Lithium battery soc chart
Electric vehicles need safe and efficient batteries as power sources. Lithium battery is used as the ideal power source of new generation electric vehicles because of its advantages such as stable working voltage, high energy density and charging efficiency, low self discharge rate, no memory, and long service life. How to accurately estimate the remaining battery power is of great significance for improving the maximum utilization rate of lithium batteries and continuously optimizing battery technology. In the research and development of electric vehicles, accurate prediction of battery SOC plays an important role in exerting the best performance of electric vehicles and predicting the driving range of electric vehicles. However, the state of charge of lithium battery can not be directly measured, and is affected by many factors, such as the rate of charge and discharge, the aging degree of the battery, and the internal resistance of the battery, which makes it difficult to measure accurately and quickly. On the basis of reading a large number of relevant literature, this paper comprehensively expounds some main prediction methods for the state of charge of lithium batteries at present, and compares the advantages and disadvantages of various methods.
1、 State of Charge (SOC) definition
SOC refers to the state of charge of the battery. From different angles such as electricity quantity and energy, SOC has many different definitions. The SOC defined by the United States Advanced Battery Association (USABC) is widely used, that is, the ratio of the residual capacity of the battery to the rated capacity under the same conditions at a certain discharge rate. The corresponding calculation formula is:
In the formula, Qm is the maximum discharge capacity of the battery when discharging at a constant current I; Q (In) is the electricity released by the battery under standard discharge current I in time t.
2、 Prediction method of state of charge of lithium battery
The state of charge of lithium battery is one of the important parameters of the battery management system, and it is also the basis for the charge and discharge control strategy and battery balance work of the entire vehicle. However, due to the complexity of the structure of lithium battery itself, its state of charge cannot be measured directly. It can only predict the state of charge according to some external characteristics of the battery, such as the internal resistance, open circuit voltage, temperature, current and other relevant parameters of the battery, using the relevant characteristic curve or calculation formula.
The estimation of the state of charge of lithium battery is nonlinear. At present, the commonly used methods include discharge experiment, open circuit voltage, ampere hour integration, Kalman filter, neural network, etc.
Discharge test method
The principle of the discharge experiment method is to keep the battery in an uninterrupted discharge state with a constant current, and calculate the discharge amount when the discharge reaches the cut-off voltage. The discharge quantity is the product of the constant current value and the discharge time used during discharge. Discharge test method is often used to estimate the state of charge of batteries under laboratory conditions, and many battery manufacturers also use discharge method to test batteries at present.
Its remarkable advantages are simple method and relatively high estimation accuracy. Its disadvantages are also very prominent: it is not allowed to measure with load, which requires a lot of measurement time. In addition, during discharge measurement, the work before the battery must be interrupted, so that the battery can be placed in the offline state, so it cannot be measured online. This method is not applicable to the electric vehicle battery that is in operation all the time and its discharge current is not constant. However, the discharge experiment method can be used in battery maintenance and parameter model determination.
Open circuit voltage method
The parameters of the battery are relatively stable after a long time of full standing, and the functional relationship between the open circuit voltage and the battery state of charge is also relatively stable. If you want to obtain the state of charge value of the battery, you only need to measure the open circuit voltage at both ends of the battery and obtain the corresponding information by comparing with the OCV-SOC curve.
The advantage of the open circuit voltage method is that it is simple to operate. The value of the state of charge can be obtained by measuring the open circuit voltage value and comparing it with the characteristic curve. However, there are many disadvantages: first of all, in order to obtain an accurate value, this method must make the battery voltage in a relatively stable state, but the battery often needs to be kept for a long time before it can be in this state, which can not meet the requirements of real-time monitoring, and is often applied to electric vehicles when parking for a long time.
When the charge discharge ratio of the battery is different, the open circuit voltage of the battery will change due to the fluctuation of the current, which will lead to the inconsistency of the open circuit voltage of the battery pack, resulting in a large deviation between the predicted residual power and the actual residual power of the battery.
Ampere hour integration
The ampere hour integration method does not consider the internal mechanism of the battery. According to some external characteristics of the system, such as current, time, temperature compensation, etc., the total amount of electricity flowing into and out of the battery is calculated by integrating the time and current, and sometimes adding some compensation coefficients, so as to estimate the state of charge of the battery. At present, the ampere hour integration method is widely used in the battery management system. The calculation formula of ampere hour integration method is as follows:
In the formula, SOC0 is the initial electric quantity value of the battery charge state; CE is the rated capacity of the battery; I (t) is the charging and discharging current of the battery at time t; T is the charging and discharging time; η It is the charge discharge efficiency coefficient, also known as the coulomb efficiency coefficient, which represents the internal power consumption of the battery during the charge discharge process. Generally, it is dominated by the charge discharge ratio and temperature correction coefficient.
The advantage of the ampere hour integration method is that it is relatively less limited by the battery itself, the calculation method is simple and reliable, and it can estimate the state of charge of the battery in real time. Its disadvantage is that the ampere hour metering method belongs to open-loop detection in control. If the current acquisition accuracy is not high, there is a certain error in the given initial state of charge. With the extension of the system operation time, the errors generated previously will gradually accumulate, thus affecting the prediction results of the state of charge. And because the ampere hour integration method only analyzes the state of charge from the external characteristics, there is a certain error in the multi link. From the calculation formula of ampere hour integration method, it can be seen that the initial charge of the battery has a great impact on the accuracy of the calculation results.
In order to improve the accuracy of current measurement, high-performance current sensors are usually used to measure current, but this increases the cost. For this reason, many scholars combine the ampere hour integral method with the open circuit voltage method. The open circuit voltage method is used to estimate the initial state of charge of the battery, and the ampere hour integration method is used for real-time estimation, and relevant correction factors are added in the formula to improve the accuracy of the calculation.
Kalman filtering method
Kalman filtering algorithm is a minimum variance estimation using the time-domain state space theory, which belongs to the category of statistical estimation. Macroscopically, it is to minimize and eliminate the impact of noise on the observation signal as much as possible. Its core is the optimal estimation, that is, the effective correction of the state variables by the system input based on the estimation.
The basic principle of the algorithm is that the state space model of noise and signal is taken as the algorithm model. During measurement, the current time observation value and the previous time estimation value are used to update the estimation of state variables. The essence of the Kalman filter algorithm to predict the state of charge of lithium battery is the ampere hour integration method, and the measured voltage value is used to correct the preliminary predicted value.
The advantage of Kalman filtering method is that it is suitable for computer to process data in real time. It has a wide range of applications and can be used in nonlinear systems. It has a good effect on the state of charge prediction of electric vehicles during driving. The disadvantage of Kalman filter method is that it depends heavily on the accuracy of the battery model. In order to improve the accuracy and precision of the prediction results of the algorithm, it is necessary to establish a reliable battery model. In addition, the algorithm of Kalman filtering method is relatively complex, so its calculation amount is relatively large, and the performance of the solver is required.
Neural network method
The purpose of neural network is to imitate human intelligent behavior, obtain the ability of data expression through parallel structure and its own strong learning ability, give corresponding output response when external excitation exists, and make it have good nonlinear mapping ability.
The principle of the neural network method applied to the detection of the state of charge of lithium batteries is that a large number of corresponding external data such as voltage and current, as well as the state of charge data of the battery are used as training samples, and the training and modification are repeated through the forward propagation of input information and the reverse propagation of error transmission in the neural network’s own learning process. When the predicted state of charge reaches the error range required by the design, The predicted value of battery state of charge is obtained by inputting new data.
The advantage of the neural network method is that it can estimate the state of charge of various batteries, and it is widely used; There is no need to establish a specific mathematical model, no need to consider the complex chemical change process inside the battery, just select appropriate samples and establish a better neural network model. The more sample data, the higher the estimation accuracy; The charge state of the battery can be determined at any time. The disadvantage of neural network method is that it requires high hardware. The accuracy of data samples, sample size and sample distribution used in training, as well as training methods, will have a great impact on the prediction of battery state of charge.
In this paper, several main methods for predicting the state of charge of lithium batteries are briefly introduced, and their advantages and disadvantages are analyzed in detail. At present, the ampere hour integration method is still the most widely used method to predict the state of charge of lithium batteries. However, due to its own limitations, the ampere hour integration method is often combined with other methods such as open circuit voltage method to complete the detection of the initial state of charge of lithium batteries.
From the perspective of development trend, more and more comprehensive factors are considered in the prediction of the state of charge of lithium batteries. The prediction methods used are often the comprehensive application of the above several methods, making the prediction results more accurate. At present, the equivalent circuit model of lithium battery has been developed continuously, which is closer to the reality, making the prediction accuracy of the state of charge further improved.