A Data-Driven Vibration Analysis Framework for Micro-Motor Fault Diagnosis and Quality Control

Vibration-Based Fault Detection in Rotating Machinery: The Effect of Accelerometer Position and Ensemble Learning

Fault Diagnosis of Automotive Gasoline Engines Using Time–Frequency Vibration Image Analysis

A Novel Vibration Centroid-Based Approach for Fault Diagnosis of Transformer Winding

Tank vibrations of a power transformer, originating primarily from winding vibration and core vibration through mechanical coupling and fluid–structure interaction, are regarded as essential carrier signals for assessing the integrity of the winding. To improve the diagnostic accuracy of winding condition, this paper presents a vibration centroid-based diagnostic model that integrates feature fusion from vibration signals. According to the frequency spectrum of vibration signals obtained using Zoom-FFT, a set of new spatial vibration feature vectors—namely vibration centroid coordinates and Boyce-Clark shape index—were defined. This approach converts spatially distributed vibration signals into compact and discriminate indicators. A diagnostic model was subsequently constructed by integrating the grey wolf optimization (GWO) algorithm with the least squares support vector machine (LSSVM), ensuring that optimal classification performance was achieved. No-load, short-circuit, and load tests were made on a 35 kV-rated oil-immersed transformer. During the experiments, the transformer winding was divided into four categories: healthy condition, winding looseness, axial deformation, and radial deformation. The proposed GWO-LSSVM-based classifier was trained and tested using the defined vibration feature vectors. The results indicate that the proposed method achieves superior performance, with a recognition rate of 98.44%, and offers high efficiency.

Real-Time Vibration Detection and Signal Analysis of Machines

This work is a condition-based maintenance low-cost real-time vibration monitoring system of industrial machines based on the ESP32 microcontroller and MPU6050 triaxial accelerometer. The axis-wise and overall RMS vibration amplitude is calculated in the system by time domain processing, which offsets them based on a specified threshold that is installed in MATLAB and which is also embedded in the profile parameters per the requirements of vibration severity. The FFT-based real-time spectral analysis is used to detect the vibration frequency components that are predominant in mechanical faults. The machine condition is automatically determined to be safe, warning or dangerous based on quantitative factors such as mean RMS vibration and danger ratio based on time. Multimachine experimental results record values of RMS vibration between 0.02 g and 0.78 g and a predominant low-frequency vibration signal between 0.3 and 3 Hz. The proposed system was implemented with classification accuracy of 91.67%, analysis latency of less than 100ms. The entire self-sacrificing system has a cost of less than 15 USD, making it fit the limited resource-based industrial settings.

Vibration Analysis of High-Speed Rotating Machinery

The saving grace of industrial systems in the present day is high-speed rotating machinery which encompasses turbines, compressors, generators and aerospace propulsion units. The successful performance of such machines largely remains the responsibility of efficient condition monitoring and fault diagnosis methods. Vibration analysis has become one of the most potent and popular in the number of these techniques. A cohesive exploration of the vibration nature of high-speed rotating machinery with its focus on signal acquisition, signal processing, feature extraction, and fault classification techniques is discussed in this paper. The process combines both experimental measurements, mathematical modeling and using advanced signal processing to detect typical mechanical faults including imbalance, misalignment, bearing flaws, shaft cracks and gear mesh anomaly. An elaborate experimental design is crafted based on an accelerometer, data collection apparatus, and spectral analysis apparatus to record the signature of vibrations at varying operation conditions. The time-domain analysis, frequency-domain abasys and time-frequency-domain analysis are used to extract diagnostic features that are usually significant. Short-Time Fourier Transform (STFT), Fast Fourier Transform (FFT), and Wavelet Transform (WT) techniques are adopted to make a fault more detectable. In addition, automated fault recognition is performed with the help of statistical indicators and classifiers based on machine learning. The findings indicate that vibration-based diagnostics have demonstrated high relative accuracy of early fault detection and reliability of the system. Comparative study shows that the hybrid signal processing solutions are better than the conventional methods in complicated operational scenarios. The given methodology has offered a systematic framework of being predictive in maintenance developed in industrial rotating machines. The results of this study help in making the operations safe, minimizing downtime and minimizing costs of maintenance. The research can be used by the researchers and practitioners who wish to adopt modern vibration monitoring systems in the rotating machines that operate at high speed.

Artificial Intelligence-Driven Quality Control in Mechanical Manufacturing: Vibration-Based Multiclass Gear Fault Detection Using LightGBM

The results show that vibration-based machine learning can support robust, near-real-time fault identification in mechanical manufacturing environments and highlights the importance of chronological validation, feature engineering over multiple time windows, and the trade-off between predictive performance and deployment efficiency.