDEs)by embedding physical laws into the network training.Despite their remarkable results,complicated problems such as irregular boundary conditions(BCs)and discontinuous or high-frequency behaviors remain persistent challenges for PINNs.For these reasons,we propose a novel two-phase framework,where a neural network is first trained to represent shape functions that can capture the irregularity of BCs in the first phase,and then these neural network-based shape functions are used to construct boundary shape functions(BSFs)that exactly satisfy both essential and natural BCs in PINNs in the second phase.This scheme is integrated into both the strong-form and energy PINN approaches,thereby improving the quality of solution prediction in the cases of irregular BCs.In addition,this study examines the benefits and limitations of these approaches in handling discontinuous and high-frequency problems.Overall,our method offers a unified and flexible solution framework that addresses key limitations of existing PINN methods with higher accuracy and stability for general PDE problems in solid mechanics.
基金financially supported by the National Natural Science Foundation of China(No.22473032)。
文摘Accelerated aging tests are widely used to rapidly evaluate the durability of materials,of which thermal-oxidative aging is the most common approach.To quantitatively predict the effects of multiple coupled factors,this study takes polyamide66 reinforced with glass fiber(PA66-GF)as a model system and proposed a high-precision paradigm for coupled thermal-oxidative aging.By integrating Arrhenius-type reaction kinetics with oxygen diffusion,a predictive formula that holistically captures the nonlinear synergistic effects of multiple factors was developed,thereby overcoming the limitations of traditional single-variable models.A systematic evaluation of the stepwise improved formulas through nonlinear fitting showed that the coefficient of determination(R^(2))increased from 0.223 to 0.803,elucidating the fundamental reason why conventional approaches fail in quantitative prediction.These formulae were further embedded as physical constraints into a physics-informed neural network(PINN),which further enhanced the predictive performance,with the proposed formula achieving a peak R^(2)of 0.946.The results highlight that robust data fitting alone is insufficient;the decisive factor for the success of PINN lies in whether the embedded formula faithfully reflects the underlying physical mechanisms.When applied to polyamide 6 reinforced with glass fiber(PA6-GF),the Formula-constrained PINN maintained a high level of accuracy(R^(2)=0.916),demonstrating its strong cross-system generalizability.In summary,this work establishes a robust hybrid physics-machine learning framework that combines high accuracy with transferability for predicting the thermal-oxidative aging behavior of composite material systems.
基金supported by the National Natural Science Foundation of China(Nos.12375310,118750356,and 62006110)Graduate Research and Innovation Projects of Hunan Province(CX20230964).
文摘Physics-informed neural networks(PINNs)are vital for machine learning and exhibit significant advantages when handling complex physical problems.The PINN method can rapidly predict ^(220)Rn progeny concentration and is very important for regulating and measuring this property.To construct a PINN model,training data are typically preprocessed;however,this approach changes the physical characteristics of the data,with the preprocessed data potentially no longer directly conforming to the original physical equations.As a result,the original physical equations cannot be directly employed in the PINN.Consequently,an effective method for transforming physical equations is crucial for accurately constraining PINNs to model the ^(220)Rn progeny concentration prediction.This study presents an equation adaptation approach for neural networks,which is designed to improve prediction of ^(220)Rn progeny concentration.Five neural network models based on three architectures are established:a classical network,a physics-informed network without equation adaptation,and a physics-informed network with equation adaptation.The transport equation of the ^(220)Rn progeny concentration is transformed via equation adaption and integrated with the PINN model.The compatibility and robustness of the model with equation adaption is then analyzed.The results show that PINNs with equation adaption converge consistently with classical neural networks in terms of the training and validation loss and achieve the same level of prediction accuracy.This outcome indicates that the proposed method can be integrated into the neural network architecture.Moreover,the prediction performance of classical neural networks declines significantly when interference data are encountered,whereas the PINNs with equation adaption exhibit stable prediction accuracy.This performance demonstrates that the proposed method successfully harnesses the constraining power of physical equations,significantly enhancing the robustness of the resultant PINN models.Thus,the use of a physics-informed network with equation adaption can guarantee accurate prediction of ^(220)Rn progeny concentration.
基金supported by the Science and Technology on Reactor System Design Technology Laboratory(No.LRSDT12023108)supported in part by the Chongqing Postdoctoral Science Foundation(No.cstc2021jcyj-bsh0252)+2 种基金the National Natural Science Foundation of China(No.12005030)Sichuan Province to unveil the list of marshal industry common technology research projects(No.23jBGOV0001)Special Program for Stabilizing Support to Basic Research of National Basic Research Institutes(No.WDZC-2023-05-03-05).
文摘The neutron diffusion equation plays a pivotal role in nuclear reactor analysis.Nevertheless,employing the physics-informed neural network(PINN)method for its solution entails certain limitations.Conventional PINN approaches generally utilize a fully connected network(FCN)architecture that is susceptible to overfitting,training instability,and gradient vanishing as the network depth increases.These challenges result in accuracy bottlenecks in the solution.In response to these issues,the residual-based resample physics-informed neural network(R2-PINN)is proposed.It is an improved PINN architecture that replaces the FCN with a convolutional neural network with a shortcut(S-CNN).It incorporates skip connections to facilitate gradient propagation between network layers.Additionally,the incorporation of the residual adaptive resampling(RAR)mechanism dynamically increases the number of sampling points.This,in turn,enhances the spatial representation capabilities and overall predictive accuracy of the model.The experimental results illustrate that our approach significantly improves the convergence capability of the model and achieves high-precision predictions of the physical fields.Compared with conventional FCN-based PINN methods,R 2-PINN effectively overcomes the limitations inherent in current methods.Thus,it provides more accurate and robust solutions for neutron diffusion equations.
基金Supported by the National Key Research and Development Program of China(No.2023YFC3008200)the Independent Research Project of Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)(No.SML2022SP505)。
文摘Physics-informed neural networks(PINNs),as a novel artificial intelligence method for solving partial differential equations,are applicable to solve both forward and inverse problems.This study evaluates the performance of PINNs in solving the temperature diffusion equation of the seawater across six scenarios,including forward and inverse problems under three different boundary conditions.Results demonstrate that PINNs achieved consistently higher accuracy with the Dirichlet and Neumann boundary conditions compared to the Robin boundary condition for both forward and inverse problems.Inaccurate weighting of terms in the loss function can reduce model accuracy.Additionally,the sensitivity of model performance to the positioning of sampling points varied between different boundary conditions.In particular,the model under the Dirichlet boundary condition exhibited superior robustness to variations in point positions during the solutions of inverse problems.In contrast,for the Neumann and Robin boundary conditions,accuracy declines when points were sampled from identical positions or at the same time.Subsequently,the Argo observations were used to reconstruct the vertical diffusion of seawater temperature in the north-central Pacific for the applicability of PINNs in the real ocean.The PINNs successfully captured the vertical diffusion characteristics of seawater temperature,reflected the seasonal changes of vertical temperature under different topographic conditions,and revealed the influence of topography on the temperature diffusion coefficient.The PINNs were proved effective in solving the temperature diffusion equation of seawater with limited data,providing a promising technique for simulating or predicting ocean phenomena using sparse observations.
基金supported by the Beijing Natural Science Foundation(Grant No.L223013)。
文摘For the diagnostics and health management of lithium-ion batteries,numerous models have been developed to understand their degradation characteristics.These models typically fall into two categories:data-driven models and physical models,each offering unique advantages but also facing limitations.Physics-informed neural networks(PINNs)provide a robust framework to integrate data-driven models with physical principles,ensuring consistency with underlying physics while enabling generalization across diverse operational conditions.This study introduces a PINN-based approach to reconstruct open circuit voltage(OCV)curves and estimate key ageing parameters at both the cell and electrode levels.These parameters include available capacity,electrode capacities,and lithium inventory capacity.The proposed method integrates OCV reconstruction models as functional components into convolutional neural networks(CNNs)and is validated using a public dataset.The results reveal that the estimated ageing parameters closely align with those obtained through offline OCV tests,with errors in reconstructed OCV curves remaining within 15 mV.This demonstrates the ability of the method to deliver fast and accurate degradation diagnostics at the electrode level,advancing the potential for precise and efficient battery health management.
基金supported in part by Multimedia University under the Research Fellow Grant MMUI/250008in part by Telekom Research&Development Sdn Bhd underGrantRDTC/241149Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R140),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘The Internet of Things(IoT)ecosystem faces growing security challenges because it is projected to have 76.88 billion devices by 2025 and $1.4 trillion market value by 2027,operating in distributed networks with resource limitations and diverse system architectures.The current conventional intrusion detection systems(IDS)face scalability problems and trust-related issues,but blockchain-based solutions face limitations because of their low transaction throughput(Bitcoin:7 TPS(Transactions Per Second),Ethereum:15-30 TPS)and high latency.The research introduces MBID(Multi-Tier Blockchain Intrusion Detection)as a groundbreaking Multi-Tier Blockchain Intrusion Detection System with AI-Enhanced Detection,which solves the problems in huge IoT networks.The MBID system uses a four-tier architecture that includes device,edge,fog,and cloud layers with blockchain implementations and Physics-Informed Neural Networks(PINNs)for edge-based anomaly detection and a dual consensus mechanism that uses Honesty-based Distributed Proof-of-Authority(HDPoA)and Delegated Proof of Stake(DPoS).The system achieves scalability and efficiency through the combination of dynamic sharding and Interplanetary File System(IPFS)integration.Experimental evaluations demonstrate exceptional performance,achieving a detection accuracy of 99.84%,an ultra-low false positive rate of 0.01% with a False Negative Rate of 0.15%,and a near-instantaneous edge detection latency of 0.40 ms.The system demonstrated an aggregate throughput of 214.57 TPS in a 3-shard configuration,providing a clear,evidence-based path for horizontally scaling to support overmillions of devices with exceeding throughput.The proposed architecture represents a significant advancement in blockchain-based security for IoT networks,effectively balancing the trade-offs between scalability,security,and decentralization.
基金funded by the National Key R&D Program of China,Grant No.2024YFF0504904.
文摘The packaging quality of coaxial laser diodes(CLDs)plays a pivotal role in determining their optical performance and long-term reliability.As the core packaging process,high-precision laser welding requires precise control of process parameters to suppress optical power loss.However,the complex nonlinear relationship between welding parameters and optical power loss renders traditional trial-and-error methods inefficient and imprecise.To address this challenge,a physics-informed(PI)and data-driven collaboration approach for welding parameter optimization is proposed.First,thermal-fluid-solid coupling finite element method(FEM)was employed to quantify the sensitivity of welding parameters to physical characteristics,including residual stress.This analysis facilitated the identification of critical factors contributing to optical power loss.Subsequently,a Gaussian process regression(GPR)model incorporating finite element simulation prior knowledge was constructed based on the selected features.By introducing physics-informed kernel(PIK)functions,stress distribution patterns were embedded into the prediction model,achieving high-precision optical power loss prediction.Finally,a Bayesian optimization(BO)algorithm with an adaptive sampling strategy was implemented for efficient parameter space exploration.Experimental results demonstrate that the proposedmethod effectively establishes explicit physical correlations between welding parameters and optical power loss.The optimized welding parameters reduced optical power loss by 34.1%,providing theoretical guidance and technical support for reliable CLD packaging.
文摘Energy efficiency stands as an essential factor when implementing deep reinforcement learning(DRL)policies for robotic control systems.Standard algorithms,including Deep Deterministic Policy Gradient(DDPG),primarily optimize task rewards but at the cost of excessively high energy consumption,making them impractical for real-world robotic systems.To address this limitation,we propose Physics-Informed DDPG(PI-DDPG),which integrates physics-based energy penalties to develop energy-efficient yet high-performing control policies.The proposed method introduces adaptive physics-informed constraints through a dynamic weighting factor(λ),enabling policies that balance reward maximization with energy savings.Our motivation is to overcome the impracticality of rewardonly optimization by designing controllers that achieve competitive performance while substantially reducing energy consumption.PI-DDPG was evaluated in nine MuJoCo continuous control environments,where it demonstrated significant improvements in energy efficiency without compromising stability or performance.Experimental results confirm that PI-DDPG substantially reduces energy consumption compared to standard DDPG,while maintaining competitive task performance.For instance,energy costs decreased from 5542.98 to 3119.02 in HalfCheetah-v4 and from1909.13 to 1586.75 in Ant-v4,with stable performance in Hopper-v4(205.95 vs.130.82)and InvertedPendulum-v4(322.97 vs.311.29).Although DDPG sometimes yields higher rewards,such as in HalfCheetah-v4(5695.37 vs.4894.59),it requires significantly greater energy expenditure.These results highlight PI-DDPG as a promising energy-conscious alternative for robotic control.
基金supported by the National Natural Science Foundation of China (Grant No.22109008)。
文摘Accurate real-time monitoring of internal temperature in lithium-ion batteries remains critical for preventing thermal runaway,as conventional approaches sacrifice either computational efficiency or cross-scenario robustness.We present a generalized fuzzy physics-informed framework that distills thermally sensitive electrochemical processes while circumventing redundant physical constraints,thereby establishing an explicit mechanism-constrained mapping between frequency-domain signals and internal temperature.This framework facilitates online thermal estimation,with dynamic validations in LiFePO_4/graphite 18650-type cells confirming real-time capability with near-instantaneous acquisition(~6 s per measurement),exceptional accuracy(±0.5℃) within the operational temperature range(30-50℃),and operational resilience across 20 %-80 % state-of-charge.The framework maintains predictive fidelity(±1.0℃ at 30℃ and ±4.0℃ at 60℃,95 % prediction intervals) across 80 %-100 % state-of-health while demonstrating adaptability to cathode materials and structural architectures.This strategy resolves the competing imperatives of physical interpretability,computational efficiency,and crossscenario generalizability,offering a universal paradigm for embedded thermal management in safetycritical applications.
基金supported by the National Key R&D Program of China(No.2023YFB4104200)the National Natural Science Foundation of China(Nos.52474067,52441411,52325402,52034010,and12131014)+2 种基金the Natural Science Foundation of Shandong Province,China(No.ZR2024ME005)Fundamental Research Funds for the Central Universities(Nos.25CX02025A and 21CX06031A)the Youth Innovation and Technology Support Program for Higher Education Institutions of Shandong Province,China(No.2022KJ070)。
文摘In this study,we propose a constraint learning strategy based on interpretability analysis to improve the convergence and accuracy of the enriched physics-informed neural network(EPINN),which is applied to simulate two-phase flow in heterogeneous porous media.Specifically,we first analyze the layerwise outputs of EPINN,and identify the distinct functions across layers,including dimensionality adjustment,pointwise construction of non-equilibrium potential,extraction of high-level features,and the establishment of long-range dependencies.Then,inspired by these distinct modules,we propose a novel constraint learning strategy based on regularization approaches,which improves neural network(NN)learning through layer-specific differentiated updates to enhance cross-timestep generalization.Since different neu ral network layers exhibit varying sensitivities to global generalization and local regression,we decrease the update frequency of layers more sensitive to local learning under this constraint learning strategy.In other words,the entire neural network is encouraged to extract more generalized features.The superior performance of the proposed learning strategy is validated through evaluations on numerical examples with varying computational complexities.Post hoc analysis reveals that gradie nt propagation exhibits more pronounced staged characte ristics,and the partial differential equation(PDE)residuals are more uniformly distributed under the constraint guidance.Interpretability analysis of the adaptive constraint process suggests that maintaining a stable information compression mode facilitates progressive convergence acceleration.
基金supported by National Natural Science Foundation of China(52274323 and 524743495)the Postdoctoral Fellowship Program of CPSF under Grant Number GZC20240231.
文摘In the electroslag remelting(ESR)process,it mainly relies on thermal experiments or analysis via mechanistic models to realize the physical fields simulation of the electromagnetic field and temperature field coupled transfer,which has the limitations of high cost,a large amount of calculating data and high computing power requirements.A novel network based on physics-informed neural network(PINN)was designed to realize the fast and high-fidelity prediction of the distribution of electromagnetic field and temperature field in ESR process.The physical laws were combined with the deep learning network through PINN,and physical constraints were embedded to achieve effective solution of partial differential equations(PDEs).PINN was used to minimize the loss function consisting of data error,physical information error and boundary condition error.The physical laws and boundary condition constraints in the ESR process were considered to maintain high PDE solution accuracy under different spatial and temporal resolutions.Automatic differentiation(Autodiff)technique and gradient descent algorithm were used to optimize the network parameters.The experimental results show that compared with the mechanistic models,PINN can effectively replace thermal experiments to realize the physical field simulation of ESR process with only a few experimental data,which can avoid the disadvantages of pure data-driven network simulation that requires a large amount of training data.Moreover,the solution of PINN has good physical interpretability and reliability of simulation results.For simulating electromagnetic field and temperature field distribution,the training time of the network is only 140 and 203 s,and the regression indicators of root mean square error can reach 12.65 and 13.76,respectively.
基金supported by the National Natural Science Foundation of China(Grant No.52409143)the Basic Scientific Research Fund of Changjiang River Scientific Research Institute for Central-level Public Welfare Research Institutes(Grant No.CKSF2025184/YT)the Hubei Provincial Natural Science Foundation of China(Grant No.2022CFB673).
文摘Accurately forecasting peak particle velocity(PPV)during blasting operations plays a crucial role in mitigating vibration-related hazards and preventing economic losses.This research introduces an approach to PPV prediction by combining conventional empirical equations with physics-informed neural networks(PINN)and optimizing the model parameters via the Particle Swarm Optimization(PSO)algorithm.The proposed PSO-PINN framework was rigorously benchmarked against seven established machine learning approaches:Multilayer Perceptron(MLP),Extreme Gradient Boosting(XGBoost),Random Forest(RF),Support Vector Regression(SVR),Gradient Boosting Decision Tree(GBDT),Adaptive Boosting(Adaboost),and Gene Expression Programming(GEP).Comparative analysis showed that PSO-PINN outperformed these models,achieving RMSE reductions of 17.82-37.63%,MSE reductions of 32.47-61.10%,AR improvements of 2.97-21.19%,and R^(2)enhancements of 7.43-29.21%,demonstrating superior accuracy and generalization.Furthermore,the study determines the impact of incorporating empirical formulas as physical constraints in neural networks and examines the effects of different empirical equations,particle swarm size,iteration count in PSO,regularization coefficient,and learning rate in PINN on model performance.Lastly,a predictive system for blast vibration PPV is designed and implemented.The research outcomes offer theoretical references and practical recommendations for blast vibration forecasting in similar engineering applications.
基金Project supported by the Basic Science Research Program through the National Research Foundation(NRF)of Korea funded by the Ministry of Science and ICT(No.RS-2024-00337001)。
文摘Enforcing initial and boundary conditions(I/BCs)poses challenges in physics-informed neural networks(PINNs).Several PINN studies have gained significant achievements in developing techniques for imposing BCs in static problems;however,the simultaneous enforcement of I/BCs in dynamic problems remains challenging.To overcome this limitation,a novel approach called decoupled physics-informed neural network(d PINN)is proposed in this work.The d PINN operates based on the core idea of converting a partial differential equation(PDE)to a system of ordinary differential equations(ODEs)via the space-time decoupled formulation.To this end,the latent solution is expressed in the form of a linear combination of approximation functions and coefficients,where approximation functions are admissible and coefficients are unknowns of time that must be solved.Subsequently,the system of ODEs is obtained by implementing the weighted-residual form of the original PDE over the spatial domain.A multi-network structure is used to parameterize the set of coefficient functions,and the loss function of d PINN is established based on minimizing the residuals of the gained ODEs.In this scheme,the decoupled formulation leads to the independent handling of I/BCs.Accordingly,the BCs are automatically satisfied based on suitable selections of admissible functions.Meanwhile,the original ICs are replaced by the Galerkin form of the ICs concerning unknown coefficients,and the neural network(NN)outputs are modified to satisfy the gained ICs.Several benchmark problems involving different types of PDEs and I/BCs are used to demonstrate the superior performance of d PINN compared with regular PINN in terms of solution accuracy and computational cost.
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The National Science and Technology Major Project under contract No.2024YFC2817003the National Natural Science Foundation of China under contract Nos 42276019 and 42249911the Guangdong Ordinary University Innovation Team Project under contract No.2023KCXTD015.
文摘Using in-situ microstructure observations from 2010 to 2018,this study investigates the performance and generalization of machine learning models in parameterizing turbulent mixing in the northwestern South China Sea.The results show that the data-driven extreme gradient boosting(XGBoost)performs better than the other four models,i.e.,random forest,neural network,linear regression and support vector machine regression.In order to further improve the generalization of machine learning-based parameterization method,we propose a physics-informed machine learning(PIML)that couples the MacKinnon-Gregg model(known as the MG model)and Osborn’s formula to the XGBoost model.The correlation coefficient(r)and root mean square error(RMSE)between the estimated and observed 1g(ε)(whereεdenotes the turbulent kinetic energy dissipation rate)from the PIML are improved by 14%and 16%,respectively.The results also show that PIML effectively improves the generalization of the XGBoost-based parameterization method,enhancing r and RMSE by 35%and 75%,respectively.
