Search Results (193)

As the main load-bearing components of engineering structures, regular health assessment of reinforced concrete (RC) columns is crucial for improving the service life and overall performance of the structures. This study focuses on the health detection problem of in-service RC columns. By combining deep learning algorithms and acoustic emission (AE) technology, the AE sources of in-service RC columns are located, and the optimal sensor layout form for the health monitoring of in-service RC columns is determined. The results show that the data cleaning method based on the k-means clustering algorithm and the voting selection concept can significantly improve the data quality. By comparing the localization performance of the Back Propagation (BP), Radial Basis Function (RBF) and Support Vector Regression (SVR) models, it is found that compared with the RBF and SVR models, the MAE of the BP model is reduced by 7.513 mm and 6.326 mm, the RMSE is reduced by 9.225 mm and 8.781 mm, and the R² is increased by 0.059 and 0.056, respectively. The BP model has achieved good results in AE source localization of in-service RC columns. By comparing different sensor layout schemes, it is found that the linear arrangement scheme is more effective for the damage location of shallow concrete matrix, while the hybrid linear-volumetric arrangement scheme is better for the damage location of deep concrete matrix. The hybrid linear-volumetric arrangement scheme can simultaneously detect damage signals from both shallow and deep concrete matrix, which has certain application value for the health monitoring of in-service RC columns. Full article

To address the challenge of efficiently identifying and providing early warnings for typical structural damages in small and medium-sized bridges during long-term service, this paper proposes an intelligent monitoring and recognition method based on ultra-weak fiber Bragg grating (UWFBG) array sensing. By deploying UWFBG strain-sensing cables across the bridge, the system enables continuous acquisition and spatial analysis of multi-point strain data. Based on this, a series of experimental scenarios simulating typical structural damages—such as single-slab loading, eccentric loading, and bearing detachment—are designed to systematically analyze strain evolution patterns before and after damage occurrence. While strain distribution maps allow for visual identification of some typical damages, the approach remains limited by reliance on manual interpretation, low recognition efficiency, and weak detection capability for atypical damages. To overcome these limitations, machine learning algorithms are further introduced to extract features from strain data and perform pattern recognition, enabling the construction of an automated damage identification model. This approach enhances both the accuracy and robustness of damage recognition, achieving rapid classification and intelligent diagnosis of structural conditions. The results demonstrate that the integration of the monitoring system with intelligent recognition algorithms effectively distinguishes different types of damage and shows promising potential for engineering applications. Full article

This study presents an advanced diagnostic methodology for assessing mechanical faults in high-performance jet engines using non-contact laser vibrometry and Fourier-based vi-bration analysis. Focusing on Pratt & Whitney F100-PW-229 engines used in F-16 aircraft, thise research identifies critical measurement locations, including the gearbox, turbine, and compressor supports. High-resolution vibration signals were collected under test bench conditions and processed using fFast Fourier tTransform (FFT) techniques to extract frequency-domain features indicative of rotor imbalances, bearing wear, and structural anomalies. Comparative analysis between nominal and degraded engines confirmed strong correlations between analytical predictions and empirical spectral patterns. Thise study introduces a signal processing framework combining time–frequency analysis with Relief-F-based feature selection, laying the groundwork for future integration with ma-chine learning algorithms. This non-intrusive, efficient diagnostic method supports early fault detection, enhances engine availability, and contributes to the development of a na-tional vibration reference database, especially vital in the absence of OEM-supplied tools. Full article

Large turbine generators rely on shaft earthing brushes to safely divert harmful shaft currents to ground, protecting bearings from electrical damage. This paper presents a novel deep learning-based diagnostic framework to detect and classify faults in shaft earthing brushes of large turbine generators. A key innovation lies in the use of FFT-derived spectrograms from both voltage and current waveforms as dual-channel inputs to the CNN, enabling automatic feature extraction of time–frequency patterns associated with different SEB fault types. The proposed framework combines advanced signal processing and convolutional neural networks (CNNs) to automatically recognize fault-related patterns in shaft grounding current and voltage signals. In the approach, raw time-domain signals are converted into informative time–frequency representations, which serve as input to a CNN model trained to distinguish normal and faulty conditions. The framework was evaluated using data from a fleet of large-scale generators under various brush fault scenarios (e.g., increased brush contact resistance, loss of brush contact, worn out brushes, and brush contamination). Experimental results demonstrate high fault detection accuracy (exceeding 98%) and the reliable identification of different fault types, outperforming conventional threshold-based monitoring techniques. The proposed deep learning framework offers a novel intelligent monitoring solution for predictive maintenance of turbine generators. The contributions include the following: (1) the development of a specialized deep learning model for shaft earthing brush fault diagnosis, (2) a systematic methodology for feature extraction from shaft current signals, and (3) the validation of the framework on real-world fault data. This work enables the early detection of brush degradation, thereby reducing unplanned downtime and maintenance costs in power generation facilities. Full article

This study presents a comprehensive investigation of acoustic emission (AE) characteristics for condition monitoring of rolling bearings in passenger ropeway systems. Through controlled laboratory experiments and field validation across multiple operational ropeways, we establish an optimized AE-based diagnostic framework. Key findings demonstrate that resonant VS150-RIC sensors outperform broadband sensors in defect detection, showing greater energy response at characteristic frequencies for inner race defects. The RMS parameter emerges as a robust diagnostic indicator, with defective bearings exhibiting periodic peaks and higher mean RMS values. Field tests reveal progressive RMS escalation preceding visible damage, enabling predictive maintenance. Furthermore, we develop a novel Paligemma LLM model for automated wear detection using AE time-domain images. The research validates the AE technology’s superiority over conventional vibration methods for low-speed bearing monitoring, providing a scientifically grounded approach for safety-critical ropeway maintenance. Full article

Large steam turbine generators are increasingly vulnerable to damage from shaft voltages and bearing currents due to the widespread adoption of modern power electronic excitation systems and more flexible operating regimes. Earthing brushes provide a critical path for discharging these shaft currents and voltages, but their effectiveness depends on the timely detection of brush degradation or faults. Conventional monitoring of shaft voltage and current is often rudimentary, typically limited to peak readings, making it challenging to identify specific fault conditions before mechanical damage occurs. This study addresses this gap by systematically analyzing shaft voltage and current signals under various controlled earthing brush fault conditions (floating brushes, worn brushes, and oil/dust contamination) in several large turbine generators. Experimental site tests identified distinct electrical signatures associated with each fault type, demonstrating that online shaft voltage and current measurements can reliably detect and classify earthing brush faults. These include unique RMS, DC, and harmonic patterns in both voltage and current signals, enabling accurate fault classification. These findings highlight the potential for more proactive maintenance and condition-based monitoring, which can reduce unplanned outages and improve the reliability and safety of power generation systems. Full article

Bearings are vital apparatuses in many industrial systems, and their failure can lead to severe damage, costly downtime, and safety risks. Therefore, early detection of bearing faults is critical to prevent catastrophic failures. However, diagnosing bearing faults in real-world conditions is challenging due to noise, which can obscure vibration signals and reduce the effectiveness of traditional diagnostic techniques. This paper portrays a unique method for bearing fault identification in high-noise environments by integrating Improved Complete Ensemble Empirical Mode Decomposition (ICEEMD) and Continuous Wavelet Transform (CWT). ICEEMD decomposes complex vibration signals into intrinsic mode functions, effectively filtering out noise and enhancing feature extraction. CWT is then applied to obtain a time–frequency representation of the cleaned signal, allowing for precise detection of transient events and frequency variations associated with faults. The proposed approach is evaluated using simulated signals, achieving a testing accuracy of 78% at −20 dB SNR, demonstrating its robustness in noisy environments. This study highlights the capability of combining ICEEMD and CWT for robust fault diagnosis in noisy industrial applications, paving the way for improved predictive maintenance strategies. Full article

Fault detection is a critical research area, especially in the automotive sector, aiming to quickly assess component conditions. Machine Learning techniques, powered by Artificial Intelligence, now represent state-of-the-art methods for this purpose. This study focuses on durability testing of Permanent Magnet Synchronous Motors for automotive applications, using Autoencoders (AEs) to predict and prevent failures. This AI-based fault detection strategy employs acceleration signals coming from electric motors tested under challenging conditions with significant variations in torque and speed. This approach goes beyond typical fault detection in steady-state conditions. Based on a review of Neural Networks, including Variational Autoencoders (VAEs), Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks, the performance of six AI architectures is compared: AE, VAE, 1D CNN AE, 1D CNN VAE, LSTM AE and LSTM VAE. The 1D CNN AE outperformed the other networks in fault detection, showing high accuracy, stability and computational efficiency. The model is integrated into an algorithm for semi-real-time fault monitoring. The algorithm effectively detects potential motor failures in real-world scenarios, including bearing faults, mechanical misalignments, and progressive wear of components, thereby proactively preventing damage and halving test bench downtime. Full article

To overcome the limitations associated with chemically synthesized nanoparticles in cancer therapy, researchers have increasingly focused on developing nanoparticles with superior biocompatibility and prolonged tumor retention using biosynthetic methods. In this study, we first identified the presence of calcium peroxide nanoparticles (CaO₂ NPs) in the blood of individuals who had ingested calcium gluconate. Furthermore, the dropwise addition of calcium gluconate to human serum resulted in the spontaneous self-assembly of CaO₂ NPs. Next, following tail vein injection of fluorescently labeled CaO₂ NPs into subcutaneous tumor-bearing nude mice, we observed that the nanoparticles exhibited prolonged accumulation at the tumor sites compared to other organs through visible-light imaging. Immunofluorescence staining demonstrated that CaO₂ NPs co-localized with vesicular transport-associated proteins, such as PV-1 and Caveolin-1, as well as the albumin-binding-associated protein SPARC, suggesting that their transport from tumor blood vessels to the tumor site is mediated by Caveolin-1- and SPARC-dependent active transport pathways. Additionally, the analysis of various organs in normal mice injected with CaO₂ NPs at concentrations significantly higher than the experimental dose showed no apparent organ damage. Hemolysis assays indicated that hemolysis occurred only at calcium concentrations of 300 µg/mL, whereas the experimental concentration remained well below this threshold with no detectable hemolytic activity. In a subcutaneous tumor-bearing nude mouse model, treatment with docetaxel-loaded CaO₂ NPs showed a 68.5% reduction in tumor volume compared to free docetaxel (DTX) alone. These novel biosynthetic CaO₂ NPs demonstrated excellent biocompatibility, prolonged retention at the tumor site, safety, and drug-loading capability. Full article

Misalignment faults in drive systems occur when the motor and load are not properly aligned, leading to deviations in the centerlines of the coupled shafts. These faults can cause significant damage to bearings, shafts, and couplings, making early detection essential. Traditional diagnostic techniques rely on vibration monitoring, which provides insights into both mechanical and electromagnetic fault signatures. However, its main drawback is the need for external sensors, which may not be feasible in certain applications. Alternatively, motor current signature analysis (MCSA) has proven effective in detecting faults without requiring additional sensors. This study investigates misalignment faults in synchronous reluctance motors (SynRMs) by analyzing both vibration and current signals under different load conditions and operating speeds. Fast Fourier transform (FFT) is applied to extract characteristic frequency components linked to misalignment. Experimental results reveal that the amplitudes of rotational frequency harmonics ($1x{f}_r$, $2x{f}_r$, and $3x{f}_r$) increase in the presence of misalignment, with $1x{f}_r$ exhibiting the most stable progression. Additionally, acceleration-based vibration analysis proves to be a more reliable diagnostic tool compared to velocity measurements. These findings highlight the potential of combining current and vibration analysis to enhance misalignment detection in SynRMs, improving predictive maintenance strategies in industrial applications. Full article

Rolling element bearings are critical components of rotating machinery, with their performance directly influencing the efficiency and reliability of industrial systems. At the same time, bearing faults are a leading cause of machinery failures, often resulting in costly downtime, reduced productivity, and, in extreme cases, catastrophic damage. This study presents a methodology that utilizes Kolmogorov–Arnold Networks—a recent deep learning alternative to Multilayer Perceptrons. The proposed method automatically selects the most relevant features from sensor data and searches for optimal hyper-parameters within a single unified approach. By using shallow network architectures and fewer features, the resulting models are lightweight, easily interpretable, and practical for real-time applications. Validated on two widely recognized datasets for bearing fault diagnosis, the framework achieved perfect F1-Scores for fault detection and high performance in fault and severity classification tasks, including 100% F1-Scores in most cases. Notably, it demonstrated adaptability by handling diverse fault types, such as imbalance and misalignment, within the same dataset. The availability of symbolic representations provided model interpretability, while feature attribution offered insights into the optimal feature types or signals for each studied task. These results highlight the framework’s potential for practical applications, such as real-time machinery monitoring, and for scientific research requiring efficient and explainable models. Full article

The catastrophic twin earthquakes that struck southern Turkey and northern Syria on 6 February 2023 caused massive casualties and extensive damage to infrastructure, with Hatay Province of Turkey bearing the brunt of the impact. To swiftly and thoroughly assess the damage caused by the earthquakes and the subsequent reconstruction efforts, this study initially investigated the application of light change ratios between the pre-earthquake monthly nighttime lights (NTLs) and the post-earthquake daily NTL data to identify earthquake damage in Hatay Province. Next, the monthly NTL data were employed to calculate the time series average lighting index (ALI). Subsequently, random noise and seasonal fluctuation were eliminated through data smoothing and seasonal decomposition techniques. Pre- and post-earthquake regression models were then utilised to establish a comprehensive framework for assessing economic recovery following the earthquake. The findings indicated that (1) the seismic damage identification method based on NTL data achieved an overall accuracy exceeding 71.55% in detecting building damage after a disaster. This method provided a swift and effective solution for rapidly assessing disaster-related destruction. (2) The reduced NTLs exhibited a strong correlation with the area of severely and moderately damaged buildings while showing a weaker correlation with areas of slightly damaged buildings. (3) The developed pre- and post-earthquake regression models demonstrated a high degree of fit, making them valuable tools for assessing regional economic recovery after the earthquake. At the county scale, such districts as Erzin and Kumlu exhibited promising signs of recovery, while such severely impacted areas as Antakya faced misconceptions of progress, primarily due to the brightening of NTLs caused by reconstruction efforts. Additionally, such districts as Dortyol and Samandag grappled with substantial short-term recovery challenges. Although the province experienced gradual economic recovery, achieving complete restoration has remained complex and time-intensive. The study offers valuable insights into earthquake damage assessment and economic recovery monitoring while serving as a scientific reference for disaster mitigation and post-disaster reconstruction efforts. Full article

Electric machines (EMs) of electrified vehicle drivetrains can be tested on drivetrain test benches at an early stage of development. In order to protect the EMs from premature damage or failure during testing, monitoring their thermal condition is important. Due to the package requirements of compact and powerful EMs with high-speed requirements and high-power densities, the heat build-up inside the motor during operation is particularly high. For this reason, fluid cooling with heat exchangers is increasingly being used in EMs. The EMs analysed in this work are water-cooled by a cooling jacket. This influences the heat flow inside the machine through heat transfer mechanisms, making it difficult to detect damage to the EMs. This paper presents a novel method for non-destructive and non-contact thermal condition monitoring of water-cooled EMs on drivetrain test benches using thermography. In an experimental setup, infrared images of an intact water-cooled EM are taken. A bearing of the EM’s rotor is then damaged synthetically, and the experiment is repeated. The infrared images are then processed and analysed using appropriate software. The analysis of the infrared images shows that the heat propagation of the motor with bearing damage differs significantly from the heat propagation of the motor without bearing damage. This means that thermography opens up another method of condition monitoring for water-cooled EMs. The results of the investigation serve as a basis for future condition monitoring of water-cooled EMs on powertrain test benches using artificial intelligence (AI). Full article

In induction motor bearings, distributed faults are prevalent, often resulting from factors such as inadequate lubrication and particle contamination. Unlike localized faults, distributed faults produce complex and unpredictable motor signal behaviors. Although existing research predominantly addresses localized faults in mains-fed motors, particularly single-point defects, a comprehensive investigation into particle contamination in bearings of inverter-fed motors is essential for a more accurate understanding of real-world bearing issues. This paper conducts a comparative analysis of vibration, stator current, speed, and acoustic signals to detect particle contamination through signal analysis across three domains: time, frequency, and time-frequency. These domains are analyzed to assess and compare the characteristics of each monitored signal in the context of bearing wear detection. The data were collected from both steady-state and startup transients of an induction motor controlled by a variable frequency drive. The experimental results highlight the most significant characteristics of each monitored signal, evaluated across the different domains of analysis. The primary conclusion indicates that, in inverter-fed motors, sound and vibration signals exhibit abnormal levels when the bearing is damaged but the related-fault signature becomes complicated. Additionally, the findings demonstrate that the analysis of startup stator current and speed signals presents the potential to detect distributed bearing damage in inverter-fed induction motors. Full article

Establishing dependable and resilient methodologies for identifying damage that may compromise the integrity of reinforced concrete (RC) infrastructures is imperative for preventing potential catastrophic failures. Continuous evaluation and Structural Health Monitoring (SHM) can play a key role in extending the lifespan of new or existing buildings. At the same time, early crack detection in critical members prevents bearing capacity loss and potential failures, enhancing safety and reliability. Furthermore, implementing discrete fibers in concrete has significantly improved the ductility and durability of Fiber-Reinforced Concrete (FRC). The present study employs a hierarchical clustering analysis (HCA) to identify damage in FRC by analyzing the raw Electromechanical Impedance (EMI) signature of piezoelectric lead zirconate titanate (PZT) transducers. The experimental program consisted of three FRC standard cylinders subjected to repeated loading. The loading procedure consists of 6 incremental steps carefully selected to gradually deteriorate FRC’s structural integrity. Additionally, three PZT patches were adhered across the height of its specimen using epoxy resin, and their EMI response was captured between each loading step. Subsequently, the HCA was conducted for each PZT transducer individually. The experimental investigation demonstrates the efficacy of HCA in detecting load-induced damage in FRC through the variations in the EMI signatures of externally bonded PZT sensors. Full article