Research on equalizing charging technology of series-connected battery packs

　　The voltage and capacity of a single battery are limited, and in many cases it is necessary to form a series battery pack for use. However, there is a balance problem among the batteries in the battery pack. How to extend the service life of the battery pack, improve the stability of the system and reduce costs are important issues before us.

　　The service life of a battery is determined by many factors, the most important of which is the physical properties of the battery itself.

　　In addition, poor battery management technology and unreasonable charging and discharging systems are also important reasons for shortened battery life. For battery packs, in addition to the above reasons, the inconsistency between single cells is also an important factor. In view of the unbalanced phenomenon of single cells in the battery charging and discharging process, the author analyzed and compared several current equalizing charging methods, proposed a lossless equalizing charging method based on actual conditions, and conducted experimental verification.

Existing equalization charging methods

　　To achieve equal charging of each single cell in a series battery pack, there are currently several main methods:

　　1. Attach a parallel balancing circuit to each single cell of the battery pack to achieve a shunt function. In this mode, when a battery reaches full charge first, the balancing device can prevent it from overcharging and convert the excess energy into heat energy to continue charging the under-charged battery. This method is simple, but it will cause energy loss and is not suitable for fast charging systems.

　　2. Before charging, discharge each cell one by one through the same load to the same level, and then perform constant current charging to ensure a more accurate equilibrium state between each cell. However, for battery packs, due to physical differences between individuals, it is difficult for each cell to achieve completely consistent ideal effects after deep discharge. Even if the same effect is achieved after discharging, new imbalances will appear during the charging process.

　　3. Timing, sequencing, and individually detecting and evenly charging the single batteries in the battery pack. When charging the battery pack, it can be ensured that each battery in the battery pack will not be overcharged or overdischarged, thus ensuring that each battery in the battery pack is in normal working condition.

　　4. Using the time-sharing principle, through the control and switching of switch components, extra current flows into the battery with a relatively low voltage to achieve the purpose of balanced charging. This method is more efficient, but the control is more complicated.

Figure 1 Principle diagram of time-sharing control equalizing charging

　　5. Use the voltage parameters of each battery as the balancing object to restore the voltage of each battery to be consistent. As shown in Figure 2, during equalization charging, the capacitor is alternately connected to two adjacent batteries through the control switch, receives charge from the high-voltage battery, and then discharges to the low-voltage battery until the voltages of the two batteries become consistent.

　　This balancing method can better solve the problem of battery pack voltage imbalance, but this method is mainly used in situations where the number of batteries is small.

Figure 2 Schematic diagram of balanced voltage charging principle

　　6. The entire system is controlled by a single-chip microcomputer, and each battery has an independent set of modules. The module performs charge management on each single battery according to the set program, and automatically disconnects after charging is completed.

　　This method is relatively simple, but it will greatly increase the cost when the number of single cells is large, and is not conducive to reducing the system volume.

Lossless charging circuit

　　This article proposes a lossless charging equalization circuit. After the equalizing charging module is started, the overcharged battery will transfer the excess power to the undercharged battery to achieve dynamic balancing. It has high efficiency and low loss, and all battery voltages are fully monitored by the equalizing charging module.

1 Circuit design

　　For a battery pack composed of N cells connected in series, the main circuit current is Ich. Each series-connected battery is connected to a balancing bypass, as shown in Figure 3. In the figure, BTi is a single battery, Si is a MOSFET, and the inductor Li is an energy storage component. Si, Li and Di form a shunt module Mi.

　　In a charging cycle, the circuit working process is divided into two stages: the voltage detection stage (time is Tv) and the equalizing stage (time is Tc). During the voltage detection phase, the balancing bypass circuit does not work, the main power supply charges the battery pack, and at the same time detects the voltage of the individual cells in the battery pack, and calculates the duty cycle of the MOSFET according to the control algorithm. In the equalizing charging stage, the triggered MOSFET in the bypass controls the switching state according to the calculated duty cycle, and performs equalizing charging on the corresponding battery. In this stage, the current flowing through each single cell is constantly changing and different.

Figure 3 Equal charging circuit

　　Except for M1 connected at both ends of B1, the composition of all bypass shunt modules is the same. In the equalizing charge bypass, due to the unidirectional conduction of the diode Di, all shunt modules will transfer excess power from the corresponding battery to the upstream battery, while M1 will transfer the excess power to the downstream battery.

2 Calculation of switching tube duty cycle

　　The state of charge (SOC) of the battery during charging can be obtained from the following empirical formula, where V is the terminal voltage of the battery.

　　SOC＝-0.24V ² +7.218V-? 53.088? (1)

　　SOC is the ratio of the battery's current capacity to its rated capacity, SOC=Q/Q _TOTAL ×100%.

　　By converting the battery voltage detected at the end of the voltage detection phase into the state of charge, and there is a corresponding relationship between the storage capacity Qest,n of a single battery and the SOC, Qest,n can be estimated.

　　In the charge balancing stage, the amount of electricity charged into a single battery from the main charger is IchTcep. Among them, Tcep is the time of the equalizing charging phase in a charging cycle. In order to achieve the balance of the storage capacity of a single battery during the equalizing charging stage, the equalizing charging target Q _tar should be:

　　 (2)

　　However, the charging conversion between the activated bypass and other batteries affects each other. The current output by a single battery to other batteries through the bypass and the charging current received are difficult to calculate with a simple formula. However, the Gauss-Seidel iteration method can solve this problem.
The desired storage capacity Q _n can be calculated using the following formula:

?(3)

　　Where, I _dis,n is the average current in a switching cycle, and I _obt,n is the current obtained from other triggered bypasses. Q _tar is the charge amount when the battery reaches equal charge after charging cycle Ts under ideal conditions, Q _n is the expected storage capacity, Q _tar = Q _n , that is, (2) and (3) are equal. Through corresponding conversion, the calculation formula of duty cycle is obtained:

(4)

_{The function f N}　　here is just a schematic function, indicating that there is a certain relationship between D _n and D ₂ ...D ₃ .

3 Experimental design

　　In order to verify the balanced charging method in this article, a battery pack composed of two single cells was used as an example to conduct experiments and analysis, mainly to verify the voltage regulation effect of the switch tube in the bypass. The control flow is shown in Figure 4.

Figure 4 Control flow

　　Since there are no off-the-shelf batteries, it is necessary to use alternative batteries to conduct experiments. During the charging process, the internal resistance and terminal voltage of the battery are constantly changing, and the battery accumulates energy during the charging process. Based on the analysis of the physical properties of the battery and related data, "resistance in series capacitance" is used to replace the single battery to conduct experiments.

　　In this experiment, two low-power NPN tubes C1815 (Q1, Q2) were selected to replace the switching tubes, and the P1.0 and P1.1 pins of the 89C51 chip were used to control the switches of Q1 and Q2. At the same time, the terminal voltages V1 and V2 of the battery are collected by the differential amplifier circuit and sent to the CPU through A/D conversion. During the whole process, the voltage is sampled every 20ms, uploaded to the host computer every 1s and saved, and the curve is automatically drawn. Figure 5 is the test circuit diagram.

Figure 5 Experimental circuit schematic diagram

　　Figure 6 is a curve drawn based on sampling values.

Figure 6 Battery terminal voltage curve during charging

Experimental results and analysis

　　It can be seen from the experimental results that the voltage difference is 1.98V at the beginning of charging. After 140s of charging, the voltage difference is about 0.2V; during the equalizing charge process, the battery voltages tend to be consistent. The equalizing charging method can shorten the inconsistency between battery packs based on the differences in individual cells, thereby improving the overall performance of the battery pack and extending its life.

　　At the same time, judging from the experimental results, this method also has unsatisfactory effects, that is, the voltage difference between the two battery terminals is large. The reasons are: First, in this experiment, a "resistance in series with a capacitor" is used to replace the battery, which is different from a real battery and cannot achieve an ideal simulation state; second, this experiment is mainly to test the balancing effect of the switching of the switching tube on the voltage. , simplified processing in many aspects and ignored some secondary factors, which also have a certain impact on the experimental results.

　　But in general, this experiment achieved the intended purpose and proved the feasibility of the lossless equalizing method.