Battery Management Algorithm for Electric Vehicles

Developing energy-saving and new energy vehicles industry is an international consensus, which is also an emerging industry in China and a key filed established by “Made in China 2025”.

A battery is a typical electrochemical system. The battery test plan established for the battery management system (BMS) studies belongs to the field of experimental science. In order to establish accurate battery models and develop high-performance BMS, it is necessary to design and imply a series of targeted tests to acquire the battery performance under diverse conditions. The quality of the test plan and the experimental data directly affects the rationality and integrity of the battery characteristics analysis, which further affects the accuracy and reliability of the battery model, and ultimately affects the control performance of the BMS. This chapter will focus on the battery system test platform construction, the design of the test methods, the data analysis, and the basic characteristics of lithium-ion batteries [1].

The complex electrochemical reactions inside the batteries are affected by many influencing factors and uncertainties. Establishing mathematical models of batteries is seen as a multidisciplinary problem, for which it has always been an important yet difficult problem in academia and industry.

Battery SOC and SOH estimation are core functions performed by the BMS. Accurate SOC and SOH estimation can ensure the safe and reliable operation of the battery system, and provide the basis for energy management and safety management of EVs. However, batteries exhibit the characteristics of limited measurable parameters, coupling feature, degradation with time, strong time-varying, and nonlinearity. The vehicle applications are also encountering the requirements of series-parallel group of inconsistent complex system, various operation conditions (wide rate charge and discharge), and all-climate (–30 to 55 °C temperature range). Battery SOC and SOH estimation with high precision and strong robustness are extremely challenging, and they have been the industry’s technical difficulties and hotspots in the international academic research. This chapter will systematically describe the basic theory and application of battery SOC and SOH estimation, discuss the performance of online SOC estimation with the known static capacity and dynamic capacity as well as the necessity of SOH and SOC collaborative estimation. A detailed algorithm flow for the practical application of BMS will also be provided.

A battery system mainly consists of battery modules, a BMS, and a battery pack case. A battery cell has maximum available capacity and SOC, the estimation of which has clear reference values and evaluation methods.

The internal mechanism of lithium-ion batteries is very complicated. There are many reasons for performance degradation, and various factors are coupled with each other, which eventually leads to an extremely challenging engineering problem [1]. The battery performance degrades throughout the whole process of use and maintenance. As the charging and discharging cycles increase, some irreversible chemical reactions occur inside the battery, resulting in an increase of internal resistance and declines of maximum available capacity, energy, and SOP, which reduces the driving mileage of EVs, even leading to some safety hazards [2‐4]. A reliable remaining useful life (RUL) prediction method can ease the user’s anxiety about the remaining driving mileage and safety issues, and ensure the safe and efficient operation of the batteries. It can also guarantee the safety and reliability of the EV during operation, reduce failure rate and operating costs, improve user experience and avoid accidents. Therefore, the battery RUL prediction is one of the core tasks of BMS. After introducing the related concepts of battery RUL prediction, this chapter summarizes and classifies the mainstream RUL prediction methods, and finally introduces two representative battery RUL prediction methods.

With the promotion and popularization of new energy vehicles, the problems of short driving range accompanied by the difficulty in starting and charging during winter in high altitude regions are becoming severe nowadays.

The simplified process of the algorithm in the theoretical design may lead to deviations in practical application. As a result, it is very important to download the algorithm to the real BMS and evaluate it according to the relevant standards and indexes, which helps designers to find and solve some practical problems that are neglected in the theoretical derivation in time and optimize the algorithm. The traditional algorithm development and evaluation methods not only consume a lot of time, manpower cost, but also limited by safety issues. In addition, it is difficult to comprehensively and systematically evaluate some actual controlled objects. Fortunately, the “V” development process based on the rapid prototyping and hardware in the loop (HIL) test can find out the problems in the algorithm and make evaluation efficiently and accurately, which improves the development efficiency. This chapter mainly focuses on the development process of BMS for EVs, and illustrates the evaluation methods of rapid prototyping simulation, HIL test algorithm, and the experiments for vehicles [1].