Aadil Arshad Ferhath

This study experimentally investigates the dynamic response of a half-car roll suspension system equipped with a magnetorheological (MR) damper under vehicle speeds of (3, 7, and 12) m s ⁻¹ and damper current levels ranging from (0 to 2) A. The objective is to evaluate the influence of variable damping on sprung-mass vibration, seat acceleration, and the body’s rotational dynamics. A custom half-car test rig incorporating a front-axle Magnetorheological damper (MR damper), accelerometers, gyroscope sensors, and a synchronised data acquisition system was developed to measure the system’s response under controlled bump-excitation conditions. The results show that increasing the applied current generally reduced sprung-mass acceleration, vibration severity, and rotational responses by increasing the damping force and the suspension system’s energy-dissipation capability. Peak, root mean square, vibration dose value, and motion sickness dose value metrics exhibited significant reductions across the investigated speed range. The effectiveness of the MR damper was particularly evident under moderate and high excitation conditions, with substantial reductions in vibration transmission. Seat acceleration measurements revealed a nonlinear relationship between damper current and occupant vibration response, with the highest current level (2 A) consistently providing the greatest reduction. Roll, pitch, and yaw responses also decreased with increasing current, indicating improved suppression of body-motion oscillations. The findings demonstrate that MR dampers can effectively reduce both vertical and rotational vibration responses in a half-car suspension system.

This study provides a systematic robustness evaluation of classical machine learning for vibration-based bearing fault detection in low-RPM internal combustion engines (1000–2000 RPM) across a controlled temperature × humidity grid, a regime underrepresented in benchmark datasets that emphasise high-speed applications. A publicly available dataset from a 658cc engine (–10 °C–45 °C, 0%–100% humidity) was analysed; vibration features were derived from the non-zero channels of a tri-axial acquisition, with the bearing-housing vibration carried primarily by channel Ch3. To prevent temporal leakage, 390 263 continuous measurements were aggregated into 89 steady-state units, each spanning 90 s, yielding a deliberately independence-preserving but low sample-to-feature ratio (89:92). Four algorithms Random Forest, Support Vector Machine, Logistic Regression, and Neural Network were evaluated using stratified 5-fold cross-validation. All models achieved apparent accuracy exceeding 95%, with Random Forest performing best (97.8% ± 2.7%), but no statistically significant differences were found (Friedman test, p = 0.732). Vibration features, particularly crest factor and root mean square, provided the greatest discriminative power, while environmental factors accounted for less than 17% combined importance. The near-perfect linear separability (98.2% with Logistic Regression) indicates that the dataset’s binary, controlled-laboratory labelling rather than intrinsic bearing-degradation physics drives the clean classification. Accordingly, the reported accuracies are apparent upper-bound estimates from an exploratory study, not expected field performance; validation on 500–1000 or more samples with progressive-degradation labelling is essential before any operational claim can be supported.

Machine learning allows systems to autonomously learn from data and improve performance without direct programming, providing robust tools for identifying patterns, forecasting outcomes, and optimising complex processes. This paper offers a comprehensive overview of machine learning, beginning with an examination of its core categories: supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, and deep learning. The review considers essential algorithms associated with each machine learning category, providing insights into their functions and real-world applications. The paper also discusses widely used evaluation metrics for regression models, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R-squared (R²), which are essential for assessing predictive performance and guiding model selection. The article examines numerous real-world applications of machine learning across various sectors, including healthcare, banking, transportation, marketing, and cybersecurity, illustrating how these technologies are transforming modern processes and delivering tangible benefits.