Battery system state of charge calculation algorithm problem

qwqwqw2088

Battery system state of charge calculation algorithm problem [Copy link]

The SOC of a battery, the full name of which is State of Charge, is also called the remaining capacity. It represents the ratio of the remaining capacity of a battery after it has been used for a period of time or has been left unused for a long time to the capacity of the battery in its fully charged state, and is usually expressed as a percentage. It is generally expressed in one byte, that is, two hexadecimal digits (the value range is 0~100), which means that the remaining capacity is 0%~100%. When SOC=0, it means that the battery is fully discharged, and when SOC=1, it means that the battery is fully charged. The SOC (state of charge) algorithm has always been one of the key technologies for the development and application of battery management systems (BMS). Regarding the calculation of SOC, the available range of SOC is determined by multiple factors. If we only look at it superficially, it is unfair to attribute the responsibility for determining the size of the SOC range to the level of BMS management. Good BMS estimation accuracy also requires more complete battery parameters to support it. First, let's clarify two concepts: What is the available range of SOC? Discharge depth? According to the definition in GBT19596 (T=25℃, first introduce the state of charge): State of charge (SOC): "The capacity that can be released in the current battery according to the specified discharge conditions accounts for the percentage of the available capacity." The SOC state range percentage is generally from 0% to 100%. However, considering the chemical battery reaction characteristics: threshold boundary, static and dynamic differences, rate differences, valuation accuracy differences, etc., SOC valuation needs to leave a buffer zone to ensure that the battery always works in a safe area. SOC available range: The SOC range minus the SOC buffer area, the remaining part is the SOC available range. As shown in Figure 1, c-d interval, 15%~95% available range. Depth of discharge DOD: "The parameter that indicates the battery discharge state is equal to the actual discharge capacity as a percentage of the available capacity." Numerical relationship: SOC=1-DOD. DOD is more of a reflection of the current battery capacity and a measure of the depth of discharge. For example, when expressing battery life, it is often used as a pre-parameter, 1C/1C DOD 80%, 3000 cycles. Attached Figure 1, borrowing the battery parameters of a world-renowned brand to illustrate the inclusive relationship between SOC range, SOC usable range, battery threshold range, and safety range.

Figure 1 Accurate correlation factors of SOC usable range: First, the integrity and accuracy of battery parameters Because lithium-ion batteries are chemical products, their energy form is the mutual conversion of chemical energy and electrical energy, and the charge and discharge curve is nonlinear. Among them, capacity, energy, and power are greatly affected by factors such as ambient temperature, temperature rise rate, current rate, and SOC state. If accurate battery performance testing is completed, the entire testing process is very time-consuming and labor-intensive. In order to respond to rapid market demand or subsidy policies, many manufacturers artificially speed up the launch of products, testing and selling products at the same time. This practice has buried hidden dangers for the engineering application of batteries. The reason for the long product testing time is reflected in every link. Only the standard cycle or operating cycle takes 3 to 6 months, which is only the factor of the battery product itself. If combined with the equipment status, the time will be longer. At present, there are no less than 12 safety tests for battery cells, and more than 16 system function tests. There are also conventional function and performance tests. If SOC at different temperatures is superimposed, the testing workload is very large. It is conceivable that from a finalized active material formula to the launch of a qualified product, the product maturity cycle takes a long time. The integrity of battery parameters depends on sufficient and multi-sample testing of individual battery cells. Generally speaking, the battery parameter model proposed based on demand is the state relationship of battery parameters under multiple dimensions, which is an all-round battery evaluation. This is also the parameter label that the product should have. From the test in the attached figure, it is generally carried out in several steps. First, its basic functions are tested, and the data that meets the use is tested first, which is similar to the grid from large to small in the spectrum chart. For example, the SOC test step is stepped by 5% or 10%. If faced with a range with higher test accuracy requirements, it is still far from enough. For key sections, it is necessary to focus on testing, at both ends of the battery charge and discharge curve, low temperature power state, etc. The above description is more for individual cells. If we stand from the perspective of the system, it is more sensitive to heat, consistency, power, and energy, and the difficulty of testing will increase accordingly. In addition, a more important point is the stability and accuracy of the test equipment. At present, many manufacturers still choose expensive imported equipment for testing in some key links. Why is this? It is mainly to ensure the accuracy and stability of the test. Fortunately, the current situation has changed in recent years. Domestic test equipment manufacturers have devoted themselves to practice and growth. A large number of excellent manufacturers such as Xingyun Technology have launched test equipment that can not only compete with similar foreign products, but also have more down-to-earth prices and thoughtful after-sales service, and have achieved good reputation and recognition. Secondly, it is the correctness and accuracy of the BMS algorithm. The accuracy of the BMS algorithm that we mentioned most is proposed for the requirements of the battery system. For the problem of optimizing the available range of SOC, it is not entirely correct to start from BMS unilaterally. As mentioned above, the integrity of battery parameters is also an important factor. A good cook cannot cook without rice, and BMS is timid in the face of missing data. When mentioning the SOC algorithm, the most common words are "estimate" and voltage "Approx". This does not contradict the SOC accuracy requirement. Because of the characteristics of the battery itself, the "current status" does change with the length of time, temperature, and C value. For example, SOC 5%, Valus Status Approx.3200-3400mV. There are certain differences in dynamic voltage and OCV value, as well as static shelf time. This is precisely the difficulty and charm of the algorithm strategy. Of course, if the display of the instrument is user-friendly, by establishing a corresponding relationship with the real SOC in the background, it can be considered as the SOC value facing the user. The accuracy of SOC estimation is different under different working conditions. Usually, we will ask BMS to achieve or be less than 5% SOC accuracy. In fact, the understanding of BMS engineering is that this accuracy represents the maximum error, not the only one. The lower limit value of SOC available range is determined through precise and detailed strategy control and accurate values Comprehensive analysis shows that the optimization of SOC available range is to determine the lower limit value of the battery under different conditions and working conditions. The cache interval of the upper limit of the battery is very small, and there is not much room for mining. The cache of the upper limit is mainly for charging safety, with the purpose of ensuring that it is not overcharged. When fast charging, the SOC is charged to 80%; when slow charging, it can reach more than 95% by relying on trickle current charging. The lower limit of the battery is mainly considered in the discharge condition. The ability of the discharge current to change will affect the power output or driving experience. At the same time, the width of its cache is still very large. Let's take an example to illustrate the relationship between the determination of the lower limit and the working condition: VOLT has an optimal life safety window (58~65%), which is a more important part of its strategy. The SOC lower limit value of this window is different according to different working conditions. In normal working mode, the lower limit is set to SOC=30%; in mountain road working mode, the lower limit is set to SOC=45%. This principle is easy to understand. When in mountain road mode, the C value of discharge or charging (energy recovery) changes greatly. In order to prevent instantaneous over-discharge (undervoltage) and overcharge (overvoltage), the battery is set to a safe state by setting the limit voltage. Differences in the available range of EV and HEV SOC

Because the tasks and roles that the battery system undertakes in EV and HEV are different, the C value requirements are different. EV emphasizes a large driving range; HEV or PHEV emphasizes dynamic power hybrid capabilities, including large current energy recovery capabilities. The difference in usage functions also determines the difference in their limits. At the same time, the higher C value of HEV or PHEV will naturally affect the service life of the battery. Therefore, the window or limit also needs to consider the life factor. As shown in the table below, the impact of battery discharge depth on life is very large.

Through the above analysis, the size of the available SOC range is still determined by the accuracy of battery parameters and the accuracy of the BMS algorithm. These two aspects are indispensable. At the same time, under the premise of ensuring battery safety, facing various working conditions, BMS strategies and algorithms cannot be one-size-fits-all, but require precise multi-level implementation. Under the premise of battery safety, make full use of battery capacity and optimize and maximize the available SOC range.