Search Results (3,133)

Fatigue has always been one of the major causes of structural failure, where repeated loading and unloading cycles reduce the fracture energy of the material, causing it to fail at stresses lower than its monotonic strength. However, predicting fatigue life is a highly challenging task and, in this context, the present study proposes a fundamentally new hybrid physics-informed and data-driven approach. Firstly, an energy-based fatigue model is developed to simulate the behavior of concrete under compressive cyclic fatigue loading. The data generated from these numerical simulations are then utilized to train machine learning (ML) models. The stress–strain curve and S-N curve of concrete under compression, obtained from the energy-based model, are validated against experimental data. For the ML models, two different algorithms are used as follows: k-Nearest Neighbors (KNN) and Deep Neural Networks (DNN), where a total of 1962 data instances generated from numerical simulations are used for the training and testing of the ML models. Furthermore, the performance of the ML models is evaluated for out-of-range inputs, where the DNN model with three hidden layers (a complex model with 128, 64, and 32 neurons) provides the best predictions, with only a 0.6% overall error. Full article

This paper proposes an intelligent control method based on Soft Actor-Critic (SAC) to address uncertainties faced by flight vehicles during flight. The method effectively reduces aerodynamic loads and enhances the reliability of structural strength under significant wind disturbances. A specific launch vehicle is taken as the research subject, and its dynamic model is established. A deep reinforcement learning (DRL) framework suitable for the attitude control problem is constructed, along with a corresponding training environment. A segmented reward function is designed: the initial stage emphasizes tracking accuracy, the middle stage, with a detrimental effect due to the high-altitude wind region, focuses on load relief, and the final stage gradually resumes following tracking accuracy on the basis of maintaining the effect of load relief. The reward function dynamically switches between stages using a time factor. The improved SAC algorithm is employed to train the agent over multiple epochs, ultimately resulting in an intelligent load relief attitude controller applicable to the launch vehicle. Simulation experiments demonstrate that this method effectively solves the attitude control problem under random wind disturbances, particularly reducing the aerodynamic loads of launch vehicles in the high-altitude wind region. Full article

Background and Objectives: This study aimed to investigate the effects of different tension loads in post-activation potentiation protocols on agility and vertical jump performance across different times of day in trained judokas, addressing a significant gap in understanding the interaction between diurnal variations and post-activation potentiation protocol responses in combat sports. Materials and Methods: Seventeen male judokas (age: 21.41 ± 1.37 years) with ≥2 years of training experience participated in the study. Participants completed three different protocols: specific warm-up, the 80% post-activation potentiation protocol, and the 100% post-activation potentiation protocol, performed both in the morning (09:00–11:00) and evening (17:00–19:00) sessions. Performance was assessed using the Illinois Agility Test and countermovement jump. Protocols were randomized and counterbalanced over a 3-week period, with a minimum 48 h recovery between sessions. Statistical analysis employed repeated measures ANOVA (3 × 2) with Greenhouse–Geisser corrections. Results: Significant differences were observed in both protocols and time interactions for agility (F = 41.691, ηp² = 0.864, p < 0.001; F = 23.893, ηp² = 0.123, p < 0.001) and countermovement jump performance (F = 7.471, ηp² = 0.410, p = 0.002; F = 38.651, ηp² = 0.530, p < 0.001). The 80% post-activation potentiation protocol demonstrated superior performance outcomes compared to both specific warm-up and 100% post-activation potentiation protocols. Evening performances were generally better than morning performances for both agility and countermovement jump; however, the protocols/time interaction was not statistically significant (p > 0.05). Conclusions: The 80% post-activation potentiation protocol was most effective for enhancing both agility and vertical jump performance in judokas, with superior results observed during evening sessions. These findings provide valuable insights for optimizing warm-up strategies in judo competition, suggesting that lower-intensity post-activation potentiation protocols might be more beneficial than maximal loading, particularly during evening competitions. Full article

Background: Understanding the balance between intensity and volume during training and competition is crucial for optimizing players’ performance and recovery in professional soccer. While worst-case scenarios (WCSs) are commonly used to assess peak match demands, little is known about how the time spent within WCS thresholds varies across congested and non-congested periods, especially when considering differences in playing time. This study examines the time spent at different percentages of WCSs during congested and non-congested periods for players with lower and higher playing times throughout training sessions and matches. Methods: Data were collected from a professional soccer team across a congested and non-congested match period. Twenty players were divided into two groups based on playing time: the top 10 playing times (PT 1–10) and the bottom 10 playing times (PT 11–20). WCS thresholds for total distance (TD) and the distance covered above 20 km·h⁻¹ (D20) were quantified in 10% increments, starting from 50% and increasing up to >100%. The time spent at each threshold was compared between periods and groups for the integrated soccer exercises performed during all training sessions. Repeated measures of ANOVA were used to analyze differences between playing time groups and periods. Results: During training, players spent significantly more time within the 50–90% WCS TD and WCS D20 thresholds during non-congested periods compared to congested periods (p < 0.05). However, no significant differences were observed in the time spent for >90% of the WCSs between periods (p > 0.05). Both PT 1–10 and PT 11–20 groups exhibited similar patterns of WCS achievement, with small effect sizes observed for a few indicators. Conclusion: Coaches should design training sessions that replicate or exceed match demands, particularly during non-congested periods. Future strategies should integrate larger-sided games with longer durations and dissociated contents to better individualize and optimize training loads, especially for non-starters. Full article

Non-Intrusive Load Monitoring (NILM) includes a set of methods orientated to disaggregating the power consumption of a household per appliance. It is commonly based on a single metering point, typically a smart meter at the entry of the electrical grid of the building, where signals of interest, such as voltage or current, can be measured and analyzed in order to disaggregate and identify which appliance is turned on/off at any time. Although this information is key for further applications linked to energy efficiency and management, it may also be applied to social and health contexts. Since the activation of the appliances in a household is related to certain daily activities carried out by the corresponding tenants, NILM techniques are also interesting in the design of remote monitoring systems that can enhance the development of novel feasible healthcare models. Therefore, these techniques may foster the independent living of elderly and/or cognitively impaired people in their own homes, while relatives and caregivers may have access to additional information about a person’s routines. In this context, this work describes an intelligent solution based on deep neural networks, which is able to identify the daily activities carried out in a household, starting from the disaggregated consumption per appliance provided by a commercial smart meter. With the daily activities identified, the usage patterns of the appliances and the corresponding behaviour can be monitored in the long term after a training period. In this way, every new day may be assessed statistically, thus providing a score about how similar this day is to the routines learned during the training interval. The proposal has been experimentally validated by means of two commercially available smart monitors installed in real houses where tenants followed their daily routines, as well as by using the well-known database UK-DALE. Full article

This study investigated the impact coefficient of a large-span steel truss arch railroad bridge under moving train loads, with the Nanning Three Banks Yongjiang Special Bridge serving as the case study. Field tests were conducted to measure the bridge’s self-vibration characteristics, dynamic deflection, and strain. A coupled vehicle–bridge vibration model was developed using the finite element software ABAQUS 2022 for the bridge and multi-body dynamics software SIMPACK 2022 for the CRH2 train. The two models were integrated to simulate the dynamic interaction between the train and bridge under different speeds and single-/double-track operations. The results demonstrate that the joint simulation of SIMPACK and ABAQUS was an effective method for the vehicle–bridge coupled vibration analysis. The key findings include the following: the deflection and stress impact coefficients increased with the train speed, where the main span exhibited larger deflection coefficients than the side span. The stress impact coefficients varied significantly across different bridge components, where the lower chord of the side span and the ties of the main span showed the highest values. While there was no substantial difference in the deflection impact coefficients between the single- and double-track operations, the stress impact coefficients showed deviations, particularly in the side span’s lower chord and ties, highlighting their sensitivity to vehicle-induced deflection. This study concluded that the bridge’s deflection impact coefficient met design specifications, but the stress impact coefficient exceeded the specified values, suggesting that stress amplification should be carefully considered in the design of similar bridges to ensure operational safety. Full article

Severe damage to cement asphalt mortar (CA mortar) can compromise the stability and safety of high-speed railway operations due to various complex factors during service. The loads from high-speed trains and temperature gradients within the ballastless track structure are significant contributors to this damage. However, most previous studies have focused on laboratory tests or numerical simulations under simple loading conditions, while few have investigated the damage evolution of CA mortar when both train loads and temperature gradients are considered simultaneously. In this paper, a finite element model of the CRTS II ballast track and a high-speed railway train dynamics model based on the damage constitutive model of CA mortar was established. The damage evolution of CA mortar through long-term cyclic numerical simulations under the combined effects of train load and temperature gradient load were investigated. By integrating the maintenance criteria for high-speed railways, the lifetime of CA mortar using the criteria of crack length and off-seam width was predicted. In addition, the material and structural properties of CA mortar were also optimized, considering the relationship between its elastic modulus and density, to enhance its lifetime. The conclusions reached are more realistic. The results indicate that the combined load causes deformation in the ballast track structure, leading to gradual damage progression from the edge to the interior of the CA mortar layer. The lifetime of CA mortar is determined by the number of days it takes for the crack length to reach the maintenance criteria. The lifetime of CA mortar under different temperature gradients ranges from 1 to 2 years. Increasing the elastic modulus and thickness of the CA mortar layer improves its lifespan. An elastic modulus of 9000 MPa and a thickness of 50 mm for the CA mortar were recommended. Full article

Purpose: Footwork speed is closely linked to explosive strength, and both percentage- (PBRT) and velocity-based resistance training (VBRT) are popular methods for developing muscle strength. This study aimed to compare the effects of PBRT and VBRT on lower limb explosive power and footwork movement speed in elite university badminton players over a 6-week training period. Methods: A total of 20 elite badminton players (12 males, 8 females) from Tongji University were randomly divided into VBRT (n = 10) and PBRT groups (n = 10). The VBRT group trained with loads determined by target speed and velocity loss, while the PBRT participants used fixed loads based on a percentage of their one-repetition maximum (1RM). Both the groups performed free-weight back squats with relative loads ranging from 65% to 95% of 1RM over 6 weeks. The pre- and post-training measurements included back squat 1RM; countermovement (CMJ), squat (SJ), and standing long jumps (SLJs); self-weighted squat jump speed (SJS); left and right touch line (LRF), full-field four-point (FF), and front and back touch net footwork (FBF). Results: (1) The baseline measurements showed no significant differences between the groups (p > 0.05). (2) Post-training, both VBRT and PBRT improved the participants’ lower limb explosive power and footwork movement (p < 0.05). (3) The VBRT group demonstrated significantly greater improvements than the PBRT group in all the measures (p < 0.05). Conclusions: VBRT was superior to PBRT in boosting lower limb explosive power and footwork speed in badminton players over 6 weeks, leading to more significant strength–related and neural adaptations. Full article

The mountainous power grids exhibit significantly lower reliability compared to conventional urban grids due to inherent structural weaknesses, dispersed load distribution, and higher failure probabilities of power supply equipment. With the ongoing construction and commissioning of electrified railways in western China, it is crucial to analyze the impact of strong shocks and random fluctuations in traction loads on mountainous power grids, and to study the reliability enhancement of these grids considering electrified railways access, to ensure their safety and the reliable, continuous power supply for the railways. Therefore, this paper proposes a method to enhance the reliability of mountainous power grids considering electrified railways access. First, stochastic fluctuation characteristics of traction loads are simulated through train traction calculations. Subsequently, the reliability level of mountainous grids is quantitatively evaluated, with a novel line vulnerability index established to identify weak grid sections. Finally, two complementary enhancement strategies are proposed: dynamic line capacity expansion and optimized backup capacity allocation. Case studies demonstrate the effectiveness of the method through comparative analysis of reliability indices before and after implementation, confirming both technical validity and practical feasibility. Full article

Structural health monitoring (SHM) is crucial for ensuring the safety and longevity of military training aircraft, which face demanding conditions such as high maneuverability, variable loads, and extreme environments, leading to structural fatigue. Traditional methods, such as modal analysis, often struggle to handle the multivariate complexity of operational conditions and data variability. Recently, deep learning has emerged as a promising alternative to overcome these limitations. However, deep learning models typically operate in a unidirectional manner, where feedback to the inputs is often neglected. In contrast, biological neurons utilize feedback mechanisms to refine and adapt their responses in natural ecosystems, enabling adaptive learning and error correction. In this context, this study proposes an innovative Convolutional Neural Network with Reversed Mapping (CNN-RM) approach to SHM, which incorporates feedback loops and self-correcting mechanisms. Before feeding the data into CNN-RM, the dataset complexity is reduced through time-series-to-images Continuous Wavelet Transform (CWT), followed by a denoising CNN (DnCNN) to mitigate complex behavior under various conditions. For application, this study utilizes a massive dataset collected from multivariate sensors installed on a decommissioned military training aircraft previously used by the British Royal Air Force and now housed in a laboratory environment. The results revealed that the overall mean of classification metrics for the CNN is 0.9673 (training) and 0.9422 (testing), while for CNN-MR, it is 0.9764 (training) and 0.9515 (testing), showing an improvement of 0.94% in training and 1.00% in testing. These results highlight significant advancements in SHM, recommending the consideration of such learning mechanisms in neural learning models. Full article

In recent years, the application of artificial neural network models has become increasingly widespread in the automotive industry; however, the sensitivity analysis of these models is often neglected. This shortfall poses significant risks in safety-critical applications, where the reliability of models under varying conditions is of critical importance. This study focuses on the sensitivity analysis of a long short-term memory neural network model, previously published by us, designed to predict the future yaw rates of vehicles. Our research aimed to determine the minimum amount of data required for effective model training and to conduct a comprehensive sensitivity analysis, examining the performance and applicability of the trained model under varying tire pressures, different passenger loads, and different passenger configurations. Additionally, we investigated whether the trained model could be applied to other vehicle types. Our results indicated that the vehicle weight distribution was the most influential factor affecting the accuracy of the model. Nonetheless, the model’s predictive error remained consistently within the safety thresholds defined by the standards under all tested conditions. Our experiments and analyses were performed using over 7.5 h of data collected under real-world conditions, which will be freely available to the research community. Full article

Classification of neurocognitive states from Electroencephalography (EEG) data is complex due to inherent challenges such as noise, non-stationarity, non-linearity, and the high-dimensional and sparse nature of connectivity patterns. Graph-theoretical approaches provide a powerful framework for analysing the latent state dynamics using connectivity measures across spatio-temporal-spectral dimensions. This study applies the graph Koopman embedding kernels (GKKE) method to extract latent neuro-markers of seizures from epileptiform EEG activity. EEG-derived graphs were constructed using correlation and mean phase locking value (mPLV), with adjacency matrices generated via threshold-binarised connectivity. Graph kernels, including Random Walk, Weisfeiler–Lehman (WL), and spectral-decomposition (SD) kernels, were evaluated for latent space feature extraction by approximating Koopman spectral decomposition. The potential of graph Koopman embeddings in identifying latent metastable connectivity structures has been demonstrated with empirical analyses. The robustness of these features was evaluated using classifiers such as Decision Trees, Support Vector Machine (SVM), and Random Forest, on Epilepsy-EEG from the Children’s Hospital Boston’s (CHB)-MIT dataset and cognitive-load-EEG datasets from online repositories. The classification workflow combining mPLV connectivity measure, WL graph Koopman kernel, and Decision Tree (DT) outperformed the alternative combinations, particularly considering the accuracy (91.7%) and F1-score (88.9%), The comparative investigation presented in results section convinces that employing cost-sensitive learning improved the F1-score for the mPLV-WL-DT workflow to 91% compared to 88.9% without cost-sensitive learning. This work advances EEG-based neuro-marker estimation, facilitating reliable assistive tools for prognosis and cognitive training protocols. Full article

Soft robots with compliant bodies offer safe human–robot interaction as well as adaptability to unstructured dynamic environments. However, the nonlinear dynamics of a soft robot with infinite motion freedom pose various challenges to operation and control engineering. This research explores the motion of a pneumatic soft robot under diverse loading conditions by conducting finite element analysis (FEA) and using machine learning. The pneumatic soft robot consists of two parallel hyper-elastic tubular chambers that convert pneumatic pressure inputs into soft robot motion to mimic an elephant trunk and its motion. The body of each pneumatic chamber consists of a series of bellows to effectively facilitate the expansion, contraction, and bending of the body. The first chamber spans the entire length of the soft robot’s body, and the second chamber spans half of it. This unique asymmetric design enables the soft robot to bend and curl in various ways. Machine learning is used to establish a forward kinematic relationship between the pressure inputs and the motion responses of the soft robot using data from FEA. Accordingly, this research employs an artificial neural network that is trained on FEA data to estimate the reachable workspace of the soft robot for given pressure inputs. The trained neural network demonstrates promising estimation accuracy with an R-squared value of 0.99 and a root mean square error of 0.783. The workspaces of asymmetric double-chamber and single-chamber soft robots were compared, revealing that the double-chamber robot offers approximately 185 times more reachable workspace than the single-chamber soft robot. Full article

The explosive growth of power load data has led to a substantial presence of abnormal data, which significantly reduce the accuracy of power system operation planning, load forecasting, and energy usage analysis. To address this issue, a novel improved GAN–Transformer model is proposed, leveraging the adversarial structure of the generator and discriminator in Generative Adversarial Networks (GANs). To provide the model with a suitable feature dataset, One-hot encoding is introduced to label different categories of abnormal power load data, enabling staged mapping and training of the model with the labeled dataset. Experimental results demonstrate that the proposed model accurately identifies and classifies mutation anomalies, sustained extreme anomalies, spike anomalies, and transient extreme anomalies. Furthermore, it outperforms traditional methods such as LSTM-NDT, Transformer, OmniAnomaly and MAD-GAN in Overall Accuracy, Average Accuracy, and Kappa coefficient, thereby validating the effectiveness and superiority of the proposed anomaly detection and classification method. Full article

Background/objectives: The study aims to investigate potential differences in vertical jump performance between elite basketball and volleyball players before and after a standard training session, in comparison to a control group from the general population. The analysis focuses on the influence of selected gene polymorphisms that may contribute to variations in the assessed performance parameters. Aims: The aim was to investigate the influence of ACE (rs4646994), ACTN3 (rs1815739), PPARA rs4253778, HIF1A (rs11549465), and AMPD1 (rs17602729) genes polymorphisms on the combined effects of post-activation potentiation (PAP), post-activation performance enhancement (PAPE), and general adaptation syndrome (GAS), as reflected in vertical jump performance, in elite basketball and volleyball players compared to a control group from the general population. Methods: The effects of PAP at the beginning of the training load (acute exercise), and the combined influences of PAPE and GAS following the training load were evaluated using parameters measured by the OptoJump Next^® system (Microgate, Bolzano, Italy). Results: A statistically significant (h, p < 0.05) negative effect of the CT genotype of the AMPD1 gene on jump height was observed in the group of athletes. The CT genotype of the AMPD1 gene negatively impacted on PAPE and GAS adaptive responses (ΔP, Δh, p < 0.001) also in the control group. A positive effect on the power during the active phase of the vertical jump was identified for the II genotype of the ACE gene and the Pro/Ser genotype of the HIF1A gene, both exclusively in the control group (ΔP, p < 0.05). Conclusion: Our findings demonstrate that different gene polymorphisms exert variable influences on the combined effects of PAPE and GAS, as reflected in vertical jump parameters, depending on the participants’ level of training adaptation. Full article