Artificial Intelligence and Data Driven Optimization of Internal Combustion Engines

c4e4c6d9_Cover
Copyrig_2022_Artificial-Intelligence-and-Data-Driven-Optimization-of-Interna
	Copyright
Contents
Contribut_2022_Artificial-Intelligence-and-Data-Driven-Optimization-of-Inter
	Contributors
Forewor_2022_Artificial-Intelligence-and-Data-Driven-Optimization-of-Interna
	Foreword
Prefac_2022_Artificial-Intelligence-and-Data-Driven-Optimization-of-Internal
	Preface
Chapter-1---Intr_2022_Artificial-Intelligence-and-Data-Driven-Optimization-o
	1 . Introduction
		1. Industrial revolution
		2. Artificial intelligence, machine learning, and deep learning
		3. Machine learning algorithms
		4. Artificial intelligence-based fuel-engine co-optimization
			4.1 Optimization of internal combustion engine
				4.1.1 Design of experiments
				4.1.2 Genetic algorithm
				4.1.3 Machine learning-based algorithms
			4.2 Optimization of fuel formulation
			4.3 Mitigation of rare combustion events
		5. Summary
		References
Chapter-2---Optimization-of-fuel-formu_2022_Artificial-Intelligence-and-Data
	2 . Optimization of fuel formulation using adaptive learning and artificial intelligence
		1. Introduction and motivation
		2. Mixed-mode combustion and fuel performance metrics
		3. A neural network model to predict fuel research octane numbers
		4. Optimization problem formulation and description of solution approaches
			4.1 Constrained optimization formulation
			4.2 Genetic algorithm
			4.3 Gaussian process–based surrogate model optimization algorithm
		5. Numerical experiments and results
		6. Discussion
		7. Summary and concluding remarks
		Acknowledgments
		References
Chapter-3---Artificial-intel_2022_Artificial-Intelligence-and-Data-Driven-Op
	3 . Artificial intelligence–enabled fuel design
		1. Transportation fuels
			1.1 Fuel representation
			1.2 Fuel formulation workflow
			1.3 Artificial intelligence modeling approaches
		2. Application of artificial intelligence to fuel formulation
			2.1 High throughput screening: finding a needle in the haystack
			2.2 Fuel property prediction by machine learning models
			2.3 Reaction discovery
			2.4 Fuel-engine co-optimization
		3. Conclusions and perspectives
		Acknowledgments
		References
Chapter-4---Engine-optimization-using_2022_Artificial-Intelligence-and-Data-
	4 . Engine optimization using computational fluid dynamics and genetic algorithms
		1. Introduction
		2. Modeling framework and acceleration strategies
			2.1 Computational fluid dynamics acceleration techniques
				2.1.1 Adaptive mesh refinement
				2.1.2 Detailed chemistry acceleration strategies
			2.2 Engine geometry generation
				2.2.1 Method of splines
				2.2.2 Method of forces
			2.3 Virtual injection model
		3. Optimization methods
			3.1 Fundamentals of genetic algorithms
			3.2 Pioneering investigations
			3.3 Multiobjective framework
			3.4 Convergence acceleration
		4. Summary and concluding remarks
		References
Chapter-5---Computational-fluid-dynamics-g_2022_Artificial-Intelligence-and-
	5 . Computational fluid dynamics–guided engine combustion system design optimization using design of experiments
		1. Introduction
		2. Methodologies
			2.1 Design space construction
			2.2 Response surface model formulation
			2.3 Model-based design optimization and verification
		3. A recent application
			3.1 Engine and fuel specifications
			3.2 Computational fluid dynamic model setup and validation
			3.3 Design variables
			3.4 Objective variables and evaluation method
			3.5 Data fitting and optimization
		4. Recommendations for best practice
			4.1 Adequate computational fluid dynamic model validation
			4.2 Efficient geometry and mesh manipulation
			4.3 Sample size
			4.4 Optimization across full engine operation range
			4.5 Computational efficiency
		5. Conclusions and perspectives
		Acknowledgments
		References
Chapter-6---A-machine-learning-genetic-al_2022_Artificial-Intelligence-and-D
	6 . A machine learning-genetic algorithm approach for rapid optimization of internal combustion engines
		1. Introduction
		2. Engine optimization problem setup
		3. Training and data examination
		4. Machine learning-genetic algorithm approach
			4.1 Optimization methodology
			4.2 Repeatability of machine learning-genetic algorithm
				4.2.1 Extension of variable domain
			4.3 Postprocessing and robustness
		5. Automated machine learning-genetic algorithm
			5.1 Hyperparameter selection
				5.1.1 Manual selection
				5.1.2 Automated strategies for selecting hyperparameters
			5.2 Problem setup
			5.3 Results
		6. Summary
		Acknowledgments
		References
Chapter-7---Machine-learning-driven-seq_2022_Artificial-Intelligence-and-Dat
	7 . Machine learning–driven sequential optimization using dynamic exploration and exploitation
		1. Introduction
		2. Active ML optimization (ActivO)
			2.1 Basic algorithm
			2.2 Query strategies
			2.3 Convergence criteria
			2.4 Dynamic exploration and exploitation
		3. Case study 1: two-dimensional cosine mixture function
		4. Case study 2: computational fluid dynamics (CFD)-based engine optimization
		5. Conclusions
		Acknowledgments
		References
Chapter-8---Artificial-intelligence-based-_2022_Artificial-Intelligence-and-
	8 . Artificial-intelligence-based prediction and control of combustion instabilities in spark-ignition engines
		1. Introduction
			1.1 Artificial intelligence applications to engine controls
			1.2 Dilute combustion instability background
		2. Case study: artificial-intelligence-enhanced modeling of dilute spark-ignition cycle-to-cycle variability
		3. Case study: neural networks for combustion stability control
			3.1 Artificial neural networks
			3.2 Spiking neural networks
		4. Case study: learning reference governor for model-free dilute limit identification and avoidance
			4.1 Constrained combustion phasing control problem
			4.2 Learning reference governor for avoiding misfire events
		5. Summary
		References
Chapter-9---Using-deep-learning-to-di_2022_Artificial-Intelligence-and-Data-
	9 . Using deep learning to diagnose preignition in turbocharged spark-ignited engines
		1. Introduction
			1.1 Fault detection
			1.2 Optimization and control
			1.3 Predicting combustion parameters (phasing and cycle-to-cycle variation) and emissions
		2. Preignition detection using machine learning algorithm
			2.1 Feed forward multilayer neural networks
			2.2 Convolutional neural networks
			2.3 Recurrent neural networks
		3. Activation functions
		4. Experiments and data extraction
		5. Machine learning methodology
		6. Model 1: Input from principal component analysis
		7. Model 2: Time series input
		8. Model metrics
		9. Results and discussion
			9.1 Training and validation losses
		10. Conclusions
		References
		Further reading
Inde_2022_Artificial-Intelligence-and-Data-Driven-Optimization-of-Internal-C
	Index
		A
		B
		C
		D
		E
		F
		G
		H
		I
		K
		L
		M
		N
		O
		P
		Q
		R
		S
		T
		U
		V
		Z