Loss modeling method for isolated equalization circuit

Abstract

The invention belongs to the technical field of energy storage and power electronics, and particularly relates to a loss modeling method based on an isolated double-layer equalization circuit. The invention designs an equalization circuit and a corresponding parasitic circuit calculation model analysis method based on a flyback converter, and the principle is as follows: firstly, providing an architecture of an equalization topology, and determining an overall framework and an equalization mode of an equalization system; then modeling and loss analysis of the converter are provided, experimental evaluation is carried out on the designed equalization circuit and the established model, an idea and a theoretical model are provided for power loss analysis in the equalization circuit, and the method has important engineering value for development of energy storage application and optimization of an equalization system.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of energy storage and power electronics, in particular to a loss modeling method for an isolated equalization circuit. BACKGROUND

[0002] In the past few years, the technological advancement of lithium batteries has made it one of the popular choices for battery energy storage system implementation. However, the need to connect many lithium battery cells in series to form a high voltage stack requires proper monitoring to ensure that all the battery cells are working within the manufacturer specified safe operating area. In addition, manufacturing or environmental factors can cause variations between the cells of a battery pack. Therefore, over multiple charge-discharge cycles, these variations can cause the cells to be damaged and compromise the safety of the application.

[0003] To overcome the risks posed by the variations between the battery cells in a battery pack, the current mainstream approach is to use battery management system (BMS) technology. The BMS is composed of multiple modules, which are responsible for not only monitoring individual cells but also equalizing their state of charge (SOC). Various equalization methods can be divided into two categories, namely dissipative and non-dissipative. Dissipative methods refer to transferring the excess energy of overcharged cells to an ohmic resistor. This method is the most cost-effective due to its simplicity, but it is not energy efficient as all the excess energy is converted to heat. In addition, this method can only equalize overcharged cells and cannot support undercharged cells. Non-dissipative equalization generally refers to transferring energy within the battery pack to equalize the state of charge of all the cells. It is clear that many topologies can support this energy transfer. Therefore, non-dissipative equalization can be further divided into adjacent cell equalization and non-adjacent cell equalization.

[0004] Adjacent cell balancing refers to the energy transfer between adjacent cells (AC2AC). This configuration is economically efficient, but the limitations of energy transfer between adjacent cells can require multiple energy conversions and complex scheduling. A commonly used method for adjacent cell balancing is the use of switched capacitors, switched inductors, and buck-boost converters. Non-adjacent cell balancing refers to the energy transfer between non-adjacent cells. This energy transfer can be achieved through a power converter with a switching matrix or using a power converter with a transformer. Non-adjacent cell balancing has a subcategory of balancing architectures. Therefore, in further classification, non-adjacent cell balancing can be achieved through direct cell-to-cell (DC2DC) and module-to-cell (M2C) energy transfer. DC2DC energy transfer refers to the transfer of energy from an overcharged cell to an undercharged cell, using a switching matrix to select a pair of cells to be balanced. M2C energy transfer refers to the transfer of energy from a cell module to an undercharged cell or from an overcharged cell to a cell module (C2M). If the total number of cells is low, then the entire battery pack can represent a module. In this case, the configuration can be considered as a group-to-cell (P2C). It must be pointed out that P2C is not common due to the high voltage of the battery pack. On the contrary, the pack is divided into many modules, and therefore, the M2C configuration is adopted.

[0005] The number of different equalizers installed in the battery pack characterizes the configuration from centralized (only one shared equalizer for the entire battery pack) to distributed (there are multiple equalizers in the battery pack). The increase in the number of different converters increases the total cost of the stack, but it provides more flexible balancing, since multiple cells can be balanced simultaneously. Therefore, the total duration of the operation is reduced. In addition, the reduction in the number of equalizers in the battery pack increases the scheduling complexity to achieve complete balancing.

[0006] From this classification, it can be observed that there are many configurations of non-dissipative cell balancing. Each one has advantages and disadvantages that must be evaluated according to the specific requirements of the application. In this document, a bidirectional M2C / C2M configuration is studied using a bidirectional flyback converter. The advantage of this configuration is that it allows the simultaneous balancing of multiple non-adjacent cells without the need for multiple conversions, which is important in fast BMS.

[0007] In this context, we provide a loss modeling method for an isolated balancing circuit. SUMMARY

[0008] The purpose of the present invention is to provide a loss modeling method for an isolated balancing circuit to solve the problems raised in the background art.

[0009] To achieve the above-mentioned purpose, the present invention aims to provide a loss modeling method for an isolated balancing circuit, comprising the following steps:

[0010] S1. The battery pack stacked in series is divided into M series modules, each module contains Q battery cells;

[0011] S2. A synchronous rectifier bidirectional flyback converter is provided on each battery cell, with its input connected to the positive and negative electrodes of the battery and its output connected to the terminals of the module to which it belongs, forming a series input / parallel output configuration;

[0012] S3. The flyback converter has two operating modes: M2C energy transfer is performed when the battery SOC is low, and C2M energy transfer is performed when the SOC is high. The bidirectional energy flow is controlled by algebraic superposition of module-side currents. All flyback converters are designed to operate in discontinuous conduction mode.

[0013] S4. Model the top-level balancing topology corresponding to a battery string with parasitic parameters, perform equivalent series modeling on the effect of leakage inductance of primary and secondary windings, and select synchronous rectification bidirectional flyback to work in DCM. In mode 1, Q ij When Qi is turned on and turned off, loops A and B are formed. Based on the KVL law, the Laplace domain equation is derived. In mode 2, Q ij When turned off, TVS tube Dz is turned on, Qi is turned on, forming loops A, B and C. The current equation is derived based on Kirchhoff's law. In mode three, all switches are turned off, forming loops B and C. The current equation is derived based on Kirchhoff's law.

[0014] S5. Calculate the losses on the transformer, including core loss and copper loss. The core loss is calculated using the improved generalized Steinmetz equation, and the copper loss is calculated by considering the winding resistance. Calculate the conduction loss and switching loss of the MOSFET switch tube, calculate the loss on the TVS tube, and calculate the total loss during energy transfer during the entire balancing process.

[0015] As a further improvement of the present technical solution, the modal modeling in step S3 includes:

[0016] The ohmic losses in the windings and the MOSFETs on both the primary and secondary sides are modeled by the formula:

[0017]

[0018] Among them, R o is the ohmic loss, R dson,s Secondary side MOS tube Q ij The on-resistance, R Cu,s is the copper wire resistance of the transformer secondary winding, I s,rms is the effective value of the current on the secondary side, R dson,p is the on-resistance of the primary side MOS tube, R Cu,p is the copper wire resistance of the transformer primary winding, I p,rmsThe current effective value for the primary side.

[0019] As a further improvement of the technical solution, the modal one modeling in step S4 comprises:

[0020] In the modal one, loop A is composed of Sij-Rs-Lk-Lm-Qij, loop B is composed of Lm-Rc, according to the KVL law of loop A and B, the loop equation in Laplace domain can be obtained:

[0021]

[0022] I B1 ·R c -(I A1 -I B1 )·s·L m =0;

[0023] Wherein, V ij is the terminal voltage of the battery string S ij , I A1 , I B1 are the currents working in the loop A, B under the modal one respectively, after solving the currents I A1 , I B1 together, the time-domain current expression is obtained by inverse Laplace transform.

[0024] As a further improvement of the technical solution, the modal two modeling in step S4 comprises:

[0025] In the modal two, loop A is composed of D z -L k -R s -L m , loop B is composed of L m -R c , loop C is composed of R c -V pack -Q i , the equation group is obtained based on Kirchhoff's law:

[0026]

[0027] Wherein, V Dz is the equivalent voltage source of D z , I A2 , I B2 , I C2 are the currents working in the loop A, B, C under the modal two respectively, I Lm,2 is the current flowing through L m in the modal two, I sw is the current flowing through the switch, and n is the number of turns ratio, after solving, the time-domain current expression is obtained by inverse Laplace transform.

[0028] As a further improvement of the present technical solution, it further comprises a step S5 of calculating the losses on the transformer, including the core losses and the copper losses, the core losses are calculated by the improved generalized Steinmetz equation, the copper losses are calculated by considering the winding resistance, the conduction losses and the switching losses of the MOSFET switch are calculated, the losses on the TVS are calculated, the total losses during the energy transfer in the whole balancing process are calculated:

[0029] The core loss calculation includes:

[0030] The core losses are calculated by the improved generalized Steinmetz equation:

[0031]

[0032] Wherein, PiGSE is the core loss, ki is the iGSE method constant, fsw is the switching frequency, Bmax is the maximum magnetic flux density, the voltage VLm,j applied on Lm by different modes is calculated, and the equivalent resistance Rc is calculated accordingly:

[0033]

[0034] As a further improvement of the present technical solution, the copper loss calculation in the step S5 includes:

[0035] The copper loss of the transformer is calculated as follows:

[0036]

[0037] Wherein, R Cu,p and R Cu,s are estimated by considering the skin effect, proximity effect and the geometric characteristics of the winding and the core.

[0038] As a further improvement of the present technical solution, the switching loss calculation in the step S5 includes:

[0039] The conduction loss of the MOSFET is calculated as follows:

[0040]

[0041] Wherein, R dson,s is the MOSFET on-resistance, I s，rms is the current effective value.

[0042] As a further improvement of the present technical solution, the switching loss calculation in the step S5 includes:

[0043] The switching loss calculation formula is as follows:

[0044]

[0045] where t vr is the time required for the voltage to rise, t if is the time required for the current to fall, I smax is the maximum current i A1 (t=t1) that is the maximum current during the entire operation of the battery cell delivering energy to the battery pack, in DCM the conduction losses of the switch are negligible.

[0046] As a further improvement of the present technical solution, the step S5 of the TVS tube loss calculation comprises:

[0047] The loss calculation formula on the TVS tube is as follows:

[0048]

[0049] where L k is the leakage inductance, I s,max is the maximum current, f sw is the switching frequency.

[0050] As a further improvement of the present technical solution, the step S5 of the total loss calculation comprises:

[0051] The total loss calculation during the entire equalization process of energy transfer is as follows:

[0052]

[0053] Theoretical calculation and estimation based on ideal analysis of the converter;

[0054] In the present invention, a flyback converter with parasitic elements will be modeled. The bidirectional flyback converter is a symmetrical converter. Therefore, to demonstrate the proposed method, the C2M energy flow is considered. To analyze the M2C energy flow, the "P" and "S" indices in the corresponding equations should be exchanged. The synchronous rectification bidirectional flyback is chosen to operate in DCM. Therefore, the converter operates as a current source and the switching losses during the high-side switch on period can be neglected.

[0055] Compared with the prior art, the present invention has the beneficial effects that:

[0056] In a loss modeling method of an isolated equalization circuit, the proposed flyback converter is a simpler dc / dc converter with isolation function, which can independently realize the equalization of non-adjacent units without complex operation, and the flyback converter operates as a direct current source, which is enabled and disabled when equalization is needed, the proposed design method allows the simple converter to operate with high output current and high efficiency, the proposed analysis can eliminate the need for current sensors in sensorless open-loop operation, the accuracy of the converter duty cycle required to estimate the equalization current, the flyback converter model considers the leakage inductance of the transformer, core loss, copper loss of the transformer and on-resistance of the semiconductor, accurate estimation of power loss and its distribution on device components provides useful insights for further efficiency improvement, compared with experimental results, the small deviation of the proposed model depends on potential errors in transformer parameter measurement, estimation of Steinmetz parameters through curve fitting, temperature influence and switching time, and has good engineering application value. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 A configuration schematic diagram of an M2C type equalization circuit based on a flyback converter is provided for the embodiment of the present application.

[0058] Figure 2 A schematic diagram of an equivalent parasitic parameter model of a bidirectional flyback equalization circuit is provided for the embodiment of the present application.

[0059] Figure 3 A conduction interval, equalization current waveform schematic diagram is provided for the embodiment of the present application.

[0060] Figure 4 A loss distribution test result diagram of a flyback converter is provided for the embodiment of the present application.

[0061] Figure 5 The overall workflow diagram of the present application DETAILED DESCRIPTION

[0062] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0063] Embodiment 1

[0064] Please refer to Figures 1-5The principle is as follows: first, the architecture of the balanced topology of the present invention is proposed to determine the overall framework and balanced mode of the balanced system; then the modeling and loss analysis of the converter are proposed, and the established model is evaluated.

[0065] For the flyback converter-based balancing circuit, the system architecture and loss modeling analysis process are as follows:

[0066] S1. The battery pack stacked in series is divided into M series modules, each module contains Q battery cells;

[0067] S2. A synchronous rectifier bidirectional flyback converter is provided on each battery cell, with its input connected to the positive and negative electrodes of the battery and its output connected to the terminals of the module to which it belongs, forming a series input / parallel output configuration;

[0068] S3. The flyback converter has two operating modes: M2C energy transfer is performed when the battery SOC is low, and C2M energy transfer is performed when the SOC is high. The bidirectional energy flow is controlled by algebraic superposition of module-side currents. All flyback converters are designed to operate in discontinuous conduction mode.

[0069] S4. Model the top-level balancing topology corresponding to a battery string with parasitic parameters, perform equivalent series modeling on the effect of leakage inductance of primary and secondary windings, and select synchronous rectification bidirectional flyback to work in DCM. In mode 1, Q ij When Qi is turned on and turned off, loops A and B are formed. Based on the KVL law, the Laplace domain equation is derived. In mode 2, Q ij When turned off, TVS tube Dz is turned on, Qi is turned on, forming loops A, B and C. The current equation is derived based on Kirchhoff's law. In mode three, all switches are turned off, forming loops B and C. The current equation is derived based on Kirchhoff's law.

[0070] S5. Calculate the losses on the transformer, including core loss and copper loss. The core loss is calculated using the improved generalized Steinmetz equation, and the copper loss is calculated by considering the winding resistance. Calculate the conduction loss and switching loss of the MOSFET switch tube, calculate the loss on the TVS tube, and calculate the total loss during energy transfer during the entire balancing process.

[0071] The first mode modeling in step S4 includes: the ohmic losses of the winding and the primary and secondary MOSFETs are modeled using equation (1).

[0072]

[0073] where Ro is the ohmic loss, Rdson,s is the on-resistance of the secondary MOS Qij, RCu,s is the copper loss of the secondary winding of the transformer, Is,rms is the current root mean square of the secondary side, similarly, Rdson,p, RCu,p, Ip,rms are the parameters of the primary side.

[0074] The modeling of the mode one in step S4 also includes: when the equalization circuit works in mode one (0 < t < t1), Qij is on, Qi is off, in this period of time, loops A, B are formed, loop A is composed of Sij-Rs-Lk-Lm-Qij, loop B is composed of Lm-Rc. According to the KVL law of loops A and B, the loop equations in the Laplace domain as shown in equations (2), (3) can be obtained:

[0075]

[0076] I B1 ·R c -(I A1 -I B1 )·s·L m =0 (3)

[0077] where IA1, IB1 are the currents of loops A, B working in mode one, respectively, and Vij is the terminal voltage of the battery string Sij.

[0078] Solving (2) and (3) together, the currents IA1, IB1 in the Laplace domain as shown in equations (4-29), (4-30) are obtained:

[0079]

[0080] Taking the inverse Laplace transform of (4), iA1(t) is obtained as:

[0081]

[0082] Similarly, taking the inverse Laplace transform of (5), iB1(t) is obtained as:

[0083]

[0084] The currents in the magnetizing inductances are expressed by equations (6), (7)

[0085]

[0086] When the equalizer works in mode two (t1 < t < t2), Qij is off, TVS Dz is on, and the current through it decreases rapidly. Meanwhile, MOSFET Qi is on, and the current through it increases rapidly. During this time, loops A, B and C are formed. Loop A consists of Dz-Lk-Rs-Lm, loop B consists of Lm-Rc, and loop C consists of Rc-Vpack-Qi. The value of the current through Lm and Lk at the end of mode one, ILm,1, constitutes the initial condition of mode two. According to Kirchhoff's law for loops A, B and C, the following equations can be obtained:

[0087]

[0088]

[0089] wherein,

[0090]

[0091] wherein VDz is the equivalent voltage source of Dz, IA2, IB2 and IC2 are the currents of loops A, B and C respectively, ILm,2 is the current through Lm in mode two, Isw is the current through the switch, and n is the turns ratio.

[0092] Combining equations (9)-(13), IA2, IB2 and IC2 can be obtained.

[0093]

[0094] wherein,

[0095]

[0096] Taking inverse Laplace transform, the time-domain forms of IA2, IB2 and IC2 can be obtained as shown in equations (17), (18) and (19).

[0097]

[0098] During mode two, the voltage of Lm is -Vpack / n, thus the current of Lm can be calculated by equation (20).

[0099]

[0100] When the equalization circuit works in mode three (t2 < t < t3), all switches are off, and the current through switch Qij is zero, but the body diode of MOSFET Q; is still in the on state. During this period, loops B and C are formed. Loop B includes Lm-Rc. Loop C includes Rc-Vpack / n-Qi. The value of the current flowing through Lm at the end of mode two forms the initial condition of mode three. According to Kirchhoff's law of loops B and C, the following equations can be obtained:

[0101] (I B3 -I C3 )·R c +L m ·I Lm,3 +I B3 ·s·L m =0 (21)

[0102]

[0103] Solving (21), (22) gives IC3and its inverse Laplace transform gives equation (23)

[0104]

[0105] After the model analysis of the equalization topology with parasitic parameters is completed, the losses existing in it can be reasonably evaluated. Among them, the losses on the transformer include core loss and copper loss. The core loss is obtained by the improved generalized Steinmetz equation (iGSE), and its expression is:

[0106]

[0107] In the formula, PiGSEis the core loss, kiis the iGSE method constant, fswis the switching frequency, and Bmaxis the maximum magnetic flux density.

[0108] Then, the voltage VLm,japplied on Lm in different modes is calculated by the following formula:

[0109]

[0110] After the core loss is calculated, the equivalent resistance Rc can be calculated:

[0111]

[0112] The copper loss of the transformer is calculated as shown in equation (29). It is worth noting that in the design stage of the transformer, the values of RCu,pand RCu,sare estimated considering the skin effect, proximity effect, and the geometric characteristics of the winding and the core.

[0113]

[0114] In addition to the losses on the transformer, the losses due to the leakage inductance of the transformer and the MOSFET switch also need to be considered. The conduction loss calculation formula of the MOSFET is as follows:

[0115]

[0116] In addition, the calculation of switching loss is shown in equation (31). In DCM, the conduction loss of the switch can be ignored. Therefore, the switching loss of the semiconductor only occurs during the off period.

[0117]

[0118] where tvr is the time required for the voltage to rise, tif is the time required for the current to fall, and Ismax is the current of iA1(t=t1), that is, the maximum current in the entire operation process of the battery unit transferring energy to the battery pack.

[0119] During mode two, the loss on the TVS tube is due to the energy stored in the leakage inductance of the transformer:

[0120]

[0121] In actual testing, the main measured current is the balancing current IEq of the circuit, so for simplicity, let

[0122]

[0123] At this point, the total loss (34) during energy transfer in the entire balancing process can be calculated,

[0124]

[0125] Example 2:

[0126] In order to verify the effectiveness and accuracy of the loss modeling method of the isolated double-layer balancing circuit provided by the present application, in this example, for a battery pack composed of 12 battery monomers, an equalization test is performed using a balancing circuit based on a flyback converter, the effectiveness of the proposed balancing method is verified, and the accuracy of the proposed loss model is verified based on the experimental circuit parameters.

[0127] wherein the experimental result data is shown in Table 1, and the experimental circuit parameters are shown in Table 2.

[0128] Table 1 Experimental data table of balancing circuit based on flyback converter

[0129]

[0130] Voltage consistency index: the initial voltage standard deviation V of the battery pack as a whole δ_in_pack= 0.2044, average V avg_in_pack = 3.746 V; voltage standard deviation after rest V δ_ev_pack = 0.0196 V, average V avg_ev_pack = 3.487 V. Analysis of the equalization effect: the voltage standard deviation is significantly reduced (AV_5= 0.1848), with a reduction of 90.4%, indicating that the equalization circuit effectively plays an equalization role and improves the inconsistency of the battery pack voltage; the average voltage after equalization decreases AV_avg= 0.259 (a reduction of 6.9%), also indicating that there is energy loss in this equalization process, and the actual equalization efficiency η2= 93.1%.

[0131] Table 2 experimental circuit parameters

[0132]

[0133] Based on the predicted values of the above loss model, in general, the analyzed loss model is accurate and reliable in predicting the equalization system. The main reason for the error is the heat loss in the circuit system, especially in the non-isolated equalization circuit. Due to the topological structure characteristics, the longer energy transfer path will cause more heat loss when the current passes through the wire in the process of energy transfer. While the isolated equalization circuit can maximize the energy loss caused by the transfer process due to its structural characteristics, but because it uses more power elements, there is a large parasitic loss in the circuit, especially the multi-winding transformer used in this paper, which has a large leakage inductance due to its structural characteristics, thereby greatly increasing the energy loss on the TVS tube.

[0134] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only preferred examples of the present application and are not intended to limit the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A loss modeling method for an isolated balancing circuit, characterized by: It includes the following steps: S1. The series-stacked battery pack is divided into M series modules, and each module contains Q battery cells; S2. A synchronous rectifier flyback converter is set on each battery cell, its input terminal is connected to the positive and negative electrodes of the battery, and the output terminal is connected to the terminal of the module to form a series input / parallel output configuration; S3. The flyback converter has two operating modes: when the battery SOC is low, M2C energy transfer is performed, and when the SOC is high, C2M energy transfer is performed. The bidirectional energy flow is controlled by the algebraic superposition of the module-side currents, and all flyback converters are designed to operate in the discontinuous conduction mode; S4. Model the top-level balancing topology corresponding to a battery string with parasitic parameters, perform equivalent series modeling on the effect of leakage inductance of primary and secondary windings, and select synchronous rectification bidirectional flyback to work in DCM. In mode 1, Q ij When Qi is turned on and turned off, loops A and B are formed. Based on the KVL law, the Laplace domain equation is derived. In mode 2, Q ij When turned off, TVS tube Dz is turned on, Qi is turned on, forming loops A, B and C. The current equation is derived based on Kirchhoff's law. In mode three, all switches are turned off, forming loops B and C. The current equation is derived based on Kirchhoff's law.

2. The loss modeling method for an isolated equalizing circuit according to claim 1, wherein: The mode 1 modeling in step S3 includes: The ohmic losses of the winding and the MOSFETs on the primary and secondary sides are modeled by the formula: Among them, R o is the ohmic loss, R dson,s Secondary side MOS tube Q ij The on-resistance, R Cu,s is the copper wire resistance of the transformer secondary winding, I s,rms is the effective value of the current on the secondary side, R dson,p is the on-resistance of the primary side MOS tube, R Cu,p is the copper wire resistance of the transformer primary winding, I p,rms is the effective value of the current on the primary side.

3. The loss modeling method for an isolated equalizing circuit according to claim 1, wherein: The mode 1 modeling in step S3 also includes: In mode 1 (0 < t < t1), Qij is turned on and Qi is turned off. During this period, loops A and B are formed. Loop A consists of Sij - Rs - Lk - Lm - Qij, and loop B consists of Lm - Rc. According to the KVL law of loops A and B, the loop equations in the Laplace domain can be obtained: I B1 ·R C -(I A1 -I B1 )·s·L m =0; Among them, V ij For battery string S ij The terminal voltage, I A1 , I B1 They are the currents of loops A and B working in the mode, and the current I is solved after combining them. A1 , I B1 , and perform inverse Laplace transform to obtain the time domain current expression.

4. The loss modeling method for an isolated equalizing circuit according to claim 1, wherein: The mode 2 modeling in step S3 includes: In mode 2, loop A is connected by D z -L k -R s -L m Loop B consists of L m -R c The loop C consists of R c -V pack -Q i Composition, based on Kirchhoff's law, we get the equation group: Among them, V Dz D z The equivalent voltage source, I A2 , I B2 , I C2 are the currents of loops A, B, and C working in mode 2, I Lm,2 In mode 2, the flow through L m Current, I sw is the current flowing through the switch, n is the turns ratio, and after solving, an inverse Laplace transform is performed to obtain the time-domain current expression.

5. The loss modeling method for an isolated equalizing circuit according to claim 1, wherein: It also includes step S5. Calculate the losses on the transformer, including core losses and copper losses. The core losses are calculated by the improved generalized Steinmetz equation, the copper losses are calculated by considering the winding resistance, calculate the conduction losses and switching losses of the MOSFET switching tubes, calculate the losses on the TVS tubes, and calculate the total losses during the energy transfer in the whole balancing process: Among them, the calculation of the core losses includes: The core losses are calculated by the improved generalized Steinmetz equation: Among them, P iGSE is the core loss, k i is the iGSE method constant, f sw is the switching frequency, B max is the maximum magnetic flux density, calculate the different modes applied to L m The voltage V Lm,j， And calculate the equivalent resistance R accordingly c :

6. The loss modeling method for an isolated equalizing circuit according to claim 1, wherein: The calculation of the copper losses in step S5 includes: The copper losses of the transformer are calculated as follows: Among them, R Cu,p and R Cu,s The value of is estimated taking into account the skin effect, proximity effect and the geometric characteristics of the winding and core.

7. The loss modeling method for an isolated equalizing circuit according to claim 1, wherein: The calculation of the switching losses in step S5 includes: The conduction loss calculation formula of the MOSFET is as follows: Among them, R dson,s is the MOSFET on-resistance, I s，rms is the effective value of current.

8. The loss modeling method for an isolated equalizing circuit according to claim 1, wherein: The calculation of the switching losses in step S5 includes: The switching loss calculation formula is as follows: Among them, t vr is the time required for the voltage to rise, t if is the time required for the current to fall, I smax is i A1 The current at (t=t1) is the maximum current during the entire operation of the battery cell transferring energy to the battery pack. In DCM, the conduction loss of the switch is negligible.

9. The loss modeling method for an isolated equalizing circuit according to claim 1, wherein: The calculation of the TVS tube losses in step S5 includes: The loss calculation formula on the TVS tube is as follows: Among them, L k is the leakage inductance, I s,max is the maximum current, f sw is the switching frequency.

10. The loss modeling method of an isolated equalizing circuit according to claim 1, characterized in that: The calculation of the total losses in step S5 includes: The total losses during the energy transfer in the whole balancing process are calculated as follows:

Images