Search Results (45)

Early detection of agricultural fires with Unmanned Aerial Vehicles (UAVs) is important for environmental safety, yet it remains difficult because ignition cues are extremely small, smoke patterns vary widely, and farmland scenes often contain strong background interference such as specular reflections. Model development is further constrained by the scarcity of data from the early ignition stage. To address these challenges, we propose a joint data and model optimization framework. We first build a hybrid dataset through an ROI-guided synthesis pipeline, in which latent diffusion models are used to insert high-fidelity, carefully screened fire samples into real farmland backgrounds. We then introduce EF-YOLO, a detector designed for high sensitivity to small targets. The network uses SPD-Conv to reduce feature loss during spatial downsampling and includes a high-resolution P2 head to improve the detection of minute objects. To reduce background clutter, a Dual-Path Frequency–Spatial Enhancement (DP-FSE) module serves as a lightweight statistical surrogate that extracts global contextual cues and local salient features in parallel, thereby suppressing high-frequency noise. Experimental results show that EF-YOLO achieves an AP_s of 40.2% on sub-pixel targets, exceeding the YOLOv8s baseline by 15.4 percentage points. With a recall of 88.7% and a real-time inference speed of 78 FPS, the proposed framework offers a strong balance between detection performance and efficiency, making it well suited for edge-deployed agricultural fire early-warning systems. Full article

Timber beams have assumed a prominent role in contemporary structural engineering, driven by sustainability requirements and the advancement of engineered wood products. Despite the evident environmental and building advantages, the performance of timber beam elements under fire conditions remains one of the main design challenges, due to the strong nonlinearity of thermal behavior, progressive charring, and degradation of mechanical properties. In this context, numerical simulations have become a central tool for the thermal and thermomechanical assessment of timber beams exposed to fire. This study presents a technical and critical review of numerical approaches applied to timber beam elements, with emphasis on finite element–based models, thermal modeling strategies, representation of charring, thermomechanical coupling, and the use of reduced-order and surrogate models. The distinctive contribution of this work lies in an integrated and critical analysis of these approaches, explicitly articulating high-fidelity numerical models with reduced-order and symbolic models, aiming at their use as complementary tools in structural design. The analysis was conducted thematically, based on literature selected from major international databases, emphasizing modeling assumptions, levels of numerical complexity, and methodological limitations. The results indicate a predominance of transient finite element (FEM) models, widespread use of two-dimensional cross-sectional analyses, increasing adoption of enthalpy-based formulations for charring, and a prevalence of sequential thermomechanical coupling strategies. In contrast, the literature reveals strong heterogeneity in thermal parameters, limited standardization of validation procedures, restricted use of probabilistic approaches, and still incipient integration of reduced-order and symbolic models. It is concluded that future advances in the field depend on the standardization of modeling strategies, the expansion of thermal property databases, and, above all, the integration of high-fidelity models with interpretable reduced-order models, capable of supporting parametric analyses and performance-based structural design methodologies. Full article

This paper presents a physics-consistent hybrid surrogate framework for simulating the mechanical behavior of reinforced concrete beams that utilize waste fired clay (WFC) as a partial substitute for cement. The main contribution is the integration of empirically observed deformation behavior with physics-informed learning to produce an interpretable, mechanically valid surrogate model. Full-field surface deformation fields were measured using Digital Image Correlation (DIC) under monotonic loading and processed through a convolutional neural network (CNN) to extract deformation- and crack-sensitive features. These features were integrated with experimentally measured stress–strain data within a Physics-Informed Neural Network (PINN) in which equilibrium and conditional constitutive monotonicity constraints were enforced through the loss function. A Non-Dominated Sorting Genetic Algorithm II (NSGA-II) was utilized as a downstream parametric exploration tool to examine trade-offs among maximum load capacity, material cost, and embodied CO₂ inside a constrained mixture-design space. Model interpretability was assessed by SHapley Additive exPlanations (SHAP), indicating that deformation-driven kinematic factors predominantly influence stress prediction, whereas WFC content and reinforcement parameters have a secondary, mixture-level impact. The resulting framework achieves enhanced predictive accuracy (R² = 0.969) relative to its individual components and operates as an offline, physics-calibrated surrogate rather than a real-time digital twin, providing a reliable and interpretable basis for structural assessment and sustainability-oriented design evaluation of WFC-modified reinforced concrete beams. Full article

The decarbonisation of urban heating systems represents a key challenge for the transition towards sustainable cities. This study investigates the field integration of a Medium-Temperature Heat Pump (MTHP) within the Osimo District Heating Network (DHN) in Italy, demonstrating how low-grade return flows (30–50 °C) can be effectively upgraded to supply temperatures of 65–75 °C, in line with 4th-generation district heating requirements. Specifically, 5256 h of MTHP operation within the DHN were analysed to validate the initial design assumptions, develop surrogate performance models, and assess the system’s techno-economic and environmental performance. The results indicate stable and reliable operation, with a weighted average Coefficient of Performance (COP) of 3.96 and a weighted average thermal output of 134.5 kW. From an economic perspective, the system achieves a payback period of approximately six years and a Levelised Cost of Heat (LCOH) of 0.0245 €/kWh. Environmentally, the MTHP enables CO₂ emission reductions of about 120 t compared with conventional gas-fired boilers. Beyond its technical performance, the study highlights the strong replicability of MTHP solutions for small- and medium-scale DHNs across Europe. The proposed approach offers urban utilities a scalable and cost-competitive pathway towards low-carbon heat supply, directly supporting municipal climate strategies and aligning with key EU policy frameworks, including the European Green Deal, REPowerEU, and the “Fit-for-55” package. Full article

The paper introduces a real-time digital-twin controller that manages evacuation routes while operating GEEM for emergency energy management during building fires. The system consists of three interconnected parts which include (i) a physics-based hazard surrogate for short-term smoke and temperature field prediction from sensor data (ii), a router system that manages path updates for individual users and controls exposure and network congestion (iii), and an energy management system that regulates the exchange between PV power and battery storage and diesel fuel and grid electricity to preserve vital life-safety operations while reducing both power usage and environmental carbon output. The system operates through independent modules that function autonomously to preserve operational stability when sensors face delays or communication failures, and it meets Industry 5.0 requirements through its implementation of auditable policy controls for hazard penalties, fairness weight, and battery reserve floor settings. We evaluate the controller in co-simulation across multiple building layouts and feeder constraints. The proposed method achieves superior performance to existing AI/RL baselines because it reduces near-worst-case egress time ($T_{95}$ and worst-case exposure) and decreases both event energy $E_{\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{t}}$ and CO₂-equivalent ${{CO}_2}_{\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{t}}$ while upholding all capacity, exposure cap, and grid import limit constraints. A high-VRE, tight-feeder stress test shows how reserve management, flexible-load shedding, and PV curtailment can achieve trade-offs between unserved critical load $U_{\mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y}}$ and emissions. The team delivers implementation details together with reporting templates to assist researchers in reaching reproducibility goals. The research shows that emergency energy systems, which integrate evacuation systems, achieve better safety results and environmental advantages that enable smart-city integration through digital thread operations throughout design, commissioning, and operational stages. Full article

Ship damage control (DC) is pivotal to platform survivability in the face of battle damage and severe accidents. The DC context features multi-hazard coupling among flooding, fire, and smoke, as well as fast system dynamics and intensive human–machine collaboration, demanding real-time predictive simulation and decision support. Conventional DC simulations fall short in multiphysics fidelity, predictive speed, and integration with onboard sensing and control. A digital twin (DT) framework for predictive shipboard DC is introduced with an explicit capability envelope, observability, and latency requirements, and a cyber-physical mapping to ship systems. Building on this foundation, a three-stage/four-level maturity model charts progression from L1 monitoring, through L2 prediction and L3 human-in-the-loop, override-enabled plan generation, to L4 closed-loop decision control, specifying capability milestones and evaluation metrics. Guided by this model, a four-layer architecture and an end-to-end roadmap are formulated, spanning multi-domain modeling, multi-source sensing and fusion, surrogate-accelerated multiphysics simulation, assisted plan generation with human approval/override, and cyber-physical closed-loop control. The framework aligns interfaces, performance targets, and verification pathways, providing actionable guidance to upgrade shipboard DC toward resilient, efficient, and human-centric operation under multi-hazard coupling. Full article

We introduce an ultra-compact, single-neuron-per-class end-to-end readout for binary classification of noisy, image-encoded sensor time series. The approach compares a linear single-unit perceptron (E2E-MLP-1) with a resonate-and-fire (RAF) neuron (E2E-RAF-1), which merges feature selection and decision-making in a single block. Beyond empirical evaluation, we provide a mathematical analysis of the RAF readout: starting from its subthreshold ordinary differential equation, we derive the transfer function $H\left(\mathrm{j}\omega \right)$, characterize the frequency response, and relate the output signal-to-noise ratio (SNR) to ${\left|H\left(\mathrm{j}\omega \right)\right|}^2$ and the noise power spectral density $S_n\left(\omega \right)\propto {\omega }^{\alpha }$ (brown, pink, and blue noise). We present a stable discrete-time implementation compatible with surrogate gradient training and discuss the associated stability constraints. As a case study, we classify walk-in-place (WIP) in a virtual reality (VR) environment, a vision-based motion encoding (72 × 56 grayscale) derived from 3D trajectories, comprising 44,084 samples from 15 participants. On clean data, both single-neuron-per-class models approach ceiling accuracy. At the same time, under colored noise, the RAF readout yields consistent gains (typically +5–8% absolute accuracy at medium/high perturbations), indicative of intrinsic band-selective filtering induced by resonance. With ∼8 k parameters and sub-2 ms inference on commodity graphical processing units (GPUs), the RAF readout provides a mathematically grounded, robust, and efficient alternative for stochastic signal processing across domains, with virtual reality locomotion used here as an illustrative validation. Full article

The traditional optimization of intumescent flame-retardant polypropylene (PP) relies on large experimental campaigns that scale poorly with compositional dimensionality, limiting the systematic exploration of tradeoffs between fire performance and material economy. We present a Multi-Objective Bayesian Optimization (MOBO) workflow that couples Gaussian Process (GP) surrogates with the q-Noisy Expected Hypervolume Improvement (qNEHVI) acquisition to co-optimize two competing objectives: maximize the Limiting Oxygen Index (LOI) and minimize total flame-retardant (FR) loading (wt.%). Two practical initialization strategies, Space-Filling Design and literature-guided sampling, are benchmarked, and convergence is monitored via dominated hypervolume and uncertainty calibration. Uniform design-space coverage yields faster hypervolume growth and better-calibrated uncertainty than literature seeding. Under a 20-experiment budget, the best formulation attains an LOI = 27.0 vol.% at 22.74 wt.% FR, corresponding to an estimated 8–14% efficiency gain, defined here as LOI improvement at comparable FR loadings relative to representative baselines. The recovered APP/PER stoichiometric ratios (1.69–2.26) are consistent with established intumescence mechanisms, indicating that a data-driven search can converge to physically meaningful solutions without explicit mechanistic priors. The proposed workflow provides a sample-efficient route to navigate multi-criteria design spaces in flame-retardant PP and is transferable to polymer systems in which performance, cost, and processing constraints must be balanced and exhaustive testing is impractical. Full article

Orbital maneuver detection is critical for space situational awareness, yet it remains challenging due to the complex and dynamic nature of satellite behaviors. This paper proposes a novel Multi-Level Firing Spiking Neural Network (MLF-SNN) for detecting orbital maneuvers based on changes in satellite orbital parameters. The MLF-SNN incorporates multiple firing thresholds and a leaky integrate-and-fire (LIF) neuron model to enhance temporal feature extraction and classification performance. The MLF-SNN encodes time-dependent input features, which include variations in orbital elements, and subsequently processes these features through a multi-layer spiking architecture. A surrogate gradient approach is adopted during training to enable end-to-end backpropagation through the spiking layers. Experimental results on real satellite data demonstrate that the proposed method achieves improved recall in maneuver detection compared to conventional approaches, effectively reducing false alarms and missed detections. The work highlights the potential of MLF-SNN in processing time-series spatial data and offers a robust solution for autonomous satellite behavior analysis. Full article

Spiking neural networks (SNNs) provide a biologically inspired, event-driven alternative to artificial neural networks (ANNs), potentially delivering competitive accuracy at substantially lower energy. This tutorial-study offers a unified, practice-oriented assessment that combines critical review and standardized experiments. We benchmark a shallow fully connected network (FCN) on MNIST and a deeper VGG7 architecture on CIFAR-10 across multiple neuron models (leaky integrate-and-fire (LIF), sigma–delta, etc.) and input encodings (direct, rate, temporal, etc.), using supervised surrogate-gradient training implemented in Intel Lava, SLAYER, SpikingJelly, Norse, and PyTorch. Empirically, we observe a consistent but tunable trade-off between accuracy and energy. On MNIST, sigma–delta neurons with rate or sigma–delta encodings achieve 98.1% accuracy (ANN baseline: 98.23%). On CIFAR-10, sigma–delta neurons with direct input reach 83.0% accuracy at just two time steps (ANN baseline: 83.6%). A GPU-based operation-count energy proxy indicates that many SNN configurations operate below the ANN energy baseline; some frugal codes minimize energy at the cost of accuracy, whereas accuracy-oriented settings (e.g., sigma–delta with direct or rate coding) narrow the performance gap while remaining energy-conscious—yielding up to threefold efficiency compared with matched ANNs in our setup. Thresholds and the number of time steps are decisive factors: intermediate thresholds and the minimal time window that still meets accuracy targets typically maximize efficiency per joule. We distill actionable design rules—choose the neuron–encoding pair according to the application goal (accuracy-critical vs. energy-constrained) and co-tune thresholds and time steps. Finally, we outline how event-driven neuromorphic hardware can amplify these savings through sparse, local, asynchronous computation, providing a practical playbook for embedded, real-time, and sustainable AI deployments. Full article

This paper presents an integrated approach for optimizing the performance of a 36-pulses converter system by using artificial intelligence (AI) techniques to be included in a Supervisory Control and Data Acquisition (SCADA) environment. The focus of the proposal is on enhancing harmonic reduction through intelligent adjustment of switching angles and coordinated control of the reinjection transformer included in the power converter topology. A key component of the proposed methodology involves a simulation-based process to determine optimal firing angles (${\mathit{\alpha }}_1$, ${\mathit{\alpha }}_2$, and ${\mathit{\alpha }}_3$), based on Selective Harmonic Elimination (SHE) theory, that minimize Total Harmonic Distortion (THD). Using MATLAB with Simulink and PLECS models, a parametric sweep of the firing angles, generating a comprehensive dataset of THD outcomes. This dataset, consisting of THD evaluations across fine-grained angle variations, serves as the training foundation for supervised machine learning models—specifically, neural network regressors—that approximate the nonlinear mapping between firing angles and harmonic distortion. These predictive models are then employed as surrogates to estimate THD rapidly and guide the selection of optimal switching angles in real time without requiring iterative numerical solvers. Optimization heuristics and predictive models are then deployed to dynamically adapt system parameters in real time under varying load conditions. The proposed method demonstrates significant improvements in power quality and operational reliability, highlighting the potential of AI-assisted SCADA systems in advanced power electronics applications. Implementation results performed on a 36-pulses voltage source converter prototype are included to illustrate the appropriateness of the proposal. Full article

Surrogate modelling is increasingly used in engineering to improve computational efficiency in complex simulations. However, traditional data-driven surrogate models often face limitations in generalizability, physical consistency, and extrapolation—issues that are especially critical in safety-sensitive fields such as fire safety engineering (FSE). To address these concerns, physics-informed surrogate modelling (PISM) integrates physical laws into machine learning models, enhancing their accuracy, robustness, and interpretability. This systematic review synthesises existing applications of PISM in FSE, classifies the strategies used to embed physical knowledge, and outlines key research challenges. A comprehensive search was conducted across Google Scholar, ResearchGate, ScienceDirect, and arXiv up to May 2025, supported by backward and forward snowballing. Studies were screened against predefined criteria, and relevant data were analysed through narrative synthesis. A total of 100 studies were included, covering five core FSE domains: fire dynamics, wildfire behaviour, structural fire engineering, material response, and heat transfer. Four main strategies for embedding physics into machine learning were identified: feature engineering techniques (FETs), loss-constrained techniques (LCTs), architecture-constrained techniques (ACTs), and offline-constrained techniques (OCTs). While LCT and ACT offer strict enforcement of physical laws, hybrid approaches combining multiple strategies often produce better results. A stepwise framework is proposed to guide the development of PISM in FSE, aiming to balance computational efficiency with physical realism. Common challenges include handling nonlinear behaviour, improving data efficiency, quantifying uncertainty, and supporting multi-physics integration. Still, PISM shows strong potential to improve the reliability and transparency of machine learning in fire safety applications. Full article

The coal cost of coal-fired units accounts for more than 70% of the total power generation cost. In addition to determining coal costs, coal blending strategies (CBS) significantly impact various types of costs, such as pollutant removal and emissions. To address these issues, we propose a framework for generating cost-effective CBS. The framework includes a unit output condition recognition module (UOCR) that enables the adaptive classification of output conditions based on historical operation datasets, performing intelligent condition recognition with Imitator and pre-trained image classification models using blending strategies and unit parameters as inputs. The cost-effective strategy generation module (CESG) employs a surrogate model to evaluate the economic viability of strategies in terms of coal and environmental costs, among other factors. It also employs UOCR as another surrogate model to validate strategy feasibility. Cost-effective strategies are generated via a population-based metaheuristic algorithm. In the case study, the UOCR achieved an average training accuracy of 96.64%, and the generated cost-effective strategies reduced costs by an average of 3.37% compared to currently implemented strategies. Full article

More than half of South Korea’s land area is covered by forests, which significantly increases the potential for extensive damage in the event of a forest fire. The majority of forest fires in South Korea are caused by humans. Over the past decade, more than half of these types of fires occurred during the spring season. Although human activities are the primary cause of forest fires, the fact that they are concentrated in the spring underscores the strong association between forest fires and meteorological factors. When meteorological conditions favor the occurrence of forest fires, certain triggering factors can lead to their ignition more easily. The purpose of this study is to analyze the meteorological factors influencing forest fires and to develop a machine learning-based prediction model for forest fire occurrence, focusing on meteorological data. The study focuses on four regions within Gangwon province in South Korea, which have experienced substantial damage from forest fires. To construct the model, historical meteorological data were collected, surrogate variables were calculated, and a variable selection process was applied to identify relevant meteorological factors. Five machine learning models were then used to predict forest fire occurrence and ensemble techniques were employed to enhance the model’s performance. The performance of the developed forest fire prediction model was evaluated using evaluation metrics. The results indicate that the ensemble model outperformed the individual models, with a higher F1-score and a notable reduction in false positives compared to the individual models. This suggests that the model developed in this study, when combined with meteorological forecast data, can potentially predict forest fire occurrence and provide insights into the expected severity of fires. This information could support decision-making for forest fire management, aiding in the development of more effective fire response plans. Full article

Spiking neural networks (SNNs), using action potentials (spikes) to represent and transmit information, are more biologically plausible than traditional artificial neural networks. However, most of the existing SNNs require a separate preprocessing step to convert the real-valued input into spikes that are then input to the network for processing. The dissected spike-coding process may result in information loss, leading to degenerated performance. However, the biological neuron system does not perform a separate preprocessing step. Moreover, the nervous system may not have a single pathway with which to respond and process external stimuli but allows multiple circuits to perceive the same stimulus. Inspired by these advantageous aspects of the biological neural system, we propose a self-adaptive encoding spike neural network with parallel architecture. The proposed network integrates the input-encoding process into the spiking neural network architecture via convolutional operations such that the network can accept the real-valued input and automatically transform it into spikes for further processing. Meanwhile, the proposed network contains two identical parallel branches, inspired by the biological nervous system that processes information in both serial and parallel. The experimental results on multiple image classification tasks reveal that the proposed network can obtain competitive performance, suggesting the effectiveness of the proposed architecture. Full article