Parameterization Strategy Research Articles

Development of a new parameterization strategy and GC parameters of halogenated hydrocarbons for PC-SAFT equation of state

Statistical Associating Fluid Theory (SAFT) equations of state (EoSs) have been extensively used in the prediction of fluid phase equilibria and thermodynamic properties. For each fluid, determining the component-dependent parameters typically involves fitting experimental data with a local optimization algorithm. SAFT-type EoSs are highly nonlinear due to the high-order functions used to describe different contribution terms. This nonlinearity leads to the presence of multiple local optima, making parameter optimization very sensitive to initial values. Hence, it is crucial to determine the starting point for the optimization process, yet little attention has been paid to how initial parameter sets are selected. In this paper, a method based on group contributions to establish an appropriate initial value for the optimization process is proposed and applied to Perturbed-Chain SAFT (PC-SAFT). The optimized PC-SAFT parameters for a total of 74 substances from 11 different chemical families have been evaluated. The fitting results for saturated pressure, liquid density, and vapor density showed overall average absolute relative deviations (AARD) of 0.050 %, 0.042 %, and 0.151 %, respectively. This paper also provided group contribution parameters for halogenated hydrocarbons to estimate PC-SAFT parameters. Additionally, an assessment of global and local optimization algorithms was conducted. The results demonstrate that the global algorithm not only requires longer computation time but also exhibits significantly lower accuracy compared to the local algorithm. The overall AARD for the global algorithm is 9.493 %, whereas for the local algorithm, it stands at 0.068 %.

Effects of rotor squealer tip with non-uniform heights on heat transfer characteristic and flow structure of turbine stage

Squealer tip has a significant influence on both aerodynamic and heat transfer characteristics of the high-pressure turbine. Among the geometric parameters of the squealer, squealer height is one of the essential parameters in the tip design. However, due to the complexity of parameterization and meshing of the squealer, the related research is usually carried out on the squealer with a constant height. In this paper, a parameterization strategy generates squealer of assigned heights at four key positions of the blade, the leading edge-pressure side, the leading edge-suction side, the trailing edge-pressure side, and the trailing edge-suction side. An in-house mesh generation platform (NuFlux) is adopted to automatically generate the structured meshes. The aerothermal performance of a transonic turbine stage is assessed using steady Reynolds-averaged Navier–Stokes simulations with the k−ω shear stress transport model for the turbulence closure. The main purpose is to obtain the squealer tip configuration with the lowest heat transfer coefficient. The results show that non-uniform squealer further reduces the cavity floor heat transfer on the basis of uniform squealer by changing the interaction process between the asymmetric vortex pair (the pressure-side corner vortex and the casing-driven scraping vortex), which provides a valuable reference for the design of the squealer tip of advanced high-pressure turbines.

Deep learning-based geological parameterization for history matching CO2 plume migration in complex aquifers

History matching is crucial for reliable numerical simulation of geological carbon storage (GCS) in deep subsurface aquifers. This study focuses on inferring highly complex aquifer permeability fields with multi- and intra-facies heterogeneity to improve the characterization of CO2 plume migration. We propose a deep learning (DL)-based parameterization strategy combined with the ensemble smoother with multiple data assimilation (ESMDA) algorithm to formulate an integrated inverse framework. The DL model is employed to parameterize non-Gaussian permeability fields using low-dimensional latent variables in a Gaussian distribution, thereby mitigating the non-Gaussianity issue faced by the ensemble-based ESMDA inverse method and simultaneously alleviating the computational burden of high-dimensional inversion. The efficacy of the integrated DL-ESMDA inverse framework is demonstrated using a 3-D GCS model, where it estimates the non-Gaussian permeability field characterized by multi- and intra-facies heterogeneity. Results show that the DL model is able to represent the highly complex and high-dimensional permeability fields using low-dimensional latent vectors. The DL-ESMDA framework sequentially updates these low-dimensional latent vectors instead of the original high-dimensional permeability field to obtain posterior estimations of the permeability field. The resulting CO2 plume migration closely matches historical measurements, suggesting a significantly improved model reliability after history matching. Additionally, a substantial reduction in uncertainty for future plume migration predictions beyond the history matching period is observed. The proposed framework provides an effective approach for reliable characterization of CO2 plume migration in highly heterogeneous aquifers, enhancing GCS project operation and risk analysis.

Practically fast finite-time stability of stochastic constrained nonlinear systems with actuator dead zone

This article addresses the challenge of achieving practically fast finite-time stabilization for stochastic constrained nonlinear systems, which are subject to both quantization effects and actuator dead zones. To tackle these issues, adaptive parameterization and partial control strategies are introduced with the aim of efficiently approximating and counteracting nonlinear disturbances. This approach ensures the robust stabilization of the controlled system within a finite time frame, despite the presence of uncertainties. Additionally, a novel barrier function is propose that mitigates the constraints usually imposed by boundary functions, while also leveraging a fuzzy logic system to manage nonlinear terms adeptly. Building on these innovations, we formulate a new theorem dedicated to practically fast finite-time control mechanisms for stochastic nonlinear systems. The efficacy of our theoretical developments is substantiated through an illustrative example involving a current-controlled DC motor system, demonstrating the practical applicability and robustness of the proposed control scheme.

Simplified Physics-Based Battery Model for Stationary Energy-Storage Applications

To perform safe battery operation and avoid unnecessary degradation, battery models are needed. How complex, or close to reality the model should be, depends on the purpose of the model and the computational power available. The possibility for accurate parameterization before model implementation is another crucial aspect for the validity. Physics-based models have the advantage to solve for internal battery states, which includes important information to avoid accelerated degradation and to evaluate state of health. But the parameterization effort can be extensive and the computational cost too high to solve for in real time. Identifying the most essential processes and if possible, simplify the model is therefore motivated.Parameter sensitivity is a powerful tool to investigate how much certain parameters and the related processes effect the output signal, i.e. battery voltage. Under normal operating conditions the cells are rarely completely discharged, and as a result the diffusion processes show low sensitivity during operation [1]. In this work we therefore explore the utilization of a physics-based model solving for the electrolyte dynamics, rather than the solid phase diffusion commonly applied in the so-called single particle model. By neglecting the diffusion processes, the second dimension in the pseudo-two-dimensional (P2D) model can be ignored and the computational complexity simplified. The parameterization strategy and validation of the model is further discussed.We evaluate our model by comparing simulations with experimental data from batteries subjected to stationary energy storage applications. The cells are commercial 18650-type NMC/Graphite cells with 2.6 Ah capacity. Different service cycles have different needs in terms of current and voltage, see examples in Figure 1. Capturing these behaviours with a physics-based model might pose different challenges and will further affect the degradation of the cells [2]. A wide range of models exist in literature to capture degradation mechanisms during battery operation [3]. To keep the computational and parameterization effort as low as possible, our methodology is not to include more physics, but instead update the parameters to the processes already included in the model [4]. We explore this strategy comparing the behaviour from different types of services, as well as batteries where the application changes, to see the model response and analyse the battery state of health, see Figure 2. Electrochemical techniques such as differential voltage analysis and electrochemical impedance spectroscopy is included to support our conclusions.Our results highlight the value of application considerations when designing battery models, rather than only focusing on physical extreme points. We present an alternative physics-based model with the ability to capture crucial degradation phenomena and validate it against operational data. We believe our findings can be of value for smarter battery operating strategies and a greater understanding of application dependant lithium-ion battery degradation.Figure 1. Some of the application duty cycles part of the experimental study. a) Current profile peak shaving b) Current profile frequency regulation c) Voltage response peak shaving d) Voltage response frequency regulation.Figure 2. Degradation trends from the experimental study. Purple cell initially performing PS following application change to FR_high cycling a) Capacity evolution of the cycled cells b) tortuosity parameter evolution obtained from model parameter fitting c) Measured impedance response during cell reference performance tests d) Calculated charge transfer resistance from measured impedance data.

ImputeVIS: An Interactive Evaluator to Benchmark Imputation Techniques for Time Series Data

With the emergence of The Internet of Things (IoT), smart sensors have become abundant in our daily lives. Failures are very common in those devices, leaving the recorded time series with missing blocks of consecutive values. A cottage industry of imputation algorithms exists, each with different performance tradeoffs. The diversity in time series features, missingness patterns, and algorithms' categories makes it challenging to select the best algorithm. In this demonstration, we showcase ImputeVIS, an analytical tool for benchmarking imputation algorithms. ImputeVIS provides an optimal configuration of those algorithms by implementing various AutoML parameterization strategies. Moreover, it uncovers the behavior of imputation algorithms by explaining the interplay between time series features and the imputation results. Its interactive web browser interface allows users to simulate real-world sensor malfunctions by contaminating time series with different missing block scenarios, deploy their imputation algorithms, and compare them against various popular imputation families.

Constellation design and performance of future quantum satellite gravity missions

Temporal aliasing is currently the largest error contributor to time-variable satellite gravity field models. Therefore, the evolution of sensor technologies has to be complemented by strategies to reduce temporal aliasing errors. The most straightforward way to improve temporal aliasing is through extended satellite constellations because they improve the observation geometry and increase the achievable temporal resolution. Therefore, strategies to optimize the design of larger satellite constellations are investigated in this contribution. A complete constellation modeling procedure is presented, starting from primary design variables (such as the required targeted resolutions) and concluding with concrete orbital elements for the individual satellites. In parallel, it is evaluated if improved instrument sensitivities based on quantum technologies (cold atom interferometry) can be fully exploited in the case of larger constellations. For this, future quantum satellite gravity missions adopting the gradiometry concept (similar to the GOCE mission) and the low-low satellite-to-satellite tracking concept (similar to GRACE/-FO) are simulated on optimized constellations with up to 6 satellites/pairs. The retrieval performance of a 6-pair mission in terms of the global equivalent water height RMS can be improved by a factor of roughly 3 compared to an inclined double-pair mission. 3D-gradiometry intrinsically has a better de-aliasing behavior but has extremely high accuracy requirements for the gradiometer (about 10 µEotvos) and the attitude reconstruction to be of any benefit. All simulations show that when incorporating improved sensor technologies, such as future quantum sensing instruments in extended constellations, temporal aliasing will remain the dominant error source by far, up to five orders of magnitude larger than the instrument errors. Therefore, improving sensor technologies has to go hand in hand with larger satellite constellations and improved space–time parameterization strategies to further reduce temporal aliasing effects.Graphical

Berkeley‐RWAWC: A New CYGNSS‐Based Watermask Unveils Unique Observations of Seasonal Dynamics in the Tropics

AbstractThe UC Berkeley Random Walk Algorithm WaterMask from CYGNSS (Berkeley‐RWAWC) is a new data product designed to address the challenges of monitoring inundation in regions hindered by dense vegetation and cloud cover as is the case in most of the Tropics. The Cyclone Global Navigation Satellite System (CYGNSS) constellation provides data with a higher temporal repeat frequency compared to single‐satellite systems, offering the potential for generating moderate spatial resolution inundation maps with improved temporal resolution while having the capability to penetrate clouds and vegetation. This paper details the development of a computer vision algorithm for inundation mapping over the entire CYGNSS domain (37.4°N–37.4°S). The sole reliance on CYGNSS data sets our method apart in the field, highlighting CYGNSS's indication of water existence. Berkeley‐RWAWC provides monthly, low‐latency inundation maps starting in August 2018 and across the CYGNSS latitude range, with a spatial resolution of 0.01° × 0.01°. Here we present our workflow and parameterization strategy, alongside a comparative analysis with established surface water data sets (SWAMPS, WAD2M) in four regions: the Amazon Basin, the Pantanal, the Sudd, and the Indo‐Gangetic Plain. The comparisons reveal Berkeley‐RWAWC's enhanced capability to detect seasonal variations, demonstrating its usefulness in studying tropical wetland hydrology. We also discuss potential sources of uncertainty and reasons for variations in inundation retrievals. Berkeley‐RWAWC represents a valuable addition to environmental science, offering new insights into tropical wetland dynamics.

Effects of hydraulic conductivity on simulating groundwater–land surface interactions over a typical endorheic river basin

Hydraulic conductivity is the key variable in subsurface configuration that is important in modeling groundwater and its interactions with land surface processes. However, an acquisition of quantitatively distributed hydraulic conductivity is challenging due to limitations on data availability, accuracy and resolution, which inhibits an advanced understanding of hydrologic and energy cycles at the basin scale. This study investigated the role of hydraulic conductivity in simulating groundwater-land surface interactions over a typical endorheic river basin in northwestern China. A 3D variably saturated groundwater model coupled to a land surface model was configured to account for interactions between groundwater dynamics and land energy fluxes. Four scenarios of hydraulic conductivity were tested and inter-compared, adopting both homogeneous and heterogeneous parameterization strategies based on Global Soil Dataset for use in Earth System Models (GSDE) and HYdrogeology MaPS (GLHYMPS). The overall objectives are to diagnose effects of hydraulic conductivity on simulating groundwater–land surface interactions, with a particular focus on differences in spatial distributions and statistical characteristics between scenarios, as well as the dynamics and feedback mechanisms between groundwater and land surface energy fluxes. Comprehensive simulations and analyses were conducted to discern spatial distributions and statistical characteristics of key hydrological variables (streamflow, water table depth, latent heat flux, sensible heat flux) under different scenarios. Results showed that the hydraulic conductivity of the saturated zone had a significant control over water table depth, while the combine hydraulic conductivity in the vadose zone and the saturated zone governed the evolution of streamflow. Moreover, the water table critical range—where the land energy budget is highly sensitive to groundwater table—was 0.1 m–5 m and the area might expand approximately 4.5 % after considering the heterogeneity of hydraulic conductivity. Compared to the homogeneous scenario, the spatial heterogeneity of hydraulic conductivity amplified water table depth variability while diminishing land energy fluxes variability.

Data-Driven Nonintrusive Model-Order Reduction for Aerodynamic Design Optimization

Fast and accurate evaluation of aerodynamic characteristics is essential for aerodynamic design optimization because aircraft programs require many years of design and optimization. Therefore, it is imperative to develop sufficiently fast, robust, and accurate computational tools for industry routine analysis. This paper presents a nonintrusive machine-learning method for building reduced-order models (ROMs) using an autoencoder neural network architecture. An optimization framework was developed to identify the optimal solution by exploring the low-dimensional subspace generated by the trained autoencoder. To demonstrate the convergence, stability, and reliability of the ROM, a subsonic inverse design problem and a transonic drag minimization problem of the airfoil were studied and validated using two different parameterization strategies. The robustness and accuracy demonstrated by the method suggest that it is valuable in parametric studies, such as aerodynamic design and optimization, and requires only a small fraction of the cost of full-order modeling.

Systematic evaluation of high-throughput PBK modelling strategies for the prediction of intravenous and oral pharmacokinetics in humans

Physiologically based kinetic (PBK) modelling offers a mechanistic basis for predicting the pharmaco-/toxicokinetics of compounds and thereby provides critical information for integrating toxicity and exposure data to replace animal testing with in vitro or in silico methods. However, traditional PBK modelling depends on animal and human data, which limits its usefulness for non-animal methods. To address this limitation, high-throughput PBK modelling aims to rely exclusively on in vitro and in silico data for model generation. Here, we evaluate a variety of in silico tools and different strategies to parameterise PBK models with input values from various sources in a high-throughput manner. We gather 2000 + publicly available human in vivo concentration–time profiles of 200 + compounds (IV and oral administration), as well as in silico, in vitro and in vivo determined compound-specific parameters required for the PBK modelling of these compounds. Then, we systematically evaluate all possible PBK model parametrisation strategies in PK-Sim and quantify their prediction accuracy against the collected in vivo concentration–time profiles. Our results show that even simple, generic high-throughput PBK modelling can provide accurate predictions of the pharmacokinetics of most compounds (87% of Cmax and 84% of AUC within tenfold). Nevertheless, we also observe major differences in prediction accuracies between the different parameterisation strategies, as well as between different compounds. Finally, we outline a strategy for high-throughput PBK modelling that relies exclusively on freely available tools. Our findings contribute to a more robust understanding of the reliability of high-throughput PBK modelling, which is essential to establish the confidence necessary for its utilisation in Next-Generation Risk Assessment.

3D Monte Carlo geometry inversion using gravity data

Diverse Monte Carlo methods have gained widespread use across a broad range of applications. However, the challenge of 3D Monte Carlo sampling remains due to the curse of dimensionality. To date, only a few works have been published regarding 3D Monte Carlo sampling. This study aims to develop an efficient 3D transdimensional Monte Carlo framework for reconstructing the spatial geometry of an anomalous body using gravity data. Our framework also can quantify the uncertainty of the shape of an anomalous body recovered from geophysical measurements. To improve the computational efficiency of 3D Monte Carlo sampling, we develop a sparse geometry parameterization strategy. This approach adequately approximates the shape of a complex 3D anomalous body using a set of simple geometries, such as ellipsoids. Each ellipsoid can be characterized by a few parameters, such as the centroid, axes, and orientations, significantly reducing the number of parameters to be sampled. During the sampling, we randomly perturb the number, locations, sizes, and orientations of the ellipsoids. To impose prior structural constraints from other geophysical methods, such as seismic imaging, we design a new method by placing a fixed layer oriented along the top boundary of the anomalous body. The fixed layer is then connected to the randomly sampled ellipsoids using an alpha shape, allowing us to estimate the geometry of the anomalous source body. The current work focuses on the reconstruction of salt bodies. We start with a synthetic spherical salt model and then conduct a more realistic study using a simplified 3D SEG/EAGE salt model, which has a much more complex geometry than the synthetic spherical model. Finally, we apply our method to the Galveston Island salt dome, offshore Texas. The numerical results demonstrate that our framework can effectively recover the shape of an anomalous body and quantify the uncertainty of the reconstructed geometry.

Perturbative Expansion in Reciprocal Space: Bridging Microscopic and Mesoscopic Descriptions of Molecular Interactions.

Determining the Fourier representation of various molecular interactions is important for constructing density-based field theories from a microscopic point of view, enabling a multiscale bridge between microscopic and mesoscopic descriptions. However, due to the strongly repulsive nature of short-ranged interactions, interparticle interactions cannot be formally defined in Fourier space, which renders coarse-grained (CG) approaches in k-space somewhat ambiguous. In this paper, we address this issue by designing a perturbative expansion of pair interactions in reciprocal space. Our perturbation theory, starting from reciprocal space, elucidates the microscopic origins underlying zeroth-order (long-range attractions) and divergent repulsive interactions from higher order contributions. We propose a systematic framework for constructing a faithful Fourier-space representation of molecular interactions, capturing key structural correlations in various systems, including simple model systems and molecular CG models of liquids. Building upon the Ornstein-Zernike equation, our approach can be combined with appropriate closure relations, and to further improve the closure approximations, we develop a bottom-up parameterization strategy for inferring the bridge function from microscopic statistics. By incorporating the bridge function into the Fourier representation, our findings suggest a systematic, bottom-up approach to performing coarse-graining in reciprocal space, leading to the systematic construction of a bottom-up classical field theory of complex aqueous systems.

Extension of the D3 and D4 London dispersion corrections to the full actinides series.

Efficient dispersion corrections are an indispensable component of modern density functional theory, semi-empirical quantum mechanical, and even force field methods. In this work, we extend the well established D3 and D4 London dispersion corrections to the full actinides series, francium, and radium. To keep consistency with the existing versions, the original parameterization strategy of the D4 model was only slightly modified. This includes improved reference Hirshfeld atomic partial charges at the ωB97M-V/ma-def-TZVP level to fit the required electronegativity equilibration charge (EEQ) model. In this context, we developed a new actinide data set called AcQM, which covers the most common molecular actinide compound space. Furthermore, the efficient calculation of dynamic polarizabilities that are needed to construct CAB6 dispersion coefficients was implemented into the ORCA program package. The extended models are assessed for the computation of dissociation curves of actinide atoms and ions, geometry optimizations of crystal structure cutouts, gas-phase structures of small uranium compounds, and an example extracted from a small actinide complex protein assembly. We found that the novel parameterizations perform on par with the computationally more demanding density-dependent VV10 dispersion correction. With the presented extension, the excellent cost-accuracy ratio of the D3 and D4 models can now be utilized in various fields of computational actinide chemistry and, e.g., in efficient composite DFT methods such as r2SCAN-3c. They are implemented in our freely available standalone codes (dftd4, s-dftd3) and the D4 version will be also available in the upcoming ORCA 6.0 program package.

A vector encoding for topology finding of structured quad-based patterns for surface structures

Topology-optimisation strategies for structural design of discrete surface structures result in unstructured patterns that require post-rationalisation to limit the number and complexity of the various structural elements. Fabrication-related objectives still demand further processing. Designers need topology exploration, before topology optimisation, following a generative-design approach that is decoupled from density tuning and shape design, to produce high-quality patterns. Such an approach allows to flexibly embed multiple design constraints, benefit from state-of-the-art form-finding algorithms and explore design trade-offs of the multiple requirements related to architecture, engineering and construction. This research investigates a parameterisation strategy to encode topological exploration of structured patterns based on quad meshes. The focus is set on singular vertices, which connect an irregular number of blocks, cables, or beams. Singularities are independent from pattern density and geometry, and have a fundamental influence on the qualitative and quantitative performance of the structure. We introduce an L-system encoding a quad-mesh grammar into a string that describes topological transformations of these singularities. Design applications to a net and a gridshell demonstrate the influence of singularities on the design of surface structures, and highlight the flexibility and generality of the approach in terms of geometrical processing and performance metrics. Beyond exploration, this parameterisation strategy opens to novel applications of search and optimisation methods for generative design of singularities in structural patterns.

Comparison of different pilot point parameterization strategies when measurements are unevenly distributed in space

The parameterization of spatially distributed hydraulic properties is one of the most crucial steps in groundwater modeling. A common approach is to estimate hydraulic properties at a set of pilot points and interpolate the values at each model cell. Despite the popularity of this method, several questions remain about the optimum number and distribution of pilot points, which are determining factors for the efficiency of the method. This study proposes a strategy for optimal pilot point parameterization that minimizes the number of parameters while maximizing the assimilation of an observed dataset unevenly distributed in space. The performance of different pilot point distributions has been compared with a synthetic groundwater model, considering regular grids of pilot points with different spacings and adaptive grids with different refinement criteria. This work considered both prior and iterative refinements, with a parameter estimation step between successive refinements. The parameter estimation was conducted with the Gauss–Levenberg–Marquardt algorithm, and the strategies were ranked according to the number of model calls to reach the target objective function. The strategy leading to the best fit with the measurement dataset at the minimum computational burden is an adaptive grid of pilot points with prior refinement based on measurement density. This strategy was successfully implemented on a regional, multilayered groundwater flow model in the south-western geological basin of France.

A novel spatial–temporal generative autoencoder for wind speed uncertainty forecasting

Wind speed interval prediction is one of the most long-standing challenges because of the high uncertainty and the complex spatial–temporal correlation between wind turbines. In this paper, based on variational Bayesian inference, we propose a novel spatial–temporal generative autoencoder (STGAE) model to capture the continuous probability distribution of each wind turbine on an arbitrary graph structure. In the encoding process, a parameterization strategy is used to expand the hypothesis space of the adjacency matrix rather than being limited by the distance between nodes. The multiple graph convolution layers with kernels approximated by Chebyshev polynomials adjust the collecting scope of spatial information. The bi-directional dependency is modeled by the multi-head attention to tap correlations of different time horizons effectively. The interval prediction at each node is yielded based on the probability densities of generated samples produced by the decoder. Using publicly available wind data from the Global Energy Forecasting Competition 2014, numerical experiments are conducted to compare the prediction reliability and accuracy between the proposed model and the recent state-of-the-art models. The forecasting results suggest that the proposed model significantly improves performance, especially the strong generalization ability makes it practical under different weather conditions.

AutoEER: automatic EEG-based emotion recognition with neural architecture search

Objective. Emotion recognition based on electroencephalography (EEG) is garnering increasing attention among researchers due to its wide-ranging applications and the rise of portable devices. Deep learning-based models have demonstrated impressive progress in EEG-based emotion recognition, thanks to their exceptional feature extraction capabilities. However, the manual design of deep networks is time-consuming and labour-intensive. Moreover, the inherent variability of EEG signals necessitates extensive customization of models, exacerbating these challenges. Neural architecture search (NAS) methods can alleviate the need for excessive manual involvement by automatically discovering the optimal network structure for EEG-based emotion recognition. Approach. In this regard, we propose AutoEER (Automatic EEG-based Emotion Recognition), a framework that leverages tailored NAS to automatically discover the optimal network structure for EEG-based emotion recognition. We carefully design a customized search space specifically for EEG signals, incorporating operators that effectively capture both temporal and spatial properties of EEG. Additionally, we employ a novel parameterization strategy to derive the optimal network structure from the proposed search space. Main results. Extensive experimentation on emotion classification tasks using two benchmark datasets, DEAP and SEED, has demonstrated that AutoEER outperforms state-of-the-art manual deep and NAS models. Specifically, compared to the optimal model WangNAS on the accuracy (ACC) metric, AutoEER improves its average accuracy on all datasets by 0.93%. Similarly, compared to the optimal model LiNAS on the F1 Ssore (F1) metric, AutoEER improves its average F1 score on all datasets by 4.51%. Furthermore, the architectures generated by AutoEER exhibit superior transferability compared to alternative methods. Significance. AutoEER represents a novel approach to EEG analysis, utilizing a specialized search space to design models tailored to individual subjects. This approach significantly reduces the labour and time costs associated with manual model construction in EEG research, holding great promise for advancing the field and streamlining research practices.

A modified Jarvis model to improve the expressing of stomatal response in a beech forest

AbstractThe Jarvis‐type model, which incorporates stress functions, is commonly used to describe the physiological behaviour of stomatal response in various vegetation species. However, the model has been criticized for its empirically formulated multiplicative equation, which may not accurately capture the mutual impact of intercorrelated stress factors, for example, vapour pressure deficit (VPD) and air temperature (Ta). This study proposed a modified Jarvis model that introduces reduction factors in the stress functions of VPD and Ta to provide the description of canopy conductance. We used sap flow data from a beech forest in the mid‐latitude region of Centre Europe to inversely estimate the canopy conductance with optimized stress functions. Our findings reveal that two recommended parameterization strategies for general deciduous broadleaf forest (DBF) significantly overestimated the transpiration rate, with a maximum value of ~2 mm/day on rainless days. This suggested that the beech forest exhibited a distinct stomatal response compared to the general DBF category. By applying boundary line analysis to fit the parameters, both the unmodified and modified Jarvis models provided better simulations of transpiration, with relatively high Nash‐Sutcliffe Efficiency (NSE) values of 0.75 and 0.77, respectively. These results indicated that modelling transpiration can be improved by refining the parameterization of canopy conductance, particularly for vegetation species with unique stomatal behaviours that deviated from the characteristics of their general vegetation type. The modified Jarvis model offers a more accurate description of canopy conductance and enhances the modelling of transpiration in vegetated areas, especially under dry environment conditions with relatively high VPD.

Migration velocities of intergranular He gas bubbles under thermal gradients in Fe by phase-field modeling

We compute the migration velocity of intergranular He bubbles in Fe under thermal gradients using phase-field (PF) modeling. Grain boundaries (GBs) with energies in the range of 0.88−1.6J/m2 are considered. To develop a suitable PF model, the influences of the PF parameterization strategy, thermal conductivity in grain boundaries, and GB diffusion coefficients on intergranular He bubble geometry are addressed. Good agreement between our simulations and a theoretical model are found for the contact angle at the junction of the bubble and GB. Under the application of a thermal gradient, we find that intergranular He bubbles migrate slower than He bubbles in an Fe matrix. We show that this is a result of the smaller temperature gradient at the bubble/GB junction when the bubble is at the GB, than when the bubble is in the matrix. A linear relationship between bubble migration velocity and temperature gradient is observed for intergranular He bubbles for all GB energies studied in this work, consistent with predictions from a theoretical model. Intergranular bubble migration velocity is also found to depend linearly on the inverse of bubble size. However, we find the shape of the bubble influences intergranular bubble migration velocity. An analytical model for bubble migration velocity that considers the shape of an intergranular bubble is proposed and shown to capture the computed PF data for He bubbles.