UID:
almahu_9949983191102882
Umfang:
1 online resource (348 pages)
Ausgabe:
First edition.
ISBN:
9780443131905
,
0443131902
Inhalt:
This book explores the application of artificial intelligence (AI) in the energy management of hybrid electric vehicles (HEVs). It addresses the growing demand for energy-efficient transportation solutions amidst rising global energy-related greenhouse gas emissions and energy crises. The authors examine various AI-based energy management strategies for HEVs, including fuzzy control, optimization, and reinforcement learning techniques, to enhance the performance of powertrains and reduce energy consumption and emissions. The book is intended for researchers, developers, and students in vehicle engineering, intelligent control, and AI, offering both foundational and advanced materials. It provides insights into the modeling of HEV systems, energy management strategies, and pattern recognition methods, aiming to inspire further research and application in this field.
Anmerkung:
Front Cover -- Application of Artificial Intelligence in Hybrid Electric Vehicle Energy Management -- Application of Artificial Intelligence in Hybrid Electric Vehicle Energy Management -- Copyright -- Contents -- Preface -- Acknowledgments -- 1 - Introduction -- 1. Motivation -- 2. Introduction on hybrid electric vehicles -- 2.1 Definition of hybrid electric vehicles -- 2.2 Classification of hybrid electric vehicles -- 2.2.1 Traditional hybrid electric vehicles -- 2.2.2 New hybrid electric vehicles -- 2.2.2.1 Lithium battery/supercapacitors HEV -- 2.2.2.2 Fuel cells/supercapacitors HEV -- 2.2.2.3 Fuel cells/lithium batteries HEV -- 3. Energy management systems -- 3.1 Rule-based EMSs -- 3.1.1 Deterministic rule based EMSs -- 3.1.2 On/off EMSs -- 3.1.3 Power follower EMSs -- 3.1.4 Fuzzy logic EMSs -- 3.2 Optimization-based EMSs -- 3.2.1 Global optimization EMSs -- 3.2.1.1 Dynamic programming -- 3.2.1.2 Stochastic dynamic programming -- 3.2.1.3 Genetic algorithm -- 3.2.2 Instantaneous optimization strategies -- 3.2.2.1 Equivalent consumption minimization strategy -- 3.2.2.2 Pontryagin's minimum principle -- 3.2.2.3 Model predictive control -- 3.3 Artificial intelligence based EMSs -- 3.3.1 Neural network -- 3.3.2 Support vector regression -- 3.3.3 Reinforcement learning -- 4. Challenges -- 4.1 Degradation modeling -- 4.2 Optimality -- 4.3 Balance between efficiency and performance -- References -- 2 - System modeling of hybrid electric vehicle -- 1. Introduction -- 1.1 Background -- 1.2 Literature review -- 1.2.1 Supercapacitor -- 1.2.2 Battery -- 1.2.3 Fuel cell -- 1.3 Safety considerations and standards -- 1.3.1 Supercapacitors -- 1.3.2 Lithium-ion batteries -- 1.3.3 PEMFCs -- 1.4 Summary -- 2. Hybrid electric vehicle -- 2.1 FC/SC HEV -- 2.2 BAT/SC HEV -- 3. Modeling of power sources -- 3.1 Modeling of fuel cell -- 3.2 Modeling of battery.
,
3.3 Modeling of supercapacitor -- 3.3.1 Circuit model -- 3.3.2 Power loaded supercapacitor -- 3.4 Vehicle speed power model -- 4. Simulation results -- 4.1 Fuel cell -- 4.2 Battery -- 4.3 Supercapacitor -- 4.4 Vehicle speed power results -- 4.5 Details of modeling resources -- 5. Conclusions -- References -- 3 - Neural network modeling for SOH of lithium-ion battery and performance degradation prediction of fuel cell -- 1. Introduction -- 1.1 Background -- 1.2 Battery SOH estimation -- 1.3 PEMFC performance degradation prediction -- 1.4 Contributions -- 2. Transfer learning based CNN for SOH estimation -- 2.1 The architecture of CNN -- 2.2 CNN model constructing -- 2.2.1 CNN training algorithm -- 2.2.2 Evaluation of CNN model -- 2.3 Improvements of CNN -- 2.3.1 Method to prevent overfitting -- 2.3.2 Adjustment strategy of learning rate -- 2.4 Transfer learning method -- 2.5 The process of the CNN-based SOH estimation -- 2.6 Results and discussions -- 2.6.1 Experimental setting -- 2.6.2 Result analysis -- 2.6.3 Performance comparison based on NASA datasets -- 3. NSGA-II-ESN for performance degradation prediction of PEMFC -- 3.1 ESN model -- 3.2 NSGA-II for ESN optimization -- 3.3 Overall modeling process -- 3.4 Result and discussion -- 3.4.1 Data description and preprocessing -- 3.4.2 Performance metrics -- 3.4.3 Materials and resources -- 3.4.4 Simulation experiments -- 4. Conclusions -- References -- 4 - Optimal fuzzy energy management for fuel cell/supercapacitor system using neural network-based driving pattern ... -- 1. Introduction -- 1.1 Background -- 1.2 Problem statement -- 1.3 Literature review -- 1.4 The contribution of method -- 2. Optimal NN driving pattern recognition -- 2.1 Feature extraction -- 2.2 Neural network driving pattern recognition -- 2.3 Feature selection and NN parameter optimization by GA.
,
2.3.1 Hybrid encoding for NN structure and parameters -- 2.3.2 Fitness function -- 2.3.3 Genetic operators for MLPNN -- 2.3.4 Optimization processes of genetic algorithm -- 2.4 Simulation results -- 2.4.1 Driving pattern recognition -- 2.4.2 Comparison and analysis -- 3. Adaptive fuzzy EMS with driving pattern recognition -- 3.1 Adaptive fuzzy energy management controller -- 3.2 Adaptive fuzzy EMS optimization by GA -- 3.2.1 The design of objective function -- 3.2.2 Genetic encoding -- 3.2.3 Genetic operators -- 3.2.4 Optimization processes for adaptive fuzzy EMS -- 3.3 Simulation results -- 3.3.1 Optimal solution of adaptive fuzzy EMS by GA -- 3.3.2 Performances analysis and comparison -- 4. Resources and limitations -- 4.1 Resources -- 4.2 Limitations -- 5. Conclusions -- 5.1 Summary -- 5.2 Further consideration -- References -- 5 - Optimal fuzzy energy management system design based on NSGA-III-SD for battery/supercapacitor HEV -- 1. Introduction -- 1.1 Background -- 1.2 Literature review -- 1.3 The contribution of method -- 2. Problem description -- 2.1 Constraint multiobjective optimization problem -- 2.2 Robust multi-objective optimization problem -- 3. NSGA- III based on similarity and diversity selection -- 3.1 The framework of the proposed NSGA-III -- 3.2 Reference points generation -- 3.3 Association with reference points -- 3.4 Fitness functions evaluation -- 3.5 Evaluation metrics -- 3.6 Main parameters of NSGA-III-SD -- 3.7 Results comparison -- 4. Fuzzy EMS based on NSGA-Ⅲ-SD -- 4.1 Framework of NSGA-Ⅲ-SD optimal fuzzy EMS -- 4.2 Fuzzy EMS design -- 4.3 Encoding of NSGA-Ⅲ-SD -- 4.4 Optimization processes -- 4.5 Fuzzy EMS experiments -- 5. Robust NSGA-III-SD optimal fuzzy EMS -- 5.1 Framework of RNSGA-III-SD fuzzy EMS -- 5.2 Dynamic robust evaluation model based on prediction -- 5.3 Robust fitness function design.
,
5.4 Robust NSGA-III-SD optimization processes -- 5.5 Experimental results -- 6. Conclusions -- References -- 6 - Q learning-based hybrid energy management strategy -- 1. Introduction -- 1.1 Background -- 1.2 Literature review -- 1.3 Motivation and innovation -- 2. Fuzzy energy management system using Q-learning -- 2.1 Framework of fuzzy energy management system utilizing Q-learning -- 2.2 EMS based on traditional Q-learning for HEV -- 2.3 Improved Q-learning-based algorithm -- 2.3.1 Q-table optimization by GA -- 2.3.2 Genetic encoding and operators -- 2.3.3 Offline training of Q-learning algorithm -- 2.3.4 Flow chart of the improved Q-learning EMS -- 2.3.5 Online Q-learning based fuzzy EMS -- 2.4 Simulation results -- 2.4.1 Off-line training -- 2.4.2 Performances comparison -- 3. Q-learning-based EMS with deterministic rule for HEV -- 3.1 QL in HEV energy control -- 3.2 Algorithm design -- 3.3 Real-time energy management -- 3.4 Simulation results -- 3.4.1 Off-line training -- 3.4.2 Real-time application -- 3.4.3 Performances comparison -- 4. Troubleshooting and limitation -- 4.1 Troubleshooting -- 4.1.1 Convergence issues -- 4.1.2 Safety range violation -- 4.1.3 Real-time adaptation -- 4.2 Limitations -- 4.2.1 Training dependency -- 4.2.2 Discretization artifacts -- 4.2.3 Deterministic rule and fuzzy rule rigidity -- 4.2.4 Sensitivity to hyperparameters -- 5. Conclusion -- 5.1 Summary and conclusions -- 5.2 Further information -- References -- 7 - Improved DDPG hybrid energy management strategy based on LSH -- 1. Introduction -- 1.1 Background -- 1.2 Problem statement -- 1.3 Literature review -- 1.4 Safety considerations and standards -- 2. Improved DDPG strategy based on LSH -- 2.1 Method description -- 2.1.1 LSH (SimHash) algorithm introduction -- 2.2 Method procedure -- 2.2.1 Initialize DDPG reinforcement learning module.
,
2.2.2 Initialize LSH counting module -- 2.2.3 Online manage HEV energy system -- 2.3 Materials and resources -- 2.4 Optimization and troubleshooting -- 2.5 Limitations -- 3. Simulation and result analysis -- 3.1 Algorithm comparison experiment -- 3.2 Results -- 4. Discussion and evaluation -- 5. Conclusion -- 5.1 Summary -- 5.2 Further information -- References -- 8 - Further idea on meta EMS for HEV -- 1. Introduction -- 2. Meta-reinforcement learning for EMS -- 2.1 Framework of meta RL EMS -- 2.2 Meta-EMS based on MAML and PPO -- 3. Simulation results -- 3.1 Adaption ability comparison -- 3.2 Control performances comparison -- 4. Conclusion -- References -- Index -- Back Cover.
Weitere Ausg.:
ISBN 9780443131899
Weitere Ausg.:
ISBN 0443131899
Sprache:
Englisch
Bookmarklink
