Wei Chen

Active suspension systems are critical for enhancing vehicle comfort, safety, and stability, yet their performance is often limited by fixed hardware designs and control strategies that cannot adapt to uncertain and dynamic operating conditions. Recent advances in Digital Twins (DTs) and Reinforcement Learning (RL) offer new opportunities for real-time, data-driven optimization across a vehicle's lifecycle. However, integrating these technologies into a unified framework for co-optimizing physical and control systems remains an open challenge. This work presents an RL-based Control Co-Design (CCD) framework for full-vehicle active suspensions using multi-generation design and DT concepts. Through integrating automatic differentiation into Deep Reinforcement Learning (DRL), we jointly optimize physical components of suspension systems and control policies under varying driver behaviors and environmental uncertainties. The DRL technique also addresses the challenge of partial observability, where only limited states can be sensed and fed back to the controller, by learning optimal control actions directly from available sensor information. The framework incorporates model updating with quantile learning to quantify data uncertainty, enabling real-time decision-making and adaptive learning from digital-physical interactions. The approach demonstrates personalized optimization of autonomous suspension systems under two distinct driving settings (mild and aggressive). The results show that the optimized systems achieve smoother trajectories and reduce control efforts by approximately 58% and 12% for mild and aggressive while improving ride comfort by approximately 17% and 28%, respectively. Contributions of this work include: (1) developing a DT-enabled CCD framework integrating DRL and uncertainty-aware model updating for full-vehicle active suspensions, (2) introducing a multi-generation design framework for self-improving systems across the whole lifecycle, and (3) demonstrating personalized optimization of active suspension systems for distinct types of drivers.

This paper presents an innovative approach to train timetable generation using Monte Carlo tree search (MCTS) integrated with a deep reinforcement learning technique. The generation and adjustment of train timetables for high-speed railways represent a complex optimisation problem with numerous rule-based constraints that traditional mathematical methods struggle to solve efficiently. Therefore, the train timetable generation problem is modelled as a discrete spatiotemporal Markov decision process, and a comprehensive MCTS-based algorithm is developed to effectively balance exploration and exploitation through a structured tree search mechanism. The result of the comparative analysis demonstrates that MCTS-based algorithms significantly outperform state-of-the-art reinforcement learning algorithms, including double deep Q-network (DDQN) and proximal policy optimisation (PPO), achieving optimal solutions 6.5 times faster with superior training stability. To validate the scalability and real-world applicability, a large-scale case study involving 120 pairs of trains on the Beijing-Shanghai High-Speed Rail corridor over an 18-hour period successfully resolved all 45,600 initial conflicts. The optimised timetables yield significant operational improvements, including a 16.4% reduction in average delay time, 22.8% improvement in track utilisation efficiency and 9.7% reduction in energy consumption. This research contributes to the advancement of intelligent railway operations optimisation and demonstrates the potential of MCTS-based approaches to transform complex transportation problems.

• Consensus-based multi-agent Bayesian optimization (BO) explores heterogeneous spaces. • Collaboration boosts efficiency, benchmarks show when simpler BO suffices. • Multi-agent BO accelerates discovery of HfTiTaNb alloys with target properties. Adaptive learning implementations for materials design are challenged by the complex, nonlinear relationships between composition and properties, particularly in high-performance applications such as high-temperature compositionally complex refractory alloys. Traditional Bayesian optimization (BO) methods, which typically rely on a single Gaussian Process (GP) surrogate, often struggle to model heterogenous behaviors across the design domain. To address this limitation, we introduce collaborative BO as a multi-agent framework for materials discovery. In the context of optimizing compositions for desired properties, each agent models a specific subregion of the design space, where subregions share similar property trends, and exchanges information with the other agents to expedite exploration and design optimization. Comparative evaluations demonstrate that, when compared to single-agent BO and other approaches discussed in this article multi-agent BO allows flexible information-sharing protocols and effectively reduces iterations of adaptive learning while reliably delivering designs that meet the targeted mechanical properties. These findings provide novel insights into the behavior of refractory multi-component alloys, using the Hf-Ti-Ta-Nb system as a case study, and illustrate the potential of adaptive multi-agent learning in efficiently screening extensive materials libraries. Moreover, the framework is broadly applicable to other problems characterized by diverse data sources, where advanced optimization strategies are essential for accelerated materials discovery.

Abstract Spinodoid architected materials have drawn significant attention due to their unique nature in stochasticity, aperiodicity, and bi-continuity. Compared to classic periodic truss-, beam-, and plate-based lattice architectures, spinodoids are insensitive to manufacturing defects, scalable for high-throughput production, functionally graded by tunable local properties, and material failure resistant due to low-curvature morphology. However, the design of spinodoids is often hindered by the curse of dimensionality with an extremely large design space of spinodoid types, material density, orientation, continuity, and anisotropy. From a design optimization perspective, while genetic algorithms are often beyond the reach of computing capacity, gradient-based topology optimization is challenged by the intricate mathematical derivation of gradient fields with respect to various spinodoid parameters. To address such challenges, we propose a data-driven multiscale topology optimization framework. Our framework reformulates the design variables of spinodoid materials as the parameters of neural networks, enabling automated computation of topological gradients. Additionally, it incorporates a Gaussian Process surrogate for spinodoid constitutive models, eliminating the need for repeated computational homogenization and enhancing the scalability of multiscale topology optimization. Compared to ‘black-box’ deep learning approaches, the proposed framework provides clear physical insights into material distribution. It explicitly reveals why anisotropic spinodoids with tailored orientations are favored in certain regions, while isotropic spinodoids are more suitable elsewhere. This interpretability helps to bridge the gap between data-driven design with mechanistic understanding. To this end, we test our design framework on several numerical experiments. We find our multiscale spinodoid designs with controllable anisotropy achieve better performance than single-scale isotropic counterparts, with clear physics interpretations.

Despite remarkable progress, multimodal foundation models still exhibit surprising deficiencies in spatial intelligence. In this work, we explore scaling up multimodal foundation models to cultivate spatial intelligence within the SenseNova-SI family, built upon established multimodal foundations including visual understanding models (i.e., Qwen3-VL and InternVL3) and unified understanding and generation models (i.e., Bagel). We take a principled approach to constructing high-performing and robust spatial intelligence by systematically curating SenseNova-SI-8M: eight million diverse data samples under a rigorous taxonomy of spatial capabilities. SenseNova-SI demonstrates unprecedented performance across a broad range of spatial intelligence benchmarks: 68.8% on VSI-Bench, 43.3% on MMSI, 85.7% on MindCube, 54.7% on ViewSpatial, 47.7% on SITE, 63.9% on BLINK, 55.5% on 3DSR, and 72.0% on EmbSpatial, while maintaining strong general multimodal understanding (e.g., 84.9% on MMBench-En). More importantly, we analyze the impact of data scaling, discuss early signs of emergent generalization capabilities enabled by diverse data training, analyze the risk of overfitting and language shortcuts, present a preliminary study on spatial chain-of-thought reasoning, and validate the potential downstream application. All newly trained multimodal foundation models are publicly released.

Polymer nanocomposites, inherently tailorable materials, are potentially capable of providing higher strength to weight ratio than conventional hard metals. However, their disordered nature makes processing control and hence tailoring properties to desired target values a challenge. Additionally, the interfacial region, also called the interphase, is a critical material phase in these heterogeneous materials and its extent depends on variety of microstructure features like particle loading and dispersion or inter-particle distances. Understanding process-structure–property (PSP) relation can provide guidelines for process and constituents’ design. Our work explores nuances of PSP relation for polymer nanocomposites with attractive pairing between particles and the bulk polymer. Past works have shown that particle functionalization can help tweak these interactions in attractive or repulsive type and can cause slow or fast decay of stiffness properties in polymer nanocomposites. In this work, we develop a material model that can represent decay for small strain elastoplastic(Young’s modulus and yield strength) properties in interfacial regions and simulate representative or statistical volume element behavior. The interfacial elastoplastic material model is devised by combining local stiffness and glass transition measurements from atomic force microscopy and fluorescence microscopy. This model is combined with a microstructural design of experiments for agglomerated nanocomposite systems. Agglomerations are particle aggregations arising from processing artifacts. Twin screw extrusion process can reduce extent of aggregation in hot pressed samples via erosion or rupture depending on screw rpms and torque. We connect this process-structure relation to structure–property relation that emerges from our study. We discover that balancing between local stress concentration zones (SCZ) and interfacial property decay governs how fast yield stress can improve by breaking down agglomeration via erosion. Rupture is relatively less effective in helping improve nanocomposite yield strength. We also observe an inflection point where incremental increase brought on by rupture is slowed due to increasing SCZ and saturation in interphase percolation.

Abstract High-dimensional structure and composition spaces pose a fundamental challenge in materials discovery due to the lack of efficient approaches for navigating the vast and complex design space. Although machine learning (ML) has aided materials discovery, most existing ML models lack the ability to quantify epistemic uncertainty arising from limited data. Developing this capability is particularly challenging for tasks involving high-dimensional design representations, such as atomic structures. In this study, building on the Bayesian optimization (BO) framework, we propose an uncertainty-aware atomistic machine learning model, uncertainty-aware PointNet, which enables automated representation learning directly from high-dimensional design inputs, such as atomic structures, and achieves principled uncertainty quantification through the use of spectral-normalized neural Gaussian process. By utilizing a constrained expected improvement acquisition function, our BO framework simultaneously considers multiple design criteria. We demonstrate the effectiveness of our approach in two materials discovery case studies: (1) identifying catalysts for the carbon dioxide reduction reaction and (2) designing transparent conducting materials. The results show that our approach achieves high prediction accuracy, facilitates interpretable feature extraction, and enables multicriteria material design using constrained BO, leading to a significant reduction of computing power and time (a 10× reduction in required simulation calculations). Beyond the demonstration examples, the developed method can accelerate materials discovery for various other applications with high-dimensional design inputs and expensive physics-based simulations.

In scientific machine learning (SciML), a key challenge is learning unknown, evolving physical processes and making predictions across spatio-temporal scales. For example, in real-world manufacturing problems like additive manufacturing, users adjust known machine settings while unknown environmental parameters simultaneously fluctuate. To make reliable predictions, it is desired for a model to not only capture long-range spatio-temporal interactions from data but also adapt to new and unknown environments; traditional machine learning models excel at the first task but often lack physical interpretability and struggle to generalize under varying environmental conditions. To tackle these challenges, we propose the Attention-based Spatio-Temporal Neural Operator (ASNO), a novel architecture that combines separable attention mechanisms for spatial and temporal interactions and adapts to unseen physical parameters. Inspired by the backward differentiation formula (BDF), ASNO learns a transformer for temporal prediction and extrapolation and an attention-based neural operator for handling varying external loads, enhancing interpretability by isolating historical state contributions and external forces, enabling the discovery of underlying physical laws and generalizability to unseen physical environments. Empirical results on SciML benchmarks demonstrate that ASNO outperforms over existing models, establishing its potential for engineering applications, physics discovery, and interpretable machine learning.

Computational science and engineering are shifting toward data-centric, optimization-based, and self-correcting solvers with artificial intelligence. This transition faces challenges such as low accuracy with sparse data, poor scalability, and high computational cost in complex system design. This work introduces Interpolating Neural Network (INN)-a network architecture blending interpolation theory and tensor decomposition. INN significantly reduces computational effort and memory requirements while maintaining high accuracy. Thus, it outperforms traditional partial differential equation (PDE) solvers, machine learning (ML) models, and physics-informed neural networks (PINNs). It also efficiently handles sparse data and enables dynamic updates of nonlinear activation. Demonstrated in metal additive manufacturing, INN rapidly constructs an accurate surrogate model of Laser Powder Bed Fusion (L-PBF) heat transfer simulation. It achieves sub-10-micrometer resolution for a 10 mm path in under 15 minutes on a single GPU, which is 5-8 orders of magnitude faster than competing ML models. This offers a new perspective for addressing challenges in computational science and engineering.

The revitalization of Hanfu, the traditional Chinese dress, has recently experienced a marked cultural revival that cannot be reduced to the sphere of fashion but rather points to the complex psychological preconditions of identity construction, heritage maintenance, and consumer choice among the representatives of different generational groups. The current study employs a composite Kano Model-Analytic Hierarchy Process (Kano-AHP) framework to explore intergenerational differences in psychological needs and decision-making priorities upon which Hanfu consumption is based. Through rigorous quantitative analysis of survey data from 552 Chinese consumers across four generational cohorts (Gen Z, Millennials, Gen X, Baby Boomers), the research categorizes nine core Hanfu attributes into Must-Be, One-Dimensional, and Attractive needs using the Kano Model, while quantifying their relative importance in purchase decisions through AHP weighting. Results reveal significant intergenerational divergence (χ² = 127.43, p < 0.001) in attribute perception and prioritization. Younger consumers (Gen Z, Millennials) prioritize identity expression and aesthetic innovation, classifying Modern Design Integration (weight: 0.248) and Community Acceptance (weight: 0.201) as primary "Attractive" delighters. Conversely, older cohorts (Gen X, Boomers) emphasize cultural authenticity and functional comfort, with Historical Accuracy (weight: 0.324) and Fabric Quality (weight: 0.284) categorized as essential "Must-Be" requirements. A unified framework provides practical advice to the designers, marketers and cultural institutions so as to develop generation-specific strategies that would synchronise product development with the generation-specific psychological profiles and decision-making calculi, thus developing sustainable cultural engagement and market growth.

Responsive materials offer solutions to complex engineering challenges by enabling systems to adapt their shapes or properties in response to external stimuli. To fully harness the potential of responsive materials, inverse design methods that integrate multiple types of stimuli and manufacturing processes are necessary. We present a unified, autonomous codesign framework that simultaneously optimizes structure, manufacturing, materials, and stimuli for responsive material systems, achieving target shape morphing under multiple stimuli without relying on human heuristics or expertise. It integrates generalized topology optimization with hybrid data-physics differentiable simulations to achieve flexible, manufacturing-aware designs for network-like responsive material systems. We showcase our framework with a multimaterial three-dimensional printing process with high material tunability, which we use to fabricate liquid crystal elastomer systems that morph into different forms in response to heat and light. The exceptional flexibility and efficiency of our method will advance shape-morphing applications spanning soft robotics to drug delivery.

Abstract Digital Twins enable real-time monitoring, state prediction, model updates, and decision-making, yet achieving real-time control under uncertainty remains challenging due to computational limitations in uncertainty quantification (UQ). Model Predictive Control (MPC), as one embodiment of the decision-making for Digital Twins, offers a promising solution, but conventional approaches suffer from oversimplifying model representation, conservative UQ, or high computational costs in estimating uncertainty. This work presents a simultaneous multistep robust MPC framework, integrating a time series deep learning model with quantile regression to perform real-time decision-making while efficiently quantifying data uncertainty. Unlike conventional MPC that relies on recursive single-step predictions, our model, Time-Series Dense Encoder (TiDE), enables multistep-ahead predictions in one shot, significantly accelerating MPC computation. With the learned quantile for UQ, our method can significantly reduce constraint violations under stochastic disturbances by reformulating the robust MPC problem into a deterministic optimization problem with safety buffers using tightened constraints. We demonstrate our method in an additive manufacturing case study, showing that TiDE is capable of learning and quantifying uncertainty for highly nonlinear dynamics, and reduces constraint violations while maintaining computational efficiency. This framework establishes a foundation for uncertainty-aware, real-time decision-making in Digital Twin applications, advancing intelligent control strategies in complex engineering systems.

Abstract Digital twins, virtual replicas of physical systems that enable real-time monitoring, model updates, predictions, and decision-making, present novel avenues for proactive control strategies for autonomous systems. However, achieving real-time decision-making in digital twins considering uncertainty necessitates an efficient uncertainty quantification (UQ) approach and optimization driven by accurate predictions of system behaviors, which remains a challenge for learning-based methods. This article presents a simultaneous multistep robust model predictive control (MPC) framework that incorporates real-time decision-making with uncertainty awareness for digital twin systems. Leveraging a multistep-ahead predictor named time-series dense encoder (TiDE) as the surrogate model, this framework differs from conventional MPC models that provide only one-step-ahead predictions. In contrast, TiDE can predict future states within the prediction horizon in one shot, significantly accelerating MPC. Furthermore, quantile regression is employed with the training of TiDE to perform flexible and computationally efficient UQ on data uncertainty. Consequently, with the deep learning quantiles, the robust MPC problem is formulated into a deterministic optimization problem and provides a safety buffer that accommodates disturbances to enhance the constraint satisfaction rate. As a result, the proposed method outperforms existing robust MPC methods by providing less conservative UQ and has demonstrated efficacy in an engineering case study involving directed energy deposition (DED) additive manufacturing. This proactive, uncertainty-aware control capability positions the proposed method as a potent tool for future digital twin applications and real-time process control in engineering systems.

Abstract Additive manufacturing has enabled the fabrication of functionally graded materials (FGMs), such as compositionally graded alloys (CGAs), offering unprecedented flexibility in structural design. CGAs hold significant potential for thermal-elastic applications, yet existing design methods often overlook temperature-dependent material properties due to the complexity of coupled physics, design-dependent temperature fields, and local constraints. To address these challenges, we propose a topology optimization (TO) framework that concurrently designs geometry and graded material composition while accounting for temperature-dependent material behaviors and nonlinear thermal analysis. Our method employs a radial basis function (RBF)-based interpolation scheme to model material properties as functions of both temperature and material composition. Additionally, we leverage automatic differentiation and adjoint sensitivity analysis for computational efficiency and extensibility to GPU acceleration. Numerical examples demonstrate the effectiveness of our approach, underscoring (1) the critical role of temperature-dependent material properties in thermal-elastic structure optimization and (2) the benefits of continuous material grading in enhancing structural performance.

The martensite start temperature (Ms) of steel plays an important guiding role in the formulation of heat treatment process. Because steel contains many alloying elements, it is very challenging to predict Ms from alloying elements. The rapid development of data-driven technology makes it convenient to accurately predict Ms. In this work, an Ms dataset of 1313 entries was collected and evaluated. By comparison, XGBoost algorithm is chosen to build a machine learning (ML) model for predicting the Ms of steel. The effects of atomic and thermal characterization on the performance of predictive models trained using machine learning algorithms are investigated. Pearson correlation coefficient and feature importance reveal the linear and nonlinear correlation of elements in the matrix. The model showed a good accuracy between predicted and actual values, and the R 2 on the test set is 0.94. Finally, R 2 of XGBoost model and JMatPro software on unknown data are 0.91 and 0.88 respectively. Successful validation shows that the model can accurately predict the Ms of various types of steel. These results indicate the potential of the model in the formulation of auxiliary heat treatment processes.

In scientific machine learning (SciML), a key challenge is learning unknown, evolving physical processes and making predictions across spatio-temporal scales. For example, in real-world manufacturing problems like additive manufacturing, users adjust known machine settings while unknown environmental parameters simultaneously fluctuate. To make reliable predictions, it is desired for a model to not only capture long-range spatio-temporal interactions from data but also adapt to new and unknown environments; traditional machine learning models excel at the first task but often lack physical interpretability and struggle to generalize under varying environmental conditions. To tackle these challenges, we propose the Attention-based Spatio-Temporal Neural Operator (ASNO), a novel architecture that combines separable attention mechanisms for spatial and temporal interactions and adapts to unseen physical parameters. Inspired by the backward differentiation formula (BDF), ASNO learns a transformer for temporal prediction and extrapolation and an attention-based neural operator for handling varying external loads, enhancing interpretability by isolating historical state contributions and external forces, enabling the discovery of underlying physical laws and generalizability to unseen physical environments. Empirical results on SciML benchmarks demonstrate that ASNO outperforms over existing models, establishing its potential for engineering applications, physics discovery, and interpretable machine learning.

The martensite start temperature (M s ) of steels plays an important role in the formulation of heat‐treatment processes. Therefore, it is of great practical importance to predict M s accurately and rapidly. In the present work, machine learning (ML) methods are used to model M s based on the M s data of 1177 steels. Moreover, its generalization performance is verified using fivefold cross validation. Three different back‐propagation (BP) neural network algorithms (genetic algorithm [GA], particle swarm optimization, mind evolutionary algorithm) are used for optimal model selection. The results indicate that, among the three BP neural network algorithms, the GA–BP model has the highest prediction accuracy on the test set. The performances of GA–BP, Thermal–Calc, and JMatPro in predicting the M s of medium‐ and low‐carbon steels and high‐carbon steels are analyzed using an unknown dataset. The results show that the GA–BP model has strong generalization ability and can predict M s relatively accurately. The influence of alloying elements on M s is analyzed using the GA–BP model and the shapley additive explanation method, which provides strategies for studying the microstructure evolution of steel or optimizing the heat‐treatment process.

Conspectus The increasing demands of sustainable energy, electronics, and biomedical applications call for next-generation functional materials with unprecedented properties. Of particular interest are emerging materials that display exceptional physical properties, making them promising candidates for energy-efficient microelectronic devices. As the conventional Edisonian approach becomes significantly outpaced by growing societal needs, emerging computational modeling and machine learning methods have been employed for the rational design of materials. However, the complex physical mechanisms, cost of first-principles calculations, and the dispersity and scarcity of data pose challenges to both physics-based and data-driven materials modeling. Moreover, the combinatorial composition–structure design space is high-dimensional and often disjoint, making design optimization nontrivial. In this Account, we review a team effort toward establishing a framework that integrates data-driven and physics-based methods to address these challenges and accelerate material design. We begin by presenting our integrated material design framework and its three components in a general context. (1) Using text mining and natural language processing techniques, our framework first extracts and organizes relevant information dispersed in the literature. (2) From this initial database of relevant materials, data-driven models can be trained and subsequently employed to perform virtual screening of the unknown materials space. This virtual screening process can identify promising materials families for further investigation, thus narrowing down the candidate space. (3) Within the identified materials families, a Bayesian optimization-based adaptive discovery workflow is applied to search for materials with optimal properties. To extend the capability of Bayesian optimization, which was previously restricted to small data and numerical variables, we developed a family of uncertainty-aware machine learning methods for mixed numerical and categorical variables. We then provide an example of applying this materials design framework to metal–insulator transition (MIT) materials, a specific type of emerging material with practical importance in next-generation memory technologies. We identify multiple new materials that may display this property in the lacunar spinel and Ruddlesden–Popper perovskite families and propose pathways for their synthesis. The classifiers used to identify new possible MIT materials also identified previously unknown features that may be used for predictive theory for this class of materials. For example, we have identified descriptors derived from ionicity and atom sizes as indicators to MIT behavior. Finally, we identify some outstanding challenges in data-driven materials design, such as material data quality issues, property–performance mismatch, and validation and deployment. We seek to raise awareness of these overlooked issues hindering material design, thus stimulating efforts toward developing methods to mitigate the gaps.

Abstract The objective was to evaluate the effect of a felt-covered filter cover in dual-flow continuous culture (DFCC) fermenters on retaining protozoa in fermenter rumen fluid. In vitro rumen simulation techniques are best studied when the microbial community in the fermenters represents native rumen protozoa communities. Protozoa serve as the foundation for methanogenesis by providing hydrogen for rumen methanogens. Filter type could impact protozoa retention in a dual-flow fermenter system. Four DFCC fermenters were run for two periods consisting of 11 days with two control fermenters equipped with a conventional, 100-micron mesh filter (MF) and two fermenters equipped with a felt-covered filter (FF) over the 100-micron mesh filter. As it is crucial to maintain the fermenter rumen fluid anaerobic and in conditions like rumen conditions for the protozoa, all fermenters were adjusted to maintain pH in the range of 5-6 using a prepared continuous culture buffer solution distributed using a flow rate system. pH was recorded at h 0, 4, 8, and 12. Fermenter temperature was monitored to maintain 39 degrees Celsius and agitated at 50 rpm to support protein turnover. The fermenters were administered 44.6 g of feed substrate consisting of 38.45% dry corn, 30% corn silage, 10% grass hay, 12% distiller’s grain, 1.30% calcium carbonate, 0.25% trace mineral mix and 8% soybean meal on a dry matter basis at 12 h intervals. Protozoa were counted by collecting two 10 mL samples of the rumen fluid from each fermenter on days 1, 3, 5, 7, 9 and 11. Each sample was preserved with 10 mL of 37% formalin was diluted to a 1:200 sample-glycerol solution. Protozoa concentration was determined by placing 1 mL of the diluted sample solution on a Sedgewick-Rafter cell counting slide. Then, a microscope was used to record the number of holotricha and number of entodinium protozoa observed in 50 cells. Data were analyzed using a MIXED procedure of SAS. There was a treatment by hour interaction (P < 0.01) observed for pH; the FF fermenter had a greater pH at h 0 and 12 compared with the MF fermenter. No treatment by day effects were observed for holotricha and entodinium per mL (P > 0.45). Mean holotricha populations were 2,360,000 and 2,430,000 protozoa/mL for the MF and FF fermenters, respectively. Mean entodinium populations were 115,907 and 90,667 protozoa/mL for the MF and FF fermenters, respectively. Ammonia concentrations were not affected by the treatment (P = 0.34). Total volatile fatty acid concentrations displayed a treatment by hour interaction (P = 0.01); the MF fermenter had a greater concentration at h 0 and 4. While minor impacts on fermentation characteristics were observed, a felt-covered filter did not affect protozoa retention in a dual-flow continuous culture fermenters.

Polycarbonate (PC) possesses uniquely high toughness among polymers, making it well-suited for use as an impact-resistant barrier material. This propensity toward energy dissipation has been associated with characteristics such as backbone flexibility, high entanglement density, and homogeneity. While recent works have enhanced our understanding of how these nanoscale mechanisms contribute to toughness in PC, it remains unclear how they are affected by the deformation mode, rate, and molecular weight of the chains. To study these effects over spatiotemporal scales that extend beyond the reach of atomistic models, we utilized a coarse-grained molecular dynamics (CGMD) model of PC developed with the energy renormalization method. We establish that yield stress rate dependence follows the Cowper–Symonds model for flow stress, the fit for which asymptotically converges to values consistent with low-rate experimental data. As a demonstration of the model’s utility, we additionally explore the effects of PC chain length on fracture behavior and show that toughness is improved through the augmentation of extensive entanglement networks that enable increased stress levels in the material. For chains 50 monomers and longer, chain length has a minimal effect on yield stress and elastic modulus, suggesting that small-strain mechanical response is dominated by nonbonded interactions. This work enables an enhanced understanding of molecular contributions to the macroscopic mechanical behavior of PC and reflects the importance of the polycarbonate chain network in modulating energy dissipation. It additionally highlights the importance of bond breaking in MD models subjected to large strain. More broadly, it represents a critical step toward the CGMD modeling of PC-based nanocomposites.

The properties of biological materials like proteins and nucleic acids are largely determined by their primary sequence. Certain segments in the sequence strongly influence specific functions, but identifying these segments, or so-called motifs, is challenging due to the complexity of sequential data. While deep learning (DL) models can accurately capture sequence–property relationships, the degree of nonlinearity in these models limits the assessment of monomer contributions to a property─a critical step in identifying key motifs. Recent advances in explainable AI (XAI) offer attention and gradient-based methods for estimating monomeric contributions. However, these methods are primarily applied to classification tasks, such as binding site identification, where they achieve limited accuracy (40–45%) and rely on qualitative evaluations. To address these limitations, we introduce a DL model with interpretable steps, enabling direct tracing of monomeric contributions. Inspired by the masking technique commonly used in vision and natural language processing domains, we propose a new metric ( I ) for quantitative analysis on datasets mainly containing distinct properties of anticancer peptides (ACP), antimicrobial peptides (AMP), and collagen. Our model exhibits 22% higher explainability than the gradient and attention-based state-of-the-art models, recognizes critical motifs (RRR, RRI, and RSS) that significantly destabilize ACPs, and identifies motifs in AMPs that are 50% more effective in converting non-AMPs to AMPs. These findings highlight the potential of our model in guiding mutation strategies for designing protein-based biomaterials.

Although polymers are most often used within bulk materials, investigating their conformations and dynamics has long been a challenging endeavor in this configuration, particularly under external forces. Addressing this, we utilize single-molecule localization microscopy as a powerful imaging tool to visualize bottlebrush poly(n-butyl acrylate) chains in the bulk state under spherical indentation, quantitatively describing changes in behavior of single polymer chains. We compare these experiments to displacement fields determined analytically and confirmed through finite element analysis. This study pioneers visualizing polymer conformational changes in their native environment in situ, offering transformative insights into polymer behavior and dynamics.

Digital Twin – a virtual replica of a physical system enabling real-time monitoring, model updating, prediction, and decision-making – combined with recent advances in machine learning, offers new opportunities for proactive control strategies in autonomous manufacturing. However, achieving real-time decision-making with Digital Twins requires efficient optimization driven by accurate predictions of highly nonlinear manufacturing systems. This paper presents a simultaneous multi-step Model Predictive Control (MPC) framework for real-time decision-making, using a multivariate deep neural network, named Time-Series Dense Encoder (TiDE), as the surrogate model. Unlike conventional MPC models which only provide one-step ahead prediction, TiDE is capable of predicting future states within the prediction horizon in one shot (multi-step), significantly accelerating the MPC. Using Directed Energy Deposition (DED) additive manufacturing as a case study, we demonstrate the effectiveness of the proposed MPC in achieving melt pool temperature tracking to ensure part quality, while reducing porosity defects by regulating laser power to maintain melt pool depth constraints. In this work, we first show that TiDE is capable of accurately predicting melt pool temperature and depth. Second, we demonstrate that the proposed MPC achieves precise temperature tracking while satisfying melt pool depth constraints within a targeted dilution range (10%–30%), reducing potential porosity defects. Compared to Proportional–Integral–Derivative (PID) controller, the MPC results in smoother and less fluctuating laser power profiles with competitive or superior melt pool temperature control performance. This demonstrates the MPC’s proactive control capabilities, leveraging time-series prediction and real-time optimization, positioning it as a powerful tool for future Digital Twin applications and real-time process optimization in manufacturing. • Simultaneous multi-step MPC accelerates real-time decision-making in Digital Twin. • Surrogate using Time-Series Dense Encoder (TiDE) enables multistep-ahead prediction. • Accurate predictions for melt pool temperature and depths using multivariate TiDE. • MPC improves melt pool temperature tracking while enforcing depth constraints in DED. • Real-time decision-making is supported by auto-differentiation. • Proactive defect mitigation enhances part quality by maintaining dilution range.

The task of entity alignment (EA) seeks to identify corresponding entities across different knowledge graphs (KGs). However, in large-scale KG alignment tasks, the complexity of the problem renders traditional entity structure representation methods, designed for small-scale KGs, ineffective. Partition-based approaches address this challenge by breaking large KGs into smaller subgraphs, but this inevitably results in a loss of structural information. Although existing methods have sought to mitigate this issue, they have largely overlooked the interplay between partitioning and entity structure representation learning. To address this, we propose a Self-Partitioning Entity Alignment (SPEA) pipeline for large-scale EA, in which partitioning and entity structure representation learning are mutually optimized. Within this framework, we introduce the Inter-Subgraph Neighbor Interaction (ISNI) for enhanced entity structure representation, the Bidirectional Margin-based Confidence (BMC) for pseudo-pairing, the Seed-oriented Cross Graph Partitioner (SCGP) for dynamic repartitioning, and the Historical Confidence Ensembling (HCE) strategy for consistent training. Extensive experiments demonstrate that SPEA significantly outperforms existing methods for large-scale EA tasks.

Abstract Artificial intelligence and machine learning frameworks have become powerful tools for establishing computationally efficient mappings between inputs and outputs in engineering problems. These mappings have enabled optimization and analysis routines, leading to innovative designs, advanced material systems, and optimized manufacturing processes. In such modeling efforts, it is common to encounter multiple information (data) sources, each varying in specifications. Data fusion frameworks offer the capability to integrate these diverse sources into unified models, enhancing predictive accuracy and enabling knowledge transfer. However, challenges arise when these sources are heterogeneous, i.e., they do not share the same input parameter space. Such scenarios occur when domains differentiated by complexity such as fidelity, operating conditions, experimental setup, and scale, require distinct parametrizations. To address this challenge, a two-stage heterogeneous multi-source data fusion framework based on the input mapping calibration (IMC) and the latent variable Gaussian process (LVGP) is proposed. In the first stage, the IMC algorithm transforms the heterogeneous input parameter spaces into a unified reference parameter space. In the second stage, an LVGP-enabled multi-source data fusion model constructs a single-source-aware surrogate model on the unified reference space. The framework is demonstrated and analyzed through three engineering modeling case studies with distinct challenges: cantilever beams with varying design parametrizations, ellipsoidal voids with varying complexities and fidelities, and Ti6Al4V alloys with varying manufacturing modalities. The results demonstrate that the proposed framework achieves higher predictive accuracy compared to both independent single-source and source-unaware data fusion models.

Recent advances in AI applied to high-throughput materials discovery, synthesis, and processing offer a pathway to accelerated breakthroughs and scaled optimization of advanced electronic materials for data-intensive computation.

Digital Twins, virtual replicas of physical systems that enable real-time monitoring, model updates, predictions, and decision-making, present novel avenues for proactive control strategies for autonomous systems. However, achieving real-time decision-making in Digital Twins considering uncertainty necessitates an efficient uncertainty quantification (UQ) approach and optimization driven by accurate predictions of system behaviors, which remains a challenge for learning-based methods. This paper presents a simultaneous multi-step robust model predictive control (MPC) framework that incorporates real-time decision-making with uncertainty awareness for Digital Twin systems. Leveraging a multistep ahead predictor named Time-Series Dense Encoder (TiDE) as the surrogate model, this framework differs from conventional MPC models that provide only one-step ahead predictions. In contrast, TiDE can predict future states within the prediction horizon in a one-shot, significantly accelerating MPC. Furthermore, quantile regression is employed with the training of TiDE to perform flexible while computationally efficient UQ on data uncertainty. Consequently, with the deep learning quantiles, the robust MPC problem is formulated into a deterministic optimization problem and provides a safety buffer that accommodates disturbances to enhance constraint satisfaction rate. As a result, the proposed method outperforms existing robust MPC methods by providing less-conservative UQ and has demonstrated efficacy in an engineering case study involving Directed Energy Deposition (DED) additive manufacturing. This proactive while uncertainty-aware control capability positions the proposed method as a potent tool for future Digital Twin applications and real-time process control in engineering systems.

Traumatic pneumothorax is a complex condition that is challenging to diagnose, particularly in hospitals, underdeveloped areas, and during mass casualty events. This study aimed to evaluate the potential of machine learning (ML) for diagnosing and assessing traumatic pneumothorax. We extracted 33 vital signs and blood gas parameters from the MIMIC-IV database, selecting 12 clinically significant features as inputs to four ML algorithms: extreme gradient boosting (XGBoost), artificial neural network (ANN), support vector machine (SVM), and k-nearest neighbors (KNN). Five-fold cross-validation was used to train and test the models, with external validation performed on the EICU database. Model performance was evaluated using AUROC, recall, and accuracy, with SHAP interpretability employed to understand feature importance. In total, 3871 participants from the MIMIC-IV database and 22,022 participants from the EICU database were analyzed. Hemoglobin, Oxygenation Index, and pH were found to be key indicators of severe traumatic pneumothorax. XGBoost exhibited the best performance, achieving an AUROC of 0.979 (95% CI: [0.966, 0.989]) on the MIMIC-IV dataset and 0.806 (95% CI: [0.740, 0.864]) on the EICU dataset. The results suggest that ML, particularly XGBoost, is faster and more convenient than traditional imaging methods, making it well-suited for emergency or mass casualty situations. ML algorithms show promise for initial diagnosis of traumatic pneumothorax, with XGBoost demonstrating strong interpretability and robust external validation.

This article reviews optimization methods that enhance adaptability, efficiency and decision making in modern manufacturing, emphasizing the transformative role of artificial intelligence (AI) and digital twin technologies. By integrating AI and machine learning algorithms within digital twin frameworks, manufacturers can facilitate real-time monitoring, quality control and dynamic process adjustments. This synergy not only boosts operational efficiency but also enables precise modelling, offering predictive insights for strategic planning and innovation. The combination of digital twins and optimization techniques supports resource optimization, balancing competing objectives and driving continuous process improvements. With both offline and online optimization approaches, digital twins enable efficient production adjustments while ensuring long-term performance and scalability. Ultimately, this review highlights digital twins as foundational technologies for smart, sustainable manufacturing, incorporating advanced optimization strategies to enhance adaptability and operational resilience. This positions optimization algorithms and digital twins as key drivers in the future of intelligent production systems.

For polymer nanocomposites, disordered microstructural nature makes processing control and tailoring properties to desired values a challenge. Understanding process-structure-property relation can provide guidelines for process and constituents design. Our work explores nuances of PSP relation for polymer nanocomposites with attractive pairing between particles and polymer bulk. In the absence of any nano or micro-scale local property measurement, we develop a material model that can represent decay for small strain elastoplastic properties in interfacial regions and simulate representative or statistical volume element behavior. This interfacial model is further combined with a microstructural design of experiments for agglomerated nanocomposite systems. Agglomerations are particle aggregations that are microstructural defects resulting from lack of processing control. Twin screw extrusion process can reduce extent of aggregation in hot pressed samples via erosion or rupture depending on screw rpms and toque. We connect this process-structure relation to structure-property relation that emerges from our study. We discover that balancing between local stress concentration zone and interfacial property decay governs how fast yield stress can improve if we break down agglomeration via erosion. Rupture is relatively less effective in helping improve nanocomposite yield strength. Additionally, we allude to yield initiation and progression in these multiphase materials. We have come up with a field quantity called local yield resistance that indicates balance stress concentration zones and interfacial effects. Yield resistance map from linear regime acts as a predictor of local yielding process and can be a useful tool for interface design for plastic deformation behavior.

Adopting a holistic approach to manufacturing that strikes a balance between economic, ecological, and social viability and well-being has the potential to address the grand challenge of climate change and achieve the sustainable development goals.

The absence of scalable and environmentally sustainable methods for producing electronic-grade graphene nanoplatelets remains a barrier to the industrial-scale application of graphene in printed electronics and conductive composites. To address this unmet need, here we report the utilization of carboxylated cellulose nanocrystals (CNCs) extracted from the perennial tall grass Miscanthus × giganteus as a biorenewable dispersant for the aqueous liquid-phase exfoliation of few-layer graphene nanoplatelets. This CNC-based exfoliation procedure was optimized using a Bayesian machine learning model, resulting in a significant graphite-to-graphene conversion yield of 13.4% and a percolating graphene thin-film electrical conductivity of 3.4 × 10 4 S m –1 . The as-exfoliated graphene dispersions were directly formulated into an aerosol jet printing ink using cellulose-based additives to achieve high-resolution printing (∼20 μm line width). Life cycle assessment of this CNC-based exfoliation method showed substantial improvements for fossil fuel consumption, greenhouse gas emissions, and water consumption compared to incumbent liquid-phase exfoliation methods for electronic-grade graphene nanoplatelets. Mechanistically, potential mean force calculations from molecular dynamics simulations reveal that the high exfoliation yield can be traced back to the favorable surface interactions between CNCs and graphene. Ultimately, the use of biorenewable CNCs for liquid-phase exfoliation will accelerate the scalable and eco-friendly manufacturing of graphene for electronically conductive applications.