A secondary recovery battery energy storage system thermal balance and state of charge balance dynamic optimization control method

Abstract

This invention discloses a dynamic optimization control method for thermal balance and state of charge (SOC) balance in a secondary recycled battery energy storage system. By controlling the current flowing through the battery, thermal balance and SOC balance in the secondary recycled battery energy storage system can be achieved. Furthermore, this invention provides a fusion control and optimization method that intelligently adjusts the coupling weight between the thermal balance target and the SOC balance target under different operating conditions for energy storage systems composed of batteries with varying degrees of age, thereby achieving dynamic optimization of the thermal balance and SOC balance control targets of the energy storage system.

Description

Technical Field

[0001] This invention belongs to the field of battery equalization control, and more specifically, relates to a dynamic optimization control method for thermal balance and charge balance of a secondary recovery battery energy storage system. Background Technology

[0002] In energy storage systems composed of recycled batteries, different batteries may exhibit manufacturing differences due to factors such as their age, production processes, and raw material quality. These differences include variations in electrochemical performance and internal resistance. Different batteries may also operate under different conditions, such as charge / discharge rates, temperature environments, and cycle counts, which can further contribute to performance variations. Prolonged use and cyclic charging / discharging lead to battery aging, with different batteries aging at different rates, potentially causing further differences. These differences are amplified during charging and discharging due to variations in battery temperature and state of charge (SOC) levels. For example, differences in internal resistance between batteries can cause larger batteries to overheat more significantly when a large current flows through the battery pack due to Joule heating. Temperature imbalances between batteries can lead to thermal runaway in the entire energy storage system, affecting its safe and stable operation. Furthermore, considering differences in remaining capacity, when all batteries in the system are charged or discharged at the same current, smaller capacity batteries will reach their threshold first, leading to overcharging or over-discharging issues, similarly impacting system stability.

[0003] To reduce temperature differences between batteries, methods for balancing the temperatures of different batteries often utilize external heat-conducting media, such as air cooling, liquid cooling, and phase change material (PCM) cooling. Chinese patent CN220189748U describes an optimized air cooling device that uses an improved heat dissipation duct structure to increase the heat dissipation area, accelerate heat dissipation, and reduce temperature differences. Chinese patent CN116315253A describes a heat dissipation method that uses water as a heat-conducting medium and designs a high-density toothed heat sink to control battery temperature. PCM refers to materials that can store or release a large amount of heat during phase change processes (such as transitions between solid and liquid states). Theoretically, phase change materials (PCMs) can effectively control battery temperature and achieve thermal balance. However, the thermal conductivity of currently developed PCMs is low and cannot fully meet practical needs. The paper "SAKhateeb, MMFarid, JRSelman and S. Al-Hallaj. Design and simulation of alithium-ion battery with a phase change material thermal management system for an electric scooter[J]. Journal of Power Sources, 2004" uses paraffin wax as a PCM to cool the battery system. Experiments show that the temperature of the central battery increases by 26.5℃-30℃, while the peripheral batteries only increase by 18.75℃-22.5℃, with a maximum temperature difference of nearly 10℃ between batteries, indicating that the thermal balance effect is not ideal. There are also control methods that combine the above-mentioned heat dissipation schemes. Chinese patent CN113013522A combines air cooling and PCM heat dissipation, using a heat dissipation module and ventilation system to achieve battery temperature control. However, PCMs with high latent heat, high specific heat, high thermal conductivity, stability, non-toxicity, and low cost are still under development. Currently, most battery thermal balance solutions are based on passive heat dissipation methods using external devices.

[0004] There are two main methods for balancing the State of Charge (SOC) of batteries: passive balancing and active balancing. Passive balancing reduces the SOC differences among batteries in a battery storage system through natural discharge. This typically involves using energy-consuming components such as parallel resistors to dissipate excess energy as heat via the Joule effect, thus achieving SOC balance. Active balancing uses a specially designed balancing circuit to control current flow, transferring excess energy from higher-voltage batteries to lower-voltage batteries, thereby balancing the SOC of the batteries within the battery pack. For example, Chinese patent CN116707099B uses an optimization algorithm based on an SOC prediction model to adjust the current flowing through each battery while avoiding over-balancing, achieving active balancing. Chinese patent CN107834655A utilizes a multi-winding transformer to first absorb excess energy magnetically at the transformer, and then release it as electrical energy to a suitable location within the battery pack.

[0005] Therefore, it can be seen that almost all of the aforementioned thermal balancing methods require external heat dissipation devices, increasing system costs. Furthermore, most current battery thermal balancing methods rely on heat transfer media to move heat from the battery elsewhere to achieve thermal equilibrium. However, the fundamental reason for temperature differences lies in the inconsistent heat generation levels caused by differences in internal resistance between batteries, and the other technologies mentioned above cannot fundamentally solve this heat source difference. In addition, because the internal resistance of recycled batteries often varies greatly, the difference in heat generation is even greater than in energy storage systems composed of ordinary batteries. Therefore, traditional thermal balancing methods are insufficient to meet the balancing requirements and need improvement.

[0006] In traditional battery energy storage systems, thermal balance control and SOC balance control are often implemented independently, with no coupling between them. However, when it comes to secondary recycling battery energy storage systems, the control loops for thermal balance and SOC balance are likely to be coupled, bringing new challenges to balance control. For example, when performing SOC balance on a battery with high internal resistance, the battery generates more Joule heat because current needs to flow through it to achieve balance, thus disrupting the battery's temperature balance.

[0007] Therefore, for energy storage systems composed of batteries of varying ages, there is an urgent need to design a dynamic optimization control method for the thermal balance and charge balance of secondary recycled battery energy storage systems. This method should be able to intelligently adjust the coupling weight between the thermal balance target and the SOC balance target under different operating conditions, so that the thermal balance and SOC balance of the energy storage system meet the control requirements. Summary of the Invention

[0008] (a) Technical problems to be solved

[0009] To address the shortcomings mentioned in the background technology, this invention discloses a dynamic optimization control method for thermal balance and state of charge (SOC) balance in a secondary recycled battery energy storage system. This method eliminates the need for additional external heat dissipation devices, achieving battery thermal balance and SOC balance solely by controlling the current flowing through the battery. Furthermore, this invention provides a fusion control and optimization method that intelligently adjusts the coupling weight between thermal balance and SOC balance targets under different operating conditions for energy storage systems composed of batteries with varying degrees of age, thereby dynamically optimizing the thermal balance and SOC balance control objectives of the energy storage system.

[0010] (II) Technical Solution

[0011] This invention discloses a dynamic optimization control method for thermal balance and charge balance in a secondary recycled battery energy storage system. Each battery is connected to a DC-DC converter to form a battery cell. N independent battery cells supply power to the output load, providing the DC bus voltage V. BUS N≥2; the duty cycles of the switching transistors in the DC-DC converters of the N battery cells are D1,D2,...,D N It is generated by a controller, with the outer loop being a voltage controller and the inner loop being a balance controller;

[0012] In the voltage controller, each battery cell has a desired voltage V. 1_ref V 2_ref ,...,V N_ref It consists of the gain coefficients l1, l2, ..., l of the balanced control loop. N The calculations show that the desired voltage is related to the current battery cell terminal voltages V1, V2, ..., V. N The error is the input signal of the voltage controller, which is then processed through the voltage controller's transfer function G. VB (z) After the action, the output shows the switching duty cycle D1, D2, ..., D of each battery cell. N The expected voltage of each battery cell is:

[0013]

[0014] In the balance controller, a reference weight λ is used. i The balance factor α between SOC and temperature i β i Combined into gain coefficient l i The expression is:

[0015] l i =λ i α i +(1-λ i )β i (3)

[0016] The output switching duty cycle is:

[0017]

[0018] Due to the differences in remaining capacity among recycled batteries, normalization parameters k1,k2,...,k need to be introduced when calculating the SOC reference value. N Assume the remaining capacity of each battery is Q1, Q2, ..., Q... N And Q max =max{Q1,Q2,...,Q N The normalization parameter is calculated as follows:

[0019]

[0020] Battery SOC reference value SOC i For battery B in the i-th battery cell i The state of charge of each battery B i The temperature is T i The calculated battery temperature reference value is

[0021] The balance controller takes the error signals of the reference value and the actual value as inputs, and obtains the balance factor α of SOC and temperature through the balance controller transfer function, respectively. i and β i The subscript i takes values ​​from 1 to N;

[0022]

[0023]

[0024] The controller first obtains the current SOC and temperature data of each battery, as well as the DC bus voltage and current, and then adjusts the balance factor α accordingly. i and β i Calculations must be performed before the optimized gain coefficient l can be used. i The duty cycles D1, D2, ..., D of each battery cell can be obtained from Equation 4. N .

[0025] Preferably, the DC-DC converter is a synchronous Boost circuit.

[0026] Preferred, SOC i The relationship between the battery current and its state of charge (SOC) calculated using the ampere-hour integration method is expressed as follows:

[0027]

[0028] i(t) is the current flowing through the battery at time t, SOC(t) is the battery's SOC at time t, and Q is the current. nom This refers to the battery capacity.

[0029] Preferably, when the two control targets of SOC and temperature on a battery are coupled, and the battery operating temperature is higher than the ambient temperature, the maximum and minimum allowable current flowing through the battery are respectively I... max I min If the current flowing through the battery is controlled to be I∈[I min ,I max Although simultaneous balance cannot be achieved at the current control moment, the SOC difference and temperature difference of the batteries can be maintained within a reasonable range. As the batteries radiate heat to the outside world, when the system is in the next moment, the battery temperature decreases accordingly, the coupling between SOC balance and thermal balance weakens, and the balance current is reselected until a certain moment when the SOC difference and temperature difference of the two batteries are reduced to a small range, thus achieving dynamic balance.

[0030] Preferably, the balance controller employs the following PSO (Particle Swarm Optimization) algorithm to optimize the reference weight λ. i :

[0031] First, determine the dimension of the solution space, which is the number of batteries N in the system, determine the number of particles W, and initialize the position of each particle, which is the reference weight λ for each battery. i The initial velocity of the particles is also a random number, determined by the reference weight λ. i By combining equations 3 and 4 with the state equations for the SOC and temperature of all batteries, the cost function J under this reference weight is obtained. The optimal reference weight is determined by comparing the magnitude of the cost function J at the position of each generation of particles, and the particle velocity and position are updated. Through a finite number of iterations, W particles search for the position in the solution space that minimizes the cost function J, thus finding the optimal duty cycle D of all battery cell switches. 1_opt D 2_opt ,...,D N_opt .

[0032] Preferably, for the gain coefficient l i Optimization requires first establishing a cost function, considering the thermodynamic dynamic equation of the i-th cell:

[0033]

[0034] Among them, T i (t) is the current battery surface temperature, T env The current ambient temperature, h is the battery's heat transfer coefficient, A is the area of ​​the battery involved in heat transfer, and C... T_i It is the battery's specific heat capacity, Ri It is resistance, I i (t) is the current flowing through the battery at the current moment, D i (t) is the duty cycle of the switching transistor at the current moment. The dynamic equation for battery SOC is:

[0035]

[0036] Among them, SOC i Q(t) is the current battery SOC, Q i It refers to battery capacity;

[0037] Assuming that the battery temperature and bus current change very little over one cycle, and replacing the instantaneous current with the average current flowing through the battery over one cycle, integrating the two differential equations in equations 8-9 over one cycle yields the discrete state equations:

[0038]

[0039] SOC i (k), T i (k), I BUS (k) corresponds to the SOC, temperature, and bus current at time k, D i (k) represents the duty cycle at time k, and τ represents the period length; let The state equations for the SOC and temperature of all batteries in the system are expressed in the following form:

[0040]

[0041] Where, at the current time k:

[0042] SOC(k)=[SOC1(k) SOC2(k)…SOC N (k)] T (12)

[0043]

[0044]

[0045] T(k) = [T1(k) T2(k) … T N (k)] T (15)

[0046]

[0047]

[0048]

[0049]

[0050] F = [θ1 θ2 … θ N ] T (20)

[0051] The input error signals of SOC and temperature in the balance controller are expressed as Equations 21-22 respectively:

[0052] M SOC (k)=M avg ·SOC(k) (21)

[0053] M T (k)=M avg ·T(k) (22)

[0054]

[0055] Therefore, the cost function for SOC and temperature is:

[0056]

[0057] Where ω SOC and ω T The weights are two symmetric positive definite matrices of dimension N×N; a scaling factor γ is introduced. SOC and γ T The cost functions of SOC and temperature are merged into a single cost function J:

[0058]

[0059] As shown in Equation 11-25, the cost function J is based on the ambient temperature T. env and duty cycle D i (k) is a multivariate function of variables.

[0060] Preferably, to simplify the problem analysis, it is assumed that the ambient temperature is approximately constant over a period of time.

[0061] (III) Beneficial Effects

[0062] 1. This invention considers the large differences in internal resistance among batteries in a secondary recycled battery energy storage system. By controlling the current flowing through the batteries, the temperature difference among the batteries in the battery pack is reduced, and active temperature balancing of the batteries can be achieved without the need for additional heat dissipation devices. In addition, this invention considers the large differences in remaining capacity among batteries in a secondary recycled battery energy storage system. By controlling the current flowing through the batteries, the difference in state of charge (SOC) of the batteries is reduced, the efficiency of the energy storage system is improved, and overcharging or over-discharging of the batteries is avoided.

[0063] 2. In addition, based on the effective integration of active temperature balancing and active SOC balancing control by setting the controller, this invention also designs a fusion cost function according to SOC and temperature requirements. Under different operating conditions, the PSO optimization algorithm searches for the reference weight that minimizes the cost function in the solution space, finds the optimal duty cycle of all battery cell switching transistors, and thus better achieves dynamic balance between battery temperature and SOC. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in this invention or the prior art, the accompanying drawings used in the embodiments will be briefly described below:

[0065] Figure 1 This is a schematic diagram of the energy storage system structure containing N battery cells in this invention;

[0066] Figure 2 The following are the balance curves of the two battery systems: (a) shows the change in battery SOC, (b) shows the change in SOC difference, (c) shows the change in battery temperature, and (d) shows the change in temperature difference.

[0067] Figure 3 This is a schematic diagram of the energy storage system balance controller in this invention, wherein (a) to (c) are the voltage control block diagram, the SOC control block diagram, and the temperature control block diagram, respectively;

[0068] Figure 4 This is a flowchart of the energy storage system balance control in this invention;

[0069] Figure 5 This is a graph showing the relationship between the cost function J and the duty cycles D1 and D2 in an embodiment of the present invention.

[0070] Figure 6 This is a flowchart of the gain coefficient optimization process in the balance controller of the energy storage system of the present invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0072] The balancing device in this invention includes a main circuit and a control circuit, such as... Figure 1As shown. The main circuit uses multiple independent recycled batteries as DC power sources. Each battery is connected to a DC-DC converter, forming a battery unit (BU). The current flowing through the battery is regulated. Figure 1 The DC-DC converter in this system is a synchronous Boost converter, but it can also be other types of DC-DC converters. The duty cycles of the switching transistors in the DC-DC converter of each battery cell are D1, D2, ..., D... N The controller generates the signal, and the driver controls the switching on and off of the switching transistor. The outputs of multiple battery cells 1 through N are connected in series to provide a suitable DC bus voltage V. BUS .

[0073] Figure 1 The circuit in the middle contains N battery cells, B1 is the energy storage battery in the first battery cell, and V B1 I is the terminal voltage of battery B1. B1 Let T1 be the current flowing through battery B1, and T1 be the temperature of battery B1. Battery B1, connected to inductor L, capacitor C, and switch S, forms battery cell 1. V1 represents the terminal voltage of battery cell 1. BUS I BUS These are the DC bus voltage and current, respectively.

[0074] The input to the energy storage system controller is the terminal voltage V of each battery. B1 V B2 ,...,V BN Current I B1 ,I B2 ,...,I BN Temperatures T1, T2, ..., T N Battery cell terminal voltages V1, V2, ..., V N DC bus voltage V BUS DC bus current I BUS The controller output is the duty cycle of the switching transistors D1, D2, ..., D. N .

[0075] Using the ampere-hour integration method, the relationship between the battery current and its state of charge (SOC) can be expressed as:

[0076]

[0077] Where i(t) is the current flowing through the battery at time t, SOC(t) is the battery SOC at time t, and Q... nom This refers to the battery capacity.

[0078] To verify the fusion controllability of battery thermal balance and SOC balance, the present invention conducted the following analysis:

[0079] Due to the significant differences in the newness and oldness of secondarily recycled batteries, there are corresponding different situations when balancing the batteries. Suppose there are two secondarily recycled batteries in a battery energy storage system and they are in the discharging state. The internal resistances of Battery 1 and Battery 2 are R1 and R2 respectively, and R1 > R2. The SOC values of the batteries are SOC1 and SOC2, and the temperatures are T1 and T2 respectively. Then the following situations exist:

[0080] Situation 1: If SOC1 > SOC2 and T1 < T2; when balancing, as long as the current flowing through Battery 1 is relatively increased, the SOC change rate of the battery is The battery heating power is making the SOC1 of Battery 1 change faster and the temperature T1 increase faster, and the SOC difference and temperature difference between the two batteries will both decrease;

[0081] Situation 2: If SOC1 > SOC2 and T1 > T2; when balancing, if the current flowing through Battery 1 is increased, then the SOC change rate of the battery can reduce the SOC difference between the two batteries and achieve battery SOC balance. However, due to the heating power brought by the larger current being making Battery 1 heat up faster and it is difficult to achieve battery thermal balance. Similarly, if the current flowing through Battery 1 is decreased, making then the battery thermal balance can be achieved, but it brings trouble to the battery SOC balance.

[0082] In the above two situations: In Situation 1, at a certain moment, only by adjusting the battery current, the SOC balance and thermal balance can be achieved; in Situation 2, due to the mutual coupling of the two control objectives of SOC and temperature, the two balances cannot be achieved simultaneously by a single control means, the control is more complex, and two control purposes need to be considered.

[0083] Suppose in Situation 2, the operating temperature of the battery is higher than the ambient temperature, and the maximum and minimum allowable currents flowing through Battery 1 are I 1max 、I 1min , and when the current flowing through Battery 1 is I 1max then When the current flowing through Battery 1 is I 1min then If the current flowing through the battery is controlled to be I1 ∈ [I 1min , I 1maxWhile simultaneous balance cannot be achieved at the current control moment, the SOC and temperature difference between the two batteries can be maintained within a reasonable range. Since the batteries can dissipate heat from the outside environment (radiative heat dissipation), as the system moves to the next moment, the battery temperature decreases accordingly, the coupling between SOC balance and thermal balance weakens, and the balance current magnitude is reselected until, at a certain moment, the SOC and temperature difference of both batteries are reduced to a small range. At this point, dynamic balance between the two can be achieved without increasing hardware costs.

[0084] Based on the above control theory Figure 2 To determine the balance control curve of an energy storage system containing two batteries within 100 seconds, allowable values ​​for SOC difference and temperature difference are set to 0.1 and 1.5℃, respectively. Figure 2 Point A in Figure (b) is a stationary point on the curve. This is because the battery temperature difference exceeds a given limit at this point, and the controller adjusts the current to reduce the weight of SOC control and instead increases the weight of temperature control. Due to the large system inertia, Figure 2 In graph (d), the curve rises continuously until it reaches its maximum value at point B. Before this point, the SOC difference exceeds a given limit, after which the controller increases the SOC control weight accordingly to achieve dynamic adjustment of the two balance objectives. Because the battery continuously exchanges heat with the external environment, the temperature difference change function has a pole with a negative real part in the complex frequency domain, exhibiting a convergent oscillation in the time domain. The controller continuously adjusts the weights of the two control objectives, ultimately achieving balance.

[0085] like Figure 3 As shown, the entire control loop is divided into two parts: the outer loop is the voltage control loop, and the inner loop is the balance control loop. The voltage control loop is responsible for controlling the DC bus voltage within a certain range, while the balance control loop achieves balance control by combining the SOC and temperature data of each battery based on the voltage control.

[0086] Voltage control block diagram as follows Figure 3 As shown in Figure (a), each battery cell has a desired voltage V. 1_ref V 2_ref ,...,V N_ref It consists of the gain coefficients l1, l2, ..., l of the balanced control loop. N Calculated. The desired voltage is related to the current battery cell terminal voltages V1, V2, ..., V. N The error is the input signal of the voltage controller, which is then processed through the voltage controller's transfer function G. VB (z) After the action, the output shows the switching duty cycle D1, D2, ..., D of each battery cell. N The expected voltage of each battery cell is:

[0087]

[0088] The balance controller uses a reference weight λ i SOC and temperature balance factor α i β i Combined into gain coefficient l i The expression is:

[0089] l i =λ i α i +(1-λ i )β i (3)

[0090] The output switching duty cycle is:

[0091]

[0092] Balance control such as Figure 3 As shown in Figures (b) and (c), since the controller aims to achieve a dynamic balance between SOC and temperature, a reference value for SOC and temperature is required for each battery. Due to the differences in remaining capacity among recycled batteries, normalization parameters k1,k2,...,k are needed when calculating the SOC reference value. N Assume the remaining capacity of each battery is Q1, Q2, ..., Q... N And Q max =max{Q1,Q2,...,Q N The normalization parameter is calculated as follows:

[0093]

[0094] Battery SOC reference value Battery temperature reference value

[0095] The balance controller takes the error signals of the reference value and the actual value as inputs, and obtains the balance control factor α of SOC and temperature through the balance controller transfer function, respectively. i (i = 1, 2, ..., N) and β i (i = 1, 2, ..., N), where N is the number of battery cells.

[0096]

[0097]

[0098] Figure 4The process for the controller to perform balancing involves first obtaining the current SOC and temperature data of each battery, as well as the DC bus voltage and current, to calculate the balancing factor. Only then can the duty cycle of each battery cell be obtained through the optimized gain coefficient.

[0099] First, the current flowing through the battery is measured using a current sensor, and the battery surface temperature is acquired using a temperature sensor. These two sets of data are then sent to the controller. The controller, which balances temperature and State of Charge (SOC), processes the acquired battery state data to estimate the battery's current SOC. The controller then needs to... Figure 4 The parameters such as the balance factor and gain coefficient are analyzed to determine the battery's state of charge (SOC) and temperature. Based on the current operating conditions, the reference weight λ is optimized. i Calculate the optimal duty cycle D1, D2, ..., D for the switching transistor in each battery cell. N To achieve the control and dynamic optimization of temperature balance and SOC balance in the battery system, ultimately achieving the optimal system efficiency.

[0100] To further achieve dynamic optimization control of temperature balance and SOC balance, such as Figure 6 As shown, this invention employs the following PSO (Particle Swarm Optimization) algorithm to optimize the reference weight λ. i .

[0101] First, regarding the gain coefficient l i Optimization requires first establishing a cost function, considering the thermodynamic dynamic equation of the i-th cell:

[0102]

[0103] Among them, T i (t) is the current battery surface temperature, T env The current ambient temperature, h is the battery's heat transfer coefficient, A is the area of ​​the battery involved in heat transfer, and C... T_i It is the battery's specific heat capacity, R i It is resistance, I i (t) is the current flowing through the battery at the current moment, D i (t) represents the duty cycle of the switching transistor at the current moment. Furthermore, the battery SOC dynamic equation is:

[0104]

[0105] Among them, SOC i Q(t) is the current battery SOC, Q i It refers to battery capacity.

[0106] To facilitate controller solution, the following approximation is made: assuming that the battery temperature and bus current change very little within one cycle, and replacing the instantaneous current with the average current flowing through the battery within one cycle, the two differential equations (8) to (9) above are integrated over time within one cycle to obtain the discrete state equation:

[0107]

[0108] SOC i (k), T i (k), I BUS (k) corresponds to the SOC, temperature, and bus current at time k, D i (k) represents the duty cycle at time k, and τ represents the period length; for convenience, let The state equations for the SOC and temperature of all batteries in the system are expressed in the following form:

[0109]

[0110] Where, at the current time k:

[0111] SOC(k)=[SOC1(k) SOC2(k)…SOC N (k)] T (12)

[0112]

[0113]

[0114] T(k) = [T1(k) T2(k) … T N (k)] T (15)

[0115]

[0116]

[0117]

[0118]

[0119] F = [θ1 θ2 … θ N ] T (20)

[0120] The input error signals of SOC and temperature in the balance controller are expressed as Equations 21-22 respectively:

[0121] M SOC (k)=M avg·SOC(k) (21)

[0122] M T (k)=M avg ·T(k) (22)

[0123]

[0124] The cost function for SOC and temperature is:

[0125]

[0126] Where ω SOC and ω T The weights are two symmetric positive definite matrices of dimension N×N; a scaling factor γ is introduced. SOC and γ T The cost functions of SOC and temperature are merged into a single cost function J:

[0127]

[0128] As shown in Equation 11-25, the cost function J is based on the ambient temperature T. env and duty cycle D i (k) is a multivariate function of variables. To simplify the problem analysis, it is assumed that the ambient temperature is approximately constant within one cycle.

[0129] Figure 5 The figure shows the cost function J versus duty cycle. When the system contains two batteries, the ambient temperature is 25℃ and the cycle length is 0.5s. Other parameters are shown in Table 1. Figure 5 This indicates that the controller needs to focus on balancing the battery temperature at this time, adjusting the duty cycle D1 of battery 1, which has a higher temperature, to be lower than the duty cycle D2 of battery 2, thereby reducing heat generation and minimizing the system cost function J.

[0130] Table 1 Detailed Battery Specifications

[0131]

[0132]

[0133] Figure 6 The diagram shows a flowchart of PSO optimization for the gain coefficient. The controller first determines the dimension of the solution space, which is the number of batteries N in the system, and the number of particles W. The position of each particle is initialized, i.e., the reference weight λ for each battery. i The initial velocity of the particles is also a random number, determined by the reference weight λ. iBy combining equations 3, 4, 11, and 20, the cost function value J under this reference weight can be obtained. The optimal reference weight is determined by comparing the magnitude of the cost function at the position of each generation of particles, and the particle velocity and position are updated. Through a finite number of iterations, W particles search for the position (reference weight) that minimizes the cost function in the solution space, thus finding the optimal duty cycle D of all battery cell switches. 1_opt D 2_opt ,...,D N_opt .

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dynamic optimization control of thermal balance and charge balance in a secondary recycled battery energy storage system, characterized in that, Each battery is connected to a DC-DC converter, forming a battery cell. These N independent battery cells power the output load, providing the DC bus voltage V. BUS N≥2; the duty cycles of the switching transistors in the DC-DC converters of the N battery cells are D1,D2,...,D N It is generated by a controller, with the outer loop being a voltage controller and the inner loop being a balance controller; In the voltage controller, each battery cell has a desired voltage V. 1_ref V 2_ref ,...,V N_ref It consists of the gain coefficients l1, l2, ..., l of the balanced control loop. N The calculations show that the desired voltage is related to the current battery cell terminal voltages V1, V2, ..., V. N The error is the input signal of the voltage controller, which is then processed through the voltage controller's transfer function G. VB (z) After the action, the output shows the switching duty cycle D1, D2, ..., D of each battery cell. N The expected voltage of each battery cell is: In the balance controller, a reference weight λ is used. i The balance factor α between SOC and temperature i β i Combined into gain coefficient l i The expression is: l i =λ i a i +(1-l i )b i (2) The output switching duty cycle is: Due to the differences in remaining capacity among recycled batteries, normalization parameters k1,k2,...,k need to be introduced when calculating the SOC reference value. N Assume the remaining capacity of each battery is Q1, Q2, ..., Q... N And Q max =max{Q1,Q2,...,Q N The normalization parameter is calculated as follows: Battery SOC reference value SOC i For battery B in the i-th battery cell i The state of charge of each battery B i The temperature is T i The calculated battery temperature reference value is The balance controller takes the error signals of the reference value and the actual value as inputs, and obtains the balance factor α of SOC and temperature through the balance controller transfer function, respectively. i and β i The subscript i takes values ​​from 1 to N; The controller first obtains the current SOC and temperature data of each battery, as well as the DC bus voltage and current, and then adjusts the balance factor α accordingly. i and β i Perform calculations, and then use the gain coefficient l i The duty cycles D1, D2, ..., D of each battery cell can be obtained from Equation 4. N .

2. The dynamic optimization control method for thermal balance and charge balance of the secondary recycled battery energy storage system according to claim 1, characterized in that, The DC-DC converter is a synchronous Boost circuit.

3. The dynamic optimization control method for thermal balance and charge balance of the secondary recycled battery energy storage system according to claim 1, characterized in that, SOC i The relationship between the battery current and its state of charge (SOC) calculated using the ampere-hour integration method is expressed as follows: i(t) is the current flowing through the battery at time t, SOC(t) is the battery's SOC at time t, and Q is the current through the battery. nom This refers to the battery capacity.

4. The dynamic optimization control method for thermal balance and charge balance of the secondary recycled battery energy storage system according to claim 3, characterized in that, When the State of Charge (SOC) and temperature are coupled as control targets for a battery, and the battery operating temperature is higher than the ambient temperature, the maximum and minimum allowable current flowing through the battery are respectively I... max I min If the current flowing through the battery is controlled to be I∈[I min ,I max Although simultaneous balance cannot be achieved at the current control moment, the SOC difference and temperature difference of the batteries can be maintained within a reasonable range. As the batteries radiate heat to the outside world, when the system is in the next moment, the battery temperature decreases accordingly, the coupling between SOC balance and thermal balance weakens, and the balance current is reselected until a certain moment when the SOC difference and temperature difference of the two batteries are reduced to a small range, thus achieving dynamic balance.

5. The dynamic optimization control method for thermal balance and charge balance of the secondary recycled battery energy storage system according to claim 4, characterized in that, The balance controller employs the following PSO (Particle Swarm Optimization) algorithm to optimize the reference weight λ. i : First, determine the dimension of the solution space, which is the number of batteries N in the system, determine the number of particles W, and initialize the position of each particle, which is the reference weight λ for each battery. i The initial velocity of the particles is also a random number, determined by the reference weight λ. i By combining equations 3 and 4 with the state equations for the SOC and temperature of all batteries, the cost function J under this reference weight is obtained. The optimal reference weight is determined by comparing the magnitude of the cost function J at the position of each generation of particles, and the particle velocity and position are updated. Through a finite number of iterations, W particles search for the position in the solution space that minimizes the cost function J, thus finding the optimal duty cycle D of all battery cell switches. 1_opt D 2_opt ,...,D N_opt .

6. The dynamic optimization control method for thermal balance and charge balance of the secondary recycled battery energy storage system according to claim 5, characterized in that, The PSO particle swarm optimization algorithm also includes: For gain coefficient l i Optimization requires first establishing a cost function, considering the thermodynamic dynamic equation of the i-th cell: Among them, T i (t) is the current battery surface temperature, T env The current ambient temperature, h is the battery's heat transfer coefficient, A is the area of ​​the battery involved in heat transfer, and C... T_i It is the battery's specific heat capacity, R i It is resistance, I i (t) is the current flowing through the battery at the current moment, D i (t) is the duty cycle of the switching transistor at the current moment. The dynamic equation for battery SOC is: Among them, SOC i Q(t) is the current battery SOC, Q i It refers to battery capacity; Assuming that the battery temperature and bus current change very little over one cycle, and replacing the instantaneous current with the average current flowing through the battery over one cycle, integrating the two differential equations in equations 8-9 over one cycle yields the discrete state equations: SOC i (k), T i (k), I BUS (k) corresponds to the SOC, temperature, and bus current at time k, D i (k) represents the duty cycle at time k, and τ represents the period length; let The state equations for the SOC and temperature of all batteries in the system are expressed in the following form: Where, at the current time k: SOC(k)=[SOC1(k) SOC2(k) … SOC N (k)] T (12) T(k)=[T1(k) T2(k) … T N (k)] T (15) F=[θ1 θ2 … θ N ] T (20) The input error signals of SOC and temperature in the balance controller are expressed as Equations 21-22 respectively: M SOC (k)=M avg ·SOC(k) (21) M T (k)=M avg ·T(k) (22) Therefore, the cost function for SOC and temperature is: Where ω SOC and ω T The weights are two symmetric positive definite matrices of dimension N×N; a scaling factor γ is introduced. SOC and γ T The cost functions of SOC and temperature are merged into a single cost function J: As shown in Equation 11-25, the cost function J is based on the ambient temperature T. env and duty cycle D i (k) is a multivariate function of variables.

7. The dynamic optimization control method for thermal balance and charge balance of a secondary recycled battery energy storage system according to claim 6, characterized in that, To simplify the problem analysis, we assume that the ambient temperature is approximately constant over a period of time.

Images