Search

Electric vehicles have been well recognized because of their contribution to the promising future of emission-free transportation. The core of electric vehicles is the Li-ion battery storage system, which plays an important role in the safety and price of these vehicles. Therefore, it is necessary to develop an effective battery management system in the field of vehicle electrification. In the management system, real-time access to state of charge and state of health information is crucial, although these states are not directly measurable. Therefore, they are solely obtained by estimation, which is based on a battery model and three measurable parameters, namely, the battery’s voltage, current, and temperature. There are many challenges in conducting estimations of the battery’s states due to both internal and external factors, such as load, temperature, and aging. Various advanced methods have been proposed and applied to cope with these difficulties. There is, however, still a conflict between the simplicity and the accuracy of the reported estimation methods. Within the scope of this thesis, a comprehensive estimation approach for both the state of charge and the state of health is proposed. This approach has been developed based on experimental results, which take into account three actual crucial factors, namely, dynamic load, variable temperature, and aging. The estimation procedure is based on multiple adaptive forgetting factors recursive least-squares approach, the correlation of the ohmic resistance to the battery capacity, and a model for the relationship of the open circuit voltage to the state of charge, the temperature, and the state of health. The accuracy and robustness of the developed estimation approach have been validated through various experiments under diverse conditions, including harsh ones. In addition to its low-level complexity, the developed approach is implementable in actual application.

Electric vehicles have been well recognized because of their contribution to the promising future of emission-free transportation. The core of electric vehicles is the Li-ion battery storage system, which plays an important role in the safety and price of these vehicles. Therefore, it is necessary to develop an effective battery management system in the field of vehicle electrification. In the management system, real-time access to state of charge and state of health information is crucial, although these states are not directly measurable. Therefore, they are solely obtained by estimation, which is based on a battery model and three measurable parameters, namely, the battery’s voltage, current, and temperature. There are many challenges in conducting estimations of the battery’s states due to both internal and external factors, such as load, temperature, and aging. Various advanced methods have been proposed and applied to cope with these difficulties. There is, however, still a conflict between the simplicity and the accuracy of the reported estimation methods. Within the scope of this thesis, a comprehensive estimation approach for both the state of charge and the state of health is proposed. This approach has been developed based on experimental results, which take into account three actual crucial factors, namely, dynamic load, variable temperature, and aging. The estimation procedure is based on multiple adaptive forgetting factors recursive least-squares approach, the correlation of the ohmic resistance to the battery capacity, and a model for the relationship of the open circuit voltage to the state of charge, the temperature, and the state of health. The accuracy and robustness of the developed estimation approach have been validated through various experiments under diverse conditions, including harsh ones. In addition to its low-level complexity, the developed approach is implementable in actual application.

Electric vehicles have been well recognized because of their contribution to the promising future of emission-free transportation. The core of electric vehicles is the Li-ion battery storage system, which plays an important role in the safety and price of these vehicles. Therefore, it is necessary to develop an effective battery management system in the field of vehicle electrification. In the management system, real-time access to state of charge and state of health information is crucial, although these states are not directly measurable. Therefore, they are solely obtained by estimation, which is based on a battery model and three measurable parameters, namely, the battery’s voltage, current, and temperature. There are many challenges in conducting estimations of the battery’s states due to both internal and external factors, such as load, temperature, and aging. Various advanced methods have been proposed and applied to cope with these difficulties. There is, however, still a conflict between the simplicity and the accuracy of the reported estimation methods. Within the scope of this thesis, a comprehensive estimation approach for both the state of charge and the state of health is proposed. This approach has been developed based on experimental results, which take into account three actual crucial factors, namely, dynamic load, variable temperature, and aging. The estimation procedure is based on multiple adaptive forgetting factors recursive least-squares approach, the correlation of the ohmic resistance to the battery capacity, and a model for the relationship of the open circuit voltage to the state of charge, the temperature, and the state of health. The accuracy and robustness of the developed estimation approach have been validated through various experiments under diverse conditions, including harsh ones. In addition to its low-level complexity, the developed approach is implementable in actual application.

Electric vehicles have been well recognized because of their contribution to the promising future of emission-free transportation. The core of electric vehicles is the Li-ion battery storage system, which plays an important role in the safety and price of these vehicles. Therefore, it is necessary to develop an effective battery management system in the field of vehicle electrification. In the management system, real-time access to state of charge and state of health information is crucial, although these states are not directly measurable. Therefore, they are solely obtained by estimation, which is based on a battery model and three measurable parameters, namely, the battery’s voltage, current, and temperature. There are many challenges in conducting estimations of the battery’s states due to both internal and external factors, such as load, temperature, and aging. Various advanced methods have been proposed and applied to cope with these difficulties. There is, however, still a conflict between the simplicity and the accuracy of the reported estimation methods. Within the scope of this thesis, a comprehensive estimation approach for both the state of charge and the state of health is proposed. This approach has been developed based on experimental results, which take into account three actual crucial factors, namely, dynamic load, variable temperature, and aging. The estimation procedure is based on multiple adaptive forgetting factors recursive least-squares approach, the correlation of the ohmic resistance to the battery capacity, and a model for the relationship of the open circuit voltage to the state of charge, the temperature, and the state of health. The accuracy and robustness of the developed estimation approach have been validated through various experiments under diverse conditions, including harsh ones. In addition to its low-level complexity, the developed approach is implementable in actual application.

Electric vehicles have been well recognized because of their contribution to the promising future of emission-free transportation. The core of electric vehicles is the Li-ion battery storage system, which plays an important role in the safety and price of these vehicles. Therefore, it is necessary to develop an effective battery management system in the field of vehicle electrification. In the management system, real-time access to state of charge and state of health information is crucial, although these states are not directly measurable. Therefore, they are solely obtained by estimation, which is based on a battery model and three measurable parameters, namely, the battery’s voltage, current, and temperature. There are many challenges in conducting estimations of the battery’s states due to both internal and external factors, such as load, temperature, and aging. Various advanced methods have been proposed and applied to cope with these difficulties. There is, however, still a conflict between the simplicity and the accuracy of the reported estimation methods. Within the scope of this thesis, a comprehensive estimation approach for both the state of charge and the state of health is proposed. This approach has been developed based on experimental results, which take into account three actual crucial factors, namely, dynamic load, variable temperature, and aging. The estimation procedure is based on multiple adaptive forgetting factors recursive least-squares approach, the correlation of the ohmic resistance to the battery capacity, and a model for the relationship of the open circuit voltage to the state of charge, the temperature, and the state of health. The accuracy and robustness of the developed estimation approach have been validated through various experiments under diverse conditions, including harsh ones. In addition to its low-level complexity, the developed approach is implementable in actual application.

Electric vehicles have been well recognized because of their contribution to the promising future of emission-free transportation. The core of electric vehicles is the Li-ion battery storage system, which plays an important role in the safety and price of these vehicles. Therefore, it is necessary to develop an effective battery management system in the field of vehicle electrification. In the management system, real-time access to state of charge and state of health information is crucial, although these states are not directly measurable. Therefore, they are solely obtained by estimation, which is based on a battery model and three measurable parameters, namely, the battery’s voltage, current, and temperature. There are many challenges in conducting estimations of the battery’s states due to both internal and external factors, such as load, temperature, and aging. Various advanced methods have been proposed and applied to cope with these difficulties. There is, however, still a conflict between the simplicity and the accuracy of the reported estimation methods. Within the scope of this thesis, a comprehensive estimation approach for both the state of charge and the state of health is proposed. This approach has been developed based on experimental results, which take into account three actual crucial factors, namely, dynamic load, variable temperature, and aging. The estimation procedure is based on multiple adaptive forgetting factors recursive least-squares approach, the correlation of the ohmic resistance to the battery capacity, and a model for the relationship of the open circuit voltage to the state of charge, the temperature, and the state of health. The accuracy and robustness of the developed estimation approach have been validated through various experiments under diverse conditions, including harsh ones. In addition to its low-level complexity, the developed approach is implementable in actual application.

Electric vehicles have been well recognized because of their contribution to the promising future of emission-free transportation. The core of electric vehicles is the Li-ion battery storage system, which plays an important role in the safety and price of these vehicles. Therefore, it is necessary to develop an effective battery management system in the field of vehicle electrification. In the management system, real-time access to state of charge and state of health information is crucial, although these states are not directly measurable. Therefore, they are solely obtained by estimation, which is based on a battery model and three measurable parameters, namely, the battery’s voltage, current, and temperature. There are many challenges in conducting estimations of the battery’s states due to both internal and external factors, such as load, temperature, and aging. Various advanced methods have been proposed and applied to cope with these difficulties. There is, however, still a conflict between the simplicity and the accuracy of the reported estimation methods. Within the scope of this thesis, a comprehensive estimation approach for both the state of charge and the state of health is proposed. This approach has been developed based on experimental results, which take into account three actual crucial factors, namely, dynamic load, variable temperature, and aging. The estimation procedure is based on multiple adaptive forgetting factors recursive least-squares approach, the correlation of the ohmic resistance to the battery capacity, and a model for the relationship of the open circuit voltage to the state of charge, the temperature, and the state of health. The accuracy and robustness of the developed estimation approach have been validated through various experiments under diverse conditions, including harsh ones. In addition to its low-level complexity, the developed approach is implementable in actual application.