Fuzzy model prediction active equalization method and equalization circuit

Abstract

The embodiment of the invention provides a fuzzy model prediction active equalization method and an equalization circuit, and relates to the technical field of power batteries. The active equalization method comprises the following steps: acquiring an SOC mean value and an SOC difference value as input values of a controller; obtaining a difference value between the reference current and the equalizing current as an equalizing error; taking the weight ratio of the equalization error to the equalization current as an output value; according to a membership function, establishing a relationship between the input value and the output value and the fuzzy discourse domain and the fuzzy subset; determining a fuzzy control rule table according to the input value and the fuzzy subset; obtaining a membership degree of an output value corresponding to the input value according to a fuzzy control rule table; performing defuzzification operation on the membership degree to obtain an actual current weight; constructing a weight matrix according to each current weight; and constructing a state equation and a cost function, and according to the weight matrix, determining an equalization current and performing active equalization. The equalization current is determined through model prediction, the equalization current value changes along with the equalization state, and the equalization efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric batteries, and in particular to a fuzzy model prediction active balancing method and a balancing circuit. Background Art

[0002] In the traditional active balancing strategy, discharge is carried out from a single cell with a high SOC value to a single cell with a lower SOC value, and the balancing current value is constant, which leads to low balancing efficiency. When the SOC difference is large, the constant balancing current will lead to a long balancing time. When the SOC difference is small, the constant balancing current will lead to over-balancing, that is, the discharged single cell is discharged too much and the charged single cell is charged too much, resulting in imbalance again. Summary of the Invention

[0003] The purpose of the present invention is to provide a fuzzy model prediction active balancing method and balancing circuit, which determines the balancing current through model prediction. The balancing current value changes with the change of the balancing state, which can improve the balancing efficiency. By fuzzifying the balancing current weight, the cost function can be flexibly adjusted according to the actual working conditions, further accelerating the balancing speed.

[0004] In order to achieve the above object, the present invention provides a fuzzy model prediction active equalization method, comprising:

[0005] Obtain the SOC mean and SOC difference as input values ​​for the controller;

[0006] Obtaining the difference between the reference current and the balancing current as the balancing error;

[0007] Taking the weight ratio of the balancing error and the balancing current as an output value;

[0008] According to the membership function, the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset is established;

[0009] Determining a fuzzy control rule table according to the input value and the fuzzy subset;

[0010] According to the fuzzy control rule table, obtaining the membership degree of the output value corresponding to the input value;

[0011] Performing a defuzzification operation on the membership degree to obtain an actual current weight;

[0012] constructing a weight matrix according to each of the current weights;

[0013] A state equation and a cost function are constructed, and according to the weight matrix, the balancing current is determined and active balancing is performed.

[0014] Optionally, obtaining the SOC mean value and the SOC difference value as input values ​​of the fuzzy controller includes:

[0015] According to formula (1), the SOC mean value is obtained:

[0016]

[0017] in, is the mean SOC, N is the number of battery cells in series, i (i=1,2,3,…,N) is the i-th battery cell, and is the SOC i is the SOC value of the i-th battery cell;

[0018] According to formula (2), the SOC difference is obtained:

[0019]

[0020] Wherein, ΔSOC is the SOC difference, min(SOC) is the minimum SOC value of the battery cell, and max(SOC) is the maximum SOC value of the battery cell.

[0021] Optionally, obtaining the current weight corresponding to each battery cell and constructing a weight matrix includes:

[0022] Construct the weight matrix according to formula (3):

[0023]

[0024] Among them, R is the weight matrix, r i is the current weight of the i-th battery cell.

[0025] Optionally, constructing a state equation and a cost function, determining a balancing current according to the weight matrix, and performing active balancing includes:

[0026] According to formula (4), the state equation is constructed.

[0027]

[0028] Among them, x(k+1) and x(k) are the sets of SOC at the k+1th and kth calculations respectively, u(k) is the balancing current at the kth calculation, and e(k) is the error at the kth calculation. is the set of SOC mean values ​​at the kth calculation.

[0029] Optionally, constructing a state equation and a cost function, determining a balancing current according to the weight matrix, and performing active balancing further includes:

[0030] Construct the cost function according to formula (5):

[0031] J=e(k) T Re(k), (5)

[0032] Among them, J is the cost function.

[0033] On the other hand, the present invention further provides a balancing circuit, which includes a plurality of single batteries connected in series, a balancing module, and a controller, wherein the controller is configured to:

[0034] Obtain the SOC mean and SOC difference as input values ​​for the controller;

[0035] Obtaining the difference between the reference current and the balancing current as the balancing error;

[0036] Taking the weight ratio of the balancing error and the balancing current as an output value;

[0037] According to the membership function, the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset is established;

[0038] Determining a fuzzy control rule table according to the input value and the fuzzy subset;

[0039] According to the fuzzy control rule table, obtaining the membership degree of the output value corresponding to the input value;

[0040] Performing a defuzzification operation on the membership degree to obtain an actual current weight;

[0041] constructing a weight matrix according to each of the current weights;

[0042] A state equation and a cost function are constructed, and according to the weight matrix, the balancing current is determined and active balancing is performed.

[0043] Optionally, obtaining the SOC mean value and the SOC difference value as input values ​​of the fuzzy controller includes:

[0044] According to formula (1), the SOC mean value is obtained:

[0045]

[0046] in, is the mean SOC, N is the number of battery cells in series, i (i=1,2,3,…,N) is the i-th battery cell, and is the SOC i is the SOC value of the i-th battery cell;

[0047] According to formula (2), the SOC difference is obtained:

[0048]

[0049] Wherein, ΔSOC is the SOC difference, min(SOC) is the minimum SOC value of the battery cell, and max(SOC) is the maximum SOC value of the battery cell.

[0050] Optionally, obtaining the current weight corresponding to each battery cell and constructing a weight matrix includes:

[0051] Construct the weight matrix according to formula (3):

[0052]

[0053] Among them, R is the weight matrix, r i is the current weight of the i-th battery cell.

[0054] Optionally, constructing a state equation and a cost function, determining a balancing current according to the weight matrix, and performing active balancing includes:

[0055] According to formula (4), the state equation is constructed.

[0056]

[0057] Among them, x(k+1) and x(k) are the sets of SOC at the k+1th and kth calculations respectively, u(k) is the balancing current at the kth calculation, and e(k) is the error at the kth calculation. is the set of SOC mean values ​​at the kth calculation.

[0058] Optionally, constructing a state equation and a cost function, determining a balancing current according to the weight matrix, and performing active balancing further includes:

[0059] Construct the cost function according to formula (5):

[0060] H=e(k) T Re(k), (5)

[0061] Among them, J is the cost function.

[0062] Through the above technical solution, the present invention provides a fuzzy model prediction active balancing method and balancing circuit. By obtaining the SOC mean and SOC difference as the input values ​​of the controller, obtaining the difference between the reference current and the balancing current as the balancing error, and using the weight ratio of the balancing error and the balancing current as the output value, according to the membership function, the relationship between the input value and the output value and the fuzzy domain and fuzzy subset is established. According to the input value and the fuzzy subset, a fuzzy control rule table is determined. According to the fuzzy control rule table, the membership of the output value corresponding to the input value is obtained, and the membership is defuzzified to obtain the actual current weight. A weight matrix is ​​constructed according to each current weight, and a state equation and cost function are constructed. According to the weight matrix, the balancing current is determined and active balancing is performed. The balancing method and balancing circuit determine the balancing current through model prediction. The balancing current value changes with the change of the balancing state, which can improve the balancing efficiency. By fuzzifying the balancing current weight, the cost function can be flexibly adjusted according to the actual working conditions, further accelerating the balancing speed.

[0063] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following detailed description, they are used to explain the embodiments of the present invention, but do not constitute a limitation of the embodiments of the present invention. In the accompanying drawings:

[0065] Figure 1 This is a flow chart of a fuzzy model prediction active equalization method according to an embodiment of the present invention;

[0066] Figure 2 is a flow chart for obtaining an SOC mean value and an SOC difference value according to an embodiment of the present invention;

[0067] Figure 3 is a flow chart of constructing a state equation and a cost function according to one embodiment of the present invention;

[0068] Figure 4 is a membership function of the SOC difference value according to one embodiment of the present invention;

[0069] Figure 5 is a membership function of the SOC mean value according to one embodiment of the present invention;

[0070] Figure 6 It is a membership function of the weight ratio in one embodiment of the present invention. DETAILED DESCRIPTION

[0071] The following describes the specific implementation of the embodiment of the present invention in detail with reference to the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the embodiment of the present invention and is not used to limit the embodiment of the present invention.

[0072] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of laws and regulations. In the embodiments of this application, certain software, components, models, and other existing solutions in the industry may be mentioned. These should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use such solutions.

[0073] As shown in the figure is a flow chart of the fuzzy model prediction active equalization method of an embodiment of the present invention. Figure 1 In the embodiment of the present invention, the balancing method may include:

[0074] In step S1, the SOC mean value and the SOC difference value are obtained as input values ​​of the controller;

[0075] In step S2, the difference between the reference current and the balancing current is obtained as the balancing error;

[0076] In step S3, the weight ratio of the balancing error and the balancing current is used as the output value;

[0077] In step S4, the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset is established according to the membership function;

[0078] In step S5, a fuzzy control rule table is determined based on the input value and the fuzzy subset;

[0079] In step S6, the membership degree of the output value corresponding to the input value is obtained according to the fuzzy control rule table;

[0080] In step S7, the membership degree is defuzzified to obtain the actual current weight;

[0081] In step S8, a weight matrix is ​​constructed according to each current weight;

[0082] In step S9, a state equation and a cost function are constructed, and according to the weight matrix, a balancing current is determined and active balancing is performed.

[0083] In steps S1 to S9, the SOC mean and SOC difference are first obtained as input values ​​for the controller. The difference between the reference current and the balancing current is then obtained as the balancing error. The weighted ratio of the balancing error and the balancing current is used as the output value. Based on the membership function, the relationship between the input and output values ​​and the fuzzy domain and fuzzy subset is established. Based on the input values ​​and fuzzy subsets, a fuzzy control rule table is determined. Based on the fuzzy control rule table, the membership of the output value corresponding to the input value is obtained. The membership is defuzzified to obtain the actual current weight. A weight matrix is ​​constructed based on each current weight, and a state equation and cost function are constructed. Based on the weight matrix, the balancing current is determined and active balancing is performed. The balancing current is determined through model prediction, and the balancing current value changes with the balancing state, which can improve balancing efficiency. By fuzzifying the balancing current weight, the cost function can be flexibly adjusted according to the actual operating conditions, further accelerating the balancing speed.

[0084] In this Figure 1 In the method shown, step S1 can be used to obtain the SOC mean and SOC difference as input values ​​of the controller. The method for obtaining the SOC mean and SOC difference can be various forms known to those skilled in the art. In one example of the present invention, the method for obtaining the SOC mean and SOC difference can include the following: Figure 2 The steps shown. Specifically:

[0085] In step S11, the SOC mean value is obtained according to formula (1):

[0086]

[0087] in, is the mean SOC, N is the number of battery cells in series, i (i=1,2,3,…,N) is the i-th battery cell, and is the SOC i is the SOC value of the i-th battery cell;

[0088] In step S12, the SOC difference is obtained according to formula (2):

[0089]

[0090] Wherein, ΔSOC is the SOC difference, min(SOC) is the minimum SOC value of the battery cell, and max(SOC) is the maximum SOC value of the battery cell.

[0091] Step S2 can be used to obtain the difference between the reference current and the balancing current as the balancing error. Step S3 can be used to output the weighted ratio of the balancing error and the balancing current. To more intuitively represent the total energy state of the battery pack and the battery cell deviation state, the SOC mean and SOC difference are obtained as controller inputs in step S1, the balancing error value is obtained in step S2, and the weighted ratio of the balancing error and the balancing current in the fuzzy predictive control in step S3 is used as the output value. This constitutes a two-dimensional fuzzy controller, ensuring control accuracy while avoiding design complexity.

[0092] Step S4 is used to establish the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset according to the membership function. The specific method for establishing the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset can be a variety of forms known to those skilled in the art. In one example of the present invention, the value ranges of the SOC difference, the SOC mean and the weight ratio are their fuzzy domains, which correspond to [0, a], [0, b], [c, d], respectively, where a = 0.1, b = 1, c = 100, d = 1000. In order to ensure the accuracy of fuzzy control, the three fuzzy domains are subdivided, that is, the domain is divided into 6 areas. After subdivision, the basic domain of SOC difference is divided into [0, 0.1a], (0.1a, 0.3a], (0.3a, 0.5a], (0.5a, 0.7a], (0.7a, 0.9a], (0.9a, a], S The basic domain of OC mean is divided into [0,0.1b], (0.1b,0.3b], (0.3b,0.5b], (0.5b,0.7b], (0.7b,0.9b], (0.9b,b]; the basic domain of weight ratio is divided into six areas: [c,c+0.1(dc)], (c+0.1(dc),c+0.3(dc)], (c+0.3(dc),c+0.5(dc)], (c+0.5(dc),c+0.7(dc)], (c+0.7(dc),(c+0.9(dc)], (c+0.9(dc),d].

[0093] The continuous quantity in the fuzzy domain is divided into multiple fuzzy subsets that can be expressed in fuzzy language. The fuzzy language variables are S (small), JS (small), M (medium), JL (large), and L (large). The relationship between the fuzzy domain and the fuzzy subset is established in the form of membership function. The membership function of SOC difference, SOC mean and weight ratio can be as follows Figure 4 、 Figure 5 、 Figure 6 shown.

[0094] Step S5 can be used to determine the fuzzy control rule table based on the input value and the fuzzy subset. The fuzzy control idea of ​​the present invention is that when the SOC mean value is large or small, it indicates that the total power of the battery pack is in a high or low state. In both cases, the weight ratio is increased, that is, the SOC error is reduced first to prevent overcharge / discharge under charge / discharge conditions; when the SOC mean value is medium, it indicates that the total power of the battery pack is moderate, the power state and the balance environment are relatively good, and the balance error and balance current input can be comprehensively considered, and the weight ratio takes a medium value; in the above case, regardless of the state of the SOC mean value, the weight ratio is increased as the SOC difference increases, that is, as the SOC deviation becomes larger, its priority for consideration increases. The number of fuzzy rules is determined by the number of fuzzy subsets of the input value. Table 1 is a specific fuzzy control rule table:

[0095] Table 1 Fuzzy control rules table

[0096]

[0097] Step S6 can be used to obtain the membership of the output value corresponding to the input value based on the fuzzy control rule table. According to Table 1, the membership of the weighted ratio can be obtained by looking up the table using the input values ​​of the SOC mean and SOC difference. Step S7 can be used to defuzzify the membership to obtain the actual current weight.

[0098] Step S8 can be used to construct a weight matrix based on each current weight. For each battery cell, the corresponding current weight can be obtained through steps S1 to S7, that is, the weight matrix can be constructed according to formula (3):

[0099]

[0100] Among them, R is the weight matrix, r i is the current weight of the i-th battery cell.

[0101] Let the balancing current of N single cells be u=[u1,u2,u3…u N ] T , the set of SOC is x=[x1,x2,x3…x N ] T , the set of SOC means is Step S9 can be used to construct a state equation and a cost function, determine the balancing current according to the weight matrix, and perform active balancing. The specific method for constructing the state equation and the cost function can be various forms known to those skilled in the art. In one example of the present invention, the specific method for constructing the state equation and the cost function can include the following: Figure 3 The steps shown. Specifically:

[0102] In step S21, the state equation is constructed according to formula (4):

[0103]

[0104] Among them, x(k+1) and x(k) are the sets of SOC at the k+1th and kth calculations respectively, u(k) is the balancing current at the kth calculation, and e(k) is the error at the kth calculation. is the set of SOC mean values ​​at the kth calculation.

[0105] In step S22, a cost function is constructed according to formula (5):

[0106] J=e(k) T Re(k), (5)

[0107] Among them, J is the cost function.

[0108] In each calculation, the possible values ​​of u(k) are traversed and substituted into step S21 to obtain e(k). The value of u(k) that minimizes the cost function is determined in step S22, which is the final balancing current, thereby performing active balancing.

[0109] On the other hand, the present invention also provides a balancing circuit, which includes a plurality of single cells connected in series, a balancing module and a controller, wherein the controller is used for any of the above-mentioned balancing methods. Specifically, the balancing method can be as follows: Figures 1 to 6 The steps shown. Figure 1 In the embodiment of the present invention, the balancing method may include:

[0110] In step S1, the SOC mean value and the SOC difference value are obtained as input values ​​of the controller;

[0111] In step S2, the difference between the reference current and the balancing current is obtained as the balancing error;

[0112] In step S3, the weight ratio of the balancing error and the balancing current is used as the output value;

[0113] In step S4, the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset is established according to the membership function;

[0114] In step S5, a fuzzy control rule table is determined based on the input value and the fuzzy subset;

[0115] In step S6, the membership degree of the output value corresponding to the input value is obtained according to the fuzzy control rule table;

[0116] In step S7, the membership degree is defuzzified to obtain the actual current weight;

[0117] In step S8, a weight matrix is ​​constructed according to each current weight;

[0118] In step S9, a state equation and a cost function are constructed, and according to the weight matrix, a balancing current is determined and active balancing is performed.

[0119] In steps S1 to S9, the SOC mean and SOC difference are first obtained as input values ​​for the controller. The difference between the reference current and the balancing current is then obtained as the balancing error. The weighted ratio of the balancing error and the balancing current is used as the output value. Based on the membership function, the relationship between the input and output values ​​and the fuzzy domain and fuzzy subset is established. Based on the input values ​​and fuzzy subsets, a fuzzy control rule table is determined. Based on the fuzzy control rule table, the membership of the output value corresponding to the input value is obtained. The membership is defuzzified to obtain the actual current weight. A weight matrix is ​​constructed based on each current weight, and a state equation and cost function are constructed. Based on the weight matrix, the balancing current is determined and active balancing is performed. The balancing current is determined through model prediction, and the balancing current value changes with the balancing state, which can improve balancing efficiency. By fuzzifying the balancing current weight, the cost function can be flexibly adjusted according to the actual operating conditions, further accelerating the balancing speed.

[0120] In this Figure 1 In the method shown, step S1 can be used to obtain the SOC mean and SOC difference as input values ​​of the controller. The method for obtaining the SOC mean and SOC difference can be various forms known to those skilled in the art. In one example of the present invention, the method for obtaining the SOC mean and SOC difference can include the following: Figure 2 The steps shown. Specifically:

[0121] In step S11, the SOC mean value is obtained according to formula (1):

[0122]

[0123] in, is the mean SOC, N is the number of battery cells in series, i (i=1,2,3,…,N) is the i-th battery cell, and is the SOC i is the SOC value of the i-th battery cell;

[0124] In step S12, the SOC difference is obtained according to formula (2):

[0125]

[0126] Wherein, ΔSOC is the SOC difference, min(SOC) is the minimum SOC value of the battery cell, and max(SOC) is the maximum SOC value of the battery cell.

[0127] Step S2 can be used to obtain the difference between the reference current and the balancing current as the balancing error. Step S3 can be used to output the weighted ratio of the balancing error and the balancing current. To more intuitively represent the total energy state of the battery pack and the battery cell deviation state, the SOC mean and SOC difference are obtained as controller inputs in step S1, the balancing error value is obtained in step S2, and the weighted ratio of the balancing error and the balancing current in the fuzzy predictive control in step S3 is used as the output value. This constitutes a two-dimensional fuzzy controller, ensuring control accuracy while avoiding design complexity.

[0128] Step S4 is used to establish the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset according to the membership function. The specific method for establishing the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset can be a variety of forms known to those skilled in the art. In one example of the present invention, the value ranges of the SOC difference, the SOC mean and the weight ratio are their fuzzy domains, which correspond to [0, a], [0, b], [c, d], respectively, where a = 0.1, b = 1, c = 100, d = 1000. In order to ensure the accuracy of fuzzy control, the three fuzzy domains are subdivided, that is, the domain is divided into 6 areas. After subdivision, the basic domain of SOC difference is divided into [0, 0.1a], (0.1a, 0.3a], (0.3a, 0.5a], (0.5a, 0.7a], (0.7a, 0.9a], (0.9a, a], S The basic domain of OC mean is divided into [0,0.1b], (0.1b,0.3b], (0.3b,0.5b], (0.5b,0.7b], (0.7b,0.9b], (0.9b,b]; the basic domain of weight ratio is divided into six areas: [c,c+0.1(dc)], (c+0.1(dc),c+0.3(dc)], (c+0.3(dc),c+0.5(dc)], (c+0.5(dc),c+0.7(dc)], (c+0.7(dc),(c+0.9(dc)], (c+0.9(dc),d].

[0129] The continuous quantity in the fuzzy domain is divided into multiple fuzzy subsets that can be expressed in fuzzy language. The fuzzy language variables are S (small), JS (small), M (medium), JL (large), and L (large). The relationship between the fuzzy domain and the fuzzy subset is established in the form of membership function. The membership function of SOC difference, SOC mean and weight ratio can be as follows Figure 4 、 Figure 5 、 Figure 6 shown.

[0130] Step S5 can be used to determine the fuzzy control rule table based on the input value and the fuzzy subset. The fuzzy control idea of ​​the present invention is that when the SOC mean value is large or small, it indicates that the total power of the battery pack is in a high or low state. In both cases, the weight ratio is increased, that is, the SOC error is reduced first to prevent overcharge / discharge under charge / discharge conditions; when the SOC mean value is medium, it indicates that the total power of the battery pack is moderate, the power state and the balance environment are relatively good, and the balance error and balance current input can be comprehensively considered, and the weight ratio takes a medium value; in the above case, regardless of the state of the SOC mean value, the weight ratio is increased as the SOC difference increases, that is, as the SOC deviation becomes larger, its priority for consideration increases. The number of fuzzy rules is determined by the number of fuzzy subsets of the input value. Table 1 is a specific fuzzy control rule table:

[0131] Table 1 Fuzzy control rules table

[0132]

[0133] Step S6 can be used to obtain the membership of the output value corresponding to the input value based on the fuzzy control rule table. According to Table 1, the membership of the weighted ratio can be obtained by looking up the table using the input values ​​of the SOC mean and SOC difference. Step S7 can be used to defuzzify the membership to obtain the actual current weight.

[0134] Step S8 can be used to construct a weight matrix based on each current weight. For each battery cell, the corresponding current weight can be obtained through steps S1 to S7, that is, the weight matrix can be constructed according to formula (3):

[0135]

[0136] Among them, R is the weight matrix, r i is the current weight of the i-th battery cell.

[0137] Let the balancing current of N single cells be u=[u1,u2,u3…u N ] T , the set of SOC is x=[x1,x2,x3…x N ] T , the set of SOC means is Step S9 can be used to construct a state equation and a cost function, determine the balancing current according to the weight matrix, and perform active balancing. The specific method for constructing the state equation and the cost function can be various forms known to those skilled in the art. In one example of the present invention, the specific method for constructing the state equation and the cost function can include the following: Figure 3 The steps shown. Specifically:

[0138] In step S21, the state equation is constructed according to formula (4):

[0139]

[0140] Among them, x(k+1) and x(k) are the sets of SOC at the k+1th and kth calculations respectively, u(k) is the balancing current at the kth calculation, and e(k) is the error at the kth calculation. is the set of SOC mean values ​​at the kth calculation.

[0141] In step S22, a cost function is constructed according to formula (5):

[0142] J=e(k) T Re(k), (5)

[0143] Among them, J is the cost function.

[0144] In each calculation, the possible values ​​of u(k) are traversed and substituted into step S21 to obtain e(k). The value of u(k) that minimizes the cost function is determined in step S22, which is the final balancing current, thereby performing active balancing.

[0145] Through the above technical solution, the present invention provides a fuzzy model prediction active balancing method and balancing circuit. By obtaining the SOC mean and SOC difference as the input values ​​of the controller, obtaining the difference between the reference current and the balancing current as the balancing error, and using the weight ratio of the balancing error and the balancing current as the output value, according to the membership function, the relationship between the input value and the output value and the fuzzy domain and fuzzy subset is established. According to the input value and the fuzzy subset, a fuzzy control rule table is determined. According to the fuzzy control rule table, the membership of the output value corresponding to the input value is obtained, and the membership is defuzzified to obtain the actual current weight. A weight matrix is ​​constructed according to each current weight, and a state equation and cost function are constructed. According to the weight matrix, the balancing current is determined and active balancing is performed. The balancing method and balancing circuit determine the balancing current through model prediction. The balancing current value changes with the change of the balancing state, which can improve the balancing efficiency. By fuzzifying the balancing current weight, the cost function can be flexibly adjusted according to the actual working conditions, further accelerating the balancing speed.

[0146] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0147] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0150] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0151] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0152] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0153] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0154] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A fuzzy model prediction active equalization method, characterized in that: The active balancing method includes: Obtain the SOC mean and SOC difference as input values ​​for the controller; Obtaining the difference between the reference current and the balancing current as the balancing error; Taking the weight ratio of the balancing error and the balancing current as an output value; According to the membership function, the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset is established; Determining a fuzzy control rule table according to the input value and the fuzzy subset; According to the fuzzy control rule table, obtaining the membership degree of the output value corresponding to the input value; Performing a defuzzification operation on the membership degree to obtain an actual current weight; constructing a weight matrix according to each of the current weights; A state equation and a cost function are constructed, and according to the weight matrix, the balancing current is determined and active balancing is performed.

2. The active equalization method according to claim 1, characterized in that: Obtain the SOC mean and SOC difference as input values ​​for the fuzzy controller, including: According to formula (1), the SOC mean value is obtained: in, is the mean SOC, N is the number of battery cells in series, i (i=1,2,3,…,N) is the i-th battery cell, and is the SOC i is the SOC value of the i-th battery cell; According to formula (2), the SOC difference is obtained: Wherein, ΔSOC is the SOC difference, min(SOC) is the minimum SOC value of the battery cell, and max(SOC) is the maximum SOC value of the battery cell.

3. The active equalization method according to claim 2, characterized in that: Obtain the current weight corresponding to each battery cell and construct a weight matrix, including: Construct the weight matrix according to formula (3): Among them, R is the weight matrix, r i is the current weight of the i-th battery cell.

4. The active equalization method according to claim 3, characterized in that: Constructing a state equation and a cost function, determining the balancing current according to the weight matrix, and performing active balancing, including: According to formula (4), the state equation is constructed. Among them, x(k+1) and x(k) are the sets of SOC at the k+1th and kth calculations respectively, u(k) is the balancing current at the kth calculation, and e(k) is the error at the kth calculation. is the set of SOC mean values ​​at the kth calculation.

5. The active equalization method according to claim 4, characterized in that: Constructing a state equation and a cost function, determining the balancing current according to the weight matrix and performing active balancing, further comprising: Construct the cost function according to formula (5): J=e(k) T Re(k), (5) Among them, J is the cost function.

6. A balancing circuit, characterized in that: The circuit includes a plurality of single batteries connected in series, a balancing module, and a controller, wherein the controller is configured to: Obtain the SOC mean and SOC difference as input values ​​for the controller; Obtaining the difference between the reference current and the balancing current as the balancing error; Taking the weight ratio of the balancing error and the balancing current as an output value; According to the membership function, the relationship between the input value and the output value and the fuzzy domain and the fuzzy subset is established; Determining a fuzzy control rule table according to the input value and the fuzzy subset; According to the fuzzy control rule table, obtaining the membership degree of the output value corresponding to the input value; Performing a defuzzification operation on the membership degree to obtain an actual current weight; constructing a weight matrix according to each of the current weights; A state equation and a cost function are constructed, and according to the weight matrix, the balancing current is determined and active balancing is performed.

7. The equalizing circuit according to claim 6, wherein: Obtain the SOC mean and SOC difference as input values ​​for the fuzzy controller, including: According to formula (1), the SOC mean value is obtained: in, is the mean SOC, N is the number of battery cells in series, i (i=1,2,3,…,N) is the i-th battery cell, and is the SOC i is the SOC value of the i-th battery cell; According to formula (2), the SOC difference is obtained: Wherein, ΔSOC is the SOC difference, min(SOC) is the minimum SOC value of the battery cell, and max(SOC) is the maximum SOC value of the battery cell.

8. The equalizing circuit according to claim 7, wherein: Obtain the current weight corresponding to each battery cell and construct a weight matrix, including: Construct the weight matrix according to formula (3): Among them, R is the weight matrix, r i is the current weight of the i-th battery cell.

9. The equalizing circuit according to claim 8, characterized in that: Constructing a state equation and a cost function, determining the balancing current according to the weight matrix, and performing active balancing, including: According to formula (4), the state equation is constructed. Among them, x(k+1) and x(k) are the sets of SOC at the k+1th and kth calculations respectively, u(k) is the balancing current at the kth calculation, and e(k) is the error at the kth calculation. is the set of SOC mean values ​​at the kth calculation.

10. The equalizing circuit according to claim 9, characterized in that: Constructing a state equation and a cost function, determining the balancing current according to the weight matrix and performing active balancing, further comprising: Construct the cost function according to formula (5): J=e(k) T Re(k), (5) Among them, J is the cost function.

Images