Lithium battery cooling method and system

Abstract

The invention relates to the technical field of lithium batteries, and particularly provides a lithium battery cooling method and system, and the method comprises the steps: constructing an energy conservation equation of an immersed fluoride solution model; obtaining a heat source power equation; obtaining a convective heat transfer equation of the lithium battery; constructing a flow equation of the fluorination liquid; predicting the flowing property of the fluorination liquid according to the Reynolds number; constructing a temperature boundary condition and a constraint condition for lithium battery temperature prediction; obtaining a physical constraint term according to the energy conservation equation, the heat source power equation, the convective heat transfer equation and the flow equation, and obtaining a loss function according to the physical constraint term; constructing a neural network model for predicting the temperature of the lithium battery according to the loss function, and predicting the temperature of the lithium battery according to the constructed neural network model; and adjusting the basic parameters of the fluorination liquid according to the predicted temperature of the lithium battery, and cooling the lithium battery. According to the invention, based on a mode of combining the data driving model and the physical constraint, the accuracy and response speed of temperature prediction of the fluorination liquid are obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium batteries, and in particular to a lithium battery cooling method and system. Background Art

[0002] Among the many liquid cooling media, fluorinated liquid as a cooling medium in liquid cooling technology has many unique advantages such as excellent thermal conductivity, good electrical insulation, fire resistance and safety, and a wide operating temperature range, which makes it have important application potential in high-performance battery packs, electric vehicles, data centers and other high-power applications.

[0003] Although fluorinated liquid has good thermal conductivity, existing fluorinated liquid cooling systems may face the problem of insufficient precision when dynamically adjusting the coolant flow or pressure. Due to factors such as the viscosity and fluidity of the fluorinated liquid, the cooling system may have difficulty in fine-tuning the flow or temperature. In some complex application scenarios, the temperature and flow rate of the coolant need to be precisely controlled to cope with different load changes. However, due to the system design and the characteristics of the fluorinated liquid itself, the adjustment accuracy is not high enough, and precise temperature control requirements cannot be achieved. Summary of the Invention

[0004] Based on this, in order to improve the adjustment accuracy of the fluorinated liquid cooling medium, the present invention provides a lithium battery cooling method and system, the specific technical solutions are as follows:

[0005] A lithium battery cooling method comprises the following steps:

[0006] Obtain basic information about the fluorinated liquid and the heat source power generated by the lithium battery, and construct an energy conservation equation for the immersion fluorinated liquid model based on the basic information and the heat source power;

[0007] Obtain basic information of the lithium battery and the temperature of the fluoride liquid, and obtain a heat source power equation based on the basic information of the lithium battery and the temperature of the fluoride liquid;

[0008] Obtain the convection heat transfer equation of lithium batteries based on the basic information of fluorinated liquid and lithium batteries;

[0009] Construct the flow equation of fluorinated liquid based on the basic information of fluorinated liquid;

[0010] Obtain the Reynolds number of the fluorinated liquid and predict the flow properties of the fluorinated liquid based on the Reynolds number;

[0011] Construct temperature boundary conditions and constraints for lithium battery temperature prediction;

[0012] Obtain physical constraints based on the energy conservation equation, heat source power equation, convection heat transfer equation, and flow equation, and obtain a loss function based on the physical constraints;

[0013] Constructing a neural network model for predicting the temperature of the lithium battery according to the loss function, and predicting the temperature of the lithium battery according to the constructed neural network model;

[0014] The basic parameters of the fluorine liquid are adjusted according to the predicted lithium battery temperature to cool the lithium battery.

[0015] The lithium battery cooling method can predict the lithium battery temperature in real time by combining the energy conservation equation, convection heat transfer equation, heat source power equation, flow equation, etc. with a neural network model. Based on the predicted lithium battery temperature, basic parameters of the fluorine liquid such as temperature and flow rate can be accurately adjusted according to changes in the load. Based on the combination of data-driven models and physical constraints, the accuracy and response speed of fluorine liquid temperature prediction are significantly improved.

[0016] Preferably, the basic information of the fluorinated liquid includes the density ρ of the fluorinated liquid f1 Specific heat capacity C of fluorinated liquid p,f1 、Temperature of fluorinated liquid f1 , the velocity μ of the fluorinated liquid and the thermal conductivity k of the fluorinated liquid f1 , the energy conservation equation is expressed as:

[0017] Among them, Q heat Indicates the heat source power.

[0018] Preferably, the basic information of the lithium battery includes the current I of the lithium battery, the open circuit voltage U of the lithium battery OCV , the actual voltage U of the lithium battery and the entropy coefficient of the lithium battery The heat source power equation is expressed as

[0019] Preferably, the basic information of the fluorinated liquid also includes the convective heat transfer coefficient h of the fluorinated liquid. f1 , the basic information of lithium batteries also includes the surface area A of the lithium battery f1 And the surface temperature T of the lithium battery battery , the convective heat transfer equation is expressed as q conv =h f1 ·A f1 ·(T f1 -T battery );

[0020] Among them, q conb Indicates the amount of heat exchange per unit time.

[0021] Preferably, the basic information of the fluorinated liquid also includes the pressure p of the fluorinated liquid, the viscosity μ of the fluorinated liquid f1 And the influence factor F of external force on fluorinated liquid, the flow equation is expressed as

[0022] A lithium battery cooling system, used to implement the lithium battery cooling method, comprising:

[0023] An energy conservation equation construction module is used to obtain basic information about the fluorinated liquid and the heat source power generated by the lithium battery, and to construct the energy conservation equation of the immersion fluorinated liquid model based on the basic information and heat source power;

[0024] A heat source power equation construction module is used to obtain basic information of the lithium battery and the temperature of the fluoride liquid, and obtain the heat source power equation based on the basic information of the lithium battery and the temperature of the fluoride liquid;

[0025] A convection heat transfer equation construction module is used to obtain the convection heat transfer equation of the lithium battery based on the basic information of the fluorinated liquid and the basic information of the lithium battery;

[0026] A flow equation construction module is used to construct the flow equation of the fluorinated liquid based on the basic information of the fluorinated liquid;

[0027] A fluorinated liquid property prediction module is used to obtain the Reynolds number of the fluorinated liquid and predict the flow properties of the fluorinated liquid based on the Reynolds number;

[0028] Boundary constraint condition construction module, used to construct temperature boundary conditions and constraints for lithium battery temperature prediction;

[0029] A loss function acquisition module is used to obtain physical constraint terms based on the energy conservation equation, the heat source power equation, the convection heat transfer equation, and the flow equation, and to obtain a loss function based on the physical constraint terms;

[0030] A neural network model building module is used to build a neural network model for predicting the temperature of the lithium battery according to the loss function, and predict the temperature of the lithium battery according to the built neural network model;

[0031] The control module is used to adjust the basic parameters of the fluorine liquid according to the predicted lithium battery temperature to cool the lithium battery.

[0032] Preferably, the energy conservation equation construction module constructs the energy conservation equation formula of the immersion fluorinated liquid model according to the basic information of the fluorinated liquid and the heat source power as follows:

[0033] The basic information of the fluorinated liquid includes the density of the fluorinated liquid ρ f1 Specific heat capacity C of fluorinated liquid p,f1 、Temperature of fluorinated liquid f1 , the velocity μ of the fluorinated liquid and the thermal conductivity k of the fluorinated liquid f1 ,q heat Indicates the heat source power.

[0034] Preferably, the heat source power equation construction module obtains the heat source power equation expression according to the basic information of the lithium battery and the temperature of the fluorinated liquid as follows:

[0035] Among them, the basic information of lithium batteries includes the current I of lithium batteries, the open circuit voltage U of lithium batteries OCV , the actual voltage U of the lithium battery and the entropy coefficient of the lithium battery

[0036] Preferably, the convection heat transfer equation construction module obtains the convection heat transfer equation of the lithium battery according to the basic information of the fluorinated liquid and the basic information of the lithium battery, which is expressed as: conv =h f1 A f1 ·(T f1 -T battery );

[0037] Among them, the basic information of fluorinated liquid also includes the convective heat transfer coefficient h of fluorinated liquid f1 , the basic information of lithium batteries also includes the surface area A of the lithium battery f1 And the surface temperature of the lithium battery T battery ,q conv Indicates the amount of heat exchange per unit time.

[0038] Preferably, the loss function acquisition module acquires the loss function expression according to the physical constraint term as follows:

[0039] in, represents the data error term, T represents the predicted temperature field and the actual temperature field of the neural network model, PhysicsLoss represents the physical constraint term, and λ represents the weight coefficient of the physical constraint term. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but rather the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0041] Figure 1 This is a schematic diagram of the overall process of a lithium battery cooling method according to one embodiment of the present invention;

[0042] Figure 2 This is a schematic diagram of the overall structure of a lithium battery cooling system in one embodiment of the present invention;

[0043] Figure 3 is a schematic structural diagram of a neural network model in one embodiment of the present invention;

[0044] Figure 4The figure is a flow chart of a specific method for supplying fluorinated liquid as a lithium battery coolant on demand in one embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with its embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and do not limit the scope of protection of the present invention.

[0046] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly attached to the other element or there may be an intermediate element. When an element is referred to as being "connected to" another element, it may be directly connected to the other element or there may be an intermediate element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only implementation methods.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0048] The "first" and "second" in the present invention do not represent specific quantities and orders, but are only used to distinguish names.

[0049] Before describing the embodiments of the present invention, a brief introduction to the prior art is given first.

[0050] With the widespread adoption of electric vehicles (EVs), energy storage systems, and various high-energy-density devices, lithium-ion batteries (Li-ion Batteries) play a crucial role in these fields. With their high energy density, long cycle life, and high charging efficiency, Li-ion Batteries have become the most mainstream energy storage technology. However, Li-ion Batteries generate significant amounts of heat during charging and discharging, particularly at high power and current densities, where this heat accumulation is particularly significant. If this heat is not dissipated promptly and effectively, it can lead to abnormally high battery temperatures, severely impacting battery performance and lifespan. Excessively high battery temperatures not only accelerate chemical reactions within the battery, causing capacity degradation, but can also cause structural damage within the battery, impacting cycle stability. More seriously, prolonged overheating can trigger thermal runaway, leading to safety incidents such as explosion and fire, posing significant risks to users and the environment. Therefore, the research and application of battery thermal management technology, particularly in high-energy-density Li-ion Batteries, has become a key technology for ensuring battery safety, extending battery life, and improving battery performance.

[0051] Battery thermal management systems can generally be divided into two categories based on the cooling medium: air cooling and liquid cooling. Air cooling systems typically use air flow to remove heat generated by the battery. While this system is simple in structure and low in cost, its cooling efficiency is significantly limited due to the poor thermal conductivity of air. Therefore, air cooling systems are more suitable for low-power or smaller-scale battery systems, such as in mobile devices and low-power electric vehicles.

[0052] When the power density and energy density of the battery are high, the cooling capacity of the air cooling system often cannot meet the demand. In contrast, liquid cooling systems are widely used in high-power, high-energy-density battery packs, especially in electric vehicles and large-scale energy storage systems. The liquid cooling system removes the heat generated by the battery through a circulating coolant (such as water or water-based coolant). Its heat conduction efficiency is high, which can effectively reduce the battery temperature and ensure that the battery operates within a safe temperature range. The liquid cooling system consists of components such as a coolant pump, pipes, and cooling plates. Through reasonably designed cooling channels, the heat is conducted and dissipated to the external environment. Compared with the air cooling system, the liquid cooling system can provide a more uniform and stable temperature control effect at higher power output, thereby ensuring the safe and efficient operation of the battery pack.

[0053] With the continuous advancement of battery technology, especially in electric vehicles and high-capacity energy storage systems, battery power and energy densities continue to increase. This directly leads to a significant increase in the heat generated by batteries during operation, making traditional thermal management solutions face greater challenges. Liquid cooling systems, due to their superior heat conduction capabilities, demonstrate significant advantages in addressing these challenges and have become the mainstream technology in lithium battery thermal management today.

[0054] Among the many liquid cooling media, fluorinated liquid as a cooling medium in liquid cooling technology has many unique advantages such as excellent thermal conductivity, good electrical insulation, fire resistance and safety, and a wide operating temperature range, which makes it have important application potential in high-performance battery packs, electric vehicles, data centers and other high-power applications. However, how to promptly remove the heat generated by the battery through efficient coolant flow to prevent the battery from overheating and ensure its long-term stable operation remains an important technical challenge in the design and optimization of battery thermal management systems. This requires researchers to not only further improve the thermal efficiency of the cooling system, but also consider how to dynamically adjust the coolant flow, flow rate or pressure to adapt to different loads and working environments, thereby achieving more precise and efficient temperature control.

[0055] The fluorinated liquid cooling system has the following disadvantages:

[0056] 1. Slow response speed.

[0057] Fluorinated fluids have low fluidity and typically high viscosity, and their fluidity may be affected, especially at higher temperatures. Therefore, fluorinated fluid cooling systems may respond slowly to sudden high heat loads. When the rate at which batteries or equipment generate heat increases significantly, the system may not be able to quickly adjust the coolant flow or pressure, causing the temperature to rise rapidly and affecting the stability of the system. This may become a bottleneck in certain applications that require a fast cooling response, such as fast-charging electric vehicles.

[0058] 2. Insufficient adjustment accuracy

[0059] Although fluorinated liquids have good thermal conductivity, existing fluorinated liquid cooling systems may face the problem of insufficient precision when dynamically adjusting the coolant flow or pressure. Due to factors such as the viscosity and fluidity of the fluorinated liquid, the cooling system may have difficulty in fine-tuning the flow or temperature. In some complex application scenarios, the temperature and flow rate of the coolant need to be precisely controlled to cope with different load changes. However, due to the system design and the characteristics of the fluorinated liquid itself, the adjustment accuracy may not be high enough to achieve precise temperature control requirements.

[0060] 3. Difficulty in adjusting coolant flow and flow rate

[0061] Due to the high viscosity of fluorinated liquid, the control system requires a large amount of energy to drive its flow, which makes it difficult to adjust the flow rate and flow velocity. In some applications with large demand changes, such as fast charging of electric vehicles or frequent load fluctuations in energy storage systems, the flow rate and flow velocity of the coolant must be adjusted quickly to adapt to the changing heat load. However, the poor fluidity of fluorinated liquid leads to a slow response to the regulation of flow rate and flow velocity, and the adjustment range is limited. This makes it difficult to achieve the goal of supplying coolant on demand and on time, especially when the cooling demand changes drastically, the system's responsiveness may be insufficient.

[0062] In order to improve the adjustment accuracy of the fluorinated liquid cooling medium, such as Figure 1 As shown, an embodiment of the present invention provides a lithium battery cooling method comprising the following steps:

[0063] S1, obtaining basic information of the fluorine liquid and the heat source power generated by the lithium battery, and constructing an energy conservation equation of the immersion fluorine liquid model based on the basic information of the fluorine liquid and the heat source power.

[0064] Here, the basic information of fluorinated liquid includes but is not limited to the density of fluorinated liquid (kg / m 3 ), specific heat capacity (J / kg·K), temperature (K), velocity (m / s), and thermal conductivity (W / m·K).

[0065] Immersion cooling involves placing electronic devices (in this case, lithium batteries or lithium battery packs) directly into an insulating liquid, such as fluorinated fluid, silicone oil, or synthetic oil. The flow of the insulating liquid removes heat generated by the components. The insulating liquid then exchanges heat with an external cooling source through a heat exchanger, powered by a pump to form a circulating cooling system. An immersion cooling system includes a liquid-cooled cabinet, piping components, a water pump, a heat exchanger, a cooling tower, and various types of sensors.

[0066] The immersion fluorinated liquid model here can be understood as a cooling system in which the lithium battery is completely immersed in the cooling medium fluorinated liquid with high specific heat capacity and convection heat transfer coefficient.

[0067] S2, obtaining basic information of the lithium battery and the temperature of the fluorine liquid, and obtaining a heat source power equation based on the basic information of the lithium battery and the temperature of the fluorine liquid.

[0068] The basic information of the lithium battery includes but is not limited to the lithium battery current, open circuit voltage, actual voltage (V), and entropy coefficient. The heat source power equation can be understood as obtaining the heat source power of the lithium battery based on the basic information of the lithium battery.

[0069] S3. Obtaining a convection heat transfer equation of the lithium battery based on basic information of the fluorinated liquid and basic information of the lithium battery.

[0070] Here, the basic information of fluorinated liquid also includes the convection heat transfer coefficient of fluorinated liquid, and the basic information of lithium battery also includes the surface area of lithium battery (m 2 ) and surface temperature (K). The convection heat transfer equation can be understood as the heat transfer rate between the fluoride solution and the lithium battery per unit time, based on the surface area and surface temperature of the lithium battery, and the convection heat transfer coefficient and temperature of the fluoride solution.

[0071] S4, constructing a flow equation of the fluorinated liquid based on basic information of the fluorinated liquid, wherein the basic information of the fluorinated liquid further includes velocity (m / s), pressure (Pa), viscosity (Pa·s) of the fluorinated liquid, and an influencing factor of an external force (such as gravity) on the fluorinated liquid.

[0072] S5, obtaining the Reynolds number of the fluorinated liquid, and predicting the flow properties of the fluorinated liquid according to the Reynolds number.

[0073] Since the cooling medium fluorinated liquid is a flowing liquid, the Reynolds number (Re) can be used to predict the nature of the fluid flow, that is, whether it is laminar flow, turbulent flow or transitional flow. Here, the density of the fluorinated liquid (kg / m 3 ), flow velocity (m / s), characteristic length (m) and dynamic viscosity (Pa·s) to obtain the Reynolds number, and the flow properties of fluorinated liquids were predicted based on the Reynolds number.

[0074] S6, construct temperature boundary conditions and constraints for lithium battery temperature prediction.

[0075] To achieve a temperature prediction for lithium batteries that complies with physical laws, it is necessary to add appropriate temperature boundary conditions and constraints. Constraints include, but are not limited to, the relationship between the fluorine solution flow rate and the pressure drop within the battery cavity, as well as the temperature range and specific heat capacity of the phase change material fluorine solution.

[0076] S7, obtaining physical constraint terms according to the energy conservation equation, the heat source power equation, the convection heat transfer equation, and the flow equation, and obtaining a loss function according to the physical constraint terms.

[0077] The physical constraints include but are not limited to physical errors of temperature changes calculated according to the thermal energy conservation equation, the heat source power equation, the convection heat transfer equation, and the flow equation.

[0078] S8, constructing a neural network model for predicting the temperature of the lithium battery according to the loss function, and predicting the temperature of the lithium battery according to the constructed neural network model.

[0079] The constructed neural network model includes an input layer, a hidden layer, and an output layer. Before performing step S9, the neural network model is first trained. Since the training of the neural network model belongs to conventional technical means in this field, it will not be described in detail here.

[0080] S9, adjusting the basic parameters of the fluorine liquid according to the predicted temperature of the lithium battery to cool the lithium battery.

[0081] Here, the adjusted basic parameters of the fluorine liquid include but are not limited to the flow rate, pressure, and temperature of the fluorine liquid. By predicting the temperature of the lithium battery to adjust the basic parameters of the fluorine liquid, it is possible to quickly respond to load changes, reduce the response time of the system, and enable instant adjustment of the coolant flow or flow rate to cope with sudden high heat loads, thereby improving the system response speed. At the same time, predicting the lithium battery temperature through a neural network model and adjusting the basic parameters of the fluorine liquid based on the predicted lithium battery temperature is also conducive to improving the regulation accuracy of the fluorine liquid.

[0082] That is to say, the lithium battery cooling method can predict the lithium battery temperature in real time by combining the energy conservation equation, convection heat transfer equation, heat source power equation, flow equation, etc. with a neural network model; based on the predicted lithium battery temperature, the basic parameters of the fluorine liquid such as temperature and flow rate can be accurately adjusted according to the change of load; based on the combination of data-driven model and physical constraints, the accuracy and response speed of fluorine liquid temperature prediction are significantly improved.

[0083] In one embodiment, the basic information of the fluorinated liquid includes the density ρ of the fluorinated liquid. f1 Specific heat capacity C of fluorinated liquid p,f1 、Temperature of fluorinated liquid f1 , the velocity μ of the fluorinated liquid and the thermal conductivity k of the fluorinated liquid f1 , the energy conservation equation is expressed as: Among them, Q heat Indicates the heat source power.

[0084] Preferably, the basic information of the lithium battery includes the current I of the lithium battery, the open circuit voltage U of the lithium battery OCV , the actual voltage U of the lithium battery and the entropy coefficient of the lithium battery The heat source power equation is expressed as

[0085] The basic information of fluorinated liquid also includes the convective heat transfer coefficient h of fluorinated liquid f1 , the basic information of lithium batteries also includes the surface area A of the lithium battery f1 And the surface temperature of the lithium battery T battery , the convective heat transfer equation is expressed as q conv =h f1 A f1 ·(T f1 -T battery ), where q conv Indicates the amount of heat exchange per unit time.

[0086] Reynolds number Where ρ is the density of the fluid (kg / m 3 ), μ is the flow velocity of the fluid (m / s), D is the characteristic length of the fluid (usually the diameter of the pipe), the unit is m, and τ is the dynamic viscosity of the fluid (Pa·s).

[0087] The basic information of fluorinated liquid also includes the pressure p of fluorinated liquid and the viscosity μ of fluorinated liquid. f1 And the external force on the fluorinated liquid factor F, μ represents the flow rate of the fluorinated liquid, the flow equation is expressed as

[0088] Constructing temperature boundary conditions and constraints for lithium battery temperature prediction includes the following steps:

[0089] First, construct the temperature boundary condition, which can be expressed as T min <T(t)<T max ; T(t) is the predicted lithium battery temperature, T min is the predicted lower limit of lithium battery temperature, T max is the predicted upper temperature limit of the lithium battery.

[0090] Then, the relationship between flow rate and pressure drop in the battery cavity is constructed, and the formula is: Where, Δp is the pressure loss (Pa), L is the pipe length (m), D is the pipe diameter (m), ρ is the density of the fluorinated liquid (kg / m 3 ), μ is the flow rate of fluorinated liquid (m / s), and f is the friction factor.

[0091] Finally, determine the temperature range and specific heat capacity of the phase change material, the formula of which includes T liquid ≤T(t)≤T boiling as well as Among them, T liquid is the lowest temperature of liquid fluoride liquid, T boiling is the boiling point of the fluorinated liquid; C p is the specific heat capacity of the fluorinated liquid (J / kg·K), Q is the absorbed heat (J), m is the mass of the fluorinated liquid (kg), and ΔT is the temperature difference between the fluorinated liquid and the lithium battery (K).

[0092] By constructing the energy conservation equation, convection heat transfer equation, heat source power equation, flow equation, and temperature boundary conditions and constraints, combined with a trained neural network model, the temperature of the lithium battery can be predicted; on this basis, according to the predicted temperature of the lithium battery, the basic parameters of the fluorine liquid such as flow rate, temperature, pressure, etc. are adjusted accordingly, which can improve the response speed of the fluorine liquid cooling system and meet the requirement of timely supply of the fluorine liquid cooling system.

[0093] In one embodiment, Figure 4As shown, the specific method for supplying lithium battery coolant fluoride liquid on demand includes the following steps:

[0094] Step 1: Place a high-precision T-type thermocouple temperature sensor within the lithium-ion battery pack or other cooling target area, ensuring that the sensor fits snugly against the battery's outer surface or the cooling target. This allows the temperature sensor to accurately monitor the battery's operating temperature in real time and generate battery surface temperature data. Due to its high temperature response speed and measurement accuracy, T-type thermocouples are ideal for tracking rapid temperature changes within lithium-ion batteries or cooling targets.

[0095] Step 2: Real-time temperature data acquired by the temperature sensor is transmitted to the data acquisition module, which consists of a signal conditioning module, an analog-to-digital converter (ADC), and a communication interface (such as a CAN bus). The signal conditioning module amplifies and filters the analog signal output by the sensor to ensure signal stability and accuracy. The ADC module converts the conditioned analog signal into a digital signal for subsequent data processing and transmission. Finally, the digital signal is sent to the system's control unit via the CAN bus.

[0096] Step 3: In this embodiment, the CAN bus is used as the communication interface, connecting multiple sensors, control units, and actuators to the same data bus. The CAN bus's advantage lies in its ability to support simultaneous communication and synchronous operation of multiple devices, significantly reducing the complexity and cost of system wiring by sharing the data bus. This bus efficiently transmits temperature data and passes control commands between different devices. Through the CAN bus, the system can effectively coordinate the temperature sensors with the PLC control system and coolant pump, ensuring efficient system operation.

[0097] Step 4: In the PLC control system, the system can dynamically adjust the cooling strategy through the preset initial temperature of the lithium battery, the target cooling temperature and the real-time temperature data, combined with the temperature control algorithm (such as PID control, fuzzy control or fuzzy PID control algorithm, etc.). Specifically, after the control system receives the real-time temperature data from the data acquisition module, it judges the current thermal state of the lithium battery based on the set temperature threshold. When the temperature of the lithium battery exceeds the preset threshold, the control system will start the cooling system to reduce the temperature of the lithium battery in time to prevent overheating from damaging the battery performance. This process involves the real-time operation of advanced control algorithms, and the flow rate of the coolant, cooling efficiency and other parameters are adjusted through PID control or fuzzy PID control to maintain the lithium battery operating within a safe temperature range.

[0098] Step 5: When the PLC control system issues a cooling request, the coolant pump receives the command and starts, precisely regulating the coolant (fluorinated liquid) flow rate through electronically controlled valves to deliver the coolant to the lithium-ion battery modules or target cooling area. The coolant pump, working in conjunction with electronically controlled valves, ensures that coolant is supplied on demand and effectively cools the lithium-ion battery. The system flexibly adjusts the coolant flow rate and cooling volume based on changes in lithium-ion battery temperature and cooling requirements to achieve precise temperature control.

[0099] Through this method, the system can achieve precise control of lithium battery temperature, improving the safety, stability, and efficiency of battery use. While ensuring lithium battery temperature control, this solution optimizes the cooling process, avoids energy waste, and extends the battery life.

[0100] In one embodiment, Figure 2 As shown, the present invention also provides a lithium battery cooling system for implementing the lithium battery cooling method, which includes an energy conservation equation building module, a heat source power equation building module, a convection heat transfer equation building module, a flow equation building module, a fluorinated liquid property prediction module, a boundary constraint condition building module, a loss function acquisition module and a control module.

[0101] The energy conservation equation construction module is used to obtain basic information of the fluorinated liquid and the heat source power generated by the lithium battery, and to construct the energy conservation equation of the immersion fluorinated liquid model based on the basic information of the fluorinated liquid and the heat source power.

[0102] Preferably, the energy conservation equation construction module constructs the energy conservation equation formula of the immersion fluorinated liquid model according to the basic information of the fluorinated liquid and the heat source power as follows: The basic information of the fluorinated liquid includes the density of the fluorinated liquid ρ f1 Specific heat capacity C of fluorinated liquid p,f1 、Temperature of fluorinated liquid f1 , the velocity μ of the fluorinated liquid and the thermal conductivity k of the fluorinated liquid f1 , Q heat Indicates the heat source power.

[0103] The heat source power equation construction module is used to obtain basic information of the lithium battery and the temperature of the fluoride liquid, and obtain the heat source power equation based on the basic information of the lithium battery and the temperature of the fluoride liquid.

[0104] Preferably, the heat source power equation construction module obtains the heat source power equation expression according to the basic information of the lithium battery and the temperature of the fluorinated liquid as follows: Among them, the basic information of lithium batteries includes the current I of lithium batteries, the open circuit voltage U of lithium batteries OCV , the actual voltage U of the lithium battery and the entropy coefficient of the lithium battery

[0105] The convection heat transfer equation construction module is used to obtain the convection heat transfer equation of the lithium battery based on the basic information of the fluorinated liquid and the basic information of the lithium battery.

[0106] Preferably, the convection heat transfer equation construction module obtains the convection heat transfer equation of the lithium battery according to the basic information of the fluorinated liquid and the basic information of the lithium battery, which is expressed as: conv =h f1· A f1 ·(T f1 -T battery ); wherein, the basic information of the fluorinated liquid also includes the convective heat transfer coefficient h of the fluorinated liquid f1 , the basic information of lithium batteries also includes the surface area A of the lithium battery f1 And the surface temperature of the lithium battery T battery ,q conv Indicates the amount of heat exchange per unit time.

[0107] The flow equation construction module is used to construct the flow equation of fluorinated liquid based on the basic information of fluorinated liquid.

[0108] Preferably, the flow equation is: Where, μ is the velocity of the fluorinated liquid (m / s), p is the pressure (Pa), and μ f1 is the viscosity of the fluorinated liquid (Pa·s), and F is the factor affecting the fluid by external forces (such as gravity).

[0109] The fluorinated liquid property prediction module is used to obtain the Reynolds number of the fluorinated liquid and predict the flow properties of the fluorinated liquid based on the Reynolds number.

[0110] Reynolds number Where ρ is the density of the fluid (kg / m 3 ), μ is the flow velocity of the fluid (m / s), D is the characteristic length of the fluid (usually the diameter of the pipe), the unit is m, and τ is the dynamic viscosity of the fluid (Pa·s).

[0111] The boundary constraint condition construction module is used to construct temperature boundary conditions and constraints for lithium battery temperature prediction.

[0112] The loss function acquisition module is used to obtain physical constraint terms based on the energy conservation equation, the heat source power equation, the convection heat transfer equation, and the flow equation, and to obtain the loss function based on the physical constraint terms.

[0113] Preferably, the loss function acquisition module acquires the loss function expression according to the physical constraint term as follows:

[0114] in, represents the data error term, T represents the predicted temperature field and the actual temperature field of the neural network model, PhysicsLoss represents the physical constraint term, and λ represents the weight coefficient of the physical constraint term. The initial value is generally 0.5-1.0.

[0115] When λ increases, the model follows the physical laws more strictly, but the data fitting accuracy may be sacrificed; when λ decreases, the model focuses on matching the observed data and may deviate from the physical laws.

[0116] Here, by simultaneously optimizing the data fitting error (data error term) and the physical law residual (physical constraint term) in the loss function, the neural network model can both match the observed data and satisfy the physical laws described by the partial differential equations.

[0117] in, T i They represent the predicted temperature value and the true temperature value of the i-th data point respectively, and N represents the number of training samples, that is, the total number of data points.

[0118] Assume that the temperature field satisfies the equation Where α is the thermal diffusion coefficient, Q is the heat source term (heat source power), and the physical constraint term can be defined as the L2 norm of the equation residual: Q j Indicates that the external heat source is at the jth time and space point (x j , t j ) strength, It represents the partial derivative of the temperature field predicted by the neural network model with respect to time t, which can be calculated by automatic differentiation. represents the Laplace operator, which can be understood as the spatial second-order derivative of the temperature field, reflecting the spatial distribution characteristics of heat diffusion. N represents the number of physical constraint points, which are usually sparsely distributed observation data points.

[0119] The neural network model building module is used to build a neural network model for predicting the temperature of the lithium battery according to the loss function, and predict the temperature of the lithium battery according to the built neural network model.

[0120] like Figure 3 As shown, the neural network model includes an input layer, a hidden layer, and an output layer.

[0121] For the input layer, the input features include the current battery temperature, battery current, coolant flow rate, ambient temperature, and other system parameters that affect temperature (such as battery load and cooling system operating status). Assuming that the input layer has n features, the mathematical expression of the input layer of the neural network is as follows:

[0122] X=(x1,x2,x3…x n )

[0123] As for the hidden layer, the hidden layer is the core layer of the entire neural network, which includes the weight matrix W i , bias term (bias) and activation function f. The weight is usually expressed as W i =(W1, W2, ..., W n ).

[0124] It is an m×n matrix, where m is the number of neurons in the hidden layer and n is the number of input features.

[0125] Set the bias term of the neural network model. The bias term (bias) is an m×1 column vector that represents the offset of each hidden layer neuron.

[0126] Set the output layer, which calculates the output through the activation function f: H i =f(bias+∑(P×W i ).

[0127] Among them, the activation function f is represented by the ReLU function, and its formula is: f(x)=max(0,x).

[0128] By constructing this neural network model, the battery pack temperature can be predicted, thereby improving the response speed of the cooling system and meeting the requirement of timely supply of the cooling system.

[0129] The control module is used to adjust the basic parameters of the fluorine liquid according to the predicted lithium battery temperature to cool the lithium battery.

[0130] The basic parameters of the fluorine liquid include but are not limited to the flow rate, pressure, and temperature of the fluorine liquid. By adjusting the basic parameters of the fluorine liquid, the operation of the liquid cooling pump is controlled to achieve cooling of the lithium battery.

[0131] In this embodiment, the lithium battery cooling system can predict the lithium battery temperature in real time by combining the energy conservation equation, the convection heat transfer equation, the heat source power equation, the flow equation, etc. with a neural network model. Based on the predicted lithium battery temperature, the basic parameters of the fluorinated liquid, such as temperature and flow rate, can be accurately adjusted according to changes in the load. Based on the combination of data-driven models and physical constraints, the accuracy and response speed of the fluorinated liquid temperature prediction are significantly improved.

[0132] As a preferred technical solution, the loss function acquisition module obtains the loss function expression according to the physical constraint term:

[0133] Among them, PhysicsLoss k is the kth physical constraint loss, λ k (t) is the weight that is dynamically adjusted with the training stage t,

[0134] λ0 represents the initial weight coefficient, which can be understood as the initial weight of the physical constraint loss. It controls the strength of the physical constraint in the early stage of training. If λ0 is large, the model will pay more attention to the optimization of the physical residual in the early stage. If λ0 is small, it may tend to data fitting too early. It can be determined through experimental parameter adjustment.

[0135] β represents the decay rate parameter, which is used to control the speed at which the weight decays over time. The larger β is, the faster the weight λ is. k The faster (t) decays, the faster the strength of the physical constraint decreases; the smaller β is, the faster λ k (t)The slower the decay, the more persistent the effects of the physical constraints.

[0136] The gradient norm of the physical loss can be understood as the gradient norm of the loss function of the kth physical constraint term with respect to the network parameters, reflecting the sensitivity of the current parameter update to the physical residual. A large gradient norm indicates that the physical constraints are not fully met and a high weight should be maintained. A small gradient norm indicates that the physical constraints are well met and the weight can be reduced to focus on data fitting.

[0137] The integral term, which can be understood as the cumulative norm of the physical loss gradient from the start of training to the current time t, is used to dynamically quantify the difficulty of optimizing the physical constraints throughout the training process. The larger the cumulative gradient, the more significant the weight decay, achieving an adaptive balance.

[0138] Introducing time-dependent weight coefficient λ into the neural network model k (t), has the following advantages:

[0139] 1. Initial strengthening of physical constraints: In the early stage of training, the gradient of the physical residual is large, the integral accumulation grows rapidly, and λ k (t) decays slowly, and the model prioritizes satisfying physical laws. For example, in a heat conduction problem, the initial forced temperature field satisfies the residual condition of the heat conduction equation.

[0140] 2. Later balanced data fitting; as the training progresses, the physical residual gradient decreases, the integral accumulation growth rate slows down, and λ k (t) decays rapidly, and the model gradually focuses on optimizing data fitting terms (such as experimental measurement errors).

[0141] 3. Adaptive stability: By integrating the gradient norm, we avoid the tediousness of manually adjusting λ and alleviate the gradient explosion or vanishing problem.

[0142] formula The weights are adaptively adjusted according to the constraint residual gradient, physical constraints are strengthened in the early stage, and data fitting is balanced in the later stage. It automatically balances data-driven and physical law constraints during training and is suitable for engineering scenarios with sparse experimental data or complex noise.

[0143] In the formula In the paper, independent weights are designed for different physical equations (such as heat conduction + fluid mechanics) to solve the problem of uneven constraint strength in multi-scale problems and realize the coupling of multiple physical fields. It establishes a dynamic trade-off between data-driven and physical knowledge-driven, and can adapt to the needs of models at different stages for data fitting and physical rationality.

[0144] In summary, the lithium battery cooling method and system of the present invention can achieve at least the following beneficial effects:

[0145] 1. The present invention ensures that coolant can be supplied in a timely manner on demand and adapt to different loads and temperature changes, which is crucial for improving cooling efficiency and battery safety.

[0146] 2. Combining physical models (such as the energy conservation equation, convection heat transfer equation, heat source power equation, and flow equation) with the prediction method of neural networks can achieve accurate prediction and control of lithium battery temperature, which is highly innovative and practical.

[0147] 3. Through real-time temperature monitoring and feedback control, the lithium battery temperature is ensured to remain within a safe range, while the regulation of the fluorine liquid flow is optimized to improve the response speed and stability of the system.

[0148] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0149] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A lithium battery cooling method, characterized in that: The lithium battery cooling method comprises the following steps: Obtain basic information about the fluorinated liquid and the heat source power generated by the lithium battery, and construct an energy conservation equation for the immersion fluorinated liquid model based on the basic information and the heat source power; Obtain basic information of the lithium battery and the temperature of the fluoride liquid, and obtain a heat source power equation based on the basic information of the lithium battery and the temperature of the fluoride liquid; Obtain the convection heat transfer equation of lithium batteries based on the basic information of fluorinated liquid and lithium batteries; Construct the flow equation of fluorinated liquid based on the basic information of fluorinated liquid; Obtain the Reynolds number of the fluorinated liquid and predict the flow properties of the fluorinated liquid based on the Reynolds number; Construct temperature boundary conditions and constraints for lithium battery temperature prediction; Obtain physical constraints based on the energy conservation equation, heat source power equation, convection heat transfer equation, and flow equation, and obtain a loss function based on the physical constraints; Constructing a neural network model for predicting the temperature of the lithium battery according to the loss function, and predicting the temperature of the lithium battery according to the constructed neural network model; The basic parameters of the fluorine liquid are adjusted according to the predicted lithium battery temperature to cool the lithium battery.

2. A lithium battery cooling method according to claim 1, characterized in that: Basic information of fluorinated liquid includes density ρ of fluorinated liquid f1 Specific heat capacity C of fluorinated liquid p,f1 、Temperature of fluorinated liquid f1 , the velocity μ of the fluorinated liquid and the thermal conductivity k of the fluorinated liquid f1 , the energy conservation equation is expressed as: Among them, Q heat Indicates the heat source power.

3. A lithium battery cooling method according to claim 2, characterized in that: The basic information of lithium batteries includes the current I of lithium batteries and the open circuit voltage U of lithium batteries. OCV , the actual voltage U of the lithium battery and the entropy coefficient of the lithium battery The heat source power equation is expressed as 4. A lithium battery cooling method according to claim 3, characterized in that: The basic information of fluorinated liquid also includes the convective heat transfer coefficient h of fluorinated liquid f1 , the basic information of lithium batteries also includes the surface area A of the lithium battery f1 And the surface temperature T of the lithium battery battery , the convective heat transfer equation is expressed as q conv =h f1 ·A f1 ·(T f1 -T battery ); Among them, q conv Indicates the amount of heat exchange per unit time.

5. A lithium battery cooling method according to claim 4, characterized in that: The basic information of fluorinated liquid also includes the pressure p of fluorinated liquid and the viscosity μ of fluorinated liquid. f1 And the influence factor F of external force on fluorinated liquid, the flow equation is expressed as 6. A lithium battery cooling system, used to implement the lithium battery cooling method according to any one of claims 1 to 5, characterized in that: The lithium battery cooling system includes: An energy conservation equation construction module is used to obtain basic information about the fluorinated liquid and the heat source power generated by the lithium battery, and to construct the energy conservation equation of the immersion fluorinated liquid model based on the basic information and heat source power; A heat source power equation construction module is used to obtain basic information of the lithium battery and the temperature of the fluoride liquid, and obtain the heat source power equation based on the basic information of the lithium battery and the temperature of the fluoride liquid; A convection heat transfer equation construction module is used to obtain the convection heat transfer equation of the lithium battery based on the basic information of the fluorinated liquid and the basic information of the lithium battery; A flow equation construction module is used to construct the flow equation of the fluorinated liquid based on the basic information of the fluorinated liquid; A fluorinated liquid property prediction module is used to obtain the Reynolds number of the fluorinated liquid and predict the flow properties of the fluorinated liquid based on the Reynolds number; Boundary constraint condition construction module, used to construct temperature boundary conditions and constraints for lithium battery temperature prediction; A loss function acquisition module is used to obtain physical constraint terms based on the energy conservation equation, the heat source power equation, the convection heat transfer equation, and the flow equation, and to obtain a loss function based on the physical constraint terms; A neural network model building module is used to build a neural network model for predicting the temperature of the lithium battery according to the loss function, and predict the temperature of the lithium battery according to the built neural network model; The control module is used to adjust the basic parameters of the fluorine liquid according to the predicted lithium battery temperature to cool the lithium battery.

7. A lithium battery cooling system according to claim 6, characterized in that: The energy conservation equation construction module constructs the energy conservation equation of the immersion fluorinated liquid model based on the basic information of the fluorinated liquid and the heat source power. The formula is expressed as follows: The basic information of the fluorinated liquid includes the density of the fluorinated liquid ρ f1 Specific heat capacity C of fluorinated liquid p,f1 、Temperature of fluorinated liquid f1 , the velocity μ of the fluorinated liquid and the thermal conductivity k of the fluorinated liquid f1 , Q heat Indicates the heat source power.

8. A lithium battery cooling system according to claim 7, characterized in that: The heat source power equation construction module obtains the heat source power equation expression based on the basic information of the lithium battery and the temperature of the fluorinated liquid as follows: Among them, the basic information of lithium batteries includes the current I of lithium batteries, the open circuit voltage U of lithium batteries OCV , the actual voltage U of the lithium battery and the entropy coefficient of the lithium battery 9. A lithium battery cooling system according to claim 8, characterized in that: The convection heat transfer equation construction module obtains the convection heat transfer equation of lithium battery according to the basic information of fluorinated liquid and lithium battery: conv =h f1 .A f1 ·(T f1 -T battery ); Among them, the basic information of fluorinated liquid also includes the convective heat transfer coefficient h of fluorinated liquid f1 , the basic information of lithium batteries also includes the surface area A of the lithium battery f1 And the surface temperature of the lithium battery T battery ,q conv Indicates the amount of heat exchange per unit time.

10. A lithium battery cooling system according to claim 9, characterized in that: The loss function acquisition module obtains the loss function expression based on the physical constraint terms: in, represents the data error term, T represents the predicted temperature field and the actual temperature field of the neural network model, PhysicsLoss represents the physical constraint term, and λ represents the weight coefficient of the physical constraint term.

Images