BMS Series-Estimation of ——SOC

let's take a look at the definition of ——SOC, which is the full name of State of Charge, and it is also called residual power. It represents the ratio of the residual dischargeable power of the battery after a period of use or long-term shelving to its fully charged power, usually expressed as a percentage. It is generally represented by one byte, that is, two hexadecimal digits (the range of values is ~1), which means that the remaining power is %~1%. When SOC=, it means that the battery is fully discharged, and when SOC=1%, it means that the battery is fully charged. SOC is closely related to our life, such as the power of common mobile phones, smart watches and electric cars.

1. characteristics of battery open circuit voltage (OCV). The open-circuit voltage of the battery refers to the potential difference between the positive and negative electrodes when the battery is at rest. Different batteries have different characteristics, so the specific battery type should be fully considered;

2. Characteristics of the relationship between charge-discharge rate and terminal voltage. When the battery is dynamic, the battery voltage we measure is actually the terminal voltage of the battery. Their relationship is relatively stable at constant temperature and constant current, but the corresponding relationship will be disturbed as long as the battery starts to work at non-constant current.

3. temperature state. Batteries of different materials will be affected by temperature, especially low temperature (so the system will have heating function); Anyway, the temperature has an influence on all parameters of the battery;

4. battery life status. The life of the battery will gradually decay during the use, and the decay mechanism mainly lies in the collapse of the anode and cathode materials and the passivation of the electrodes, which leads to the loss of effective lithium ions. The total power will also approach from BOL(Beginning of Life) to EOL(End of Life). Therefore, it is necessary to consider whether to use the total capacity at BOL moment or the actual total capacity at the current life when calculating SOC.

5. Series and parallel combination of batteries. In our actual use, the batteries are definitely not single, all because of the series-parallel combination of batteries, which makes the estimation of SOC more complicated. There are inevitably differences in ohmic resistance, polarization resistance, self-discharge rate and initial capacity between different batteries.

1. AH integration method. The classical SOC estimation generally adopts ampere-hour integration method (also called current integration method or Coulomb counting method). The ampere-hour integration method is to estimate the SOC of the battery based on the initial time SOC. By calculating the integral of the charging and discharging current and the corresponding time in a certain time, the percentage of the changed power is calculated, and finally the difference between the initial SOC and the changed SOC, that is, the remaining power, is obtained. The ampere-hour integration method equates the battery to a closed system, and only studies the external characteristics of the battery, that is, it only needs to monitor the power in and out of the battery in real time and then accumulate the power, so the remaining power of the battery at any time can be obtained.

However, the ampere-hour integration method, because there are errors in ampere-hour measurement, the cumulative error will get bigger and bigger with the increase of service time, so it is not effective to estimate the SOC of the battery by this method alone.

2. Open circuit voltage method. The principle is that there is a relatively fixed functional relationship between the open circuit voltage and SOC when the battery is left standing for a long time, so the SOC can be estimated according to the open circuit voltage. The accuracy is high when the battery is left standing for a long time, but it is not applicable in actual working conditions. Therefore, the open-circuit voltage method is generally combined with other methods to predict SOC. The most commonly used method in the industry is open circuit voltage+ampere-hour integration method.

3. Kalman filtering method. Kalman filter has been widely used in aerospace, communication, navigation, control, image processing and other fields. It is a very mainstream direction to estimate the SOC of power battery by Kalman filter. The core of the algorithm is to make the optimal estimation of the state of the dynamic system, and the criterion is the minimum covariance. When applied to the battery, the state and observation equations must be established first, and the SOC is the state component. Here, the KF algorithm can be used to estimate the SOC, and the KF algorithm can be used to estimate the unknown state in the model, which has relatively high accuracy and robustness. KF algorithm can make the estimation result close to the true value after many updates, and can correct the initial value of capacity well, and has strong anti-interference ability. This method can theoretically realize the dynamic estimation of the system, so it is also considered as one of the reliable and effective methods in the research field.

4. BP neural network method. BP network can learn and store a large number of input-output pattern mapping relationships without revealing the mathematical equations describing this mapping relationship in advance. Its learning rule is to use the steepest descent method, and constantly adjust the weights and thresholds of the network through reverse propagation to minimize the sum of squares of the network errors. The topological structure of BP neural network model includes input layer, hide layer and output layer. In practice, due to the complexity of the algorithm, the hardware requirements are extremely high, so it is still a long way to apply this method to embedded BMS products.

At present, the estimation accuracy of SOC is still a world-class problem, and we hope to break through it as soon as possible. For related enterprises, it is also a standard to measure the technical level of an enterprise. Of course, it takes a lot of manpower and material resources to estimate the breakthrough of SOC. (PS- having learned so much, I'd better use the basic method to calibrate honestly, and it's almost enough.)

I wish you to go away for half your life and come back as a teenager …