Optimization, Learning, and Control for Interdependent Complex Networks (Advances in Intelligent Systems and Computing, 1123)

Preface
Contents
About the Editor
1 Panorama of Optimization, Control, and Learning Algorithms for Interdependent SWEET (Societal, Water, Energy, Economic, and Transportation) Networks
	1.1 Introduction
	1.2 Part I: Theoretical Algorithms for Optimization, Learning, and Data Analytics in Interdependent Complex Networks
		1.2.1 Chapter 2: Promises of Fully Distributed Optimization for IoT-Based Smart City Infrastructures: Theory and Applications
		1.2.2 Chapter 3: Evolutionary Computation, Optimization, and Learning Algorithms for Data Science
		1.2.3 Chapter 4: Applications of Nature-Inspired Algorithms for Dimension Reduction: Enabling Efficient Data Analytics
		1.2.4 Chapter 5: Feature Selection in High-Dimensional Data
		1.2.5 Chapter 6: An Introduction to Advanced Machine Learning: Meta-Learning Algorithms, Applications, and Promises
	1.3 Part II: Application of Optimization, Learning, and Control in Interdependent Complex Networks
		1.3.1 Chapter 7: Predictive Analytics in Future Power Systems: A Panorama and State-of-the-Art of Deep Learning Applications
		1.3.2 Chapter 8: Bilevel Adversary-Operator Cyberattack Framework and Algorithms for Transmission Networks in Smart Grids
		1.3.3 Chapter 9: Toward Operational Resilience of Smart Energy Networks in Complex Infrastructures
		1.3.4 Chapter 10: Control of Cooperative Unmanned Aerial Vehicles: Review of Applications, Challenges, and Algorithms
		1.3.5 Chapter 11: An Optimal Approach for Load-Frequency Control of Islanded Microgrids Based on Non-linear Model
		1.3.6 Chapter 12: PV Design for Smart Cities and Demand Forecasting Using Truncated Conjugate Gradient Algorithm
	References
Part I Theoretical Algorithms for Optimization, Learning, and Data Analytics in Interdependent Complex Networks
	2 Promises of Fully Distributed Optimization for IoT-Based Smart City Infrastructures
		2.1 Introduction
			2.1.1 Motivation
			2.1.2 Related Works
			2.1.3 Contribution
			2.1.4 Organization
		2.2 A Novel Holistic Framework for Interdependent Operation of Power Systems and Electrified Transportation networks
		2.3 Definition of Agents and Their Corresponding Features
			2.3.1 Power System-Specific Agents
			2.3.2 Transportation Network-Specific Agents
			2.3.3 Coupling Agents
		2.4 General Optimization Problem
			2.4.1 Problem Formulation
			2.4.2 Optimality Conditions
		2.5 Consensus+Innovations Based Distributed Algorithm
			2.5.1 Distributed Decision Making: General Distributed Update Rule
			2.5.2 Agent-Based Distributed Algorithm
		2.6 Conclusions
		Appendix 1: Convergence Analysis
		References
	3 Evolutionary Computation, Optimization, and Learning Algorithms for Data Science
		3.1 Introduction
			3.1.1 Overview
			3.1.2 Motivation
			3.1.3 Curse of Dimensionality
			3.1.4 Nature-Inspired Computation
			3.1.5 Nature-Inspired Meta-Heuristic Computation
			3.1.6 Nature-Inspired Evolutionary Computation
				3.1.6.1 Evolutionary-Based Memetic Algorithms
			3.1.7 Organization
		3.2 Feature Extraction Techniques
		3.3 Bio-Inspired Evolutionary Computation
			3.3.1 Overview of Evolutionary Algorithms
			3.3.2 Genetic Algorithm vs. Genetic Programming
				3.3.2.1 Genetic Algorithm
				3.3.2.2 Genetic Programming
			3.3.3 Artificial Bee Colony Algorithm
			3.3.4 Particle Swarm Optimization Algorithm
			3.3.5 Ant Colony Optimization (ACO)
			3.3.6 Grey Wolf Optimizer (GWO)
			3.3.7 Coyote Optimization Algorithm (COA)
			3.3.8 Other Optimization Algorithms
		3.4 Conclusion
		References
	4 Applications of Nature-Inspired Algorithms for Dimension Reduction: Enabling Efficient Data Analytics
		4.1 Introduction
			4.1.1 Overview
			4.1.2 Organization
		4.2 Application of Evolutionary Algorithms
			4.2.1 Feature Extraction Optimization
				4.2.1.1 Feature Selection for Image Classification
			4.2.2 Feature Selection for Network Traffic Classification
			4.2.3 Feature Selection Benchmarks
		4.3 Discussion
		4.4 Conclusion
		References
	5 Feature Selection in High-Dimensional Data
		5.1 Overview
		5.2 Intrinsic Characteristics of High-Dimensional Data
			5.2.1 Large Number of Features
			5.2.2 Small Number of Samples
			5.2.3 Class Imbalance
			5.2.4 Label Noise
			5.2.5 Intrinsic Characteristics of Microarray Data
		5.3 Feature Selection
		5.4 Filter Methods
			5.4.1 Similarity-Based Methods
				5.4.1.1 Relief and ReliefF
				5.4.1.2 Fisher Score
				5.4.1.3 Laplacian Score
			5.4.2 Statistical-Based Methods
				5.4.2.1 Correlation-Based Feature Selection (CFS)
				5.4.2.2 Low Variance
				5.4.2.3 T-Score
				5.4.2.4 Information Theoretical-Based Methods
				5.4.2.5 FCBF
				5.4.2.6 Minimum-Redundancy-Maximum-Relevance (mRMR)
				5.4.2.7 Information Gain
		5.5 Wrapper Methods
			5.5.1 ABACOH and ACO
			5.5.2 PSO
			5.5.3 IBGSA
		5.6 Hybrid Method
		5.7 Embedded Methods
		5.8 Ensemble Techniques
		5.9 Practical Evaluation
			5.9.1 Dataset
			5.9.2 Performance Evaluation Criteria
			5.9.3 Data Normalization
			5.9.4 Analysis of Filter Algorithms
			5.9.5 Analysis of Hybrid-Ensemble Methods
				5.9.5.1 Hybrid-Ensemble 1
				5.9.5.2 Hybrid-Ensemble 2
		5.10 Summary
		References
	6 An Introduction to Advanced Machine Learning: Meta-Learning Algorithms, Applications, and Promises
		6.1 Introduction
		6.2 Machine Learning: Challenges and Drawbacks
		6.3 Meta-Learning Algorithms
			6.3.1 Model-Based MTL
			6.3.2 Metric-Based Learning
			6.3.3 Gradient Decent-Based Learning
		6.4 Promises of Meta-Learning
			6.4.1 Few-Shot Learning
			6.4.2 One-Shot Learning
			6.4.3 Zero-Shot Learning
		6.5 Discussion
		6.6 Conclusion
		References
Part II Application of Optimization, Learning and Control in Interdependent Complex Networks
	7 Predictive Analytics in Future Power Systems: A Panorama and State-Of-The-Art of Deep Learning Applications
		7.1 Introduction
			7.1.1 Motivation
			7.1.2 Classification of Power Systems Forecasting Models
				7.1.2.1 Classification Based on the Domain of Application in Power Systems
				7.1.2.2 Classification Based on Timescale
			7.1.3 Organization of the Chapter
		7.2 Forecasting in Power Systems Using Classical Approaches
			7.2.1 Time Series Data
			7.2.2 Statistical Forecasting Approaches
				7.2.2.1 Naïve Model Approach
				7.2.2.2 Exponential Smoothing
				7.2.2.3 Autoregressive Moving Average (ARMA) Models
				7.2.2.4 Autoregressive Moving Integrated Average (ARIMA) Models
			7.2.3 Machine Learning Forecasting Approaches
				7.2.3.1 Support Vector Regression
				7.2.3.2 Gaussian Process Regression
			7.2.4 Shortcomings of Classical Approaches
		7.3 Forecasting in Power Systems Using Deep Learning
			7.3.1 Deep Learning
				7.3.1.1 Recurrent Neural Network
				7.3.1.2 Long Short-Term Memory Network
				7.3.1.3 Other Relevant Models
			7.3.2 Deep Learning Applications
				7.3.2.1 Load Forecasting
				7.3.2.2 Generation Forecasting
				7.3.2.3 Electricity Price Forecasting and Electric Vehicle Charging
			7.3.3 Deep Learning Strengths and Shortcomings
				7.3.3.1 Strengths
				7.3.3.2 Shortcomings
		7.4 Case Study: Multi-Timescale Solar Irradiance Forecasting Using Deep Learning
			7.4.1 Data
				7.4.1.1 Global Horizontal Irradiance
				7.4.1.2 Exogenous Input Variables
				7.4.1.3 Data Preprocessing and Postprocessing
			7.4.2 Model Architecture and Training
			7.4.3 Results
				7.4.3.1 Single Time Horizon Model
				7.4.3.2 Multi-Time-Horizon Model
		7.5 Summary and Future Work
			7.5.1 Deterministic Versus Probabilistic Forecasting
			7.5.2 Other Potential Applications
		References
	8 Bi-level Adversary-Operator Cyberattack Framework and Algorithms for Transmission Networks in Smart Grids
		8.1 Introduction
			8.1.1 Overview
		8.2 DC Power Flow Model
		8.3 False Data Injection Attacks Based on DC State Estimation
		8.4 Attacker\'s Problem: Finding the Optimal Set of Target Transmission Lines using MILP
			8.4.1 Identifying Feasible Attacks
		8.5 Operator\'s Problem: Bad Data Detection to Prevent Outages Caused by Cyberattack
		8.6 Case Studies
			8.6.1 Feasibility of Line Overflow
			8.6.2 Targeted Attack on Line 15
			8.6.3 Severe Attack on an Area
		8.7 Conclusion
		References
	9 Toward Operational Resilience of Smart Energy Networks in Complex Infrastructures
		9.1 Introduction
			9.1.1 Overview
		9.2 Resilience Enhancement Scheme
		9.3 Real-Time Decision Making Process
		9.4 Optimization Model
			9.4.1 Pre-event Preparation Strategy
			9.4.2 Mid-Event Monitoring
			9.4.3 Post-event Restoration Problem
		9.5 Simulation Results
		9.6 Conclusion
		References
	10 Control of Cooperative Unmanned Aerial Vehicles: Review of Applications, Challenges, and Algorithms
		Abbreviations
		10.1 Introduction
		10.2 Applications and Literature Review
			10.2.1 Search and Rescue
			10.2.2 Surveillance
			10.2.3 Localization and Mapping
			10.2.4 Military Applications
				10.2.4.1 Reconnaissance Strategy
				10.2.4.2 Penetrating Strategy
		10.3 Challenges
		10.4 Algorithms
			10.4.1 Consensus Strategies
				10.4.1.1 Graph Theory Basics in Communication Systems
				10.4.1.2 Consensus Control Theory
				10.4.1.3 Consensus Recent Researches
			10.4.2 Flocking Based Strategies
				10.4.2.1 Flocking Control Theory
				10.4.2.2 Flocking Recent Researches
			10.4.3 Guidance Law Based Cooperative Control
				10.4.3.1 Guidance Law Based Recent Researches
		10.5 Summary and Conclusion
		Bibliography
	11 An Optimal Approach for Load-Frequency Control of Islanded Microgrids Based on Nonlinear Model
		Nomenclature
		11.1 Introduction
		11.2 Dynamic Model of Microgrid
		11.3 The Proposed Intelligent Control Method
		11.4 Simulation and Results
		11.5 Conclusion
		References
	12 Photovoltaic Design for Smart Cities and Demand Forecasting Using a Truncated Conjugate Gradient Algorithm
		Abbreviations
		12.1 Introduction
		12.2 Objectives and Targets
		12.3 Literature Review
		12.4 Rule-Based Neural Network Structure
		12.5 The Proposed Model
		12.6 Results and Discussion
		12.7 Conclusion
		References
Index