Adel Boudiaf

Timely and accurate fault diagnosis in rotary machinery is essential for ensuring system reliability and minimizing unplanned downtime. While deep learning approaches, particularly Convolutional Neural Networks (CNNs), have demonstrated strong performance in vibration-based fault classification, their limited interpretability poses challenges for adoption in safety-critical environments. To address this, the present study introduces a hybrid diagnostic framework that integrates CNN-based transfer learning with interpretable supervised classification, aiming to enhance both predictive accuracy and model transparency. A key innovation of this work lies in the dual-stage feature selection process, combining Analysis of Variance (ANOVA) and Permutation Feature Importance (PFI) to refine deep features extracted from a pre-trained VGG19 network. This strategy improves both dimensionality reduction and classification performance in a statistically grounded, model-agnostic manner. Furthermore, SHapley Additive exPlanations (SHAP) are employed to interpret the predictions, offering insight into the most influential features driving the classification decisions. Experimental evaluation on the Case Western Reserve University (CWRU) bearing dataset confirms the effectiveness of the proposed approach, achieving 100% classification accuracy using ten-fold cross-validation. By uniting high performance with transparent decision-making, the framework demonstrates strong potential for explainable and reliable fault diagnosis in industrial settings.

In this paper, we propose a federated deep reinforcement learning framework, named FL-DDQN, designed to enhance irrigation management in smart farming environments. The proposed framework addresses key challenges posed by fragmented data collected from small-scale farming devices, as well as the complexities associated with varying climate and soil patterns. Specifically, our framework optimizes the selection of clients in the federated learning process through deep reinforcement learning. The evaluation of the proposed framework is conducted across different multifarm configurations, showcasing its scalability and adaptability. Furthermore, the FL-DDQN framework facilitates collaborative training of a weather and climate forecasting model using a hybrid convolutional neural network (CNN)–long short-term memory architecture. This model is enhanced by incorporating regulation and normalization techniques, which help mitigate the effects of distribution shifts and pattern changes in dynamic farming environments. The results demonstrate high accuracy, achieving a low mean absolute error for soil moisture (0.0118), temperature (1.0200), and relative humidity (4.3958). Additionally, we integrate autoencoders based on CNNs to detect anomalies in irrigation system sensor readings by evaluating the reconstruction error. The proposed framework achieves a significantly lower reconstruction error compared to recent state-of-the-art methods, confirming its robustness in anomaly detection.

The quality control of steel products' surface is of utmost importance, where several inspection techniques and technologies were proposed over the last few years. Traditional manual inspection procedures are facing several limitations and often fall short in ensuring flawlessness. Vision-based strategies for automatic steel surface inspection have emerged as powerful and effective tools to solve various industrial-related problems, including products quality control. Therefore, the current study aims to improve the recognition rate of steel surface defects classification system by introducing a novel classifier combination approach. The proposed system utilizes two distinct feature sets, namely LCCMSP and DCP, which were carefully selected based on a comprehensive comparative study of 19 state-of-the-art texture descriptors, considering both accuracy and time consumption. These generated features are individually fed to two classifiers, SVM and RF, leading to the creation of four base classifiers. In the final step, the Bayesian fusion rule is applied to integrate the outputs of these classifiers, ultimately providing the definitive classification decision. To evaluate the proposed system, two steel surface defects datasets, NEU-CLS and X-SDD, are utilized. The experimental results demonstrate that the proposed combination approach surpasses classical combination methods achieving remarkable outcomes compared to existing steel surface defects classification approaches. This highlights the effectiveness and superiority of the proposed system in accurately identifying and classifying steel surface defects while maintaining fast execution time.

The bearings are the most important mechanical elements of rotating machinery. They are employed to support and rotate the shafts in rotating machinery. An unexpected defect of the bearing may cause significant economic losses. For that reason, the condition monitoring of these bearings has become a fundamental axis of development and industrial research. The focus of this paper is to combine tow conventional methods: Hilbert Transform (HT) and Discrete Wavelet Transform (DWT) to develop a better method for detection and diagnosis the bearing faults. This new method applied on real measurement signals collected from an experimental vibration system. The monitoring results indicate that the proposed method improves the bearing faults diagnosis compared to other common techniques.