A battery temperature control method

Abstract

The application provides a battery temperature control method, relates to the technical field of power grid energy storage batteries, and can timely adjust the battery temperature according to the battery temperature change. The specific technical scheme is as follows: a simulation model of a controlled battery object is constructed according to state parameters of the controlled battery object, and the simulation model is used to determine state data of the controlled battery object; the controlled battery object comprises a liquid metal battery; an equivalent thermal resistance network model of the controlled battery object is constructed according to a heat transfer process of the controlled battery object, and thermal parameters of the equivalent thermal resistance network model are determined based on the state data; a simplified control model is constructed based on the equivalent thermal resistance network model and the thermal parameters, and the simplified control model is used to predict temperature changes of the controlled battery object under different auxiliary heating powers; and rolling optimization is performed according to the state data and the simplified control model, so as to determine a target auxiliary heating power, and the target auxiliary heating power is used to control the temperature of the controlled battery object.

Description

Technical Field

[0001] This application relates to the field of grid energy storage battery technology, and in particular to a battery temperature control method. Background Technology

[0002] With the advent of dual-carbon energy and the continued growth of global energy demand, batteries (such as liquid metal batteries), as a device for large-scale static energy storage applications at the grid level, are gradually moving towards large-scale production and industrialization. Battery temperature directly affects its performance and safety, and therefore needs to be controlled.

[0003] However, the proportional-integral-derivative (PID) control strategy is currently commonly used to control battery temperature. This strategy cannot regulate battery temperature in a timely manner, resulting in response lag, which accelerates the corrosion and aging of internal battery materials, reduces battery life, and in severe cases may even lead to sealing failure or even fire and explosion, posing a serious threat to battery operational safety. Summary of the Invention

[0004] This application provides a battery temperature control method that can adjust the battery temperature in a timely manner according to changes in battery temperature.

[0005] To achieve the above objectives, the embodiments of this application adopt the following technical solutions: Firstly, a battery temperature control method is provided. This method includes: constructing a high-fidelity multiphysics finite element simulation model of the controlled battery object based on its state parameters. The high-fidelity multiphysics finite element simulation model is used to determine the state data of the controlled battery object. The state parameters include at least geometric parameters, electrochemical parameters, and thermodynamic parameters. The state data includes at least the temperature of the controlled battery object, the air temperature, the insulation layer temperature, and the battery's heat generation power. The controlled battery object includes a liquid metal battery. Based on the heat transfer process of the controlled battery object, an equivalent thermal resistance network model of the controlled battery object is constructed, and the thermal parameters of the equivalent thermal resistance network model are determined based on the state data. Based on the equivalent thermal resistance network model and the thermal parameters, a simplified control model is constructed. The simplified control model is used to predict the temperature change of the controlled battery object under different auxiliary heating powers. Based on the state data and the simplified control model, rolling optimization is performed to determine the target auxiliary heating power, which is used to control the temperature of the controlled battery object.

[0006] This scheme firstly, by combining the geometric, electrochemical, and thermodynamic parameters of the controlled battery, a high-fidelity multiphysics finite element simulation model is accurately constructed. This model can precisely simulate the internal thermal state of a real battery, providing accurate state data for subsequent control. Next, by constructing an equivalent thermal resistance network model, the complex heat transfer process is simplified into a predictive model, and thermal parameters are determined. This allows the system to construct a simplified control model based on the equivalent thermal resistance network model and thermal parameters. The simplified control model can predict the temperature changes of the controlled battery under different auxiliary heating powers and perform rolling optimization based on the prediction results to determine the target auxiliary heating power. This enables proactive responses to battery temperature changes, rather than passively waiting for temperature deviations to occur before adjustment. Thus, it solves the problem of delayed battery temperature regulation response, allowing timely adjustment of the battery temperature based on temperature changes, improving the real-time performance and accuracy of temperature control. Simultaneously, because the battery temperature can be controlled in a timely and precise manner, it avoids prolonged operation at excessively high temperatures, effectively slowing down the corrosion and aging of internal materials, extending battery life, and reducing the safety hazards of sealing failure or even battery explosion caused by temperature runaway, thereby improving the safety and reliability of battery operation.

[0007] In one possible implementation of the first aspect, the high-fidelity multiphysics finite element simulation model includes a one-dimensional electrochemical simulation model and a three-dimensional thermally coupled simulation model; the one-dimensional electrochemical simulation model is used to determine the heat generation power of the battery, and the three-dimensional thermally coupled simulation model is used to determine the evolution of the battery temperature field, which includes the evolution of the temperature of the controlled battery object, the air temperature, and the temperature of the insulation layer.

[0008] In another possible implementation of the first aspect, the equivalent thermal resistance network model adopts a lumped parameterization method, which equates the heat transfer process of the controlled battery object to a thermal resistance network including battery nodes, air nodes, and insulation layer nodes; the thermal parameters include the equivalent heat capacity of each node and the equivalent thermal resistance between each node.

[0009] In another possible implementation of the first aspect, the thermal parameters of the equivalent thermal resistance network model are determined based on state data, including: determining the thermal parameters of the equivalent thermal resistance network model based on state data using recursive least squares (RLS) with a forgetting factor; wherein the initial thermal parameters of the equivalent thermal resistance network model are determined by a thermal quasi-static method.

[0010] In another possible implementation of the first aspect, the simplified control model is a state-space equation obtained by discretizing the equivalent thermal resistance network model; the simplified control model includes a state matrix, an input matrix, and a disturbance matrix; the state matrix, the input matrix, and the disturbance matrix are determined based on thermal parameters.

[0011] In another possible implementation of the first aspect, rolling optimization is achieved through a model predictive controller. The rolling optimization process includes: within the prediction time domain, the model predictive controller predicts the temperature prediction trajectory corresponding to different candidate auxiliary heat power sequences based on state data and a simplified control model; the prediction time domain is the time length predicted by the model predictive controller; each candidate auxiliary heat power sequence and its corresponding temperature prediction trajectory are evaluated according to a multi-objective cost function, and the candidate auxiliary heat power sequence that minimizes the value of the multi-objective cost function is determined as the target auxiliary heat power; wherein, the multi-objective cost function includes a temperature tracking error term, an energy loss term, and a power smoothing term.

[0012] In another possible implementation of the first aspect, a disturbance estimator is used to correct the target temperature inside the model-predicted controller based on the deviation between the actual temperature of the controlled battery object and the target temperature. The corrected target temperature is then used to evaluate the temperature prediction trajectory through a multi-objective cost function.

[0013] In another possible implementation of the first aspect, the disturbance estimator is a first-order low-pass filter disturbance estimator; using the disturbance estimator, the target temperature inside the model predictor controller is corrected based on the deviation between the actual temperature and the target temperature of the controlled battery object, including: using the first-order low-pass filter disturbance estimator, the equivalent disturbance estimate at the current sampling time is recursively calculated based on the deviation between the actual temperature and the target temperature of the controlled battery object at the current sampling time and the equivalent disturbance estimate at the previous sampling time; the corrected target temperature is obtained by subtracting the equivalent disturbance estimate at the current sampling time from the target temperature.

[0014] In a second aspect, an electronic device is provided, comprising: a memory and one or more processors; the memory and the processors are coupled; wherein the memory stores computer program code, the computer program code including computer instructions, which, when executed by the processor, cause the electronic device to perform the battery temperature control method described in any of the first aspects.

[0015] Thirdly, a computer-readable storage medium is provided, including computer instructions that, when executed on an electronic device, cause the electronic device to perform the battery temperature control method described in any of the first aspects above.

[0016] Fourthly, a computer program product is provided, which, when run on a computer, causes the computer to execute the battery temperature control method described in any of the first aspects above.

[0017] It is understood that the beneficial effects achieved by the electronic device described in the second aspect, the computer-readable storage medium described in the third aspect, and the computer program product described in the fourth aspect can be referred to in the beneficial effects of the first aspect and any of its possible design embodiments, and will not be repeated here. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of a controlled battery object provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a battery management system provided in an embodiment of this application; Figure 3 A schematic flowchart illustrating a battery temperature control method provided in an embodiment of this application; Figure 4 A schematic diagram of an analog equivalent circuit provided in an embodiment of this application; Figure 5 A schematic diagram of the battery heat generation power curve of a liquid metal battery under constant auxiliary heating power during the 0.2C charge-discharge process provided in the embodiments of this application; Figure 6 A schematic diagram showing the battery temperature results of the PID battery temperature control method and the battery temperature control method of the embodiment of this application when the liquid metal battery is charged and discharged at 0.2C. Figure 7 A schematic diagram showing the auxiliary heating power results of the PID battery temperature control method and the battery temperature control method of the embodiment of this application when the liquid metal battery is charged and discharged at 0.2C. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. In the description of this application, unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can represent A or B. "And / or" in this application is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple. Furthermore, to facilitate a clear description of the technical solutions in the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish identical or similar items with substantially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that "first" and "second" are not necessarily different. Meanwhile, in the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is being used as an example, illustration, or description. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of terms such as "exemplary" or "for example" is intended to present related concepts in a concrete manner for ease of understanding.

[0020] With the emergence of dual-carbon energy and the further development of global energy demand, renewable energy sources, represented by solar and wind power, have become important power generation methods. Their intermittency and instability necessitate energy storage technologies to smooth peak flows, ensure grid supply and demand balance, and guarantee safe and stable operation. As a rapidly developing energy storage method, electrochemical energy storage technologies, including lithium-ion battery energy storage, sodium-ion battery energy storage, and flow battery energy storage, play an irreplaceable role in energy storage systems. Liquid metal batteries, as batteries specifically designed for large-scale grid-scale static energy storage, are gradually moving towards large-scale industrialization.

[0021] Liquid metal batteries (LMBs) are a novel type of energy storage battery that utilizes a fully liquid electrochemical system. Their operating temperature ranges from 300°C to 700°C. A typical structure, from top to bottom, consists of a low-density metal anode, a molten salt electrolyte layer, and a high-density metal cathode. Under thermodynamic equilibrium, they automatically form a stable three-layer self-assembled configuration, exhibiting significant advantages such as intrinsic safety, low cost, long lifespan, and high efficiency. Due to their unique operating temperature, maintaining the molten state of the anode, cathode, and electrolyte requires ensuring the battery's operating temperature remains within a reasonable range, thus placing high demands on the battery's thermal management system.

[0022] However, current battery temperature control strategies cannot regulate battery temperature in a timely manner, resulting in a lag in response.

[0023] For example, current temperature control strategies for liquid metal batteries (or auxiliary heating strategies) generally employ passive proportional-integral-derivative (PID) error feedback. The PID control scheme adjusts the auxiliary heating power by comparing the difference between the measured real-time temperature and the target temperature using proportional, integral, and derivative operations. PID performs well in temperature control scenarios with low thermal inertia and simple operating conditions. Although the PID control strategy is very easy to implement in terms of algorithm, it adjusts based on real-time temperature error feedback, which is a typical model-free feedback control. Essentially, it passively compensates for hysteresis based on the already generated temperature error. This passive adjustment exhibits significant hysteresis in temperature control and cannot be adjusted in a timely manner, leading to major drawbacks when applied to liquid metal battery systems with large thermal inertia. For example, when the system faces changing operating conditions, the heat generation and absorption characteristics of the battery may change significantly. Since the PID control strategy cannot obtain the change in the internal heat generation power of the battery to judge the trend of battery temperature change, it can only wait for the temperature to change before adjusting the power. Therefore, it cannot adjust the battery temperature in time according to the battery temperature change, resulting in a response lag problem. This lag may lead to temperature overshoot, high-frequency oscillation of heater power, and unnecessary energy loss.

[0024] Therefore, this application provides a battery temperature control method, which includes: constructing a high-fidelity multiphysics finite element simulation model of the controlled battery object based on its state parameters; the high-fidelity multiphysics finite element simulation model is used to determine the state data of the controlled battery object; the state parameters include at least geometric parameters, electrochemical parameters, and thermodynamic parameters; the state data includes at least the temperature of the controlled battery object, the air temperature, the insulation layer temperature, and the battery heat generation power; the controlled battery object includes a liquid metal battery. Based on the heat transfer process of the controlled battery object, an equivalent thermal resistance network model of the controlled battery object is constructed, and the thermal parameters of the equivalent thermal resistance network model are determined based on the state data. Based on the equivalent thermal resistance network model and the thermal parameters, a simplified control model is constructed, which is used to predict the temperature change of the controlled battery object under different auxiliary heating powers. Based on the state data and the simplified control model, rolling optimization is performed to determine the target auxiliary heating power, which is used to control the temperature of the controlled battery object.

[0025] This scheme first constructs a high-fidelity multiphysics finite element simulation model to accurately simulate the internal thermal state of a real battery, providing accurate state data for subsequent control. Next, an equivalent thermal resistance network model simplifies the complex heat transfer process into a predictive model and determines thermal parameters. This allows the system to construct a simplified control model based on the equivalent thermal resistance network model and thermal parameters. The simplified control model can predict the temperature changes of the controlled battery under different auxiliary heating powers and perform rolling optimization based on the prediction results to determine the target auxiliary heating power. This enables proactive response to battery temperature changes, rather than passively waiting for temperature deviations to occur before adjustment, thus solving the problem of lag in battery temperature regulation response. It allows for timely adjustment of the battery temperature based on temperature changes, improving the real-time performance and accuracy of temperature control. This timely response to temperature regulation avoids temperature overshoot, high-frequency oscillations in heater power, and unnecessary energy loss caused by lag in temperature regulation response.

[0026] Meanwhile, because it can regulate the battery temperature in a timely and precise manner, it avoids the battery from operating at excessively high temperatures for a long time, effectively slows down the corrosion and aging of internal materials, extends battery life, and reduces the safety hazards of sealing failure or even battery explosion caused by temperature runaway, thus improving the safety and reliability of battery operation.

[0027] First, the controlled battery object involved in the embodiments of this application is introduced, such as... Figure 1As shown, this application uses a battery pack (i.e., the controlled battery object) consisting of four liquid metal batteries as an example for illustration. Liquid metal batteries differ from traditional lithium-ion batteries; their operating temperature ranges from 300°C to 700°C, and their heat generation characteristics are primarily reversible, exhibiting an initial heat absorption followed by heat release during operation. The thermal inertia of this system is much greater than that of lithium-ion batteries, and its thermodynamic parameters are prone to drift during operation. Currently, most control strategies for lithium-ion batteries focus on controlling the liquid cooling flow rate. However, the unique high-temperature operating environment of liquid metal batteries dictates that their control objective must maintain the operating temperature and cannot excessively pursue energy conservation. Low temperatures can cause the molten material to solidify, leading to battery failure. Therefore, methods for controlling the liquid cooling flow rate cannot be directly applied to liquid metal batteries. Thus, reasonable auxiliary heating power management is needed to ensure that the liquid metal battery maintains its operating temperature.

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described in this application are merely illustrative of the battery temperature control method of this application and are not intended to limit this application. Furthermore, the technical features involved in the various embodiments of this application described below can be combined with each other as long as they do not conflict with each other.

[0029] The battery temperature control method provided in this application can be applied to battery management systems, such as... Figure 2 As shown, the battery management system may include a controller, sensors, and an auxiliary heating device. The controller, as the executing entity, can be a temperature control module within the battery management system or an embedded controller, or other device with data processing capabilities. The controller can acquire temperature data of the controlled battery object collected by the sensors, including battery temperature, ambient air temperature, insulation layer temperature, and environmental temperature. The auxiliary heating device is used to heat the controlled battery object according to the target auxiliary heating power output by the controller. Through the coordinated operation of this battery management system, the battery temperature control method provided in this embodiment can be realized.

[0030] For example, consider a liquid metal battery as the controlled battery object. Figure 3 As shown, the method includes steps S1 to S4.

[0031] Step S1: Based on the state parameters of the controlled battery object, construct a high-fidelity multiphysics finite element simulation model of the controlled battery object. This high-fidelity multiphysics finite element simulation model is used to determine the state data of the controlled battery object.

[0032] Specifically, firstly, the state parameters of the controlled battery object are determined, including the battery's geometric parameters, electrochemical parameters, and thermodynamic parameters. Then, based on these parameters, a high-fidelity multiphysics finite element simulation model of the controlled battery object is established to simulate real battery behavior with high fidelity.

[0033] For example, based on the state parameters of the controlled battery object, a one-dimensional electrochemical-three-dimensional thermal coupling simulation model of the controlled battery object is established using a general multiphysics simulation software based on the finite element method (such as the COMSOL Multiphysics platform). It can be considered that the high-fidelity multiphysics finite element simulation model is a collaborative work of the one-dimensional electrochemical simulation model and the three-dimensional thermal coupling simulation model.

[0034] The one-dimensional electrochemical simulation model includes a charge conservation sub-model, a mass transfer conservation sub-model, and an electrode kinetics sub-model. The one-dimensional electrochemical simulation model is used to calculate the heat generation power of the battery and transfer this heat generation power to the three-dimensional thermal model.

[0035] The three-dimensional thermal coupling simulation model includes the energy conservation equation and the heat conduction equation (including convective heat transfer and radiative heat transfer). The three-dimensional thermal coupling simulation model is used to calculate the temperature field evolution of the battery.

[0036] In this way, the heat generation power is calculated by a one-dimensional electrochemical simulation model and transmitted to a three-dimensional thermally coupled simulation model. The three-dimensional thermal model calculates the temperature field evolution. The two work together to achieve high-fidelity simulation of battery heat generation, heat transfer and temperature distribution, providing accurate state data of the controlled battery object for subsequent control. This state data includes at least the temperature of the controlled battery object, air temperature, insulation layer temperature and battery heat generation power.

[0037] In this embodiment, a high-fidelity multiphysics finite element simulation model is constructed as a virtual observation platform. The heat generation power is calculated by a one-dimensional electrochemical model and the temperature field evolution is calculated by a three-dimensional thermal model. This can accurately simulate the internal thermal state of a real battery and provide a reliable data foundation for subsequent temperature control strategy research.

[0038] Step S2: Based on the heat transfer process of the controlled battery object, construct an equivalent thermal resistance network model, and determine the thermal parameters of the equivalent thermal resistance network model based on the state data.

[0039] Specifically, based on the heat transfer process of the controlled battery object (i.e., the battery pack) using the high-fidelity multiphysics model in step 1, the heat transfer data of the controlled battery object (including convective heat transfer, radiative heat transfer, battery heat generation power, and temperature distribution, etc.) can be considered as an equivalent thermal resistance network with multiple temperature nodes, which can then be used for model predictive control. This method simulates the equivalent circuit idea, simplifying the battery system into a thermal resistance network through lumped parameterization, and using a low-order model to predict the temperature evolution of the controlled battery object.

[0040] For example, Figure 4 This is a schematic diagram of a simulated equivalent circuit provided in an embodiment of this application; the equivalent circuit is used to simulate the dynamic temperature response of the controlled battery object. Wherein, T b T a T c It refers to the temperature of the battery, air, and insulation layer, analogous to node voltage; T amb It refers to ambient temperature, analogous to a reference voltage source; R ba R ac R bc and R out Represents the thermal resistance between the battery, air, insulation layer, and external environment; analog resistance; C b C a C c These represent the equivalent heat capacity of the battery, air, and insulation layer, respectively, analogous to a capacitor.

[0041] The multiple temperature nodes can include: battery nodes, air nodes surrounding the battery nodes, and insulation layer nodes. Based on the heat transfer relationships between different temperature nodes, the following set of differential equations (i.e., the equivalent thermal resistance network model) is established.

[0042] (1); (2); (3); The physical meanings of each parameter in equations (1) to (3) are as follows: C b C a C c These represent the equivalent heat capacities of the battery, air, and insulation layer, respectively. T b T a T c It refers to the temperature of the battery, air, and insulation layer. R ba and R ac The distribution is the equivalent thermal resistance for convective heat transfer between the battery and air, and between the insulation layer and air; R bc It is the equivalent thermal resistance of radiative heat transfer between the battery and the insulation layer.

[0043] R out This represents the insulation layer's exposure to the external environment (temperature T). amb The equivalent thermal resistance for heat dissipation; Q b The heat generation power of the battery is provided in real time by the one-dimensional electrochemical simulation model in step 1; P heat It is the auxiliary heating power, which is the manipulated variable of the control system and is solved by subsequent model prediction controller optimization.

[0044] In equations (1) to (3) above, the equivalent heat capacity C and equivalent thermal resistance R are collectively referred to as thermal parameters. Their initial values ​​(i.e., the initial thermal parameters θ(0) of the equivalent thermal resistance network model) can be obtained by measuring using the quasi-static method. For example, when the battery pack is in thermal equilibrium or under slow temperature rise conditions, a known constant auxiliary heating power is applied, and the steady-state temperature difference and dynamic temperature rise rate of each node are measured simultaneously. Then, the equivalent heat capacity of each node and the equivalent thermal resistance between nodes are deduced from the thermal equilibrium equation and used as the initial values ​​for the online operation of the model, i.e., θ(0).

[0045] The thermal parameters may include three equivalent heat capacities: C b C a C c ; and four equivalent thermal resistances: R ba R ac R bc R out .

[0046] In addition, considering that factors such as battery aging and environmental changes may cause thermal parameters to drift slowly, in order to maintain model accuracy, the recursive least squares (RLS) method with a forgetting factor is used to determine and update thermal parameters during online operation, so that the model can accurately predict the temperature changes of the controlled battery object.

[0047] The online RLS identification method is as follows: (4); Where Y(k) represents the actual temperature of each node. θ(k) represents the thermal parameters to be identified, such as the equivalent heat capacity C and equivalent thermal resistance R in the equivalent thermal resistance network model. (k) represents the data matrix, which can be composed of known quantities such as historical temperature difference, heat generation power, and auxiliary heat power. E(k): model residuals, which are expected to approach zero.

[0048] The RLS recursive update formula is: (5); (6); (7); Where K(k) represents the Kalman gain vector. I represents the identity matrix. P(k) represents the covariance matrix. λ represents the forgetting factor (0 < λ ≤ 1), used to reduce the weight of old data and enhance the algorithm's ability to track parameter changes. P(k) and λ are determined based on practical engineering experience.

[0049] In this way, through the above-mentioned RLS online identification, the equivalent thermal resistance network model can update the thermal parameters in a timely manner, enabling the equivalent thermal resistance network model to accurately follow the actual temperature changes of the controlled battery object, and providing reliable model parameters for the subsequent step S3.

[0050] An equivalent thermal resistance network model was established for the working characteristics of liquid metal batteries, simplifying the complex battery system into three temperature nodes: battery, air, and insulation layer. This reduced computational complexity and improved computational efficiency. At the same time, a recursive least squares method was introduced to identify and update thermal parameters online, enabling the prediction model to adapt to the drift of thermodynamic parameters caused by material aging and environmental changes during long-term battery operation, thus avoiding the failure of subsequent prediction models.

[0051] Step S3: Based on the equivalent thermal resistance network model and thermal parameters, a simplified control model is constructed. This simplified control model is used to predict the temperature change of the controlled battery object under different auxiliary heating powers.

[0052] Among them, the model prediction algorithm requires the mathematical model to be transformed into a standard discrete state-space matrix structure. Therefore, based on the continuous-time differential equations corresponding to the equivalent thermal resistance network model, the augmented matrix method can be used to transform them into a standard discrete-time state-space matrix form, forming a simplified mathematical model suitable for model predictive control.

[0053] For example, the equivalent thermal resistance network model output in step S2 is a system of differential equations, in which each thermal parameter (heat capacity Cb, Ca, Cc and thermal resistance Rba, Rac, Rbc) has been updated online by RLS. To adapt to the digital model predictive controller, the equivalent thermal resistance network model needs to be discretized.

[0054] In step S2, equations (1) to (3) are discretized using the sampling time with forward Euler discretization, and the relationship between heat capacity and mass, and specific heat capacity C is used. i =m i c i (i=b,a,c, where m) i For the node equivalent quality, c i Using the equivalent specific heat capacity, we obtain the recursive formulas for the temperatures of each node (i.e., the simplified state-space model guided by control): (8); (9); (10); in, m b c b ，m a c a ，m c c c This is the equivalent heat capacity C in step 2. b C a C c The two are completely equivalent. The heat generation power Q of the controlled battery... b (k) consists of two parts: Ohmic heat and reversible entropy heat. Its algebraic expression is: (11); Among them, I b R is the battery operating current. ohm (k) is the ohmic internal resistance at the current temperature. Vocv / T This is the temperature coefficient of the open-circuit voltage. Q is... b (k) can be used as a known perturbation term in the temperature recursion equation (8), the value of which is provided by the one-dimensional electrochemical model in step 1, or directly calculated online by equation (11).

[0055] Equations (8) to (10) above represent the discretization of the thermal equations for the battery, air, and temperature nodes, respectively. The above discrete recursive equations are then organized into the following standard state-space model: (12); The matrices and vectors in equation (12) are defined as follows: The state vector x(k), containing the temperatures of the battery, air, and insulation layer nodes, is the controlled state variable of the system. ; The control input vector u(k) represents the auxiliary heating power, which is a variable that the controller can manipulate. ; The disturbance variable v(k) includes the battery's heat generation power and the external ambient temperature, where the battery's heat generation power Q b (k) is provided by the one-dimensional electrochemical model from step 1, with ambient temperature T. amb Obtained by sensor measurements.

[0056] ; The state matrix A is used to describe the dynamic heat conduction inside the battery pack system (which is directly composed of thermal parameters): ; Input matrix B u Represents the control variable P heat Effects on the system: ; Wherein, the perturbation matrix B v Q can be used to represent the heat generation power of a controlled battery object. b With external ambient temperature T amb Impact: ; Since the RLS updates the thermal parameters in step S2, the state matrix A and input matrix B in step S3... u Perturbation matrix B v The simplified control model is then adaptively adjusted to ensure that it always accurately matches the actual thermal dynamics of the battery pack, providing an accurate prediction basis for the model predictive control in step S4.

[0057] Step S4: Based on the state data and simplified control model, perform rolling optimization to determine the target auxiliary heating power, which is used to control the temperature of the controlled battery object.

[0058] The goal of step S4 is to solve, at each sampling time, a sequence of auxiliary heat power that minimizes the multi-objective cost function (solved using a quadratic programming (QP) algorithm) based on the current system state and the online updated prediction model (the simplified control model in step S3), and apply it as the target auxiliary heat power to the controlled battery object, thereby achieving precise closed-loop control of the battery temperature.

[0059] For example, the specific process of scrolling optimization in step S4 is as follows: First, read the current state data from step S1, such as the node temperature T. b (k), T a (k), T c (k) and battery heat generation power Q b (k); and obtain the ambient temperature T from the sensor. amb (k). This constitutes the current state x(k) = [T] b (k),T a (k),T c (k)]^T and perturbation v(k)=[Q b (k),T amb (k)]^T.

[0060] Next, the latest thermal parameter C is obtained from the RLS identification algorithm in step S2. b C a Cc R ba R ac R bc R out Step S3 refreshes the state matrix A and input matrix B based on the latest thermal parameters. u and perturbation matrix B v .

[0061] To eliminate steady-state errors caused by prediction model mismatch and external disturbances, a first-order low-pass filter disturbance estimator is introduced: (13); Among them, T mean T is the current measured temperature of the battery. ref The temperature predicted by the model can be considered as the set target temperature. α, γ These are the filter coefficients. γ Let α represent the forgetting rate and α represent the step size. The target temperature T inside the model predictive control (MPC) is predicted using the estimated equivalent perturbation correction model. target The corrected target temperature T is obtained. target_x : (14); When the system has a persistent deviation, the corrected T target It will automatically adjust to force the temperature to track the original set value. In this way, by introducing a first-order low-pass filter perturbation estimator to track the prediction model error and correct the target temperature to eliminate steady-state error, precise control of battery temperature management can be achieved.

[0062] Subsequently, a future battery temperature prediction trajectory is constructed. For example, let the prediction time domain length (e.g., N) be... p The control time domain length is N. c (N) c ≤N p The control sequence to be optimized is P. heat ={P heat (k),P heat (k+1),...,P heat (k+N c -1)}. Iterative prediction is performed using the discrete state-space equation (12) from step S3, and the battery temperature sequence T is extracted from the predicted state. pred (i)=[1,0,0]x(k+i|k),i=1,2,...,N p .

[0063] Then, a multi-objective cost function J is constructed, and the temperature prediction value T is used. pred (i) and control sequence P heatSubstitute the following objective cost function: (15); Among them, W Q The weight representing the temperature difference indicates the system's ability to track the target temperature; W R W represents the weight of energy loss; S This represents the weight for power smoothing, preventing abrupt changes in auxiliary heating power; T pred T represents the predicted temperature (i.e., the predicted temperature value). target P represents the target temperature. pred This represents the predicted power. Multi-objective control, including precise temperature control and energy loss management, is achieved through constraints imposed by the cost function.

[0064] By solving the objective cost function using quadratic programming (QP), the control sequence P that minimizes J is obtained. heat *. That is, the target auxiliary heating power.

[0065] Among them, only P heat The first element of the output is applied to the auxiliary heating device of the controlled battery object, thereby controlling the temperature changes of the battery, air, and insulation layer. The entire rolling optimization process is repeated at the next sampling time k+1.

[0066] In this way, by constructing a multi-objective cost function that takes into account temperature tracking accuracy, energy loss, and auxiliary heating power smoothness, and by using rolling optimization and a quadratic programming problem to solve for the optimal auxiliary heating power, the system can find the optimal auxiliary heating power sequence that comprehensively optimizes temperature tracking, energy consumption, and power smoothness in each control cycle based on the latest battery state data and the latest prediction model, and execute it immediately. This drives the battery temperature to accurately follow the target temperature, thereby achieving precise, energy-saving, and smooth temperature control of the liquid metal battery pack.

[0067] For example, Figure 5 This is a schematic diagram of the battery heat generation power curve during the charge-discharge process of a liquid metal battery with constant auxiliary heating power at 0.2C (rate), as provided in the embodiments of this application. Figure 5 As shown, during the charging phase (0~20000s), the battery exhibits continuous heat generation, with the heat generation power (W) gradually increasing and stabilizing over time (s), and the battery temperature rising slowly. At the moment of switching between charging and discharging, the battery heat generation power drops sharply from a positive value to a negative value, indicating strong heat absorption, which causes a significant temperature change. After entering the discharging phase (20000~40000s), the heat generation power gradually recovers from a negative value to a positive value, and the battery temperature shows a trend of first decreasing and then increasing.

[0068] For example, Figure 6This diagram illustrates the battery temperature results of the liquid metal battery provided in this application embodiment using a PID battery temperature control method under 0.2C charge / discharge conditions, compared to the battery temperature control method in this application embodiment. Figure 6 As shown, both control methods can adjust the battery temperature (°C) to the target value, but the adaptive model predictive control (AMPC) method (i.e., the battery temperature control method of this application embodiment) has better temperature control accuracy and stability than the PID method. Throughout the charging and discharging process, the curve under AMPC control shows that the battery temperature (°C) remains stable near the target temperature over time (s), with almost no overshoot or fluctuations; while the curve under PID control shows that under thermal disturbances such as charging and discharging switching, the battery temperature (°C) exhibits significant temperature drops and multiple oscillations over time (s).

[0069] For example, Figure 7 This diagram illustrates the auxiliary heating power results of the PID battery temperature control method used in the liquid metal battery provided in this application embodiment under 0.2C charge / discharge conditions, compared to the battery temperature control method in this application embodiment. Figure 7 As shown, under strong thermal disturbance conditions such as charging and discharging switching, the AMPC control method can adjust the auxiliary heating power (W) more quickly and suppress temperature fluctuations as time (s) changes. In contrast, the PID control requires multiple power adjustments to eliminate temperature deviations and has a slow response in auxiliary heating power (W).

[0070] This application provides a battery temperature control device, which can be understood as a collection of functional modules that implement the above-described battery temperature control method. This battery temperature control device can be integrated into an electronic device to perform the various functions or steps executed by the electronic device in the above method embodiments.

[0071] This application provides an electronic device that may include a memory and one or more processors. The memory and processors are coupled. The memory stores computer program code, including computer instructions. When the processor executes the computer instructions, the electronic device can perform various functions or steps described in the method embodiments above.

[0072] This application also provides a computer-readable storage medium including computer instructions that, when executed on an electronic device, cause the electronic device to perform various functions or steps performed by the electronic device in the above method embodiments.

[0073] This application also provides a computer program product that, when run on a computer, causes the computer to perform various functions or steps performed by the electronic device in the above method embodiments.

[0074] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0075] The electronic devices, computer-readable storage media, or computer program products provided in this application are all used to perform the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

Claims

1. A battery temperature control method, characterized in that, include: Based on the state parameters of the controlled battery object, a high-fidelity multiphysics finite element simulation model of the controlled battery object is constructed. The high-fidelity multiphysics finite element simulation model is used to determine the state data of the controlled battery object. The state parameters include at least geometric parameters, electrochemical parameters, and thermodynamic parameters; the state data includes at least the temperature of the controlled battery object, air temperature, insulation layer temperature, and battery heat generation power; the controlled battery object includes a liquid metal battery; Based on the heat transfer process of the controlled battery object, an equivalent thermal resistance network model of the controlled battery object is constructed, and the thermal parameters of the equivalent thermal resistance network model are determined based on the state data. Based on the equivalent thermal resistance network model and the thermal parameters, a simplified control model is constructed. The simplified control model is used to predict the temperature change of the controlled battery object under different auxiliary heating powers. Based on the state data and the simplified control model, rolling optimization is performed to determine the target auxiliary heating power, which is used to control the temperature of the controlled battery object.

2. The method according to claim 1, characterized in that, The high-fidelity multiphysics finite element simulation model includes a one-dimensional electrochemical simulation model and a three-dimensional thermally coupled simulation model. The one-dimensional electrochemical simulation model is used to determine the heat generation power of the battery, and the three-dimensional thermally coupled simulation model is used to determine the evolution of the battery temperature field, which includes the evolution of the temperature of the controlled battery object, the air temperature, and the temperature of the insulation layer.

3. The method according to claim 1, characterized in that, The equivalent thermal resistance network model adopts a lumped parameterization method to equate the heat transfer process of the controlled battery object to a thermal resistance network including battery nodes, air nodes, and insulation layer nodes; the thermal parameters include the equivalent heat capacity of each node and the equivalent thermal resistance between each node.

4. The method according to claim 3, characterized in that, Determining the thermal parameters of the equivalent thermal resistance network model based on the state data includes: Based on the state data, the thermal parameters of the equivalent thermal resistance network model are determined by recursive least squares (RLS) with a forgetting factor. The initial thermal parameters of the equivalent thermal resistance network model are determined by the thermal quasi-static method.

5. The method according to claim 1, characterized in that, The simplified control model is a state-space equation obtained by discretizing the equivalent thermal resistance network model; the simplified control model includes a state matrix, an input matrix, and a disturbance matrix; the state matrix, input matrix, and disturbance matrix are determined based on the thermal parameters.

6. The method according to claim 1, characterized in that, The rolling optimization is achieved through a model prediction controller; the rolling optimization process includes: Within the prediction time domain, the model prediction controller predicts the temperature prediction trajectory corresponding to different candidate auxiliary heat power sequences based on the state data and the simplified control model; the prediction time domain is the time length predicted by the model prediction controller. The candidate auxiliary heat power sequence and its corresponding temperature prediction trajectory are evaluated according to the multi-objective cost function. The candidate auxiliary heat power sequence that minimizes the value of the multi-objective cost function is determined as the target auxiliary heat power. The multi-objective cost function includes a temperature tracking error term, an energy loss term, and a power smoothing term.

7. The method according to claim 6, characterized in that, Prior to the scrolling optimization, the following is also included: Using a disturbance estimator, the target temperature inside the model predictor is corrected based on the deviation between the actual temperature and the target temperature of the controlled battery object. The corrected target temperature is then used to evaluate the temperature prediction trajectory through the multi-objective cost function.

8. The method according to claim 7, characterized in that, The disturbance estimator is a first-order low-pass filter disturbance estimator; the step of using the disturbance estimator to correct the target temperature inside the model predictive controller based on the deviation between the actual temperature and the target temperature of the controlled battery object includes: Using a first-order low-pass filter perturbation estimator, the equivalent perturbation estimate at the current sampling time is recursively calculated based on the deviation between the actual temperature and the target temperature of the controlled battery object at the current sampling time and the equivalent perturbation estimate at the previous sampling time; the corrected target temperature is obtained by subtracting the equivalent perturbation estimate at the current sampling time from the target temperature.

9. An electronic device, characterized in that, include: A memory, and one or more processors; the memory and the processors are coupled to each other; wherein the memory stores computer program code, the computer program code including computer instructions, which, when executed by the processor, cause the electronic device to perform the battery temperature control method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, Includes computer instructions that, when executed on an electronic device, cause the electronic device to perform the battery temperature control method as described in any one of claims 1-8.

Images