A study on rolling bearing fault diagnosis using RIME-VMD

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 8, 2025

To address the challenges of feature extraction in Variational Mode Decomposition (VMD) for rolling bearing fault diagnosis, this paper proposes a method optimized by RIME algorithm, called RIME-VMD. First, under various conditions, algorithm is employed to determine optimal combination decomposition components and penalty factors VMD. Next, kurtosis values each decomposed Intrinsic Function (IMF) are calculated, component with most prominent features selected noise reduction through reconstruction. Finally, sample entropy reconstructed signal calculated as input into Support Vector Machine (SVM) rapid identification diagnosis types. Simulation results indicate that, compared Whale Optimization Algorithm VMD (WOA-VMD), (RIME-VMD) achieves shorter search times higher efficiency. It facilitates faster parameters enhancing robustness detection enabling rapid, efficient faults. The findings study offer guidance reference future research on diagnosis.

Language: Английский

---

Unlocking the power of knowledge for few-shot fault diagnosis: A review from a knowledge perspective

Information Sciences, Journal Year: 2025, Volume and Issue: unknown, P. 121996 - 121996

Published: Feb. 1, 2025

Language: Английский

---

Machine learning aided photovolatic property predictions, design and library generation of indeno-fluorene donors with lowest exciton bindings

Solar Energy, Journal Year: 2025, Volume and Issue: 291, P. 113399 - 113399

Published: March 6, 2025

Language: Английский

---

Precision assembly error analysis of parts based on multi-constraint surface matching

Frontiers in Mechanical Engineering, Journal Year: 2025, Volume and Issue: 10

Published: Jan. 20, 2025

Existing assembly analysis methods often fail to accurately capture the complexities involved in precision of real-world parts. This paper introduces an advanced error method based on multi-constraint surface matching, aimed at overcoming these limitations. The proposed approach incorporates interference-free constraints and force stability develop positioning model that better reflects realistic process. To solve model, Spatial Pyramid Matching with chaotic mapping is employed for population initialization, thereby enhancing diversity. A nonlinear control mechanism further introduced dynamically adjust inertia weight, a simulated annealing integrated into particle swarm optimization algorithm enhance efficiency matching ultimately achieves high-precision completes comprehensive analysis. effectiveness enhanced performance methodology are validated through vibratory bowl feeder, demonstrating its potential significantly improve accuracy manufacturing contexts.

Language: Английский

---

Deep Neural Network for Valve Fault Diagnosis Integrating Multivariate Time-Series Sensor Data

Actuators, Journal Year: 2025, Volume and Issue: 14(2), P. 70 - 70

Published: Feb. 5, 2025

Faults in valves that regulate fluid flow and pressure industrial systems can significantly degrade system performance. In where multiple are used simultaneously, a single valve fault reduce overall efficiency. Existing diagnosis methods struggle with the complexity of multivariate time-series data unseen scenarios. To overcome these challenges, this study proposes method based on one-dimensional convolutional neural network (1D CNN) for diagnosing location severity faults multi-valve system. An experimental setup was constructed 17 sensors, including 8 sensors at inlets outlets 4 valves, 5 along main pipe. Sensor were collected to observe sensor values corresponding behavior, foreign objects varying sizes inserted into simulate different severities. These train evaluate proposed model. The achieved robust prediction accuracy (MAE: 0.0306, RMSE: 0.0629) compared existing networks, performing both trained It identified faulty quantified severity, demonstrating generalization capabilities.

Language: Английский

---

A multi-source domain adaption intelligent fault diagnosis method based on asymmetric adversarial training

Measurement Science and Technology, Journal Year: 2025, Volume and Issue: 36(3), P. 036123 - 036123

Published: Feb. 18, 2025

Abstract To enhance the cross-domain diagnostic ability of model, domain adaptation method is adopted. When using traditional adaption methods to extract invariant characteristics axial flow fan faults, source and target domains will be close each other, thereby distribution trained changed. fault gather at classification boundary, model incorrectly classify some samples. In addition, single can lead poor generalization ability. resolve above issues, a multi-source intelligent diagnosis based on asymmetric adversarial training proposed. this method, used realize unidirectional movement from domain; triplet-center loss expand inter-class distance shorten intra-class in are extracted different domains, they inputted their respective classifiers, then aligning outputs classifier cosine similarity. improve strategy weights The industrial actual data verification results indicate that effective solving relevant practical problems.

Language: Английский

---

Harmonic Interference Resilient Backscatter Communication with Adaptive Pulse-Width Frequency Shifting

Electronics, Journal Year: 2025, Volume and Issue: 14(5), P. 946 - 946

Published: Feb. 27, 2025

The last few decades have witnessed the rapid development of passive backscatter technologies, which envision promising cost-efficient ambient Internet Things (IoT) for various applications, such as distributed solar sensor networks. However, limited by harmonic interference caused conventional frequency-shifting-based control methods, existing communication technologies cannot support growing scale network. To tackle this issue, we propose a resilient frequency-shifting technique to compress harmonics during communication. Different from tags that shift frequency with square waves constant pulse width, dynamically modify width wave different parts waves. Furthermore, lightweight coding algorithm enhance compatibility our system applications. We implement off-the-shelf components and conduct comprehensive experiments evaluate performance. results demonstrate can reduce BER (bit error rate) 70%.

Language: Английский

---

A Three-Channel Feature Fusion Approach Using Symmetric ResNet-BiLSTM Model for Bearing Fault Diagnosis

Symmetry, Journal Year: 2025, Volume and Issue: 17(3), P. 427 - 427

Published: March 12, 2025

For mechanical equipment to operate normally, rolling bearings—which are crucial parts of rotating machinery—need have their faults diagnosed. This work introduces a bearing defect diagnosis technique that incorporates three-channel feature fusion and is based on enhanced Residual Networks Bidirectional long- short-term memory networks (ResNet-BiLSTM) model. The can effectively establish spatial-temporal relationships better capture complex features in data by combining the powerful spatial extraction capability ResNet bidirectional temporal modeling BiLSTM. Specifically, one-dimensional vibration signals first transformed into two-dimensional images using Continuous Wavelet Transform (CWT) Markov Transition Field (MTF). upgraded ResNet-BiLSTM network then used extract combine original signal along with from two types images. Finally, experimental validation performed datasets. results show compared other state-of-the-art models, computing cost greatly reduced, params flops at 15.4 MB 715.24 MB, respectively, running time single batch becomes 5.19 s. fault accuracy reaches 99.53% 99.28% for datasets, successfully realizing classification task.

Language: Английский

---

Experimental investigation of vibrational signal in a fault induced Francis runner

Frontiers in Mechanical Engineering, Journal Year: 2025, Volume and Issue: 11

Published: March 25, 2025

Hydro turbines are prone to failure and the detection of fault in turbine is essential ensure reliability power plant. This study investigates vibrational signals a fault-induced Francis using an experimental test setup identify trends that could be helpful diagnosis faults. By analyzing signal, aims correlate turbine's dynamic behavior. Faults have been introduced by adding masses blades, tests conducted under two different conditions: dry wet testing conditions for both normal faulty blades. The operating condition determined with help pressure, flow, RPM sensors. speed varied variable frequency drive. For acquisition vibration signals, NI-LabVIEW system employed along uniaxial sensor located at bearing. obtained data analyzed Fast Fourier Transform (FFT) algorithm wavelet transform frequency-domain characteristics. While studying comparing fundamental shaft, it found faults can either increase or decrease amplitude resonant peak system, but other frequencies remains almost unaffected.

Language: Английский

---

Research on Fault Detection and Classification of Industrial Equipment Based on Machine Learning

Journal of Computer and Communications, Journal Year: 2025, Volume and Issue: 13(04), P. 226 - 243

Published: Jan. 1, 2025

Language: Английский

---