Immersion cooling system

Abstract

Immersion cooling system, characterized in that the immersion cooling system comprises: an immersion control system and an electrochemical energy storage system; wherein the immersion control system comprises a pump control unit, a circulation pump and an immersion fluid storage unit, wherein the electrochemical energy storage system comprises a predetermined number of cluster immersion tanks, the cluster immersion tanks consisting of a corresponding number of honeycomb cells; wherein the electrochemical energy storage system acquires predicted temperature information for each cluster immersion tank and sends the predicted temperature information to the immersion cooling system, wherein the predicted temperature information of the cluster immersion tank comprises the predicted temperature information of all honeycomb cells in the cluster immersion tank; wherein the predicted temperature information is determined by cluster control units corresponding to each cluster immersion tank;wherein the pump control unit determines a corresponding target flow rate information based on the predicted temperature information and sends the target flow rate information to the circulation pump; wherein the circulation pump transports an immersion fluid in the immersion fluid storage unit, based on the target flow rate information, into each cluster immersion vessel of the electrochemical energy storage system to achieve the cooling treatment of batteries in each cluster immersion vessel.

Description

TECHNICAL AREA The present application relates to the field of electrochemical energy storage, in particular an immersion cooling system. STATE OF THE ART With the increasing use of electrochemical energy storage technologies in areas such as peak load coverage in the power grid and the integration of renewable energies, the operational stability, safety, and lifespan of energy storage systems are becoming increasingly important. Temperature is a crucial factor influencing the rate of power degradation and the safety risks of electrochemical energy storage batteries (such as lithium or sodium batteries). If the heat generated in the battery modules during charging and discharging cannot be dissipated and distributed evenly in a timely manner, significant temperature gradients occur both within and between the modules. This, in turn, leads to problems such as local overcharging or over-discharging, uneven capacity loss, and an increased risk of thermal instability. Current technical solutions for immersion cooling batteries typically dissipate heat using a circulation pump that moves the immersion fluid through the battery units. However, the control logic still relies primarily on simple feedback based on temperature thresholds; that is, the pump speed or flow rate is only adjusted once the battery temperature reaches a predetermined upper limit. This "retroactive" control model results in an inherent delay in battery temperature management: by the time the temperature sensor detects an anomaly, local overheating or excessive temperature differences between the electrical cores have often already occurred.It is then difficult to suppress sudden temperature fluctuations in time, which can easily lead to reduced battery homogeneity, a shortened cycle life and even an increased risk of thermal runaway. CONTENT OF THE PRESENT APPLICATION The purpose of this application is to provide an immersion cooling system that enables predictive control and dynamic optimization of the immersion fluid flow rate. This completely eliminates the significant temperature fluctuations and response delays caused by control delays, thereby achieving precise temperature control under all operating conditions. A first aspect of the embodiments of the present application provides an immersion cooling system, wherein the immersion cooling system comprises: an immersion control system and an electrochemical energy storage system; wherein the immersion control system comprises a pump control unit, a circulation pump, and an immersion fluid storage unit; wherein the electrochemical energy storage system comprises a predetermined number of cluster immersion tanks, the cluster immersion tanks consisting of a corresponding number of honeycomb cells; wherein the electrochemical energy storage system acquires predicted temperature information for each cluster immersion tank and sends the predicted temperature information to the immersion cooling system, wherein the predicted temperature information of the cluster immersion tank comprises the predicted temperature information of all honeycomb cells in the cluster immersion tank;wherein the predicted temperature information is determined by cluster control units corresponding to each cluster immersion tank; wherein the pump control unit determines a corresponding target flow rate information based on the predicted temperature information and sends the target flow rate information to the circulation pump; wherein the circulation pump delivers an immersion fluid in the immersion fluid storage unit, based on the target flow rate information, to each cluster immersion tank of the electrochemical energy storage system to achieve the cooling treatment of batteries in each cluster immersion tank. Optionally, the submersible control system further includes the following: a nitrogen liquid reservoir and a pressure sensor; wherein the pressure sensor detects a pressure value of the submersible liquid reservoir and transmits the pressure value to the pump control unit; wherein, based on the pressure value and a predetermined pressure value, the pump control unit controls an electromagnetic nitrogen refill valve of the nitrogen liquid reservoir and a pressure relief valve of the submersible liquid reservoir to maintain pressure equalization in the submersible liquid reservoir. Optionally, the electrochemical energy storage system further includes the following: battery management systems corresponding to each cluster immersion tank; wherein the battery management systems detect a power change rate during charging and discharging of the battery in the cluster immersion tank and send the power change rate to the cluster control unit of the corresponding cluster immersion tank; wherein the cluster control unit determines a current operating state of the cluster immersion tank based on the power change rate and a predetermined rate of change, and determines a target temperature prediction model corresponding to the cluster immersion tank based on the current operating state in order to predict the predicted temperature information of the cluster immersion tank;wherein, if the current operating state is a steady-state operating state, the corresponding target temperature prediction model is a predetermined mathematical prediction model to accurately predict future temperatures of all honeycomb cells in the cluster immersion tank; wherein, if the current operating state is a transient operating state, the corresponding target temperature prediction model is a predetermined neural network prediction model to rapidly predict future temperatures of all honeycomb cells in the cluster immersion tank. Optionally, if the power rate of change is less than or equal to the predetermined rate of change, the cluster control unit determines the current operating state of the cluster immersion tank as the steady-state operating state; where, if the power rate of change is greater than the predetermined rate of change, the cluster control unit determines the current operating state of the cluster immersion tank as the transient operating state. Optionally, the electrochemical energy storage system further includes the following: a temperature sensor; wherein the temperature sensor monitors temperature data of each honeycomb cell in the cluster immersion tank and transmits the temperature data to the cluster control unit; wherein the cluster control unit inputs the temperature data of each honeycomb cell in an identical cluster immersion tank into the target temperature prediction model to obtain the predicted temperature information for each honeycomb cell in the identical cluster immersion tank. Optionally, the electrochemical energy storage system further comprises the following: a central branch line and an edge branch line; wherein the central branch line is connected to each honeycomb cell in a high-temperature area of ​​the cluster immersion tank and is used to transport the immersion fluid to each honeycomb cell in the high-temperature area; wherein the edge branch line is connected to each honeycomb cell in a low-temperature area of ​​the cluster immersion tank and is used to transport the immersion fluid to each honeycomb cell in the low-temperature area. Optionally, the cluster control unit determines an opening degree of a smart branch valve corresponding to the central branch line and an opening degree of a smart branch valve corresponding to the peripheral branch line, based on the predicted temperature information of each honeycomb cell; the cluster control unit controls the opening of the smart branch valves based on the opening degree of the smart branch valve corresponding to the central branch line and the opening degree of the smart branch valve corresponding to the peripheral branch line, so that the immersion fluid conveyed by the circulation pump is conveyed to each honeycomb cell via the central branch line and the peripheral branch line. Optionally, the cluster control unit determines a maximum temperature corresponding to the high-temperature area based on the predicted temperature information of all honeycomb cells in the high-temperature area and determines the opening degree of the intelligent branch valve corresponding to the central branch line based on the highest temperature of the high-temperature area; wherein the cluster control unit determines a maximum temperature corresponding to the low-temperature area based on the predicted temperature information of all honeycomb cells in the low-temperature area and determines the opening degree of the intelligent branch valve corresponding to the edge branch line based on the highest temperature of the low-temperature area. Optionally, all honeycomb cells in each cluster immersion tank are arranged horizontally in a 1×N arrangement. Optionally, the electrochemical energy storage system further comprises the following: a collection groove and a drain line; wherein the collection groove is used to convey the immersion fluid to the drain line after completion of the cooling treatment; wherein the drain line is used to convey the immersion fluid into the immersion fluid storage tank. Beneficial effects: The embodiments of the present application provide an immersion cooling system, wherein the immersion cooling system comprises: an immersion control system and an electrochemical energy storage system; wherein the immersion control system comprises a pump control unit, a circulation pump, and an immersion fluid storage unit; wherein the electrochemical energy storage system comprises a predetermined number of cluster immersion tanks, the cluster immersion tanks consisting of a corresponding number of honeycomb cells; wherein the electrochemical energy storage system acquires predicted temperature information for each cluster immersion tank and sends the predicted temperature information to the immersion cooling system, wherein the predicted temperature information of the cluster immersion tank comprises the predicted temperature information of all honeycomb cells in the cluster immersion tank;wherein the predicted temperature information is determined by cluster control units corresponding to each cluster immersion tank; wherein the pump control unit determines a corresponding target flow rate information based on the predicted temperature information and sends the target flow rate information to the circulation pump; wherein the circulation pump delivers an immersion fluid in the immersion fluid storage unit, based on the target flow rate information, to each cluster immersion tank of the electrochemical energy storage system to achieve the cooling treatment of batteries in each cluster immersion tank. The present application uses a cluster controller in an electrochemical energy storage system to predict the future temperature of the cluster's immersion tank. The predicted temperature information obtained in this way is sent to the pump controller of the immersion control system, enabling the pump controller to quickly regulate the flow rate of the circulation pump. This allows for efficient control of the battery temperature within the cluster's immersion tank, preventing sudden temperature spikes and the resulting problems caused by delayed control. By predicting the battery temperature development and implementing timely dynamic control, the present application enables rapid cooling of the battery temperature within the cluster's immersion tank, thus ensuring precise temperature control under all operating conditions. BRIEF DESCRIPTION OF THE DRAWING To more clearly illustrate the technical solutions of the embodiments, the accompanying drawings, which are to be used in the embodiments of the present application, are briefly described below. Of course, the accompanying drawings described below are only some of the embodiments of the present application, and other accompanying drawings can be derived from them by a person with normal technical knowledge without any creative effort. Fig. 1 shows a schematic representation of an immersion cooling system according to an embodiment of the present application. DETAILED DESCRIPTION The technical solutions in the embodiments of the present application are clearly and completely described below in conjunction with the attached drawings in the embodiments of the present application, and it is clear that the described embodiments represent only a part of the embodiments of the present application and not all embodiments. Based on the embodiments of the present application, all other embodiments that could be obtained by a person skilled in the art without inventive step fall within the scope of protection of the present application. The embodiments of the present application provide an immersion cooling system. As shown in Fig. 1, the present application provides a schematic representation of an immersion cooling system, wherein the immersion cooling system comprises: an immersion control system and an electrochemical energy storage system; wherein the immersion control system comprises a pump control unit, a circulation pump, and an immersion fluid storage unit, and wherein the electrochemical energy storage system comprises a predetermined number of cluster immersion tanks, the cluster immersion tanks consisting of a corresponding number of honeycomb cells. The electrochemical energy storage system acquires predicted temperature information for each cluster immersion tank and sends the predicted temperature information to the immersion cooling system, wherein the predicted temperature information of the cluster immersion tank includes the predicted temperature information of all honeycomb cells in the cluster immersion tank; wherein the predicted temperature information is determined by cluster control units corresponding to each cluster immersion tank. In this embodiment, the electrochemical energy storage system comprises a plurality of cluster immersion tanks, each cluster containing a plurality of honeycomb cells, each honeycomb cell containing a battery module. When the battery module delivers electrical energy, its temperature rises. Since excessively high temperatures reduce the battery module's lifespan, cooling is necessary. In this embodiment, each cluster immersion tank is equipped with a corresponding cluster control unit. The cluster control unit receives temperature information from the honeycomb cells containing the battery modules and, based on this, predicts the future temperature of the honeycomb cells, thus obtaining the predicted temperature information for all honeycomb cells.After the cluster control units, corresponding to each cluster immersion tank in the electrochemical energy storage system, have determined the predicted temperature information, the cluster control units send the predicted temperature information to the immersion control system so that the immersion cooling system transfers the immersion fluid to the electrochemical energy storage system to lower the temperature of the batteries in the cluster immersion tanks. The pump control unit determines a corresponding target flow rate information based on the predicted temperature information and sends the target flow rate information to the circulation pump. In this embodiment, the immersion control system is used to supply the electrochemical energy storage system with the immersion fluid. The immersion control system comprises a pump control unit, a circulation pump, and an immersion fluid reservoir, as shown in Fig. 1. The pump control unit is used to regulate the flow rate of the circulation pump and the pressure in the immersion fluid reservoir. Specifically, after receiving the predicted temperature information sent by the cluster control unit, the pump control unit determines the highest temperature from the predicted temperature information and, based on a predetermined mapping relationship, determines the target flow rate corresponding to the highest temperature, generating target flow rate information to ensure that the battery module corresponding to the highest temperature is effectively cooled.This target flow rate information is sent to the circulation pump so that it transports the immersion fluid to the electrochemical energy storage system at the target flow rate. The predetermined mapping relationship is the correspondence between each temperature and each flow rate; this mapping relationship can be determined through experimental investigations by the personnel. The circulation pump transports an immersion fluid in the immersion fluid storage unit, based on the target flow rate information, into each cluster immersion tank of the electrochemical energy storage system to achieve the cooling treatment of batteries in each cluster immersion tank. After the immersion cooling system's circulation pump receives the target flow rate information, it draws immersion fluid from the system's reservoir and delivers it at the target flow rate to the inlet lines corresponding to each cluster immersion tank of the electrochemical energy storage system. From there, the immersion fluid is conveyed via the inlet lines to the honeycomb cells, effectively cooling the batteries within these cluster immersion tanks. The inlet lines are controlled by the corresponding intelligent master valves to allow independent regulation of each cluster immersion tank.The immersion fluid reservoir serves as a refill and buffer unit for the immersion cooling system to compensate for volume changes of the immersion fluid during circulation and to ensure a stable fluid level in the immersion cooling system. Furthermore, in this embodiment, the circulation pump features a magnetic drive, eliminating the need for a conventional mechanical shaft seal. The speed is continuously adjustable from 800 to 1200 rpm, which eliminates the risk of shaft seal leakage and, through speed adjustment, meets the flow requirements of various immersion fluids. For example, lower speeds are selected for oil-based fluids to reduce viscosity resistance, while higher speeds are used for fluorine-based fluids to increase heat exchange efficiency. Additionally, the circulation pump can be configured as a pump group, meaning it consists of two parallel-connected pumps with identical specifications, each capable of independent operation, and automatically responding to failures via a control unit.Under normal operating conditions, the system can operate with a single pump or with both pumps running in parallel. If one of the pumps experiences a problem such as insufficient flow, abnormal speed, or failure, the pump control unit automatically closes the inlet and outlet valves of the failed pump within 0.5 seconds of receiving the fault signal and activates the backup pump to ensure continuous operation of the circulation system without the risk of cooling interruption. The intelligent valves in the main line feature a composite seal consisting of two O-rings and a PTFE corrugated hose. In conjunction with appropriately installed leakage sensors, the sealing condition of the valves is monitored in real time to ensure that no immersion fluid leaks occur at the valves, thereby limiting the annual immersion fluid loss in the entire system to less than 0.3%. In another embodiment of the present application, the submersible control system further comprises: a nitrogen liquid reservoir and a pressure sensor; wherein the pressure sensor detects a pressure value of the submersible liquid reservoir and transmits the pressure value to the pump control unit; wherein the pump control unit, based on the pressure value and a predetermined pressure value, controls an electromagnetic nitrogen refill valve of the nitrogen liquid reservoir and a pressure relief valve of the submersible liquid reservoir in order to maintain pressure equalization in the submersible liquid reservoir. In this embodiment, the submersible control system also includes a nitrogen liquid reservoir, as shown in Fig. 1. Specifically, a nitrogen connection is provided at the neck of the submersible liquid reservoir, which is connected to a liquid reservoir for high-purity nitrogen, wherein the purity of the nitrogen in the nitrogen liquid reservoir is at least 99.9%. The nitrogen liquid reservoir is further equipped with an electromagnetic nitrogen refill valve, which is used to control the supply rate of nitrogen from the nitrogen liquid reservoir to the submersible liquid reservoir. By introducing nitrogen into the submersible liquid reservoir, a slight overpressure of 0.5 kPa to 1 kPa is maintained in the submersible liquid reservoir and in the entire closed circulation system.This pressure range effectively inhibits the evaporation rate of any immersion fluid (including oil-based and fluorine-based types), while simultaneously reducing the oxidation risk for oil-based fluids and preventing additional stress on system seals due to high pressure. By combining a closed-loop pressure control system to compensate for volume changes due to temperature fluctuations with the use of a magnetically driven pump and a high-precision sealing valve, the annual loss rate of the immersion fluid in the immersion cooling system is reduced to less than 0.3%, significantly reducing operating costs, eliminating the risk of fluid leaks, and increasing system safety. Furthermore, a pressure sensor is installed in the immersion cooling system, as shown in Fig. 1. This pressure sensor is located on the top of the immersion fluid reservoir and achieves an accuracy of ±0.1 kPa. This pressure sensor detects the pressure value in the immersion cooling system in real time and transmits it to the pump control unit. If the value displayed by the pressure sensor exceeds a predetermined pressure value, for example, 1 kPa, this indicates an expansion of the immersion fluid in the immersion cooling system. In this case, the pump control unit closes the electromagnetic nitrogen refill valve and simultaneously provides a slight pressure relief via the pressure relief valve integrated in the immersion fluid reservoir to maintain a stable pressure. If the pressure value falls below another predetermined lower limit, for example, 0.5 kPa, this indicates a decrease in the system volume.The pump control unit activates the electromagnetic nitrogen refill valve and refills with high-purity nitrogen gas until the pressure is back within the set range, in order to ensure volume compensation and pressure control in a closed control loop in the event of temperature fluctuations. In another embodiment of the present application, the electrochemical energy storage system further comprises: battery management systems corresponding to each cluster immersion tank; wherein the battery management systems detect a power change rate during charging and discharging of the battery in the cluster immersion tank and transmit the power change rate to the cluster control unit of the corresponding cluster immersion tank; wherein the cluster control unit determines a current operating state of the cluster immersion tank based on the power change rate and a predetermined rate of change, and determines a target temperature prediction model corresponding to the cluster immersion tank based on the current operating state in order to predict the predicted temperature information of the cluster immersion tank;wherein, if the current operating state is a steady-state operating state, the corresponding target temperature prediction model is a predetermined mathematical prediction model to accurately predict future temperatures of all honeycomb cells in the cluster immersion tank; wherein, if the current operating state is a transient operating state, the corresponding target temperature prediction model is a predetermined neural network prediction model to rapidly predict future temperatures of all honeycomb cells in the cluster immersion tank. In this embodiment, each cluster immersion tank is further equipped with a battery management system, as shown in Fig. 1. This battery management system is used to detect the power change rate during charging and discharging of the batteries in the corresponding cluster immersion tank and to transmit this data to the corresponding cluster control unit. Based on this power change rate and a predetermined rate of change, the cluster control unit determines the current operating state of the cluster immersion tank. Subsequently, based on temperature prediction models for various operating states, a target temperature prediction model for the current operating state of the cluster immersion tank is determined. In particular, if the current operating state is a steady-state operating state, the predetermined mathematical prediction model corresponding to this steady-state operating state is determined as the target temperature prediction model. Preferably, an RC thermal network model is chosen as the predetermined mathematical prediction model. Specifically, based on the heat exchange structure of the cluster immersion tank and the experimentally determined convective heat exchange coefficients, an integrated RC thermal network model is formed, encompassing "heat generation of the battery - structural heat conduction - convective heat exchange of the fluid". After verification through several groups of steady-state operating states, this model is embedded in an algorithm library of the control unit to obtain the predetermined mathematical prediction model.This model is used to predict the heat generation rate and temperature distribution trends, enabling a precise prediction of the future temperature of the honeycomb cells. Based on the predicted temperature information obtained from this mathematical model, the cluster controller can pre-control the opening degree of the smart valves within the relevant range. Similarly, the pump controller can pre-control the speed of the circulation pump to maintain a stable system temperature within the set range, e.g., 25°C to 35°C, and to prevent temperature fluctuations. Furthermore, this embodiment provides a method for creating an RC model, in which a thermal property test must first be performed on individual battery cells and modules before the RC model is created. Specifically, a constant charge and discharge power with a gradient of 0.1 Pn is set in the range of 0.1 Pn to 1 Pn, covering the SOC range of 20% to 100%, and the following data are recorded: data on the heat generation rate, data on the temperature response at multiple nodes, data on thermal resistance and heat capacity, and data on the heat diffusion properties.These data are mainly used for the creation and parameterization of the RC thermal network model, with the heat generation rate data supporting the model's heat generation module and the temperature response, thermal resistance, heat capacity and thermal diffusion data serving to adjust the parameters of the equivalent nodes to ensure that the model accurately reflects the actual thermal conductivity properties of the battery. The heat generation rate data is acquired as follows: Based on the heat generation of cells and modules per unit of time, curves are generated to show the change in the heat generation rate at different power levels and different SOC values. The temperature response data at multiple nodes is acquired as follows: Thermal sensors are positioned at the cathode and anode tabs of the cells, in the center of the side faces, and at the center, edges, and surface of the modules to record real-time temperatures at each node at a frequency of 1 Hz under different power levels and ambient temperatures (15°C to 45°C), thus generating temperature response curves.The data on thermal resistance and heat capacity are acquired as follows: By combining a steady-state heating procedure and a transient cooling procedure, the equivalent volume heat capacity of the individual cells and modules, as well as the contact thermal resistance between the cells, are measured; simultaneously, the changes in contact thermal resistance under different insertion forces are recorded. The data on heat diffusion properties are acquired as follows: Using pulse heating tests, the diffusion rates of heat within the cells and between the modules are recorded in order to determine the heat conduction time and the temperature decay laws. Subsequently, experiments are conducted to determine the parameters of convective heat exchange in the immersion fluid. Specifically, using the immersion fluid intended for the system as the medium, the flow channels and the honeycomb structure in a cluster immersion tank are simulated. Within a flow rate range of 0.4 m / s to 1.2 m / s, the convective heat exchange coefficients between the immersion fluid and the battery modules are tested at various flow rates using a combination of fluid mechanics simulation and experiments. This aims to establish a mapping between the flow rate and the heat exchange coefficient, thereby providing parameter support for the RC heat network model. The convective heat exchange coefficients and the flow rate mapping obtained from this experiment are primarily used to optimize the heat exchange module within the RC heat network model. Based on the experimental data above regarding the thermal properties of battery cells and modules, a basic model is created in which the battery module is represented as an RC thermal network with multiple nodes. Each node corresponds to a region of the battery module, the nodes are interconnected by equivalent thermal resistances, and each node is equipped with an equivalent heat capacity to simulate heat accumulation characteristics. The parameters for the equivalent thermal resistance and equivalent heat capacity are calibrated using the experimental data.Using the least-squares method, the temperature response curves are fitted to back-calculate the equivalent thermal resistance between each node and the equivalent heat capacity of the nodes, ensuring a ≥95% fit of the model to the actual heat conduction properties. Subsequently, taking into account the heat exchange structure of the cluster immersion tank, such as honeycomb cells, baffles, and branch lines, and incorporating the convective heat exchange coefficients obtained from experiments on the convective heat exchange parameters of the immersion fluid, an integrated RC heat network model was developed, encompassing "heat generation of the battery - structural heat conduction - convective heat exchange of the fluid". In particular, if the current operating state is a transient one, the predetermined neural network prediction model corresponding to this transient state is selected as the target temperature prediction model. Preferably, an LSTM neural network model is chosen as the predetermined neural network prediction model. The trained and optimized LSTM neural network model is linked to an operating state detection module and embedded in the control unit to enable fast response control under the transient operating state. That is, in a transient operating state where the temperature will change significantly, fast temperature control is of paramount importance. Therefore, a predetermined neural network prediction model with a higher prediction speed is used to obtain the predicted temperature information of the cluster immersion tank more quickly.This allows the cluster control unit to quickly adjust the opening degree of the smart valve based on this predicted temperature information, and the pump control unit can quickly control the speed of the circulation pump to achieve predictive temperature control and suppress temperature spikes and gradient changes under the transient operating condition. Furthermore, this embodiment provides a method for training optimization of an LSTM neural network model. Specifically, an experimental platform for a simulated energy storage system is set up to simulate the transient and steady-state operating conditions at a power change rate of -5 kW / min to +5 kW / min and to acquire multidimensional data for the creation of a dataset. The property dimensions include: the rate of change of the battery charging and discharging power, the real-time temperature in each zone (15°C to 50°C), the opening degree of the smart valves in the branch lines (0 to 100%), the speed of the circulation pump (800 rpm to 1200 rpm), and the system pressure (0.3 kPa to 1.2 kPa). The acquired data are preprocessed, with anomalous data such as sensor failures being removed.Missing values ​​are filled in using linear interpolation, and the data are split in a 7:2:1 ratio into training, validation, and test datasets, which are used for training and optimizing the LSTM neural network model. Based on this optimized LSTM neural network model, a temperature prediction error of ≤ ±0.5 °C can be guaranteed over a 30-second prediction period, enabling temperature fluctuations to be predicted 5 to 8 seconds in advance. The bimodal prediction and control method described in this embodiment enables intelligent switching between steady-state and transient operating conditions. The application of the predetermined mathematical prediction model at constant voltage ensures accurate temperature prediction. The application of the predetermined neural network prediction model in transient operating conditions allows for a predictive and rapid response, reducing the control response time to less than 3 seconds. Temperature changes can be predicted 5 to 8 seconds in advance, effectively suppressing temperature fluctuations caused by sudden power changes and ensuring adaptation to the complex operating conditions of the energy storage system. In another embodiment of the present application, if the power rate of change is less than or equal to the predetermined rate of change, the cluster control unit determines the current operating state of the cluster immersion tank as the steady-state operating state; wherein, if the power rate of change is greater than the predetermined rate of change, the cluster control unit determines the current operating state of the cluster immersion tank as the transient operating state. In this embodiment, the cluster control unit compares the power rate of change transmitted by the battery management system with a predetermined rate of change. Preferably, the predetermined rate of change is 2 kW / min. If the power rate of change is less than or equal to the predetermined rate of change, the cluster immersion chamber containing the battery management system is assumed to be in a steady-state operating condition. If the power rate of change is greater than the predetermined rate of change, the cluster immersion chamber containing the battery management system is assumed to be in a transient operating condition. In another embodiment of the present application, the electrochemical energy storage system further comprises: a temperature sensor; wherein the temperature sensor monitors temperature data of each honeycomb cell in the cluster immersion tank and transmits the temperature data to the cluster control unit; wherein the cluster control unit inputs the temperature data of each honeycomb cell in an identical cluster immersion tank into the target temperature prediction model in order to obtain the predicted temperature information for each honeycomb cell in the identical cluster immersion tank. In this embodiment, temperature sensors are distributed throughout each honeycomb cell of each cluster immersion tank. Specifically, each honeycomb cell has four temperature sensing points, each capturing temperature data at different positions of the battery modules within the cell. The temperature sensors transmit the captured temperature data to the cluster control unit. The cluster control unit inputs the temperature data from each honeycomb cell into a target temperature prediction model based on the current operating state to generate a temperature forecast and thus obtain the predicted temperature information for each honeycomb cell in the cluster immersion tank. This allows the cluster control unit to precisely control the smart valves based on the predicted temperature information for each honeycomb cell. Furthermore, this embodiment enables distributed fault diagnosis based on real-time temperature data acquired by the temperature sensors. Specifically, the cluster control unit detects anomalies based on the temperature data of each honeycomb cell. If the temperature at a detection point exceeds a preset threshold (e.g., 45°C) or the temperature difference within the same honeycomb cell exceeds 3°C, this is interpreted as a local cooling anomaly. The cluster control unit locates the fault area, closes the corresponding intelligent inlet and outlet valves of the cluster immersion tank, activates the backup cooling circuit, and simultaneously triggers a fault alarm to enable precise maintenance and operation. In another embodiment of the present application, the electrochemical energy storage system further comprises: a central branch line and an edge branch line; wherein the central branch line is connected to each honeycomb cell in a high-temperature region of the cluster immersion tank and is used to convey the immersion fluid to each honeycomb cell in the high-temperature region; wherein the edge branch line is connected to each honeycomb cell in a low-temperature region of the cluster immersion tank and is used to convey the immersion fluid to each honeycomb cell in the low-temperature region. In this embodiment, each cluster immersion tank of the electrochemical energy storage system is further equipped with a central branch line and a peripheral branch line, as shown in Fig. 1. Specifically, the inlet pipe at the connection of the cluster immersion tank is divided into a central branch line and a peripheral branch line by means of corrosion-resistant T-pieces. Furthermore, the cluster immersion tank is divided into a high-temperature zone and a low-temperature zone according to the operating temperatures in the different areas, with the average temperature of the honeycomb cells in the high-temperature zone being higher than the average temperature of the honeycomb cells in the low-temperature zone. The central branch line extends to below the honeycomb cells corresponding to the central high-temperature zone inside the tank, while the peripheral branch lines run along both sides of the tank bottom to below the honeycomb cells corresponding to the peripheral low-temperature zone. Each branch line has a precise liquid supply opening with a diameter of 8 mm to 10 mm for each honeycomb cell. The spacing corresponds to the distance between the honeycomb cells and is oriented towards the inlet side of the honeycomb cells to ensure a uniform supply of immersion fluid to each individual cell. The ends of the branch lines are sealed to prevent backflow and the formation of dead volume.At the same time, the outer walls of the branch lines are provided with an anti-deposition and anti-adhesion coating to reduce the influence of adhesions of oil-based fluid contaminants or residues of fluorine-based fluids on the flow rate during long-term operation. Furthermore, the branch lines are made of the same material as the inlet pipes, which is compatible with all types of immersion fluids (including oil-based and fluorinated fluids) (preferably PTFE or 316L stainless steel). The pipe diameter is designed according to the flow rate requirements of the respective area and the type of fluid, with the diameter of the central branch line being set at 25 mm to 32 mm to accommodate a flow rate of 0.8 m / s to 1.2 m / s. For fluorinated fluids, the upper limit is preferably selected, while for oil-based fluids, the lower to middle limit can be chosen. The diameter of the peripheral branch lines is set at 15 mm to 20 mm to accommodate a flow rate of 0.4 m / s to 0.6 m / s and to meet the basic flow requirements of all fluid types.This prevents variations in the flow rate due to mismatched pipe diameters. The branch lines are routed along the inside of the base of the cluster immersion tank and secured with a snap-fit ​​connection. They are located 50 mm below the base of the honeycomb cells, thus not interfering with the installation of either the battery modules or the honeycomb cells. Simultaneously, this ensures a uniform flow of the immersion fluid from the bottom, facilitating the return flow of the oil-based fluid and reducing dead volume. In another embodiment of the present application, the cluster control unit determines an opening degree of a smart branch valve corresponding to the central branch line and an opening degree of a smart branch valve corresponding to the peripheral branch line, based on the predicted temperature information of each honeycomb cell; wherein the cluster control unit controls the opening of the smart branch valves based on the opening degree of the smart branch valve corresponding to the central branch line and the opening degree of the smart branch valve corresponding to the peripheral branch line, such that the immersion fluid conveyed by the circulation pump is conveyed to each honeycomb cell via the central branch line and the peripheral branch line. In this embodiment, smart valves for the branch lines are installed on the two branch lines outside the cluster immersion tank, near the smart valve of the main line. The lines downstream of the valves are connected to the branch lines inside the cluster immersion tank via sealing fittings. The fittings have a seal consisting of two O-rings that interact with the sealing seats at the openings in the tank wall to ensure that the tightness inside the tank is not compromised by the arrangement of the lines. Based on the predicted temperature information of the individual honeycomb cells, the cluster control unit sends commands to the intelligent valves of the two branch lines to adjust their respective opening degrees and thus regulate the flow rate of the immersion fluid in the central and peripheral branch lines. Since the central branch line typically corresponds to the high-temperature range and the peripheral branch line to the low-temperature range, the opening degree of the central branch line's valve is set to a wider range to increase the flow rate. The opening degree of the peripheral branch line's valve is appropriately reduced to control the flow rate, resulting in a flow rate of 0.8 m / s to 1.2 m / s in the central high-temperature range and 0.4 m / s to 0.6 m / s in the peripheral low-temperature range. The exact values ​​can be dynamically fine-tuned depending on the fluid type.Simultaneously, the intelligent valve in the main line regulates the total fluid flow to the cluster immersion tank to ensure that the sum of the flow rates in both branch lines corresponds to the total heat generation of the batteries in the cluster immersion tank and the heat exchange properties of the fluid, thus preventing excessively high or low total flow rates. During operation, the intelligent valve adjusts its opening degrees in real time according to commands from the control unit, with a response time of ≤ 3 seconds, to achieve precise adjustment of the flow rates in each area. By varying the flow rates, heat exchange is enhanced in the high-temperature areas while simultaneously reducing energy consumption in the peripheral areas. In another embodiment of the present application, the cluster control unit determines a maximum temperature corresponding to the high-temperature range based on the predicted temperature information of all honeycomb cells in the high-temperature range and determines the opening degree of the intelligent branch valve corresponding to the central branch line based on the highest temperature of the high-temperature range; wherein the cluster control unit determines a maximum temperature corresponding to the low-temperature range based on the predicted temperature information of all honeycomb cells in the low-temperature range and determines the opening degree of the intelligent branch valve corresponding to the edge branch line based on the highest temperature of the low-temperature range. In this embodiment, after receiving the predicted temperature information, the cluster control unit filters out all predicted temperature information from the honeycomb cells belonging to the high-temperature zone in the cluster immersion tank. From this, it determines the highest temperature among all predicted temperatures in the high-temperature zone and, based on a predetermined mapping between the predicted temperature and the valve opening degree, determines the valve opening degree corresponding to this highest temperature. Based on the valve opening degree for the high-temperature zone, the cluster control unit sets the intelligent valve of the corresponding central branch line to this valve opening degree. Simultaneously, the cluster control unit filters out all predicted temperature information from the honeycomb cells belonging to the low-temperature zone within the cluster immersion tank. From this, it determines the highest temperature among all predicted temperatures in the low-temperature zone and, based on a predetermined mapping between the predicted temperature and the valve opening degree, determines the valve opening degree corresponding to this highest temperature. Based on the valve opening degree for the low-temperature zone, the cluster control unit sets the intelligent valves of the corresponding branch lines to this valve opening degree. In another embodiment of the present application, all honeycomb cells in each cluster immersion tank are arranged horizontally in a 1×N arrangement. In this embodiment, all honeycomb cells in the cluster immersion tank are arranged in a horizontal 1×N configuration, meaning all honeycomb cells are arranged horizontally in a row within the cluster immersion tank, replacing the conventional vertical N×1 configuration. Since each honeycomb cell contains a corresponding battery module, the battery modules are also arranged in a horizontal 1×N configuration within the cluster immersion tank. The purpose of this horizontal arrangement is twofold: firstly, to improve ease of maintenance: after opening the container's side doors, all modules in each row are positioned close to the container doors. No other components need to be moved during maintenance or module replacement, and sufficient working space is available, reducing maintenance time for individual battery modules by more than 40%; and secondly, to improve temperature uniformity.Since there is no stacking of upper and lower layers, heat build-up in the upper battery modules, which would otherwise be caused by heat dissipation from the lower battery modules, is avoided. Combined with complete encapsulation by the immersion fluid, the temperature difference between the battery modules can be reduced to below 2°C, resulting in better temperature control than with conventional vertical matrix arrangements. Furthermore, the battery modules in the cluster immersion tank are fully connected in series, with a 15 mm to 20 mm gap between them to allow for fluid flow and maintenance. The entire system is mounted horizontally in the center of the tank using an insulating bracket. The bracket is positioned 50 mm above the pipework running along the bottom of the tank to match the installation height of the branch lines. This ensures that the underside of the modules is precisely aligned with the fluid inlet ports, prevents any overlap of the battery modules between the upper and lower layers, and completely eliminates vertical temperature stratification caused by gravity.A gap of 8 mm to 10 mm is maintained between the surface of the battery modules and the inner wall of the honeycomb cells to ensure a smooth flow of the immersion fluid that surrounds the entire battery module and to accommodate the horizontal fluid flow direction. The honeycomb cells have a structure adapted to the contour of the battery modules, consisting of a single chamber of regular hexagons. The cross-section of these individual chambers is 12 mm to 15 mm larger than that of the battery modules to ensure the necessary spacing. PP or 316L stainless steel are used as materials, which are resistant to corrosion from various immersion fluids (including oil-based, fluorinated, etc.). The entire system is attached to the inner wall of the cluster immersion tank and to the module mounts using snap fasteners, achieving a one-to-one fit with the battery modules, meaning each individual honeycomb cell encloses a single battery module.The honeycomb cells cover the entire module area inside the container; the density of the honeycomb cells is the same in the central high-temperature area and in the peripheral low-temperature area, with heat exchange being adjusted solely by varying the flow rates in the lower branch lines. Flow openings are provided at the bottom of the honeycomb cell, corresponding to the fluid supply openings of the branch lines in the container, and their diameter matches that of the fluid supply openings, i.e., 8 mm to 10 mm. This ensures the precise introduction of the immersion fluid into the interior of the cell and simultaneously adapts to the flow characteristics of the oil-based fluid to prevent fluid residue. Furthermore, the honeycomb cells play a key role in structural protection and ease of maintenance: Regarding structural protection, their corrosion-resistant material protects the module mounts from corrosion caused by immersion fluids and simultaneously forms a physical barrier that prevents damage between the modules from operational vibrations and impacts, thus increasing the overall system's structural stability. With respect to ease of maintenance, the snap-fit ​​structure can be assembled and disassembled independently. Combined with a horizontal 1×N arrangement and the design with side-opening containers, adjacent battery modules do not need to be moved during maintenance. It is sufficient to remove the relevant honeycomb cell individually to service or replace the battery module.This further increases ease of maintenance and works synergistically with the advantage of a maintenance time reduced by more than 40% compared to a single battery module. In another embodiment of the present application, the electrochemical energy storage system further comprises: a collecting groove and a drain line; wherein the collecting groove is used to convey the immersion liquid to the drain line after completion of the cooling treatment; wherein the drain line is used to convey the immersion liquid into the immersion liquid storage tank. In this embodiment, a circumferential U-shaped collecting groove is arranged in the upper region of the cluster immersion tank, as shown in Fig. 1. The material of the collecting groove corresponds to the material of the honeycomb cells; PP or 316L stainless steel is used, which is resistant to corrosion by various immersion fluids. The groove width is 50 mm to 60 mm, and the depth is 30 mm to 40 mm. A guide surface inclined at 3° is provided on the inner wall, which directs the immersion fluid treated after heat exchange to the lowest point of the collecting groove. In particular, the collecting groove is arranged around the honeycomb cells, which are horizontally arranged in a 1×N configuration, and covers the upper outlet area of ​​all honeycomb cells. The immersion fluid, cooled within the honeycomb cells, flows along the guide surface into the collecting groove after naturally overflowing the top surface, thus preventing fluid from remaining on the top of the battery module. The lowest point of the collecting groove is precisely aligned with the outlet opening of the predetermined single outlet line on the side wall of the cluster immersion tank. The pipe diameter of the outlet opening is 5 mm to 10 mm larger than the pipe diameter of the inlet pipe to meet the flow requirements after merging. A seal consisting of two O-rings is also used at the connection point, which is connected to the outlet line outside the cluster immersion tank via a sealing flange.This reduces the number of openings in the vessel wall and ensures a tight seal under slight overpressure. This collection structure requires no branch lines or T-pieces; the integrated, circumferential design alone achieves a uniform merging of the fully cooled immersion fluid, thus avoiding sudden flow resistances that could impair temperature equalization. At the same time, it accommodates the properties of the low-viscosity, oil-based fluid, reducing the risk of fluid residue and, together with the gradient branch lines at the inlet, creating a "divide and merge" flow system. In another embodiment of the present application, the electrochemical energy storage system further comprises the following: vertical guide vanes. In particular, the guide vanes consist of a thin plate made of the same material as the honeycomb cells and with a thickness of 2 mm to 3 mm. They are mounted vertically inside the honeycomb cells and evenly spaced along the direction of fluid flow, i.e., from the lower inlet to an upper return, with the spacing strictly controlled to 15 mm. The guide vanes are inclined at an angle of 30° to the inner wall of the honeycomb cell, with the angle of inclination directed towards the lower inlet side, i.e., the underside of the vanes is inclined in the direction of the fluid supply. This guides the immersion fluid entering the honeycomb cell along the vanes, creating a swirling flow that breaks up the stratified flow regime.The number of baffles in each honeycomb cell depends on the module height; typically, 6 to 7 baffles are arranged per 100 mm of module height, extending from the bottom to the top of the module. This ensures that a deflection is created across the entire module height, thus preventing thermal stratification between the upper and lower areas. In summary, the cluster immersion tank is designed according to the principle of "liquid supply via the lower branch line, liquid flow through the honeycomb cells, flow deflection by the baffles, heat exchange with the battery modules, and return in the upper area." Combined with the horizontal arrangement of the modules, this design ensures sufficient contact between the immersion fluid and the modules, as well as a uniform temperature distribution. In a further embodiment of the present application, a mechanically passive bypass valve and a pressure sensor are provided on the circulating main line between the circulation pump and the intelligent valve of the main line, wherein the opening pressure of the bypass valve is set to 0.15 MPa. If the pressure in the pipeline exceeds 0.15 MPa due to a system malfunction, such as an accidentally closed valve or an anomaly in the pump assembly, the bypass valve opens automatically and forms an emergency pressure relief channel to prevent damage to pipelines and cluster immersion tanks from the high pressure. Simultaneously, the pressure sensor transmits the pressure value to the pump control unit, which then issues a corresponding alarm signal and prompts the operating personnel to take action. The honeycomb cells, inclined guide vanes, and U-shaped flow channel structure described in the present application further disrupt the horizontal layered flow, while the gradient flow rate design enables differentiated heat exchange between high-temperature and low-temperature areas. Combined with decentralized temperature monitoring, this ultimately limits the temperature difference across the entire battery module to below 2 °C, representing a reduction of more than 30% compared to conventional matrix arrangements. This effectively slows down battery performance degradation and avoids the risk of local thermal runaway. The present method combines three key safety measures: a redundant parallel pump design, a mechanical passive bypass valve emergency pressure relief, and decentralized fault diagnostics. Together, these ensure automatic switching in the event of pump group failures, emergency protection in case of overpressure, and the precise localization and isolation of local faults. This guarantees uninterrupted operation of the cooling system, increases the overall reliability of the energy storage system, and reduces operating and maintenance costs. Furthermore, the arrangement of various branch lines and the monitoring of temperature data enable gradient flow rate control. This control dynamically adjusts the flow rate to the heat generation characteristics, thus avoiding unnecessary energy consumption. A bimodal prediction algorithm precisely regulates the operating parameters. In combination with the adjustable speed of the circulation pumps, energy consumption is reduced by 15% to 20% compared to conventional immersion cooling systems, thereby meeting both cooling capacity and energy efficiency requirements. Each embodiment in this description is described step by step, and each embodiment focuses on the differences from other embodiments, and it is sufficient to refer to each embodiment for identical and similar parts of each embodiment. Although preferred embodiments of the present application have been described, a person skilled in the art, once they understand the basic inventive concepts, may make further changes and modifications to these embodiments. Therefore, the attached claims are to be interpreted as encompassing the preferred embodiments as well as any changes and modifications that fall within the scope of the present application. In this application, the terms “an embodiment”, “some embodiments”, “schematic example”, “specific example”, or “a particular example”, “an embodiment”, “some examples”, “exemplary”, “specific examples”, or “some examples” mean that certain features, structures, materials, or properties described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this description, the schematic expressions of the above terms need not refer to the same embodiments or examples. Furthermore, the described specific features, structures, materials, or properties may be combined appropriately in one or more embodiments. Furthermore, the terms "have," "include," or any other variant are intended to cover their exclusive inclusion, such that a process, procedure, object, or terminal comprising a set of elements includes not only those elements but also other elements not expressly listed or belonging to such process, procedure, object, or terminal. Without further limitation, the fact that an element is defined by the phrase "with a..." does not preclude the presence of another identical element in the process, method, object, or terminal comprising that element. The preceding section has presented in detail an immersion cooling system provided by the present application. The principles and embodiments of the present application have been explained in detail using specific examples; the explanations of the preceding embodiments serve only to facilitate a better understanding of the method and the core idea of ​​the present application. At the same time, a person skilled in the art in this field will perceive changes to the specific embodiments and the scope of application based on the concept of the present application. In summary, the content of this description should not be understood as limiting the scope of the present application.

Claims

Immersion cooling system, characterized in that the immersion cooling system comprises: an immersion control system and an electrochemical energy storage system; wherein the immersion control system comprises a pump control unit, a circulation pump and an immersion fluid storage unit, wherein the electrochemical energy storage system comprises a predetermined number of cluster immersion tanks, the cluster immersion tanks consisting of a corresponding number of honeycomb cells; wherein the electrochemical energy storage system acquires predicted temperature information for each cluster immersion tank and sends the predicted temperature information to the immersion cooling system, wherein the predicted temperature information of the cluster immersion tank comprises the predicted temperature information of all honeycomb cells in the cluster immersion tank; wherein the predicted temperature information is determined by cluster control units corresponding to each cluster immersion tank;wherein the pump control unit determines a corresponding target flow rate information based on the predicted temperature information and sends the target flow rate information to the circulation pump; wherein the circulation pump transports an immersion fluid in the immersion fluid storage unit, based on the target flow rate information, into each cluster immersion vessel of the electrochemical energy storage system to achieve the cooling treatment of batteries in each cluster immersion vessel. Immersion cooling system according to claim 1, characterized in that the immersion control system further comprises: a nitrogen liquid reservoir and a pressure sensor; wherein the pressure sensor detects a pressure value of the immersion liquid reservoir and transmits the pressure value to the pump control unit; wherein the pump control unit, based on the pressure value and a predetermined pressure value, controls an electromagnetic nitrogen refill valve of the nitrogen liquid reservoir and a pressure relief valve of the immersion liquid reservoir to maintain pressure equalization in the immersion liquid reservoir. Immersion cooling system according to claim 1, characterized in that the electrochemical energy storage system further comprises: battery management systems corresponding to each cluster immersion vessel; wherein the battery management systems detect a power change rate during charging and discharging of the battery in the cluster immersion vessel and send the power change rate to the cluster control unit of the corresponding cluster immersion vessel; wherein the cluster control unit determines a current operating state of the cluster immersion vessel based on the power change rate and a predetermined rate of change, and determines a target temperature prediction model corresponding to the cluster immersion vessel based on the current operating state in order to predict the predicted temperature information of the cluster immersion vessel;wherein, if the current operating state is a steady-state operating state, the corresponding target temperature prediction model is a predetermined mathematical prediction model to accurately predict future temperatures of all honeycomb cells in the cluster immersion tank; wherein, if the current operating state is a transient operating state, the corresponding target temperature prediction model is a predetermined neural network prediction model to rapidly predict future temperatures of all honeycomb cells in the cluster immersion tank. Immersion cooling system according to claim 3, characterized in that, if the power change rate is less than or equal to the predetermined change rate, the cluster control unit determines the current operating state of the cluster immersion tank as the steady-state operating state; wherein, if the power change rate is greater than the predetermined change rate, the cluster control unit determines the current operating state of the cluster immersion tank as the transient operating state. Immersion cooling system according to claim 3, characterized in that the electrochemical energy storage system further comprises: a temperature sensor; wherein the temperature sensor monitors temperature data of each honeycomb cell in the cluster immersion vessel and transmits the temperature data to the cluster control unit; wherein the cluster control unit inputs the temperature data of each honeycomb cell in an identical cluster immersion vessel into the target temperature prediction model in order to obtain the predicted temperature information for each honeycomb cell in the identical cluster immersion vessel. Immersion cooling system according to claim 1, characterized in that the electrochemical energy storage system further comprises: a central branch line and an edge branch line; wherein the central branch line is connected to each honeycomb cell in a high-temperature area of ​​the cluster immersion tank and is used to convey the immersion fluid to each honeycomb cell in the high-temperature area; wherein the edge branch line is connected to each honeycomb cell in a low-temperature area of ​​the cluster immersion tank and is used to convey the immersion fluid to each honeycomb cell in the low-temperature area. Immersion cooling system according to claim 6, characterized in that the cluster control unit determines an opening degree of a smart branch valve corresponding to the central branch line and an opening degree of a smart branch valve corresponding to the peripheral branch line, based on the predicted temperature information of each honeycomb cell; wherein the cluster control unit controls the opening of the smart branch valves based on the opening degree of the smart branch valve corresponding to the central branch line and the opening degree of the smart branch valve corresponding to the peripheral branch line, such that the immersion fluid conveyed by the circulation pump is conveyed to each honeycomb cell via the central branch line and the peripheral branch line. Immersion cooling system according to claim 7, characterized in that the cluster control unit determines a maximum temperature corresponding to the high-temperature range based on the predicted temperature information of all honeycomb cells in the high-temperature range and determines the opening degree of the intelligent branch valve corresponding to the central branch line based on the highest temperature of the high-temperature range; wherein the cluster control unit determines a maximum temperature corresponding to the low-temperature range based on the predicted temperature information of all honeycomb cells in the low-temperature range and determines the opening degree of the intelligent branch valve corresponding to the edge branch line based on the highest temperature of the low-temperature range. Immersion cooling system according to claim 1, characterized in that all honeycomb cells in each cluster immersion container are arranged horizontally in a 1 × N arrangement. Immersion cooling system according to claim 1, characterized in that the electrochemical energy storage system further comprises: a collecting groove and a drain line; wherein the collecting groove is used to convey the immersion fluid to the drain line after completion of the cooling treatment; wherein the drain line is used to convey the immersion fluid into the immersion fluid storage tank.

Images