4.2.2　Calculation of SOC

Text（课文）

State of charge（SOC）calculation is one of the most important functions of a BMS.Traditionally，SOC is calculated by counting how much current（or electrons）has moved in or out of the battery:

SOC（new）＝SOC（old）-I×TS/（A·h）nominal（4.1）

However，this calculation is subject to many uncertainties:

◆Measurement accuracy:Current sensors have measurement accuracy and resolution.Over time，the error could be cumulated in SOC.

◆The amplifier circuit also has accuracy associated with the precision of resistors and power supplies.

◆MCU calculation rounding errors

◆MCU A/D resolution

◆Battery initial SOC which may not be known to the MCU

◆Battery loss that cannot be counted for the calculation of SOC using the above equation

◆Self-discharge cannot be counted for by the SOC algorithm.

◆Current has harmonics which could cause measurement error due to discrete sampling by the MCU.

◆Noises in the measurement loop and in the amplification circuits

◆Battery aging could impact how much energy can be charged to the battery.Hence a percentage of SOC may not truly represent how much energy is available.

Therefore，other approaches must be developed to supplement the traditional SOC calculations.One of the most popular approaches is to use battery terminal voltage to calibrate the battery SOC.In this method，the battery is tested to develop a relationship between terminal voltage and SOC for different charge/discharge current and temperature.A lookup table is then built to find SOC based on measurement of battery voltage，current，and temperature.This SOC is then compared with the calculated SOC from the Coulomb counting method.If the discrepancy exceeds a certain threshold，then attention needs to be given and some further diagnostics needs to happen.A typical SOC and voltage relationship is shown in Figure 4.6.This method can not take into account the influence of battery aging.

Figure 4.6　Battery SOC versus terminal voltage

The second approach is to reset SOC when the battery is fully charged.Some batteries are not allowed to be fully charged but modern lithium ion batteries can be charged full every time without affecting the battery life.

New words and expressions（单词和短语）

1.New words（单词）

2.Expressions（短语）

Notes to the text（难点解析）

★1.However，this calculation is subject to many uncertainties.然而，这个计算受很多不确定因素的影响。

★2.Battery loss that cannot be counted for the calculation of SOC using the above.使用上述方程不能把电池损耗算进SOC的计算中。

Exercises（练习）

◆1.Translate the following passages（expressions）into Chinese（英译汉）

（1）Therefore，other approaches must be developed to supplement the traditional SOC calculations.

（2）Battery aging could impact how much energy can be charged to the battery.

（3）take into account

（4）rounding errors

◆2.Translate the following passages（expressions）into English（汉译英）

（1）受支配

（2）测量误差

（3）一种流行的方法是用电池端电压来校准电池SOC。

◆3.Directions:Answer the following questions briefly according to the text

Can you tell us several uncertainties in SOC calculation？

Reading material（阅读材料）

Lastly，advanced mathematical models can be used to predict battery SOC，such as Kalman Filtering，etc.The Kalman filter makes an optimum trade-off between believing the sensor reading and believing the model to achieve the best possible state estimate.The key is to filter the noise and combine with a good mathematical model.Kalman Filtering treats the measurement noise as a random variable and removes measurement noise from the system.In Kalman Filtering，we estimate signal X for each K（state）using

XK＝KKZK＋（1-KK）ZK-1（4.2）

where ZK is the measurement value that has error，and KK is the Kalman Gain which needs to be determined for different systems.A good mathematical model is a critical part of the Kalman Filter based SOC estimation.