CN107839500B - Lithium battery pack balance control method and system for dynamically correcting SOC - Google Patents

Abstract

The invention provides a lithium battery pack balance control method and system for dynamically correcting SOC, including: obtaining the SOC of each single lithium battery in the lithium battery pack; calculating the range _rsoc of the battery pack; comparing the range _rsoc The size of the preset range threshold; select the average value of the SOC of all single lithium batteries

As the equilibrium target SOC, for SOC below The single lithium battery is charged for equalization, and the SOC is higher than The single lithium battery performs discharge balance, where dSOC is the balance control band. The invention has the advantages of high modularization degree, fast equalization speed, improved battery use efficiency and extended battery life.

Description

A lithium battery pack balance control method and system with dynamic correction of SOC

technical field

The invention belongs to the field of new energy vehicle batteries, and in particular relates to a lithium battery pack balance control method and system for dynamically correcting SOC.

Background technique

Affected by environmental pollution issues and policies and regulations, new energy vehicles represented by hybrid and pure electric vehicles have received increasing attention. Due to the advantages of high energy density, low self-discharge rate, long life and no pollution, lithium batteries have gradually replaced lead-acid batteries as the main power battery energy in the field of electric vehicles.

When using lithium batteries in electric vehicles, in order to meet the requirements for load capacity and battery life, lithium battery packs must meet certain capacity and voltage requirements. For this reason, single lithium batteries are often connected in parallel to form battery units to solve the problem of insufficient capacity of a single battery; at the same time, a higher voltage is obtained by connecting a single lithium battery unit in series to form a battery pack. Lithium battery packs used in series are often affected by inconsistencies between individual cells or battery cells. The inconsistency of lithium batteries can be defined as the inconsistency of important parameters such as voltage, capacity, internal resistance and self-discharge rate between single lithium batteries of the same specification and the same type. The existence of inconsistency in the battery pack will limit its available power, especially at the end of high-current discharge, the voltage of the battery with large internal resistance drops too fast, which will cause damage to the battery pack, so it is necessary to discharge current of the battery pack. throttling, thereby limiting its output power, resulting in a drop in available power. At the same time, when the battery is discharged with a high current for a long time as the number of times of charging and discharging increases, the aging degree of each single battery is different, resulting in aggravating the inconsistency of each battery.

In order to deal with inconsistencies in lithium battery packs, balancing technology is usually integrated into the battery management system BMS. Balancing technology generally refers to the capacity utilization, output power and service life of battery packs to avoid or reduce battery pack inconsistency problems. Special technical means introduced in the face of adverse effects. The current equalization technology has become the key technology in the battery management system BMS. Efficient equalization measures can improve the effective use capacity of the entire battery and prolong the service life of the battery. The current research on equalization technology is mainly carried out from two aspects: equalization control strategy and equalization circuit topology design. The research on the balance strategy of the battery pack focuses on the establishment of the evaluation index of the inconsistency of the individual cells in the battery pack, and based on this, an effective balance control method is proposed; while the balance circuit topology design focuses on high efficiency, simple control structure, Design and improvement of relatively low-cost equalization circuit structures.

In terms of balancing strategy, there has been a relatively in-depth research on using voltage as the balancing variable in the prior art, while the use of battery state of charge (SOC) as the balancing variable is still a research hotspot, and SOC is used as the balancing variable. Variables can achieve a better balance effect, and the system is also easy to control, which better reflects the real state of the battery pack. However, there are technical problems in the prior art that the SOC estimation accuracy is not high and the accuracy is not real-time.

In terms of equalization circuit topology, passive equalization technology has been applied in actual products due to mature technology and simple structure. However, passive equalization is not suitable in the field of pure electric vehicles that pursue energy efficiency. For active equalization technology, there are still the following problems:

(1) Long equilibration time, which is a common problem in existing equilibration systems. The equilibration time of most equilibration systems is more than an hour, and some even reach several hours.

(2) The balance based on external voltage in the existing common balance technology has been widely studied, but due to the difference in the capacity of the single cells, the external characteristics of the charge and discharge voltage of each single cell are inconsistent, especially in the single cell. In the later stage of charging, the voltage of the single cell rises rapidly, so that the balance criterion is unstable when the external voltage of the battery is used as the criterion for the consistency of the battery pack. At the same time, the study also found that the method has no obvious effect on the increase of the available capacity of the battery pack before and after the balance.

(3) The practicability needs to be improved, the circuit design is relatively complex and the integration is not enough, and the modular expansion cannot be conveniently carried out with the increase of the number of battery cells in series in the battery pack.

SUMMARY OF THE INVENTION

The purpose of the present invention is to overcome the above-mentioned deficiencies in the prior art, and to provide a method and system for the balance control of lithium battery packs with dynamic correction of SOC, which realizes the balance control of series-connected lithium ion battery packs, has a high degree of modularity, and is balanced Fast speeds, improved battery efficiency and extended battery life.

The lithium battery pack balance control method for dynamically correcting SOC proposed by the present invention is characterized in that the method comprises the following steps:

Step 21, obtaining the SOC of each single lithium battery in the lithium battery pack;

Step 22, calculate the range T _soc of the battery pack;

Step 23, compare the size of the range _rsoc and the preset range threshold, when the range _rsoc is greater than the preset range threshold, enter step 24, when the range _rsoc is less than or equal to the predetermined range. When the preset range threshold is described, go to step 25;

作为均衡的目标SOC，对SOC低于的单体锂电池进行充电均衡，对SOC高于

的单体锂电池进行放电均衡，其中，dSOC为均衡控制带；Step 24, select the average value of the SOC of all single lithium batteries

As the equilibrium target SOC, for SOC below The single lithium battery is charged for equalization, and the SOC is higher than

The single lithium battery is discharged for equalization, where dSOC is the equalization control band;

Step 25, end.

Preferably, step 21 specifically includes:

Step 11, determine whether the battery is in a working state, if so, go to Step 12, if not, go to Step 17;

Step 12: Calculate the SOC _i of the current state using formula 2, where SOC _i is the SOC of the current state of the battery, SOC ₀ is the initial SOC when the battery starts to work, _CN is the rated capacity of the battery, I is the battery current, η is the charge-discharge efficiency, when charging, η is a negative number, and when discharging, η is a positive number;

Step 13, determine whether the battery is in a working state, if so, go back to step 12, if not, go to step 14;

Step 14, measure the open-circuit voltage OCV ₁ , look up the correspondence table between SOC and OCV, and obtain the SOC _{1 corresponding to the open-circuit voltage OCV 1 ;}

Step 15, calculate the difference e between SOC _i and SOC ₁ , when the absolute value of e is greater than the preset error threshold, go to step 16, when the absolute value of e is less than or equal to the preset error threshold, output SOC _i ; enter step 11;

Step 16, calculate the corrected SOC, output the corrected SOC correction, and update the correspondence table between SOC and OCV at the same time; go to step 11;

Step 17 , measure the open circuit voltage OCV ₂ , look up the correspondence table between SOC and OCV, obtain the SOC ₂ corresponding to the open circuit voltage OCV ₂ , and output the SOC ₂ ;

Preferably, before step 11, an initial correspondence table between SOC and OCV is established by using an interpolation method.

Preferably, step 16 specifically includes:

Calculate the revised _Sn (i+1) by formula 5, output _Sn (i+1), and update the corresponding _Sn (i) in the correspondence table between SOC and OCV to be _Sn (i+1);

S _n (i+1)=S _n (i)-F(n, e) (0≤n≤50) (Formula 5)

Among them, _Sn (i) represents the value in the table after the current i-th update, _Sn (i+1) represents the value in the table after the i+1-th update, F(n, e) is the correction coefficient, the The coefficient is a function related to n and e, which is expressed by Equation 6;

F(n, e)=a*e*n (Formula 6)

where a is an adjustable constant representing the correction rate, n is the number of points for interpolation, and e is the difference between SOC _i and SOC ₁ .

Preferably, in different segments of the SOC, the value of a is different.

Preferably, in step 24, when the battery is discharged and balanced, the battery that is about to enter the discharge cut-off is charged, so that its SOC is consistent with the SOC of other batteries, regardless of whether the battery is within the cut-off band; when the battery is charged and balanced When the battery is about to enter the end of charging, the equalization circuit is activated to make the SOC of the battery fluctuate in the vicinity, so that finally all the batteries can reach the state of SOC=1 at the same time.

The present invention proposes a lithium battery pack balance control system for dynamically correcting the SOC of the above method. The control system is respectively an upper-level system and a lower-level system, wherein the upper-level system includes a PC and a main control MCU, and the lower-level system includes a plurality of sub-sub-systems, each of which is a sub-sub-system. The system is used for balance control of a small battery pack. Each sub-sub-system includes a secondary MCU and a balance module, which is characterized by:

The main control MCU is used to collect data fed back by each secondary MCU in the lower-level system, transmit the data to the upper-level PC, and forward the commands sent by the PC to the corresponding secondary MCU;

PC, used to receive data sent by the main control MCU, and used to send commands to the main control MCU;

The secondary MCU is used to collect the voltage, current and temperature data of each battery in the small battery pack; feed back the collected data to the main control MCU; calculate the SOC of each single battery; Equalization; when equalization is required, the equalization module of the small battery pack is controlled to equalize the batteries that need to be equalized;

The equalization module is used to equalize the batteries that need to be equalized according to the control of the secondary MCU in the small battery pack.

Preferably, the equalization module adopts a bidirectional flyback transformer.

Preferably, the bidirectional flyback transformer adopts the LTC3300-1 chip.

Preferably, for a single battery with a high SOC that needs to be balanced, the battery is turned on to the secondary transformer switch, and all other switches including the primary switch are turned off. There is current flowing through the secondary winding of the bidirectional flyback transformer. When the electric energy is stored in the secondary winding in the form of magnetic energy; after the SOC in the battery drops to meet the requirements, the secondary change switch is turned off, the primary switch is turned on, the energy is transferred from the secondary winding to the primary winding, and the magnetic energy is converted into Electric energy, thereby transferring excess energy to other batteries in the battery pack; for single cells with low SOC that need to be balanced, open the corresponding primary switch, disconnect all secondary switches, and the primary winding of the bidirectional flyback transformer. When there is current passing through, the electric energy is stored in the primary winding in the form of magnetic energy on the primary side; when enough electric energy is charged, the switch of the primary is turned off, the secondary switch corresponding to the lowest SOC is turned on, and the secondary switch is turned on. The secondary winding, the energy is transferred from the primary winding to the secondary winding, the magnetic energy is converted back into electrical energy and charged into the battery, the SOC of the single battery rises, and the overall SOC of the battery pack returns to a consistent value.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

Figure 1 is a graph of the experimental results of the aging phenomenon of lithium batteries;

Fig. 2 is the flow chart of the improved SOC estimation algorithm;

Fig. 3 is the principle block diagram of the improved SOC estimation algorithm verification platform;

Fig. 4 is the test result graph of the improved SOC estimation algorithm;

Fig. 5 is the judgment flow based on the SOC equalization strategy;

Figure 6 is a schematic diagram of a bidirectional flyback transformer equalization circuit;

Fig. 7 is the equivalent circuit model of the flyback transformer;

Fig. 8 is the equivalent magnetic circuit diagram of the flyback transformer;

Fig. 9 is the current waveform diagram under two working modes;

Fig. 10 is the overall block diagram of the balanced control system;

Fig. 11 is the execution flow chart of main control MCU;

Fig. 12 is the execution flow chart of secondary MCU;

Figure 13 is a schematic diagram of a power supply circuit;

Figure 14 is a schematic diagram of a single cell gate switch circuit;

Figure 15 is a schematic diagram of a voltage inversion circuit;

Figure 16 is a schematic diagram of a voltage-frequency conversion circuit;

Figure 17 is a schematic diagram of a voltage polarity reversal circuit;

Figure 18 is a schematic diagram of the connection of two voltage devices;

Figure 19 is a schematic diagram of an active equalization circuit.

Detailed ways

The present invention will be described in further detail below with reference to the accompanying drawings.

1. Improved SOC estimation algorithm

Under the existing technical conditions, the inconsistency of lithium battery packs is mainly generated in the production process. At the same time, the working environment of lithium battery packs on electric vehicles is generally harsh, resulting in increased inconsistency. This inconsistency is usually manifested by the parameters of battery voltage, capacity and internal resistance, and has a very large impact on the life and performance of the battery pack. Before the balance control of the battery pack, how to evaluate the consistency state of the lithium battery pack is a prerequisite. The quantitative evaluation of the consistency provides important data support for the balance and maintenance of the battery pack.

Electric vehicles characterize the cruising range of the vehicle by estimating the state of charge (SOC) of the battery pack. As a state parameter of the battery pack capacity, SOC reflects the remaining capacity state of the battery pack, which is numerically defined as the ratio of the remaining capacity of the battery to the total rated capacity of the battery:

In Equation 1, Q _R is the charge capacity remaining in the battery and C is the nominal (rated) charge capacity of the battery.

From the definition method of SOC given above, SOC represents the ratio between the remaining capacity of the battery and the available capacity of the battery. If SOC is used as the evaluation basis for inconsistency, it can avoid the existence of available capacity among the individual cells. The goal is to allow all single cells to reach the cut-off voltage for charging and discharging at the same time, ensuring that the available capacity of the battery pack is utilized as much as possible. Ensuring the SOC consistency of all single cells is actually equivalent to ensuring the consistency of the depth of discharge of all cells, thereby avoiding the performance degradation or even scrapping of the entire battery pack caused by the excessive aging of a single cell in the battery pack. question. The difference of SOC can reflect the inconsistency of the battery, and the estimated SOC value can also be obtained in real time during the operation of the battery pack, so the balancing strategy based on SOC can suppress the inconsistency of the battery pack in real time.

When using SOC estimation, the biggest difficulty is the accuracy and real-time performance of SOC estimation. In the early stage of lithium battery discharge, the SOC of the battery changes very little. If it cannot be accurately estimated, the error will be too large due to the accumulation of SOC estimation errors at the end of battery discharge. At this time, if it is rebalanced, the challenge to the equalization control circuit is quite large, and the effect of equalization will be greatly reduced. In addition, the real-time estimation of SOC requires a certain amount of calculation. When the number of batteries in the battery pack is large, in order to ensure the real-time performance of SOC, it is necessary for the balancing system to use an MCU with certain computing power.

There is no way to directly measure the remaining power inside the battery, and it is only possible to estimate the remaining power of the battery through the amount of data that the battery can measure, such as voltage, current, and internal resistance. However, the chemical characteristics inside the battery are very complex, so there is no linear or other simple functional relationship between these data and SOC. The currently proposed estimation methods for SOC always have more or less defects. It also makes the estimation of the SOC state of the battery pack a key and difficult point in the battery management system. The SOC estimation method in the prior art is as follows:

(1) An hour law

The ampere-hour method is the most commonly used method for estimating battery SOC parameters. The core idea of this method can be expressed by formula 2:

Among them, SOC _i is the SOC of the current state of the battery; _CN is the rated capacity of the battery; I is the battery current; η is the charge-discharge efficiency. By calculating the integral of the current over time, the amount of power lost over a period of time can be obtained. With the known initial SOC ₀ , the SOC at a certain moment can be obtained. The principle and implementation of this method are very simple, but it has the following defects: first, the current measurement accuracy is required to be high. If the current measurement is inaccurate, it will lead to SOC calculation errors, which will accumulate for a long time, and the errors will gradually be amplified; The estimation of SOC ₀ also requires a certain method; finally, when the temperature is high or the current fluctuates greatly, the error of this method is relatively large. Therefore, in order to obtain a reliable SOC estimation value, a high-performance current sensor is required to obtain an accurate current value, and sufficient data is required for estimating the initial state.

(2) Open circuit voltage method

The electromotive force of the battery can be considered to be composed of three parts, including the open-circuit voltage (OCV) of the battery, the ohmic voltage of the battery and the polarization voltage of the battery. When the battery is in a non-working state, and after a long period of standing, the ohmic voltage and polarization voltage of the battery will drop to 0. At this time, the open circuit voltage OCV of the battery is equal to the terminal voltage of the battery, which is the electromotive force of the battery, so The SOC can be estimated from a relationship curve between OCV and SOC. In real-time, there is a fairly good linear relationship between OCV-SOC on lead-acid batteries, and this method can be used to estimate SOC more accurately. However, for lithium batteries, the linear relationship between the two is not so obvious, so it is necessary to establish a more complex relationship control table.

The obvious disadvantage of the open-circuit voltage method is that the battery needs to be fully rested before the measurement, which takes several hours or even a dozen hours, which makes the measurement difficult; secondly, the length of the resting time is also difficult to determine; As the battery ages, the correspondence between the open circuit voltage OCV and SOC also changes. These reasons make it impossible to obtain the SOC online using the open-circuit voltage method in practical use. The open circuit voltage method can be combined with the ampere-hour method as a method for obtaining the initial SOC value in the ampere-hour method. However, the relationship between the open circuit voltage and SOC will change with the aging of the battery, and it is impossible to always re-measure the relationship between the two in actual use. Therefore, a method that can dynamically correct the corresponding relationship between the open circuit voltage OCV and SOC is required. algorithm.

(3) Load voltage method

The principle of the load voltage method is the same as that of the open circuit voltage method, and it is proposed to overcome the disadvantage that the open circuit voltage method cannot estimate the battery SOC online. Its principle is as follows: if the internal resistance r and the working current I of the battery can be obtained, the balance electromotive force EMF of the battery can be calculated according to the following formula by measuring the voltage U across the load R.

EMF=U+I*r (Formula 3)

From the analysis of the open circuit voltage method, it can be known that the corresponding relationship between EMF and SOC is the relationship between OCV and SOC in the open circuit voltage method, so after knowing the EMF, the SOC of the battery can be correspondingly obtained.

In theory, this method does overcome the shortcoming that the open-circuit voltage method cannot measure SOC in real time, but in practice, this method still has obvious defects: First, there are many factors that affect the internal resistance r of the battery, and the internal resistance of the battery It is inherently inconsistent, and the internal resistance between individual cells may vary greatly, so it is difficult to accurately obtain the internal resistance of the battery; secondly, the method itself is based on the open-circuit voltage method, so the open-circuit The problems faced by the voltage method cannot be avoided by this method. The load voltage method is rarely used to obtain SOC online in the field of electric vehicles, but is often used as a basis for judging the battery charge and discharge cut-off.

(4) Internal resistance method

The basic idea of the internal resistance method is consistent with the open circuit voltage method. A large number of experimental studies have shown that there is a close relationship between the AC impedance or DC internal resistance of a lithium battery and the SOC of the battery. If such a relationship can be determined through some battery samples , then the SOC of the battery can be obtained by detecting the internal resistance of the battery.

The internal resistance of the battery can be divided into AC impedance and DC internal resistance. The AC impedance reflects the resistance of the battery to the alternating current, which can be measured by an AC impedance meter. Similarly, the DC internal resistance indicates the resistance of the battery to DC, which can be obtained by detecting the change in voltage and current in a short period of time.

However, in the previous analysis of the load voltage method, it was also mentioned that the relationship between the internal resistance of the battery and the SOC is very complicated. It is not only affected by the SOC, but also by many factors such as temperature and battery health. The relationship between resistance and SOC is difficult to really determine. At the same time, the internal resistance of the battery is often very small, only at the milliohm level, which requires very high measurement accuracy, and the measurement error has a great impact on the results. Therefore, this method is rarely used in practical applications of electric vehicles.

(5) Neural network method

The neural network method is to estimate the battery SOC by training data through a large number of samples under the premise of building a network model. The battery is a highly nonlinear system. The neural network method has nonlinear basic characteristics and can simulate the nonlinear dynamic characteristics of the battery well. Therefore, the neural network method has a good effect in estimating SOC. The neural network method to estimate SOC needs to train a large amount of sample data, in which the training sample data and the training method will affect the estimation result. Another defect is that the neural network method requires a lot of resources, which puts forward more requirements for the design of the battery management system. High requirements often require the use of higher-performance control chips, which greatly increases the cost.

(6) Kalman filter method

The Kalman filter method takes the battery system as a nonlinear dynamic system, in which the SOC of the battery is only one state in the system, establishes the battery model accordingly, lists the state equation and the observation equation according to the model, and uses the extended Kalman filter. method to estimate battery SOC. The basic idea of this method is to make the optimal estimation on the minimum variance for the state of the dynamic system. It solves the problems of inaccurate initial SOC estimation and accumulated error in the ampere-hour method. If the battery model can be established accurately, the Kalman filter method can accurately estimate the SOC of the battery. However, this method also has the following problems: firstly, the accuracy of estimating battery SOC depends on the accuracy of the battery model; secondly, due to the application of a large number of matrix operations in the Kalman filter method, it also requires the operating speed of the system processor. improved.

By summarizing the existing SOC estimation methods, the following are commonly used: ampere-hour method, open-circuit voltage method, internal resistance method, neural network method and Kalman filter method. The above methods have their own unique characteristics in different use environments and for different power batteries, but they also have different defects and shortcomings. Table 1 summarizes the advantages and disadvantages of common SOC estimation methods.

Table 1 Summary of common SOC estimation methods

To sum up, the open-circuit voltage method in the common SOC algorithm needs to estimate the SOC under the non-working state of the battery, so it is difficult to meet the requirements of electric vehicle power online estimation of SOC, and it is suitable to estimate SOC in combination with other methods. The ampere-hour method has the problems of increasing dependence on the initial value and accumulative error, and cannot deal with the self-discharge problem of the battery. The Kalman filter method solves the problem of inaccurate SOC initial value estimation and cumulative error in the ampere-hour method. At the same time, the main problem is that it is highly dependent on the battery model, and only an accurate battery model can be established. Accurately estimate the SOC of the battery; and because a large number of matrix operations are used in the Kalman filter method, it requires a higher speed of the system processor. The main problem of estimating SOC with the neural network method is that a large amount of sample data needs to be trained, so the estimation error will be affected by the training sample data and the training method. put forward higher requirements.

In actual use, the lithium battery will age with the increase of the number of cycles. The aging phenomenon of lithium battery is manifested in the change of internal parameters of the battery, and the most important thing is the attenuation of battery capacity. Such attenuation can be seen by comparing the relationship between the open circuit voltage and SOC of lithium batteries with different cycles. In the present invention, the 3400mAh lithium iron phosphate battery was tested between the OCV and SOC of a new battery and after 500 cycles. The results of the experiment are shown in Figure 1.

It can be seen from Figure 1 that there is a certain correspondence between the open circuit voltage OCV and SOC of the lithium battery, but at the same time, the gradual aging of the lithium battery will cause the relationship between the open circuit voltage OCV and the battery SOC to gradually change. With the increase of charge-discharge cycles, the battery gradually ages, and the capacity of the battery decreases continuously. It can also be seen from Figure 1 that at the same open circuit voltage OCV, the OCV-SOC curves of the new battery and the aged battery correspond to different SOCs. Therefore, when using the open circuit voltage as the basic estimation algorithm of SOC, it is necessary to add a certain correction algorithm to deal with the change of the relationship between OCV and SOC.

Considering the influence of the aging factor of the lithium battery, and because the embedded chip used in the lithium battery management system is usually relatively weak in computing performance, the improved SOC estimation algorithm proposed in the present invention integrates the ampere-hour method and the open circuit voltage method, while adding A certain dynamic correction algorithm is developed to keep the SOC estimate within an acceptable error range to deal with the effects of lithium battery aging.

As mentioned above, the ampere-hour method is a commonly used method, and its source of error lies in the estimation of the initial SOC and the measurement accuracy of the current during use. The measurement accuracy of the current can be improved by replacing the high-precision measuring device, and the high-precision initial SOC can be measured by the open-circuit voltage method. However, due to the aging effect of the battery, the corresponding relationship between the open circuit voltage of the battery and the SOC changes dynamically, so the core of the improved SOC estimation algorithm proposed by the present invention is to dynamically maintain the corresponding relationship table between SOC and OCV. During the use of the battery, if the difference between the measured value of the battery SOC and the value in the corresponding relationship table is greater than a preset error threshold, the corresponding relationship table is revised.

In the present invention, the charging and discharging process of the battery is divided into three stages: before charging and discharging, during charging and discharging, and after charging and discharging. Since the open-circuit voltage measurement requires the battery to stand for a long time, the open-circuit voltage method is only suitable for estimating the SOC when the battery is not in use, that is, before or after charging and discharging. The SOC of these two stages is obtained by looking up the correspondence table between SOC and OCV. During charging and discharging, the SOC cannot be directly obtained, but the amount of electricity discharged or charged can be calculated by the ampere-hour method, and then obtained with the SOC before charging and discharging. After charging and discharging, the SOC after charging and discharging can be calculated by the ampere-hour method. After standing at the same time, a theoretical SOC can also be obtained by looking up the table by the open circuit voltage method. There will be a deviation between the calculated SOC and the theoretical SOC. If the difference is greater than the preset error threshold, the corresponding relationship table between SOC and OCV is updated.

The flowchart of the improved SOC estimation algorithm of the present invention is shown in FIG. 2 . Specifically:

Step 11, determine whether the battery is in a working state, if so, go to Step 12, if not, go to Step 17;

Step 12: Calculate the SOC _i of the current state using formula 2, where SOC _i is the SOC of the current state of the battery, SOC ₀ is the initial SOC when the battery starts to work, _CN is the rated capacity of the battery, I is the battery current, η is the charge-discharge efficiency, when charging, η is a negative number, and when discharging, η is a positive number;

Step 13, determine whether the battery is in a working state, if so, go back to step 12, if not, go to step 14;

Step 14, measure the open-circuit voltage OCV ₁ , look up the correspondence table between SOC and OCV, and obtain the SOC _{1 corresponding to the open-circuit voltage OCV 1 ;}

Step 15, calculate the difference e between SOC _i and SOC ₁ , when the absolute value of e is greater than the preset error threshold, go to step 16, when the absolute value of e is less than or equal to the preset error threshold, output SOC _i ; enter step 11;

Step 16, calculate the corrected SOC, output the corrected SOC correction, and update the correspondence table between SOC and OCV at the same time; go to step 11;

Step 17 , measure the open circuit voltage OCV ₂ , look up the correspondence table between SOC and OCV, obtain the SOC ₂ corresponding to the open circuit voltage OCV ₂ , and output the SOC ₂ ;

For example, if the battery is discharged, its open circuit voltage can be measured before discharge. Then, by looking up the correspondence table between SOC and OCV, you can know the SOC at this time, which is recorded as S1. When the battery starts to discharge, the recording current starts to calculate the consumed power, and finally the total discharged power is obtained, which is recorded as Sc. Through calculation, the power at the last moment can be known, S2=S1-Sc. When the discharge is over, the theoretical SOC is obtained by measuring the open circuit voltage and the data in the comparison table, which is recorded as S3. The error value Sb is defined as the difference between S3 and S2, ie Sb=S3−S2. When Sb exceeds the preset error threshold, the correction process is started to correct the data in the correspondence table between SOC and OCV.

When starting to use the algorithm, before step 11, an initial SOC and OCV correspondence table needs to be established. In order to obtain such a table, and considering the limitations of space and computing power, the interpolation method is chosen to obtain such a table. 20 points can be selected, that is, the value of the open circuit voltage is measured at each 5% SOC. If you need to be denser, you can also choose 50 points or more, depending on the needs and available memory on the chip.

The corrected SOC and OCV correspondence table in step 16 can be completed by formulas four and five.

e=SOC _Table -SOC _Cal (Formula 4)

S _n (i+1)=S _n (i)-F(n, e) (0≤n≤50) (Formula 5)

The SOC _Table in Formula 4 is the original SOC in the table, while SOC _Cal is the calculated SOC, and finally an e is used to represent the difference between the two. S _n (i) in Formula 5 represents the value in the table after the current i-th update, and _Sn (i+1) is the value in the i+1-th table after the update. F(n, e) is the correction coefficient, which is a function related to n and e. What is adopted in the present invention is the representation of a primary function:

F(n, e)=a*e*n (Formula 6)

Among them, a is an adjustable constant, indicating the correction rate. It can be seen from Figure 1 that the voltage correspondences in different SOC segments are different, so the effects of aging under different SOCs are also different. Therefore, the correction rate in different segments should be different, that is, a should take different values in different segments. n is the number of points to interpolate. If necessary, the functional form of F(n) can also be changed, and the updated functional form can be modified according to different battery characteristics.

In order to verify the validity and practicability of the revised SOC estimation algorithm proposed by the present invention, it is necessary to verify whether the correspondence table between the SOC and OCV calculated after the battery has been charged and discharged for many times is consistent with the current real correspondence table of the battery, and the error is what range. To this end, the present invention designs an experimental platform to verify the effectiveness of the SOC estimation algorithm. The principle block diagram of the platform is shown in Figure 3 . The invention takes 18650 lithium battery cells with a rated capacity of 3400mAh and a rated voltage of 3.7V as the test object. In order to see obvious attenuation in the experiment, the number of battery charge and discharge cycles needs to be more, and 500 cycles were cycled in this experiment. At the same time, in order to avoid the influence of temperature on the test results, the temperature was controlled at 25 degrees Celsius during the test. In order to restore the battery to a resting state, it is necessary to rest the battery for more than 30 minutes when measuring the open circuit voltage. At the same time, in order to reduce the influence of the voltage drop of the internal resistance during discharge, the battery uses a low-speed discharge mode during discharge, so the voltage curve can be directly used to estimate the open-circuit voltage curve.

The results of the test are shown in Figure 4. As can be seen from Figure 4, there is a large gap between the two curves of the new battery and the aging battery. If the correction algorithm is not added, the relationship between the voltage and capacity of the new battery will still be used after the battery is aging, which will cause the error in the estimated SOC value to increase as the battery continues to age, which is very unfavorable for extending the battery. service life. When the correction method is added, the calculated curve is in good agreement with the curve of the aging battery, which indicates that the algorithm in the present invention can dynamically correct the corresponding relationship between the voltage and the capacity as the battery ages. This can greatly help reduce the error in SOC estimation caused by battery aging.

2. Equilibrium control strategy

As mentioned above, the SOC estimation algorithm is one of the core technologies of the balancing strategy in the balancing technology. After the above-mentioned revised SOC estimation algorithm is determined, the balancing strategy adopted in the present invention will be introduced below. The balance strategy and the balance circuit topology design are the two most important key points in the balance control system. The balance circuit topology needs to cooperate with a reasonable balance strategy to achieve a good balance effect. The two are complementary parts of the balance control system. In the present invention, an equalization circuit based on a bidirectional flyback transformer is adopted, and a set of equalization strategy is designed for the circuit at the same time.

According to the above analysis and research, in order to achieve a good balance effect and better achieve the design goal of prolonging the life of the battery pack in the present invention, SOC is used as the variable of the balance control in the present invention, and a quantitative standard needs to be adopted in the balance control system. to enable or disable the equalization function. In order to evaluate and identify the inconsistent state of the battery pack, the present invention analyzes two different quantitative indicators, the variance δ _soc ² and the range r _soc , and the expressions of the two are consistent with those used in statistical mathematics, as follows:

r _soc =max(SOC(i))-min(SOC(i)), i=1...n (Equation 9)

From the perspective of statistics and mathematics, the degree of dispersion of the SOC values between the individual cells in the battery pack is represented by the variance, that is, the smaller the variance is, the smaller the dispersion degree of the individual cells in the battery pack relative to the average battery SOC is. The smaller the difference of SOC values in the battery pack; on the contrary, if the variance is larger, it means that the dispersion degree of each single cell in the battery pack relative to the average SOC value is higher, and the SOC value of each single cell is more different. Therefore, if the variance is used as the quantitative standard for evaluating the consistency of lithium battery packs, good results can be obtained theoretically. However, the amount of calculation of variance or standard deviation is quite large. Considering that the embedded chips used in the balance control system are often not powerful in computing performance, and the consistency of the battery pack needs to be evaluated frequently or even in real time, so using Variance is not suitable as a quantitative criterion for systematic consistency evaluation.

The range indicates the difference between the maximum SOC and the minimum SOC in the battery pack. When the value is small, it means that the difference between the maximum SOC and the minimum SOC in the battery pack is small, that is to say, the SOC of each single battery is small. Distributed in a small range, it can show that the consistency of the battery pack is better. When the extreme value is large, the gap between the maximum SOC and the minimum SOC of the battery pack is large, and the SOC of the battery pack may be distributed in a relatively wide range, indicating that the consistency of the battery pack may be poor. At the same time, it is not necessary to take into account the SOC conditions of all the cells of the battery pack when calculating the range, and it is only necessary to find the maximum and minimum SOC values, which greatly reduces the amount of calculation.

It can be known from the above analysis that the variance and range of SOC can reflect the consistency of the battery pack, so it is appropriate to use the variance and range as the evaluation criteria to quantify the consistency of the battery pack. However, considering the computing power of the balanced control system, the range can greatly reduce the calculation of the system and speed up the response speed of the system, so the range is more suitable for use in the present invention than the variance.

作为均衡的目标SOC，使用均值作为目标可以提高均衡的效率并且充分发挥充放电均衡的优势。本发明在使用SOC作为均衡控制手段时，设置了一个均衡控制带(dSOC)以防止均衡的波动，本发明中采用了1％作为截止带控制dSOC。接着对SOC高于

的单体锂电池进行放电均衡，同样的对低于

的电池组进行充电均衡。该过程可以通过流程图示意，如图5所示。具体为：In the SOC-based balancing strategy, the purpose of battery pack balancing is mainly achieved by reducing the SOC difference between batteries. The SOC of each single lithium battery is used as the main control object, and the battery is reduced by charging and discharging the single battery. The difference between SOC. The balancing process used in the present invention is as follows: at the beginning of balancing, the SOCs of all batteries are measured, and one of them is selected as the target SOC for balancing. Usually, however, the mean is chosen

As the target SOC of the equilibrium, using the mean value as the target can improve the efficiency of the equilibrium and give full play to the advantages of the charge-discharge equilibrium. In the present invention, when SOC is used as the balance control means, a balance control band (dSOC) is set to prevent the fluctuation of the balance. In the present invention, 1% is used as the cut-off band to control the dSOC. Then for SOC higher than

The single lithium battery discharges equalization, and the same is less than

charge equalization of the battery pack. The process can be illustrated by a flowchart, as shown in FIG. 5 . Specifically:

Step 21: Obtain the SOC of each single lithium battery in the lithium battery pack, and specifically obtain the SOC of each single lithium battery through the above-mentioned improved SOC estimation algorithm;

Step 22: Calculate the range _rsoc of the battery pack, that is, calculate the difference _rsoc between the maximum SOC and the minimum SOC in each single lithium battery of the battery pack;

Step 23, compare the size of the range _rsoc and the preset range threshold, when the range _rsoc is greater than the preset range threshold, enter step 24, when the range _rsoc is less than or equal to the predetermined range. When the preset range threshold is described, go to step 25;

作为均衡的目标SOC，对SOC低于的单体锂电池进行充电均衡，对SOC高于

的单体锂电池进行放电均衡，其中，dSOC为均衡控制带，本发明中采用1％作为截止带控制dSOC；Step 24, select the average value of the SOC of all single lithium batteries

As the equilibrium target SOC, for SOC below The single lithium battery is charged for equalization, and the SOC is higher than

The single lithium battery is discharged and balanced, wherein dSOC is the balance control band, and 1% is used as the cut-off band to control the dSOC in the present invention;

Step 25, end.

Some special nodes need attention in this process. When the battery is in a state of discharge, the battery with poor physique will enter the discharge in advance to reach the cut-off voltage, while the battery with good physique will have a part of the power remaining, which leads to the capacity of the battery pack not being fully used. In order to solve this problem and at the same time protect the life of each battery so that it cannot be in an over-discharge state, the solution in the present invention is to charge the battery that is about to enter the end of discharge so that its SOC is consistent with the SOC of other batteries , regardless of whether the cell is within the cutoff band or not. In this way, the batteries in the entire battery pack can reach the state of SOC=0 at the same time, and the capacity of the battery pack can be fully utilized.

Similarly, when the battery is in the charging state, the battery with poor physical fitness will enter the fully charged state in advance and reach the cut-off voltage, so that the capacity of the battery pack is also wasted. Therefore, in order to prevent these batteries from entering the cut-off voltage in advance, the present invention The countermeasures taken in the battery are: when it is detected that a battery is about to enter the end of charging, the equalization circuit is activated to make the SOC of the battery fluctuate near the battery SOC, so that finally all the batteries can reach the state of SOC=1 at the same time.

3. Equalization circuit

The present invention selects a bidirectional flyback transformer equalizing circuit. The main reason is that the equalizing circuit has large current and fast equalizing speed, and can realize two-way equalization. A schematic of a typical bidirectional flyback transformer equalization circuit is demonstrated in Figure 6.

The essence of flyback transformer equalization is to realize the bidirectional transfer of energy between battery cells through the mutual conversion of electrical energy and magnetic energy. When a single cell of the battery pack has more energy than other batteries, the flyback transformer is used as the energy transfer medium, and the excess energy of the battery is transferred to the entire battery pack; When there are few other batteries, the flyback transformer is also used as the energy transfer medium to input the energy of the whole battery pack to the single battery, so as to prevent the battery energy from being too low and causing harm to the battery. This structure has a balanced way of two directions, as follows.

(1) Equalization of single cell to battery pack (top equalization)

After the equalization control system detects the SOC of all the single cells, for the single cell with higher SOC that needs to be balanced, the battery is turned on to the secondary transformer switch, and all other switches including the primary switch are disconnected, and the flyback transformer is turned off. There is current passing through the secondary winding of the battery, and the electric energy is stored in the secondary winding in the form of magnetic energy; after the SOC in the battery drops to the required value, the secondary switch is turned off and the primary switch is turned on, so that the energy will flow from the secondary The secondary winding is transferred to the primary winding, and the magnetic energy is converted into electrical energy and transferred to the entire battery pack, thus controlling the battery with the highest SOC, while transferring excess energy to other batteries in the battery pack.

(2) Balance of battery pack to single cell (bottom balance)

After the balance control system detects the SOC of all single cells, for the single cells with low SOC that need to be balanced, the corresponding primary switch is turned on, and all secondary switches are disconnected, and there will be current in the primary winding of the flyback transformer. At this time, the electric energy is stored in the primary winding in the form of magnetic energy on the primary side; when enough electric energy is charged, the primary switch is turned off, and the secondary change switch corresponding to the lowest SOC is opened, and the secondary winding is turned on. In this way, the energy is transferred from the primary winding to the secondary winding, and the magnetic energy is converted back to electric energy and charged into the battery. Through this process, the SOC of the single battery can be recovered, and the overall SOC of the battery pack can also return to a relatively consistent value.

Circuit and magnetic circuit analysis of the flyback transformer can wait until the equivalent circuit model shown in Figure 7 and the equivalent magnetic circuit diagram shown in Figure 8.

When the MOS switch is turned on, it can be seen from the left side of Figure 8 that the magnetoresistance _RM in the iron core is connected in parallel with the leakage magnetoresistance R _S , the magnetic equal potential is N ₁ i ₁ , and the magnetic flux Φ _M generated on _RM passes through After passing through the iron core and disconnecting the MOS tube, the current i ₂ flowing in the secondary winding generates a magnetic flux Φ _M . It can be seen in the right part of Fig. 8 that at the moment of the transformation again N ₁ i ₁ =N ₂ i ₂ . If N ₁ : N ₂ =1:1, the equivalent circuit model in FIG. 7 is transformed from the equivalent magnetic circuit model of the flyback transformer, and the leakage inductance L _S of the primary and secondary in the figure is the same. Due to the existence of an air gap in a flyback transformer, its iron core has a small inductance value, and the leakage inductance L _S in a flyback transformer cannot be ignored. After the MOS tube is turned on, i ₀ flows through the primary _LS and _LM , and no current flows through the secondary at this time. When the MOS tube is turned off, all the energy of _LS in the leakage _inductance of the input side and a part of the energy of LM are consumed in the absorption network (clamping circuit: in the flyback transformer, it can reduce the voltage stress on the switching tube), and _A part of the energy remaining in the _LM and all the energy in the secondary LS are output through the secondary.

In Figure 6, when the MOS transistor S is turned on, the input voltage U _i is applied to both ends of the primary winding of the transformer. According to Lenz's law, it can be known that the secondary winding generates a positive and negative induced electromotive force at this time, and the diode D ₂ cannot be turned on. Therefore current cannot flow in the secondary circuit. At this time, the primary winding of the transformer is equivalent to an inductance. Assuming that the inductance of the primary winding is Lp and the inductance of the secondary winding is Ls, the current flowing through the primary winding during the conduction period of the MOS transistor is:

At t=t _on , the primary winding current reaches its maximum value:

When the MOS tube is turned off, according to Lenz's law, it can be known that the voltage polarity of the secondary winding turns to upper positive and lower negative. At this time, the diode D2 is turned _on , the magnetic energy stored in the transformer is converted into electric energy, and there is current in the secondary winding. flowing, the current is:

When t=t _off , the current of the secondary winding reaches the minimum value _Ismin . When I _smin = 0, the energy stored in the magnetic field during the conduction period of the MOS transistor is completely released, and this process is called the intermittent operation mode of the flyback transformer; when I _smin >0, the stored energy in the magnetic field during the conduction period of the MOS transistor is called The energy in the transformer is not completely released, this process is called the continuous operation mode of the flyback transformer. The current waveforms of the above two operating modes are the same as shown in Figure 9.

The magnetic flux in the magnetic core of the transformer must return to its original position at the end of each cycle. This principle is called the magnetic flux reset principle. There is residual magnetism in continuous operation mode. In theory, the magnetic flux at the end of each cycle should be maintained. However, due to the iron loss in the magnetism and the copper loss in the coil winding, the temperature rises during use, causing the initial value of the magnetic flux to be shifted and unable to be reset, which eventually leads to the change of the magnetic flux into a nonlinear In this area, the inductance decreases, the current value increases, the magnetic core is easy to reach a saturation state, the transformer cannot work normally, resulting in great insecurity of the circuit. The transformer is smaller in continuous operation mode and allows larger primary and secondary currents. Therefore, the flyback transformer used in the present invention works in an intermittent working mode.

The primary side of the flyback transformer used in the present invention is connected to a battery pack consisting of 6 single lithium iron phosphate batteries in series, each secondary side is connected to each single battery, and the rated voltage of each single battery is 3.6V Therefore, the working voltage of the primary side of the transformer is about 18~24V, the working voltage range of the secondary side is 4.2~3.0, the working efficiency of the transformer is designed to be 80%, and the working frequency is 10KHz.

(1) Maximum duty cycle

Under normal circumstances, the output efficiency of the transformer increases with the increase of the duty cycle, but when the duty cycle exceeds 50%, the circuit will oscillate. Although this phenomenon can be improved by adding a harmonic compensation module to the circuit, if the appropriate components are not selected and arranged reasonably, the harmonic compensation module in the circuit may not work at this time, resulting in the work of the circuit. The state remains unstable at duty cycles greater than 50%. Therefore, the maximum duty cycle of the transformer is generally between 40% and 50%, and the maximum duty cycle of the present invention is finally selected to be 45%. The actual duty cycle used is also obtained through simulation.

(2) turns ratio

The turns ratio N of the transformer is the ratio of the number of turns Np of the primary winding of the transformer to the number of turns Ns of the secondary winding of the transformer.

At the beginning of the design, the primary and secondary turns of the transformer cannot be directly known. In the present invention, the transformation ratio N of the transformer is directly determined according to the reflected voltage of the transformer shown in Formula 13 as:

In the above formula, the reflected voltage V _OR indicates that when there is current flowing through the secondary winding, an opposite voltage is formed on the primary winding, V _o indicates the output voltage, and V _f is the voltage drop of the MOS tube. V _OR can be calculated by the following formula.

According to Formula Fourteen and Formula fifteen, N=5.2 can be calculated, and N=5 is taken.

Experiments have shown that when the duty cycle is greater than 40%, the equalization current is too large, which does not meet the hardware conditions. When the duty cycle is 20%, the secondary side current is less than 5A, which is inconsistent with the goal of the present invention. Therefore, the top equalization is performed. The duty cycle should be between 25% and 35%. At the same time, when the duty cycle is greater than 35%, the transformer works in continuous mode, so it is not suitable, so the duty cycle should be selected between 20% and 30% in the bottom equalization. To sum up, choosing a duty cycle of 25% or 30% is a suitable range for both top and bottom equalization.

4. Balance control system

The present invention adopts the K64 chip newly launched by NXP as the control chip at the MCU end, and the latest embedded chip with powerful performance, which provides a good performance guarantee for the development of the SOC estimation algorithm and the equalization strategy of the present invention. At the same time, a voltage measurement module, a current measurement module, a temperature measurement module and an equalization module are designed with the chip as the core. Among them, the equalization module uses the LTC3300-1 chip specially designed for the active equalization circuit of the bidirectional transformer. With the advantages of the integrated chip, the equalization module has further improved the circuit complexity and cost control.

Lithium battery packs often need to combine hundreds of single-cell lithium-ion batteries. Considering the flexibility of battery pack distribution, the entire battery pack needs to be divided into many small battery packs and installed in different battery boxes. At the same time, in order to expand the capacity , the new battery can be easily expanded into the battery pack in the future, and the present invention adopts the design of modular expansion. The overall design block diagram is shown in Figure 10.

The whole system is divided into upper and lower levels in design. The secondary MCU implements the function of monitoring the batteries of this group, including: collecting data such as voltage, current and temperature of each battery in the group; feeding back the collected data to the main control MCU; calculating the SOC of each single battery; According to the SOC, it is judged whether equalization is required; when equalization is required, the equalization module of the group is controlled to equalize the batteries that need to be equalized. The main control MCU is responsible for collecting the data fed back by the lower-level MCU, and at the same time transmitting the data to the upper-level PC, so as to facilitate the data collection and debugging of the entire battery pack. The main control MCU can also forward the commands sent by the PC to the corresponding battery pack. The execution flow of the main control MCU is shown in FIG. 11 , and the execution flow of the secondary MCU is shown in FIG. 12 .

Adopting such a hierarchical and modular design has the following advantages:

(1) Improve the scalability of the system. Lithium battery packs need to be combined with different numbers of batteries when used. For example, in order to sell, car manufacturers often classify the price of the same model according to different cruising ranges. Different cruising ranges need to integrate different quantities in the battery pack. If different balance control systems are used for the same vehicle model, it will not only increase the cost of early research and development, but also need to invest more energy and cost in the maintenance of different balance control systems. Therefore, the use of hierarchical design can enhance the scalability of the system.

(2) Improve the real-time performance of the system. The MCU in each small battery group only needs to manage the batteries in this group, which greatly reduces the calculation amount of each subordinate MCU and enhances the real-time performance of the balanced control system.

(3) Improve the compatibility of the system. If the same large battery pack needs to use small battery packs from different manufacturers, it is only necessary to re-debug or design the equalization module of the group, avoiding large-scale modifications to the entire system.

(4) Increase the reliability of the balanced control system. The modular design can avoid the paralysis of the entire system. When the balancing module in a small battery pack fails, other small battery packs can still work normally. This not only protects the batteries in other small battery packs, but also prevents more serious accidents.

In terms of specific hardware circuits, the present invention adopts the following design methods:

power circuit

The function of the power supply module is to provide the normal working power supply for the secondary MCU. The circuit schematic diagram of the power supply module designed by the present invention is shown in FIG. 13 . The normal working voltage of the main control chip K64 of the secondary MCU is between 1.71V and 3.6V. Usually, the power supply of the chip is guaranteed to be about 3.3V. The balancing system of the present invention also needs to use a voltage of 5V. Since both the secondary MCU and the equalization module are set in the small battery pack, power can be drawn directly from the small battery pack. A small battery pack usually consists of 6 single cells, and the working voltage of each battery is between 3.0V and 4.2V, so the terminal voltage of the small battery pack is between 18V and 25.2V. This system adopts the LM2576 voltage conversion chip of NI Company. The LM2576 chip can accept voltage input from 7 to 40V, output voltage of 5V, and can drive a load of 3A. The linearity and load adjustment capabilities are very strong. At the same time, the LM2576 also integrates a frequency compensator and a fixed frequency oscillator. Components can complete a good voltage output. In Figure 13, a 5V to 3.3V circuit is also included, because the usual working voltage of K64 is 3.3V, so an ASM1117-3.3V stabilized power supply module needs to be used to convert 5V to 3.3V and output it to K64 for power supply. The two power indicator lights added to the circuit are used to indicate whether the two power sources are working normally.

Voltage acquisition circuit

The accurate voltage acquisition circuit is not only related to the normal use and monitoring of the battery pack, but also a necessary guarantee for accurate balance judgment of the battery. Therefore, when lithium batteries are connected in series to form a battery pack, it is necessary to accurately measure the voltage of each cell in the battery pack.

The battery voltage acquisition methods commonly used in series lithium battery packs include common mode measurement method and differential mode measurement method. Among them, the common mode measurement is to measure the voltage of the battery pack with the proportional attenuation of the precision resistance relative to the same reference level, and then subtract it in turn to obtain the voltage of each single cell. The advantage of this method is that the circuit is simple, but the measurement accuracy of this method depends on the voltage divider resistance, which is easily affected by temperature and produces serious cumulative errors. Therefore, this method is only suitable for a small number of batteries in series and does not require high measurement accuracy. the occasion. It is not suitable for situations that rely on and voltage to calculate accurate SOC.

In the present invention, the method of differential mode measurement is selected, and in this method, each battery is sequentially selected for measurement by a certain method. This method is suitable for occasions where there are many batteries connected in series and high precision is required. Another advantage of using this differential mode measurement method is that when the acquisition of a certain channel fails, it does not affect the normal operation of other channels. In addition, compared with the acquisition method using an integrated chip, this sub-channel measurement method only needs to repair the corresponding faulty channel in the event of a failure, rather than replacing the entire chip, which is of great benefit for reducing later maintenance costs. of.

(1) Single battery selection circuit

The method of differential mode measurement needs to be able to gate each single cell. In the present invention, MOSFET PS7241-2A is used as the single cell gate switch for voltage acquisition. PS7241 series devices are composed of light-emitting diode (input side) and normally open contact MOS tube (output side). Each PS7241-2A contains two independent gate switches. The device is characterized by low operating current, high withstand voltage and very fast response speed. The schematic diagram of the single cell gate switch circuit is shown in Figure 14.

The resistors R1 to R4 in Figure 14 are current-limiting and voltage-dividing resistors, which are used to limit the magnitude of the current in the circuit during the measurement process. When it is necessary to measure the voltage of a single battery, it is only necessary to pull up the two pins 1 and 3 on the corresponding PS7241 through K64. At this time, the voltage at both ends of the corresponding battery cell will change from 6 and 8 of the PS7241. output on two pins. Taking the circuit in Figure 14 as an example, when you need to measure the voltage of bat1, pull up pins 1 and 3 in PS1. At this time, the output from 6 and 8 is the voltage across bat1, of which pin 6 outputs the positive pole of the battery. Pin 8 is the negative pole of the battery. When the battery bat2 needs to be measured, the 3-pin in PS1 and the 1-pin in PS2 are gated, and the voltage across the bat2 is output from the 6-pin of PS1 and the 8-pin of PS2, of which the 8-pin of PS2 is the battery bat2. Positive, PS1 pin 6 is the negative of battery bat1.

(2) Voltage inversion circuit

In the above analysis, it has been pointed out that when measuring bat1 and bat2, the voltages on the corresponding pins of CAP_1 and CAP_2 are opposite. In fact, all odd-numbered batteries and even-numbered batteries have opposite voltages during measurement. Therefore, the present invention uses two more PS7241s, and designs a voltage inversion circuit as shown in FIG. 15 . Through this circuit, the odd-numbered battery and the even-numbered battery can be measured to output the same voltage direction, which is convenient for the subsequent AD circuit to measure the voltage value of each single battery. The reason for not using a relay is that the response of the relay is not as fast as that of the PS7241, and the voltage drop caused by the relay will also affect the accuracy of the result.

When testing an odd-numbered battery, CAP2 is positive for the battery and CAP1 is negative for the battery. At this time, it is necessary to control K_AD_1 to be low voltage and K_AD_2 to be high voltage, even if the 8 pins of PS5 and PS6 in the figure are turned on, while 6 is not conductive Pass. When testing even-numbered batteries, CAP1 is battery positive and CAP2 is battery negative. At this time, control K_AD_1 to high level and K_AD_2 to low level. At this time, pins 6 of PS5 and PS6 are turned on, but 8 is not turned on. Through this circuit, AD_P can always be connected to the positive electrode of the battery, and AD_N is always connected to the negative electrode of the battery.

(3) Voltage-frequency conversion circuit

The voltage-frequency conversion circuit (VFC) can convert the input voltage signal into a frequency signal output. The frequency signal output by the VFC circuit is linearly proportional to the input voltage signal, that is, the higher the voltage, the faster the output frequency. . VFC circuits are widely used in various circuits, including signal frequency modulation, phase modulation, AD conversion circuits, etc. The VFC circuit has the advantages of strong anti-interference ability, convenient isolation, stable performance, high sensitivity and small nonlinear error. At the same time, when the analog signal is digitally processed, the resolution and precision of the VFC circuit are higher than those of the AD conversion circuit, and the cost of the VFC circuit is usually lower under the premise of the same precision.

The voltage-frequency conversion circuit designed in the present invention consists of two parts. The first part is the operational amplifier circuit, and the core of this part is an OP07C operational amplifier. OP07C has the characteristics of low noise and non-chopping zero stabilization. For most usage scenarios, the OP07C does not require external components for offset zeroing and frequency calibration. In addition, the OP07C features low bias current, high open-loop gain, and a wide operating temperature range. In order to ensure the stability and anti-interference ability of the VFC circuit, the present invention sets the magnification of the operational amplifier to 2 times.

The second part of the voltage-frequency conversion circuit is the VFC circuit, the core of which is the AD7740 chip. AD7740 is a low-cost, compact voltage-to-frequency conversion chip. This chip can work between 3.0V to 3.6V or 4.75V to 5.25V, and the operating current can be as low as 0.9mA. AD7740 supports a very wide operating temperature range, relies on few external original components, and has accurate voltage conversion frequency. The chip integrates a 2.5V reference reference, and also supports the use of externally input VDD as a reference voltage. The chip also has a synchronous clock input pin - CLKIN, which can support up to 1MHz frequency input. In the present invention, the output clock of K64 is used as the synchronous clock of AD7740, which reduces unnecessary original devices and reduces circuit complexity.

When the analog voltage varies between 0V and VREF, the signal output frequency of the AD7740 varies linearly between 0.1 and 0.9 times FCLKIN. The conversion formula of its voltage and frequency is as follows:

Finally, the voltage-frequency conversion circuit is given as shown in Figure 16.

It is worth noting that the OP07C needs positive and negative voltages for power supply, so a reverse polarity circuit is also required to convert the +5V voltage into -5V voltage, which is used to provide the OP07C with a negative power supply. In the present invention, the MAX660 charge pump reverse polarity switching integrated voltage regulator is used to realize this function, and the circuit diagram is shown in Figure 17.

Current acquisition circuit

Accurate current measurement is an essential condition for SOC estimation using the ampere-hour method, and a Hall current sensor is used to measure the current in the present invention. The principle of the Hall current sensor is that when the primary current flows through a long wire, a magnetic field is generated around the wire, and the magnitude of the magnetic field is proportional to the magnitude of the current. The device is used to measure and amplify the output, and its output voltage can reflect the size of the primary current.

The advantage of the Hall current sensor is that it has a wide measurement range, can measure current and voltage of arbitrary waveforms, and can faithfully reflect even transient peak current and voltage signals. The response speed of the Hall current sensor is extremely fast, which can reach the response speed of us-level. At the same time, the accuracy of the Hall current sensor is very high, which can achieve a measurement accuracy of better than 1%, and the degree of measurement thread is also good, and it can also work for a long time without failure, usually guaranteeing continuous work for several hours. In addition, the Hall element can be small and convenient to use.

The Hall current sensor selected by the present invention has a measurement range of ±100A and an operating voltage of 5V. When the current in the circuit is zero, the voltage output is 2.5V; when the current in the single circuit is -100A, the voltage output is 0V; when the current in the circuit is 100A, the output is 5V. For the sampling circuit adopting OP07C, the output of the Hall current sensor adopted in the present invention is within the normal working range.

temperature acquisition circuit

The temperature acquisition circuit is used to collect the temperature value of each small battery pack. The temperature has a great influence on the operation of the lithium battery. On the one hand, the real-time temperature value can ensure the accuracy of the estimation of the SOC of the lithium battery, on the other hand Temperatures that can be too high are also essential for the safe operation of the system.

The invention adopts the temperature acquisition circuit designed based on the NTC thermistor NTC10KB3950K to realize the temperature acquisition of the small battery pack, and the circuit has the characteristics of high measurement accuracy, simple structure and good stability. The accuracy of NTC10KB3950K can reach 1%, the resistance is 32.5K at 0 degrees, the corresponding voltage is 0.29V, and the resistance is 1.063K at 85 degrees, and the corresponding voltage is 3.26V. The calculation between resistance and voltage is given by:

Then according to the voltage and temperature correspondence table of NTC10K-3950, the temperature value in the small battery pack can be found.

Equalization circuit

The equalization control circuit is one of the cores of the present invention. The invention designs a transformer equalization circuit based on LTC3300-1 chip. The LTC3300-1 is a fault-protected controller IC for transformer-based bidirectional active balancing of multi-cell battery packs. The device integrates all required gate drive circuits, high-precision battery sensing, fault detection circuits, and a watchdog with built-in timer. Each LTC3300-1 can balance up to 6 series Li-Ion cells with a 36V input common-mode voltage. The charge of any chosen cell can be efficiently transferred back and forth between itself and 12 or more adjacent cells. The SPI interface of the LTC3300-1 can be connected to multiple LTC3300-1 devices in series without optocoupler isolation, thereby achieving charge balance for each battery in a long series of batteries. A series-connected LTC3300-1 can operate independently at the same time, thus allowing equalization management of all cells in a battery pack simultaneously and independently.

The equalizer circuit corresponding to each LTC3300-1 operates independently. The primary side of the transformer is connected to each single battery through a MOS tube, and the secondary side of the transformer is connected to the entire battery pack through a MOS. The LTC3300-1 supports two methods of transformer equalization. One is that each transformer has its own primary and secondary sides of the transformer; the other is that all transformers have their own primary side, but share a secondary side connected to the battery pack. The schematic diagram of the connection of the two transformers on the LTC3300-1 is shown in Figure 18.

In Figure 18(a), the equalizing transformer for each single cell has independent primary and secondary sides. The primary side is connected to the single cell through a MOS transistor, and the secondary side is connected to the battery pack through a MOS transistor. ; In Figure 18(b), each equalizing transformer has only a separate primary side, which is connected to the battery cell through a MOS tube, and the secondary side is shared by all transformers. Taking into account the modular design goals of the first system and the ease of maintenance, the present invention adopts all the forms shown in Fig. 18(a).

The LTC3300-1 can be connected to a maximum of 6 batteries for equalization. Figure 19 shows the connection method of channel 2. The solutions of other circuits are similar to this circuit. In the figure, the C2 pin is linked to the positive pole of bat2. I2P and I2S are used to measure the current on the primary and secondary sides of the transformer, respectively. G2P and G2S are used to control the on-off of the MOS tubes on the primary and secondary sides. BAT+, BAT- are linked to the positive and negative poles of the entire battery pack.

It is to be understood that the embodiments described herein may be implemented by hardware, software, firmware, middleware, microcode, or any combination thereof. For a hardware implementation, the processing unit may be implemented in one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays ( FPGA), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof. When the embodiments are implemented in software, firmware, middleware or microcode, program code or code segments, they may be stored in a machine-readable medium such as a storage component.

It will be apparent to those skilled in the art that the present invention is not limited to the details of the above-described exemplary embodiments, but that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments are to be regarded in all respects as illustrative and not restrictive, and the scope of the invention is to be defined by the appended claims rather than the foregoing description, which are therefore intended to fall within the scope of the claims. All changes within the meaning and scope of the equivalents of , are included in the present invention. Any reference signs in the claims shall not be construed as limiting the involved claim.

In addition, it should be understood that although this specification is described in terms of embodiments, not each embodiment only includes an independent technical solution, and this description in the specification is only for the sake of clarity, and those skilled in the art should take the specification as a whole , the technical solutions in each embodiment can also be appropriately combined to form other implementations that can be understood by those skilled in the art.

Claims (9)

1. A lithium battery pack balance control method for dynamically correcting SOC, characterized in that the method comprises the steps: Step 21, obtaining the SOC of each single lithium battery in the lithium battery pack; Step 22, calculate the range r _soc of the battery pack; Step 23, compare the size of the range _rsoc and the preset range threshold, when the range _rsoc is greater than the preset range threshold, enter step 24, when the range _rsoc is less than or equal to the predetermined range. When the preset range threshold is described, go to step 25; Step 24, select the average value of the SOC of all single lithium batteries

As the equilibrium target SOC, for SOC below

The single lithium battery is charged for equalization, and the SOC is higher than

The single lithium battery is discharged for equalization, where dSOC is the equalization control band; Step 25, end; Step 21 specifically includes: Step 11, determine whether the battery is in a working state, if so, go to Step 12, if not, go to Step 17; Step 12: Calculate the SOC _i of the current state using formula 2, where SOC _i is the SOC of the current state of the battery, SOC ₀ is the initial SOC when the battery starts to work, _CN is the rated capacity of the battery, I is the battery current, η is the charge-discharge efficiency, when charging, η is a negative number, and when discharging, η is a positive number; Step 13, determine whether the battery is in a working state, if so, go back to step 12, if not, go to step 14; Step 14, measure the open-circuit voltage OCV ₁ , look up the correspondence table between SOC and OCV, and obtain the SOC _{1 corresponding to the open-circuit voltage OCV 1 ;} Step 15, calculate the difference e between SOC _i and SOC ₁ , when the absolute value of e is greater than the preset error threshold, go to step 16, when the absolute value of e is less than or equal to the preset error threshold, output SOC _i ; enter step 11; Step 16, calculate the corrected SOC, output the corrected SOC correction, and update the correspondence table between SOC and OCV at the same time; go to step 11; Step 17 , measure the open circuit voltage OCV ₂ , look up the correspondence table between SOC and OCV, obtain the SOC ₂ corresponding to the open circuit voltage OCV ₂ , and output the SOC ₂ ; 2 . The balance control method for a lithium battery pack with dynamic correction of SOC according to claim 1 , wherein: before step 11 , an initial correspondence table between SOC and OCV is established by means of an interpolation method. 3 . 3. The lithium battery pack balance control method for dynamically correcting SOC according to claim 1, wherein step 16 specifically comprises: Calculate the revised _Sn (i+1) by formula 5, output _Sn (i+1), and update the corresponding _Sn (i) in the correspondence table between SOC and OCV to be _Sn (i+1); S _n (i+1)=S _n (i)-F(n, e) (0≤n≤50) (Formula 5) Among them, _Sn (i) represents the value in the table after the current i-th update, _Sn (i+1) represents the value in the table after the i+1-th update, F(n, e) is the correction coefficient, the The coefficient is a function related to n and e, which is expressed by Equation 6; F(n, e)=a*e*n (Formula 6) where a is an adjustable constant representing the correction rate, n is the number of points for interpolation, and e is the difference between SOC _i and SOC ₁ . 4 . The balance control method for a lithium battery pack with dynamic correction of SOC according to claim 3 , wherein the value of a is different in different sections of the SOC. 5 . 5. The lithium battery pack balance control method for dynamically correcting SOC according to claim 1, wherein: In step 24, when the battery is in the discharge balance, the battery that is about to enter the discharge cut-off is charged, so that its SOC is consistent with the SOC of other batteries, regardless of whether the battery is within the cut-off band; when the battery is charged and balanced, the For the battery that is about to enter the end of charging, the equalization circuit is activated to make the SOC of the battery fluctuate in the vicinity, so that finally all the batteries can reach the state of SOC=1 at the same time. 6. A lithium battery pack equalization control system for realizing the dynamic correction SOC of the method according to any one of claims 1-5, the control system comprises an upper-level system and a lower-level system, wherein the upper-level system comprises a PC and a main control MCU, The subordinate system includes a plurality of sub subordinate systems, each sub subordinate system is used for balancing control of a small battery pack, and each sub subordinate system includes a secondary MCU and a balancing module, and is characterized in that: The main control MCU is used to collect data fed back by each secondary MCU in the lower-level system, transmit the data to the upper-level PC, and forward the commands sent by the PC to the corresponding secondary MCU; PC, used to receive data sent by the main control MCU, and used to send commands to the main control MCU; The secondary MCU is used to collect the voltage, current and temperature data of each battery in the small battery pack; feed back the collected data to the main control MCU; calculate the SOC of each single battery; Equalization; when equalization is required, the equalization module of the small battery pack is controlled to equalize the batteries that need to be equalized; The equalization module is used to equalize the batteries that need to be equalized according to the control of the secondary MCU in the small battery pack. 7 . The balance control system for a lithium battery pack with dynamic correction of SOC according to claim 6 , wherein the balance module adopts a bidirectional flyback transformer. 8 . 8 . The lithium battery pack balance control system for dynamically correcting SOC according to claim 7 , wherein the bidirectional flyback transformer adopts LTC3300-1 chip. 9 . 9. The lithium battery pack balance control system for dynamically correcting SOC according to claim 7, wherein: For a single battery with a high SOC that needs to be balanced, turn on the secondary transformer switch of the battery, disconnect all other switches including the primary switch, and a current flows through the secondary winding of the bidirectional flyback transformer. The form of magnetic energy is stored in the secondary winding; after the SOC in the battery drops to meet the requirements, the secondary switch is turned off, the primary switch is turned on, the energy is transferred from the secondary winding to the primary winding, and the magnetic energy is converted into electrical energy, thereby The excess energy is transferred to other cells in the battery pack; For a single battery with a low SOC that needs to be balanced, the switch corresponding to the primary is turned on, all secondary switches are turned off, and current flows through the primary winding of the bidirectional flyback transformer. At this time, the electrical energy is stored in the form of magnetic energy on the primary side. into the primary winding; when enough electric energy is charged, turn off the primary switch, open the secondary change switch corresponding to the lowest SOC, turn on the secondary winding, the energy is transferred from the primary winding to the secondary winding, and the magnetic energy is converted back to electric energy After charging into the battery, the SOC of the single battery rises, and the overall SOC of the battery pack returns to a consistent value.

Images