Surrogate Model Research Articles

An mRNA-LNP-based Lassa virus vaccine induces protective immunity in mice.

The mammarenavirus Lassa virus (LASV) causes the life-threatening hemorrhagic fever disease, Lassa fever. The lack of licensed medical countermeasures against LASV underscores the urgent need for the development of novel LASV vaccines, which has been hampered by the requirement for a biosafety level 4 facility to handle live LASV. Here, we investigated the efficacy of mRNA-lipid nanoparticle (mRNA-LNP)-based vaccines expressing the LASV glycoprotein precursor (LASgpc) or nucleoprotein (LCMnp) of the prototypic mammarenavirus, lymphocytic choriomeningitis virus (LCMV), in mice. Two doses of LASgpc- or LCMnp-mRNA-LNP administered intravenously (i.v.) protected C57BL/6 mice from a lethal challenge with a recombinant (r) LCMV expressing a modified LASgpc (rLCMV/LASgpc2m) inoculated intracranially. Intramuscular (i.m.) immunization with two doses of LASgpc- or LCMnp-mRNA-LNP significantly reduced the viral load in C57BL/6 mice inoculated i.v. with rLCMV/LASgpc2m. High levels of viremia and lethality were observed in CBA mice inoculated i.v. with rLCMV/LASgpc2m, which were abrogated by i.m. immunization with two doses of LASgpc-mRNA-LNP. The protective efficacy of two i.m. doses of LCMnp-mRNA-LNP was confirmed in a lethal hemorrhagic disease model of FVB mice i.v. inoculated with wild-type rLCMV. In all conditions tested, negligible and high levels of LASgpc- and LCMnp-specific antibodies were detected in mRNA-LNP-immunized mice, respectively, but robust LASgpc- and LCMnp-specific CD8+ T cell responses were induced. Accordingly, plasma from LASgpc-mRNA-LNP-immunized mice did not exhibit neutralizing activity. Our findings and surrogate mouse models of LASV infection, which can be studied at a reduced biocontainment level, provide a critical foundation for the rapid development of mRNA-LNP-based LASV vaccines.IMPORTANCELassa virus (LASV) is a highly pathogenic mammarenavirus responsible for several hundred thousand infections annually in West African countries, causing a high number of lethal Lassa fever (LF) cases. Despite its significant impact on human health, clinically approved, safe, and effective medical countermeasures against LF are not available. The requirement of a biosafety level 4 facility to handle live LASV has been one of the main obstacles to the research and development of LASV countermeasures. Here, we report that two doses of mRNA-lipid nanoparticle-based vaccines expressing the LASV glycoprotein precursor (LASgpc) or nucleoprotein (LCMnp) of lymphocytic choriomeningitis virus (LCMV), a mammarenavirus genetically closely related to LASV, conferred protection to recombinant LCMV-based surrogate mouse models of lethal LASV infection. Notably, robust LASgpc- and LCMnp-specific CD8+ T cell responses were detected in mRNA-LNP-immunized mice, whereas no virus-neutralizing activity was observed.

Bayesian Optimization for Efficient Prediction of Gas Uptake in Nanoporous Materials.

The discovery and optimization of novel nanoporous materials (NPMs) such as Metal-Organic Frameworks (MOFs) are crucial for addressing global challenges. Traditional experimental approaches for optimizing these materials are time-consuming and resource-intensive. This research paper presents a strategy using Bayesian optimization (BO) to efficiently navigate the NPMs for gas storage. For a MOF dataset drawn from 19 different sources, we present a quantitative evaluation of BO. In our study, we employed machine learning (ML) techniques to conduct regression analysis on many models. Following this, we identified the three ML models that exhibited the highest accuracy, which were subsequently chosen as surrogates in our investigation, includingthe conventional Gaussian Process (GP) model. We found that GP with expected improvement (EI) as the acquisition function but without a gamma prior which is standard in Bayesian Optimization python library outperforms other surrogate models. Additionally, it should be noted that while the machine learning model that exhibits superior performance in predicting the target variable may be considered the best choice, it may not necessarily serve as the most suitable surrogate model for BO. This observation has significant importance and warrants further investigation. This comprehensive framework accelerates the pace of materials discovery.

Bayesian inversion of tilt data using a machine-learned surrogate model for pressurised fractures

Bayesian inversion of tilt data using a machine-learned surrogate model for pressurised fractures

A constrained machine learning surrogate model to predict the distribution of water-in-oil emulsions in electrostatic fields

Accurately describing the evolution of water droplet size distribution in crude oil is fundamental for evaluating the water separation efficiency in dehydration systems. Enhancing the separation of an aqueous phase dispersed in a dielectric oil phase, which has a significantly lower dielectric constant than the dispersed phase, can be achieved by increasing the water droplet size through the application of an electrostatic field in the pipeline. Mathematical models, while being accurate, are computationally expensive. Herein, we introduced a constrained machine learning (ML) surrogate model developed based on a population balance model. This model serves as a practical alternative, facilitating fast and accurate predictions. The constrained ML model, utilizing an extreme gradient boosting (XGBoost) algorithm tuned with a genetic algorithm (GA), incorporates the key parameters of the electrostatic dehydration process, including droplet diameter, voltage, crude oil properties, temperature, and residence time as input variables, with the output being the number of water droplets per unit volume. Furthermore, we modified the objective function of the XGBoost algorithm by incorporating two penalty terms to ensure the model’s predictions adhere to physical principles. The constrained model demonstrated accuracy on the test set, with a mean squared error of 0.005 and a coefficient of determination of 0.998. The efficiency of the model was validated through comparison with the experimental data and the results of the population balance mathematical model. The analysis shows that the initial droplet diameter and voltage have the highest influence on the model, which aligns with the observed behaviour in the real-world process.

Atmospheric Chemistry Surrogate Modeling With Sparse Identification of Nonlinear Dynamics

AbstractModeling atmospheric chemistry is computationally expensive and limits the widespread use of chemical transport models. This computational cost arises from solving high‐dimensional systems of stiff differential equations. Previous work has demonstrated the promise of machine learning (ML) to accelerate air quality model simulations but has suffered from numerical instability during long‐term simulations. This may be because previous ML‐based efforts have relied on explicit Euler time integration—which is known to be unstable for stiff systems—and have used neural networks which are prone to overfitting. We hypothesize that parsimonious models combined with modern numerical integration techniques can overcome this limitation. Using a small‐scale mechanism to explore the potential of these methods, we have created a machine‐learned surrogate by (a) reducing dimensionality using singular value decomposition to create an interpretably‐compressed low‐dimensional latent space and (b) using Sparse Identification of Nonlinear Dynamics (SINDy) to create a differential‐equation‐based representation of the underlying dynamics in the compressed latent space with reduced stiffness. The root mean square error of ML model prediction for ozone concentration over 9 days is 37.8% of the root mean square concentration across all simulations in our testing data set. The surrogate model is 10× faster with 12× fewer integration timesteps compared to reference model and is numerically stable in all tested simulations. Overall, we find that SINDy can be used to create fast, stable, and accurate surrogates of a simple photochemical mechanism. In future work, we will explore the application of this method to more detailed mechanisms and their use in large‐scale simulations.

Data-assisted training of a physics-informed neural network to predict the separated Reynolds-averaged turbulent flow field around an airfoil under variable angles of attack

AbstractPhysics-informed neural networks are a promising method to yield surrogate models of flow fields. We present a metamodeling technique for variable geometries based on physics-informed neural networks. The method was applied to the DU99W350 airfoil at a Reynolds number of $$1\times 10^{5}$$ 1 × 10 5 . Using our technique, the angle of attack was introduced as an additional input parameter of the network and the model was trained to predict the Reynolds-averaged velocity and pressure fields around the airfoil for arbitrary angles of attack between 10.0 and 17.5$$^{\circ }$$ ∘ . Furthermore, we present an effective method to generate the training points for the parameterized geometry. The model was trained with data from simulations for a limited set of angles of attack. Additionally, satisfaction of the a priori known boundary conditions as well as the Reynolds-averaged Navier–Stokes equations was attained. A sensitivity analysis concerning the Reynolds number, the amount and distribution of training data, and the turbulence model was conducted showing the superiority of the pseudo-Reynolds stress method and the demand for labeled training data in the domain. The trained network was capable of predicting the flow separation progressing with angle of attack on the suction surface and exhibited excellent agreement with numerically simulated results, even in the proximity of the wall for interpolations as well as extrapolations from the labeled data set. Our study demonstrates that physics-informed neural networks can be used to obtain accurate flow field surrogate models of variable geometries.

Task-oriented machine learning surrogates for tipping points of agent-based models

We present a machine learning framework bridging manifold learning, neural networks, Gaussian processes, and Equation-Free multiscale approach, for the construction of different types of effective reduced order models from detailed agent-based simulators and the systematic multiscale numerical analysis of their emergent dynamics. The specific tasks of interest here include the detection of tipping points, and the uncertainty quantification of rare events near them. Our illustrative examples are an event-driven, stochastic financial market model describing the mimetic behavior of traders, and a compartmental stochastic epidemic model on an Erdös-Rényi network. We contrast the pros and cons of the different types of surrogate models and the effort involved in learning them. Importantly, the proposed framework reveals that, around the tipping points, the emergent dynamics of both benchmark examples can be effectively described by a one-dimensional stochastic differential equation, thus revealing the intrinsic dimensionality of the normal form of the specific type of the tipping point. This allows a significant reduction in the computational cost of the tasks of interest.

Combining simulation and experimental data via surrogate modelling of continuum dislocation dynamics simulations

Abstract Several computational models have been introduced in recent years to
yield comprehensive insights into microstructural evolution analyses. However, the
identification of the correct input parameters to a simulation that corresponds to a
certain experimental result is a major challenge on this length scale. To complement
simulation results with experimental data (and vice versa) is not trivial since, e.g.,
simulation model parameters might lack a physical understanding or uncertainties
in the experimental data are neglected. Computational costs are another challenge
mesoscale models always have to face, so comprehensive parameter studies can be
costly. In this paper, we introduce a surrogate model to circumvent continuum
dislocation dynamics simulation by a data-driven linkage between well-defined input
parameters and output data and vice versa. We present meaningful results for a
forward surrogate formulation that predicts simulation output based on the input
parameter space, as well as for the inverse approach that derives the input parameter
space based on simulation as well as experimental output quantities. This enables, e.g.,
a direct derivation of the input parameter space of a continuum dislocation dynamics
simulation based on experimentally provided stress-strain data.

Focused Test Generation for Autonomous Driving Systems

Testing Autonomous Driving Systems (ADSs) is crucial to ensure their reliability when navigating complex environments. ADSs may exhibit unexpected behaviours when presented, during operation, with driving scenarios containing features inadequately represented in the training dataset. To address this shift from development to operation, developers must acquire new data with the newly observed features. This data can be then utilised to fine tune the ADS, so as to reach the desired level of reliability in performing driving tasks. However, the resource-intensive nature of testing ADSs requires efficient methodologies for generating targeted and diverse tests. In this work, we introduce a novel approach, DeepAtash-LR , that incorporates a surrogate model into the focused test generation process. This integration significantly improves focused testing effectiveness and applicability in resource-intensive scenarios. Experimental results show that the integration of the surrogate model is fundamental to the success of DeepAtash-LR . Our approach was able to generate an average of up to 60 × more targeted, failure-inducing inputs compared to the baseline approach. Moreover, the inputs generated by DeepAtash-LR were useful to significantly improve the quality of the original ADS through fine tuning.

Integrative BNN-LHS Surrogate Modeling and Thermo-Mechanical-EM Analysis for Enhanced Characterization of High-Frequency Low-Pass Filters in COMSOL

Integrative BNN-LHS Surrogate Modeling and Thermo-Mechanical-EM Analysis for Enhanced Characterization of High-Frequency Low-Pass Filters in COMSOL

Prediction and optimisation of gasoline quality in petroleum refining: The use of machine learning model as a surrogate in optimisation framework

Prediction and optimisation of gasoline quality in petroleum refining: The use of machine learning model as a surrogate in optimisation framework

Integrating Graph Neural Network-Based Surrogate Modeling with Inverse Design for Granular Flows

Integrating Graph Neural Network-Based Surrogate Modeling with Inverse Design for Granular Flows

The research on the failure mechanism and dynamic reliability of lubricated clearance

In the operation of mechanisms with lubricated clearances, uncertain parameters such as the temperature of lubricant, driving speed, and clearance value will lead to dynamic variations for oil film thickness of lubricated clearance. The lubrication may fail when the oil film thickness of bearing surfaces is insufficient, thereby significantly affecting the performance of the mechanism. This paper presents a novel method for the dynamic reliability of lubricated clearances by improved response surface method (IRSM). Firstly, the mathematical model of lubricated clearance is established, and the oil film load of lubricated clearance is defined, then the dynamic model of mechanism considering lubricated clearances is built through Lagrange method. Secondly, considering the failure mechanism of the lubricated clearance, the surrogate model is established using the IRSM with minimum oil film thickness as the failure criterion. Finally, taking the 9-link mechanism with multiple lubricated clearances as a numerical example, the dynamic minimum oil film thickness of the lubricated clearance considering uncertain parameters is computed. The reliability of the lubricated clearance is investigated by the surrogate model, and the applicability of the model is verified by the Monte Carlo method (MCM). The results demonstrate that the IRSM has higher accuracy compared to the traditional response surface method (TRSM). This paper provides a theoretical basis for the study of failure mechanism and reliability of lubricated clearances, and also has practical significance for engineering applications.

Transient surrogate modeling of modally reduced structures with discontinuous loads and damping

AbstractIn this work, a surrogate model for structural, transient and discontinuously excited finite element method simulations is developed. This allows to reduce the computational effort of repeated calculations of identical models under different load cases. The architecture of the surrogate combines fully connected neural network layers with long short-term memory layers. For the reproduction of different damping ratios, a categorical variable is added to the continuous input data. Based on a recursive flow of the predicted data back to the input layer, long-term dependencies do not vanish due to short-input sequences. The system dimension is reduced by applying the model-order reduction technique for modal decomposition. The high accuracy of the surrogate and the reduction of computational costs are shown on an academic example of a cantilever beam and a real-world example of a robot. The advantages of our approach are illustrated in comparison with state-of-the-art surrogates for transient finite element analysis. By using the surrogate proposed in this study, oscillations due to discontinuous excitation of mechanical structures can be reproduced. For this purpose, only short-input sequences are necessary since the excitation of the oscillations does not have to be part of the input sequence during the whole duration of the oscillations. Due to the categorical variable for the damping ratio, the surrogate can account for the influence of different damping in parameter studies.

Detection and classification of human respiration under building debris model using VHF/UHF waves

AbstractDetecting live humans in buildings that have collapsed due to disasters and identifying their condition of health is of great importance for search and rescue operations. Although various methods have been used for this purpose, there are still critical challenges to ensure accurate and rapid life-saving operations. Immediate detection of the presence of living humans under debris combined with the assessment of their vital signs is a crucial factor. This research endeavors to introduce a previously unexplored method: the use of artificial neural network-based techniques to detect human respiration under building debris by generating novel simulation-derived electromagnetic data. To achieve this, a realistic three-dimensional debris model was integrated into an electromagnetic simulation program and complemented by a phantom simulating anterior–posterior body movements indicative of respiration. Measurements of magnitude and phase between 150 and 650 MHz were performed under different conditions. Using surrogate models based on artificial neural networks, noise with different signal-to-noise ratios within the selected frequencies was introduced. These models were used to perform two different steps. Firstly, the presence of respiration of living humans trapped under debris was successfully detected with a success rate of 99.97%. Secondly, the difficult task of classifying the respiration patterns of the human was accomplished with an impressive accuracy of 99.69%, providing a solid basis for proof of concept.

A Method for Straightening Distorted Giga-Cast Large Thin-Walled Components

Giga-casting, a revolutionary approach for manufacturing large, single-piece car body components from aluminium, has emerged as a potential game-changer in the automotive industry. However, these large, thin-walled castings are prone to distortions during solidification and heat treatment processes. Straightening these distortions is crucial to ensure structural integrity, facilitate downstream assembly, and maintain aesthetic qualities. This paper proposes a novel method for straightening giga-cast components using a multi-pin straightening machine. The machine’s versatility stems from its ability to adapt to various geometries through multiple strategically controlled straightening pins. This paper introduces the concept of a “straightening stroke decision algorithm” to achieve precise straightening and overcome the challenges of complex shapes. This algorithm determines the stroke length for each pin, combining a polynomial model representing the global stiffness of the component with a machine learning model that captures the stiffness changes arising from the current geometry. The effectiveness of the proposed approach is evaluated through comprehensive numerical experiments using finite element analyses. The straightening performance is assessed for the straightening algorithm with different machine learning models (deep neural network and XGBoost) and compared to a traditional optimisation method. The proposed surrogate models decided the straightening strokes so that the maximum remaining distortion became 0.02% of the largest dimension of each target geometry. The results of the numerical experiment showed that the proposed straightening method is suitable for straightening distortion in large thin-walled components.

Uncertainty quantification and global sensitivity analysis of bearing capacity of push-out specimen considering randomness in bond-slip behaviour

As an artificial composite material, the uncertainty in mechanical property of concrete significantly impacts the bond-slip behaviour of the shaped steel concrete structure. To investigate the effect of the uncertainty of concrete mechanical properties on the bearing capacity of push-out specimens, based on the data of a small number of push-out test with identical design parameters, the improved Bootstrap method is utilized to execute sample augmentation for the bond stress-slip curves, and the stochastic bond slip constitutive model is established after considering five parameters (K1, …, K5) in the deterministic bond-slip model as random variables. Simultaneously, considering the randomness in elastic modulus of concrete (Ec), uncertainty quantification (UQ) of the bearing capacities of the push-out specimens is conducted using the arbitrary polynomial chaos (aPC) method. Subsequently, the global sensitivity analysis of the ultimate and residual bearing capacities is accomplished through the aPC surrogate model. The results indicate that the probability distribution types of the parameters in the stochastic bond-slip constitutive model can be arbitrary when considering the randomness of concrete material, and the aPC method with at least a third-order expansion is recommended to execute UQ for the push-out test due to the strong nonlinearty between the bearing capacities and the random parameters. The ultimate bearing capacity is primarily affected by the parameter K2, followed by K4, K3, and K5. Whereas the residual bearing capacity is predominantly influenced by the parameter K4, followed by K5, the remaining parameters exhibit negligible influence. In contrast, K1 and Ec have the greatest influence on equivalent plastic strain of the concrete near the loading end of the push-out specimen.

Surrogate information transfer and fusion in high-dimensional expensive optimization problems

The Kriging surrogate model is less frequently employed in high-dimensional expensive problems than is the radial basis function (RBF) model. This discrepancy is attributed to the challenge of hyperparameter tuning within the Kriging covariance function, which leads to relatively worse predictive performance. However, the Kriging model still plays a crucial role in providing the predicted uncertainty for new point infill that cannot be replaced by the RBF model. To leverage the advantages of both models, a surrogate information transfer and fusion algorithm is presented. Surrogate information transfer introduces global sensitivity information from the constructed RBF model to the hyperparameter optimization of the Kriging model, reducing the dimensions of hyperparameter tuning and improving its approximation performance. Surrogate information fusion combines the RBF and Kriging models, with the RBF model providing more accurate predictions of unknown points for exploitation, while the Kriging model provides predicted uncertainty for swarm updating and new point infilling to ensure exploration of the solution space. Compared with several state-of-the-art algorithms, the proposed algorithm is evaluated on eight benchmark problems and a in real-world optimization case. The experimental results demonstrate the effectiveness of the surrogate information transfer and fusion methods and the significant superiority of the proposed algorithm over the compared algorithms.

Direct torque control for PMSM based on the RBFNN surrogate model of electromagnetic torque and stator flux linkage

Direct Torque Control (DTC) is widely used in motion control of motors. Classical DTC utilizes the Park transformation(dq-model) or observer to estimate the stator fluxlinkage and electromagnetic torque. However, the dq-model and observer lack descriptions of the cogging torque and magnetic saturation in permanent magnet synchronous motors (PMSM). This paper proposes a novel method based on the surrogate model to estimate stator fluxlinkage and torque for PMSM DTC. First, a finite element model (FEM) was established based on the parameters of PMSM. Using the FEM and Latin hypercube sampling (LHS), a surrogate model for estimating the stator fluxlinkage and torque of PMSM was constructed. The stator fluxlinkage and torque surrogate model, active flux observer, and dq current model under different control methods are constructed, respectively. The servo control parameters were tuned using the extended symmetric optimum algorithm, and simulation and experiments were conducted. Finally, electromechanical coupling models with a ball screw feed system were built based on various control and feedback methods. Simulation and experiment results indicate that the surrogate model reduces torque ripple, stator fluxlinkage offset, and feed system tracking error.

Efficient reliability assessment of slope stability with insightful sensitivity analysis concerning the safety factor

The limit state functions (LSFs) of slope stability are nonlinear, implicit, or derived from numerical simulations, and additional surrogate models are usually used for its reliability problems. Alternatively, an efficient numerical procedure using a variant of the HLRF-BFGS algorithm is proposed to address this issue without additional surrogate models. This proposed procedure simultaneously conducts reliability assessment and sensitivity analysis concerning the safety factor by suggested calculations of partial derivatives using the finite difference method (FDM). Two highly nonlinear numerical examples are presented to validate its convergence and to investigate the step length of the FDM. Results show that the proposed procedure can get considerable accuracy while the HLRF algorithm cannot converge, and its required iterations are about 30% and 81% less than those of the limited step-size iteration algorithm. In addition, a step length not greater than 0.1 is suggested after investigations. Application to a classical two-layered slope is illustrated. Reliability results show that the proposed procedure can get the same accuracy as those of surrogate models after only 5 iterative steps. Sensitivity results show that the cohesion of the upper soil is the only sensitivity factor, and the most possible potential slide surface lies in the upper soil.