Federated Learning-Based Anomaly Detection Framework for Enhancing Security in Smart Grid Environments / Chaima Talhi

Public ISBD Titre : Federated Learning-Based Anomaly Detection Framework for Enhancing Security in Smart Grid Environments Type de document : document électronique Auteurs : Chaima Talhi ; Nesrine Dehli, Auteur ; Goudjil, Lakhdar, Directeur de thèse Editeur : Setif:UFA Année de publication : 2025 Importance : 1 vol (87 f .) Format : 29 cm Langues : Anglais (eng) Catégories : Thèses & Mémoires:Informatique Mots-clés : Smart Grid Anomaly Detection Federated Learning Machine Learning Deep Learning Index. décimale : 004 Informatique Résumé : The growing integration of smart grids into modern energy infrastructures presents both unprecedented opportunities for intelligent power management and serious concerns regarding data privacy and cyber-security. Traditional anomaly detection methods, although effective, often rely on centralized data collection, thereby increasing the risk of exposing sensitive user information. To overcome these limitations, this thesis introduces FED-XID, a novel Federated Learning-based framework for privacy-preserving anomaly detection in smart grid systems. The framework enables decentralized model training using the XGBoost algorithm and incorporates embedded Intrusion Detection Systems (IDS) at the edge level, ensuring localized monitoring while safeguarding user data confidentiality. In addition, advanced deep learning techniques based on Temporal Convolutional Networks (TCN) and Long Short-Term Memory (LSTM) are applied to handle missing or incomplete smart meter data. The proposed model is both robust and efficient, leveraging a hybrid edge–cloud architecture and a lightweight classification core to ensure high performance, low latency, and scalable deployment in real-world smart grid environments. FED-XID achieved an AUC of 93.60 and a training time of only 25.97 seconds, demonstrating strong detection capability and computational efficiency. Note de contenu : Sommaire Abbreviations 1 General Introduction 4 1 Smart Grids: A Technological Shift Toward Sustainable Energy 7 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.2 Conceptual Foundations and Architecture of Smart Grids . . . . . . . . . . 8 1.2.1 Evolution and Definition of Smart Grids . . . . . . . . . . . . . . . 8 1.2.2 Smart Grids vs Legacy Power Systems . . . . . . . . . . . . . . . . 8 1.2.3 Smart Grid Architecture . . . . . . . . . . . . . . . . . . . . . . . . 9 1.2.4 Key Components of Smart Grids : . . . . . . . . . . . . . . . . . . 14 1.3 Key Challenges in Smart Grid Management . . . . . . . . . . . . . . . . . 16 1.3.1 Demand-Response . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 1.3.2 Load balancing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 1.3.3 Fault Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 1.3.4 Cyber-security Threats: . . . . . . . . . . . . . . . . . . . . . . . . 18 1.4 Role of Edge Computing in Smart Grids . . . . . . . . . . . . . . . . . . . 20 1.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2 Artificial Intelligence Techniques in Smart Grids 23 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.2 Machine Learning (ML) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.2.1 Definition and Core Principles . . . . . . . . . . . . . . . . . . . . . 24 2.2.2 Machine Learning Techniques: . . . . . . . . . . . . . . . . . . . . . 24 2.2.3 Applications of Machine Learning in smart grid . . . . . . . . . . . 26 2.3 Deep Learning (DL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.3.1 Definition and Core Principles . . . . . . . . . . . . . . . . . . . . . 27 2.3.2 Enhancing Smart Grid Security . . . . . . . . . . . . . . . . . . . . 29 2.4 Federated Learning (FL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.4.1 Overview of Federated Learning (FL) . . . . . . . . . . . . . . . . . 32 2.4.2 Federated Learning vs Traditional Machine Learning . . . . . . . . 32 2.4.3 Key Components of Federated Learning . . . . . . . . . . . . . . . 34 2.4.4 Types of Federated Learning . . . . . . . . . . . . . . . . . . . . . . 36 2.4.5 Applications of Federated Learning in Smart Grids . . . . . . . . . 37 2.4.6 Key FL Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.4.7 Benefits of Federated Learning in Smart Grids . . . . . . . . . . . . 39 2.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 3 Literature Review: Recent research in Federated Learning-Based Anomaly Detection 42 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 4 The Proposed FED-XID Approach 50 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 4.2 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.3 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.4 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.5 Overview of the Proposed FED-XID Approach . . . . . . . . . . . . . . . 52 4.5.1 Step-by-Step Workflow of the FED-XID . . . . . . . . . . . . . . . 54 4.6 Predictive Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 4.6.1 XGBoost-Based Local Anomaly Detection on Edge Devices . . . . . 57 4.7 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4.7.1 Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4.7.2 Data Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 4.7.3 Data Pre-processing . . . . . . . . . . . . . . . . . . . . . . . . . . 59 4.7.4 Assessing Model Performance . . . . . . . . . . . . . . . . . . . . . 65 4.8 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.8.1 Centralized Versus Decentralized XGBoost Classifier . . . . . . . . 70 4.8.2 Comparison with Other State-of-the-Art . . . . . . . . . . . . . . . 70 4.8.3 Comparison with Centralized FVC model . . . . . . . . . . . . . . 71 4.8.4 Discussion and Challenges Encountered . . . . . . . . . . . . . . . . 71 4.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 General Conclusion and Future Work 74 A Implementation Stack : Language, Libraries, and FL Tool 84 Côte titre : MAI/1010 Federated Learning-Based Anomaly Detection Framework for Enhancing Security in Smart Grid Environments [document électronique] / Chaima Talhi ; Nesrine Dehli, Auteur ; Goudjil, Lakhdar, Directeur de thèse . - [S.l.] : Setif:UFA, 2025 . - 1 vol (87 f .) ; 29 cm. Langues : Anglais (eng) Catégories : Thèses & Mémoires:Informatique Mots-clés : Smart Grid Anomaly Detection Federated Learning Machine Learning Deep Learning Index. décimale : 004 Informatique Résumé : The growing integration of smart grids into modern energy infrastructures presents both unprecedented opportunities for intelligent power management and serious concerns regarding data privacy and cyber-security. Traditional anomaly detection methods, although effective, often rely on centralized data collection, thereby increasing the risk of exposing sensitive user information. To overcome these limitations, this thesis introduces FED-XID, a novel Federated Learning-based framework for privacy-preserving anomaly detection in smart grid systems. The framework enables decentralized model training using the XGBoost algorithm and incorporates embedded Intrusion Detection Systems (IDS) at the edge level, ensuring localized monitoring while safeguarding user data confidentiality. In addition, advanced deep learning techniques based on Temporal Convolutional Networks (TCN) and Long Short-Term Memory (LSTM) are applied to handle missing or incomplete smart meter data. The proposed model is both robust and efficient, leveraging a hybrid edge–cloud architecture and a lightweight classification core to ensure high performance, low latency, and scalable deployment in real-world smart grid environments. FED-XID achieved an AUC of 93.60 and a training time of only 25.97 seconds, demonstrating strong detection capability and computational efficiency. Note de contenu : Sommaire Abbreviations 1 General Introduction 4 1 Smart Grids: A Technological Shift Toward Sustainable Energy 7 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.2 Conceptual Foundations and Architecture of Smart Grids . . . . . . . . . . 8 1.2.1 Evolution and Definition of Smart Grids . . . . . . . . . . . . . . . 8 1.2.2 Smart Grids vs Legacy Power Systems . . . . . . . . . . . . . . . . 8 1.2.3 Smart Grid Architecture . . . . . . . . . . . . . . . . . . . . . . . . 9 1.2.4 Key Components of Smart Grids : . . . . . . . . . . . . . . . . . . 14 1.3 Key Challenges in Smart Grid Management . . . . . . . . . . . . . . . . . 16 1.3.1 Demand-Response . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 1.3.2 Load balancing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 1.3.3 Fault Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 1.3.4 Cyber-security Threats: . . . . . . . . . . . . . . . . . . . . . . . . 18 1.4 Role of Edge Computing in Smart Grids . . . . . . . . . . . . . . . . . . . 20 1.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2 Artificial Intelligence Techniques in Smart Grids 23 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.2 Machine Learning (ML) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.2.1 Definition and Core Principles . . . . . . . . . . . . . . . . . . . . . 24 2.2.2 Machine Learning Techniques: . . . . . . . . . . . . . . . . . . . . . 24 2.2.3 Applications of Machine Learning in smart grid . . . . . . . . . . . 26 2.3 Deep Learning (DL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.3.1 Definition and Core Principles . . . . . . . . . . . . . . . . . . . . . 27 2.3.2 Enhancing Smart Grid Security . . . . . . . . . . . . . . . . . . . . 29 2.4 Federated Learning (FL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.4.1 Overview of Federated Learning (FL) . . . . . . . . . . . . . . . . . 32 2.4.2 Federated Learning vs Traditional Machine Learning . . . . . . . . 32 2.4.3 Key Components of Federated Learning . . . . . . . . . . . . . . . 34 2.4.4 Types of Federated Learning . . . . . . . . . . . . . . . . . . . . . . 36 2.4.5 Applications of Federated Learning in Smart Grids . . . . . . . . . 37 2.4.6 Key FL Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.4.7 Benefits of Federated Learning in Smart Grids . . . . . . . . . . . . 39 2.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 3 Literature Review: Recent research in Federated Learning-Based Anomaly Detection 42 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 4 The Proposed FED-XID Approach 50 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 4.2 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.3 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.4 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.5 Overview of the Proposed FED-XID Approach . . . . . . . . . . . . . . . 52 4.5.1 Step-by-Step Workflow of the FED-XID . . . . . . . . . . . . . . . 54 4.6 Predictive Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 4.6.1 XGBoost-Based Local Anomaly Detection on Edge Devices . . . . . 57 4.7 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4.7.1 Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4.7.2 Data Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 4.7.3 Data Pre-processing . . . . . . . . . . . . . . . . . . . . . . . . . . 59 4.7.4 Assessing Model Performance . . . . . . . . . . . . . . . . . . . . . 65 4.8 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.8.1 Centralized Versus Decentralized XGBoost Classifier . . . . . . . . 70 4.8.2 Comparison with Other State-of-the-Art . . . . . . . . . . . . . . . 70 4.8.3 Comparison with Centralized FVC model . . . . . . . . . . . . . . 71 4.8.4 Discussion and Challenges Encountered . . . . . . . . . . . . . . . . 71 4.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 General Conclusion and Future Work 74 A Implementation Stack : Language, Libraries, and FL Tool 84 Côte titre : MAI/1010

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MAI/1010	MAI/1010	Mémoire	Bibliothèque des sciences	Anglais	Disponible Disponible

---

1 (1 - 1 / 1)