Abstract
State of charge (SOC) and state of health (SOH) are key issues for the application of batteries, especially the absorbent glass mat valve regulated lead-acid (AGM VRLA) type batteries used in the idle stop start systems (ISSs) that are popularly integrated into conventional engine-based vehicles. This is due to the fact that SOC and SOH estimation accuracy is crucial for optimizing battery energy utilization, ensuring safety and extending battery life cycles. The dual extended Kalman filter (DEKF), which provides an elegant and powerful solution, is widely applied in SOC and SOH estimation based on a battery parameter model. However, the battery parameters are strongly dependent on operation conditions such as the SOC, current rate and temperature. In addition, battery parameters change significantly over the life cycle of a battery. As a result, many experimental pretests investigating the effects of the internal and external conditions of a battery on its parameters are required, since the accuracy of state estimation depends on the quality of the information regarding battery parameter changes. In this paper, a novel method for SOC and SOH estimation that combines a DEKF algorithm, which considers hysteresis and diffusion effects, and an auto regressive exogenous (ARX) model for online parameters estimation is proposed. The DEKF provides precise information concerning the battery open circuit voltage (OCV) to the ARX model. Meanwhile, the ARX model continues monitoring parameter variations and supplies information on them to the DEKF. In this way, the estimation accuracy can be maintained despite the changing parameters of a battery. Moreover, online parameter estimation from the ARX model can save the time and effort used for parameter pretests. The validation of the proposed algorithm is given by simulation and experimental results.
Highlights
The growing awareness of global warming, fossil fuel depletion and fuel cost escalation have resulted in opportunities for the development of the automotive industry in terms of increasing vehicle efficiency and reducing carbon dioxide (CO2 ) emissions
The State of charge (SOC) and state of health (SOH) estimation performance has been tested with batteries with different aging levels under the varying temperature conditions and reliable results have been obtained under extensive variations in temperature, SOC and current
The combined method of an online parameter estimation algorithm by the auto regressive exogenous (ARX) model and a SOC and SOH estimation by the dual extended Kalman filter (DEKF) is implemented in the Labview environment, and the voltage and current sequences obtained during the dynamic charge discharge test are used as input data
Summary
The growing awareness of global warming, fossil fuel depletion and fuel cost escalation have resulted in opportunities for the development of the automotive industry in terms of increasing vehicle efficiency and reducing carbon dioxide (CO2 ) emissions. This method is an off-line method that requires a large number of tests to determine the battery model parameters at different SOCs and SOHs of a VRLA battery As a result, this method is not practical to use. Online parameters estimation can provide a lot of benefits in terms of reducing the time and labor for pretests and in increasing the accuracy of SOC and SOH estimation even when the parameters of a battery change. The SOC and SOH estimation performance has been tested with batteries with different aging levels under the varying temperature conditions and reliable results have been obtained under extensive variations in temperature, SOC and current. To test the performance in estimating the SOH, three different initial values have been given to verify its convergence characteristics by using a 440 h test profile
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
