Search Results (1,819)

Given the intricate operating environment of steel truss bridges, data anomalies are frequently initiated by faults in the sensor monitoring system itself during the monitoring process. This paper utilizes a steel truss bridge as a case study in engineering, with a primary focus on the deflection of the main girder. The paper establishes an Attention Mechanism-based Bidirectional Long Short-Term Memory Neural Network (ATT-BiLSTM) model, with the objective of accurately repairing abnormal monitoring data. Firstly, correlation heat maps and Gray correlation are employed to detect anomalies in key measurement point data. Subsequently, the ATT-BiLSTM and Support Vector Machine (SVR) models are established to repair the anomalous monitoring data. Finally, various evaluation indexes, including Pearson’s correlation coefficient, mean squared error, and coefficient of determination, are utilized to validate the repairing accuracy of the ATT-BiLSTM model. The findings indicate that the repair efficacy of ATT-BiLSTM on anomalous data surpasses that of SVR. The repaired data exhibited a tendency to decrease in amplitude at the anomalous position, while maintaining the prominence of the data at abrupt deflection change points, thereby preserving the characteristics of the data. The repair rate of anomalous data attained 93.88%, and the mean square error of the actual complete data was only 0.0226, leading to substantial enhancement in the integrity and reliability of the data. Full article

To address the inefficiency of traditional concrete crack detection methods and the heavy reliance of supervised learning on extensive labeled data, in this study, an intelligent assessment method of concrete damage based on pseudo-label semi-supervised learning and fractal geometry theory is proposed to solve two core tasks: one is binary classification of pixel-level cracks, and the other is multi-category assessment of damage state based on crack morphology. Using three-channel RGB images as input, a dual-path collaborative training framework based on U-Net encoder–decoder architecture is constructed, and a binary segmentation mask of the same size is output to achieve the accurate segmentation of cracks at the pixel level. By constructing a dual-path collaborative training framework and employing a dynamic pseudo-label refinement mechanism, the model achieves an F1-score of 0.883 using only 50% labeled data—a mere 1.3% decrease compared to the fully supervised benchmark DeepCrack (F1 = 0.896)—while reducing manual annotation costs by over 60%. Furthermore, a quantitative correlation model between crack fractal characteristics and structural damage severity is established by combining a U-Net segmentation network with the differential box-counting algorithm. The experimental results demonstrate that under a cyclic loading of 147.6–221.4 kN, the fractal dimension monotonically increases from 1.073 (moderate damage) to 1.189 (failure), with 100% accuracy in damage state identification, closely aligning with the degradation trend of macroscopic mechanical properties. In complex crack scenarios, the model attains a recall rate (Re = 0.882), surpassing U-Net by 13.9%, with significantly enhanced edge reconstruction precision. Compared with the mainstream models, this method effectively alleviates the problem of data annotation dependence through a semi-supervised strategy while maintaining high accuracy. It provides an efficient structural health monitoring solution for engineering practice, which is of great value to promote the application of intelligent detection technology in infrastructure operation and maintenance. Full article

Alkaline electrolytic water hydrogen generation, a key driver in the growth of hydrogen energy, heavily relies on high-efficiency and high-purity ion exchange membranes. In this study, three-dimensional (3D) wrinkled reduced graphene oxide (WG) nanosheets obtained through a simple thermal reduction process and two-dimensional (2D) graphene oxide act as building blocks, with ethylenediamine as a crosslinking stabilizer, to construct a unique 3D/2D 2 nm-tunneling structure between the GO and WG sheets through via an amide connection at a WG/GO ratio of 1:1. Here, the wrinkled graphene (WG) undergoes a transition from two-dimensional (2D) graphene oxide (GO) into three-dimensional (3D) through the adjustment of surface energy. By increasing the interlayer spacing and the number of ion fluid channels within the membranes, the E-W/G membrane has achieved the rapid passage of hydroxide ions (OH⁻) and simultaneous isolation of produced gas molecules. Moreover, the dense 2 nm nano-tunneling structure in the electrolytic water process enables the E-W/G membrane to attain current densities >99.9% and an extremely low gas crossover rate of hydrogen and oxygen. This result suggests that the as-prepared membrane effectively restricts the unwanted crossover of gases between the anode and cathode compartments, leading to improved efficiency and reduced gas leakage during electrolysis. By enhancing the purity of the hydrogen production industry and facilitating the energy transition, our strategy holds great potential for realizing the widespread utilization of hydrogen energy. Full article

Unexpected machine failures in industrial environments lead to high maintenance costs, unplanned downtime, and safety risks. This study proposes a proactive predictive model using a hybrid of eXtreme Gradient Boosting (XGBoost) and Neural Networks (NN) to forecast machine failures. A synthetic dataset capturing recent breakdown history and time since last failure was used to simulate industrial scenarios. To address class imbalance, SMOTE and class weighting were applied, alongside a focal loss function to emphasize difficult-to-classify failures. The XGBoost model was tuned via GridSearchCV, while the NN model utilized ReLU-activated hidden layers with dropout. Evaluation using stratified 5-fold cross-validation showed that the NN achieved an F1-score of 0.7199 and a recall of 0.9545 for the minority class. XGBoost attained a higher PR AUC of 0.7126 and a more balanced precision–recall trade-off. Sample predictions demonstrated strong recall (100%) for failures, but also a high false positive rate, with most prediction probabilities clustered between 0.50–0.55. Additional benchmarking against Logistic Regression, Random Forest, and SVM further confirmed the superiority of the proposed hybrid model. Model interpretability was enhanced using SHAP and LIME, confirming that recent breakdowns and time since last failure were key predictors. While the model effectively detects failures, further improvements in feature engineering and threshold tuning are recommended to reduce false alarms and boost decision confidence. Full article

Background/Objectives: Current dosing recommendations for piperacillin/tazobactam suggest adjustments only for patients with creatinine clearance (CrCl) below 40 mL/min, potentially neglecting the variability in drug exposure among patients with a CrCl greater than 40 mL/min. This study aimed to develop a population pharmacokinetic (PK) model for piperacillin/tazobactam and explore optimal dosage regimens tailored by renal function and pathogen susceptibility. Methods: Twelve healthy adults received a single intravenous dose of piperacillin/tazobactam (4 g/0.5 g). Population PK models were developed using nonlinear mixed-effects modeling. Monte Carlo simulations were conducted to identify optimal dosing regimens across various renal functions and MIC levels, guided by pharmacodynamic targets defined as the percentage of time that free drug concentrations exceed the minimum inhibitory concentration (fT_>MIC). Results: PK profiles of both drugs were best described by two-compartment models. Estimated glomerular filtration rate (eGFR) adjusted by body surface area and body weight were identified as significant covariates influencing drug clearance and peripheral volume of distribution. Simulations showed that the standard dosing regimen (4/0.5 g q6h with 30 min infusion) achieved a 90% probability of target attainment (PTA) for 50%fT_>MIC at MIC values up to 4 mg/L in patients with normal renal function. However, this regimen often did not achieve a 90% PTA for stringent targets (100%fT_>MIC, 100%fT_>4MIC) or higher MICs, particularly in patients with eGFR ≥ 130 mL/min. Conclusions: These findings suggest current dosing regimens may be inadequate and highlight the potential of alternative strategies, such as extended or continuous infusion, which warrant further investigation in clinical populations to optimize therapeutic outcomes. Full article

People frequently choose not to be in an intimate relationship, but the reasons behind this choice vary. In the current study, we analyzed a dataset pooled from previous studies, consisting of 3226 Greek-speaking participants, 357 of whom were voluntarily single, to estimate the occurrence of different types of voluntary singlehood. We found that the largest subgroup, accounting for more than 60% of cases, consisted of individuals who indicated that they preferred to be single because they had different priorities. This was followed by those who indicated that they had given up on trying to attain an intimate relationship, comprising more than 26% of cases. Furthermore, about 13% of voluntarily single participants indicated that they were in this group for “other” reasons. Additionally, we found that participants in the different priorities group were single significantly longer than participants in the group who had given up on finding an intimate relationship. No significant sex differences were detected in the occurrence rates of the two types of voluntary singlehood. Moreover, younger participants were significantly more likely to indicate that they had different priorities than that they had given up on finding intimate relationships. Full article

In today’s industrial applications, cutting fluids have attained prime importance due to their all-round features, including increase of tool life by lubrication of the tool at the tool–workpiece interface. This study compares the effects of mango seed oil and pongamia oil on cutting force and surface morphology during the turning of AISI 304 steel. The design of experiments was applied using Taguchi’s method with an L9 array of experiments. During the experiment, it was discovered that mango seed and pongamia-based cutting fluid exhibited the lowest contact angles of 22.1° and 24.4°, respectively, at a 97:3 volumetric concentration of deionized water and eco-friendly oil. The use of mango seed oil as a cutting fluid with MQL (Minimum Quantity Lubrication) resulted in the lowest surface roughness of 0.809 µm, compared to 0.921 µm with pongamia-based cutting fluid. The cutting force was reduced by a maximum of 28.68% using mango seed-based cutting fluid, compared to pongamia-based cutting fluid. ANOVA analysis revealed that feed rate had the maximum influence on the optimization of output parameters for mango seed cutting fluid. For pongamia-based cutting fluid, feed rate had the maximum influence on cutting force, while the depth of cut had the strongest influence on surface roughness. Full article

This study introduces a novel model reference adaptive control (MRAC) framework that incorporates fractional-order gradients (FGs) to regulate the displacement of an inverted pendulum-cart system. Fractional-order gradients have been shown to significantly improve convergence rates in domains such as machine learning and neural network optimization. Nevertheless, their integration with fractional-order error models within adaptive control paradigms remains unexplored and represents a promising avenue for research. The proposed control scheme extends the classical MRAC architecture by embedding Caputo fractional derivatives into the adaptive law governing parameter updates, thereby improving both convergence dynamics and control flexibility. To ensure optimal performance across multiple criteria, the controller parameters are systematically tuned using a multi-objective Particle Swarm Optimization (PSO) algorithm. Two fractional-order error models (FOEMs) incorporating fractional gradients (FOEM2-FG, FOEM3-FG) are investigated, with their stability formally analyzed via Lyapunov-based methods under conditions of sufficient excitation. Validation is conducted through both simulation and real-time experimentation on a physical pendulum-cart setup. The results demonstrate that the proposed fractional-order MRAC (FOMRAC) outperforms conventional MRAC, proportional-integral-derivative (PID), and fractional-order PID (FOPID) controllers. Specifically, FOMRAC-FG achieved superior tracking performance, attaining the lowest Integral of Squared Error (ISE) of $2.32\times {10}^{-5}$ and the lowest Integral of Squared Input (ISI) of 6.40 in simulation studies. In real-time experiments, FOMRAC-FG maintained the lowest ISE ($5.11\times {10}^{-6}$). Under real-time experiments with disturbances, it still achieved the lowest ISE ($1.06\times {10}^{-5}$), highlighting its practical effectiveness. Full article

The interconnections among water, energy, and food (WEF) systems are growing increasingly complex, making it essential to understand their evolutionary mechanisms and coordination barriers to enhance regional resilience and sustainability. In this study, we investigated the WEF system in Northeast China by constructing a comprehensive indicator system encompassing resource endowment and utilization efficiency. The coupling coordination degree (CCD) of the WEF system was quantitatively assessed from 2001 to 2022. An obstacle degree model was employed to identify key constraints, while grey relational analysis was used to evaluate the driving influence of individual indicators. Furthermore, a co-evolution model based on logistic growth and competition–cooperation dynamics was developed to simulate system interactions. The results reveal the following: (1) the regional WEF-CCD increased from 0.627 in 2001 to 0.769 in 2022, reaching the intermediate coordination level, with the CCDs of the food, water, and energy subsystems rising from 0.39 to 0.62, 0.38 to 0.60, and 0.40 to 0.55, respectively, highlighting that the food subsystem had the most stable and significant improvement; (2) Jilin Province attained the highest WEF-CCD, 0.850, in 2022, while that for Heilongjiang remained the lowest, at 0.715, indicating substantial interprovincial disparities; (3) key indicators, such as food self-sufficiency rate, electricity generation, and ecological water use, functioned as both core constraints and major drivers of system performance; (4) co-evolution modeling revealed that the food subsystem exhibited the fastest growth, followed by water and energy (${\alpha }_3$ >${ \alpha }_1$ > ${ \alpha }_2$ > 0), with mutual promotion between water and energy subsystems and inhibitory effects from the food subsystem, ultimately converging toward a stable equilibrium state; and (5) interprovincial co-evolution modeling indicated that Jilin leads in WEF system development, followed by Liaoning and Heilongjiang, with predominantly cooperative interactions among provinces driving convergence toward a stable and coordinated equilibrium despite structural asymmetries. This study proposes a transferable, multi-method analytical framework for evaluating WEF coordination, offering practical insights into bottlenecks, key drivers, and co-evolutionary dynamics for sustainable resource governance. Full article

Underwater acoustic (UWA) communication supports many critical applications but still faces several physical-layer signal processing challenges. In response, recent advances in machine learning (ML) and deep learning (DL) offer promising solutions to improve signal detection, modulation adaptability, and classification accuracy. These developments highlight the need for a systematic evaluation to compare various ML/DL models and assess their performance across diverse underwater conditions. However, most existing reviews on ML/DL-based UWA communication focus on isolated approaches rather than integrated system-level perspectives, which limits cross-domain insights and reduces their relevance to practical underwater deployments. Consequently, this systematic literature review (SLR) synthesizes 43 studies (2020–2025) on ML and DL approaches for UWA communication, covering channel estimation, adaptive modulation, and modulation recognition across both single- and multi-carrier systems. The findings reveal that models such as convolutional neural networks (CNNs), long short-term memory networks (LSTMs), and generative adversarial networks (GANs) enhance channel estimation performance, achieving error reductions and bit error rate (BER) gains ranging from ${10}^{-3}$ to ${10}^{-6}$. Adaptive modulation techniques incorporating support vector machines (SVMs), CNNs, and reinforcement learning (RL) attain classification accuracies exceeding 98% and throughput improvements of up to 25%. For modulation recognition, architectures like sequence CNNs, residual networks, and hybrid convolutional–recurrent models achieve up to 99.38% accuracy with latency below 10 ms. These performance metrics underscore the viability of ML/DL-based solutions in optimizing physical-layer tasks for real-world UWA deployments. Finally, the SLR identifies key challenges in UWA communication, including high complexity, limited data, fragmented performance metrics, deployment realities, energy constraints and poor scalability. It also outlines future directions like lightweight models, physics-informed learning, advanced RL strategies, intelligent resource allocation, and robust feature fusion to build reliable and intelligent underwater systems. Full article

Speaker profiling systems are often evaluated on a single corpus, which complicates reliable comparison. We present a fully reproducible evaluation pipeline that trains Convolutional Neural Networks (CNNs) and Long-Short Term Memory (LSTM) models independently on three speech corpora representing distinct recording conditions—studio-quality TIMIT, crowdsourced Mozilla Common Voice, and in-the-wild VoxCeleb1. All models share the same architecture, optimizer, and data preprocessing; no corpus-specific hyperparameter tuning is applied. We perform a detailed preprocessing and feature extraction procedure, evaluating multiple configurations and validating their applicability and effectiveness in improving the obtained results. A feature analysis shows that Mel spectrograms benefit CNNs, whereas Mel Frequency Cepstral Coefficients (MFCCs) suit LSTMs, and that the optimal Mel-bin count grows with corpus Signal Noise Rate (SNR). With this fixed recipe, EfficientNet achieves 99.82% gender accuracy on Common Voice (+1.25 pp over the previous best) and 98.86% on VoxCeleb1 (+0.57 pp). MobileNet attains 99.86% age-group accuracy on Common Voice (+2.86 pp) and a 5.35-year MAE for age estimation on TIMIT using a lightweight configuration. The consistent, near-state-of-the-art results across three acoustically diverse datasets substantiate the robustness and versatility of the proposed pipeline. Code and pre-trained weights are released to facilitate downstream research. Full article

Low regression rate is the most critical issue for the development and application of hybrid rocket motors (HRMs). Paraffin-based fuels are potential candidates for HRMs due to their high regression rates but adding polymers to improve strength results in insufficient regression rates for HRMs applications. In this work, Computational Fluid Dynamics (CFD) modeling and analysis were used to investigate the mixing and combustion of gaseous fuels and oxidizers in HRMs for various fuel grains and injector combinations. In addition, the regression rate characteristics and combustion efficiency were evaluated using a ground test. The results showed that the swirling flow with both high mixing intensity and high velocity could be formed by using the swirl injector. The highest mixing degree attained for the star-swirl grain and swirl injector was 86%. The reported combustion efficiency calculated by the CFD model attained a maximum of 93% at the nozzle throat. In addition, a spatially averaged regression rate of 1.40 mm·s⁻¹ was achieved for the star-swirl grain and swirl injector combination when the mass flux of N₂O was 89.94 kg·m⁻²·s⁻¹. This is around 191% higher than the case of non-swirling flow. However, there were obvious local regression rate differences between the root of the star and the slot. The regression rate increase was accompanied by a decrease in the combustion efficiency for the strong swirling flow condition due to the remarkable higher mass flow rate of gasified fuels. It was shown that the nano-sized aluminum was unfavorable for the combustion efficiency, especially under extreme fuel-rich conditions. Full article

Optimizing hydrodynamic performance and dissolved oxygen (DO) distribution is essential for improving water quality management in industrial recirculating aquaculture systems. This study combines experimental measurements and data analysis to evaluate the effects of the inlet pipe flow rate (Q), deployment distance ratio (d/r), deployment angle (θ), inlet pipe structure on hydrodynamics and the dissolved oxygen distribution across various tank layers. The flow field distribution in the tanks was measured using Acoustic Doppler Velocimetry (ADV), and the hydrodynamic characteristics, including average velocity (v_avg) and the velocity uniformity coefficient (DU₅₀), were quantitatively analyzed. The dissolved oxygen content at different tank layers was recorded using an Aquameter GPS portable multi-parameter water quality analyzer. The findings indicate that average velocity (v_avg) and the velocity uniformity coefficient (DU₅₀) are key determinants of the hydrodynamic characteristic of circular aquaculture tanks. Optimal hydrodynamic performance occurs for the vertical single-pipe porous configuration at Q = 9 L/s, d/r = 1/4, and θ = 45°,the average velocity reached 0.0669 m/s, and the uniformity coefficients attained a maximum value of 40.4282. In a vertical single-pipe porous structure, the tank exhibits higher dissolved oxygen levels compared to a horizontal single-pipe single-hole structure. Under identical water inflow rates and deployment distance ratios, dissolved oxygen levels in the surface layer of the circular aquaculture tank are significantly greater than that in the bottom layer. The results of this study provide valuable insights for optimizing the engineering design of industrial circular aquaculture tanks and addressing the dissolved oxygen distribution across different water layers. Full article

This study aimed to examine the 6-month hospital readmission rate for Transition Care Program (TCP) clients and its association with community goal attainment. This was a single-site retrospective cohort study of TCP clients admitted from 2014 to 2019. Goals were set at TCP entry and coded as goals ‘within the home’ or ‘in the community’. Hospital readmissions were tracked using electronic health records. Logistic regression, area under the curve, and number needed to treat were the primary analyses performed. Of 747 (66.8% female and 33.2% male) client episodes, 164 (22%) resulted in a hospital readmission. Clients who were not readmitted to hospital set and achieved a higher number of community-based goals (1.08 vs. 0.8, p = 0.01 and 0.8 vs. 0.6, p = 0.001). Utilising a logistic regression model, each additional community goal achieved was associated with a 30% reduction in risk of readmission to the hospital (OR: 0.69, 95%CI: 0.5–0.8; p = 0.002), adjusted for age, sex, MBI change, number of home goals achieved, hospital length of stay and number of comorbidities. Achieving community-based goals can reduce the risk of hospital readmission by 30% after adjusting for demographic and clinical variables. Full article

Objective: Gut dysbiosis has been implicated in the pathology of inflammatory bowel disease (IBD). There is some evidence to suggest that the use of antibiotic treatment can incite an early clinical response or remission when used in conjunction with standard-of-care (SOC) therapy to treat IBD-related flares. Furthermore, antibiotics have been historically investigated for use as a bridge when initiating biologic therapy while waiting for peak biologic treatment effect to occur. This study investigated and compared the time to clinical response when treated with combination antibiotics, metronidazole monotherapy, or SOC therapy in pediatric patients with an active IBD flare. Methods: This study was a retrospective, Institution Review Board-approved, single-centered cohort study which included patients who were less than 18 years of age with a confirmed diagnosis of IBD who received conventional treatment alone or with either combination antibiotic therapy or metronidazole monotherapy to treat an active IBD flare between March 2013 and January 2024. Patients were excluded if they received antibiotic therapy to treat an active infection, had positive stool cultures or enteric pathogen polymerase chain reaction panel, or had colonic disease limited to the rectum. Results: Fifty-nine patients were included and divided into metronidazole monotherapy (n = 18), SOC therapy (n = 20), and combination antibiotics (n = 21). The primary outcome of days to clinical response was not significantly different across all groups; however, patients who received combination antibiotics achieved the fastest time to clinical response (median (IRQ))—4 days (1, 65), compared to 7.5 days (1, 119) for the SOC group and 9 days (2, 217) for the metronidazole group. Secondary outcomes of achievement of clinical response, remission, or failure were determined to be non-significant between all groups. Conclusions: There is no significant difference in time to clinical response, attaining clinical response or remission, or treatment failure rate for patients treated with combination antibiotics, metronidazole monotherapy, or SOC. However, results of this study suggest that the use of combination antibiotics plus SOC may lead to a faster time to clinical response and remission compared to SOC therapy alone. Further studies are warranted to elucidate the role of antimicrobial therapy in management of pediatric IBD. Full article