Manufacturers of cutting and machining machines face increasing pressure to optimize performance and sustainability while complying with evolving regulations. Traditional machine learning approaches are often limited by biased and repetitive datasets collected during real operations. This article presents a real-time simulation framework for generating large synthetic datasets to train predictive machining models. A mechanistic model with probabilistic parameters is validated on experimental data and integrated into the simulator, enabling neural networks to predict process metrics such as vibrations, cutting forces, and product quality prior to machining. The framework further supports large-scale optimal control by testing setpoint control strategies for virtual prototyping. This approach allows manufacturers to enhance efficiency, reduce waste, and improve product quality while minimizing operational risks.

It is now well established that a laser pulse can demagnetize a ferromagnet. However, for a long time, it has not had an analytic theory because it falls into neither nonlinear optics (NLOs) nor magnetism. Here we attempt to fill this gap by developing a nonlinear optical theory centered on the spin moment, instead of the more popular susceptibility. We first employ group theory to pin down the lowest order of the nonzero spin moment in a centrosymmetric system to be the second order, where the second-order density matrix contains four terms of sum frequency generation and four terms of difference frequency generation (DFG). By tracing over the product of the density matrix and the spin matrix, we are now able to compute the light-induced spin moment. We apply our theory to FePt and FePd, two most popular magnetic recording materials with identical crystal and electronic structures. We find that the theory can clearly distinguish the difference between those two similar systems. Specifically, we show that FePt has a stronger light-induced spin moment than FePd, in agreement with our real-time ultrafast demagnetization simulation and the experimental results. Among all the possible NLO processes, DFGs produce the largest spin moment change, a manifestation of optical rectification. Our research lays a solid theoretical foundation for femtomagnetism, so the light-induced spin moment reduction can now be computed and compared among different systems, without time-consuming real-time calculations, representing a significant step forward.

Real-time simulation and abnormal condition recognition of series compressor system based on digital twin

Gaden-RT: A real time and interactive gas dispersion simulator for mobile robotics

A novel real-time aeroelastic hybrid simulation system of section model wind tunnel testing based on adaptive extended Kalman filter

Enhancing stability and convergence of offline real-time hybrid simulation via global model-based numerical substructuring approach

The rapid integration of wind power plants (WPPs) into modern electrical power systems (MEPSs) is crucial to global decarbonization, but it introduces significant technical challenges. Variability, intermittency, and forecasting uncertainty compromise frequency stability, voltage regulation, and grid reliability, particularly at high levels of renewable energy integration. To address these issues, adaptive control strategies have been proposed at the turbine, plant, and system levels, including reinforcement learning-based optimization, cooperative plant-level dispatch, and hybrid energy schemes with battery energy storage systems (BESS). At the same time, interoperability frameworks based on international standards, notably IEC 61850 and IEC 61400-25, provide the communication backbone for vendor-independent coordination; however, their application remains largely limited to monitoring and protection, rather than holistic adaptive operation. Real-Time Automation Controllers (RTACs) emerge as promising platforms for unifying monitoring, operation, and protection functions, but their deployment in large-scale WPPs remains underexplored. Validation of these frameworks is still dominated by simulation-only studies, while real-time digital simulation (RTDS) and hardware-in-the-loop (HIL) environments have only recently begun to bridge the gap between theory and practice. This review consolidates advances in adaptive control, interoperability, and validation, identifies critical gaps, including limited PCC-level integration, underutilization of IEC standards, and insufficient cyber–physical resilience, and outlines future research directions. Emphasis is placed on holistic adaptive frameworks, IEC–RTAC integration, digital twin–HIL environments, and AI-enabled adaptive methods with embedded cybersecurity. By synthesizing these perspectives, the review highlights pathways toward resilient, secure, and standards-compliant renewable power systems that can support the transition to a low-carbon future.

DECODE: An open-source cloud-based platform for the noninvasive management of peripheral artery disease.

A high-fidelity reduced-order modeling framework for accurate and real-time performance simulation of lead-acid batteries considering buoyancy-driven electrolyte stratification

A Review of Real-time Military Training Simulator Based on Improving War Scenario Using AI Tools

Real-time simulation for deep-sea mining system with sea trial validation

Abstract Autonomous robotic application of shotcrete requires not only precise actuation and control, but also a minimum understanding of the sprayed-concrete placement process. However, modeling shotcrete is inherently complex due to its dynamic, multiphase, and stochastic nature. To address this challenge, we present a real-time simulation model based on physics-informed particle systems that captures the key characteristics of shotcrete spraying and accumulation. The model accounts for total rebound, cohesive failure (material detachment), and interactions with reinforcement elements commonly found in construction scenarios. This approach enables a flexible and computationally efficient simulation model that provides visual feedback to the user and structured data output (i.e., height-field) suitable for downstream analysis. Material distribution on a receiving target surface and the total rebound were systematically analyzed and compared with empirical data found in the literature, demonstrating a strong correlation between simulation outcome and experimental observations. This model establishes a foundation for the integration of artificial intelligence (AI) technologies and hybrid, simulation-based planning and control methods—such as Digital Twins—to enhance robotic shotcrete performance and autonomous decision-making.

This paper presents the design and implementation of a torque vectoring control strategy aimed at improving the dynamic stability and handling performance of a Formula Student electric vehicle. The approach dynamically distributes torque between individual wheels based on real-time vehicle parameters such as speed, yaw rate, lateral acceleration, steering input, throttle position, and track width. A comprehensive algorithm is developed to compute dynamic load distribution using the vehicle's mass, center of gravity height, and lateral dynamics. Yaw correction is applied to counteract instability during aggressive cornering, ensuring enhanced responsiveness and precise directional control. The torque output is constrained within safety thresholds to prevent motor overload and maintain operational stability. The proposed system is implemented using MATLAB/Simulink with the Powertrain Blockset, enabling a modular and real-time simulation environment. Simulation results demonstrate the effectiveness of the control logic in improving vehicle stability, minimizing yaw deviation, and maintaining control in potential off-track conditions, offering a viable solution for high-performance and safety-critical Formula Student applications.

The validation of sensor systems, particularly lidar, is crucial in advancing autonomous vehicle technology. Despite their robust perception capabilities, certain weather conditions and object characteristics can challenge detection performance, leading to potential safety concerns. This study investigates corner cases where object detection may fail due to physical constraints. Utilizing virtual testing environments like Carla and ROS2, simulations analyze how reflection characteristics affect detectability by implementing weather models into a real-time simulation. Results reveal challenges in detecting black objects compared to white ones, particularly in adverse weather conditions. A time-sensitive corner case was analyzed, revealing that while bad weather and wet roads restrict the safe driving speed range, complete deactivation of the driving assistant at certain speeds may be unnecessary despite current manufacturer practices. The study underscores the importance of considering such factors in future safety protocols to mitigate accidents and ensure reliable autonomous driving systems.

This study investigates the effectiveness of a feedback-based proportional–integral (PI) regulator in the control system of a doubly fed induction generator (DFIG) used in wind energy applications, with a focus on enhancing the reliability and sustainability of renewable power generation. The primary objective is to assess how the feedback-based PI regulator can improve the efficiency and stability of rotor-side converter control, thereby ensuring consistent power quality and resilient operation under variable environmental and loading conditions. A novel experimental setup was developed by integrating a laboratory-scale DFIG system with real-time digital simulation tools, enabling a realistic assessment of dynamic performance. Various operating scenarios, including wind speed fluctuations and generator parameter variations, were analyzed to evaluate the regulator’s ability to minimize power ripples, ensure voltage stability, reduce total harmonic distortion (THD), and mitigate torque ripple—all of which contribute to more sustainable and efficient energy conversion. Comparative analyses using performance indicators such as power ripple, steady-state error, and overshoot demonstrate that the feedback-based PI regulator outperforms conventional control methods reported in the literature. The experimental results confirm that the proposed control strategy not only enhances dynamic performance and operational robustness but also contributes to the long-term sustainability of wind energy systems by improving energy efficiency, reducing losses, and supporting grid stability. Overall, this work promotes sustainability by advancing control techniques that optimize renewable energy utilization and strengthen the reliability of clean power technologies.

Abstract DC microgrids are growing more popular because of their versatility and effective energy resource sharing. However, control mechanisms that depend on communication between numerous power units are typically necessary to provide a stable and well-coordinated operation. This paper proposes an artificial neural network (ANN)-powered adaptive management approach to control a widely distributed PV/battery-based DC nanogrid system without requiring direct communication linkages. It comprises of many Nano Grid (NG) groups, each of which can function autonomously by supplying electricity to neighbouring units when required. An ANN-based droop control technique is used to achieve dynamic load balancing across linked nanogrids using state of charge (SOC) and DC bus voltage data. The intelligent controller seamlessly transitions between maximum power point tracking (MPPT) mode and current regulation mode to optimize energy transfer and preserve grid stability. The system’s modular extension and adaptability, coupled with the decentralized nature of nanogrids, allow it to scale efficiently. It increases the efficiency of power distribution by dynamically modifying energy flow in response to existing conditions. Additionally, by enabling intelligent energy coordination without requiring direct data exchange between nanogrids, the method streamlines and improves reliability. The simulation results show improved voltage control, faster transient response, and reliable power sharing without communication. Hardware-in-the-loop (HIL) validation is performed using an OP4510 real-time simulator to confirm the effectiveness of the proposed method.

Accurately characterizing dislocation behavior - the driving force behind nucleation and growth of recrystallized grains - has long been a formidable challenge for traditional cellular automata methods. We are proud to unveil a groundbreaking, machine learning-enhanced cellular automaton framework that fundamentally transforms the mapping of dislocation substructures while expertly modeling static recrystallization (SRX) behavior. By integrating the dislocation escape assumption during recovery, we decisively eliminate spatial resolution limitations and capture the intricate mesoscopic dynamics of dislocation evolution with unprecedented precision. Our innovative approach has demonstrated extraordinary effectiveness in predicting the recrystallization kinetics of a typical austenitic alloy, bolstered by strong experimental validation. The deep learning-based dislocation implantation module, SRX-net, stands out as a game-changer, surpassing traditional techniques such as random forests and U-net by showcasing exceptional capabilities in identifying complex intracrystalline substructures and managing uneven strain concentrations. The proposed advanced simulations yield critical insights into dislocation density variations, highlighting significant local fluctuations driven by dislocation escape. Importantly, while the migration and accumulation of dislocations may falter in meeting nucleation conditions during grain growth, our model excels in accurately predicting average SRX grain sizes without introducing any unphysical artifacts. This revolutionary framework dramatically reduces time-to-solution, empowering comprehensive parametric studies and enabling near real-time recrystallization simulations, thus setting a bold new standard for industrial applications.

Antimicrobial Resistance (AMR) is a global concern demanding high-throughput and precise AMR surveillance strategies. This review provides a comprehensive list of Artificial Intelligence (AI) driven frameworks widely employed in the early detection, structural characterization, and designing of novel inhibitors to block the resistance pathways critical for AMR. Deep learning algorithms including DeepGO, DeepGOPlus, DeepGO-SE, PFresGO, DPFunc, ProtENN and graph-based architectures of GraphSite, GrASP enables precise functional annotation of resistance-associated proteins. AI-guided protein modeling performed by AlphaFold, RoseTTAFold, ProtGPT-2, ESMFold etc. generates high resolution 3D conformations, further utilized in performing molecular docking via tools like AutoDock, DeepDocking and DeepChem and analyzed with tools like DeepDriveMD, TorchMD, and PRITHVI, which can perform real-time molecular dynamics simulations. Identification of relevant resistant biomarkers from mass-spectrometry profiles can also be achieved with the help of DeepNovo, Casanovo, or Prosit. Tools like DeepARG, HMD-ARG, and BacEffluxPred enables identification of unannotated resistance genes from metagenomic samples. Natural Language Processing (NLP) and Large Language-based models (LLM) facilitate identification of resistant determinants via literature mining enabling regulatory network mapping and rational inhibitor design. Furthermore, AI-mediated de-novo inhibitor design is achieved using Variational Autoencoders (VAE), Generative Adversarial Networks (GAN), diffusion and flow-matching based frameworks serve as potential options for enhancing diagnostic interventions against resistant phenotypes. AI-based protein-protein interaction predictors include DeepInteract, Pred_PPI, PLIP, DeepAIPs-Pred, DeepAIPs-SFLA, SBSM-Pro, Deep Stacked-AVPs, and pNPs-CapsNet help in understanding how resistance proteins interact with each other enabling precise identification of AMR-modulating peptides and supports the modeling of novel antibiotics for blocking interactions and disrupting resistance pathways.

We propose a computational framework to extract nonlinear response functions from real-time simulations in the presence of more than one external field. We apply this approach to the calculation of sum frequency generation (SFG) and difference frequency generation (DFG). SFG and DFG are second-order nonlinear processes where two lasers with frequencies \omega_1 ω 1 and \omega_2 ω 2 combine to produce a response at frequency \omega = \omega_1 ± \omega_2 ω = ω 1 ± ω 2 . Compared with other nonlinear responses such as second-harmonic generation, SFG and DFG allow for tunability over a larger range. Moreover, the optical response can be enhanced by selecting the two laser frequencies in order to match specific electron-hole transitions. To assess the approach, we calculate the SFG and DFG of two-dimensional crystals, h-BN and MoS _2 2 monolayers, from real-time solution of an effective Schrödinger equation. Within the effective Schrödinger equation, one can select from various levels of theory for the effective one-particle Hamiltonian to account for local-field effects and electron-hole interactions. We compare results obtained within the independent particle picture and including many-body effects. Such comparison allows us to identify and characterize excitonic features in the obtained spectra. Additionally, we demonstrate that our approach can also extract higher-order response functions, such as field-induced second-harmonic generation. We provide an example using the h-BN bilayer.

This paper provides an evaluation methodology and prototyped and tested a virtual Intelligent Electronic Device (vIED) for digital substations in a Real Time (RT) simulation. Based on the IEC 61850 standard and virtualization technology, the proposed framework aims to enhance protection, automation, and control systems for application in today’s substations. A new and very specific method of testing was conceived and used as an instrument for verifying the given framework and its potential based on different methods of communications, scalability, and functionality options. The analysis also examines the effectiveness of the SDCommNet for vIED frameworks about operational network re-scaling capacity, besides testing the impact of nominal and transient data traffic on the global performance of the system. Empirical findings demonstrate that the vIED framework always meets essential system characteristics, such as response time and network transfer latency, irrespective of configurations. Such findings act as a perfect foundation for future development of design and testing protocols for the virtualization of substation systems.