A heat dissipation control method, equipment and medium for a single-phase immersion liquid-cooled charging pile

Abstract

This invention discloses a heat dissipation control method, device, and medium for a single-phase immersion liquid-cooled charging pile, relating to the field of battery thermal management technology. The method includes: performing high-precision scanning and dynamic feature extraction on the fiber Bragg grating array built into the cold plate according to a zoned flow control signal to obtain a temperature field matrix and a dynamic thermal feature set; performing multi-dimensional fusion analysis and electrochemical coupling analysis on the temperature field matrix and dynamic thermal feature set to generate a thermal coupling parameter set; performing real-time risk quantification assessment and risk topology mapping on the thermal coupling parameter set to output a risk topology dataset; and dynamically modulating the cooling intensity based on the risk topology dataset to obtain a zoned flow redistribution command. This invention achieves refined monitoring of the contact thermal resistance at the cold plate-battery interface through high-precision pressure field reconstruction technology, significantly improving the accuracy and reliability of pressure distribution detection and effectively eliminating the risk of localized overheating caused by assembly gaps.

Description

Technical Field

[0001] This invention relates to the field of battery thermal management technology, and in particular to a heat dissipation control method, equipment and medium for a single-phase immersion liquid-cooled charging pile. Background Technology

[0002] In recent years, with the development of fast charging technology for electric vehicles towards higher power density, the heat dissipation system of charging piles faces severe challenges. Traditional heat dissipation solutions are mainly divided into two categories: air cooling and liquid cooling. Air cooling technology reduces temperature through forced convection, while liquid cooling technology improves heat dissipation efficiency through cold plates or immersion designs. In liquid cooling systems, cold plate heat dissipation conducts heat through contact between metal channels and the battery surface, while immersion liquid cooling directly immerses the heat-generating components in insulating coolant. However, these solutions still have significant limitations in dealing with the dynamic deformation and thermal runaway risks during battery fast charging.

[0003] The current technical problem lies in the inability to compensate for the dynamic contact thermal resistance at the battery-cold plate interface in real time. During fast charging, the battery undergoes periodic expansion due to lithium ion insertion / extraction, causing continuous changes in the contact pressure distribution of the cold plate. This can easily lead to a chain of problems, such as increased temperature field inhomogeneity, triggering forced power reduction by the BMS, imbalance in coolant flow rate distribution, and insufficient flow in low-pressure areas leading to boiling risks. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a heat dissipation control method for a single-phase immersion liquid-cooled charging pile to solve the problems of uneven heat dissipation and thermal runaway risk caused by battery deformation during fast charging.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a heat dissipation control method for a single-phase immersion liquid-cooled charging pile, comprising,

[0008] Real-time acquisition of contact pressure data between the battery and the cold plate, followed by visualization reconstruction to generate a pressure distribution matrix;

[0009] Construct a traffic decision model, perform intelligent strategy mapping and dynamic threshold optimization on the pressure distribution matrix, and generate partitioned traffic control signals;

[0010] Based on the zoned flow control signal, high-precision scanning and dynamic feature extraction are performed on the built-in fiber Bragg grating array of the cold plate to obtain the temperature field matrix and dynamic thermal feature set.

[0011] A multi-dimensional fusion analysis and electrochemical coupling analysis of the temperature field matrix and dynamic thermal feature set are performed to generate a thermal coupling parameter set.

[0012] Perform real-time risk quantification assessment and risk topology mapping on the thermally coupled parameter set, and output a risk topology dataset;

[0013] Dynamically modulate cooling intensity based on the risk topology dataset and obtain partition traffic redistribution instructions.

[0014] As a preferred embodiment of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile of the present invention, the specific steps for generating the pressure distribution matrix are as follows:

[0015] Real-time acquisition of contact pressure data between the battery and the cold plate, signal conditioning and dynamic compensation, and generation of a temperature-compensated pressure signal array;

[0016] Noise suppression and physical quantity calibration conversion are performed on the temperature-compensated pressure signal array to generate a physical pressure value array.

[0017] Spatial interpolation expansion and gradient continuity verification are performed on the physical pressure value array to generate a contact pressure feature field;

[0018] Pressure threshold partitioning and statistical feature integration are performed on the contact pressure feature field to generate a pressure distribution matrix.

[0019] As a preferred embodiment of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile of the present invention, the construction of the flow decision model refers to constructing the flow decision model based on the sensor data layer, the strategy reasoning layer and the dynamic optimization layer.

[0020] As a preferred embodiment of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile of the present invention, the specific steps for generating the zoned flow control signal are as follows:

[0021] The sensor data layer extracts and vectorizes multi-dimensional features from the pressure distribution matrix to generate a pressure feature vector.

[0022] The strategy inference layer performs intelligent strategy mapping and flow ratio allocation on the pressure feature vector to generate an initial flow ratio coefficient array.

[0023] The dynamic optimization layer performs multi-source feedback optimization and dynamic threshold adjustment on the initial flow ratio coefficient array to generate partition flow control signals.

[0024] As a preferred embodiment of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile of the present invention, the specific steps for obtaining the temperature field matrix and dynamic thermal feature set are as follows:

[0025] Based on the partitioned flow control signal, synchronous scanning and spectrum-temperature conversion of the fiber grating array are performed to generate a preliminary temperature data array.

[0026] Spatial interpolation expansion and edge optimization processing are performed on the initial temperature data array to generate a high-resolution temperature field matrix;

[0027] The dynamic characteristics of the high-resolution temperature field matrix are analyzed to generate a dynamic thermal feature set.

[0028] As a preferred embodiment of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile of the present invention, the specific steps for generating the thermal coupling parameter set are as follows:

[0029] Multi-source data spatiotemporal alignment and feature fusion are performed on dynamic hot feature sets to generate spatiotemporally aligned datasets;

[0030] Perform thermo-electric property correlation and parameter mapping on the spatiotemporally aligned dataset, and output an intermediate set of coupling parameters;

[0031] Based on the intermediate set of coupling parameters, the lithium plating probability is calculated and a safety margin assessment is performed, and the thermal coupling parameter set is output.

[0032] As a preferred embodiment of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile of the present invention, the specific steps for outputting the risk topology dataset are as follows:

[0033] The probability field is reconstructed and the risk region is extracted from the thermally coupled parameter set to generate a risk probability field matrix;

[0034] The risk probability field matrix is ​​optimized for regional continuity and the risk intensity is calibrated to output a comprehensive risk field matrix.

[0035] Perform risk region clustering analysis and hierarchical topology construction on the comprehensive risk field matrix, and output a risk topology dataset.

[0036] In a preferred embodiment of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile described in this invention, the specific steps for obtaining the zone flow redistribution instruction are as follows:

[0037] Hierarchical cooling intensity mapping and dynamic battery state constraints are applied to the risk topology dataset, and a cooling intensity modulation strategy table is output.

[0038] The cooling intensity modulation strategy table is processed by actuator signal conversion and safety parameter encapsulation, and outputs partition flow redistribution instructions.

[0039] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in the first aspect of the present invention.

[0040] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in the first aspect of the present invention.

[0041] The beneficial effects of this invention are as follows: by using high-precision pressure field reconstruction technology to achieve refined monitoring of the contact thermal resistance of the cold plate-battery interface, the accuracy and reliability of pressure distribution detection are significantly improved, and the risk of local overheating caused by assembly gaps is effectively eliminated; by using electrochemical-thermal coupling risk modeling to achieve early prediction of lithium plating risk, the timeliness and accuracy of identifying thermal runaway hazards are greatly improved, and a full-link intelligent heat dissipation control system from the physical layer to the electrochemical layer is constructed. Attached Figure Description

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart of a heat dissipation control method for a single-phase immersion liquid-cooled charging pile.

[0044] Figure 2 A flowchart for generating the pressure distribution matrix.

[0045] Figure 3 A flowchart generated for the thermal coupling parameter set.

[0046] Figure 4 A flowchart generated for the risk topology dataset. Detailed Implementation

[0047] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0048] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0050] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a heat dissipation control method for a single-phase immersion liquid-cooled charging pile, comprising the following steps:

[0051] S1. Real-time acquisition of contact pressure data between the battery and the cold plate, and visualization reconstruction to generate a pressure distribution matrix;

[0052] Specifically, such as Figure 2 The steps are as follows:

[0053] S1.1. Real-time acquisition of contact pressure data between the battery and the cold plate, and signal conditioning and dynamic compensation to generate a temperature-compensated pressure signal array;

[0054] It should be noted that the original analog voltage signal is continuously acquired at a fixed sampling frequency by a micro piezoelectric thin film sensor array embedded on the surface of the cold plate. The original analog voltage signal is converted into a digital signal by a 16-bit analog-to-digital converter. Temperature drift compensation processing is performed on the digital signal. Furthermore, the temperature sensor data inside the cold plate is read synchronously. According to the sensor output voltage and the standard pressure gauge reading under different temperature conditions, a quantitative correspondence table of temperature-voltage-pressure is fitted by the least squares method. The output voltage of each sensing channel is compensated in real time. The compensated digital voltage signal is then processed by a Butterworth low-pass filter to filter out high-frequency noise components caused by mechanical vibration and retain the low-frequency pressure signal. Finally, the temperature-compensated pressure signal array is output.

[0055] S1.2. Perform noise suppression and physical quantity calibration conversion on the temperature-compensated pressure signal array to generate a physical pressure value array;

[0056] It should be noted that the waveform is smoothed by taking the arithmetic mean of multiple consecutive data points in the pressure signal array after temperature compensation. Furthermore, a fixed-length sliding window is set, and the window moves point by point along the signal sequence. Each time, the average value of all data points in the window is calculated as the output value of the window center point, thereby eliminating high-frequency fluctuations and retaining the signal trend components. Subsequently, the sensor calibration curve is applied. The sensor calibration curve is established through laboratory calibration experiments to convert the voltage value of each sensing unit into the corresponding pressure physical quantity, and finally outputs the physical pressure value array.

[0057] S1.3. Spatial interpolation expansion and gradient continuity verification are performed on the physical pressure value array to generate a contact pressure characteristic field;

[0058] It should be noted that the physical pressure value array is divided into rectangular units by four adjacent pressure sensing points, and each unit is further divided into several equally spaced grid points. Based on the known pressure values ​​at the four vertices of the unit, linear interpolation is first performed along the X direction for each row of boundary points to obtain the values ​​of intermediate auxiliary points, and then secondary interpolation is performed along the Y direction for the values ​​of intermediate auxiliary points to obtain the grid point pressure values. After traversing and processing all sensor areas unit by unit, the discrete sensor point data is transformed into a continuous spatial pressure distribution field. Gradient continuity verification is performed on the continuous spatial pressure distribution field, the absolute value of the pressure gradient between adjacent grid points is calculated, and abnormal jump points exceeding the pressure gradient threshold are corrected using the median replacement method. Finally, a contact pressure characteristic field with spatial continuity and conforming to physical laws is output.

[0059] It should be noted that the pressure gradient threshold is set based on the shear stress limit of the laminar boundary layer in fluid mechanics, and the example value range is 3-8 MPa / mm.

[0060] S1.4. Perform pressure threshold partitioning mapping and statistical feature integration on the contact pressure feature field to generate a pressure distribution matrix.

[0061] It should be noted that the pressure value of each grid point in the contact pressure feature field is compared with the preset first pressure threshold and second pressure threshold. Areas with pressure values ​​lower than the first pressure threshold are marked as low-pressure areas, areas with pressure values ​​between the first and second pressure thresholds are marked as medium-pressure areas, and areas with pressure values ​​higher than the second pressure threshold are marked as high-pressure areas. The area ratio of low-pressure areas and the global pressure variance are calculated simultaneously. The partitioning results and statistical feature values ​​are written together into the matrix header file, and finally, a pressure distribution matrix containing pressure partitioning information and statistical features is output.

[0062] It should be noted that the first pressure threshold and the second pressure threshold are set based on the contact thermal resistance safety limit of the cold plate-battery interface.

[0063] S2. Construct a traffic decision model, perform intelligent strategy mapping and dynamic threshold optimization on the pressure distribution matrix, and generate partitioned traffic control signals.

[0064] S2.1. Based on the sensor data layer, strategy reasoning layer, and dynamic optimization layer, a traffic decision model is constructed.

[0065] As should be noted, the traffic flow decision model is constructed as follows: In the PyTorch framework, a one-dimensional convolutional network is invoked using the nn.Conv1d parameter and initialized. A spatial attention mechanism is then applied after the convolutional layer to enhance the feature response of key regions through self-attention. A LayerNorm layer is used for feature normalization, completing the construction of the sensing data layer. The Transformer encoder architecture is invoked using the nn.TransformerEncoder parameter and initialized. Process constraint rules are embedded in the encoder, and a safety threshold is injected into the network as prior knowledge through a gating mechanism. Residual connections are used to prevent gradient vanishing, completing the construction of the policy inference layer. A Long Short-Term Memory (LSTM) network is invoked using the nn.LSTM parameter and initialized. A real-time feedback regulator is embedded in the LSM network, and the traffic flow coefficient is dynamically corrected through a temporal difference learning algorithm. A temperature-pressure coupling loss function is simultaneously integrated for backpropagation optimization, completing the construction of the dynamic optimization layer and obtaining the traffic flow decision model.

[0066] It should be noted that the safety threshold is set based on the battery thermal safety boundary and pressure fluctuation tolerance. Example value range: real-time battery surface temperature 45℃-55℃, spatial variance of contact pressure between battery and cold plate is 0.3MPa-0.8MPa.

[0067] The pre-training process of the flow decision model is as follows: Supervised learning using historical running datasets is adopted to complete parameter training by optimizing the comprehensive performance of temperature uniformity and energy consumption indicators. Furthermore, a training set containing a large number of pressure distribution matrix samples and regional flow control signals is constructed. The maximum temperature difference on the battery surface and the reduction of total power consumption are used as joint optimization objectives. Five-fold cross-validation is used during training to prevent overfitting. Finally, training is terminated when the loss decrease is lower than the minimum convergence threshold for 5 consecutive rounds, and the flow decision model is output.

[0068] It should be noted that the minimum convergence threshold is set based on the balance between overfitting risk and training efficiency, with example values ​​ranging from (0.1%, 0.5%).

[0069] S2.2 The sensor data layer extracts and vectorizes the multi-dimensional features of the pressure distribution matrix to generate a pressure feature vector;

[0070] It should be noted that the low-pressure area ratio and the global pressure variance are read from the header file of the pressure distribution matrix. At the same time, the number of connected regions and the spatial distribution density of different pressure zones in the pressure distribution matrix are counted. The eight feature parameters, namely low-pressure area ratio, pressure variance, high-pressure area ratio, medium-pressure area ratio, number of low-pressure connected regions, number of high-pressure connected regions, low-pressure area density, and high-pressure area density, are combined into an eight-dimensional vector in a fixed order, and finally the pressure feature vector with a standardized structure is output.

[0071] S2.3 The strategy reasoning layer performs intelligent strategy mapping and flow ratio allocation on the pressure feature vector to generate an initial flow ratio coefficient array;

[0072] It should be noted that the pressure feature vector is input into a pre-constructed decision rule base, which contains a mapping table between pressure features and flow coefficients. The basic flow coefficient is obtained by looking up the mapping table. Furthermore, firstly, pressure distribution matrices under different operating conditions and corresponding optimal flow control records are collected to form a training dataset. Then, a multivariate regression method is used to analyze the quantitative relationship between the pressure feature vector and the basic flow coefficient. Next, the correlation analysis results are transformed into a lookup table structure with pressure feature values ​​as input indices and optimal flow coefficients as output values. Finally, cross-validation is used to iteratively optimize the mapping relationship of the decision rule base, controlling the basic flow coefficient and the optimal flow coefficient below a preset flow threshold. The mapping table is generated by discretizing the pressure characteristic values ​​into intervals and obtaining the mode of the corresponding flow coefficient in each interval as the mapping output value. Then, based on the low-pressure zone density value in the pressure characteristic vector, the corresponding zone coefficient is adjusted proportionally according to the change in low-pressure zone density (e.g., the zone coefficient increases by 0.1 for every 10% increase in density). At the same time, combined with the pressure variance value, the high-pressure zone coefficient is corrected accordingly based on the degree of change in pressure variance (e.g., the high-pressure zone coefficient decreases by 0.05 for every 0.5 increase in variance). Finally, the flow ratio coefficients of sixteen zones are generated, forming an initial flow ratio coefficient array.

[0073] It should be noted that the preset flow rate threshold is set based on the thermal safety boundary of the lithium-ion battery, and the example value range is 0.08.

[0074] S2.4 The dynamic optimization layer performs multi-source feedback optimization and dynamic threshold control on the initial flow ratio coefficient array to generate partition flow control signals.

[0075] It should be noted that the initial flow ratio coefficient array receives real-time temperature data of the battery surface and pressure variance values ​​in the pressure distribution matrix; when the real-time temperature data of the battery surface exceeds the temperature control threshold, the flow ratio coefficients of all zones are adjusted upwards; when the pressure variance value exceeds the pressure control threshold, the flow ratio coefficients of the high-pressure zone are corrected downwards; the adjusted flow ratio coefficients are converted into pulse width modulation duty cycle signals through the flow-control signal conversion relationship, and finally sixteen zone flow control signals are generated.

[0076] It should be noted that the temperature control threshold is set based on the thermal safety boundary of the lithium-ion battery, and the pressure control threshold is set based on the pressure fluctuation tolerance of the contact interface.

[0077] The process of establishing the flow-control signal conversion relationship is as follows: Under laboratory conditions, a precision flow meter is used to measure the coolant flow rate of the solenoid valve under different control signal intensities, and the corresponding data of signal intensity and flow rate are recorded; a correspondence table of signal intensity and flow rate is established based on the measured data, and the flow control signal is generated by looking up the table; finally, the relationship table is written into the non-volatile memory of the controller to form a fixed flow-control signal conversion relationship.

[0078] S3. Based on the zoned flow control signal, perform high-precision scanning and dynamic feature extraction on the built-in fiber grating array of the cold plate to obtain the temperature field matrix and dynamic thermal feature set.

[0079] S3.1 Based on the partitioned flow control signal, perform synchronous scanning and spectrum-temperature conversion of the fiber optic array to generate a preliminary temperature data array;

[0080] It should be noted that the zone flow control signal triggers the wavelength demodulator to start, synchronously acquiring the reflection spectra of all fiber Bragg grating sensors built into the cold plate. The reflection spectra are converted using a wavelength-temperature correspondence table calibrated in the laboratory. The wavelength-temperature correspondence table is constructed by measuring the center wavelength offset of the fiber Bragg grating sensors under different temperature environments and recording the corresponding data. The center wavelength offset of each sensor is converted into a temperature value using the wavelength-temperature correspondence table. The converted temperature values ​​are then processed by Kalman filtering to eliminate fluid pulsation interference, and finally, a preliminary temperature data array containing the temperature values ​​of all sensors is output.

[0081] S3.2 Perform spatial interpolation expansion and edge optimization processing on the preliminary temperature data array to generate a high-resolution temperature field matrix;

[0082] It should be noted that the initial temperature data array achieves grid refinement through spatial interpolation expansion processing. Taking the rectangular area formed by four adjacent temperature sensor points as the reference area, the temperature value of each interpolation point within the cell is determined based on the known temperature values ​​of the four vertices of the reference area using a linear weighting method. This process is repeated cell by cell until the entire sensor area is covered, forming a continuous spatial temperature distribution. Subsequently, edge optimization processing is performed on the regularized temperature distribution matrix generated after spatial interpolation expansion processing. By calculating the temperature difference between each grid point and its directly adjacent points, outlier points exceeding the temperature gradient anomaly threshold are replaced with the median temperature of the surrounding points, eliminating discontinuities and abrupt changes in the distribution field. The final output is a high-resolution temperature field matrix with physical rationality and spatial continuity.

[0083] It should be noted that the temperature gradient anomaly threshold is set based on the maximum allowable rate of temperature change in the heat conduction theory, with an example value range of 1.5℃ / mm - 3.0℃ / mm.

[0084] S3.3 Perform dynamic thermal characteristic analysis on the high-resolution temperature field matrix to generate a dynamic thermal feature set.

[0085] It should be noted that the high-resolution temperature field matrix calculates the temperature difference between adjacent grid points. When the temperature difference exceeds the maximum allowable temperature gradient for preventing local thermal runaway in the lithium-ion battery thermal safety specifications, it is marked as a high-risk area. Simultaneously, the global temperature change rate value is acquired, and combined with the coordinate information of the high-risk area to form a structured data set, and finally outputs a dynamic thermal feature set.

[0086] S4. Perform multi-dimensional fusion analysis and electrochemical coupling analysis on the temperature field matrix and dynamic thermal feature set to generate a thermal coupling parameter set;

[0087] Specifically, such as Figure 3 The steps are as follows:

[0088] S4.1 Perform multi-source data spatiotemporal alignment and feature fusion on the dynamic hot feature set to generate a spatiotemporal aligned dataset;

[0089] It should be noted that the temperature field matrix, high-risk area coordinate set, and temperature change rate scalar included in the dynamic thermal feature set are synchronized with the voltage, current, and SOC data provided by the battery management system through timestamp matching, with the timestamp accuracy controlled at the millisecond level. The synchronized temperature field matrix, voltage, current, and SOC data are integrated according to the time series to form a spatiotemporally aligned dataset containing four types of parameters: temperature field, voltage, current, and SOC, ensuring that all data have a unified time base and spatial reference framework.

[0090] S4.2 Perform thermo-electric property correlation and parameter mapping on the spatiotemporal aligned dataset, and output the intermediate set of coupling parameters;

[0091] It should be noted that the temperature field matrix in the spatiotemporally aligned dataset is aggregated into average temperature values ​​by region, and then combined with voltage and current data at the same time stamp to form time series triplets. Subsequently, the sliding window method is used to analyze the synchronicity between temperature change trends and current change trends, and the consistency ratio between temperature rise / fall and current increase / decrease within the window is used as the correlation index. Finally, the thermal-electric coupling coefficient is generated based on the correlation index. When the correlation in the high-voltage area is higher than the thermal-electric correlation judgment threshold, it is marked as a strong coupling area. The final output is an intermediate set of coupling parameters containing the equivalent thermal impedance distribution, the thermal-electric correlation scalar, and the coupling coefficient.

[0092] It should be noted that the threshold for determining the thermo-electric correlation is set based on the synchronicity of thermo-electric changes in historical charge and discharge data, with an example value range of (0.75, 0.88).

[0093] S4.3. Based on the intermediate set of coupling parameters, calculate the lithium plating probability and perform a safety margin assessment, outputting the thermal coupling parameter set, expressed as follows:

[0094] ;

[0095] in, for The probability value of lithium plating at that location. for Temperature gradient value at that location, for The equivalent thermal resistance at that point.

[0096] It should be noted that the equivalent thermal impedance distribution and temperature gradient data in the intermediate set of coupling parameters are processed through a risk probability function to generate a lithium plating probability value for each grid point. At the same time, a linear correspondence is established between the lithium plating probability value and the safety margin number, specifically, the safety margin equals one minus the lithium plating probability value; when the lithium plating probability value is zero, the safety margin is 1.0; when the lithium plating probability value is 0.5, the safety margin is 0.5; and when the lithium plating probability value is 1.0, the safety margin is zero. The final output is a thermal coupling parameter set containing the lithium plating probability distribution, the safety margin index, and the original coupling parameters.

[0097] S5. Perform real-time risk quantification assessment and risk topology mapping on the thermally coupled parameter set, and output the risk topology dataset;

[0098] Specifically, such as Figure 4 The steps are as follows:

[0099] S5.1 Perform probability field reconstruction and risk region extraction on the thermal coupling parameter set to generate a risk probability field matrix;

[0100] It should be noted that the lithium plating probability distribution in the thermal coupling parameter set is processed by Gaussian filtering to eliminate local noise interference and smooth the probability distribution field; continuous regions with probability values ​​exceeding the risk probability judgment threshold are extracted from the smoothed probability distribution field, marked as risk candidate regions, and their boundary coordinates are recorded; the final output is a risk probability field matrix containing the probability values ​​of all grid points and a list of risk candidate region boundaries.

[0101] It should be noted that the risk probability determination threshold is set based on the critical probability of lithium plating reaction occurring in lithium-ion batteries, with an example value range of (0.3, 0.5).

[0102] S5.2 Perform regional continuity optimization and risk intensity calibration on the risk probability field matrix, and output a comprehensive risk field matrix;

[0103] It should be noted that the risk probability field matrix improves the regional continuity in the lithium plating risk probability distribution through morphological processing, filling discontinuous regions and connecting adjacent risk areas. Furthermore, regions where the lithium plating risk probability value reaches or exceeds the risk probability judgment threshold are expanded, extending each risk point outwards to fill the tiny gaps between regions. Subsequently, an erosion operation is performed on the expanded regions to eliminate redundant edges generated by the expansion and preserve the complete shape of the main risk area. Finally, neighbor analysis is used to merge adjacent risk areas, forming continuous risk bands and eliminating isolated points. Ultimately, the regional continuity in the lithium plating risk probability distribution is optimized. The processed lithium plating risk probability values ​​are categorized into low-probability and high-probability intervals. The low-probability interval refers to regions in the lithium plating risk probability field matrix where the lithium plating risk probability value is below the risk probability judgment threshold, indicating a low lithium plating risk and corresponding to a safe operating state. The high-probability interval refers to regions where the lithium plating risk probability value is equal to or higher than the risk probability judgment threshold, indicating a higher lithium plating risk and corresponding to a high-risk state requiring intervention. Low probability intervals correspond to low risk indicators, and high probability intervals correspond to high risk indicators; the final output is a comprehensive risk field matrix that includes risk intensity calibration results and probability distribution information.

[0104] S5.3 Perform risk region clustering analysis and hierarchical topology construction on the comprehensive risk field matrix, and output the risk topology dataset.

[0105] It should be noted that the comprehensive risk field matrix is ​​processed using spatial clustering methods, grouping grid points with similar risk intensity values ​​and adjacent locations into the same risk cluster. Each risk cluster is further divided into different levels based on its risk intensity value. Risk clusters with risk intensity values ​​below the risk probability judgment threshold are marked as warning level (low risk), while those with risk intensity values ​​equal to or higher than the risk probability judgment threshold are marked as emergency level (high risk). Based on the spatial distribution of risk clusters, an adjacency table and a list of center point coordinates are generated. The final output is a risk topology dataset containing a hierarchical risk area map, an adjacency table, and a list of risk center point coordinates.

[0106] S6. Dynamically modulate the cooling intensity based on the risk topology dataset and obtain the partition traffic redistribution instruction.

[0107] S6.1. Perform hierarchical cooling intensity mapping and dynamic battery state constraints on the risk topology dataset, and output a cooling intensity modulation strategy table.

[0108] It should be noted that in the risk topology dataset, emergency-level risk areas are mapped to high-intensity cooling levels, and warning-level risk areas are mapped to medium-intensity cooling levels. The cooling intensity mapping results are adjusted based on battery state of charge and battery health data. When the battery state of charge value is low, the cooling intensity coefficient of all areas is reduced proportionally, and when the battery health value is low, the increase in cooling intensity of emergency-level risk areas is reduced proportionally. The final output is a cooling intensity modulation strategy table containing the cooling intensity coefficient and adjustment parameters of each area.

[0109] S6.2 Perform actuator signal conversion and safety parameter encapsulation on the cooling intensity modulation strategy table, and output the partition flow redistribution instruction.

[0110] It should be noted that the cooling intensity coefficient in the cooling intensity modulation strategy table is processed through a flow-control signal conversion relationship calibrated in the laboratory. This flow-control signal conversion relationship is obtained by precisely measuring the coolant flow rate of the solenoid valve under different control signals and establishing a corresponding table, converting the cooling intensity requirement of each zone into the corresponding control signal value. Simultaneously, the coordinates of the highest-risk area, the global cooling intensity adjustment range, and the suggested charging power adjustment range parameters are extracted from the cooling intensity modulation strategy table and encapsulated into a safety control parameter set. The final output is a zone flow redistribution command containing the zone control signal and the safety control parameter set. The final output zone flow redistribution command controls the opening of the solenoid valves in each zone through pulse width modulation duty cycle control, achieving dynamic flow distribution: higher flow is allocated to high-risk areas (e.g., 300% of the baseline flow), while low-risk areas maintain the baseline flow. Combined with the safety control parameter set, the cooling intensity increase is dynamically limited when the battery condition is abnormal. By monitoring temperature field changes in real time and providing feedback to optimize flow distribution, the temperature of each zone is ensured to remain stable within the target range.

[0111] This embodiment also provides a computer device applicable to the heat dissipation control method of a single-phase immersion liquid-cooled charging pile, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the heat dissipation control method of the single-phase immersion liquid-cooled charging pile as proposed in the above embodiment.

[0112] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0113] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the heat dissipation control method for a single-phase immersion liquid-cooled charging pile as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0114] In summary, this invention achieves refined monitoring of the contact thermal resistance at the cold plate-battery interface through high-precision pressure field reconstruction technology, significantly improving the accuracy and reliability of pressure distribution detection and effectively eliminating the risk of local overheating caused by assembly gaps; and enables early prediction of lithium plating risk through electrochemical-thermal coupling risk modeling, greatly improving the timeliness and accuracy of identifying potential thermal runaway hazards, thus constructing a full-link intelligent heat dissipation control system from the physical layer to the electrochemical layer.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A heat dissipation control method for a single-phase immersion liquid-cooled charging pile, characterized in that: include, Real-time acquisition of contact pressure data between the battery and the cold plate, followed by visualization reconstruction to generate a pressure distribution matrix; Construct a traffic decision model, perform intelligent strategy mapping and dynamic threshold optimization on the pressure distribution matrix, and generate zoned traffic control signals; Based on the zoned flow control signal, high-precision scanning and dynamic feature extraction are performed on the built-in fiber Bragg grating array of the cold plate to obtain the temperature field matrix and dynamic thermal feature set. A multi-dimensional fusion analysis and electrochemical coupling analysis of the temperature field matrix and dynamic thermal feature set are performed to generate a thermal coupling parameter set. Perform real-time risk quantification assessment and risk topology mapping on the thermally coupled parameter set, and output a risk topology dataset; Dynamically modulate cooling intensity based on the risk topology dataset and obtain partition traffic redistribution instructions.

2. The heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in claim 1, characterized in that: The specific steps for generating the pressure distribution matrix are as follows: Real-time acquisition of contact pressure data between the battery and the cold plate, followed by signal conditioning and dynamic compensation to generate a temperature-compensated pressure signal array; Noise suppression and physical quantity calibration conversion are performed on the temperature-compensated pressure signal array to generate a physical pressure value array. Spatial interpolation expansion and gradient continuity verification are performed on the physical pressure value array to generate a contact pressure feature field; Pressure distribution matrix is ​​generated by integrating pressure threshold partitioning and statistical features of the contact pressure feature field.

3. The heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in claim 2, characterized in that: The traffic decision model is constructed based on a sensor data layer, a strategy reasoning layer, and a dynamic optimization layer.

4. The heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in claim 3, characterized in that: The specific steps for generating the partition flow control signal are as follows: The sensor data layer extracts and vectorizes multi-dimensional features from the pressure distribution matrix to generate a pressure feature vector. The strategy inference layer performs intelligent strategy mapping and flow ratio allocation on the pressure feature vector to generate an initial flow ratio coefficient array. The dynamic optimization layer performs multi-source feedback optimization and dynamic threshold adjustment on the initial flow ratio coefficient array to generate partition flow control signals.

5. The heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in claim 4, characterized in that: The specific steps for obtaining the temperature field matrix and dynamic thermal feature set are as follows: Based on the partitioned flow control signal, synchronous scanning and spectrum-temperature conversion of the fiber grating array are performed to generate a preliminary temperature data array. Spatial interpolation expansion and edge optimization processing are performed on the initial temperature data array to generate a high-resolution temperature field matrix; The dynamic characteristics of the high-resolution temperature field matrix are analyzed to generate a dynamic thermal feature set.

6. The heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in claim 5, characterized in that: The specific steps for generating the thermal coupling parameter set are as follows: Multi-source data spatiotemporal alignment and feature fusion are performed on dynamic hot feature sets to generate spatiotemporally aligned datasets; Perform thermo-electric property correlation and parameter mapping on the spatiotemporally aligned dataset, and output an intermediate set of coupling parameters; Based on the intermediate set of coupling parameters, the lithium plating probability is calculated and a safety margin assessment is performed, and the thermal coupling parameter set is output.

7. The heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in claim 6, characterized in that: The specific steps for outputting the risk topology dataset are as follows. The probability field is reconstructed and the risk region is extracted from the thermally coupled parameter set to generate a risk probability field matrix; The risk probability field matrix is ​​optimized for regional continuity and the risk intensity is calibrated to output a comprehensive risk field matrix; Perform risk region clustering analysis and hierarchical topology construction on the comprehensive risk field matrix, and output a risk topology dataset.

8. The heat dissipation control method for a single-phase immersion liquid-cooled charging pile as described in claim 7, characterized in that: The specific steps for obtaining the partition traffic redistribution instruction are as follows: Hierarchical cooling intensity mapping and dynamic battery state constraints are applied to the risk topology dataset, and a cooling intensity modulation strategy table is output. The cooling intensity modulation strategy table is processed by actuator signal conversion and safety parameter encapsulation, and outputs partition flow redistribution instructions.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the heat dissipation control method for the single-phase immersion liquid-cooled charging pile according to any one of claims 1 to 8.

Images