Search Results (8,462)

Predictive maintenance based on vibration monitoring can significantly improve gearbox reliability in heavy-industry environments. Although it is well established in vibration engineering that operating regimes influence vibration levels, the contribution of this work lies in providing an integrated, data-driven diagnostic linkage between continuously acquired IoT vibration indicators and key process/operational variables to identify and quantify the dominant drivers of vibration escalation. This study deployed wireless IoT sensors for continuous acquisition of RMS vibration and lubrication temperature in gearboxes operating in cement and mining plants and applied multivariate machine learning models to detect anomalies and identify root causes. We compared boosted multilayer feedforward neural networks, boosted trees, and k-nearest neighbors using RMS vibration and process variables including mill feed, lubrication pressures, and temperature. The boosted neural network delivered superior validation performance and isolated low or near-zero mill feed during operation as the primary driver of elevated RMS vibration, with lubrication instability acting as a secondary interacting factor. This shifts the diagnosis from a generic “high vibration during transients” statement to actionable operational mitigations—minimum feed set-points, controlled ramping logic, and lubrication pressure governance—supported by multivariate evidence. Our results motivate further validation with k-fold and out-of-time tests. Full article

A dedicated experimental rig is presented for a half-journal bearing operating under highly loaded, well-controlled hydrodynamic lubrication conditions relevant to turbomachinery. The apparatus combines pressure measurements in the film, distributed temperature measurements in the shaft and bush, and ultrasonic film-thickness measurements that map the circumferential film-thickness profile across the lubrication region. Experiments are reported for normal loads of 5–20 kN and shaft speeds of 1000–4000 rpm with controlled oil supply conditions. The measured pressure and temperature trends are consistent with established hydrodynamic lubrication behaviour. The film thickness measurements confirm full-film operation across the tested operating envelope, while indicating increased uncertainty in regions affected by cavitation. A correlation for the temperature rise due to viscous heating is proposed as a compact representation of the data. The rig design and accompanying measurements provide a benchmark-quality data set intended for validation and development of thermal elasto-hydrodynamic lubrication (TEHL)/computational fluid dynamics (CFD) models under high load and speed conditions. Full article

Flavor-relevant trace volatiles and microbial communities were examined in six sauce-aroma high-temperature Daqu samples. Headspace solid-phase microextraction coupled with gas chromatography–mass spectrometry (HS-SPME-GC-MS) quantified 210 trace volatile compounds across 14 chemical classes. Orthogonal partial least squares discriminant analysis (OPLS-DA) with variable importance in projection (VIP) screening was integrated with sensory scoring, correlation analysis, and molecular docking to an olfactory receptor model. Volatile profiles showed clear stratification in total abundance. Pyrazines dominated the high-total group. Tetramethylpyrazine served as a major driver. Sensory evaluation indicated that aroma explained overall quality best. (E)-2-pentenal and dimethyl trisulfide showed significant positive associations with aroma and overall scores. In the olfactory receptor, the polar residue module that provides directional constraints for Daqu odor activation was formed by Ser75, Ser92, Ser152, Ser258, Thr74, Thr76, Thr98, Thr200, Gln99, and Glu94. The hydrogen-bond or charge network was further reinforced by Arg150, Arg262, Asn194, His180, His261, Asp182, and Gln181. The core discriminant set comprised acetic acid, hexanoic acid, (E)-2-pentenal, nonanal, decanal, dimethyl trisulfide, trans-3-methyl-2-n-propylthiophane, 2-hexanone oxime, ethyl linoleate, propylene glycol, 2-ethenyl-6-methylpyrazine, 4-methylquinazoline, 5-methyl-2-phenyl-2-hexenal, and 1,2,3,4-tetramethoxybenzene. Sequencing revealed higher bacterial diversity than fungal. Bacillus and Kroppenstedtia were dominant bacterial genera. Aspergillus, Paecilomyces, Monascus, and Penicillium were major fungal genera. Correlation patterns suggested that Bacillus and Monascus were positively linked to acetic acid and 1,2,3,4-tetramethoxybenzene. Together, these results connected chemical fingerprints, sensory performance, receptor-level plausibility, and microbial ecology. Concrete targets are provided for quality control of high-temperature Daqu. Full article

Wind energy plays a critical role in the European Union’s decarbonization strategy, including Romania’s growing renewable energy capacity. This study proposes a deep learning-based method for forecasting hourly wind energy production in Romania using feedforward neural networks (FFNNs) and recurrent neural networks (RNNs), trained on a dataset spanning from 1 January to 31 December 2023. The dataset includes hourly wind energy output data (mean = 850.6 MW, std = 694.0 MW) and 13 meteorological variables (e.g., average wind speed = 4.7 km/h, temperature = 14.4 °C). A total of 1296 models were trained and evaluated, with the best-performing RNN model achieving a coefficient of determination of R² = 0.9680 and a mean absolute error (MAE) of 81.03 MW. The top three models all exceeded R² = 0.966, demonstrating strong generalization on unseen data. The models were also validated using two external time intervals outside the training/testing sets, confirming robustness. These results show that deep learning models can provide highly accurate, data-driven predictions of wind energy output, supporting grid stability and informed decision-making amid renewable energy variability. Full article

The induction of triploidy, a strategy to mitigate unwanted pre-harvest sexual maturation and a genetic containment measure for escaped farmed Atlantic salmon (Salmo salar), may give rise to challenges because of the distinct environmental and dietary requirements of sterile triploid fish. Smoltification is a critical phase in the life cycle of Atlantic salmon, so knowledge about parr–smolt transformation in triploids is important for the salmon farming industry. This study covered an investigation of hepatic expression of growth hormone receptor (GHrec) and insulin-like growth factor-I (IGF-I) genes, both of which are intimately involved in the regulation of osmoregulation and growth. Additionally, hepatic presence and location of IGF-I mRNA were examined using RNAscope^®, an advanced in situ hybridization technique. Triplicate groups of juvenile diploid and triploid salmon were reared at low temperature (10 °C) and fed either a standard diet or one enriched with hydrolyzed fish proteins from the start of feeding onwards. Liver samples were collected from three fish per tank each month from October to December (2454–3044 degree-days post-start feeding), the period encompassing smoltification, and hepatic expression of IGF-I and GHrec genes was quantified by real-time PCR. The results indicated that neither ploidy nor diet significantly influenced IGF-I or GHrec gene expression, suggesting that, under our conditions, triploidy and diet did not adversely affect this molecular pathway linked to growth and osmoregulation. IGF-I gene expression exhibited significant temporal variation, correlating with the progression of smoltification, while GHrec gene expression showed a similar, albeit non-significant, trend. Triploids exhibited IGF-I and GHrec gene expression patterns comparable to diploids, and both the temporal changes and lack of difference between triploids and diploids were mirrored in the quantification of IGF-I mRNA within the liver cells. The potential applicability to a commercial aquaculture setting requires further investigation. Full article

Artificial neural networks are reliable machine learning models that have been applied to a multitude of practical and scientific applications in recent decades. Among these applications, there are examples from the areas of physics, chemistry, medicine, etc. To effectively apply them to these problems, it is necessary to adapt their parameters using optimization techniques. However, in order to be effective, optimization techniques must know the range of values for the parameters of the artificial neural network, so that they can adequately train the artificial neural network. In most cases, this is not possible, as these ranges are also significantly affected by the inputs to the artificial neural network from the objective problem it is called upon to solve. This situation usually results in artificial neural networks becoming trapped in local minima of the error function or, even worse, in the phenomenon of overfitting, where although the training error achieves low values, the artificial neural network exhibits low performance in the corresponding test set. To address this limitation, this work proposes a novel two-stage training approach in which a simulated annealing (SA)-based preprocessing stage is employed to automatically identify optimal parameter value intervals before the application of any optimization method to train the neural network. Unlike similar approaches that rely on fixed or heuristically selected parameter bounds, the proposed preprocessing technique explores the parameter space probabilistically, guided by a temperature-controlled acceptance mechanism that balances global exploration and local refinement. The proposed method has been successfully applied to a wide range of classification and regression problems and comparative results are presented in detail in the present work. Full article

The blood-brain barrier (BBB) is one of the most selective physiological interfaces in the human body. Transendothelial electrical resistance (TEER) has become a widely adopted quantitative metric for assessing its in vitro structural and functional integrity. Although TEER measurements are routinely incorporated into BBB-on-chips, the absence of harmonized electrode architectures, measurement settings, and reporting standards continues to undermine reproducibility and translational reliability among laboratories. This systematic review provides the first comprehensive classification and critical comparison of electrode configurations used for TEER assessment, specifically within BBB-on-chip systems. Eligible studies were analyzed and categorized according to electrode design, fabrication method, integration strategy, and operational constraints. We critically evaluated six principal electrode architectures, highlighting their performance trade-offs in terms of uniformity of current distribution, long-term stability, scalability, and compatibility with dynamic shear conditions. Furthermore, we propose a bioinspired TEER reporting framework that consolidates essential metadata, including electrode specification, temperature control, viscosity effects, and blank resistance correction. Our analysis proposes screen-printed and hybrid silver-indium tin oxide (ITO) electrodes as promising candidates for next-generation BBB platforms. Moreover, our review provides a structured roadmap for standardizing TEER electrode design and reporting practices to facilitate interlaboratory consistency and accelerate the adoption of BBB-on-chip systems as truly biomimetic platforms for predictive neuropharmacological workflows. Full article

High-precision phase difference measurement based on field-programmable gate arrays (FPGA) has important application requirements in fields such as high-stability time-frequency transmission, signal synchronization, and precision testing. Addressing the limitations of traditional methods in terms of temperature stability and measurement accuracy, this paper proposes two high-precision phase difference measurement schemes based on the FPGA platform. An eight-parallel-multi-carry chain time-to-digital converter (TDC) and digital dual-mixer time difference (DDMTD) measurement modules are constructed to perform high-precision phase difference measurements on the phase-shifted output signal of the MMCM dynamic phase-shifted module. Results show that at room temperature (25 °C), the single-carry chain TDC exhibits better measurement accuracy than the DDMTD, and the single-carry chain TDC’s measurement error range of 4.7–6.0 ps is superior to the DDMTD’s 20–75 ps error range. Under different temperature conditions, the eight-parallel-multi-carry chain TDC consistently demonstrates superior measurement accuracy, resolution, and temperature stability compared to the single-carry chain TDC. In terms of measurement accuracy, under room temperature conditions, in three sets of phase difference tests (178.5714 ps, 357.1428 ps, and 535.7142 ps), the measurement error of the eight-parallel-multi-carry chain TDC was controlled within 4.6 ps, which is better than the 4.7–6.0 ps error range of the single-carry chain TDC. Average resolution: The average resolution of the single-carry chain TDC was 6.329 ps, while the average resolution of the eight-parallel-multi-carry chain TDC improved to 0.833 ps. Temperature stability: Within the temperature range of 10 °C to 100 °C, the temperature coefficient of the single-carry chain TDC was 0.002127 ps/°C, while the temperature coefficient of the eight-parallel-multi-carry chain TDC decreased to 0.000564 ps/°C. This paper also summarizes the advantages and limitations of the above methods in terms of implementation complexity and robustness, providing a reference for the optimized design of high-precision phase difference measurement technology for FPGA platforms. Full article

Data loss caused by sensor malfunctions in bridge Structural Health Monitoring (SHM) systems poses a critical risk to structural safety assessment. Although deep learning has advanced data imputation, standard “black-box” models often fail to capture the underlying deterioration mechanisms governed by physical laws. To address this limitation, we propose SP-VMD-CNN-GRU, a prior-knowledge-guided framework that integrates environmental thermal mechanisms with deep representation learning for bridge crack data imputation. Deviating from empirical parameter selection, we utilize the Granger causality test to statistically validate temperature as the primary driver of crack evolution. Leveraging this prior knowledge, we introduce a Shared Periodic Variational Mode Decomposition (SP-VMD) method to isolate temperature-dominated annual and daily periodic components from noise. These physically validated components serve as inputs to a hybrid CNN-GRU network, designed to simultaneously capture spatial correlations across sensor arrays and long-term temporal dependencies. Validated on real-world monitoring data from the Luo’an River Grand Bridge, our framework achieves the highest coefficient of determination ($R^2$) of 0.9916 and the lowest Mean Absolute Percentage Error (MAPE) of 12.95%. Furthermore, statistical validation via Diebold–Mariano and Model Confidence Set tests proves that our physics-guided approach significantly surpasses standard baselines (TCN, LSTM), demonstrating the critical value of integrating prior knowledge into data-driven SHM. Full article

As human space exploration advances towards establishing sustainable Martian habitats, achieving autonomous food production is a critical requirement. The potato (Solanum tuberosum L.), with its notable environmental resilience and nutritional efficiency, is a prime candidate crop. This study develops a conceptual framework for Martian potato cultivation by systematically analyzing the profound disparities between Martian conditions and plant physiology. We identify and evaluate seven fundamental challenges: atmospheric composition, extreme temperatures, water scarcity, soil properties, nutrient deficiencies, absent microbiota, and radiation/gravity effects. To address these challenges, we propose a phased, testable roadmap comprising four stages: (I) screening and bio-engineering of multi-stress-tolerant potato genotypes; (II) phased domestication via Earth-based analog experiments to define adaptability thresholds; (III) deployment of a controlled cultivation module within a Martian habitat, integrating targeted technological interventions; and (IV) conceptual exploration of extra-habitat agricultural potential. The primary contribution of this work is a structured set of hypotheses and key performance indicators for each stage, translating visionary goals into a defined research agenda to guide future empirical work in extraterrestrial agronomy. Full article

Increased global energy consumption contributes to higher operational costs in the energy sector and results in environmental deterioration. This study evaluates the effectiveness of integrating Internet of Things (IoT) sensors and machine learning techniques to predict adaptive thermostat setpoints to support behavior-aware Heating, Ventilation, and Air Conditioning (HVAC) operation in residential buildings. The dataset was collected over two years from 2080 IoT devices installed in 370 zones in two buildings in Halifax, Canada. Specific categories of real-time information, including indoor and outdoor temperature, humidity, thermostat setpoints, and window/door status, shaped the dataset of the study. Data preprocessing included retrieving data from the MySQL database and converting the data into an analytical format suitable for visualization and processing. In the machine learning phase, deep learning (DL) was employed to predict adaptive threshold settings (“from” and “to”) for the thermostats, and a gradient boosted trees (GBT) approach was used to predict heating and cooling thresholds. Standard metrics (RMSE, MAE, and R²) were used to evaluate effective prediction for adaptive thermostat setpoints. A comparative analysis between GBT ”from” and “to” models and the deep learning (DL) model was performed to assess the accuracy of prediction. Deep learning achieved the highest performance, reducing the MAE value by about 9% in comparison to the strongest GBT model (1.12 vs. 1.23) and reaching an R² value of up to 0.60, indicating improved predictive accuracy under real-world building conditions. The results indicate that IoT-driven setpoint prediction provides a practical foundation for behavior-aware thermostat modeling and future adaptive HVAC control strategies in smart buildings. This study focuses on setpoint prediction under real operational conditions and does not evaluate automated HVAC control or assess actual energy savings. Full article

Olive cultivation in warm regions is likely to be vulnerable to the expected temperature increases associated with climate change. The objectives of this study were to evaluate the effects of experimental warming at the end of late winter and spring on the timing of phenological stages, the duration of developmental periods, plant growth, and biomass production. The experiment was conducted in control (T0) and warmed (+4 °C, T+) open-top chambers (OTCs) during 2018 and 2019 using two olive cultivars (‘Arbequina’, ‘Coratina’) in northwest Argentina. Warming generally led to statistically significant earlier inflorescence development, flowering, fruit set, and pit hardening, with the responses being more pronounced as the spring progressed. Earlier vegetative bud opening occurred due to warming in 2018, but not in 2019. Additionally, no differences in shoot elongation or aboveground biomass were observed due to warming at the end of spring in either 2018 or 2019. Fruit set was reduced by warming, particularly in ‘Coratina’. Overall, the experimental results show that reproductive development is more sensitive to warming than vegetative growth in young olive trees. Further studies should be conducted in a larger number of olive cultivars and regions to improve our ability to predict responses to global warming. Full article

The aging and failure of transformer bushing seals under multi-factor effects are significant causes of oil leakage incidents. However, their failure mechanisms under combined environmental stressors remain inadequately understood. This study presents a comprehensive investigation into the aging behavior and failure mechanisms of nitrile rubber (NBR) and fluoroelastomer (FKM) sealing materials subjected to single and multi-factor aging conditions, including thermo-oxidative, hygrothermal, hygrothermal–compression, and hygrothermal–compression–salt environments. NBR undergoes severe degradation under multi-factors, dominated by additive loss and molecular chain crosslinking. At high temperatures, large-scale molecular chain scission occurs, along with increased compression set, microscopic morphological damage, and filler precipitation. In contrast, FKM exhibits excellent stability thanks to its C-F main chain. Stress synergy significantly accelerates the failure of both materials. These findings highlight the need for multivariate analysis to support reliable condition assessment and lifetime prediction and to inform sealing material selection and proactive grid maintenance. Full article

Background: The piezoelectric saw is a technology used in osteotomies, providing precise and minimally invasive cuts, especially in areas close to vital structures. Despite its advantages, limitations such as prolonged surgical time and restrictions in use for larger bones have motivated the development of ultrasonic tips with more efficient geometries. Methods: A laboratory trial was conducted with 40 ultrasonic tips (n = 40), divided into 2 groups: the test group (n = 20), with an ultrasonic tip featuring a multidirectional angular cutting-tooth geometry, and the control (n = 20), with a straight-tooth ultrasonic tip. Two operators performed standardized osteotomies on synthetic bone blocks, with monitoring of variables including cutting time (in seconds), maximum block and blade temperature (in °C), and bone mass loss (in grams). Sample randomization was block-based, and blade coding ensured operator and evaluator blinding. Results: The results showed a statistically significant reduction of approximately 26% in cutting time with the multidirectional ultrasonic tips (Test = 52.85 s; Control = 71.55 s; p < 0.001), regardless of the operator. No significant differences were observed between groups regarding maximum bone temperature (Test = 30.45 °C; Control = 29.40 °C; p = 0.337), blade temperature variation (Test = 5.30 °C; Control = 4.10 °C; p = 0.337), overall temperature variation (Test = −0.19 °C; Control = 0.06 °C; p = 0.285), or bone mass loss (Test = 0.1355 g; Control = 0.0350 g; p = 0.387). A significant interaction between operator and blade type in some variables, such as bone temperature variation (p = 0.001), reinforces the influence of technical experience on the results. Conclusions: The multidirectional angular geometry of the ultrasonic tip significantly improves cutting efficiency without compromising thermal safety, representing a promising advancement for optimizing osteotomies in surgical settings. The use of this new geometry may enhance productivity, particularly in complex procedures, and deserves future clinical investigation to expand its applicability across different surgical specialties, including orthopedics. Full article

It is now crucial to accurately monitor the state of health (SoH) of batteries in a setting where the use of electric vehicles (EVs) and renewable energy technologies is still growing. To solve this issue and evaluate the SoH, this paper makes use of deep learning technology. The suggested method incorporates voltage, current, and temperature data, which are important indications of the SoH and can potentially be obtained directly from the battery management system (BMS). Although deep neural networks (DNNs) have previously been employed for SoH estimation, our study distinguishes itself by implementing a robust, completely configurable DNN application in MATLAB/Simulink R2019a. This design enables the adjustment of activation functions, layer depth, and neuron count to adapt to different battery aging conditions. To achieve optimal performance, numerous configurations were examined, highlighting the relevance of hyperparameter setting. Our technique avoids traditional feature engineering while providing a practical, adaptive, and accurate SoH estimate framework appropriate for real-world integration. The precision of the improved model was then verified against a Li-ion battery dataset with various discharge profiles given by the national aeronautics and space administration (NASA). The collected findings revealed that the proposed method is more accurate and robust than other regularly used models. The DNN model achieved a Mean absolute error (MAE) of 1.433% and a Coefficient of determination of 0.99998, outperforming previous methods such as CNN-BiGRU, which reported an MAE of 2.448% in a recent publication. This study demonstrates the reliable performance of the DNN in predicting the SoH of Li-ion cells. Full article