Heat balance control method of liquid cooling charging module

Abstract

The invention discloses a thermal balance control method for a liquid cooling charging module, and relates to the technical field of power electronics, and the method comprises the steps: deploying a multi-dimensional distributed temperature sensing array in a key heating region of the liquid cooling charging module; original data are acquired through the temperature acquisition module and transmitted to the temperature field reconstruction module, a three-dimensional dynamic temperature field model is constructed by using a spatial interpolation algorithm, and temperature values, temperature gradients and temperature difference distribution of all regions are output in real time; the thermal balance control module compares the model output data with a preset temperature threshold value and a temperature difference threshold value, and if the threshold value exceeding condition exists, a regulation and control instruction is sent to a partition cooling execution unit; the subarea cooling execution unit comprises three independent branches (corresponding to key areas), and each branch realizes accurate adjustment through an independent electromagnetic flow valve and a variable frequency water pump; and the steps are circulated until the temperatures of all the key areas reach the standard and the temperature difference of the adjacent areas meets the standard, so that the heat balance precision and the operation safety of the liquid cooling charging module are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power electronics, in particular to a charging thermal balance control method. BACKGROUND

[0002] With the evolution of new energy vehicles to an 800V high-voltage platform, super-charging power has broken through 600kW, and the efficiency of traditional air cooling (temperature rise > 25℃) is difficult to meet the demand. The heat dissipation efficiency of liquid cooling technology is improved by 80% compared with air cooling, supporting 600A continuous output without derating, and has become a necessity for high-power charging.

[0003] At present, the liquid cooling scheme is divided into "semi-liquid cooling" (only charging terminal liquid cooling) and "full-liquid cooling" (full-system liquid cooling). Semi-liquid cooling has become the mainstream due to its low cost and mature technology. The popularization of liquid cooling technology promotes the improvement of charging module power density and prolongs the equipment life by more than 30%.

[0004] In the liquid cooling system, key components such as charging guns and cable interfaces are prone to local overheating due to uneven current distribution. Continuous charging for 10 minutes can cause a cable temperature rise of 15℃, and if not balanced in time, it may cause insulation failure or component aging. Therefore, thermal balance control is the core of ensuring charging safety and system reliability.

[0005] Therefore, there is an urgent need for a thermal balance control method for liquid-cooled charging modules. To address the issue of insufficient temperature control accuracy, existing liquid cooling systems rely on single-point temperature sensors, which are difficult to capture complex temperature field distribution. When the liquid-cooled super-charging pile is running at high power, the temperature difference between the charging gun head and the cable connection can reach 5-8℃, leading to the risk of local overheating. SUMMARY

[0006] To solve the above technical problems, a thermal balance control method for a liquid-cooled charging module is provided, which solves the problem of insufficient temperature control accuracy.

[0007] To achieve the above purpose, the technical scheme adopted by the present application is as follows: A thermal balance control method for a liquid-cooled charging module, comprising: Deploying a multi-dimensional temperature sensor array in the key heat generating area of the liquid-cooled charging module, the temperature sensor array comprising a plurality of distributed temperature sensors arranged according to predetermined positions; Real-time acquisition of the original temperature data of the temperature sensor array by a temperature acquisition module and transmission to a temperature field reconstruction module, processing of the original temperature data based on a spatial interpolation algorithm, construction of a three-dimensional dynamic temperature field model of the key heat generating area, and real-time output of the temperature value, temperature gradient and temperature difference distribution of each area; The output data of the three-dimensional dynamic temperature field model is transmitted to the thermal equilibrium control module, which presets a temperature threshold range and a temperature difference threshold. The thermal equilibrium control module compares the output data of the three-dimensional dynamic temperature field model with the preset threshold. If the temperature value of a certain area exceeds the maximum temperature threshold or the temperature difference between two adjacent key areas exceeds the temperature difference threshold, the thermal equilibrium control module sends a control command to the partitioned cooling execution unit of the liquid cooling system. The partitioned cooling execution unit includes at least three independent cooling branches, corresponding to the charging gun head, cable connection node, and cable mid-section, respectively. Each cooling branch is equipped with an independent electromagnetic flow valve and a variable frequency water pump. Repeat the above steps until the temperature values ​​of all key heating areas in the three-dimensional dynamic temperature field model are within the preset temperature threshold range, and the temperature difference between adjacent key areas is less than or equal to the temperature difference threshold.

[0008] Preferably, the temperature sensing array comprises multiple distributed temperature sensors, which are placed in predetermined positions, specifically including: Patches of temperature sensors are arranged at intervals along the circumference of the inner conductor contact surface of the charging gun head, and an isomorphic sensing ring is added at a predetermined distance in the axial depth of the conductor to form a three-dimensional monitoring system; temperature sensors are arranged around the outer periphery of the conductor contact area on the bonding surface between the insulating outer shell and the metal shell of the charging gun head. Miniature temperature sensors are uniformly embedded in the circumferential direction at the crimping point between the cable shielding layer and the gun head metal adapter; sensors are arranged on both sides of the welding point between the inner copper conductor of the cable core and the inner conductor of the gun head; a flow-temperature composite sensor is installed inside the liquid cooling channel inlet of the connection node. Along the length of the cable, a predetermined length is designated as a monitoring unit. Distributed fiber optic temperature sensors are deployed inside the shielding layer, in the middle of the insulation layer, and on the surface of the inner conductor of each unit, forming a three-line parallel axial sensing chain. Redundant temperature sensors are added at the midpoint and both ends of the entire cable.

[0009] Preferably, the temperature sensing array includes multiple distributed temperature sensors, which are placed in predetermined positions and further include: High-precision temperature sensors are installed at the main inlet and main outlet of the liquid cooling system; temperature sensors are installed before and after each branch valve that branches the flow to the nozzle and cable; and temperature sensors are installed at the inlet and outlet of the heat exchanger on the heat dissipation circuit of the cooling medium. Temperature sensors are placed on the heat dissipation substrates of the IGBTs and rectifier bridge power devices in the charging module, corresponding to the center position of each chip; temperature sensors are also placed in the inlet and outlet chambers of the liquid cooling cavity inside the module.

[0010] Preferably, the step of acquiring the raw temperature data of the temperature sensing array in real time through the temperature acquisition module and transmitting it to the temperature field reconstruction module specifically includes: The temperature acquisition module collects raw temperature data from all distributed sensors and performs multi-level preprocessing, including removing outliers that exceed the sensor's range, marking sensors with multiple consecutive anomalies as faulty and enabling redundant backup data; using a 5th-order moving average filtering algorithm to smooth high-frequency noise, and superimposing Kalman filtering on dynamic heating areas; binding preset spatial coordinates based on sensor IDs and synchronizing the acquisition times of different sensors through timestamps. The preprocessed temperature data is transmitted to the temperature field reconstruction module via industrial Ethernet and fiber optic cable. The transmission frame format includes sensor ID, three-dimensional coordinates, temperature value, acquisition timestamp, and data confidence level. Data with a transmission delay of more than 50ms is marked as invalid and a retransmission mechanism is triggered.

[0011] Preferably, the step of processing the original temperature data based on a spatial interpolation algorithm to construct a three-dimensional dynamic temperature field model of the key heat-generating region, and outputting the temperature values, temperature gradients, and temperature difference distributions of each region in real time, specifically includes: The three-dimensional spatial mesh modeling and temperature field reconstruction module is based on the CAD model of the key heat-generating area, constructs a three-dimensional spatial mesh, and performs mesh division; maps the actual deployment location of the sensor to the mesh node, assigns adiabatic boundary conditions to the boundaries of the gun head shell and cable insulation layer, and assigns convective heat transfer boundary conditions to the liquid cooling channel wall. An improved spatial interpolation algorithm is used to fill the grid temperature values ​​using an improved Kriging interpolation algorithm that incorporates thermal characteristic weights. Spherical variograms and exponential variograms are defined for different thermal characteristics of metallic conductors and insulating materials, respectively. The weights are dynamically adjusted according to the heat dissipation power of the sensor's region. Constraints are set on the temperature difference between adjacent grid points. The dynamic temperature field is updated in real time. The temperature field reconstruction module performs a full mesh update periodically, corrects the mesh node temperature based on the latest collected data, and calculates the temperature gradient vector of each node using the finite difference method. For areas with temperature gradients greater than 2℃ / mm, the calculation is refined to generate local high temperature gradient cloud maps. Areas with temperature differences exceeding the threshold are automatically identified and marked. Output a standardized dataset of the three-dimensional dynamic temperature field model, including the highest, lowest, and average temperatures of each key region; the temperature difference matrix between any two points; a thermogram of the temperature gradient distribution; and a temperature field prediction curve.

[0012] Preferably, the step of transmitting the output data of the three-dimensional dynamic temperature field model to the thermal equilibrium control module, wherein the thermal equilibrium control module presets a temperature threshold range and a temperature difference threshold, specifically including: A three-dimensional dynamic temperature field model transmits standardized data frames to the thermal equilibrium control module via an industrial bus, including the highest temperature value and coordinates of each key area; the temperature difference matrix between adjacent areas; the coordinate set of areas where the temperature gradient exceeds the threshold; and temperature field prediction data. The control module performs CRC verification on the received data; if the verification fails, it triggers immediate retransmission. The control module parses the data according to priority. The first-level data is the temperature of the inner conductor of the gun head and the temperature of the connection node; the second-level data is the axial temperature difference distribution of the cable and the temperature difference between the inlet and outlet of the liquid cooling circuit; the third-level data is the temperature field prediction curve and the overall temperature difference distribution statistics.

[0013] Preferably, the preset temperature threshold range and temperature difference threshold of the thermal equilibrium control module specifically include: Threshold grading settings: The thermal equalization control module has multiple built-in threshold parameter tables, which are dynamically switched according to the charging power. The thresholds are divided into absolute temperature thresholds, such as the inner conductor contact surface of the gun head, the connection node between the cable and the gun head, and the liquid cooling circuit inlet; temperature difference thresholds, such as the circumferential temperature difference between the gun head and the connection node, the same cross section of the cable, and the inlet and outlet temperature difference of the liquid cooling circuit; and gradient thresholds, such as the temperature gradient of any adjacent grid points. Threshold dynamic correction: Based on ambient temperature sensor data, the threshold is corrected in real time. When the ambient temperature is higher than the predetermined temperature, the maximum temperature threshold of the nozzle and the connection node is lowered; when the ambient temperature is lower than 0℃, the minimum temperature threshold is raised.

[0014] Preferably, the thermal equilibrium control module compares the output data of the three-dimensional dynamic temperature field model with a preset threshold. If the temperature value of a certain area exceeds the maximum temperature threshold, or the temperature difference between two adjacent key areas exceeds the temperature difference threshold, then it sends a control command to the partitioned cooling execution unit of the liquid cooling system. Specifically, this includes: Threshold comparison: Check whether the absolute temperature of each key area exceeds the corresponding threshold range, mark the over-temperature area and the over-temperature range; calculate the temperature difference between adjacent key areas to determine whether it exceeds the temperature difference threshold; scan the grid area with excessive temperature gradient, if there is an area with a gradient greater than 2℃ / mm for 3 consecutive cycles, it is determined to be a local heat concentration. The control module generates corresponding control commands based on the anomaly level. Level 1 anomalies only adjust the cooling circuit flow rate of the corresponding area, with the adjustment range being a predetermined proportion of the base flow rate. Level 2 anomalies simultaneously adjust the cooling circuits of the target area and adjacent areas, increasing the main circuit flow rate by a predetermined proportion and activating the overclocking mode of the variable frequency water pump in the corresponding area. Level 3 anomalies activate the emergency cooling mode, adjusting the flow rate of the target area to the maximum value, running the main circuit water pump at full power, and simultaneously sending a power reduction request to the charging main control module. The control command adopts a standardized frame structure, which includes the target branch ID, target flow, adjustment duration, and execution priority. The partition cooling execution unit returns the execution status within 10ms after receiving the command. If the control module does not receive feedback within 30ms, it will resend the command and mark the branch as a suspected fault. After the command is executed, the control module tracks the temperature field data for the next cycle. If the deviation between the measured temperature in the over-threshold area and the preset threshold decreases by less than a predetermined proportion, the predetermined proportion of flow rate adjustment is added until the temperature returns to the threshold range.

[0015] Preferably, the repeated steps described above, until the temperature values ​​of all key heat-generating regions in the three-dimensional dynamic temperature field model are within a preset temperature threshold range, and the temperature difference between adjacent key regions is less than or equal to the temperature difference threshold, specifically include: With a control cycle of 100ms, temperature acquisition, field model reconstruction, threshold comparison, control command output, and execution feedback are executed sequentially within each cycle. When the system is normal and all parameters are within the threshold, the basic sampling frequency and control cycle are maintained. When a level 1 anomaly is detected, the cycle is automatically shortened to 50ms to increase the control density. Under level 3 anomaly conditions, a high-frequency response mode is triggered until the anomaly level drops to level 2 or below.

[0016] Active control can be stopped and the system switched to steady-state monitoring mode when the following conditions are met: within three consecutive control cycles, the highest temperature of all key areas is less than or equal to the preset highest threshold, and the lowest temperature is greater than or equal to the preset lowest threshold. The temperature difference between adjacent key areas is less than or equal to the temperature difference threshold for five consecutive cycles; there are no grid areas with gradient exceeding the limit in the temperature field model, and the temperature prediction curve shows no risk of exceeding the threshold within the next 2 seconds.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention utilizes a multi-dimensional temperature sensing array to deploy distributed sensors at predetermined locations in key heat-generating areas such as the charging gun head and cable connection nodes, achieving three-dimensional coverage of the core area. The multi-layer ring sensor on the inner conductor contact surface of the charging gun head and the axial sensing chain in the middle section of the cable can simultaneously collect temperature data from different spatial locations, providing raw data support for subsequent temperature field reconstruction, solving the problem of blind spots in local temperature monitoring, and avoiding overheating misjudgments caused by single-point data deviations. By processing the raw data using spatial interpolation algorithms, a three-dimensional dynamic temperature field model is constructed that can output the temperature values, temperature gradients, and temperature difference distributions of each region in real time. Compared with traditional static temperature measurement methods, it can intuitively present the temperature difference changes between the gun head and the cable connection node. Moreover, the model is updated every 100ms, ensuring rapid capture of temperature fluctuations. This provides a visualized and predictable basis for thermal equilibrium control, solving the limitations of existing technologies where the temperature field distribution is fuzzy and the temperature difference trend cannot be predicted. The thermal balance control module presets differentiated threshold ranges and temperature difference thresholds. After comparing temperature field data, it sends control commands according to the over-temperature amplitude and temperature difference level, avoiding a one-size-fits-all, coarse adjustment. At the same time, the three independent cooling branches of the zoned cooling execution unit (corresponding to the nozzle, connection node, and cable midsection) are each equipped with an independent electromagnetic flow valve and variable frequency water pump. This allows for precise flow adjustment for temperature anomalies in different areas. When the nozzle overheats, the flow of the corresponding branch is increased separately without having to link the entire cooling system. This reduces energy waste and improves the adjustment response speed, solving the problems of insufficient control precision and difficulty in multi-area coordination in existing technologies.

[0018] By cyclically executing the monitoring, modeling, control, and feedback process until the temperature of all key areas is within the threshold and the temperature difference between adjacent areas is less than or equal to the set value, a complete closed-loop control link is formed. Even if disturbances such as sudden changes in charging power or fluctuations in ambient temperature occur, the system can maintain temperature stability by shortening the control cycle and pre-adjusting the flow rate, thus avoiding the risk of local overheating. At the same time, the emergency handling when abnormal convergence occurs improves the operational reliability and safety of the liquid-cooled charging module. Attached Figure Description

[0019] Figure 1 This is a flowchart of a thermal equalization control method for a liquid-cooled charging module. Detailed Implementation

[0020] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0021] Reference Figure 1 As shown, a thermal equalization control method for a liquid-cooled charging module includes: A multi-dimensional temperature sensing array is deployed in the key heat-generating area of ​​the liquid-cooled charging module. The temperature sensing array contains multiple distributed temperature sensors placed in predetermined positions. The temperature acquisition module acquires the raw temperature data of the temperature sensing array in real time and transmits it to the temperature field reconstruction module. The raw temperature data is processed based on the spatial interpolation algorithm to construct a three-dimensional dynamic temperature field model of the key heat-generating area and output the temperature value, temperature gradient and temperature difference distribution of each area in real time. The output data of the three-dimensional dynamic temperature field model is transmitted to the thermal equilibrium control module, which presets a temperature threshold range and a temperature difference threshold. The thermal equilibrium control module compares the output data of the three-dimensional dynamic temperature field model with the preset threshold. If the temperature value of a certain area exceeds the maximum temperature threshold or the temperature difference between two adjacent key areas exceeds the temperature difference threshold, the thermal equilibrium control module sends a control command to the partitioned cooling execution unit of the liquid cooling system. The partitioned cooling execution unit includes at least three independent cooling branches, corresponding to the charging gun head, cable connection node, and cable mid-section, respectively. Each cooling branch is equipped with an independent electromagnetic flow valve and a variable frequency water pump. Repeat the above steps until the temperature values ​​of all key heating areas in the three-dimensional dynamic temperature field model are within the preset temperature threshold range, and the temperature difference between adjacent key areas is less than or equal to the temperature difference threshold.

[0022] It should be noted that the process involves real-time acquisition of raw data by a multi-dimensional temperature sensor array (sensing layer), spatial information fusion and three-dimensional dynamic field construction by a temperature field reconstruction module (data processing and modeling layer), state judgment and generation of control commands based on preset thresholds by a thermal equilibrium control module (decision layer), and precise flow / pressure regulation by a zoned cooling execution unit (execution layer). The final effect is then fed back to the system through the temperature sensor array. This closed-loop architecture ensures the real-time performance, accuracy, and adaptability of the control.

[0023] Distributed multi-sensor arrays, which deploy multiple sensors in "critical heat-generating areas" rather than relying on a few points, are the foundation for achieving precise thermal management; they can capture local hotspots, temperature gradients, and spatial differences.

[0024] Three-dimensional dynamic temperature field reconstruction uses spatial interpolation algorithms to transform discrete point temperature data into a continuous and visualized three-dimensional temperature field model. This not only outputs the temperature values ​​of each region in real time, but more importantly, it can calculate the temperature gradient (reflecting the direction and intensity of heat transfer) and temperature difference distribution (identifying areas of thermal imbalance), providing a global thermal state picture that far exceeds single-point information for control decisions.

[0025] The independent cooling circuit design sets up at least three independent and controllable cooling circuits for key sub-regions with different thermal characteristics in the charging module (gun head - contact resistance / arc heat, connection node - contact resistance / concentrated loss heat, cable mid-section - ohmic loss heat), and monitors the differences in heat generation mechanism, heat dissipation path and heat capacity of different parts.

[0026] The flexible actuators, including the electromagnetic flow valve which provides rapid flow on / off and regulation capabilities, and the variable frequency water pump which can flexibly change the system pressure and total / branch basic flow, combine to give the system the ability to make high-precision and rapid dynamic adjustments to coolant flow and pressure. This provides the hardware guarantee for achieving zoned differentiated cooling, such as instantly increasing the flow rate for overheated nozzles and maintaining or reducing the flow rate for cable sections with moderate temperatures.

[0027] The temperature sensing array comprises multiple distributed temperature sensors, which are placed in predetermined locations, specifically including: Patches of temperature sensors are arranged at intervals along the circumference of the inner conductor contact surface of the charging gun head, and an isomorphic sensing ring is added at a predetermined distance in the axial depth of the conductor to form a three-dimensional monitoring system; temperature sensors are arranged around the outer periphery of the conductor contact area on the bonding surface between the insulating outer shell and the metal shell of the charging gun head. Miniature temperature sensors are uniformly embedded in the circumferential direction at the crimping point between the cable shielding layer and the gun head metal adapter; sensors are arranged on both sides of the welding point between the inner copper conductor of the cable core and the inner conductor of the gun head; a flow-temperature composite sensor is installed inside the liquid cooling channel inlet of the connection node. Along the length of the cable, a predetermined length is designated as a monitoring unit. Distributed fiber optic temperature sensors are deployed inside the shielding layer, in the middle of the insulation layer, and on the surface of the inner conductor of each unit, forming a three-line parallel axial sensing chain. Redundant temperature sensors are added at the midpoint and both ends of the entire cable.

[0028] High-precision temperature sensors are installed at the main inlet and main outlet of the liquid cooling system; temperature sensors are installed before and after each branch valve that branches the flow to the nozzle and cable; and temperature sensors are installed at the inlet and outlet of the heat exchanger on the heat dissipation circuit of the cooling medium. Temperature sensors are placed on the heat dissipation substrates of the IGBTs and rectifier bridge power devices in the charging module, corresponding to the center position of each chip; temperature sensors are also placed in the inlet and outlet chambers of the liquid cooling cavity inside the module.

[0029] It should be noted that in the conductor contact area: circumferential and axial layered sensing rings (patch-type sensors) are used on the conductor contact surface inside the gun head to form a three-dimensional monitoring grid; this captures local overheating caused by contact resistance, circumferential temperature difference caused by uneven current distribution, and thermal conduction attenuation characteristics in the conductor's longitudinal direction; axial layered monitoring can identify whether heat is abnormally diffused to the insulation layer. Welding points / crimping points are placed on both sides of the welding point and at the crimping part of the shielding layer to monitor the microscopic thermal changes at the metal connection interface, such as sudden changes in contact resistance caused by poor welding and local Joule heating caused by loose crimping. The temperature difference on both sides reflects the direction of heat diffusion.

[0030] Sensors are placed between the insulating shell and the metal shell on the bonding surface of the outer shell to monitor the structural stress caused by the difference in the thermal expansion coefficients of the materials (excessive temperature difference may cause cracking), and at the same time indirectly assess the aging state of the insulating material (abnormal temperature indicates partial discharge or thermal failure). The cable's three-wire parallel sensing chain (shielding layer / insulation layer / inner conductor) includes the shielding layer, which monitors eddy current loss heating; the middle of the insulation layer, which detects local insulation defects (such as uneven thermal conduction caused by air bubbles); and the inner conductor, which reflects the current-carrying temperature rise and heat dissipation efficiency. The axial temperature gradient of the three layers can be used to construct a cable heat transfer model.

[0031] The fluid thermal management closed-loop control calculates the actual heat dissipation of the branch (Q=cmΔT) by using composite temperature and flow sensors before and after the valve, and dynamically calibrates the cooling efficiency by combining flow data, and identifies pipeline blockage or valve failure. By using sensors at the inlet / outlet and heat exchanger, a system-level thermal balance equation is established to assess the performance degradation of the radiator.

[0032] The step of acquiring the raw temperature data of the temperature sensing array in real time through the temperature acquisition module and transmitting it to the temperature field reconstruction module specifically includes: The temperature acquisition module collects raw temperature data from all distributed sensors and performs multi-level preprocessing, including removing outliers that exceed the sensor's range, marking sensors with multiple consecutive anomalies as faulty and enabling redundant backup data; using a 5th-order moving average filtering algorithm to smooth high-frequency noise, and superimposing Kalman filtering on dynamic heating areas; binding preset spatial coordinates based on sensor IDs and synchronizing the acquisition times of different sensors through timestamps. The preprocessed temperature data is transmitted to the temperature field reconstruction module via industrial Ethernet and fiber optic cable. The transmission frame format includes sensor ID, three-dimensional coordinates, temperature value, acquisition timestamp, and data confidence level. Data with a transmission delay of more than 50ms is marked as invalid and a retransmission mechanism is triggered.

[0033] The three-dimensional spatial mesh modeling and temperature field reconstruction module is based on the CAD model of the key heat-generating area, constructs a three-dimensional spatial mesh, and performs mesh division; maps the actual deployment location of the sensor to the mesh node, assigns adiabatic boundary conditions to the boundaries of the gun head shell and cable insulation layer, and assigns convective heat transfer boundary conditions to the liquid cooling channel wall. An improved spatial interpolation algorithm is used to fill the grid temperature values ​​using an improved Kriging interpolation algorithm that incorporates thermal characteristic weights. Spherical variograms and exponential variograms are defined for different thermal characteristics of metallic conductors and insulating materials, respectively. The weights are dynamically adjusted according to the heat dissipation power of the sensor's region. Constraints are set on the temperature difference between adjacent grid points. The dynamic temperature field is updated in real time. The temperature field reconstruction module performs a full mesh update periodically, corrects the mesh node temperature based on the latest collected data, and calculates the temperature gradient vector of each node using the finite difference method. For areas with temperature gradients greater than 2℃ / mm, the calculation is refined to generate local high temperature gradient cloud maps. Areas with temperature differences exceeding the threshold are automatically identified and marked. Output a standardized dataset of the three-dimensional dynamic temperature field model, including the highest, lowest, and average temperatures of each key region; the temperature difference matrix between any two points; a thermogram of the temperature gradient distribution; and a temperature field prediction curve.

[0034] A three-dimensional dynamic temperature field model transmits standardized data frames to the thermal equilibrium control module via an industrial bus, including the highest temperature value and coordinates of each key area; the temperature difference matrix between adjacent areas; the coordinate set of areas where the temperature gradient exceeds the threshold; and temperature field prediction data. The control module performs CRC verification on the received data; if the verification fails, it triggers immediate retransmission. The control module parses the data according to priority. The first-level data is the temperature of the inner conductor of the gun head and the temperature of the connection node; the second-level data is the axial temperature difference distribution of the cable and the temperature difference between the inlet and outlet of the liquid cooling circuit; the third-level data is the temperature field prediction curve and the overall temperature difference distribution statistics.

[0035] It should be noted that data acquisition and preprocessing are as follows: At the hardware level, physically unreachable values ​​(such as -50℃ or 300℃) are directly eliminated based on the ADC range threshold; at the logic level, invalid jumps with a sudden change of >50% are judged based on a historical sliding window (such as 10 cycles); at the system level, sensor failure is marked after 3 consecutive abnormalities, and the system automatically switches to a redundant node (such as a backup point at the midpoint of the cable). Dynamic filtering strategy: In the steady-state region (temperature rise rate < 1℃ / s), a 5th-order moving average filter (window width 0.5s) is used to suppress random noise. In the transient region (such as the instant of IGBT switching), Kalman filtering is used to fuse the heat conduction model predictions to track drastic changes at the 200℃ / s level.

[0036] Spatiotemporal synchronization, hardware timestamp, adopts IEEE1588 precision clock protocol to compress the sampling time synchronization accuracy to ≤100μs; coordinate binding optimization, in order to cope with micro displacements caused by mechanical vibration (such as gun head insertion and removal), establishes an elastic mapping relationship between sensor ID and grid coordinates, allowing ±2mm dynamic offset compensation.

[0037] Temperature field reconstruction, 3D mesh modeling:

[0038] Improved Kriging Algorithm: Thermal characteristic weighted interpolation is used, and the variogram function is selected according to the material type. The spherical function fits the rapid thermal diffusion of metals, and the exponential function matches the slowly varying temperature field of insulators. A heat dissipation power weight factor is introduced to calculate the power consumption ratio of the current region, and the search radius is increased in the high power consumption region. Temperature gradient constraints are applied, and a gradient threshold is introduced to increase the smoothness of interpolation. Adaptive computing with a base update rate of 100Hz full grid refresh (10ms cycle). High temperature gradient region triggering: when ||∇T||>2℃ / mm, the local mesh is refined to 0.2mm, and the calculation frequency is increased to 1kHz; Cloud map generation is optimized by using GPU parallel computing to control the gradient cloud map rendering latency to within 5ms.

[0039] Key design considerations for the transmission protocol:

[0040] Control module three-level response: Level 1 data → If the nozzle conductor temperature is >150℃, an emergency response will be initiated, with the solenoid valve fully open and the water pump frequency increased. Level 1 data → If the connection node ΔT > 10℃, then an emergency response is initiated, the solenoid valve is fully opened and the water pump frequency is increased; Secondary data → If the cable axial ΔT > 5°C, then adjust the gradient to adjust the flow distribution of the cable segment; Level 3 data → If the predicted temperature rise rate is >50℃ / s, then preventative power reduction should be implemented.

[0041] The preset temperature threshold range and temperature difference threshold of the thermal equilibrium control module specifically include: Threshold grading settings: The thermal equalization control module has multiple built-in threshold parameter tables, which are dynamically switched according to the charging power. The thresholds are divided into absolute temperature thresholds, such as the inner conductor contact surface of the gun head, the connection node between the cable and the gun head, and the liquid cooling circuit inlet; temperature difference thresholds, such as the circumferential temperature difference between the gun head and the connection node, the same cross section of the cable, and the inlet and outlet temperature difference of the liquid cooling circuit; and gradient thresholds, such as the temperature gradient of any adjacent grid points. Threshold dynamic correction: Based on ambient temperature sensor data, the threshold is corrected in real time. When the ambient temperature is higher than the predetermined temperature, the maximum temperature threshold of the nozzle and the connection node is lowered; when the ambient temperature is lower than 0℃, the minimum temperature threshold is raised.

[0042] Threshold comparison: Check whether the absolute temperature of each key area exceeds the corresponding threshold range, mark the over-temperature area and the over-temperature range; calculate the temperature difference between adjacent key areas to determine whether it exceeds the temperature difference threshold; scan the grid area with excessive temperature gradient, if there is an area with a gradient greater than 2℃ / mm for 3 consecutive cycles, it is determined to be a local heat concentration. The control module generates corresponding control commands based on the anomaly level. Level 1 anomalies only adjust the cooling circuit flow rate of the corresponding area, with the adjustment range being a predetermined proportion of the base flow rate. Level 2 anomalies simultaneously adjust the cooling circuits of the target area and adjacent areas, increasing the main circuit flow rate by a predetermined proportion and activating the overclocking mode of the variable frequency water pump in the corresponding area. Level 3 anomalies activate the emergency cooling mode, adjusting the flow rate of the target area to the maximum value, running the main circuit water pump at full power, and simultaneously sending a power reduction request to the charging main control module. The control command adopts a standardized frame structure, which includes the target branch ID, target flow, adjustment duration, and execution priority. The partition cooling execution unit returns the execution status within 10ms after receiving the command. If the control module does not receive feedback within 30ms, it will resend the command and mark the branch as a suspected fault. After the command is executed, the control module tracks the temperature field data for the next cycle. If the deviation between the measured temperature in the over-threshold area and the preset threshold decreases by less than a predetermined proportion, the predetermined proportion of flow rate adjustment is added until the temperature returns to the threshold range.

[0043] It should be noted that the multi-dimensional threshold is:

[0044] Absolute temperature threshold: 175±5℃ for the contact surface of the gun head conductor, based on the recrystallization temperature of CuCrZr alloy (210℃), with a 15% safety margin to prevent high-temperature softening from causing a doubling of contact resistance; The cable connection node (160±8℃) is based on the liquidus temperature of the tin-silver-copper brazing filler metal at the solder joint (220℃) to avoid impedance sudden change caused by the melting of the brazing filler metal; The inlet of the liquid cooling circuit (65±3℃) is below the boiling point of ethylene glycol-based coolant (75℃@1.5Bar) to prevent cavitation from damaging the pump body. Temperature difference threshold: ΔT ≤ 7℃, the thermal expansion difference between the copper conductor (17ppm / ℃) and the aluminum alloy shell (23ppm / ℃) is controlled within 0.008mm / mm, avoiding the fatigue limit of shear stress exceeding 125MPa; The circumferential temperature of the cable is ΔT≤5℃ to ensure the uniform expansion of the multi-stranded conductor and prevent local tearing of the insulation layer (tensile strength of silicone rubber <8MPa). Gradient threshold: 2℃ / mm in the metal region: exceeding this value will induce copper grain boundary slip and accelerate creep fracture (refer to the Larson-Miller parametric model). When the insulation layer temperature gradient is 1℃ / mm and the gradient is >1.2℃ / mm, the vulcanization aging rate of silicone rubber increases by 300% (verified by the Arrhenius equation).

[0045] Dynamic correction mechanism, ambient temperature compensation model:

[0046] In the formula, This is the adjusted maximum temperature; The reference temperature is ; k is a correction factor used to adjust the threshold according to changes in ambient temperature. This refers to the ambient temperature, i.e., the current external temperature. The reference temperature is the base temperature used to calculate the correction factor. High temperature correction ( >40℃), coefficient k=0.012, the threshold is reduced by 1.2% for every 1℃ increase in temperature (the positive temperature coefficient of copper resistivity of 0.4% / ℃ leads to a doubling of heat generation). Low temperature protection ( <0℃), threshold increase minimum value This offsets the heat transfer attenuation caused by the increase in coolant viscosity (viscosity spikes by 300% at -20℃). Power adaptive switching logic, with built-in parameter table categorized by power level: 200kW setting: T_max = 175℃ (base value); 480kW range: T_max=165℃ (8℃ decrease for every 100kW increase, matching the I²R loss curve); 800kW setting: T_max=152℃ (with thermal shock compensation for silicon carbide devices).

[0047] A three-level response mechanism is used, with the first level being an anomaly (local adjustment). The target branch flow rate is ±(15-25%), and the trigger condition is a single-point over-temperature of ≤10℃ without diffusion. PID closed-loop control is employed, with an adjustment time of 50ms. Level 2 anomaly (regional coordination): Main flow rate increases by 30-50%, adjacent branch line coupling adjustment ±20%, water pump overfrequency 5%; response conditions: temperature over-temperature 10-20℃ or thermal gradient lasting for 3 cycles. Level 3 anomaly (system-level intervention): Target area flow rate 200%+, main pump full power 120%, synchronously send power reduction request (slope 120kW / s); Blocking conditions: over-temperature > 20℃ or gradient > 3℃ / mm.

[0048] Verification of the control effect, defining temperature drop efficiency:

[0049] If η < 0.5: Add flow rate

[0050] If η < 0.3 twice consecutively: trigger fault diagnosis, indicating valve body jamming or sensor failure; In the formula, Temperature drop efficiency represents the efficiency with which the temperature decreases. This is the temperature value measured for the nth time; This is the temperature value measured for the (n+1)th time. The set target temperature; For old traffic; Command transmission: Standardized command frame design, 28-byte enhanced structure, target branch ID (1 byte) is the high 4 bits of the encoded area (0001=head / 0010=node), and the low 4 bits are the device type; The target flow rate (4 bytes) is in IEEE 754 floating-point format with a resolution of 0.01 L / min; The execution priority (1 byte) is graded from 0 to 254. When it is greater than 200, the hardware interrupt preemption mechanism is triggered. Dynamic CRC32 checksum (4 bytes) is used to prevent replay attacks by incorporating timestamps. Three-tier protection for fault diagnosis: Communication layer: No response within 30ms → Automatic retransmission (3 failures are marked with a 0xF0 fault code); Execution layer: Flow feedback deviation > 15% → activate high-frequency micro-vibration (50μm amplitude to break the jam); Effect layer: Continuous adjustment is ineffective → switch to backup cooling circuit + upload black box data.

[0051] The partitioned cooling execution unit includes at least three independent cooling branches, corresponding to the charging gun head, cable connection node, and cable mid-section, respectively. Each cooling branch is equipped with an independent electromagnetic flow valve and a variable frequency water pump. It should be noted that the charging gun head cooling circuit is divided as follows: The electromagnetic flow valve adopts a high-frequency response proportional valve (response time ≤8ms), with a flow range of 0.5-8L / min, and supports pulse modulation to prevent scaling; the valve body integrates a 25Bar pressure-resistant sealing structure, meets 3000V insulation requirements, and the valve core is coated with diamond-like carbon (DLC) to reduce friction loss. Variable frequency water pump, brushless DC magnetic drive pump with power of 50-200W, PWM speed regulation accuracy of ±1%, adopts shaftless impeller design, and has built-in flow-temperature composite sensor to achieve microsecond-level flow closed-loop feedback. Cooling channels, three-dimensional spiral microchannels (channel depth 0.8mm / wall thickness 1.2mm) are embedded in the metal cavity of the gun head, and the surface is sprayed with a nano-zirconia ceramic coating to improve the heat exchange rate by 30%. Thermal grease (thermal conductivity 8W / mK) is filled between the channels and the conductor to eliminate contact thermal resistance. Cable connection node cooling branch: electromagnetic flow valve, dual redundant parallel valve group (main valve + backup valve), leakage class Class VI (≤3 bubbles / min). The valve body uses a copper-nickel alloy shell, which improves its resistance to sulfide corrosion. The variable frequency water pump has a vibration-resistant structure that has passed the 5G@200Hz vibration test. The impeller edge is embedded with a micro-serrated structure to effectively prevent fiber impurities from entangled in the coolant. Cooling implementation scheme: conductor crimping area, copper-based micro-needle rib array for heat dissipation (needle diameter 0.3mm / spacing 0.5mm, heat dissipation density 500W / cm²); solder joint, liquid metal thermal pad (gallium indium tin alloy, thermal resistance <0.05K / W) to fill gaps; Mid-section cooling branch of cable: Electromagnetic flow valve with low pressure loss design (ΔP < 0.2 Bar @ 5 L / min) and supports CANFD bus control. The valve body flow channel is CFD optimized, reducing turbulence intensity by 45%. Variable frequency water pumps and long-stroke plunger pumps adapt to cable bending and deformation, and have a built-in flow-resistance curve compensation algorithm to dynamically adjust the output pressure. Cooling channel, sandwich flow channel structure: Outer layer: 2mm shielding layer cooling channel (spiral guide groove to enhance heat exchange); Middle layer: Insulating layer filled with thermally conductive silicone (thermal conductivity 4W / mK); Inner layer: 1.5mm conductor direct cooling cavity (micro-turbulence column design); Flexible connection, the corrugated pipe adopts multi-layer stainless steel + Kevlar braided layer, with a minimum bending radius of 8 times the pipe diameter.

[0052] The repeated steps described above, until the temperature values ​​of all key heat-generating regions in the three-dimensional dynamic temperature field model are within the preset temperature threshold range, and the temperature difference between adjacent key regions is less than or equal to the temperature difference threshold, specifically include: With a control cycle of 100ms, temperature acquisition, field model reconstruction, threshold comparison, control command output, and execution feedback are executed sequentially within each cycle. When the system is normal and all parameters are within the threshold, the basic sampling frequency and control cycle are maintained. When a level 1 anomaly is detected, the cycle is automatically shortened to 50ms to increase the control density. Under level 3 anomaly conditions, a high-frequency response mode is triggered until the anomaly level drops to level 2 or below.

[0053] Active control can be stopped and the system switched to steady-state monitoring mode when the following conditions are met: within three consecutive control cycles, the highest temperature of all key areas is less than or equal to the preset highest threshold, and the lowest temperature is greater than or equal to the preset lowest threshold. The temperature difference between adjacent key areas is less than or equal to the temperature difference threshold for five consecutive cycles; there are no grid areas with gradient exceeding the limit in the temperature field model, and the temperature prediction curve shows no risk of exceeding the threshold within the next 2 seconds.

[0054] It should be noted that the base cycle (100ms) executes a total of 5 stages under steady state: Temperature acquisition (0-20ms): Simultaneously reads data from 128 sensors, using TDMA protocol to avoid signal collisions; Field model reconstruction (20-70ms): Reconstruct the temperature field based on an improved FFT algorithm (1 million grid points, accuracy ±0.5℃). Threshold comparison (70-85ms): Parallel comparison of 3 types of thresholds (absolute temperature / temperature difference / gradient); Command output (85-95ms): Generates a standardized control frame (28 bytes, CRC32 checksum). Execution feedback (95-100ms): Verify valve action and flow rate compliance.

[0055] Level 1 response (50ms cycle), triggered by single-area over-temperature ≤10℃ or ΔT>th threshold, compressing field reconstruction time to 30ms (only updating hot spot area mesh), enabling FPGA hardware acceleration threshold comparison (time reduced to 5ms), and enabling high-priority CANFD channel (baud rate 5Mbps) for instruction transmission. Level 3 emergency response (10ms cycle), triggered by gradient > 3℃ / mm or temperature rise rate > 100℃ / s, sensor sampling switches to parallel ADC mode (5 channels synchronized, time taken 0.8ms), pre-compiled control instruction library is enabled (hash retrieval time taken 0.1ms), valve and pump control adopts PWM hard interrupt (response delay < 0.5ms).

[0056] Steady-state transition: Absolute temperature stability, satisfying the following for 3 consecutive cycles: It can cover the sensor error band (±0.2℃) and the steady-state error of the control system; Thermal uniformity verification, temperature difference across 5 consecutive cycles: ; In the formula, The temperature difference between adjacent critical regions i and j; The maximum allowable temperature difference threshold; ensure that the thermal stress is below the yield strength of the aluminum alloy (125 MPa); The risk prediction LSTM prediction model must meet the following requirements: ; In the formula, It is the derivative of temperature T with time t, i.e., the rate of temperature change; The time interval is set to 2 seconds; historical temperature rise curves are integrated with the rate of change of ambient temperature. Degradation protection strategy with a step-by-step exit mechanism: In the first cycle, the sampling rate is reduced to 50% (redundant sensors are turned off); in the second cycle, local mesh encryption calculation is stopped; in the final cycle, the control chip switches to low-power mode (power consumption <3W). The state rollback logic states that if any indicator in the steady-state monitoring exceeds the limit, the control cycle will be restored instantly to 100ms, triggering the second-level response pre-cooling mode (flow increased by 30%).

[0057] Safety Boundaries and Failure Prevention Data filtering employs sliding window variance detection (window size = 10 periods), with outlier removal rules as follows:

[0058] In the formula, The temperature value of the k-th data point; This represents the average value of the temperature data. The standard deviation of the temperature data; This is the second derivative of temperature with respect to time, i.e., the rate of change of the rate of temperature change; Sensor fault tolerance: In case of single-point failure, data is reconstructed based on Kriging interpolation (error < 1.2℃); in case of area failure, a conservative control strategy is activated (flow rate is automatically increased to a safe baseline). Limit protection, infinite loop blocking, 10 consecutive ineffective adjustments → trigger system reset (error code 0x7F) Judgment basis: ;

[0059] In the formula, The temperature value of the k-th data point; The target temperature is set; this formula represents the sum of the absolute values ​​of the differences between the measured temperature Tk and the set temperature Tset in 10 consecutive measurements. The cooling failure contingency plan sends a power reduction request if the temperature continues to exceed the limit. If there is no response within 3 seconds, the contactor is hard-cut off to achieve forced circulation of coolant. The contactor action time is <20ms (compliant with IEC60947 standard).

[0060] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A thermal equalization control method for a liquid-cooled charging module, characterized in that, include: A multi-dimensional temperature sensing array is deployed in the key heat-generating area of ​​the liquid-cooled charging module. The temperature sensing array contains multiple distributed temperature sensors placed in predetermined positions. The temperature acquisition module acquires the raw temperature data of the temperature sensing array in real time and transmits it to the temperature field reconstruction module. The raw temperature data is processed based on the spatial interpolation algorithm to construct a three-dimensional dynamic temperature field model of the key heat-generating area and output the temperature value, temperature gradient and temperature difference distribution of each area in real time. The output data of the three-dimensional dynamic temperature field model is transmitted to the thermal equilibrium control module, which presets a temperature threshold range and a temperature difference threshold. The thermal equilibrium control module compares the output data of the three-dimensional dynamic temperature field model with the preset threshold. If the temperature value of a certain area exceeds the maximum temperature threshold or the temperature difference between two adjacent key areas exceeds the temperature difference threshold, it sends a control command to the partitioned cooling execution unit of the liquid cooling system. The partitioned cooling execution unit includes at least three independent cooling branches, corresponding to the charging gun head, cable connection node, and cable mid-section, respectively. Each cooling branch is equipped with an independent electromagnetic flow valve and a variable frequency water pump. Repeat the above steps until the temperature values ​​of all key heating areas in the three-dimensional dynamic temperature field model are within the preset temperature threshold range, and the temperature difference between adjacent key areas is less than or equal to the temperature difference threshold.

2. The thermal equalization control method for a liquid-cooled charging module according to claim 1, characterized in that, The temperature sensing array comprises multiple distributed temperature sensors, which are placed in predetermined locations, specifically including: Patches of temperature sensors are arranged at intervals along the circumference of the inner conductor contact surface of the charging gun head, and an isomorphic sensing ring is added at a predetermined distance in the axial depth of the conductor to form a three-dimensional monitoring system; temperature sensors are arranged around the outer periphery of the conductor contact area on the bonding surface between the insulating outer shell and the metal shell of the charging gun head. Miniature temperature sensors are uniformly embedded in the circumferential direction at the crimping point between the cable shielding layer and the gun head metal adapter; sensors are arranged on both sides of the welding point between the inner copper conductor of the cable core and the inner conductor of the gun head; a flow-temperature composite sensor is installed inside the liquid cooling channel inlet of the connection node. Along the length of the cable, a predetermined length is designated as a monitoring unit. Distributed fiber optic temperature sensors are deployed inside the shielding layer, in the middle of the insulation layer, and on the surface of the inner conductor of each unit, forming a three-line parallel axial sensing chain. Redundant temperature sensors are added at the midpoint and both ends of the entire cable.

3. The thermal equalization control method for a liquid-cooled charging module according to claim 2, characterized in that, The temperature sensing array includes multiple distributed temperature sensors, which are placed in predetermined locations and also include: High-precision temperature sensors are installed at the main inlet and main outlet of the liquid cooling system; temperature sensors are installed before and after each branch valve that branches the flow to the nozzle and cable; and temperature sensors are installed at the inlet and outlet of the heat exchanger on the heat dissipation circuit of the cooling medium. Temperature sensors are placed on the heat dissipation substrates of the IGBTs and rectifier bridge power devices in the charging module, corresponding to the center position of each chip; temperature sensors are also placed in the inlet and outlet chambers of the liquid cooling cavity inside the module.

4. The thermal equalization control method for a liquid-cooled charging module according to claim 3, characterized in that, The step of acquiring the raw temperature data of the temperature sensing array in real time through the temperature acquisition module and transmitting it to the temperature field reconstruction module specifically includes: The temperature acquisition module collects raw temperature data from all distributed sensors and performs multi-level preprocessing, including removing outliers that exceed the sensor's range, marking sensors with multiple consecutive anomalies as faulty and enabling redundant backup data; using a 5th-order moving average filtering algorithm to smooth high-frequency noise, and superimposing Kalman filtering on dynamic heating areas; binding preset spatial coordinates based on sensor IDs and synchronizing the acquisition times of different sensors through timestamps. The preprocessed temperature data is transmitted to the temperature field reconstruction module via industrial Ethernet and fiber optic cable. The transmission frame format includes sensor ID, three-dimensional coordinates, temperature value, acquisition timestamp, and data confidence level. Data with a transmission delay of more than 50ms is marked as invalid and a retransmission mechanism is triggered.

5. The thermal equalization control method for a liquid-cooled charging module according to claim 4, characterized in that, The process of processing the raw temperature data based on the spatial interpolation algorithm to construct a three-dimensional dynamic temperature field model of the key heat-generating areas, and to output the temperature values, temperature gradients, and temperature difference distributions of each area in real time, specifically includes: The three-dimensional spatial mesh modeling and temperature field reconstruction module is based on the CAD model of the key heat-generating area, constructs a three-dimensional spatial mesh, and performs mesh division; maps the actual deployment location of the sensor to the mesh node, assigns adiabatic boundary conditions to the boundaries of the gun head shell and cable insulation layer, and assigns convective heat transfer boundary conditions to the liquid cooling channel wall. An improved spatial interpolation algorithm is used to fill the grid temperature values ​​using an improved Kriging interpolation algorithm that incorporates thermal characteristic weights. Spherical variograms and exponential variograms are defined for different thermal characteristics of metallic conductors and insulating materials, respectively. The weights are dynamically adjusted according to the heat dissipation power of the sensor's region. Constraints are set on the temperature difference between adjacent grid points. The dynamic temperature field is updated in real time. The temperature field reconstruction module performs a full mesh update periodically, corrects the mesh node temperature based on the latest collected data, and calculates the temperature gradient vector of each node using the finite difference method. For areas with temperature gradients greater than 2℃ / mm, the calculation is refined to generate local high temperature gradient cloud maps. Areas with temperature differences exceeding the threshold are automatically identified and marked. Output a standardized dataset of the three-dimensional dynamic temperature field model, including the highest, lowest, and average temperatures of each key region; the temperature difference matrix between any two points; a thermogram of the temperature gradient distribution; and a temperature field prediction curve.

6. The thermal equalization control method for a liquid-cooled charging module according to claim 5, characterized in that, The step of transmitting the output data of the three-dimensional dynamic temperature field model to the thermal equilibrium control module, wherein the thermal equilibrium control module presets a temperature threshold range and a temperature difference threshold, specifically including: A three-dimensional dynamic temperature field model transmits standardized data frames to the thermal equilibrium control module via an industrial bus, including the highest temperature value and coordinates of each key area; the temperature difference matrix between adjacent areas; the coordinate set of areas where the temperature gradient exceeds the threshold; and temperature field prediction data. The control module performs CRC verification on the received data; if the verification fails, it triggers immediate retransmission. The control module parses the data according to priority. The first-level data is the temperature of the inner conductor of the gun head and the temperature of the connection node; the second-level data is the axial temperature difference distribution of the cable and the temperature difference between the inlet and outlet of the liquid cooling circuit; the third-level data is the temperature field prediction curve and the overall temperature difference distribution statistics.

7. The thermal equalization control method for a liquid-cooled charging module according to claim 6, characterized in that, The preset temperature threshold range and temperature difference threshold of the thermal equilibrium control module specifically include: Threshold grading settings: The thermal equalization control module has multiple built-in threshold parameter tables, which are dynamically switched according to the charging power. The thresholds are divided into absolute temperature thresholds, such as the inner conductor contact surface of the gun head, the connection node between the cable and the gun head, and the liquid cooling circuit inlet; temperature difference thresholds, such as the circumferential temperature difference between the gun head and the connection node, the same cross section of the cable, and the inlet and outlet temperature difference of the liquid cooling circuit; and gradient thresholds, such as the temperature gradient of any adjacent grid points. Threshold dynamic correction: Based on ambient temperature sensor data, the threshold is corrected in real time. When the ambient temperature is higher than the predetermined temperature, the maximum temperature threshold of the nozzle and the connection node is lowered; when the ambient temperature is lower than 0℃, the minimum temperature threshold is raised.

8. The thermal equalization control method for a liquid-cooled charging module according to claim 7, characterized in that, The thermal equilibrium control module compares the output data of the three-dimensional dynamic temperature field model with the preset threshold. If the temperature value of a certain area exceeds the maximum temperature threshold, or the temperature difference between two adjacent key areas exceeds the temperature difference threshold, it sends a control command to the partitioned cooling execution unit of the liquid cooling system. Specifically, this includes: Threshold comparison: Check whether the absolute temperature of each key area exceeds the corresponding threshold range, mark the over-temperature area and the over-temperature range; calculate the temperature difference between adjacent key areas to determine whether it exceeds the temperature difference threshold; scan the grid area with excessive temperature gradient, if there is an area with a gradient greater than 2℃ / mm for 3 consecutive cycles, it is determined to be a local heat concentration. The control module generates corresponding control commands based on the anomaly level. Level 1 anomalies only adjust the cooling circuit flow rate of the corresponding area, with the adjustment range being a predetermined proportion of the base flow rate. Level 2 anomalies simultaneously adjust the cooling circuits of the target area and adjacent areas, increasing the main circuit flow rate by a predetermined proportion and activating the overclocking mode of the variable frequency water pump in the corresponding area. Level 3 anomalies activate the emergency cooling mode, adjusting the flow rate of the target area to the maximum value, running the main circuit water pump at full power, and simultaneously sending a power reduction request to the charging main control module. The control command adopts a standardized frame structure, which includes the target branch ID, target flow, adjustment duration, and execution priority. The partition cooling execution unit returns the execution status within 10ms after receiving the command. If the control module does not receive feedback within 30ms, it will resend the command and mark the branch as a suspected fault. After the command is executed, the control module tracks the temperature field data for the next cycle. If the deviation between the measured temperature in the over-threshold area and the preset threshold decreases by less than a predetermined proportion, the predetermined proportion of flow rate adjustment is added until the temperature returns to the threshold range.

9. The thermal equalization control method for a liquid-cooled charging module according to claim 8, characterized in that, The repeated steps described above, until the temperature values ​​of all key heat-generating regions in the three-dimensional dynamic temperature field model are within the preset temperature threshold range, and the temperature difference between adjacent key regions is less than or equal to the temperature difference threshold, specifically include: With a control cycle of 100ms, temperature acquisition, field model reconstruction, threshold comparison, control command output, and execution feedback are executed sequentially within each cycle. When the system is normal and all parameters are within the threshold, the basic sampling frequency and control cycle are maintained. When a level 1 anomaly is detected, the cycle is automatically shortened to 50ms to increase the control density. Under level 3 anomaly conditions, a high-frequency response mode is triggered until the anomaly level drops to level 2 or below.

10. The thermal equalization control method for a liquid-cooled charging module according to claim 9, characterized in that, The process of repeating the above steps until the temperature values ​​of all key heat-generating regions in the three-dimensional dynamic temperature field model are within the preset temperature threshold range, and the temperature difference between adjacent key regions is less than or equal to the temperature difference threshold, further includes: Active control can be stopped and the system switched to steady-state monitoring mode when the following conditions are met: within three consecutive control cycles, the highest temperature of all key areas is less than or equal to the preset highest threshold, and the lowest temperature is greater than or equal to the preset lowest threshold. The temperature difference between adjacent key areas is less than or equal to the temperature difference threshold for five consecutive cycles; there are no grid areas with gradient exceeding the limit in the temperature field model, and the temperature prediction curve shows no risk of exceeding the threshold within the next 2 seconds.

Images