Xuezhe Wei

Online SOC Estimation of High-power Lithium-ion Batteries Used on HEVs

Battery management systems (BMS) in hybrid electric vehicles (HEVs) should be able to online estimate the present state of charge (SOC) of the battery pack accurately. In this paper, we proposed a SOC estimating method for battery packs based on the well-known extended Kalman filter (EKF). The underlying dynamic behavior of the battey pack was described by a model which was based on an equivalent circuit comprising of two capacitors and three resistors. Measurements of current and voltage in two different tests were applied to validate the proposed method. By comparing the SOC estimated by model based EKF to the SOC estimated by coulomb counting, we got the results showing that the methodologies we proposed were able to perform a good estimation of SOC of the battery pack. Moreover, a corresponding BMS including hardware and software based on our estimating method was designed.

State and Parameter Estimation of a HEV Li-ion Battery Pack Using Adaptive Kalman Filter with a New SOC-OCV Concept

A new methodology of defining the relationship between SOC (State of Charge) and OCV (Open Circuit Voltage) relationship of the Li-ion battery pack used on HEVs (Hybrid Electric Vehicles) which is independent of the battery condition was proposed. This methodology could avoid the problems resulting from the defects that the conventional SOC-OCV relationship differs between batteries and different working conditions. Based on the new definition, a state and parameter estimator of the Li-ion battery pack based on the Sage-Husa adaptive Kalman filter was proposed. This estimator recruited an equivalent circuit model to describe the dynamic characteristics of the battery pack. The estimator could estimate the SOC, the battery actual capacity and the inner resistance on-board. The implementation of the estimator on a FPGA platform was also introduced. Testing results show that the new definition and the estimator work very well in any specific working condition.

Studies on the medium-frequency impedance arc for Lithium-ion batteries considering various alternating current amplitudes

An impedance measurement approach with various current amplitudes is proposed to investigate the impedance behavior of power Lithium-ion battery in the frequency domain. Notably, the impedance arc is divided into amplitude-independent response in the high-frequency range and amplitude-dependent response in medium-frequency range. A shrinking of semi-circle diameter of the impedance arc at medium-frequency part of measured spectrum is distinctly observed with rising current amplitude. Thus, a series of particular experiments are performed to facilitate the interpretation of the phenomenon. The possible influencing factors for the battery, including its state of charge, internal pressure, and temperature rise during the high-rate short-time charge/discharge, are illustrated and excluded for analysis, respectively. Based on the analysis of Butler-Volmer equation and Arrhenius empirical equation, a formula, which contains temperature and current known to influence the charge-transfer resistance, is derived logically. It indicates that the charge-transfer resistance reduces as the current amplitude increases at a fixed temperature, which fundamentally dominates the shrinking of the semi-circle at medium frequencies. Therefore, the medium-frequency impedance arc is able to represent the chemical reaction process, even under large current excitation (within a certain range).

A new electrochemical impedance spectroscopy model of a high-power lithium-ion battery

This paper presents an electrochemical impedance spectroscopy battery model including an electrical double layer capacitance, which can comprehensively depict the internal state of the battery. Based on the porous electrode theory, this model can obtain a complete spectrum of cell impendence. In addition, the EIS data recorded by an appropriate measurement instrument are consistent with the records of theoretical impedance derived from the model in terms of room temperature. To establish the relationship between internal parameters and external spectrum, a virtual simulation experiment is conducted to illustrate the effect of key parameters on EIS. According to the simulation experimental results, the model can be used to identify internal parameters based on the practical external spectrum. In the last part of the paper, a preliminary parameter identification strategy is proposed to study the relationship between battery internal parameters and external conditions.

A Semi-Empirical Capacity Degradation Model of EV Li-Ion Batteries Based on Eyring Equation

To ensure a reliable application of li-ion batteries on electric vehicles (EVs), life prediction of li- ion cells is very essential. The main objective of the present study is to propose a semi-empirical capacity fading model of li-ion batteries being practical for predicting the degradation of battery capacity under real working conditions. The capacity fading of li-ion cells is considered to be mainly caused by side reaction, which consumes the active materials. The main contribution of the study is that the model is not founded only empirically based on the experiment results, but also with the theoretical description of side reaction based on Eyring equation. By this combined model, it is expected to get an accurate and meanwhile practical model for performance prediction. The model takes DOD, temperature and current rate as three main factors affecting ageing process. The ageing tests were designed and implemented to get the parameters of the capacity fading model, and one validation experiment was used to evaluate the prediction performance of the model. There is a good agreement between the actual capacity degradation and predicted capacity degradation.

State Estimation of Lithium Ion Battery Based on Electrochemical Impedance Spectroscopy with On-Board Impedance Measurement System

Lithium ion battery is widely used in electric vehicles. To estimate battery state-of-charge, state-of-health and internal temperature accurately is not only in laboratory but also on-board significant. As a useful tool to study electrochemical system, numbers of estimation methods based on electrochemical impedance spectroscopy arise. The paper focuses on the relation between the battery states and impedance to develop a decoupled state estimation method to satisfy on-board applications. To overcome shortcomings of bulky and high cost of electrochemical impedance spectroscopy equipment in laboratory, an impedance measurement system is designed. Validation results indicate that the battery states can be estimated properly with the impedance based method and the developed measurement system works well, which facilitates the implementation of the impedance based state estimation on board.

Principle Elaboration and System Structure Validation of Wireless Power Transfer via Strongly Coupled Magnetic Resonances

Theoretical analysis of wireless power transfer (WPT) via strongly coupled magnetic resonances (SCMR) is given on the coupled-mode theory (CMT) and circuit theory (CT). Furthermore, differences and similarities between these two theories are described. Some kinds of practical system structures are analyzed by CT, and the limitation of SCMRs application is elaborated. Aiming at different practical system structures, the matching methods are explored and the conclusions are validated by simulation.

A High-voltage Safety Protection Method for Electric Vehicle Based on FPGA

Currently, the voltages of power supply on electric vehicle are far higher than the level that people can endure. Aiming at the high voltage safety of electric vehicle, a solution based on FPGA is introduced, which gives a real time detection for creepage of power supply and persons getting an electric shock on electric vehicle, and which cuts off the power supply immediately when the danger is detected. In the solution, a kind of voltage-sampling circuit is provided for the power supply on electric vehicle while the device of FPGA is used to count the insulated resistance to ground of power supply and control the switches. Using FPGA to realize the controlling function in place of MCU can increase the speed and stability of system. The result of experiment indicates that, applied on electric vehicle, this solution takes effect on the detection and protection for the creepage of power supply and persons getting an electric shock.

A Hardware-in-the-Loop System for Development of Automotive Battery Management System

Battery management system (BMS) plays a vital role in the development of hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (PHEVs) and pure electric vehicles (PEVs). The vehicle management relies highly on the information transmitted from BMS. In this study, a battery pack-level hardware-in-the-loop (HIL) system is developed to verify the functions of BMS, including the cell voltage and temperature monitoring, battery voltage and current measurement, State of Charge (SOC) estimation, power relay and cooling fan control etc. By emulating different types of input signals to BMS, and at the same time detecting the output signals from BMS, the HIL system could verify and evaluate the BMS before it goes to mass-production, and thereby accelerate the BMS development and reduce the development costs.

A Simulation of Lithium-Ion Battery Ohmic Resistance Identification

In electric vehicle applications, it is necessary to estimate the internal resistance of Lithium-ion battery, which is feasible through internal resistance identification based on equivalent circuit model. But in the actual battery management system, current sampling tends to lag behind the voltage sampling, which affects the identification accuracy apparently. In this article, the ohmic resistance of the lithium-ion battery is evaluated based on equivalent circuit model and least squares identification algorithm. Influence by desynchronized sampling of current and voltage is simulated. And a reasonable value of sampling step is introduced.