Handbook of Artificial Intelligence Techniques in Photovoltaic Systems: Modeling, Control, Optimization, Forecasting and Fault Diagnosis

Front Cover
Handbook of Artificial Intelligence Techniques in Photovoltaic Systems: Modelling, Control, Optimization, Forecasting and  ...
Copyright
Contents
Authors biographies
Preface
Acknowledgments
Chapter 1: Solar radiation and photovoltaic systems: Modeling and simulation
	1.1. Introduction
	1.2. Solar radiation
	1.3. Photovoltaics
		1.3.1. Photovoltaic effect
		1.3.2. Photovoltaic technologies
			1.3.2.1. Crystalline silicon
			1.3.2.2. Thin films
			1.3.2.3. Emerging technologies
		1.3.3. Modeling and simulation
			1.3.3.1. Solar cell modeling
			1.3.3.2. PV module modeling
			1.3.3.3. PV array modeling under Matlab-Simscape
		1.3.4. Photovoltaic systems
			1.3.4.1. Stand-alone (off-grid) PV systems
			1.3.4.2. Grid-connected PV systems
			1.3.4.3. Hybrid photovoltaic systems
	1.4. Main issues of photovoltaic systems
	1.5. Summary
	References
Chapter 2: Artificial intelligence techniques: Machine learning and deep learning algorithms
	2.1. Introduction
	2.2. Artificial intelligence
	2.3. Machine learning
		2.3.1. Supervised learning
			2.3.1.1. Linear regression
			2.3.1.2. Logistic regression
			2.3.1.3. Support vector machine
			2.3.1.4. K-Nearest neighbors
			2.3.1.5. Decision trees
			2.3.1.6. Random forest
			2.3.1.7. Naïve Bayes
			2.3.1.8. Neural networks
		2.3.2. Unsupervised learning
			2.3.2.1. K-means
			2.3.2.2. Fuzzy c-means
			2.3.2.3. Hierarchical clustering
			2.3.2.4. Principal component analysis
		2.3.3. Reinforcement learning
	2.4. Ensemble learning
		2.4.1. Bagging
		2.4.2. Boosting
		2.4.3. Stacking
	2.5. Deep learning
		2.5.1. Convolutional neural networks
		2.5.2. Long short-term memory
		2.5.3. Gated recurrent unit
		2.5.4. Bidirectional LSTM and GRU
		2.5.5. Generative adversarial network
	2.6. Advantages and disadvantages of ML, EL, and DL algorithms
	2.7. Summary
	References
Chapter 3: Forecasting of solar radiation using machine learning and deep learning algorithms
	3.1. Introduction
	3.2. Solar radiation forecasting based on data-driven methods
	3.3. Datasets description and preparation
		3.3.1. Dataset 3.1
		3.3.2. Dataset 3.2
		3.3.3. Dataset 3.3
		3.3.4. Dataset 3.4
	3.4. Application of machine learning and deep learning algorithms for forecasting of solar radiation
		3.4.1. Machine learning for forecasting of daily global horizontal irradiation (GHI)
		3.4.2. Application of deep learning algorithms for in-plane solar irradiance forecasting
		3.4.3. Multistep ahead forecasting of 5min in-plane solar irradiance using stacked LSTM
		3.4.4. Application of neural networks for solar radiation forecasting from meteorological parameters
	3.5. Summary
	References
Chapter 4: Forecasting of photovoltaic output power using machine learning and deep learning algorithms
	4.1. Introduction
	4.2. Forecasting of photovoltaic power based on data-driven methods
	4.3. Datasets description and preparation
		4.3.1. Dataset 4.1
		4.3.2. Dataset 4.2
		4.3.3. Dataset 4.3
		4.3.4. Dataset 4.4
	4.4. Machine learning for forecasting of photovoltaic output power
	4.5. Neural networks for forecasting of photovoltaic output power from meteorological parameters and historical power
	4.6. Ensemble learning for forecasting of photovoltaic power for one-step ahead
	4.7. Deep learning for forecasting of photovoltaic power (one-step and multistep ahead)
		4.7.1. One-step ahead forecasting
		4.7.2. Multistep ahead forecasting
	4.8. Uncertainty quantification and interval prediction
		4.8.1. Uncertainty quantification
		4.8.2. Interval prediction
	4.9. Summary
	References
Chapter 5: Optimization of photovoltaic systems based on artificial intelligence techniques
	5.1. Introduction
	5.2. Maximum power point tracking methods
	5.3. Photovoltaic array reconfiguration methods
	5.4. Maximum power point tracking using artificial intelligence techniques
		5.4.1. Tracking the maximum power point under uniform conditions using AI techniques
		5.4.2. Tracking the maximum power point under non-uniform conditions using hybrid methods
		5.4.3. Tracking the MPP under PSCs using evolutionary algorithms
	5.5. PV module reconfiguration based on dynamic techniques and AI techniques
	5.6. Summary
	References
Chapter 6: Machine learning and deep learning algorithms for fault diagnosis of photovoltaic systems
	6.1. Introduction
	6.2. Type of faults in photovoltaic arrays
	6.3. Protection devices for photovoltaic systems
	6.4. Fault detection and diagnosis methods
		6.4.1. Visual inspection and thermography methods
		6.4.2. Electrical methods
	6.5. Datasets description and preparation
		6.5.1. Dataset 6.1
		6.5.2. Dataset 6.2
		6.5.3. Dataset 6.3
	6.6. Feature selection and extraction
		6.6.1. Recursive feature selection
		6.6.2. Principal component analysis
	6.7. Fault detection in photovoltaic arrays
		6.7.1. Fault detection based on analytical models
		6.7.2. Fault detection using machine learning algorithms
		6.7.3. Fault detection using deep learning algorithms
	6.8. Fault classification of photovoltaic arrays
		6.8.1. Fault classification using machine learning algorithms
		6.8.2. Fault classification using ensemble learning algorithms
		6.8.3. Fault classification using deep learning algorithms
	6.9. Summary
	References
Chapter 7: Control and optimal management of grid-connected photovoltaic systems and micro-grids using artificial inte
	7.1. Introduction
	7.2. Control of grid-connected and hybrid photovoltaic systems
	7.3. Application of AI techniques for management and control of hybrid micro-grids
	7.4. Energy management of micro-grids
		7.4.1. Micro-grid with an electrical vehicle station
		7.4.2. Dataset 7.1
		7.4.3. Energy management system based on artificial neural networks
	7.5. Power sharing in micro-grids
		7.5.1. Virtual impedance for reactive power sharing
		7.5.2. Reactive power sharing analysis for meshed microgrid
		7.5.3. Virtual impedance adaptation using ant colony optimization algorithm
	7.6. Summary
	References
Chapter 8: Internet of things (IoT) and embedded systems for photovoltaic systems
	8.1. Introduction
	8.2. Programmable electronic boards and IDE
		8.2.1. Arduino
		8.2.2. Raspberry Pi
		8.2.3. Field programmable gate arrays
	8.3. Internet of things (IoT)
	8.4. Literature review of the application of FPGA, Arduino, and Raspberry Pi for photovoltaic systems
	8.5. Real-time applications
		8.5.1. Smart PV monitoring system and fault diagnosis based on the IoT technique and machine learning algorithms
		8.5.2. Implementation of intelligent off-grid photovoltaic system simulator using an FPGA board
		8.5.3. Implementation of an intelligent MPPT algorithms using an FPGA board
		8.5.4. Co-simulation of an intelligent fault diagnosis of a photovoltaic array using XSG
		8.5.5. Implementation of PV fault diagnosis techniques using thermography images and Raspberry Pi 4
	8.6. Summary
	References
Appendices
	Appendix A: PV module modeling
	Appendix B: Machine learning and deep learning functions
	Appendix C: Error metrics
	Appendix D: Confusion matrix
	Appendix E: Evaluation error metrics in Python
	Appendix F: K-fold cross validation
	Appendix G
		G.1. Arduino code for reading sensor connected to port A0
	Appendix H: XSG models
		H.1. I-V model block based on XSG
		H.2. Extraction parameters block on XSG
		H.3. NN model block based on XSG
	Appendix I
		I.1. Generate SMS and sending email
		I.2. Sending email
Index
Back Cover