Search Results (11,531)

Rolling bearings are critical components in rotating machinery, and their failures may lead to significant economic losses and safety hazards. However, early fault signals are often weak and masked by strong background noise, making accurate fault diagnosis extremely challenging. To address this issue, this paper proposes a fault diagnosis method for rolling bearings based on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN), an autoregressive (AR) model, and observer-based eigenvalue extraction, combined with a particle swarm optimization-based kernel extreme learning machine (PSO-KELM). Targeting rotating machinery with rolling bearings, the approach begins by applying ICEEMDAN as a preprocessing step to decompose non-stationary vibration signals into multiple intrinsic mode functions (IMFs), from which all essential fault-related information is extracted. The preprocessed vibration signal is then reconstructed. Subsequently, an AR model is used to establish a state-space representation for the observer, which processes the reconstructed signal and generates a residual output by comparing it with the actual mechanical signal. Features are then extracted from the residual signal, including its mean, variance, maximum and minimum values, kurtosis, waveform factor, pulse factor, and clearance factor. These features serve as inputs to the PSO-KELM classifier for fault diagnosis. To validate the method, real vibration data from electric motor bearings were employed in a case study, covering normal conditions and three typical fault types: outer race fault, inner race fault, and rolling element fault. The results demonstrate that the proposed method effectively enables fault feature extraction and accurate identification of bearing conditions. Full article

Considering the synergistic optimization design of battery thermal safety and system economy in extreme environments, a hybrid lithium-ion battery thermal management system (BTMS) employing composite phase change material (CPCM) with liquid cooling is proposed by comparing four BTMSs of pure air cooling, pure CPCM, pure liquid cooling, and the hybrid cooling using CPCM and liquid cooling. The proposed hybrid cooling system demonstrates the capability to maintain the maximum battery temperature at 45.27 °C under extreme operating conditions, including elevated ambient temperatures of 40 °C combined with 5C discharge rate. Notably, this thermal regulation performance is achieved without requiring additional power input, highlighting the energy-efficient design of the system. Further, to address the critical challenge of thermal runaway prevention under summer extreme temperature up to 50 °C, an artificial neural network (ANN) model is established for the hybrid cooling, integrated with the non-dominated sorting genetic algorithm II (NSGA-II) algorithm, leading to the maximum temperature controlled at 48.68 °C and minimum system power consumption of 158 W, achieving a 12.1% reduction in thermal fluctuation amplitude and a 5.9% reduction in power consumption compared to initial design and optimal solutions, respectively. The proposed BTMS introduces the NSGA-II-ANN model for multi-objective collaborative optimization to solve the contradiction between thermal safety and energy consumption under extreme working conditions, enhancing the safety measures of power batteries and economic viability for electric vehicles. Full article

To support multimedia and deep learning applications running on containers within a server, both processor cores and main memory space are critical resources for performance tuning. With the growing memory demands of applications to maintain intermediate data, installing additional dynamic random-access memory (DRAM) modules increases not only hardware costs but also the static and dynamic energy consumption of a server. In this study, both DRAM and non-volatile memory (NVM) are leveraged to provide short access latency and large main memory capacity for a server running multiple containers with diverse applications. Contention for memory space and processor cores among containers is jointly modeled as part of the performance optimization problem for the hybrid memory system of the server. Our memory and computing resource scheduling algorithms are thus developed to judiciously balance the usage of cores and DRAM space among tasks, while NVM is utilized to increase the degree of parallelism to reduce the Makespan of task batches. Benchmark programs were used to generate the input task set, and experimental results show that our solution outperforms others by achieving at least an 18.34% reduction in Makespan when 100 distinct containerized tasks are executed on a system with 512 gigabytes (GB) of NVM, 32 GB of DRAM, and eight cores. Full article

In the domain of engineering applications, this study addresses the challenge of achieving a unified estimation of the suspension relative velocity and road gradient during vehicle operation while mitigating the significant costs associated with automotive sensors. This paper proposes a simplified system model that integrates discretized vehicle vertical dynamics and longitudinal kinematics, intentionally excluding wheel dynamics. Utilizing the front axle vertical velocity, vehicle speed, and longitudinal and vertical accelerations as inputs, an estimator is employed in conjunction with the Extended Kalman filter algorithm to concurrently predict the relative velocity of the vehicle suspension, the sprung mass velocity, and the road gradient. The feasibility of the proposed methodology is corroborated through simulation experiments. Furthermore, real-world road tests validate the efficacy and timeliness of the joint estimation approach based on a “2 + 1” sensor arrangement. This methodology not only optimizes sensor system configuration and reduces engineering costs but also provides substantial technical support for further advancements in vehicle parameter estimation and suspension control applications. Full article

The chlorophyll content of plant leaves measured by the soil plant analysis development (SPAD) is an important indicator for measuring crop growth status and irrigation effect. The rapid, non-destructive and efficient estimation of crop SPAD values is of great significance to the field management of crops. In this study, the canopy hyperspectral reflectance and SPAD values of winter wheat were obtained, and the spectral curve was changed through four spectral processing methods, including first-order differential (FD), second-order differential (SD), multivariate scattering correction (MSC), and Savitzky–Golay smoothing (SG) to improve the correlation between canopy spectral reflectance and SPAD. Furthermore, to investigate and evaluate the performance of various vegetation indices (VIs) in estimating SPAD values for winter wheat, existing published indices were optimized using random band combinations derived from multiple canopy spectral transformations. The optimized vegetation index was used as the input variable of the model, and six machine learning algorithms, including random forest (RF), long short-term memory network (LSTM), multilayer perceptron (MLP), deep recurrent neural network (Deep-RNN), gated recurrent unit (GRU), and convolutional neural network (CNN), were used to construct the winter wheat SPAD values estimation model, and the model was verified. The experimental results demonstrate that, when utilizing an equivalent number of optimized vegetation indices as input, the GRU-based model achieves higher estimation accuracy compared to other models. Specifically, the coefficient of determination (R²) is improved by 0.12 compared to the RF model, by 0.03 compared to the LSTM model, by 0.12 compared to the MLP model, by 0.02 compared to the Deep-RNN model, and by 0.02 compared to the CNN model. At the same time, the GRU model also has a lower root mean square error (RMSE) and relative error (RE) of 7.37 and 24.90%, respectively. This study provides valuable hyperspectral remote sensing technology support for the implementation of winter wheat SPAD values estimation in the field. Full article

Accurate time series forecasting of sea surface temperature (SST) is essential for understanding the ocean climate system and large-scale ocean circulation, yet it remains challenging due to regime-dependent variability and correlated errors across heterogeneous prediction models. This study addresses these challenges by formulating SST ensemble time series forecasting aggregation as a stochastic, sample-adaptive weighting problem. We propose a diffusion-conditioned ensemble framework in which heterogeneous base forecasters generate out-of-sample SST predictions that are combined through a noise-conditioned weighting network. The proposed framework produces convex, sample-specific mixture weights without requiring iterative reverse-time sampling. The approach is evaluated on short-horizon global SST forecasting using the Global Ocean Data Assimilation System (GODAS) reanalysis as a representative multivariate dataset. Under a controlled experimental protocol with fixed input windows and one-step-ahead prediction, the proposed method is compared against individual deep learning forecasters and conventional global pooling strategies, including uniform averaging and validation-optimized convex weighting. The results show that adaptive, diffusion-weighted aggregation yields consistent improvements in error metrics over the best single-model baseline and static pooling rules, with more pronounced gains in several mid- to high-latitude regimes. These findings indicate that stochastic, condition-dependent weighting provides an effective and computationally practical framework for enhancing the robustness of multivariate time series forecasting, with direct applicability to global SST prediction from large-scale geophysical reanalysis data. Full article

The objective of this work is to propose a simulation strategy for production planning that is compatible with the dynamism of natural gas processing, especially under an open-market arrangement, in which several scheduling simulations must be performed within short time horizons. In such contexts, traditional first-principles-based approaches, although accurate, require prohibitive computational times, motivating the need for an alternative simulation strategy. This work thus proposes a data-driven model built with the aid of machine learning and applied in a case study with historical data from the largest gas processing site in Brazil: Cabiúnas Petrobras asset. Main plant flowrates were selected: 18 targets and 44 input candidates—1282 observations from three and a half years of operation. Principal Component Analysis was used for order reduction, keeping the 22 main principal components. A forward neural network (2 hidden layers and 225 neurons per layer) was built from training/test sets randomly selected and optimized hyperparameters—learning rate (0.001533) and batch size (8). Training converged in roughly 200 epochs (Adam optimizer), with early stop triggered by the validation set. A mean absolute error of 0.0017 (test set) and R² = 0.72 were found, a promising result considering plant complexity and data simplicity. Results showed a particularly good fit for lighter products (sales gas and natural gas liquid), also indicating an opportunity for further work by including inputs related to liquid fractionation. Full article

For arch dam seismic safety evaluation, the finite element equivalent stress method has been widely used, and existing studies have realized mature equivalent stress calculation along the foundation surface path. However, from the scientific research perspective, there is a lack of a full dam surface equivalent stress characterization method for arch dams under seismic action; from the engineering practice perspective, the traditional path method cannot fully reflect the overall stress distribution of the dam, leading to insufficient comprehensive safety evaluation. To accurately assess the impact of seismic action on the overall structural safety of arch dams and address the above limitations, this study develops a methodology for calculating equivalent stress across the entire dam surface of arch dams under seismic action. Taking a concrete arch dam as the research object, a seismic wave input method based on viscoelastic artificial boundaries is employed. Three-dimensional finite element analysis of the arch dam is performed using ABAQUS, integrated with Python-based secondary development to extract stress along the integration path of each arch ring layer and calculate sectional internal forces. The equivalent stress of each arch ring layer integration path is then processed using the material mechanics method to obtain the equivalent stress distribution across the entire dam surface. A comparative analysis is conducted between the equivalent stress on the entire dam surface and that along paths on the foundation surface regarding the seismic dynamic response and behavioral patterns of the dam. The results demonstrate that the full dam surface equivalent stress approach not only accurately captures the extreme tensile and compressive stress values in the downstream foundation area but also identifies stress extrema in the upstream dam crest region, thereby achieving comprehensive characterization of the dam stress field under seismic action and enhancing both the efficiency and accuracy of equivalent stress calculations for arch dams. This method provides more comprehensive and reliable data support for seismic design optimization and reinforcement of arch dams. Compared with the traditional foundation surface path method, the proposed method achieves 100% identification of the whole dam surface stress extremum areas, with a maximum relative error of only 1.62% in the overlapping calculation area. Full article

We present GANimate, a lightweight method for animating talking faces that leverages recent advances in latent-space manipulation of Generative Adversarial Networks (GANs). Unlike existing approaches based on computationally intensive diffusion models, transformers, or complex 3DMM representations, which are impractical for mobile and other low-resource edge devices due to high memory and compute demands, GANimate is designed for efficient operation on low-memory, low-compute edge devices. The model operates on 2D lip landmarks extracted from standard mobile vision-sensor inputs and requires no pre-training, making it easily integrable with any lip-landmark generator. Through an optimization process in the GAN feature latent space, these landmarks act as geometric constraints to animate a static portrait, producing realistic and expressive lip movements. To maintain stability and visual coherence across frames, we employ a Kalman filter to detect and track lip landmarks during video synthesis, enabling adaptive refinement and improved temporal consistency. The result is a compact and modular framework that bridges the gap between performance and accessibility in talking face synthesis, delivering high-quality and stable animations with minimal computational overhead. GANimate represents an important step toward lifelike, real-time avatars suitable for sensor-enabled and mobile human–computer interaction. Full article

Surface electromyography (sEMG) signals are inevitably affected by noise during acquisition, thereby degrading signal quality and analytical reliability. Most existing denoising methods combine signal decomposition with thresholding, and their performance depends on empirically set decomposition parameters and threshold estimation. However, in high-rate repetitive motions such as rapid keystrokes, sustained high-duty-cycle muscle activation biases universal-threshold noise estimation, leading to unreliable thresholds. To overcome these issues, an sEMG denoising method that integrates the Walrus Optimizer (WO) with Variational Mode Decomposition (VMD) is proposed. WO is employed to optimize key VMD parameters, including the number of modes K and the penalty factor $\alpha$. Based on this method, an improved threshold estimation strategy is developed to accommodate high-duty-cycle sEMG during rapid keystrokes. It reduces thresholding-induced over-attenuation of meaningful myoelectric components. The dataset included 18 participants with sEMG recorded from six muscles during rapid keystroke tasks (10 trials per participant; 20 keystrokes per trial). Across input signal-to-noise ratios (SNRs) of 0, 5, 10, 15 dB, the proposed method achieved a median SNR improvement (ΔSNR) ranging from 2.75 to 6.65 dB and a median root-mean-square error (RMSE) reduction rate (ΔRMSE%) ranging from 27% to 53%, while maintaining spectral fidelity with a median of median frequency variation rate (ΔMDF%) below 3.48%.These results indicate that the proposed method provides an efficient and reliable solution for sEMG signal processing in rapid keystroke analysis. Full article

This paper investigates the performance of a multi-user multiple-input single-output (MU-MISO) broadcast channel under the practical constraints of imperfect, delayed, and quantized channel state information at the transmitter (CSIT). Conventional interference alignment (IA) strategies—classified into spatial (SIA), temporal (TIA), and message-domain (MIA) techniques— typically designed for specific, idealized CSI regimes and often rely on successive interference cancellation (SIC) at the receiver. However, the iterative structure of SIC is highly susceptible to error propagation, particularly under CSI uncertainty and high-order modulation. We propose Deep-STMIA, a novel end-to-end deep learning framework that jointly optimizes interference management across the space, time, and message domains. Using a neural network-based autoencoder architecture with structural message-domain regularization, Deep-STMIA learns to mitigate the catastrophic effects of error propagation and adapts to a continuum of CSIT conditions. Simulation results demonstrate that Deep-STMIA matches the performance of degrees-of-freedom (DoF) optimal benchmarks in extreme CSI regimes and significantly outperforms state-of-the-art baselines, such as rate-splitting multiple access (RSMA), in practical imperfect CSIT scenarios. Full article

Tomato is the most widely consumed vegetable worldwide and serves as an important source of vitamins and minerals. Using the Bacillus species as biocontrol agents and plant growth promoters is a sustainable approach to optimize production and mitigate the effects of root-infecting phytopathogenic fungi, thereby reducing reliance on chemical inputs. This study evaluated the effectiveness of a Bacillus sp.-based bioinoculant, produced in a 7 L bioreactor, for controlling root phytopathogens and enhancing tomato yields under field conditions. The trial was conducted at an experimental field of the Universidad Nacional Agraria La Molina (Lima, Peru) using a randomized complete block design with four blocks. Treatment means were compared using Tukey’s multiple range test (α = 0.05) to evaluate treatment effects. The treatments included three concentrations of the bioinoculant (10%, 20%, and 30%) derived from an initial concentration of 1 × 10⁸ CFU/mL of a Bacillus halotolerans IcBac2.1 strain sourced from the LEMyB laboratory strain collection, a commercial biological product (1 × 10⁹ CFU/g), and uninoculated control. Applications were made for the following four key stages of crop development: 10 days after germination, when transplanting through root dipping, 7 days after transplanting, and at the onset of flowering. In all treated groups, applications were directed to the plant crown, whereas the control group received no treatment. The evaluated variables included plant height (cm), stem diameter (mm), root disease incidence (%), chlorophyll index (SPAD), °Brix, pH, vitamin C (mg/100 g), total protein (mg/100 g) and crop yield (t/ha). The greatest plant growth-promoting effects were observed in plants inoculated with the 20% bioinoculant and in the commercial product treatment, as evidenced by increased plant height, greater fruit diameter, caliber, and length, as well as lower root disease incidence (2.86% and 1.43%, respectively). In addition, yields were highest in these treatments (29.9 and 25.2 t ha⁻¹, respectively) compared with 14.5 t ha⁻¹ in the control. These results indicate that a 20% B. halotolerans-based bioformulation, similar to the commercial formulation, promotes plant growth, improves agronomic performance, and reduces root disease incidence in tomato crops. Full article

This study investigates the structure of innovation and entrepreneurship competence among university students in China. Based on an analysis of 33 policy texts on innovation and entrepreneurship education from 2010 to 2022, it constructs a structural model of university students’ innovation and entrepreneurship competence comprising the knowledge layer, ability layer, and literacy layer by employing the Onion Model. From the perspective of policy instruments, a two-dimensional competence–policy instrument analytical framework is established. The analysis reveals that the articulation of university students’ innovation and entrepreneurship competence in policy texts exhibits distinct stage-wise evolutionary characteristics. Furthermore, the current policy support system suffers from three structural imbalances: an over-reliance on supply-side policy instruments, with insufficient synergy from environmental and demand-side instruments; weak support from environmental and demand-side instruments for certain key competencies; and an emphasis on explicit knowledge over implicit literacy in the cultivation logic. Consequently, this study proposes a shift in the policy paradigm from factor input to system generation. Recommendations include optimizing the mix of policy instruments, improving the precision of interventions by environmental and demand-side instruments targeting key competencies, and reconstructing the cultivation system based on the different generative logics of explicit and implicit competence. Full article

Background: Type 2 diabetes (T2D) is associated with acute and chronic complications, entailing significant use of healthcare resources. Clinical guidelines recommend holistic management and recognize the critical role of obesity and cardio-renal protection in T2D. This study aims to estimate the clinical, quality of life, and economic benefits of optimized weight, metabolic, and cardiovascular management of T2D-related complications in Spain. Methods: An estimation model was built incorporating the risk of complications associated with changes in glycated hemoglobin (HbA1c), weight and high-sensitivity C-reactive protein (hs-CRP), considering incidence of complications and healthcare costs in Spain. A literature review was performed to identify these clinical inputs. The potential reduction in the annual number of complications and their associated disability-adjusted life years (DALYs) and costs were estimated for reductions of 1% HbA1c, 5% weight and 0.5 mg/L hs-CRP in three T2D patient subgroup scenarios. Probabilistic sensitivity analyses were conducted to validate the results and determine their potential range. Results: Combined reduction of HbA1c, weight and hs-CRP was estimated to prevent 19.16–20.80% T2D complications per year. This led to an estimated range of 1317–6568 avoided DALYs, and potential annual savings between €242.77M and €821.68M depending on the T2D patient subgroup. Savings per patient and year ranged from €196.86 to €296.75 for the three scenarios analyzed. Sensitivity analysis validated these results. Conclusions: Integrated management of patients with T2D, controlling HbA1c levels, weight, and cardiovascular benefit, can improve patient outcomes, reduce incidence of complications, prevent quality of life worsening, and result in cost savings for the Spanish national healthcare system. Full article

Dynamic Vision Sensor (DVS) is an event-based imaging technology inspired by biological photoreceptors, which holds great promise for edge computing. The event streams produced by DVS are often contaminated by Background Activity (BA) noise and hot-pixel noise, which degrade downstream processing. Existing filters typically use fixed parameters, resulting in poor adaptability to changing illumination. In this paper, we propose a lightweight Adaptive Event-based Filtering Spiking Neural Network (AEFSNN) to address these limitations. Inspired by homeostatic plasticity, AEFSNN dynamically adjusts neuronal thresholds by monitoring the input-to-output spike ratio, allowing the network to autonomously converge to an optimal operating point across different lighting conditions. Furthermore, we introduce a novel neuronal wake-up mechanism that inhibits processing neurons until triggered by valid input, which effectively suppresses redundant events generated by neighboring activity. Experiments show that AEFSNN is more robust under varying illumination. Compared with current filters, our method increases the Signal-to-Noise Ratio (SNR) of the output data by 1.42–2.33 dB. Additionally, the filtered data improves classification accuracy on downstream tasks, validating its practical value for neuromorphic vision systems. Full article