Railway vehicle power battery immersion liquid cooling thermal management device and system

Abstract

The application relates to the technical field of battery heat management, and discloses a rail passenger car power battery immersion type liquid cooling heat management device and system.The system comprises a parameter identification module, an online ohmic internal resistance and polarization internal resistance separated based on real-time working data and surface temperature, and a total heating power calculated; a temperature estimation module, a highest temperature estimation value of a battery cell inside reconstructed in combination with the total heating power and internal equivalent thermal physical parameters; a flow prediction module, a temperature deviation converted into expected heat dissipation power based on a cooling time constant, and a target volume flow prediction value calculated through feedforward; and a cooperative optimization control module, a circulating driving power of a pump adaptively adjusted according to dynamic dynamic viscosity changes caused by fluid temperature to achieve a target flow.The application realizes accurate quantification of a battery heating source and dynamic tracking of a core temperature, improves the response speed of the system to high heat flux density, and significantly reduces the pump power consumption of the cooling system in a full temperature range.

Description

Technical Field

[0001] This invention belongs to the field of battery thermal management technology, and particularly relates to an immersion liquid-cooled thermal management device and system for power batteries of rail passenger vehicles. Background Technology

[0002] During actual operation of rail vehicles, the power battery system frequently undergoes high-rate charging and discharging conditions, generating enormous instantaneous heat flux density, leading to a rapid rise in battery temperature. This rapid temperature rise not only significantly increases the risk of thermal runaway but also accelerates the aging of electrode materials and electrolyte decomposition, seriously threatening the battery's safety performance and lifespan. Traditional air-cooling solutions have extremely limited heat dissipation capacity due to the low thermal conductivity and small specific heat capacity of air, making it unable to dissipate the heat generated by high-power charging and discharging in a timely manner. While indirect cold-plate liquid cooling solutions improve heat exchange efficiency through the liquid medium, they are still limited by the interfacial contact thermal resistance between the battery and the cold plate and the length of the heat transfer path, making it difficult to cope with the unique challenge of instantaneous high heat flux density in rail vehicles. At the same time, temperature sensors can only be placed on the surface of the battery casing and cannot directly obtain the highest temperature point of the core area of ​​the battery cell, forcing thermal management strategies to adopt overly conservative threshold settings, resulting in a bulky cooling system, increased energy consumption, and an inability to accurately match actual heat load requirements.

[0003] Immersion liquid cooling technology achieves direct, efficient heat exchange by completely submerging the battery in insulating coolant, significantly improving heat dissipation. However, existing control strategies generally rely on surface temperature feedback mechanisms, exhibiting significant dynamic response lag and prone to temperature overshoot—for example, during sudden changes in charging / discharging power, surface temperature changes lag behind the internal core temperature, resulting in insufficient or excessive cooling. Some methods attempt to estimate heat dissipation based on simplified models, but fail to achieve online dynamic separation of the battery's ohmic and polarization resistances, and neglect the influence of the reversible entropy heat component in the electrochemical reaction, leading to systematic biases in the calculation of heat generation power. In the flow control stage, there is a lack of feedforward prediction capability that combines the desired cooling time constant with the dynamic characteristics of convective heat transfer, making it impossible to quickly generate target flow commands based on real-time thermal states. Furthermore, the dynamic viscosity of the coolant changes significantly with temperature fluctuations, but pump drive systems mostly use constant power output or simple proportional-integral-derivative control, failing to adaptively adjust the drive power based on real-time viscosity, resulting in additional energy loss and reducing the overall vehicle energy efficiency. Summary of the Invention

[0004] The purpose of this invention is to provide an immersion liquid-cooled thermal management device and system for power batteries of rail passenger vehicles, in order to solve the above-mentioned problems.

[0005] This invention is implemented as follows: a submerged liquid-cooled thermal management system for a power battery in a rail passenger vehicle, comprising: a parameter identification module configured to acquire real-time operating data of the power battery and surface temperature collected by a temperature sensor array, online separation of the ohmic internal resistance and polarization internal resistance of the power battery, and calculation of the real-time total heat generation power of the power battery; a temperature estimation module configured to reconstruct the estimated maximum internal temperature of the battery cell by combining the total heat generation power with the internal equivalent thermophysical parameters of the power battery; a flow prediction module configured to convert the temperature deviation into the desired heat dissipation power based on the cooling time constant, and to calculate the target volumetric flow rate prediction value by combining the fluid convection heat transfer temperature rise model and dynamically extracted convection heat transfer thermal resistance; and a collaborative optimization control module configured to adaptively adjust the output to achieve the cycle drive power required to reach the target volumetric flow rate prediction value based on the dynamic dynamic viscosity change caused by the fluid temperature.

[0006] A further technical solution involves calculating the real-time total heat generation power of the power battery in the parameter identification module. This process includes: determining the irreversible Joule heat generation power of the power battery based on the real-time charging and discharging operating current, the separated ohmic internal resistance and polarization internal resistance; determining the reversible thermodynamic entropy heat generation power of the power battery based on the direction and magnitude of the real-time charging and discharging operating current, the open-circuit voltage temperature coefficient of the power battery, and the estimated maximum internal temperature of the cell in the previous cycle; and physically superimposing the irreversible Joule heat generation power and the reversible thermodynamic entropy heat generation power to obtain the real-time total heat generation power of the power battery.

[0007] A further technical solution, in the temperature estimation module, includes the following process for reconstructing the estimated maximum internal temperature of the battery cell: based on the total heat generation power, and the internal heat conduction dissipation power determined by the estimated maximum internal temperature of the battery cell in the previous cycle, the surface weighted average temperature, and the equivalent internal thermal resistance of the power battery, a dynamic heat flow balance relationship is constructed inside the power battery; combined with the internal equivalent heat capacity parameters of the power battery, the dynamic rate of change of the maximum internal temperature of the battery cell is calculated according to the dynamic heat flow balance relationship, and the dynamic rate of change is iteratively accumulated over a preset calculation period to update and output the estimated maximum internal temperature of the battery cell in the current cycle.

[0008] A further technical solution, in the flow prediction module, the process of feedforward calculation of the target volume flow prediction value includes: generating a corresponding temperature regulation compensation power based on the temperature deviation between the allowable safety limit temperature inside the power battery and the estimated maximum temperature inside the cell in the current cycle, combined with the cooling time constant; and superimposing the temperature regulation compensation power with the total heat generation power to obtain the desired heat dissipation power.

[0009] A further technical solution, in the flow prediction module, the process of feedforward calculation of the target volumetric flow rate prediction value also includes: using the target volumetric flow rate and the average fluid temperature of the previous cycle to interpolate and look up a table to obtain the equivalent convective heat transfer resistance of the current cycle, and combining the expected heat dissipation power, the surface weighted average temperature, and the initial fluid temperature at the inlet, to calculate the predicted outlet temperature of the coolant when it flows out of the power battery according to the convective heat transfer physical boundary constraint equation; based on the expected heat dissipation power, and the enthalpy change relationship between the coolant density, constant pressure specific heat capacity, and the temperature difference between the inlet and outlet fluids, to solve and output the target volumetric flow rate prediction value.

[0010] A further technical solution, in the collaborative optimization module, the process of adaptively adjusting the cyclic drive power includes: obtaining the reference viscosity constant and flow activation energy of the insulating coolant, and calculating the temperature-corrected dynamic viscosity of the fluid based on the physical characteristic that the fluid viscosity decreases nonlinearly with temperature, combined with the real-time average fluid temperature; calculating the net hydrodynamic work required to overcome the viscous frictional resistance of the fluid in the cooling channel based on the obtained fluid dynamic viscosity, the structural resistance parameters of the cooling channel, and the predicted value of the target volumetric flow rate, and outputting the cyclic drive power in combination with the mechanical-electrical energy conversion efficiency of the cyclic drive mechanism.

[0011] A submerged liquid-cooled thermal management device for a rail passenger vehicle's power battery includes a memory, a processor, and a communication bus. The memory and the processor are connected via the communication bus. The memory stores a computer program, which, when executed by the processor, enables the functional configuration of each module in the aforementioned submerged liquid-cooled thermal management system for the rail passenger vehicle's power battery.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0013] 1. Precise source heat sensing and real-time core temperature tracking completely avoid control lag. This invention eliminates systematic deviations in heat generation power at the source by online decoupling and separating ohmic and polarization internal resistances and incorporating the reversible entropy heat of electrochemical reactions. Combined with a first-order lumped parameter thermal model to dynamically reconstruct the highest internal temperature of the battery cell that cannot be directly measured, it overcomes the dynamic response lag of traditional methods that rely solely on the surface temperature feedback of the casing, effectively avoiding the risk of temperature overshoot under intensive charging and discharging conditions.

[0014] 2. Feedforward flow dynamic prediction and viscosity adaptive adjustment significantly reduce vehicle energy consumption. The flow control abandons the traditional passive feedback regulation and predicts the target volume flow rate based on the cooling time constant. At the same time, the collaborative optimization control module corrects the nonlinear decay of coolant dynamic viscosity caused by temperature fluctuations in real time, and adaptively matches the pump's cycle drive power, avoiding energy loss caused by constant power operation, and achieving refined dual optimization of heat dissipation performance and system energy efficiency. Attached Figure Description

[0015] Figure 1 This invention provides an overall physical and communication architecture diagram of an immersion liquid-cooled thermal management system for a rail passenger vehicle's power battery.

[0016] Figure 2 The main flowchart of the control method provided by the present invention;

[0017] Figure 3 The core data flow and causal closed-loop diagram of the algorithm provided in this invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the traditional thermal management process of power batteries for rail passenger vehicles, there are systematic biases in the estimation of heat generation power under intensive charge and discharge conditions. These biases stem from the failure to separate ohmic internal resistance and polarization internal resistance online, and the absence of reversible entropy heat components in the calculation model. Furthermore, the perception of the highest internal temperature of the battery cell is lagging, as temperature sensors can only be placed on the cell surface and cannot directly acquire the internal thermal state. The adjustment response of cooling flow is slow, lacking a feedforward prediction mechanism based on the expected cooling time constant and the dynamic characteristics of convective heat transfer. In addition, the dynamic viscosity of the coolant changes significantly with temperature, and existing pump drive systems do not adaptively adjust according to real-time viscosity, resulting in a mismatch between the cycle drive power and actual demand. These problems collectively constrain the thermal management system in terms of safety, response speed, and energy efficiency, with increased risk of temperature overshoot, insufficient decision-making basis for thermal management strategies, and forced increases in system size and energy consumption.

[0020] For example, in scenarios involving frequent start-stop operations of rail vehicles in high-temperature environments, the battery pack undergoes high-rate charge-discharge cycles, resulting in a surge in instantaneous heat flux density. Traditional immersion liquid cooling systems rely on feedback from surface temperature sensors for regulation, but the surface temperature is lower than the actual maximum temperature inside the battery cell, causing a delay in the activation of thermal management strategies. Cooling flow regulation is based solely on historical temperature deviations for feedback control, failing to anticipate changes in heat load, leading to a mismatch between heat dissipation capacity and heat generation power. Simultaneously, the rising coolant temperature causes a decrease in dynamic viscosity, while the pump drive system continues to operate at a fixed power, resulting in a circulation flow exceeding actual needs and additional energy consumption. In this scenario, the data collected by the temperature sensor array cannot reflect the internal heat distribution of the battery cell, the lag in the flow control module's response causes the battery temperature to continue to rise, and the energy consumption of the pump drive unit is not optimized according to changes in fluid properties, leading to a reduction in overall system efficiency.

[0021] If the above problems are not solved, the battery may cause local thermal runaway due to temperature overshoot under continuous high heat flux density conditions, which will shorten the cycle life of the battery cells. The thermal management system will need to compensate for insufficient heat dissipation by increasing the volume of the heat sink or increasing the flow rate of the coolant, which will further increase the weight and energy consumption of the vehicle. At the same time, a conservative thermal management strategy will limit the power output capability of the battery, affecting the operational reliability and energy utilization efficiency of the rail passenger vehicle.

[0022] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0023] The immersion liquid-cooled thermal management system for power batteries in rail passenger vehicles is a thermal management solution specifically designed for rail passenger vehicles. Its core is to directly immerse the power battery in insulating coolant, achieving efficient heat exchange through direct contact between the coolant and the battery surface, and supplemented by intelligent control strategies to maintain the power battery within the optimal operating temperature range, ensuring its safe and stable operation.

[0024] like Figure 1 , Figure 2 and Figure 3 As shown, an embodiment of the present invention provides a submerged liquid-cooled thermal management system for a rail passenger vehicle's power battery, comprising:

[0025] The parameter identification module is configured to acquire real-time operating data of the power battery and surface temperature collected by a temperature sensor array, separate the ohmic internal resistance and polarization internal resistance of the power battery online, and calculate the total real-time heat generation power of the power battery. The power battery refers to the battery pack that provides driving power to the rail vehicle, typically composed of multiple cells connected in series and parallel. It generates a large amount of heat during charging and discharging. Real-time operating data refers to various electrical parameters, such as charging and discharging current and voltage, collected in real time by sensors during the operation of the power battery. This data is the basis for evaluating the battery status and heat generation. The temperature sensor array is a collection of temperature sensors arranged at multiple locations on the surface of the power battery, used to accurately monitor the temperature distribution on the battery casing surface, providing important input information for the thermal management system. Ohmic internal resistance refers to the instantaneous impedance generated by the resistive characteristics of the material itself when current flows through the power battery; its heat generation is proportional to the square of the current. Polarization internal resistance refers to the additional impedance generated by the limitations of electrochemical reaction kinetics (such as charge transfer and mass diffusion) during the charging and discharging process; its heat generation is related to the current and polarization voltage. Total heat generation power refers to all the heat generated by the power battery during operation, including ohmic internal resistance, polarization internal resistance, and reversible entropy heat. It is the total heat load that the thermal management system needs to remove.

[0026] The temperature estimation module is configured to reconstruct the estimated maximum internal temperature of the battery cell by combining the total heat generation power with the internal equivalent thermophysical parameters of the power battery. Internal equivalent thermophysical parameters refer to a series of physical quantities describing the internal thermal characteristics of the power battery, such as equivalent heat capacity and equivalent thermal conductivity. These parameters are crucial for accurately estimating the internal temperature of the battery cell. The estimated maximum internal temperature of the battery cell refers to the temperature value of the area with the highest temperature inside a single battery cell, calculated through the model. This value is usually higher than the surface temperature and is a key indicator for judging the thermal safety of the battery.

[0027] The flow prediction module is configured to convert temperature deviation into expected heat dissipation power based on the cooling time constant, and combine this with a fluid convection heat transfer temperature rise model and dynamically extracted convective heat transfer thermal resistance to calculate the predicted target volumetric flow rate. The cooling time constant is a characteristic parameter describing the time required for the battery temperature to drop from its current value to the target value under cooling, reflecting the system's dynamic response speed. Temperature deviation refers to the difference between the currently estimated maximum internal temperature of the battery and the preset safe or optimal operating temperature limit. Expected heat dissipation power refers to the amount of heat the thermal management system needs to remove from the battery to bring it to the target temperature within the preset cooling time constant. The fluid convection heat transfer temperature rise model describes the physical model of the coolant carrying away heat and increasing its temperature as it flows over the battery surface. Convection heat transfer thermal resistance refers to the resistance encountered during convective heat transfer between the coolant and the battery surface, reflecting the heat transfer efficiency; its value changes dynamically with fluid state and heat transfer conditions. The target volumetric flow rate prediction value refers to the volumetric flow rate of the coolant that the thermal management system should obtain through feedforward calculation based on the current thermal state of the battery and the expected heat dissipation requirements.

[0028] The collaborative optimization control module is configured to adaptively adjust the output to achieve the required circulating drive power to reach the predicted target volumetric flow rate based on the dynamic dynamic viscosity changes caused by fluid temperature. The dynamic dynamic viscosity changes caused by fluid temperature refer to the alteration of the dynamic viscosity of the insulating coolant with temperature variations. This change affects the coolant's flow resistance, thus influencing the required circulating drive power. Circulating drive power refers to the energy required to drive the coolant to circulate throughout the liquid cooling system, typically provided by the cooling pump.

[0029] This embodiment proposes an immersion liquid-cooled thermal management system for power batteries in rail passenger vehicles, aiming to address the thermal management challenges of power batteries under intensive charge and discharge conditions. Through the collaborative work of multiple modules, the system achieves accurate sensing of the battery's thermal state, intelligent prediction of cooling requirements, and efficient operation of the cooling system.

[0030] The parameter identification module is the foundation of this system. Its function is to accurately acquire real-time operating data of the power battery and perform refined analysis of the battery's internal heat sources based on surface temperatures collected by a temperature sensor array. For example, this module can continuously monitor the battery's charging and discharging current and voltage, and, combined with surface temperature data, estimate the battery's overall internal resistance by establishing a simplified equivalent circuit model. Based on this, the total heat generation power of the battery under the current operating conditions can be calculated. As one possible implementation, this module can employ an online identification method based on voltage response. By analyzing the battery's transient voltage response under a specific current pulse, its equivalent internal resistance can be estimated, and thus the total heat generation power can be calculated. However, this method may treat ohmic internal resistance and polarization internal resistance as a whole, or may not fully consider reversible entropy heat, leading to some deviation in the heat generation power estimation.

[0031] The temperature estimation module is responsible for reconstructing the estimated maximum internal temperature of the battery cell by combining the total heat generation power output from the parameter identification module with the internal equivalent thermophysical parameters of the power battery. Since the internal temperature of the battery cell is difficult to measure directly, this module establishes a heat conduction model inside the battery and uses the known surface temperature and internal heat generation to estimate the core temperature. Specifically, this module can use a lumped-parameter thermal model to simplify the battery cell into an object with specific heat capacity and thermal resistance. By inputting the total heat generation power and the battery surface temperature, and combining this with the preset internal equivalent heat capacity and equivalent thermal resistance from the core to the surface, the average internal temperature of the battery cell can be calculated. However, this simplified model may not accurately capture the complex temperature distribution inside the battery cell, especially under high-rate charging and discharging or uneven local heating conditions, which may lead to inaccurate estimates of the maximum temperature.

[0032] The flow prediction module aims to calculate the required target volumetric flow rate of the coolant based on the real-time thermal state of the power battery and the desired cooling effect. This module first converts the temperature deviation into the desired heat dissipation power based on the cooling time constant. For example, when the estimated maximum internal temperature of the battery cell exceeds a safety threshold, the system calculates the total heat that needs to be removed from the battery within a certain time, i.e., the desired heat dissipation power, according to a preset cooling rate requirement. Subsequently, the module combines a fluid convection heat transfer temperature rise model and dynamically extracted convection heat transfer resistance to calculate the predicted target volumetric flow rate. As one implementation, this module can utilize a simplified convection heat transfer model, assuming the specific heat capacity and density of the coolant are constant, and calculate the flow rate required to achieve the desired heat dissipation power by measuring the inlet temperature and the estimated outlet temperature of the coolant. However, this method may not fully consider that the convection heat transfer resistance of the coolant changes dynamically with factors such as flow rate, temperature, and battery surface condition as it flows through the battery pack, thus affecting the accuracy of the flow prediction.

[0033] The collaborative optimization control module is responsible for adaptively adjusting the output of the circulation drive power required to achieve the predicted target volumetric flow rate based on the dynamic dynamic viscosity changes caused by fluid temperature. Coolant viscosity is a key factor affecting its flow resistance, and viscosity changes significantly with temperature. This module estimates the current dynamic viscosity of the coolant by monitoring the average temperature of the coolant in real time and consulting pre-stored viscosity-temperature curves or using a simplified linear interpolation model. Subsequently, based on the estimated viscosity, the predicted target volumetric flow rate, and the geometric parameters of the cooling channels, it calculates the pump power required to drive the coolant circulation and adjusts the cooling pump's operating status accordingly. However, this method based on simplified models or lookup tables may not accurately reflect the nonlinear variation of coolant viscosity with temperature, especially over a wide temperature range. This could lead to insufficiently precise pump power adjustment, resulting in additional energy consumption or failure to accurately achieve the target flow rate.

[0034] The following example will provide a more detailed explanation of the above technical solution:

[0035] Suppose that during the operation of a rail passenger vehicle, the power battery pack is undergoing a high-power discharge condition, causing its internal temperature to rise rapidly. At this time, the immersion liquid-cooled thermal management system of this application begins to function. First, the parameter identification module continuously acquires real-time charging and discharging current, voltage, and other operating data of the power battery, and combines this with the surface temperature collected by a temperature sensor array arranged on the battery casing surface. This module processes this data, for example, by analyzing the battery's voltage response under different currents, to preliminarily estimate the battery's equivalent internal resistance, and based on this, calculates the battery's current total heat generation power. For example, if only a simple... The model is used to calculate the heating power, where R is an average internal resistance value. However, this preliminary estimate may have some errors because it does not distinguish between ohmic internal resistance and polarization internal resistance, nor does it fully consider the influence of reversible entropy heat.

[0036] Next, the temperature estimation module receives this preliminary total heat generation power value and, combined with the internal equivalent thermophysical parameters of the power battery, such as the battery's heat capacity and thermal conductivity, begins to reconstruct the estimated maximum internal temperature of the cell. Specifically, this module can use a simplified heat conduction model to model the heat generation and outward conduction process inside the battery. For example, by assuming that heat is generated uniformly inside the battery and conducted to the surface with constant thermal resistance, the average internal temperature of the cell is calculated by combining the surface temperature and the total heat generation power. However, this simplified model may not accurately reflect the actual heat transfer path and dynamic thermal resistance between the cell core and the casing surface, resulting in an inaccurate estimate of the maximum internal temperature of the cell.

[0037] Subsequently, the flow prediction module calculates the current temperature deviation based on the estimated maximum internal temperature of the battery cell output by the temperature estimation module and the preset safety limit temperature of the power battery. To ensure the battery temperature returns to a safe range within a preset cooling time constant, this module converts this temperature deviation into the desired heat dissipation power. For example, if the internal battery temperature is too high, the system calculates how much heat needs to be removed to reduce the temperature to a safe level in a short time. Based on this, the module combines a fluid convection heat transfer temperature rise model and dynamically extracted convection heat transfer thermal resistance to calculate the predicted target volumetric flow rate of the coolant. For example, the module can calculate the required flow rate based on the coolant's inlet temperature, estimated outlet temperature, and constant specific heat capacity and density, using the principle of energy conservation. However, if the dynamic changes in convection heat transfer thermal resistance are not accurately captured, or the coolant outlet temperature prediction is inaccurate, the calculated target flow rate may not fully meet the actual heat dissipation requirements.

[0038] Finally, the collaborative optimization control module receives the predicted target volumetric flow rate. Since the dynamic viscosity of the coolant changes significantly with temperature, thus affecting the pump's drive power requirements, this module monitors the average temperature of the coolant in real time. For example, it estimates the current dynamic viscosity of the coolant by consulting a pre-stored viscosity-temperature table or using a simple linear function. Then, based on this estimated viscosity, the predicted target volumetric flow rate, and the geometry of the cooling channels, the module calculates the required circulation drive power to circulate the coolant and adjusts the cooling pump's speed or voltage accordingly. However, if the viscosity estimation is inaccurate or the pump's efficiency model is inadequate, the pump's output power may not match the actual demand, resulting in energy waste or poor cooling performance. Through the collaborative work of these modules, this system can comprehensively and dynamically perceive and manage the thermal state of the power battery, ensuring it remains within its optimal temperature range under various operating conditions.

[0039] The immersion liquid-cooled thermal management system for rail vehicle power batteries proposed in this embodiment demonstrates significant technological contributions compared to existing technologies. Traditional thermal management solutions, such as air cooling or indirect plate-type liquid cooling, often suffer from limited heat dissipation efficiency and slow response when dealing with the instantaneous high heat flux density generated under high-power charging and discharging conditions of rail vehicle power batteries. This application introduces a parameter identification module to achieve online separation of the ohmic internal resistance and polarization internal resistance of the power battery and calculates a more accurate total heat generation power. Compared to existing methods that estimate heat generation power based on simplified models or fixed parameters, this application can more accurately capture the battery's true heat source, providing a more reliable basis for subsequent thermal management decisions.

[0040] Furthermore, the temperature estimation module of this application can reconstruct the estimated maximum internal temperature of the battery cell by combining the accurate total heat generation power with the internal equivalent thermophysical parameters of the power battery. This overcomes the limitations of existing technologies that rely solely on surface temperature sensors for feedback control and solves the problem of lag in sensing the maximum internal temperature of the battery cell. Through accurate estimation of the core temperature of the battery cell, the thermal management strategy can make decisions based on the actual thermal state inside the battery, avoiding conservative control or temperature overshoot caused by inaccurate information.

[0041] Furthermore, the flow prediction module converts temperature deviation into desired heat dissipation power based on the cooling time constant, and combines this with a fluid convection heat transfer temperature rise model and dynamically extracted convection heat transfer thermal resistance to calculate the predicted target volumetric flow rate. Compared with the surface temperature-based feedback regulation commonly used in existing technologies, this feedforward control strategy significantly improves the system's response speed to instantaneous high heat flux density. By predicting cooling demand and calculating the target flow rate in advance, the system can adjust the coolant circulation more quickly, effectively avoiding the response lag and temperature overshoot problems common in traditional feedback control.

[0042] Finally, the collaborative optimization control module adaptively adjusts the output based on the dynamic dynamic viscosity changes caused by fluid temperature to achieve the required circulating drive power to reach the predicted target volumetric flow rate. Existing pump drive systems often use constant power or simple PID output, failing to fully consider the significant impact of coolant viscosity changes with temperature, leading to additional energy consumption. This application incorporates the dynamic viscosity changes of the coolant into the control logic, achieving refined management of the circulating drive power, thereby optimizing the energy efficiency of the cooling system and reducing the overall vehicle's operating energy consumption. In summary, this embodiment, through the collaborative innovation of the above modules, forms a more intelligent, efficient, responsive, and energy-saving immersion liquid-cooled thermal management system for rail passenger vehicle power batteries, significantly improving the battery system's thermal safety, response speed, and energy efficiency.

[0043] This application further proposes that, in the parameter identification module, the process of calculating the real-time total heat generation power of the power battery includes: determining the irreversible Joule heat generation power of the power battery based on the real-time charge / discharge operating current, the separated ohmic internal resistance and polarization internal resistance; determining the reversible thermodynamic entropy heat generation power of the power battery based on the direction and magnitude of the real-time charge / discharge operating current, the open-circuit voltage temperature coefficient of the power battery, and the estimated maximum internal temperature of the cell in the previous cycle; and physically superimposing the irreversible Joule heat generation power and the reversible thermodynamic entropy heat generation power to obtain the real-time total heat generation power of the power battery; the specific formula is as follows:

[0044]

[0045] in, This represents the real-time total heat dissipation power of the power battery. For real-time charging and discharging operating current, and when the discharge state is set. A positive value indicates that the device is in a charging state. It is a negative value; The ohmic internal resistance separation value identified online; The polarization internal resistance separation value identified online; This is an estimate of the highest internal temperature of the cell in the previous cycle; This represents the rate of change of the open-circuit voltage of the power battery with temperature.

[0046] Total heating power This refers to all the heat generated by the power battery during operation due to internal electrochemical reactions and resistive effects. It forms the basis for the thermal management system's heat dissipation control, and accurate... This is key to achieving precise temperature control. Real-time charge / discharge operating current. This refers to the magnitude and direction of the current flowing through the battery during actual operation. It is typically measured in real-time using a high-precision Hall current sensor or shunt connected in series in the battery circuit, or directly acquired by the current sampling module integrated within the Battery Management System (BMS). The online identification of the ohmic internal resistance separation value... This refers to the resistance component corresponding to the instantaneous voltage drop caused by the pure resistive effect inside the power battery when current flows through it. Its identification methods can include the current interruption method, which involves momentarily interrupting the current during battery charging and discharging, measuring the instantaneous voltage drop, and calculating it using Ohm's law; or the pulse discharge method, which involves applying a short-duration (e.g., 10ms) pulse current to the battery, measuring the voltage response at the start of the pulse, and calculating the ohmic internal resistance. The polarization internal resistance separation value is identified online. This refers to the resistance component corresponding to the voltage hysteresis caused by electrochemical reaction kinetics and concentration polarization effect during the charging and discharging process of a power battery. Its identification can be achieved using the current interruption method, analyzing the voltage change curve over time during the slow voltage recovery phase after measuring the ohmic internal resistance, and separating the polarization internal resistance through equivalent circuit model fitting; or using the pulse discharge method, analyzing the portion of the voltage that changes slowly over time in the voltage response curve after measuring the ohmic internal resistance, and obtaining the polarization internal resistance through equivalent circuit model fitting. The estimated highest internal temperature of the cell in the previous cycle is also included. This refers to the predicted maximum internal temperature of the battery cell obtained through a temperature estimation model in the previous calculation cycle. It is typically output and stored by the temperature estimation module at the end of the previous calculation cycle. The rate of change of open-circuit voltage with temperature. This refers to the rate at which the terminal voltage of a power battery changes with battery temperature under open-circuit conditions, reflecting the reversible entropic thermal effect of the electrochemical reactions within the battery. This parameter can be calibrated experimentally by measuring the battery's open-circuit voltage at different temperatures, fitting the measured data to obtain a curve showing the relationship between open-circuit voltage and temperature, and then calculating its derivative. Alternatively, a lookup table method can be used: detailed temperature-open-circuit voltage characteristic tests are performed on specific battery models in the laboratory beforehand to generate a lookup table; during system operation, the corresponding open-circuit voltage value is retrieved from the table based on the real-time temperature. value.

[0047] This solution achieves precise quantification of the heat source of a power battery by constructing a comprehensive heat generation power calculation model that includes Ohmic heat, polarization heat, and reversible entropy heat. Specifically, it calculates the real-time charging and discharging operating current. The squared terms are multiplied by the online identified ohmic resistance separation values ​​respectively. Separation value from polarization internal resistance This method accurately characterizes the Joule heat generated by current flowing through the battery's internal impedance, effectively distinguishing the heat generated by different physical processes within the battery, and overcoming the calculation errors caused by treating internal resistance as a single variable in traditional models. Specifically, it introduces a factor that includes the real-time charging and discharging operating current. The estimated highest internal temperature of the battery cell in the previous cycle. Rate of change of open-circuit voltage with temperature The product term precisely characterizes the reversible entropic thermal effect of the battery during charging and discharging, i.e., the endothermic or exothermic phenomena accompanying the chemical reaction of the battery. By superimposing the above three parts of heat, the true total heat generation power of the battery under different operating conditions can be reflected in real time and dynamically. This refined method of calculating heat generation power provides high-precision input data for the parameter identification module, thereby ensuring that the entire thermal management system can be adjusted based on an accurate heat generation benchmark, effectively avoiding problems of insufficient heat dissipation or overcooling caused by deviations in heat generation power estimation.

[0048] In one specific implementation, the parameter identification module can be integrated into the main control unit of the battery management system (BMS), which typically includes a high-performance microprocessor. When the railcar is operating under a high-power discharge condition, a high-precision current sensor connected to the main circuit of the battery pack acquires the real-time discharge current. The current is set to 100A (discharge state is set to positive). At this point, the main control unit, combining the real-time voltage response and operating current data during battery operation, triggers an internal online identification algorithm. Specifically, the algorithm relies on a preset first-order RC Thevenin equivalent circuit model. The microprocessor discretizes the data in the continuous time domain based on a preset sampling period, constructing a difference equation in Autoregressive Moving Average (ARMA) format: .in, This represents the voltage difference between the terminals after the depolarization open-circuit voltage. , , These are the model parameters to be identified. The microprocessor runs recursive least squares with a forgetting factor (FF-RLS) to iteratively update the parameter matrix in real time. Subsequently, based on the explicit algebraic mapping relationship between the parameter matrix and the properties of the physical model components, the system uses formulas... The pure ohmic resistance was calculated and then obtained using the formula. The polarization resistance is calculated by stripping away the polarization resistance. Through the rigorous mapping from the physical topology to the discrete mathematical model described above, the main control unit accurately decouples and estimates the current ohmic resistance separation value under the current operating conditions. The polarization internal resistance separation value is 1.0 mΩ (i.e., 0.001 Ω). The Ω is 0.5mΩ (i.e., 0.0005Ω), and the irreversible Joule heating power is instantaneously calculated by the microprocessor to be 10W (i.e., ) and 5W (i.e. Simultaneously, the microprocessor reads the estimated maximum internal temperature of the cell, which was stored and output by the temperature estimation module at the end of the previous control cycle. The value is 300K, combined with the pre-calibrated and stored open-circuit voltage as a function of temperature. (At this point, the corresponding value is 0.0001 V / K). The reversible thermodynamic entropy heat power used to characterize the dynamic endothermic and exothermic effects is calculated to be 3 W (i.e., The main control unit dynamically superimposes these three heat sources internally and ultimately calculates the total heat output of the current power battery cell in real time. 18W (i.e.) All of this data is dynamically updated in the microprocessor according to operating conditions, thereby ensuring that the entire thermal management system can be adjusted in real time based on an accurate heating baseline.

[0049] The above technical solution achieves precise quantification of the heat source of a power battery by constructing a comprehensive heat generation power calculation model that includes ohmic heat, polarization heat, and reversible entropy heat. This overcomes the estimation bias caused by traditional simple heat generation models that ignore the complex electrochemical characteristics inside the battery and fail to finely distinguish between internal resistance and reversible entropy heat components. By identifying ohmic and polarization internal resistance online and incorporating the reversible entropy heat effect, the solution can reflect the actual heat generation power of the battery under different operating conditions in real time and dynamically. This provides high-precision input data for subsequent temperature estimation and flow prediction, ensuring that the entire thermal management system can be adjusted based on an accurate heat generation benchmark. This effectively avoids insufficient heat dissipation or overcooling caused by deviations in heat generation power estimation, significantly improving the accuracy and response speed of the thermal management strategy.

[0050] This application further proposes that, in the temperature estimation module, the process of reconstructing the estimated value of the highest internal temperature of the battery cell includes: based on the total heat generation power, and the internal heat conduction dissipation power jointly determined by the estimated value of the highest internal temperature of the battery cell in the previous cycle, the surface weighted average temperature, and the equivalent internal thermal resistance of the power battery, a dynamic heat flow balance relationship is constructed inside the power battery; combined with the internal equivalent heat capacity parameters of the power battery, the dynamic rate of change of the highest internal temperature of the battery cell is calculated according to the dynamic heat flow balance relationship, and the dynamic rate of change is iteratively accumulated based on a preset calculation cycle to update the estimated value of the highest internal temperature of the battery cell in the current cycle; the specific calculation formula includes:

[0051] Calculate the rate of change of the highest internal temperature of the battery cell with respect to time:

[0052]

[0053] in, This refers to the internal equivalent heat capacity of the power battery. This represents the rate of change of the highest internal temperature of the battery cell with respect to time. Total heating power; The weighted average temperature of the power battery casing surface is obtained in real time based on the temperature sensor array; It is the equivalent internal thermal resistance for heat transfer from the core of the battery cell winding to the surface of the casing; This is an estimate of the highest internal temperature of the cell in the previous cycle; The internal equivalent heat capacity of a power battery characterizes the amount of heat that a cell can store when its temperature increases by one degree Celsius. This parameter can be obtained through experimental testing, data provided by the battery manufacturer, or thermodynamic calculations based on the battery materials and structure. For example, it can be calibrated by conducting adiabatic charge-discharge experiments on the battery and measuring its temperature rise relative to the input energy; or it can be calculated by thermally modeling the internal structure of the battery using finite element analysis software. The rate of change of the highest temperature inside the battery cell with respect to time is the direct output of the above thermal balance equation, reflecting the instantaneous rate of change of temperature inside the battery cell. The total heat generation power is calculated from the ohmic internal resistance, polarization internal resistance, and reversible entropy heat component separated online by the parameter identification module mentioned above. It represents all the heat generated by the power battery under the current operating state. The weighted average temperature of the power battery casing surface, acquired in real time based on a temperature sensor array, is a key externally measurable thermal parameter. This temperature can be obtained by arranging multiple thermocouples or thermistors on the battery casing surface and performing weighted averaging on the collected data to reflect the overall surface thermal state of the battery. The specific logic of the weighted averaging process is as follows: weights are assigned based on the predicted heat flux density of the sensor area or the physical distance from the battery cell tabs. The closer the core measuring point is to the tab, the higher the weight coefficient (for example, the weight of the measuring point at the tab end is set to 0.6, and the weight of the measuring point at the side wall end is set to 0.4), thereby more sensitively capturing surface hot spot characteristics. The equivalent internal thermal resistance from the core of the battery cell to the surface of the casing represents the resistance encountered when heat is transferred from the core region inside the cell to the surface of the battery casing. This parameter can be theoretically calculated based on the battery's structural design, the thermal conductivity of the materials, and the internal heat conduction path, or identified and calibrated through experimental methods such as transient thermal response testing. The highest internal temperature of the cell in the previous cycle is used as the initial condition for iterative calculation, ensuring the continuity and dynamism of temperature estimation.

[0054] Update the current estimate of the highest internal temperature of the battery cell:

[0055]

[0056] in, This is the estimated highest internal temperature of the cell in the current cycle; For the calculation period; This is the estimated highest internal temperature of the cell in the current cycle. It is the final output of this module and represents the predicted temperature of the hottest spot inside the cell at the end of the current calculation cycle. The calculation cycle refers to the time interval required to perform one temperature estimation calculation. This cycle is usually set according to the system's response speed requirements and computing resource limitations; for example, it can be set to 100 milliseconds or 1 second.

[0057] The temperature estimation module is a core component of the immersion liquid-cooled thermal management system. Its main function is to overcome the limitations of traditional surface temperature measurement methods by accurately modeling and calculating the internal thermal state of the power battery, thereby providing information on the highest internal temperature that more closely approximates the battery's actual thermal risk. This module can be a standalone hardware unit, such as an embedded controller or dedicated signal processor, or a software logic unit integrated into the main controller. Calculating the rate of change of the highest internal temperature of the cell is a crucial step in understanding the battery's internal thermal dynamics, quantifying the upward or downward trend of the highest internal temperature per unit time. This rate of change is typically calculated based on the battery's internal thermal balance equation, comprehensively considering the dynamic processes of heat generation and dissipation from the battery's interior to the exterior. Time integration is used to solve for the current estimate of the highest internal temperature of the cell, aiming to update the estimate of the highest internal temperature of the cell in real time by accumulating the rate of temperature change. This method can transform instantaneous temperature change trends into continuous temperature states, thereby achieving dynamic tracking of the battery's internal temperature.

[0058] This application's solution achieves accurate reconstruction from externally measurable physical quantities to the internal core temperature state by constructing an internal thermal balance model of the battery cell. Specifically, the temperature estimation module receives the total heat generation power calculated by the parameter identification module. And the weighted average temperature of the power battery casing surface acquired in real time through a temperature sensor array. Simultaneously, this is combined with the pre-calibrated or identified internal equivalent heat capacity of the power battery. The equivalent internal thermal resistance of the battery cell core to the casing surface And the estimated highest internal temperature of the cell in the previous cycle. First, the rate of change of the highest temperature inside the battery cell with respect to time was calculated. This calculation process precisely describes the dynamic balance between heat generation inside the battery cell and its conduction to the casing surface using a thermal balance equation. Subsequently, this rate of change is used... The estimated highest internal temperature of the battery cell in the previous cycle. and preset calculation cycle The current estimated maximum internal temperature of the battery cell is obtained by iteratively updating the value through time integration. This iterative calculation mechanism enables the system to track the dynamic changes in the internal temperature of the battery cell in real time, overcoming the lag inherent in relying solely on surface temperature feedback. Through this approach, the solution combines the precise heating power provided by the parameter identification module with the battery's internal thermophysical parameters, effectively compensating for the limitations of traditional methods that cannot directly measure the internal temperature of the battery cell. This provides the subsequent flow prediction module with more accurate and real-time information on the battery's internal thermal state, significantly improving the entire immersion liquid-cooled thermal management system's ability to predict and control battery thermal risks.

[0059] In one specific implementation, the temperature estimation module described above can be integrated into the main controller of the rail passenger vehicle's battery management system (BMS). This main controller typically includes a high-performance microprocessor (such as an ARM Cortex series processor) and sufficient memory. During the current control cycle of system operation, the processor receives the total heat output power input from the upstream parameter identification module. The power battery is rated at 18W. Multiple thermistors or thermocouples arranged on the surface of the battery casing simultaneously collect the casing temperature in real time, and the BMS performs weighted averaging to obtain key externally measurable thermal parameters. The value is 295K. When the system initially powers on and enters its first calculation cycle (cold start), When historical state records are lacking, the microprocessor calculates the surface-weighted average temperature under the current static state. direct assignment to This serves as the initial benchmark for iterative calculations. The software algorithm of the temperature estimation module pre-stores the internal equivalent heat capacity, determined through offline adiabatic charge-discharge experiments. and the equivalent internal thermal resistance of heat transfer from the core to the shell surface The processor first uses the estimated internal maximum temperature obtained in the previous calculation cycle, based on the thermal balance equation. Dividing the difference between the current surface temperature and the internal thermal resistance, the current power dissipated to the outside is calculated to be 10W (i.e., Then, subtracting the dissipated power of 10W from the total heating power of 18W yields the internal net heat accumulation rate of 8W. This is then divided by the heat capacity of 800J / K to accurately calculate the rate of change of the current instantaneous highest internal temperature of the cell with respect to time. The value is 0.01 K / s. Then, within the preset calculation period... Internally, the processor uses numerical integration methods such as the Euler method to perform time-series iterative accumulation, through... The dynamic update calculation ultimately solves for and outputs the estimated maximum internal temperature of the battery cell for the current cycle. The value is 300.01K. Through this periodic iterative calculation, the system can overcome the lag of traditional surface measurements and continuously output high-precision cell core temperature.

[0060] Through the above technical solution, this application overcomes the response lag and temperature estimation bias problems caused by traditional thermal management systems that rely solely on battery surface temperature for control. By accurately constructing an internal thermal balance model of the battery cell and combining it with online identification of total heat generation power and equivalent thermophysical parameters inside the battery, real-time, high-precision estimation of the maximum internal temperature of the battery cell is achieved. This precise perception of internal temperature enables the thermal management system to identify internal thermal risks of the battery earlier and more accurately. Therefore, under conditions of instantaneous high heat flux density, it can promptly activate or adjust cooling strategies to prevent overheating inside the battery cell, effectively ensuring the safe operation of the power battery and extending its service life. Compared to feedback control relying solely on surface temperature, this solution provides more forward-looking thermal state information, significantly improving the response speed and control accuracy of the thermal management system, and providing a solid guarantee for the efficient and safe operation of rail vehicle power batteries under complex conditions.

[0061] This application further proposes that, in the flow prediction module, the process of calculating the predicted target volumetric flow rate includes:

[0062] The first step is to generate a corresponding temperature regulation compensation power based on the temperature deviation between the allowable safety limit temperature inside the power battery and the estimated highest internal temperature of the cell in the current cycle, combined with the cooling time constant. This temperature regulation compensation power is then added to the total heat generation power to obtain the desired heat dissipation power. Specifically, the formula for calculating the desired heat dissipation power is as follows:

[0063]

[0064] in, The desired heat dissipation power required by the system to bring the power battery close to the safety limit; Total heating power; This refers to the internal equivalent heat capacity of the power battery. This is the preset desired cooling time constant; This is the estimated highest internal temperature of the cell in the current cycle; Define the permissible safe temperature limit inside the power battery; calculate the expected heat dissipation power. The aim is to determine how much heat the system needs to remove from the battery to achieve its highest internal temperature. Within the preset desired cooling time constant Temperature reaching the safety limit This calculation not only considers the current heat generation capacity of the power battery. Temperature deviation was also introduced. and time constant This allows for a forward-looking and dynamic calculation of heat dissipation power. The desired cooling time constant. The safe temperature limit can be preset based on the battery type, application scenario, and temperature response speed requirements. For example, it can be determined through experimental testing, simulation analysis, or empirical values. This setting is typically provided by the battery manufacturer or set according to the system's safe operating requirements; for example, it can be set to the battery's maximum permissible operating temperature or a value slightly below that temperature.

[0065] The second step involves interpolating the target volumetric flow rate and average fluid temperature from the previous cycle to obtain the equivalent convective heat transfer resistance for the current cycle. Combined with the desired heat dissipation power, surface weighted average temperature, and initial fluid temperature at the inlet, the predicted outlet temperature of the coolant as it exits the power battery is calculated based on the convective heat transfer physical boundary constraint equations. The formula for calculating the predicted outlet temperature of the coolant as it exits the power battery is as follows:

[0066]

[0067] in, The weighted average temperature of the power battery casing surface is obtained in real time based on the temperature sensor array; The initial temperature of the fluid at the system inlet; The predicted actual outlet temperature of the coolant as it flows out of the battery pack; The equivalent convective heat transfer resistance from the battery casing to the coolant fluid is calculated; the predicted outlet temperature of the coolant as it flows out of the power battery is then estimated. This is based on the convective heat transfer equilibrium relationship. This step utilizes the weighted average temperature of the power battery casing surface. Initial fluid temperature at system inlet And the equivalent convective heat transfer thermal resistance from the battery casing to the coolant fluid. Combined with the desired heat dissipation power This is used to predict the outlet temperature of the coolant after absorbing heat from the battery. Equivalent convective heat transfer thermal resistance. This reflects the heat transfer efficiency between the battery surface and the coolant, and its value can be estimated through experimental measurement, numerical simulation, or based on empirical formulas. For example, it can be determined dynamically or offline based on the coolant flow rate, physical properties, and the battery geometry and surface characteristics. Weighted average temperature of the power battery casing surface. and the initial temperature of the fluid at the system inlet The temperature is then obtained in real time through the corresponding temperature sensor.

[0068] The third step involves calculating the predicted target volumetric flow rate based on the desired heat dissipation power, the enthalpy change relationship between the coolant density, constant-pressure specific heat capacity, and the temperature difference between the inlet and outlet fluids. The formula for calculating the predicted target volumetric flow rate is as follows:

[0069]

[0070] in, The calculated target volumetric flow rate prediction value; The density of the insulating coolant; Calculate the constant-pressure specific heat capacity of the insulating coolant; predict the target volumetric flow rate. Based on the principle of energy conservation, the desired heat dissipation power is... This is converted into the required coolant flow rate. The calculation utilizes the density of the insulating coolant. and constant pressure specific heat capacity and the temperature difference between the inlet and outlet of the coolant. Density of insulating coolant and constant pressure specific heat capacity These are inherent physical properties of the coolant, which can be obtained by consulting material handbooks or conducting experimental measurements. For example, for a specific insulating coolant, these parameters can be considered constants within a certain temperature range, or corrected for using temperature-dependent functions.

[0071] The flow prediction module is a key component of the thermal management system. Its main function is to proactively calculate the required coolant flow rate based on the real-time thermal state of the power battery and preset cooling targets. This module is typically implemented by software algorithms and runs on the controller or processor of the thermal management system, responsible for translating the battery's thermal parameters into specific cooling execution commands.

[0072] This application's solution, by constructing a flow prediction module, achieves closed-loop feedforward calculation from the internal thermal state of the power battery to the flow control of the cooling system, effectively solving the problems of lag or insufficient heat dissipation in traditional thermal management systems. This module first utilizes the ohmic internal resistance and polarization internal resistance separated online by the parameter identification module and calculates the total heat generation power. And the estimated maximum internal temperature of the cell reconstructed by the temperature estimation module. This input ensures accurate perception of the battery's thermal state. Based on this, the desired heat dissipation power is calculated in the first step. The highest internal temperature of the current battery cell With safety limit temperature The deviation between them, combined with the preset expected cooling time constant This is transformed into dynamic heat dissipation demand. This approach not only considers the current heat generation power of the battery but also smooths temperature deviations through a time constant, ensuring the stability and foresight of the heat dissipation power calculation. Subsequently, the second step, based on the convective heat transfer balance relationship and combined with the weighted average temperature of the power battery casing surface... Initial fluid temperature at system inlet and equivalent convective heat transfer thermal resistance The predicted outlet temperature of the coolant flowing out of the power battery was calculated. This prediction method based on physical properties can accurately reflect the temperature rise of the coolant after flowing through the battery pack, providing accurate temperature difference parameters for subsequent flow rate calculations. Finally, the third step, based on the law of conservation of energy, utilizes the expected heat dissipation power... Temperature difference between coolant inlet and outlet The proportional relationship directly yields the predicted target volumetric flow rate of the coolant. This calculation method directly links the battery's heat load requirements with the physical properties of the coolant, enabling precise quantification of cooling flow. The entire process forms a logically rigorous predictive chain, allowing the thermal management system to proactively and accurately adjust the flow based on the real-time heating status of the power battery, thereby significantly improving the response speed and safety of thermal management.

[0073] In one specific implementation, the flow prediction module can be deployed in the central controller of the thermal management system. The central controller dynamically executes the prediction chain every 100 milliseconds by executing a fixed feedforward closed-loop control algorithm. The controller first retrieves the estimated maximum internal temperature of the cell for the current cycle from the temperature estimation module. Compare it with the preset safety limit temperature in the non-volatile memory. A comparison revealed an internal temperature deviation of 2.01K. The central controller then combined this with the preset desired cooling time constant. and internal equivalent heat capacity The calculated temperature regulation compensation power is 16.08W (i.e. This is then dynamically added to the total heat dissipation power of 18W to obtain the desired heat dissipation power required for the system to bring the power battery close to the safety limit. The value is 34.08W. Subsequently, the controller obtains the initial fluid temperature at the system inlet from the temperature sensor at the coolant inlet. The equivalent convective heat transfer resistance for the current cycle is dynamically extracted by interpolating the target volumetric flow rate and average fluid temperature from the previous cycle and looking up a table. Based on the physical boundary constraint equations of convective heat transfer, the controller executes... The feedforward formula is used to calculate the predicted outlet temperature of the coolant when it flows out of the battery pack. Approximately 293.2 K. Finally, based on the law of conservation of energy, the controller combines the known stored intrinsic physical properties of the insulating coolant (density) Specific heat capacity under constant pressure The expected heat dissipation power of 34.08W is compared with the predicted temperature difference between the fluid inlet and outlet (i.e., Substituting these values ​​into the enthalpy change equation, the system ultimately solves and outputs the accurate predicted target volumetric flow rate under the current operating conditions. Approximately This predicted value is then proactively and forward-lookingly transmitted to the downstream control module.

[0074] Through the above technical solution, this application can quantitatively map the real-time thermal state of the power battery to the coolant flow rate, realizing feedforward active adjustment of the coolant flow rate. This predictive control strategy effectively avoids the response lag problem existing in traditional feedback control, enabling the thermal management system to respond more quickly and accurately to the instantaneous thermal load changes of the power battery. By accurately calculating the required heat dissipation power and coolant flow rate, the system can avoid over-cooling or under-cooling, thereby optimizing the energy consumption of the cooling system while ensuring the thermal safety of the power battery, and improving the overall operating efficiency and lifespan of the rail vehicle's power battery system.

[0075] This application further proposes that the process of adaptively adjusting the cyclic drive power in the collaborative optimization module includes:

[0076] The baseline viscosity constant and flow activation energy of the insulating coolant are obtained. Based on the physical characteristic that fluid viscosity decreases nonlinearly with temperature, and combined with the real-time average fluid temperature, the temperature-corrected dynamic viscosity of the fluid is calculated. The formula for calculating the temperature-corrected dynamic viscosity of the fluid is as follows:

[0077]

[0078] in, The dynamic viscosity of the fluid after temperature changes. The reference viscosity constant is used. The activation energy for the flow of insulating coolant. The universal ideal gas constant is... This represents the real-time average temperature of the immersion liquid cooling fluid.

[0079] The collaborative optimization control module calculates the dynamic viscosity of the fluid after correction for real-time average fluid temperature using the Arrhenius relation. This aims to obtain the dynamic viscosity of the coolant in real time, reflecting its true physical characteristics as it changes with temperature. In immersion liquid cooling systems, coolant viscosity is a key parameter affecting flow resistance and pump power consumption. Traditional fixed viscosity assumptions cannot accurately reflect actual operating conditions, leading to control deviations. Calculations using the Arrhenius relation tightly couple the fluid's thermophysical properties with real-time temperature conditions, providing precise input for subsequent power regulation. The collaborative optimization control module can integrate a microcontroller or digital signal processor (DSP) that pre-stores parameters from the Arrhenius relation (such as the reference viscosity constant). Activation energy of flow Universal ideal gas constant It also receives the real-time average temperature of the immersion liquid-cooled fluid from fluid temperature sensors (such as thermocouples or thermistors). The processor performs Arrhenius equation calculations based on the received temperature value and outputs the current dynamic viscosity of the fluid. Alternatively, the collaborative optimization control module can also be implemented using a lookup table method. A lookup table containing the dynamic dynamic viscosity values ​​of the fluid at different temperatures is pre-built in memory. When the real-time average fluid temperature is obtained... Subsequently, the collaborative optimization control module quickly obtains the corresponding fluid dynamic viscosity from the lookup table using an interpolation algorithm (such as linear interpolation or spline interpolation). This approach can reduce the burden of real-time computation and improve response speed.

[0080] Based on the obtained fluid dynamic viscosity, structural resistance parameters of the cooling channel (equivalent total flow length and equivalent hydraulic diameter), and predicted target volumetric flow rate, the net hydrodynamic work required to overcome the viscous frictional resistance of the fluid within the cooling channel is calculated. Combined with the electromechanical conversion efficiency of the circulating drive mechanism, the circulating drive power is output. The formula for calculating the circulating drive power required for the predicted target volumetric flow rate is as follows:

[0081]

[0082] in, For cyclic drive power, For fluid dynamic viscosity, This is the equivalent total flow length of the cooling channel. For the target volumetric flow rate prediction value, This is the equivalent effective cross-sectional area of ​​the cooling channel. The equivalent hydraulic diameter of the flow channel. The overall mechanical-electrical energy conversion efficiency of the actuator.

[0083] The adaptive adjustment of the pump's output to achieve the predicted target volumetric flow rate requires precise calculation and adjustment of the pump's cyclic drive power based on real-time changes in the fluid's dynamic viscosity and the expected target volumetric flow rate. Its core lies in achieving dynamic matching between pump power output and actual fluid resistance, avoiding excessive or insufficient power due to viscosity variations, thereby optimizing system energy efficiency. The collaborative optimization control module calculates the fluid's dynamic viscosity... Then, the target volumetric flow rate prediction value obtained from the flow prediction module is combined. And the preset cooling channel geometry parameters (equivalent total flow length) Equivalent effective cross-sectional area The equivalent hydraulic diameter of the flow channel The combined mechanical-electrical energy conversion efficiency of the actuator and the actuator Substituting the values ​​into the formula for calculating the cyclic drive power, we obtain the required cyclic drive power. Subsequently, the collaborative optimization control module sends commands to the pump driver via a digital-to-analog converter (DAC) or pulse-width modulation (PWM) signal to precisely control the pump's speed or output torque, achieving the calculated cyclic drive power. Alternatively, the collaborative optimization control module can employ a model predictive control (MPC) based strategy. This is done when obtaining the fluid's dynamic viscosity. and target volume flow rate prediction Then, the MPC controller predicts the dynamic behavior of the system over a future period and optimizes the pump's cycle drive power based on the prediction results. This controller not only considers the current parameters but also the system's inertia and response delay. Through iterative optimization, it ensures that the pump's output power can smoothly and accurately maintain the fluid flow rate at the target volumetric flow rate prediction value while minimizing energy consumption.

[0084] The collaborative optimization control module of this application receives the target volumetric flow rate prediction value output by the flow prediction module. Subsequently, to ensure the pump drive system can achieve the target flow rate with optimal energy efficiency, the dynamic viscosity of the submerged liquid cooling fluid is first corrected in real time using the Arrhenius relation. Specifically, the collaborative optimization control module obtains the real-time average temperature of the submerged liquid cooling fluid. And using preset Arrhenius relational parameters (such as the reference viscosity constant) Activation energy of flow Universal ideal gas constant ), calculate the dynamic viscosity of the fluid at the current temperature. This step is crucial because it overcomes the limitations of traditional thermal management systems that treat coolant viscosity as a constant, accurately reflecting the actual change in coolant viscosity with temperature, thus providing a precise physical basis for subsequent power calculations. Subsequently, the collaborative optimization control module utilizes the calculated fluid dynamic viscosity... Combined with the target volume flow rate prediction value provided by the flow prediction module and the equivalent total flow length of the cooling channel. Equivalent effective cross-sectional area The equivalent hydraulic diameter of the flow channel The combined mechanical-electrical energy conversion efficiency of the actuator Substituting the values ​​into the formula for calculating the cycle drive power, the required cycle drive power to achieve the target volumetric flow rate can be accurately calculated. In this way, the collaborative optimization control module can adaptively adjust the pump's output power to match the flow resistance determined by the current fluid viscosity. This scheme, together with the aforementioned parameter identification module, temperature estimation module, and flow prediction module, forms a closed-loop and feedforward optimized thermal management system. The parameter identification module isolates the internal resistance online and calculates the total heat generation power. This provides accurate heat source information for the temperature estimation module; the temperature estimation module reconstructs the highest internal temperature of the battery cell. This provides the critical temperature status for the flow prediction module; the flow prediction module is based on the expected cooling time constant. Using a convective heat transfer model, the target volumetric flow rate is predicted by feedforward calculation. Building upon this, the collaborative optimization control module further considers the dynamic changes in coolant viscosity with temperature, and finely adjusts the pump's drive power. This collaborative working mechanism ensures that the system can not only quickly respond to changes in battery heating and predict the required cooling flow rate, but also dynamically adjust the pump power to adapt to changes in fluid physical properties during cooling tasks. This significantly improves the energy efficiency of the entire thermal management system while ensuring battery thermal safety.

[0085] As a specific implementation, the collaborative optimization control module can employ a high-performance embedded controller, such as a microcontroller with a floating-point unit, to convert the predicted target volumetric flow rate into a hardware-driven signal for optimal energy efficiency. During actual cooling operations, the controller acquires signals from PT100 platinum resistance temperature sensors at the coolant inlet and outlet in real time via an analog-to-digital converter (ADC) interface, calculating the real-time average temperature of the immersion liquid cooling fluid. The value is 291.6K (i.e., the average of inlet 290K and predicted outlet 293.2K). The controller loads the pre-stored insulating coolant reference viscosity constant from non-volatile memory. Activation energy of flow and the universal ideal gas constant Substituting the values ​​into the Arrhenius relation, a nonlinear temperature correction calculation is performed to accurately determine the dynamic viscosity of the fluid after changes caused by the current fluid temperature. The value is 0.002 Pa·s (i.e., 2 mPa·s). To accurately overcome the viscous frictional resistance of the fluid within the cooling channel, the controller further incorporates the inherent geometric resistance parameter of the flow channel—the equivalent total flow length. Equivalent effective cross-sectional area The equivalent hydraulic diameter of the flow channel and the upstream input flow forecast value and the overall mechanical-electrical energy conversion efficiency of the actuator. Substitute these values ​​into the cycle drive power calculation formula. Using microprocessor floating-point operations, the cycle drive power required to maintain the target volumetric flow rate is accurately calculated. Approximately Before the controller generates the PWM control signal, the system has a limiting protection logic. When the calculated cyclic drive power... Exceeding the rated maximum power of the actuator (pump) At that time, the controller forcibly clamps the output power to [the specified value]. Simultaneously, a heat dissipation saturation warning is reported to the vehicle controller via the communication bus, requesting a reduction in the charge / discharge rate of the power battery. Ultimately, the central controller uses this calculated power to retrieve the internally stored power-speed characteristic curve mapping table of the pump, converting it into an electrical signal. This signal is then generated by its integrated PWM output module and output to the DC motor driver, thereby precisely adjusting the actual speed of the circulating pump. When the fluid temperature rises, causing the viscosity to decrease non-linearly with temperature, the controller calculates a lower resistance power consumption and adaptively reduces the pump output, ensuring minimal energy loss from the circulating drive throughout its entire lifespan across temperature fluctuations.

[0086] Through the above technical solution, this application effectively solves the problem of reduced energy efficiency of pump drive systems caused by changes in the physical properties of coolant with temperature. By calculating the dynamic dynamic viscosity of the fluid in real time and adaptively adjusting the circulating drive power accordingly, the system can ensure that the pump's output power always matches the actual fluid flow resistance. This avoids the power excess or deficiency caused by viscosity changes in traditional control strategies, thereby significantly reducing the energy consumption of the pump drive system. Combined with the aforementioned parameter identification, temperature estimation, and flow prediction functions, this solution not only achieves precise thermal management of the power battery but also further optimizes the overall energy efficiency of the system, improving the economy and sustainability of the immersion liquid-cooled thermal management system for rail passenger vehicle power batteries.

[0087] Traditional rail vehicle power batteries generate significant heat and experience rapid temperature rise under intensive charge-discharge conditions, directly threatening safety and lifespan. Traditional air-cooled or indirect plate-type liquid cooling solutions have limited heat dissipation efficiency, struggling to handle instantaneous high heat flux densities. Furthermore, temperature sensors, only surface-mounted, cannot accurately capture the highest internal temperature, leading to conservative thermal management strategies and large system size and energy consumption. While immersion liquid cooling technology improves heat exchange capacity, control strategies still largely rely on surface temperature feedback, resulting in response lag and temperature overshoot. Some methods calculate heat dissipation based on simplified models, failing to consider the online separation of ohmic and polarization resistance, or the reversible entropy heat component, leading to inaccurate power estimation. Flow control lacks feedforward prediction based on the desired cooling time constant and convective heat transfer dynamics, hindering rapid matching of heat dissipation requirements. Simultaneously, coolant viscosity varies significantly with temperature; existing pump drive systems often operate with constant power or simple PID output, failing to adaptively adjust based on real-time viscosity, resulting in additional energy consumption and impacting overall vehicle energy efficiency.

[0088] In response, this application proposes an immersion liquid-cooled thermal management device for a rail passenger vehicle's power battery, including a memory, a processor, and a communication bus. The memory and the processor are connected via the communication bus. The memory stores a computer program, which, when executed by the processor, enables the functional configuration of each module in the aforementioned immersion liquid-cooled thermal management system for a rail passenger vehicle's power battery.

[0089] This application's solution constructs a hardware-software integrated computing platform, solidifying the complex battery thermal management logic into an executable computer program, thus achieving digital closed-loop control of the immersion liquid cooling system. Specifically, when the processor executes the program in memory, it can call the parameter identification module in real time. Using the acquired real-time operating data and surface temperature, it can separate the ohmic internal resistance and polarization internal resistance online and calculate the total heat generation power. This identification method based on real-time parameters overcomes the estimation bias caused by traditional static models, providing an accurate input basis for subsequent thermal management. Since the ohmic internal resistance reflects the instantaneous impedance characteristics during battery charging and discharging, while the polarization internal resistance reflects the electrochemical reaction kinetics limitation, by separating the two online and incorporating the reversible entropy heat component, the total heat generation power can be accurately quantified, thereby avoiding the failure of thermal management strategies caused by errors in heat generation power estimation.

[0090] Furthermore, by executing a temperature estimation module and combining the total heat generation power with the internal equivalent thermophysical parameters, the estimated maximum internal temperature of the battery cell is reconstructed. This extrapolation from surface temperature to internal temperature solves the temperature monitoring lag problem caused by the limitations of sensor placement, ensuring that the thermal management strategy can make decisions based on the core temperature. Specifically, this module, based on the internal heat conduction model of the battery cell, uses surface temperature data and total heat generation power to accurately calculate the temperature distribution in the core region of the battery cell by dynamically updating the equivalent heat capacity and thermal resistance parameters. This yields an estimated maximum internal temperature higher than the surface temperature, effectively eliminating the risk of conservative control or temperature overshoot caused by temperature sensing lag.

[0091] Furthermore, the flow prediction module converts temperature deviation into desired heat dissipation power based on the cooling time constant. Combining this with a fluid convection heat transfer temperature rise model and dynamically extracted convection heat transfer resistance, it calculates the predicted target volumetric flow rate. This feedforward control strategy can respond to heat load changes in advance, avoiding temperature overshoot caused by traditional feedback regulation. In practice, the system determines the desired heat dissipation power requirement corresponding to the temperature deviation based on the preset cooling time constant. Then, by monitoring the temperature difference between the coolant inlet and outlet in real time, it dynamically extracts the changing characteristics of the convection heat transfer resistance, thereby calculating the predicted target volumetric flow rate that meets the instantaneous high heat flux density heat dissipation requirements. This significantly improves the response speed and matching accuracy of cooling flow rate regulation.

[0092] Meanwhile, the collaborative optimization control module adaptively adjusts the circulation drive power based on the dynamic dynamic viscosity changes caused by fluid temperature. This dynamic compensation mechanism for changes in the physical properties of the coolant effectively reduces the additional energy consumption of the pump drive system, enabling the thermal management system to achieve refined management of energy efficiency while ensuring safety. Since the dynamic viscosity of the coolant exhibits non-linear characteristics with temperature changes, this module dynamically calculates the current viscosity value by monitoring the average coolant temperature in real time and combining it with pre-stored viscosity-temperature characteristic data. Based on this, it optimizes the pump drive power output, ensuring that the target volumetric flow rate is achieved while minimizing energy loss in the circulation drive power.

[0093] Through the above technical solutions, this application can identify battery internal resistance and heat generation power online, accurately reconstruct the core temperature of the battery cell, calculate cooling flow rate in advance, and collaboratively optimize pump drive power, thereby improving the thermal safety, response speed, and energy efficiency of the rail vehicle battery system. For example, under high-power discharge conditions in rail vehicles, the system can quickly identify internal temperature anomalies and adjust the cooling flow rate in advance to avoid temperature overshoot; at the same time, it can adaptively adjust the pump power according to changes in coolant viscosity to reduce ineffective energy consumption, ultimately achieving intelligent, precise, and efficient control of the thermal management strategy.

[0094] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A submerged liquid-cooled thermal management system for a power battery in a rail passenger vehicle, characterized in that, include: The parameter identification module is configured to acquire real-time operating data of the power battery and surface temperature based on temperature sensor array, separate the ohmic internal resistance and polarization internal resistance of the power battery online, and calculate the real-time total heat generation power of the power battery. The temperature estimation module is configured to combine the total heat generation power with the internal equivalent thermophysical parameters of the power battery to reconstruct the estimated value of the highest internal temperature of the cell. The flow prediction module is configured to convert the temperature deviation into the expected heat dissipation power based on the cooling time constant, and combine the fluid convection heat transfer temperature rise model and the dynamically extracted convection heat transfer thermal resistance to calculate the target volume flow prediction value. The collaborative optimization control module is configured to adaptively adjust the output to achieve the cycle drive power required to reach the target volumetric flow rate prediction value based on the dynamic dynamic viscosity changes caused by fluid temperature.

2. The immersion liquid-cooled thermal management system for rail passenger vehicle power batteries according to claim 1, characterized in that, In the parameter identification module, the process of calculating the real-time total heat generation power of the power battery includes: Based on the real-time charging and discharging operating current and the separated ohmic internal resistance and polarization internal resistance, the irreversible Joule heating power of the power battery is determined. Based on the direction and magnitude of the real-time charging and discharging operating current, the open-circuit voltage temperature coefficient of the power battery, and the estimated maximum internal temperature of the cell in the previous cycle, the reversible thermodynamic entropy heat power of the power battery is determined. The irreversible Joule heating power and the reversible thermodynamic entropy heating power are physically superimposed to obtain the real-time total heating power of the power battery.

3. The immersion liquid-cooled thermal management system for rail passenger vehicle power batteries according to claim 2, characterized in that, In the temperature estimation module, the process of reconstructing the estimated maximum internal temperature of the battery cell includes: Based on the total heat generation power, and the internal heat conduction dissipation power determined by the estimated highest internal temperature of the cell in the previous cycle, the surface weighted average temperature, and the equivalent internal thermal resistance of the power battery, a dynamic heat flow balance relationship inside the power battery is constructed. Combining the internal equivalent heat capacity parameters of the power battery, the dynamic rate of change of the highest temperature inside the cell is calculated based on the dynamic heat flow balance relationship. The dynamic rate of change is then iteratively accumulated over a preset calculation period to update and output the estimated value of the highest temperature inside the cell for the current period.

4. The immersion liquid-cooled thermal management system for rail passenger vehicle power batteries according to claim 3, characterized in that, In the flow prediction module, the process of feedforward calculation of the target volume flow prediction value includes: Based on the temperature deviation between the allowable safe limit temperature inside the power battery and the estimated maximum temperature inside the cell in the current cycle, the corresponding temperature regulation compensation power is generated in combination with the cooling time constant. The temperature regulation compensation power is then added to the total heat generation power to obtain the desired heat dissipation power.

5. The immersion liquid-cooled thermal management system for rail passenger vehicle power batteries according to claim 4, characterized in that, In the flow prediction module, the process of feedforward calculation of the target volumetric flow rate prediction value further includes: By interpolating the target volumetric flow rate and average fluid temperature of the previous cycle, the equivalent convective heat transfer resistance of the current cycle is obtained. Combined with the expected heat dissipation power, surface weighted average temperature and initial fluid temperature at the inlet, the predicted outlet temperature of the coolant when it flows out of the power battery is calculated according to the physical boundary constraint equation of convective heat transfer. Based on the expected heat dissipation power, and the relationship between the density of the coolant, the specific heat capacity at constant pressure, and the enthalpy change between the inlet and outlet fluid temperatures, the predicted value of the output target volumetric flow rate is calculated.

6. The immersion liquid-cooled thermal management system for rail passenger vehicle power batteries according to claim 5, characterized in that, In the aforementioned collaborative optimization control module, the process of adaptively adjusting the cyclic drive power includes: The reference viscosity constant and flow activation energy of the insulating coolant are obtained, and the dynamic dynamic viscosity of the fluid after temperature correction is calculated based on the physical property that the fluid viscosity decreases nonlinearly with temperature and the real-time average fluid temperature. Based on the obtained fluid dynamic viscosity, structural resistance parameters of the cooling channel, and predicted target volumetric flow rate, the net hydrodynamic work required to overcome the viscous frictional resistance of the fluid in the cooling channel is calculated, and the cycle drive power is output by combining the mechanical-electrical energy conversion efficiency of the cycle drive mechanism.

7. A submersible liquid-cooled thermal management device for a rail passenger vehicle's power battery, characterized in that, It includes a memory, a processor, and a communication bus, wherein the memory and the processor are connected via the communication bus; The memory stores a computer program, which, when executed by a processor, implements the functional configuration of each module in the immersion liquid-cooled thermal management system for rail passenger vehicle power batteries as described in any one of claims 1 to 6.

Images