Search Results (661)

Combined sewer overflow (CSO) pollution threatens urban water environments, yet optimizing integrated green–grey infrastructure solutions remains computationally intensive, often making robust, large-scale multi-algorithm comparisons impractical. This study’s primary contribution is the development of an innovative physics-guided optimization framework that overcomes this computational barrier. By coupling a deep learning surrogate (trained on 60,000 scenarios generated in 7.7 h) with evolutionary algorithms, this framework provides a 6.2- to 7.7-fold acceleration in total project time (approximately 13 h vs. 80–100 h) compared to direct SWMM optimization. This significant speedup enabled a comprehensive comparative analysis of four multi-objective evolutionary algorithms (MOEAs), which established NSGA-II’s superiority in discovering a larger and more diverse set of optimal trade-off solutions. The physics-guided surrogate achieved an R² of 0.9965 and a Mean Absolute Error (MAE) corresponding to 0.5% of the baseline overflow volume. The validated framework successfully identified Permeable Pavement as the dominant control variable and a critical knee-point scenario. This solution, requiring a 426 million CNY investment, achieved a 67.0% overflow volume reduction and a 74.4% COD load reduction under the 5-year design storm. Furthermore, the optimized system demonstrated high resilience to extreme events, contrasting sharply with the failure of a cost-minimized approach, which underscores the importance of designing for resilience. This framework provides urban planners with a validated, efficient, and reliable methodology for designing resilient, cost-effective CSO control systems. Full article

The accurate and timely extraction of water bodies from high-resolution remote sensing imagery is vital for environmental monitoring, yet segmenting small, scattered, and irregularly shaped water bodies in complex rural landscapes remains a persistent challenge. While state-of-the-art deep learning models have advanced segmentation accuracy, they often achieve this at the cost of substantial computational overhead, limiting their practical application for large-scale monitoring. To address this trade-off between precision and efficiency, this paper introduces HWBENet, a novel hybrid network for water body extraction. HWBENet is built upon a lightweight MobileNetV3 encoder to ensure computational efficiency while preserving strong feature extraction capabilities. Its core innovation lies in two specifically designed modules. First, the Contextual Information Mining Module (CIMM) is proposed to enhance the network’s ability to learn and fuse both global scene-level context and fine-grained local details, which is crucial for identifying fragmented water bodies. Second, an Edge Refinement Module (ERM) is integrated into the decoder, which uniquely leverages transformer mechanisms to sharpen boundary details by effectively fusing prior feature information with up-sampled features. Extensive experiments on challenging rural water body datasets demonstrate that HWBENet strikes a superior balance between accuracy and computational cost. The experimental results validate the finding that HWBENet is an efficient, accurate, and scalable solution, offering significant practical value for large-scale hydrological mapping in complex rural environments. Full article

This study focuses on highway service areas. Building upon prior research that identified key influencing factors through surveys and ISM–MICMAC analysis, it constructs a tripartite evolutionary game model involving the government, service area operators, and carbon reduction technology providers based on stakeholder theory. Combined with MATLAB simulations, the model reveals the dynamic patterns of the carbon reduction system. The results indicate that government strategies exert the strongest influence on the system and catalyze the other two parties, followed by service area operators. Carbon reduction technology providers adopt a more cautious stance in decision-making. Government actions shape system evolution through a “cost-benefit-incentive” triple mechanism, with its strategies exhibiting significant spillover effects on other actors. Enterprise behavior is markedly influenced by Xinjiang’s regional characteristics, where the core barriers to corporate carbon reduction lie in the costs of proactive equipment and technological investments. The willingness of technology providers to cooperate primarily depends on two drivers: incremental baseline benefits and enhanced economies of scale. The core trade-off in government decision-making lies between the cost of strong regulation (Cg1) and the cost of environmental governance under weak regulation (Cg2). An increase in Cg1 prolongs the government’s convergence time by 233.3% and indirectly suppresses the willingness of enterprises and technology providers due to weakened subsidy capacity. Enterprises are relatively sensitive to the investment costs of carbon reduction equipment and technology, with convergence time extending by 120%. Technology providers are highly sensitive to incremental baseline returns (Rt), with stabilization time extending by 500%. Compared to existing research, this model quantitatively reveals the “cost-benefit-incentive” triple transmission mechanism for carbon reduction coordination in “grid-end” regions, identifying key parameters for strategic shifts among stakeholders. Based on this, corresponding policy recommendations are provided for all three parties, offering precise and actionable directions for the sustainable advancement of carbon reduction efforts in service areas. The research conclusions can provide a replicable collaborative framework for decarbonizing transportation infra-structure in grid-end regions with high clean energy endowments. Full article

The paper manufacturing industry faces increasing challenges in balancing operational costs with service quality under uncertain market conditions. This research presents a bi-objective credibility-based expected value model for integrated production–distribution planning that simultaneously minimizes total costs and maximizes service-level performance. The model considers multiple paper grades, production facilities, warehouses, and customer zones while handling demand uncertainty through credibility theory. Three additional constraints are introduced: service time limitations, capacity expansion decisions, and quality assurance requirements. The Torabi–Hassini (TH) method is employed to solve the bi-objective optimization problem effectively. Computational experiments demonstrate the model’s capability to provide balanced trade-off solutions between cost efficiency and service quality, achieving service-level improvements of 8–13% with cost increases of 5–9% compared to cost-only optimization, and cost reductions of 10–15% compared to service-only optimization. The results show that the credibility-based expected value approach provides robust and practical solutions for paper manufacturing supply chain optimization. Full article

Background: Gait event detection from inertial sensors offers scalable insights into locomotor health, with applications in clinical monitoring and mobile health. However, supervised methods are limited by scarce annotations, device variability, and sensor placement shifts. This in silico study evaluates self-supervised learning (SSL) as a resource-efficient strategy to improve robustness and generalizability. Methods: Six public smartphone and wearable inertial measurements unit (IMU) datasets (WISDM, PAMAP2, KU-HAR, mHealth, OPPORTUNITY, RWHAR) were harmonized within a unified deep learning pipeline. Models were pretrained on unlabeled windows using contrastive SSL with sensor-aware augmentations, then fine-tuned with varying label fractions. Experiments systematically assessed included (1) pretraining scale, (2) label efficiency, (3) augmentation contributions, (4) device/placement shifts, (5) sampling-rate sensitivity, and (6) backbone comparisons (CNN, TCN, BiLSTM, Transformer). Results: SSL consistently outperformed supervised baselines. Pretraining yielded accuracy gains of ΔF1 +0.08–0.15 and reduced stride-time error by −8 to −12 ms. SSL cut label needs by up to 95%, achieving competitive performance with only 5–10% labeled data. Sensor-aware augmentations, particularly axis-swap and drift, drove the strongest transfer gains. Robustness was maintained across sampling rates (25–100 Hz) and device/placement shifts. CNNs and TCNs offered the best efficiency–accuracy trade-offs, while Transformers delivered the highest accuracy at greater cost. Conclusions: This computational analysis across six datasets shows SSL enhances gait event detection with improved accuracy, efficiency, and robustness under minimal supervision, establishing a scalable framework for human performance and sports medicine in clinical and mobile health applications. Full article

Efficient coordination of multiple mobile robots is essential in automated systems, especially when robots must follow predefined paths while avoiding collisions. This paper proposes a centralized optimization framework using Genetic Algorithms to optimize the velocity profiles of a system of robots without altering their paths. The goal is to minimize task completion time and energy consumption while ensuring collision avoidance. Three Genetic Algorithm-based methods are introduced: Maximum Velocity Optimization, Slow-Down Segment Single-Objective Optimization and Slow-Down Segment Multi-Objective Optimization. The first method adjusts each robot’s maximum velocity along its entire path, whereas the second introduces a slow-down segment only at the start of its path. While these two approaches only optimize task completion time, the third method contains a multi-objective formulation, producing solutions that balance time and energy. Methods such as Brute-Force and Prioritized Planning were used as baseline methods for comparison. Simulation results indicate that the proposed strategies significantly outperform the baseline methods. Furthermore, the second method achieves better results than the first by introducing more targeted velocity adjustments, while the third further enhances flexibility by offering a range of trade-offs between task completion time and energy consumption. Scalability and computational cost remain critical challenges, especially as the number of robots increases. Full article

Joint overheating in low-voltage distribution cabinets presents a major safety risk, often leading to insulation failure, accelerated aging, and even fires. Conventional threshold-based inspection methods are limited in detecting early temperature evolution and lack predictive capabilities. To address this, a short-term temperature prediction method for electrical joints based on deep learning is proposed. Using a self-developed sensing device and Raspberry Pi edge nodes, multi-source data—including voltage, current, power, and temperature—were collected and preprocessed. Comparative experiments with ARIMA, GRU, and LSTM models demonstrate that the LSTM achieves the highest prediction accuracy, with an RMSE, MAE, and MAPE of 0.26 °C, 0.21 °C, and 0.54%, respectively. Furthermore, a lightweight version of the model was optimized for edge deployment, achieving a comparable accuracy (RMSE = 0.27 °C, MAE = 0.21 °C, MAPE = 0.67%) while reducing the inference latency and memory cost. The model effectively captures temperature fluctuations during 6 h prediction tasks and maintains stability under different cabinet scenarios. These results confirm that the proposed edge-enabled lightweight LSTM model achieves a balanced trade-off between accuracy, real-time performance, and efficiency, providing a feasible technical solution for intelligent temperature monitoring and predictive maintenance in low-voltage distribution systems. Full article

Infrared (IR) thermal sensors on CMOS-SOI-MEMS platforms enable scalable, low-cost thermal imaging but require optimized optical, thermal, and mechanical performance. This paper presents a multiphysics modeling framework to study the integration of Metasurface absorbers into a Thermal CMOS-SOI-MEMS IR sensor. Using finite-difference time-domain (FDTD) simulations, we demonstrate near-unity absorption at targeted wavelengths (e.g., 4.26 µm for CO₂ sensing, 10 µm for thermal imaging) compared to conventional absorbers. The absorbed power, calculated from blackbody irradiance, drives thermal finite element analysis (FEA), confirming high thermal isolation and maximized temperature rise (ΔT) while quantifying the thermal time constant’s sensitivity to Metasurface mass. An analytical RC circuit model, validated against 3D FEA, accurately captures thermal dynamics for rapid design iterations. Mechanical modal and harmonic analyses verify structural integrity, with natural frequencies above 20 kHz, ensuring resilience against mechanical resonances and environmental vibrations. This holistic framework quantifies trade-offs between optical efficiency, thermal responsivity, and mechanical stability, providing a predictive tool for designing high-performance, uncooled IR sensors compatible with CMOS processes. Full article

This paper addresses the simultaneous feeder routing and conductor sizing problem in unbalanced three-phase distribution systems, formulated as a nonconvex mixed-integer nonlinear program (MINLP) that minimizes the equivalent annualized expansion cost—combining investment and loss costs—under voltage, ampacity, and radiality constraints. The model captures nonconvex voltage–current–power couplings, $\Delta /Y$ load asymmetries, and discrete conductor selections, creating a large combinatorial design space that challenges heuristic methods. An exact MINLP formulation in complex variables is implemented in Julia/JuMP and solved with the Basic Open-source Nonlinear Mixed Integer programming (BONMIN) solver, which integrates branch-and-bound for discrete variables and interior-point methods for nonlinear subproblems. The main contributions are: (i) a rigorous, reproducible formulation that jointly optimizes routing and conductor sizing; (ii) a transparent, replicable implementation; and (iii) a benchmark against minimum spanning tree (MST)-based and metaheuristic approaches, clarifying the trade-off between computational time and global optimality. Tests on 10- and 30-node rural feeders show that, although metaheuristics converge faster, they often yield suboptimal solutions. The proposed MINLP achieves globally optimal, technically feasible results, reducing annualized cost by 14.6% versus MST and 2.1% versus metaheuristics in the 10-node system, and by 17.2% and 2.5%, respectively, in the 30-node system. These results highlight the advantages of exact optimization for rural network planning, providing reproducible and verifiable decisions in investment-intensive scenarios. Full article

This paper develops a Sustainable Artificial Intelligence-Driven Wind Power Forecasting System (SAI-WPFS) to enhance the integration of renewable energy while minimizing the environmental footprint of deep learning computations. Although deep learning models such as CNN, LSTM, and GRU have achieved high accuracy in wind power forecasting, existing research rarely considers the computational energy cost and associated carbon emissions, creating a gap between predictive performance and sustainability objectives. Moreover, limited studies have addressed the need for a balanced framework that jointly evaluates forecast precision and eco-efficiency in the context of large-scale renewable deployment. Using real-time data from the Dumat Al-Jandal Wind Farm, Saudi Arabia’s first utility-scale wind project, this study evaluates multiple deep learning architectures, including CNN-LSTM-AM and GRU, under a dual assessment framework combining accuracy metrics (MAE, RMSE, R²) and carbon efficiency indicators (CO₂ emissions per computational hour). Results show that the CNN-LSTM-AM model achieves the highest forecasting accuracy (MAE = 29.37, RMSE = 144.99, R² = 0.74), while the GRU model offers the best trade-off between performance and emissions (320 g CO₂/h). These findings demonstrate the feasibility of integrating sustainable AI into wind energy forecasting, aligning technical innovation with Saudi Vision 2030 goals for zero-carbon cities and carbon-efficient energy systems. Full article

As the importance of energy efficiency and smart readiness in the building sector has been on the rise, the financial evaluation of smart-ready technologies (SRTs) remains a gap in this field. This study introduces a methodology that comparatively evaluates the cost–benefit relationship between 11 different SRTs across three European countries—Cyprus, Italy and The Netherlands. Key performance indicators (KPIs) for energy-focused aspects such as Country-Specific Energy Savings Potential (CSESP) and Seasonal Smart Efficiency Coefficient (SSEC) and financial aspects such as Smart Readiness Cost Index (SRCI), Labor Cost Impact Factor (LCIF), Return on Smart Investment (RoSI), and Smart Investment Break-Even Period (SIBEP) were used to quantify the performance of the SRTs. The results indicate that regional labor rates, energy pricing, and climatic conditions—as well as relative technology cost–benefit tradeoffs—play a significant role in the economic viability of smart-ready devices. Having low labor costs and energy pricing, Cyprus exhibited the most cost-effective outcomes among the three countries. Italy showed strong returns although the initial investments were higher. The Netherlands was observed to benefit the most from heating-oriented technologies. The study comes to the conclusion that regionally specific methods are necessary for the adoption of SRTs and that techno-economic performance cannot be assessed separately from local market dynamics. The proposed framework supports stakeholders and policymakers in smart building investment and planning by offering a scalable method for device-level benchmarking. These indicators are developed specifically for this study and are not part of the official EU SRI (Smart Readiness Indicator) methodology. Their inclusion supports device-level evaluation and complements ongoing efforts toward SRI standardization. This research directly addresses Sustainable Development Goal (SDG) 7 on Affordable and Clean Energy, as well as SDG 11 on Sustainable Development, by evaluating how smart-ready technologies can contribute to energy efficiency and decarbonization in buildings. Based on the results, further research is needed to expand the indicator framework to additional technologies, include building typology effects, and integrate dynamic factors such as CO₂ pricing and real-time tariffs. Full article

The design of battery modules for Electric Vehicles (EVs) and stationary Energy Storage Systems (ESSs) plays a pivotal role in advancing sustainable energy technologies. This paper presents a comprehensive overview of the critical considerations in battery module design, including system requirements, cell selection, mechanical integration, thermal management, and safety components such as the Battery Disconnect Unit (BDU) and Battery Management System (BMS). We discuss the distinct demands of EV and ESS applications, highlighting trade-offs in cell chemistry, form factor, and architectural configurations to optimize performance, safety, and cost. Integrating advanced cooling strategies and robust electrical connections ensures thermal stability and operational reliability. Additionally, the paper describes a prototype battery module, a BDU, and the hardware and software architectures of a prototype BMS designed for a Hardware/Model-in-the-Loop framework for the real-time monitoring, protection, and control of battery packs. This work aims to provide a detailed framework and practical insights to support the development of high-performance, safe, and scalable battery systems essential for transportation electrification and grid energy storage. 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

The increasing adoption of high-performance DC motor control in embedded systems has driven the development of cost-effective solutions that extend beyond traditional software-based optimization techniques. This work presents a refined hardware-centric approach implementing real-time particle swarm optimization (PSO) directly executed on STM32 microcontroller for DC motor speed control, departing from conventional simulation-based parameter-tuning methods. Novel hardware-optimized composition of an interval type-2 fuzzy logic controller (FLC) and a PID controller is developed, designed for resource-constrained embedded systems and accounting for processing delays, memory limitations, and real-time execution constraints typically overlooked in non-experimental studies. The hardware-in-the-loop implementation enables real-time parameter optimization while managing actual system uncertainties in controlling DC micro-motors. Comprehensive experimental validation against conventional PI, PID, and PIDF controllers, all optimized using the same embedded PSO methodology, reveals that the proposed FT2-PID controller achieves superior performance with 28.3% and 56.7% faster settling times compared to PIDF and PI controllers, respectively, with significantly lower overshoot at higher reference speeds. The proposed hardware-oriented methodology bridges the critical gap between theoretical controller design and practical embedded implementation, providing detailed analysis of hardware–software co-design trade-offs through experimental testing that uncovers constraints of the low-cost microcontroller platform. Full article

Under the tightening carbon reduction policies, port microgrids face the challenge of optimizing the installed capacity of multiple power generation types to reduce operating costs and increase renewable energy penetration. We develop a bi-level cost-optimization framework in which the upper level decides long-term capacities (PV, wind, gas turbine, bio-fuel unit, and battery energy storage), and the lower level dispatches a multi-energy port microgrid (electricity–heat–cold) on an hourly basis with frequency regulation services. To ensure rigor and reproducibility, we (i) move the methodology upfront and formalize all constraints, (ii) provide a dedicated data–preprocessing pipeline for multi-region 50/60 Hz frequency time series, and (iii) map a policy intensity index to a carbon price and/or an annual cap used in the objective/constraints. The bi-level MILP is solved by a column-and-constraint generation algorithm with optimality gap control. We report quantitative metrics—annualized total cost, CO₂ emissions (t), renewable shares (%), and regulation cycles—across scenarios. Results show consistent cost–carbon trade-offs and robust capacity shifts toward storage and biofuel as policy tightens. All inputs and scripts are organized for exact replication. Full article