BMS Management System, Battery Working Time Prediction System and Prediction Method

Abstract

The present application discloses a BMS management system, a prediction system and a prediction method for battery working time, belonging to the technical field of battery management. A BMS management system, a prediction system and a prediction method for battery working time include a battery pack, a balancing module, a switching module, a monitoring module and a control module. Among them, the control module obtains the SOC of each battery based on the charge and discharge information of each battery, generates several pairwise paired balance groups based on the SOC of each battery, and connects the balance groups to the corresponding balancing circuits through control switches to balance the batteries within the balance groups. In the technical solution proposed in the present application, different from the previous practice of configuring different levels of balancing circuits for each battery and starting balancing when needed, a common balancing circuit including multiple different balancing levels is innovatively designed.

Description

Technical Field

[0001] The present application relates to the technical field of battery management, and more specifically, to a BMS management system, a prediction system for battery working time, and a prediction method. Background Art

[0002] As a widely used energy storage technology at present, batteries have been widely applied in many fields such as power grid load balancing, vehicle driving, and industrial production. In order to ensure the safety and service life of battery production, currently the smallest single battery is designed to be relatively small in terms of voltage, capacitance, and load capacity. To meet the actual needs of various fields, multiple small batteries are usually combined into multiple battery packs by connecting them in series to increase the voltage and in parallel to increase the current, and then these battery packs are integrated into a complete energy storage system.

[0003] The performance of battery monomers is affected by multiple factors such as manufacturing processes and application environments, so there are certain differences. After multiple charge and discharge cycles, the performance differences between battery monomers will further expand, which will not only lead to a decline in the overall performance of the battery, but also shorten its service life, and thus have a negative impact on the performance and service life of the entire battery pack. Taking the charging process as an example, when a single battery in the battery pack reaches its capacity limit, charging cannot continue for this battery, so the overall performance of the battery pack will be limited by this battery with the smallest capacity.

[0004] To solve this problem, an equalization circuit is currently commonly used to balance the load of the battery. Taking two series-connected batteries as an example, when one battery is fully charged and the other is not, the equalization circuit will transfer the current of the fully charged battery to the other battery to achieve the balance of the charging states of the two. However, in the existing equalization circuit, the key lies in the need to be equipped with an energy storage element. Due to the voltage difference between different batteries, different levels of energy storage elements are required for balancing. At the same time, since the aging rates of each battery are inconsistent and difficult to predict, if an equalization circuit is set for any two batteries, it will lead to an overly complex circuit structure. In addition, even if an equalization circuit is set for two batteries, the energy storage element on the equalization circuit may not be able to adapt to the voltage difference between the two batteries. Therefore, in a complex battery pack, the equalization ability of each battery is often insufficient, which further affects the energy storage performance of the battery pack. Summary of the Invention

[0005] This section of the present application is used to briefly introduce concepts, which will be described in detail in the following detailed implementation section. This section of the present application is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] As the first aspect of this application, to solve the technical problems mentioned in the above background art section, this application provides a BMS battery management system, including:

[0007] A battery pack, composed of a number of batteries connected in series and parallel;

[0008] An equalization module, provided with a number of equalization circuits with different equalization levels;

[0009] A switch module, provided with a number of control switches, and the control switches control the connection between the batteries and the equalization circuits;

[0010] A monitoring module, used to monitor the charge and discharge information of each battery;

[0011] A control module, which is respectively signal-connected to the battery pack, the equalization module, the switch module and the monitoring module;

[0012] Among them, the control module obtains the SOC of each battery based on the charge and discharge information of each battery, generates a number of pairwise paired balance groups based on the SOC of each battery, and connects the balance groups to the corresponding equalization circuits with control switches to equalize the batteries within the balance groups.

[0013] In the technical solution proposed in this application, different from the previous method of configuring different-level equalization circuits for each battery and starting equalization when needed, a public equalization circuit containing a number of different equalization levels is innovatively designed. When it is monitored that there is a battery in the battery pack that needs to be equalized, only through the switch module, the two batteries that need to be equalized are accurately connected, and then the equalization operation is performed. This solution makes clever use of the characteristic that the aging rates of the batteries in the battery pack are inconsistent, but the number of batteries that actually need to be equalized is relatively small. By adopting this design of the public equalization circuit, the overall complexity of the equalization circuits in the battery pack is effectively reduced. At the same time, this solution can also flexibly adapt to the equalization requirements of each battery in the battery pack, ensuring the performance and stability of the battery pack.

[0014] Further, the monitoring module includes:

[0015] An original information monitoring unit, used to monitor the voltage, current and internal resistance of each battery;

[0016] An SOC calculation unit, which calculates the SOC value of each battery in real time based on the Kalman filtering method.

[0017] In the technical solution disclosed in this application, the Kalman filtering method is adopted to realize the real-time high-precision calculation of the SOC (State of Charge) value of each battery. This method not only has high precision, but also can continuously correct the prediction method as the battery aging process progresses, so as to ensure the continuous accuracy of the SOC value calculation.

[0018] Although this solution uses a balancing module and a switching module to precisely control the process of each battery accessing the balancing circuit, in fact, the number of balancing circuits still needs to be appropriate and should not be set too many. If it is assumed that each battery can access any one of the balancing circuits, then each battery needs to be equipped with a switch connected to the balancing circuit. However, if the number of balancing circuits is too large, the design of the switching module will inevitably become too complex. To more effectively reduce the number of balancing circuits in the balancing module, the present application specifically proposes the following technical solutions:

[0019] Further, the control module includes:

[0020] An information acquisition unit for obtaining the SOC values of each battery and establishing a first SOC sequence of the battery pack;

[0021] An information update unit for obtaining the rate information of the SOC of each battery and establishing a second SOC sequence of the battery pack;

[0022] A balancing record unit for obtaining the connection information of each balancing circuit and obtaining the idle number of the balancing circuits;

[0023] A model construction unit for establishing a particle swarm balancing model of the battery pack based on the first SOC sequence, the second SOC sequence, and the idle number of the balancing circuits, so as to solve the division method of the balance group and the corresponding relationship between each balance group and the balancing circuit;

[0024] An execution unit: controlling the switching module to connect the balance group to the corresponding balancing circuit according to the corresponding relationship between the balance group and the balancing circuit.

[0025] In the technical solution proposed by the present application, the first SOC sequence, the second SOC sequence, and the idle number of the balancing circuits are combined to construct a particle swarm model. This model can comprehensively consider the SOC values of each battery and their change rates, so as to accurately solve the optimal balancing scheme. This scheme aims to minimize the demand for the number of balancing circuits while achieving the best balancing efficiency.

[0026] Although the first SOC sequence and the second SOC sequence can indicate the current charging situation of each battery and the charging situation in a future period of time, it is difficult to accurately find the correlation relationship of the charging of each battery from them. Therefore, it is difficult to accurately construct a particle swarm balancing model. To solve this problem, the present application provides the following technical solutions:

[0027] Further, the first SOC sequence H1 is: H1 = S1, S2, S3,... S i ... S n ; The second SOC sequence H2 is: H2 = V1, V2, V3,... V i ... Vn ;

[0028] Among them, S i represents the SOC value of the i-th battery, V i represents the change rate of the SOC of the i-th battery, i represents the battery index, and n represents the total number of batteries;

[0029] The model construction unit preprocesses H1 and H2 to obtain battery distribution information;

[0030] Taking H1 as the ordinate and H2 as the abscissa, a plane rectangular coordinate system is established to obtain the battery distribution map M.

[0031] In the technical solution provided by this application, a plane rectangular coordinate system is established with H1 as the ordinate and H2 as the abscissa to obtain the battery distribution map. Therefore, the battery distribution map can more intuitively reflect the differences in charging and discharging of each battery, so as to better establish the particle swarm equilibrium model.

[0032] In practice, the number of idle balancing circuits is generally not equal to the number of batteries that need to be balanced. Therefore, in order to ensure their equality, the number of batteries that need to be balanced needs to be adjusted. If the SOC difference is directly used as the pairing basis, it is difficult to establish a particle swarm model from the overall perspective of the battery pack, which easily leads to poor active balancing effect of the batteries and cannot make all batteries fully charged as much as possible. To address this problem, this application provides the following technical solution:

[0033] Further, the construction process of the particle swarm equilibrium model includes the following steps:

[0034] Step 1: Extract the center points of the batteries from the battery distribution map M;

[0035] Step 2: Taking the center point as the reference point, using the time when the battery at the reference point is fully loaded as the reference time, dividing lines are drawn on the battery distribution map M according to the reference time. The batteries on the right side of the dividing line need to be positively loaded, and the batteries on the left side of the dividing line need to be negatively loaded;

[0036] Step 3: Taking the dividing line as the reference, translating w1 positions to the right to generate the first dividing line; translating w2 positions to the left to generate the second dividing line;

[0037] Step 4: Construct a particle swarm model based on w1, w2, and R.

[0038] In the technical solution provided by this application, a reference point is established based on the reference time, then the dividing line is established, and then the first dividing line and the second dividing line are generated by spreading to the right and left from the dividing line; furthermore, taking this as the selected balancing mode of the solution, it can consider the overall balancing effect of the battery pack to make all batteries reach the fully loaded state as much as possible.

[0039] Furthermore, step 1 includes the following steps:

[0040] Step 11: Extract all data points from the battery distribution map M. Each data point is represented by (V i , S i ), where i = 1, 2,..., n;

[0041] Step 12: For each data point i, use the Gaussian kernel function and bandwidth h to calculate the density estimate value of data point i;

[0042]

[0043] where exp is the exponential function used to calculate the weight; (V, S) is the point in the battery distribution map M for which the density needs to be calculated:

[0044] Step 13: Calculate the density estimate value of each point, obtain the partitioning method of the battery distribution map R to get the density grid map, and map the density value of each point into the density grid map to obtain the density map U;

[0045] Step 14: Obtain the region with the maximum density value in the density map U of the battery, take this region as the central region, and extract the central point from the central region.

[0046] In this solution, the Gaussian kernel function is used to calculate the central point. Therefore, regardless of how the data points are distributed, the Gaussian kernel function will smooth them in the same way, resulting in an isotropic density estimation result, and then accurately locating the position of the central point. Thus, when locating the central point in this solution, it is not actually locating a specific battery, but finding the central point related to the SOC distribution among all batteries. Although this central point is a virtual point, it can best reflect the future load expectation of the batteries in the battery pack.

[0047] In the existing particle swarm construction algorithms, generally, the balancing circuit and each battery are used as the basis for selection to allocate the best battery pack to each balancing circuit. This solution overly focuses on the matching relationship between the battery pack and the balancing circuit, easily leading to the final result falling into a local optimal solution and being unable to find a global solution.

[0048] Furthermore, step 4 includes the following steps:

[0049] Step 41: Establish a feasible solution matrix L;

[0050] In each element of the feasible solution matrix L, a is w1 and b is w2; in the feasible solution matrix L, each element represents a partitioning scheme of the battery pack, and for each partitioning scheme of the battery pack, an optimal partitioning method of the balancing group can be generated;

[0051] Step 42: Set the objective function f(x);

[0052]

[0053] Where α is the first parameter, β is the second parameter, U is the value of the remaining balance circuit after normalization, T is the normalized value of the remaining working time of the battery, Q1 is the maximum charge and discharge amount of all batteries under the current charging scheme, and Q2 is the theoretical charging amount for all batteries to be fully charged;

[0054] Step 43: Set the number of honey sources and the iteration threshold;

[0055] Step 44: Replace the old honey source with a new one in a greedy selection manner; repeat the above steps and perform m consecutive iterations until the best honey source is found.

[0056] In the technical solution provided by this application, when constructing the feasible solution matrix L, the values of w1 and w2 are used as the sorting basis. In this way, when updating the honey source, the moving relationship between the first boundary line and the second boundary line can be well measured, and then the best solution can be found quickly and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The drawings constituting a part of this application are used to provide a further understanding of this application, making other features, objectives, and advantages of this application more obvious. The schematic embodiments and descriptions thereof of this application are used to explain this application and do not constitute an improper limitation to this application.

[0058] In addition, throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic and the elements and components are not necessarily drawn to scale.

[0059] In the drawings:

[0060] Figure 1 is a schematic structural diagram of the BMS battery management system.

[0061] Figure 2 is a schematic diagram of the battery connected to the balance circuit through the switch module.

[0062] Figure 3 is a schematic diagram of the battery connected to the balance circuit.

[0063] Figure 4 is a scatter plot of the battery distribution map;

[0064] Figure 5 is a schematic diagram of drawing the first dividing line o, the second dividing line q, and the dividing line in the battery distribution map. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0065] Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not used to limit the protection scope of the present application.

[0066] In addition, it should be noted that for the convenience of description, only parts related to the relevant invention are shown in the drawings. Without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0067] The present application will be described in detail below with reference to the drawings and in combination with embodiments.

[0068] Embodiment 1, referring to Figure 1 , a BMS battery management system, including: a battery pack, a balancing module, a switching module, and a control module. Among them, the battery pack is composed of a number of batteries connected in series and in parallel. For the series and parallel connection methods of the batteries inside the battery pack, no specific restrictions are set, but flexible configuration is carried out according to actual needs and existing battery pack designs. However, it is clear that the battery pack in this solution is composed of multiple batteries, and during the charging or discharging process, all batteries need to maintain a synchronous working state. Therefore, once any one of the batteries in the battery pack is fully discharged or charged, the entire battery pack will stop working, showing a typical short-board effect.

[0069] To address this problem, a balancing module is introduced to achieve balanced management of each battery in the battery pack. Specifically, the present application designs multiple balancing circuits with different balancing levels. These balancing circuits are all active balancing circuits, and components such as capacitors or inductors are configured on the circuits (the specific configuration method will not be elaborated in detail here). When these balancing circuits are connected to two batteries with different SOCs, they can perform balancing processing on these two batteries so that they can complete the charging or discharging process synchronously.

[0070] Regarding the specific design scheme of the balancing circuit, since it belongs to the category of existing technologies, the design structure will not be elaborated herein. However, it is worth mentioning that all the balancing circuits in this solution are active balancing circuits, and their core function is to connect two batteries with different remaining power levels through the balancing circuit, so as to achieve synchronous charging or discharging of the two batteries.

[0071] In this solution, the balancing module includes multiple balancing circuits, but these balancing circuits are not directly connected to two batteries, but are flexibly connected to any two batteries through the switching module. Specifically, the switching module is composed of multiple control switches, such as Figure 2As shown, it can control the connection state between the left pin and any pin on the right. In this way, with the control module, any two batteries can be connected to the required balancing circuit. Figure 2 Only the case where the positive electrode of the battery is connected to the switch module is shown. For the circuit structure on the other side, only mirror copying is needed.

[0072] For the sake of easy understanding, Figure 3 a circuit diagram example of a balancing module for balancing 3 batteries is provided.

[0073] The monitoring module is used to monitor the charge and discharge information of each battery. In each battery of the battery pack, various sensors are set, and thus the voltage, current, and internal resistance of each battery can be obtained, and in this way, the charge and discharge information of each battery is obtained. In this application, to achieve active balancing, the most crucial part is to obtain the SOC value of the battery. The SOC value cannot be directly measured and needs to be calculated directly using the detected data. Specifically:

[0074] The monitoring module includes: a raw information monitoring unit and an SOC calculation unit. The raw information monitoring unit is used to monitor the voltage, current, and internal resistance of each battery; the SOC calculation unit calculates the SOC value of each battery in real time based on the Kalman filtering method.

[0075] The Kalman filtering method is a prior art. For the sake of easy understanding, the general method of the Kalman filtering method is provided below:

[0076] Determine the state equation and the observation equation: The state equation describes the change law of the battery state over time, usually including influencing factors such as current and temperature.

[0077] The observation equation describes how to estimate the battery state by measuring parameters such as the voltage and current of the battery.

[0078] Set the initial state of the system, including the initial value of SOC.

[0079] Set the initial variance matrix of the system, indicating the uncertainty of the initial state.

[0080] According to the state equation and the control quantity (such as current, temperature, etc.) at the current moment, predict the battery state and variance matrix at the next moment.

[0081] According to the observed value and the predicted value, use the Kalman gain matrix to update the state and variance matrix of the system. The observed value is usually obtained by measuring parameters such as the voltage and current of the battery.

[0082] By continuously repeating the prediction and update steps, a more accurate SOC estimated value can be obtained. As new data is continuously observed, the Kalman filtering algorithm can update the SOC estimated value in real time.

[0083] The control module obtains the SOC of each battery based on the charge and discharge information of each battery, generates several paired balance groups based on the SOC of each battery, and connects the balance groups to the corresponding equalization circuits through control switches to equalize the batteries within the balance groups.

[0084] Specifically: The control module includes: an information acquisition unit, an information update unit, an equalization record unit, a model construction unit, and an execution unit. Among them, the information acquisition unit is used to obtain the SOC value of each battery and establish the first SOC sequence of the battery pack; the information update unit obtains the rate information of the SOC of each battery and establishes the second SOC sequence of the battery pack.

[0085] Because the SOC calculation unit can calculate the SOC value of each battery in real time and the growth rate of the SOC, the growth rate of the SOC selects the growth rate in the most recent unit time, and the length of the unit time is set according to requirements. In this application, it is set to 1 second. Therefore, the required first SOC sequence and second SOC sequence can be directly obtained. The equalization record unit obtains the connection information of each equalization circuit to obtain the idle quantity of the equalization circuit.

[0086] The model construction unit establishes a particle swarm equalization model of the battery pack based on the first SOC sequence, the second SOC sequence, and the idle quantity of the equalization circuit to solve the division method of the balance groups and the corresponding relationship between each balance group and the equalization circuit.

[0087] The execution unit: According to the corresponding relationship between the balance group and the equalization circuit, the switch module connects the balance group to the corresponding equalization circuit.

[0088] The execution unit is a specific execution component. In the case of knowing the corresponding relationship, it only needs to control the corresponding control switch in the switch module to close or open. Therefore, the execution unit will not be described here.

[0089] Further, the first SOC sequence H1 is: H1 = S1, S2, S3,... S i …S n ; the second SOC sequence H2 is: H2 = V1, V2, V3,... V i 、…V n ; where S i represents the SOC value of the i-th battery, V i represents the change rate of the SOC of the i-th battery, i represents the index of the battery, and n represents the total number of batteries.

[0090] The above are the representation methods of the first SOC sequence and the second SOC sequence.

[0091] The model construction unit preprocesses H1 and H2 to obtain battery distribution information.

[0092] Taking H1 as the vertical coordinate and H2 as the horizontal coordinate, a plane rectangular coordinate system is established to obtain the battery distribution map M.

[0093] The battery distribution map M is the information that needs to be calculated by the model construction unit after receiving the first SOC sequence and the second SOC sequence. In the battery distribution map M, it contains the distribution information of the batteries in terms of SOC. In fact, because the charging power and discharging power of the battery pack are basically constant during use, the change rate of the SOC of each battery is basically equal and will not change. Therefore, in the battery distribution map M, each battery will actually follow the law of normal distribution and concentrate near the change rate of the average SOC, and then gradually extend upward or downward as the charging and discharging proceed. As Figure 4 shown.

[0094] Furthermore: The construction process of the particle swarm equilibrium model includes the following steps:

[0095] Step 1: Extract the center points of the batteries from the battery distribution map M.

[0096] Step 1 includes the following steps:

[0097] Step 11: Extract all the data points from the battery distribution map M, and each data point is represented by (V i , S i ), where i = 1, 2,..., n;

[0098] Step 12: For each data point i, use the Gaussian kernel function and bandwidth h to calculate the density estimate value of the data point i;

[0099]

[0100] where exp is the exponential function used to calculate the weight; (V, S) is the point in the battery distribution map M for which the density needs to be calculated:

[0101] Step 13: Calculate the density estimate value of each point, set the smallest cell to obtain the density grid map, and map the density value of each point into the density grid map to obtain the density map U;

[0102] The smallest cell here is the smallest difference in SOC and the difference in SOC rate, and the specific situation is set according to the accuracy. If a lower accuracy is required, the cell is set smaller, and vice versa.

[0103] Step 14: Obtain the region with the largest density value in the density map U of the battery, take this region as the central region, and extract the center point from the central region.

[0104] The central area consists of grids with the maximum and equal density values, and the center point is the geometric center of this area.

[0105] Step 2: Using the center point as a reference point, take the time when the battery at the reference point is fully loaded as the reference time. According to the reference time, divide the separator line on the battery distribution map M. The batteries on the right side of the separator line need to be positively loaded, and the batteries on the left side of the separator line need to be negatively loaded.

[0106] In Step 1, the center point is calculated. Although there may not necessarily be a battery at the center point, it can be regarded as an ideal battery position. Taking the SOC value and the SOC rate at this center point as a reference, the time required to fully charge this point can be calculated, and then the reference time can be obtained. By obtaining all the points that are fully charged within the reference time, a separator line can be obtained, and this separator line is a straight line.

[0107] Specifically, since it is necessary to charge the SOC to the maximum value (100%), and there is also the reference time t, for each point, the following equation can be obtained:

[0108] S0 = S + tV, where S0 and t are constants, S0 is the maximum value to be reached during charging, and t is the reference time. After simple transformation of the equation, we can get: S = tV - S0; thus, S = tV - S0 is the function expression of the separator line.

[0109] For example: As shown by the point (10, 2), when the SOC reaches 20, it means the charging is completed, and thus the reference time is 5 seconds; for the point one unit to the right of the point (10, 2), the point that can reach 20 within 5S is: (5, 3), and for the point one unit to the left of the point (10, 2), the point that can reach 20 within 5S is: (15, 1). In this way, it can be seen that (15, 1), (10, 2), and (5, 3) are all on a straight line. Correspondingly, the separator line is also a straight line.

[0110] Step 3: Using the separator line p as a reference, translate it w1 positions to the right to generate the first dividing line; translate it w2 positions to the left to generate the second dividing line.

[0111] As Figure 5 shown, for the batteries between the first dividing line o and the second dividing line q, although there are certain differences, generally speaking, the difference in the expected charging time is relatively small. Therefore, for the batteries between the first dividing line and the second dividing line, it is not necessary to consider whether to perform balancing.

[0112] The batteries on the right side of the first demarcation line and the batteries on the left side of the second demarcation line need to be actively balanced, and the batteries on both sides can be paired up exactly two by two. Generally speaking, the batteries on the right have a faster charging rate, so they are distributed in the upper right corner. On the contrary, the batteries on the left have a slower charging rate, so they are distributed in the lower left corner. Pairing the two up for active balancing can bring the SOC information of the two batteries closer to the inside of the first demarcation line and the second demarcation line.

[0113] Step 4: Construct a particle swarm model based on w1, w2, u, and R.

[0114] Step 4 includes the following steps:

[0115] Step 41: Establish a feasible solution matrix L;

[0116] In each element of the feasible solution matrix L, a is w1, b is w2, and the minimum number of batteries on the right side of the first demarcation line or the left side of the second demarcation line is z; in the feasible solution matrix L, each element represents a battery group division scheme; in the feasible solution matrix L, each element represents a battery group division scheme, and for each battery group division scheme, an optimal balance group division method can be generated.

[0117] The key of the particle swarm algorithm lies in the construction method of the feasible solution matrix L. In this solution, the arrangement of the elements in the feasible solution matrix L is set according to the values of w1 and w2. For example, the meaning of (1, 1, u) is that if both w1 and w2 are equal to 1, then the minimum number of batteries on the right side of the first demarcation line or the left side of the second demarcation line is z, and z < u.

[0118] In this way, z batteries need to be paired. Given the batteries to be paired, how to reasonably select a suitable balancing circuit is the prior art and will not be elaborated further here. In this solution, the batteries farthest from the center are mainly preferentially selected for active balancing. Therefore, in practice, for each value of w1 and w2, one distribution scheme of the balancing circuit can be obtained. And from top to bottom, w1 gradually increases, from left to right w2 gradually increases, and at the same time, z will gradually decrease. The relationship between the elements in the feasible solution matrix L reflects the implicit state of the gradual change of w1 and w2, and thus the optimal solution can be found with a smaller number of iterations.

[0119] Step 42: Set the objective function f(x);

[0120]

[0121] Among them, α is the first parameter, β is the second parameter, U is the normalized value of the remaining balancing circuit, T is the normalized value of the remaining working time of the battery; Q1 is the maximum charge and discharge capacity of all batteries under the current charging scheme, and Q2 is the theoretical charge capacity of all batteries when they are fully charged.

[0122] The objective function provided in this application is f(x), and the value of the objective function needs to be as large as possible during the iteration process. In the objective function, α and β are parameter items, which are used to adjust the first item and the second item in the objective function to have the required weight influence. It indicates the relationship between the number of remaining balancing circuits and the remaining working time of the battery after the current balancing circuit allocation method is adopted. If the two are closer, the value will be larger. In this way, when allocating balancing circuits, all balancing circuits will not be allocated immediately, but will be gradually introduced during the charging and discharging process to ensure that there are a certain number of free balancing circuits in the middle and late stages of charging and discharging to play a balancing role; It is used to ensure that the balancing effect can be increased as much as possible, and all batteries can be charged or discharged at the same time. Q1 and Q2 are actually theoretical values. Q2 is the cumulative sum of the power of all batteries, and Q1 is the maximum charge and discharge of all batteries in the current state. This is generally related to the worst battery. For example, in the current state, about 10 seconds later, a battery will be fully charged. At this time, the charging of all batteries will be stopped, so the remaining batteries cannot continue to charge even if they are not fully charged, and then Q1 can be calculated.

[0123] Step 43: Set the number of honey sources and the iteration threshold;

[0124] Step 44: Use a greedy selection method to replace the old nectar source with a new nectar source; repeat the above steps for m consecutive iterations until the best nectar source is found.

[0125] The particle swarm algorithm is a common technical means in this field. When the feasible solution matrix and the objective function f(x) are known, technicians in this field can construct the required particle swarm algorithm.

[0126] In the particle swarm algorithm provided in this application, balancing circuits may be used at different stages. As mentioned above, balancing circuits have multiple levels. When setting feasible solutions, how to allocate balancing circuits? This application provides the following solutions:

[0127] Although the aging rates of the batteries in the battery pack are not consistent, in general, the aging conditions show a normal distribution, so during the charging process, the difference in power between the batteries remains basically unchanged. In this way, the balancing circuit can be used in proportion each time the balancing circuit is adjusted.

[0128] Embodiment 2: A battery working time prediction system, comprising:

[0129] The BMS management system described in Embodiment 1;

[0130] A prediction unit that calculates the predicted working time of the battery pack based on the current active balancing scheme.

[0131] Specifically, by using the BMS management system described in Embodiment 1, the current balancing scheme of the battery pack can be obtained. In this way, the battery that can complete charging or discharging fastest in the current state can be obtained, and then the predicted working time of the battery pack can be obtained.

[0132] For example: After the active balancing is completed, the current SOC value and SOC rate of each battery can be obtained. In this way, we calculate the time when the SOC value reaches the peak value (the maximum value during charging and the minimum value during discharging) fastest, and then this time is the predicted working time of the battery pack.

[0133] Embodiment 3: A battery working time prediction method that uses the battery working time prediction system described in Embodiment 2 to calculate the predicted working time of the battery pack.

[0134] The above description is only some preferred embodiments of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present application is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, technical solutions formed by mutually replacing the above features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present application.

Claims

1. A BMS battery management system, characterized in that: Including: A battery pack, which is composed of a plurality of batteries connected in series and parallel; An equalization module, which is provided with a plurality of equalization circuits with different equalization levels; A switching module, which is provided with a plurality of control switches, and the control switches control the connection between the batteries and the equalization circuits; A monitoring module, which is used to monitor the charge and discharge information of each battery; A control module, which is respectively connected to the battery pack, the equalization module, the switching module and the monitoring module by signals; Among them, the control module obtains the SOC of each battery based on the charge and discharge information of each battery, generates a plurality of pairwise paired balance groups based on the SOC of each battery, and connects the balance groups to the corresponding equalization circuits by control switches to equalize the batteries within the balance groups; The control module includes: An information acquisition unit, which is used to obtain the SOC value of each battery and establish the first SOC sequence of the battery pack; An information update unit, which obtains the rate information of the SOC of each battery and establishes the second SOC sequence of the battery pack; An equalization record unit, which obtains the connection information of each equalization circuit and gets the idle number of the equalization circuits; A model construction unit, which establishes a particle swarm equalization model of the battery pack based on the first SOC sequence, the second SOC sequence and the idle number of the equalization circuits to solve the division method of the balance groups and the corresponding relationship between each balance group and the equalization circuits; An execution unit: According to the corresponding relationship between the balance groups and the equalization circuits, the switching module connects the balance groups to the corresponding equalization circuits.

2. The BMS battery management system according to claim 1, wherein: The monitoring module includes: An original information monitoring unit, which is used to monitor the voltage, current and internal resistance of each battery; An SOC calculation unit, which calculates the SOC value of each battery in real time based on the Kalman filtering method.

3. The BMS battery management system according to claim 1, characterized in that: The first SOC sequence H1 is: H1 = S1, S2, S3, … S i … S n ; The second SOC sequence H2 is: H2 = V1, V2, V3, … V i 、… V n ; Among them, S i represents the SOC value of the i-th battery, V i represents the change rate of the SOC of the i-th battery, i represents the battery index, and n represents the total number of batteries; The model construction unit preprocesses H1 and H2 to obtain the battery distribution information; Taking H1 as the ordinate and H2 as the abscissa to establish a plane rectangular coordinate system to obtain the battery distribution map M.

4. The BMS battery management system according to claim 3, characterized in that: The construction process of the particle swarm equalization model includes the following steps: Step 1: Extract the center points of the batteries from the battery distribution map M; Step 2: Taking the center point as the reference point, using the time when the battery at the reference point is fully loaded as the reference time, dividing the dividing line on the battery distribution map M according to the reference time, the batteries on the right side of the dividing line need to be positively loaded, and the batteries on the left side of the dividing line need to be negatively loaded; Step 3: Taking the dividing line as the reference, translating w1 positions to the right to generate the first dividing line; translating w2 positions to the left to generate the second dividing line; Step 4: Construct a particle swarm model based on w1, w2 and R.

5. The BMS battery management system according to claim 4, wherein: Step 1 includes the following steps: Step 11: Extract all data points from the battery distribution map M, and each data point is represented by (V i , S i ), where i = 1, 2,..., n; Step 12: For each data point i, use the Gaussian kernel function and bandwidth h to calculate the density estimation value of the data point i; Among them, exp is: the exponential function, which is used to calculate the weight; (V, S) is the point in the battery distribution map M that needs to calculate the density: Step 13: Calculate the density estimation value of each point, obtain the division method of the battery distribution map R to get the density grid map, and map the density value of each point into the density grid map to get the density map U; Step 14: Obtain the region with the largest density value in the density map U of the battery, take this region as the central region, and extract the center point from the central region.

6. The BMS battery management system according to claim 5, characterized in that: Step 4 includes the following steps: Step 41: Establish a feasible solution matrix L; In each element of the feasible solution matrix L, a is w1, b is w2, and the minimum number of batteries on the right side of the first dividing line or on the left side of the second dividing line is z; in the feasible solution matrix L, each element represents a battery pack division scheme; Step 42: Set the objective function f(x); Wherein, α is the first parameter, β is the second parameter, U is the value of the remaining balancing circuit after normalization, T is the normalized value of the remaining working time of the battery, Q1 is the maximum charge and discharge amount of all batteries under the current charging scheme, and Q2 is the theoretical charging amount for all batteries to be fully charged; Step 43: Set the number of nectar sources and the iteration threshold; Step 44: Replace the old nectar source with a new nectar source by means of greedy selection; repeat the foregoing steps and perform m consecutive iterations until the best nectar source is found.

7. A prediction system for battery working time, characterized in that: Including: The BMS management system according to any one of claims 1 to 6; A prediction unit that calculates the predicted working time of the battery pack based on the current active balancing scheme.

8. A method for predicting the working time of a battery, characterized in that: Calculate the predicted working time of the battery pack by using the prediction system for the battery working time described in claim 7.

Images