Mouaad BOUGOFFA

Mr. Mouaad BOUGOFFA

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 April 2025 in Machines
Verified
Journal Article
Open access
PDF file
MBMouaad BOUGOFFA
Mouaad BOUGOFFA
Samir BENMOUSSASamir BENMOUSSA
Samir BENMOUSSA
Mohand DjeziriMohand Djeziri
Mohand Djeziri
OPOlivier Palais
Olivier Palais
Photovoltaic (PV) systems are integral to global renewable energy generation, yet their efficiency and reliability are frequently compromised by undetected faults, leading to significant energy losses, increased maintenance costs, and reduced operational lifespans. To address these challenges, this study proposes a novel hybrid deep learning framework that combines Stacked Sparse Auto-Encoders (SSAE) for autonomous feature extraction with an Optimized-Multi-Layer Perceptron (OMLP) for precise fault classification. The SSAE extracts high-dimensional fault features from raw operational data, while the OMLP leverages these features to classify faults with exceptional accuracy. The model was rigorously validated using real-world PV datasets, encompassing diverse fault types such as partial shading, open circuits, and module degradation under dynamic environmental conditions. Results demonstrate state-of-the-art performance, with the model achieving 99.82% accuracy, 99.7% precision, 99.4% sensitivity, and 100% specificity, outperforming traditional machine learning and deep learning approaches. These findings highlight the framework’s robustness and reliability in real-world applications. By significantly enhancing fault detection accuracy and computational efficiency, the proposed approach optimizes PV system performance, reduces operational costs, and supports sustainable energy production. This study concludes that the hybrid SSAE-Optimized MLP model represents a scalable and efficient solution for improving the reliability and longevity of renewable energy infrastructure, setting a new benchmark for intelligent maintenance strategies in the field.

28 January 2024
Verified
Conference Paper
Closed access
MBMouaad BOUGOFFA
Mouaad BOUGOFFA
BSBenmoussa Samir
Benmoussa Samir
DMDjeziri Mohand
Djeziri Mohand
This paper introduces an advanced fault diagnosis methodology for Solar Photovoltaic (PV) systems, employing Multilayer Perceptron (MLP) neural networks driven by Artificial Intelligence (AI) techniques. The critical task of fault diagnosis is addressed through the effective utilization of historical PV array data. The developed MLP framework exhibits exceptional accuracy in detecting common faults, particularly module degradation, crucial for ensuring sustained system integrity and performance. The research involves the meticulous creation of detailed model parameters, enabling precise simulation of PV system behavior under varying conditions. Leveraging historical data for MLP-based training, the approach adeptly captures diverse system states, facilitating accurate simulations reflective of real-world scenarios. Furthermore, the integration of fault diagnosis techniques enhances the model's efficiency in detecting and classifying system faults, thereby enabling timely maintenance interventions. To underscore its effectiveness, we conducted a comprehensive comparison between our MLP-based method and a prior AI technique. This comparative analysis highlights the superior performance and robustness of our proposed methodology in fault detection and diagnosis. Validation through extensive simulations and real-world data showcases the model's remarkable accuracy in predicting system behavior and diagnosing faults. The findings underscore the significant contribution of this AI-driven MLP methodology to PV system research, providing a robust framework for fault detection and diagnosis essential for ensuring system reliability and optimizing performance.

11 December 2024
Verified
Conference Paper
Closed access
MBMouaad BOUGOFFA
Mouaad BOUGOFFA
SBSamir Benmoussa
Samir Benmoussa
MDMohand Djeziri
Mohand Djeziri
TCThierry Contaret
Thierry Contaret
This study presents a methodology for diagnosing faults in photovoltaic (PV) systems using Convolutional Neural Networks (CNNs). By leveraging CNNs' ability to extract intricate features from PV data, including voltage, current, irradiation, and temperature, the proposed approach enables the accurate identification of various fault types such as short circuits, open circuits, wear, and shading. The contribution of this paper enhances PV system performance and reliability through early fault identification and a comprehensive framework for fault analysis. By effectively distinguishing between normal operating conditions and different fault occurrences, this methodology supports a more sustainable and efficient solar energy future. Through data preprocessing, optimization of the CNN architecture, and incorporation of a multi-class classification layer, the model achieved a high accuracy in fault detection, with an average detection and classification accuracy of 99.7%. The rigorous validation and testing employed in this study affirm the effectiveness and reliability of the CNN-based approach. Overall, this methodology offers a promising strategy for improving fault diagnosis in PV systems, advancing the reliability and efficiency of solar energy applications.

16 August 2024 in IFAC-PapersOnLine
Verified
Journal Article
Closed access
MBMouaad BOUGOFFA
Mouaad BOUGOFFA
SBSamir Benmoussa
Samir Benmoussa
MDMohand Djeziri
Mohand Djeziri
TCThierry Contaret
Thierry Contaret

11 December 2024
Verified
Conference Paper
Closed access
MBMouaad BOUGOFFA
Mouaad BOUGOFFA
SBSamir Benmoussa
Samir Benmoussa
MDMohand Djeziri
Mohand Djeziri
TCThierry Contaret
Thierry Contaret
Wind energy is a key contributor to renewable energy production, with wind turbines playing a critical role in its generation. However, the operational efficiency of wind turbines is often compromised by blade faults, leading to reduced performance, expensive maintenance, and potential safety hazards. Traditional methods of diagnosing faults in wind turbine blades are limited in both accuracy and efficiency. This study proposes a novel fault diagnosis approach using Convolutional Neural Networks (CNNs), a powerful deep learning technique for data analysis. The dataset comprises four sets of vibration signals representing both healthy and faulty wind turbine blades, covering fault conditions such as blade cracks, surface degradation, imbalance, and blade twist. Our proposed CNN-based method aims to accurately detect and classify these faults, enabling proactive maintenance and improving the overall reliability and efficiency of wind energy systems.