Search Results (556)

Glass fiber-reinforced polymer (GFRP) composites exhibit significant advantages over conventional structural webbing materials, including lightweight and corrosion resistance. This study investigates the localized compression performance of the proposed GFRP grid web–concrete composite beam through experimental and numerical analyses. Three specimen groups with variable shear-span ratios (λ = 1.43, 1.77) and local stiffener specimens were designed to assess their localized compressive behavior. Experimental results reveal that a 19.2% reduction in shear-span ratio enhances ultimate load capacity by 22.93% and improves stiffness by 66.85%, with additional performance gains of 77.53% in strength and 94.29% in stiffness achieved through local stiffener implementation. In addition, finite element (FE) analysis demonstrated a strong correlation with experimental results, showing less than 5% deviation in ultimate load predictions while accurately predicting stress distributions and failure modes. FE parametric analysis showed that increasing the grid thickness and decreasing the grid spacing within a reasonable range can considerably enhance the localized compression performance. The proposed analytical model, based on Winkler elastic foundation theory, predicts ultimate compression capacities within 10% of both the experimental and numerical results. However, the GFRP grid strength adjustment factor β_g should be further refined through additional experiments and numerical analyses to improve reliability. Full article

Background: Unlike most cancers, breast cancer poses a persistent risk of distant recurrence—often years after initial treatment—making long-term risk stratification uniquely challenging. Current tools fall short in predicting late metastatic events, particularly for early-stage patients. Methods: We present an interpretable machine learning (ML) pipeline to predict distant recurrence-free survival at 5, 10, and 15 years, integrating Bayesian network-based causal feature selection, deep feed-forward neural network models (DNMs), and SHAP-based interpretation. Using electronic health record (EHR)-based clinical data from over 6000 patients, we first applied the Markov blanket and interactive risk factor learner (MBIL) to identify minimally sufficient predictor subsets. These were then used to train optimized DNM classifiers, with hyperparameters tuned via grid search and benchmarked against models from 10 traditional ML methods and models trained using all predictors. Results: Our best models achieved area under the curve (AUC) scores of 0.79, 0.83, and 0.89 for 5-, 10-, and 15-year predictions, respectively—substantially outperforming baselines. MBIL reduced input dimensionality by over 80% without sacrificing accuracy. Importantly, MBIL-selected features (e.g., nodal status, hormone receptor expression, tumor size) overlapped strongly with top SHAP contributors, reinforcing interpretability. Calibration plots further demonstrated close agreement between predicted probabilities and observed recurrence rates. The percentage performance improvement due to grid search ranged from 25.3% to 60%. Conclusions: This study demonstrates that combining causal selection, deep learning, and grid search improves prediction accuracy, transparency, and calibration for long-horizon breast cancer recurrence risk. The proposed framework is well-positioned for clinical use, especially to guide long-term follow-up and therapy decisions in early-stage patients. Full article

The process of designing reinforced concrete (RC) buildings has traditionally relied on manually evaluating a limited number of layout alternatives—a time-intensive process that may not always yield the most functionally efficient solution. This research introduces a parametric algorithmic model for the automated optimization of RC buildings with solid slab systems. The model automates and optimizes the layout process, yielding measurable improvements in spatial efficiency while maintaining compliance with structural performance criteria. Unlike prior models that address structural or architectural parameters separately, the proposed framework integrates both domains through a unified generative design approach within a BIM environment, enabling automated evaluation of structurally viable and architecturally coherent slab layouts. Developed within the parametric visual programming environment in Dynamo for Revit, the model employs a generative design (GD) engine to explore and refine various design alternatives while adhering to structural constraints. By leveraging a BIM-based framework, this method enhances efficiency, optimizes resource utilization, and systematically balances structural and architectural requirements. The model was validated through three case studies, demonstrating cost reductions between 2.7% and 17%, with material savings of up to 13.38% in concrete and 20.87% in reinforcement, achieved within computational times ranging from 120 to 930 s. Despite the current development being limited to vertical load scenarios and being most suitable for regular slab-based configurations, the results demonstrated the model’s effectiveness in optimizing grid dimensions and reducing material quantities and costs, and highlighted its ability to streamline early-stage design processes. Full article

Geogrids are frequently utilized in engineering for reinforcement; yet, they are vulnerable to construction damage when employed on coarse-grained soil subgrades. In contrast, waste tire grids are more appropriate for subgrade reinforcement owing to their rough surfaces, integrated steel meshes, robust transverse ribs, extended degradation cycles, and superior durability. Based on the limit equilibrium theory, this study developed formulae for calculating the internal and external frictional resistance, as well as the end resistance of waste tires, to ascertain the interface bearing properties and calculation techniques of waste tire grids. Based on this, a mechanical model for the ultimate pull-out resistance of waste-tire-reinforced soil was developed, and its validity was confirmed through a series of pull-out tests on single-sided strips, double-sided strips, and tire grids. The results indicated that the tensile strength of one side of the strip was approximately 43% of that of both sides, and the rough outer surface of the tire significantly enhanced the tensile performance of the strip; under identical normal stress, the tensile strength of the single-sided tire grid was roughly nine times and four times greater than that of the single-sided and double-sided strips, respectively, and the grid structure exhibited superior anti-deformation capabilities compared to the strip structure. The average discrepancy between the calculated values of the established model and the theoretical values was merely 2.38% (maximum error < 5%). Overall, this research offers technical assistance for ensuring the safety of subgrade design and promoting environmental sustainability in engineering, enabling the effective utilization of waste tire grids in sustainable reinforcement applications. Full article

This comprehensive review maps the fast-evolving landscape in which artificial intelligence (AI) and deep-learning (DL) techniques converge with the Internet of Things (IoT) to manage energy, comfort, and sustainability across smart environments. A PRISMA-guided search of four databases retrieved 1358 records; after applying inclusion criteria, 143 peer-reviewed studies published between January 2019 and April 2025 were analyzed. This review shows that AI-driven controllers—especially deep-reinforcement-learning agents—deliver median energy savings of 18–35% for HVAC and other major loads, consistently outperforming rule-based and model-predictive baselines. The evidence further reveals a rapid diversification of methods: graph-neural-network models now capture spatial interdependencies in dense sensor grids, federated-learning pilots address data-privacy constraints, and early integrations of large language models hint at natural-language analytics and control interfaces for heterogeneous IoT devices. Yet large-scale deployment remains hindered by fragmented and proprietary datasets, unresolved privacy and cybersecurity risks associated with continuous IoT telemetry, the growing carbon and compute footprints of ever-larger models, and poor interoperability among legacy equipment and modern edge nodes. The authors of researches therefore converges on several priorities: open, high-fidelity benchmarks that marry multivariate IoT sensor data with standardized metadata and occupant feedback; energy-aware, edge-optimized architectures that lower latency and power draw; privacy-centric learning frameworks that satisfy tightening regulations; hybrid physics-informed and explainable models that shorten commissioning time; and digital-twin platforms enriched by language-model reasoning to translate raw telemetry into actionable insights for facility managers and end users. Addressing these gaps will be pivotal to transforming isolated pilots into ubiquitous, trustworthy, and human-centered IoT ecosystems capable of delivering measurable gains in efficiency, resilience, and occupant wellbeing at scale. Full article

This study explores the mechanical behavior of polymer and composite specimens fabricated using fused deposition modeling (FDM), focusing on three material configurations: acrylonitrile butadiene styrene (ABS), carbon fiber-reinforced polyphthalamide (PPA/Cf), and a sandwich-structured composite. A systematic experimental plan was developed using the Box–Behnken design (BBD) to investigate the effects of material type (MT), infill pattern (IP), and printing direction (PD) on tensile and flexural strength. Experimental results showed that the PPA/Cf material with a “Cross” IP printed “Flat” yielded the highest mechanical performance, achieving a tensile strength of 75.8 MPa and a flexural strength of 102.3 MPa. In contrast, the lowest values were observed in ABS parts with a “Grid” pattern and “Upright” orientation, recording 37.8 MPa tensile and 49.5 MPa flexural strength. Analysis of variance (ANOVA) results confirmed that all three factors significantly influenced both outputs (p < 0.001), with MT being the most dominant factor. Machine learning (ML) algorithms, Bayesian linear regression (BLR), and Gaussian process regression (GPR) were employed to predict mechanical performance. GPR achieved the best overall accuracy with R² = 0.9935 and MAPE = 11.14% for tensile strength and R² = 0.9925 and MAPE = 12.96% for flexural strength. Comparatively, the traditional BBD yielded slightly lower performance with MAPE = 13.02% and R² = 0.9895 for tensile strength. Validation tests conducted on three unseen configurations clearly demonstrated the generalization capability of the models. Based on actual vs. predicted values, the GPR yielded the lowest average prediction errors, with MAPE values of 0.54% for tensile and 0.45% for flexural strength. In comparison, BLR achieved 0.79% and 0.60%, while BBD showed significantly higher errors at 1.76% and 1.32%, respectively. Full article

This article presents a review on the applications of basalt fibers and their composites in infrastructures. The characteristics and advantages of high-performance basalt fibers and their composites are firstly introduced. Then, the article discusses strengthening using basalt fiber sheets and BFRP bars or grids, followed by concrete structures reinforced with BFRP bars, asphalt pavements, and cementitious composites reinforced with chopped basalt fibers in terms of mechanical behaviors and application examples. The load-bearing capacity of the strengthened structures can be increased by up to 60%, compared with those without strengthening. The lifespan of the concrete structures reinforced with BFRP can be extended by up to 50 years at least in harsh environments, which is much longer than that of ordinary reinforced concrete structures. In addition, the fatigue cracking resistance of asphalt can be increased by up to 600% with basalt fiber. The newly developed technologies including anchor bolts using BFRPs, self-sensing BFRPs, and BFRP–concrete composite structures are introduced in detail. Furthermore, suggestions are proposed for the forward-looking technologies, such as long-span bridges with BFRP cables, BFRP truss structures, BFRP with thermoplastic resin matrix, and BFRP composite piles. Full article

The integration of high-penetration renewable energy sources (RESs) into transmission networks introduces profound uncertainty that challenges traditional infrastructure planning approaches. Existing transmission expansion planning (TEP) models either rely on static scenario sets or over-conservative worst-case assumptions, failing to capture the operational stress triggered by rare but structurally impactful renewable behaviors. This paper proposes a novel bi-level optimization framework for transmission planning under adversarial uncertainty, coupling a distributionally robust upper-level investment model with a lower-level operational response embedded with physics and market constraints. The uncertainty space was not exogenously fixed, but instead dynamically generated through a physics-informed spatiotemporal generative adversarial network (PI-ST-GAN), which synthesizes high-risk renewable and load scenarios designed to maximally challenge the system’s resilience. The generator was co-trained using a composite stress index—combining expected energy not served, loss-of-load probability, and marginal congestion cost—ensuring that each scenario reflects both physical plausibility and operational extremity. The resulting bi-level model was reformulated using strong duality, and it was decomposed into a tractable mixed-integer structure with embedded adversarial learning loops. The proposed framework was validated on a modified IEEE 118-bus system with high wind and solar penetration. Results demonstrate that the GAN-enhanced planner consistently outperforms deterministic and stochastic baselines, reducing renewable curtailment by up to 48.7% and load shedding by 62.4% under worst-case realization. Moreover, the stress investment frontier exhibits clear convexity, enabling planners to identify cost-efficient resilience strategies. Spatial congestion maps and scenario risk-density plots further illustrate the ability of adversarial learning to reveal latent structural bottlenecks not captured by conventional methods. This work offers a new methodological paradigm, in which optimization and generative AI co-evolve to produce robust, data-aware, and stress-responsive transmission infrastructure designs. Full article

The increasing deployment of photovoltaic and energy storage systems (ESSs) in modern power grids has highlighted the critical challenge of component degradation, which significantly impacts system efficiency, operational costs, and long-term reliability. Conventional energy dispatch and optimization approaches fail to adequately mitigate the progressive efficiency loss in PV modules and battery storage, leading to suboptimal performance and reduced system longevity. To address these challenges, this paper proposes a quantum-enhanced degradation pathway optimization framework that dynamically adjusts operational strategies to extend the lifespan of PV storage systems while maintaining high efficiency. By leveraging quantum-assisted Monte Carlo simulations and hybrid quantum–classical optimization, the proposed model evaluates degradation pathways in real time and proactively optimizes energy dispatch to minimize efficiency losses due to aging effects. The framework integrates a quantum-inspired predictive maintenance algorithm, which utilizes probabilistic modeling to forecast degradation states and dynamically adjust charge–discharge cycles in storage systems. Unlike conventional optimization methods, which struggle with the complexity and stochastic nature of degradation mechanisms, the proposed approach capitalizes on quantum parallelism to assess multiple degradation scenarios simultaneously, significantly enhancing computational efficiency. A three-layer hierarchical optimization structure is introduced, ensuring real-time degradation risk assessment, periodic dispatch optimization, and long-term predictive adjustments based on PV and battery aging trends. The framework is tested on a 5 MW PV array coupled with a 2.5 MWh lithium-ion battery system, with real-world degradation models applied to reflect light-induced PV degradation (0.7% annual efficiency loss) and battery state-of-health deterioration (1.2% per 100 cycles). A hybrid quantum–classical computing environment, utilizing D-Wave’s Advantage quantum annealer alongside a classical reinforcement learning-based optimization engine, enables large-scale scenario evaluation and real-time operational adjustments. The simulation results demonstrate that the quantum-enhanced degradation optimization framework significantly reduces efficiency losses, extending the PV module’s lifespan by approximately 2.5 years and reducing battery-degradation-induced wear by 25% compared to conventional methods. The quantum-assisted predictive maintenance model ensures optimal dispatch strategies that balance energy demand with system longevity, preventing excessive degradation while maintaining grid reliability. The findings establish a novel paradigm in degradation-aware energy optimization, showcasing the potential of quantum computing in enhancing the sustainability and resilience of PV storage systems. This research paves the way for the broader integration of quantum-based decision-making in renewable energy infrastructure, enabling scalable, high-performance optimization for future energy systems. Full article

Generative Artificial Intelligence (GenAI)-assisted Virtual Reality (VR) heritage serious game design constitutes a complex adaptive socio-technical system in which natural language prompts act as control levers shaping designers’ cognition and action. However, the systemic effects of prompt type on agency construction, decision boundaries, and process strategy remain unclear. Treating the design setting as adaptive, we captured real-time interactions by collecting think-aloud data from 48 novice designers. Nine prompt categories were extracted and their cognitive effects were systematically analyzed through the Repertory Grid Technique (RGT), principal component analysis (PCA), and Ward clustering. These analyses revealed three perception profiles: tool-based, collaborative, and mentor-like. Strategy coding of 321 prompt-aligned utterances showed cluster-specific differences in path length, first moves, looping, and branching. Tool-based prompts reinforced boundary control through short linear refinements; collaborative prompts sustained moderate iterative enquiry cycles; mentor-like prompts triggered divergent exploration via self-loops and frequent jumps. We therefore propose a stage-adaptive framework that deploys mentor-like prompts for ideation, collaborative prompts for mid-phase iteration, and tool-based prompts for final verification. This approach balances creativity with procedural efficiency and offers a reusable blueprint for integrating prompt-driven agency modelling into GenAI design workflows. Full article

This study explores strategic deception and counter-deception in multi-agent reinforcement learning environments for a police officer–robber game. The research is motivated by real-world scenarios where agents must operate with partial observability and adversarial intent. We develop a suite of progressively complex grid-based environments featuring dynamic goals, fake targets, and navigational obstacles. Agents are trained using deep Q-networks (DQNs) with game-theoretic reward shaping to encourage deceptive behavior in the robber and intent inference in the police officer. The robber learns to reach the true goal while misleading the police officer, and the police officer adapts to infer the robber’s intent and allocate resources effectively. The environments include fixed and dynamic layouts with varying numbers of goals and obstacles, allowing us to evaluate scalability and generalization. Experimental results demonstrate that the agents converge to equilibrium-like behaviors across all settings. The inclusion of obstacles increases complexity but also strengthens learned policies when guided by reward shaping. We conclude that integrating game theory with deep reinforcement learning enables the emergence of robust, deceptive strategies and effective counter-strategies, even in dynamic, high-dimensional environments. This work advances the design of intelligent agents capable of strategic reasoning under uncertainty and adversarial conditions. Full article

Multi-agent reinforcement learning has demonstrated excellent performance in complex decision-making tasks such as electronic games, power grid management, and autonomous driving. However, its vulnerability to adversarial attacks may impede its widespread application. Currently, research on adversarial attacks in reinforcement learning primarily focuses on single-agent scenarios, while studies in multi-agent settings are relatively limited, especially regarding how to achieve optimized attacks with fewer steps. This paper aims to bridge the gap by proposing a heuristic exploration-based attack method named the Search for Key steps and Key agents Attack (SKKA). Unlike previous studies that train a reinforcement learning model to explore attack strategies, our approach relies on a constructed predictive model and a T-value function to search for the optimal attack strategy. The predictive model predicts the environment and agent states after executing the current attack for a certain period, based on simulated environment feedback. The T-value function is then used to evaluate the effectiveness of the current attack. We select the strategy with the highest attack effectiveness from all possible attacks and execute it in the real environment. Experimental results demonstrate that our attack method ensures maximum attack effectiveness while greatly reducing the number of attack steps, thereby improving attack efficiency. In the StarCraft Multi-Agent Challenge (SMAC) scenario, by attacking 5–15% of the time steps, we can reduce the win rate from 99% to nearly 0%. By attacking approximately 20% of the agents and 24% of the time steps, we can reduce the win rate to around 3%. Full article

The rapid growth of electric vehicle (EV) adoption necessitates advanced energy management strategies to ensure sustainable, reliable, and efficient operation of charging infrastructure. This study proposes a hybrid AI-based framework for optimizing residential EV charging systems through the integration of Reinforcement Learning (RL), Linear Programming (LP), and real-time grid-aware scheduling. The system architecture includes smart wall-mounted chargers, a 120 kWp rooftop solar photovoltaic (PV) array, and a 60 kWh lithium-ion battery energy storage system (BESS), simulated under realistic load conditions for 800 residential units and 50 charging points rated at 7.4 kW each. Simulation results, validated through SCADA-based performance monitoring using MATLAB/Simulink and OpenDSS, reveal substantial technical improvements: a 31.5% reduction in peak transformer load, voltage deviation minimized from ±5.8% to ±2.3%, and solar utilization increased from 48% to 66%. The AI framework dynamically predicts user demand using a non-homogeneous Poisson process and optimizes charging schedules based on a cost-voltage-user satisfaction reward function. The study underscores the critical role of intelligent optimization in improving grid reliability, minimizing operational costs, and enhancing renewable energy self-consumption. The proposed system demonstrates scalability, resilience, and cost-effectiveness, offering a practical solution for next-generation urban EV charging networks. Full article

This research introduces a comprehensive AI-driven framework for secure and efficient load management in multi-station electric vehicle (EV) charging networks, responding to the increasing demand and operational difficulties associated with widespread EV adoption. The suggested architecture has three main parts: a Smart Load Balancer (SLB), an AI-driven intrusion detection system (AIDS), and a Real-Time Analytics Engine (RAE). These parts use advanced machine learning methods like Support Vector Machines (SVMs), autoencoders, and reinforcement learning (RL) to make the system more flexible, secure, and efficient. The framework uses federated learning (FL) to protect data privacy and make decisions in a decentralized way, which lowers the risks that come with centralizing data. The framework makes load distribution 23.5% more efficient, cuts average wait time by 17.8%, and predicts station-level demand with 94.2% accuracy, according to simulation results. The AI-based intrusion detection component has precision, recall, and F1-scores that are all over 97%, which is better than standard methods. The study also finds important gaps in the current literature and suggests new areas for research, such as using graph neural networks (GNNs) and quantum machine learning to make EV charging infrastructures even more scalable, resilient, and intelligent. Full article

Given the increasing energy demand and the environmental consequences of fossil fuel consumption, the shift toward sustainable energy sources has become a global priority. Renewable Energy Communities (RECs)—comprising citizens, businesses, and legal entities—are emerging to democratise access to renewable energy. These communities allow members to produce their own energy, sharing or selling any surplus, thus promoting sustainability and generating economic value. However, scaling RECs while ensuring profitability is challenging due to renewable energy intermittency, price volatility, and heterogeneous consumption patterns. To address these issues, this paper presents a Machine Learning as a Service (MLaaS) framework, where each REC microgrid has a customised Reinforcement Learning (RL) agent and electricity price forecasts are included to support decision-making. All the conducted experiments, using the open-source simulator Pymgrid, demonstrate that the proposed agents reduced operational costs by up to 96.41% compared to a robust baseline heuristic. Moreover, this study also introduces two cost-saving features: Peer-to-Peer (P2P) energy trading between communities and internal energy pools, allowing microgrids to draw local energy before using the main grid. Combined with the best-performing agents, these features achieved trading cost reductions of up to 45.58%. Finally, in terms of deployment, the system relies on an MLOps-compliant infrastructure that enables parallel training pipelines and an autoscalable inference service. Overall, this work provides significant contributions to energy management, fostering the development of more sustainable, efficient, and cost-effective solutions. Full article