Battery bump self-adaptive flexible two-phase liquid cold and hot out-of-control early warning method and system

Abstract

The invention discloses a battery bump self-adaptive flexible two-phase liquid cold and hot out-of-control early warning method and system, and relates to the technical field of battery liquid cooling heat dissipation. Comprising the following steps of S1, collecting fitting state data of a cold plate and a battery in real time, and performing data preprocessing; s2, judging whether the battery swells or not, and identifying a swelling abnormal event; s3, self-adaptive flexible regulation and control are carried out on the battery bump working condition according to the bump abnormal event; evaluating the liquid cooling heat conduction efficiency after self-adaptive flexible regulation and control, and carrying out local heat conduction performance optimization regulation; and S4, after the local heat conduction performance optimization adjustment is completed, evaluating the collaborative heat dissipation capability margin, carrying out liquid cooling and heating runaway early warning, and implementing thermal runaway intervention measures. The problems that under the dynamic deformation working conditions of battery swelling and the like, an existing liquid cooling heat dissipation system is difficult to keep efficient heat conduction and stable attachment and lacks a real-time sensing and active early warning mechanism, so that local overheating and thermal runaway risks are difficult to restrain in time are solved.

Description

A method and system for early warning of thermal runaway caused by adaptive flexible two-phase liquid cooling in battery swelling Technical Field

[0001] This invention relates to the field of battery liquid cooling technology, specifically to a method and system for early warning of thermal runaway caused by adaptive flexible two-phase liquid cooling in battery bulging. Background Technology

[0002] With the rapid development of electric vehicles and energy storage systems, battery thermal management has placed higher demands on efficient and intelligent heat dissipation technologies. Liquid cooling, with its excellent thermal conductivity and temperature control efficiency, has become the core technology for current battery thermal management systems. Existing liquid cooling systems are continuously evolving towards a combination of technologies, including indirect cooling plates, direct immersion, and composite phase change technologies. Through synergy with phase change thermal storage materials, they achieve peak temperature reduction and continuous cooling in high heat flux density scenarios.

[0003] For example, invention patent CN119475763A discloses a liquid cooling method and apparatus for a battery pack, relating to the field of liquid cooling. The method includes constructing a simulation model based on the battery cells and coolant in the battery pack; performing discharge simulation on the simulation model to obtain the temperature distribution of the battery cells and the flow rate and temperature of the coolant; determining the flow rate and temperature of the coolant based on the influence relationship between the maximum temperature and maximum temperature difference and the flow rate and temperature of the coolant when both are at their minimum values; and liquid cooling of the battery pack based on the flow rate and temperature of the coolant. Through simulation of the battery pack, the simulation model simulates the power generation state of the actual battery pack. Adjusting the flow rate and temperature of the coolant in the simulation model to minimize both the maximum temperature and maximum temperature difference of the battery pack, liquid cooling of the battery pack is achieved based on the flow rate and temperature values ​​of the simulation model. The entire process has a short cycle and does not require verification with an actual battery pack, thus reducing costs.

[0004] For example, invention patent CN117371304A discloses a method for optimizing the structure of a liquid-cooled lithium-ion battery pack, belonging to the field of liquid cooling heat dissipation technology for electric vehicle power batteries. The method includes: establishing a geometric model of the liquid-cooled lithium-ion battery pack, setting design variables and analysis objectives, and performing mesh generation; the design variables include: inlet flow velocity, inlet temperature, ambient temperature, channel thickness, and channel width; the analysis objectives are the highest temperature and the maximum temperature difference; using a regression model to describe the relationship between the design variables and the analysis objectives, and using response surface methodology to analyze and obtain the regression equation; substituting the regression equation into a particle swarm optimization algorithm to find the optimal structure, and selecting the optimal structural parameters based on the Pareto optimal solution. This invention utilizes response surface methodology to establish a model, which facilitates and quickly finds the optimal comprehensive value to reduce analysis time. The optimized parameters can achieve the best cooling performance, reduce the overall battery temperature, and improve the uniformity of temperature distribution.

[0005] However, under long-term cycling or extreme operating conditions, cell bulging is still difficult to avoid. Under dynamic deformation conditions such as battery bulging, existing liquid cooling systems are unable to maintain efficient heat conduction and stable adhesion, and lack real-time sensing and active early warning mechanisms, making it difficult to suppress the risk of local overheating and thermal runaway in a timely manner.

[0006] Therefore, in order to address the above problems, there is an urgent need for an adaptive flexible two-phase liquid-cooled early warning method and system for battery bulging thermal runaway. Summary of the Invention

[0007] Addressing the shortcomings of existing technologies, this invention provides an adaptive flexible two-phase liquid cooling thermal runaway early warning method and system for battery bulging. It solves the problem that under dynamic deformation conditions such as battery bulging, existing liquid cooling systems are unable to maintain efficient heat conduction and stable adhesion, lack real-time sensing and active early warning mechanisms, resulting in the difficulty in timely suppressing the risk of local overheating and thermal runaway.

[0008] Technical solution

[0009] To achieve the above objectives, the present invention provides the following technical solution: a battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method, comprising the following steps: S1, real-time acquisition of cold plate and battery bonding state data, and data preprocessing operation on the cold plate and battery bonding state data; S2, based on the preprocessed cold plate and battery bonding state data, determining whether the battery is bulging, and identifying bulging abnormal events; S3, for bulging abnormal events, adaptive flexible control of the battery bulging condition; evaluating the liquid cooling thermal conductivity based on the adaptive flexible control of the cold plate and battery bonding state data, and optimizing and adjusting the local thermal conductivity performance based on the liquid cooling thermal conductivity efficiency; S4, after the local thermal conductivity performance optimization and adjustment is completed, evaluating the collaborative heat dissipation capacity margin using the cold plate and battery bonding state data, and issuing a liquid cooling thermal runaway early warning and implementing thermal runaway intervention measures based on the collaborative heat dissipation capacity margin.

[0010] Furthermore, the specific process of real-time acquisition of cold plate and battery bonding status data and data preprocessing of cold plate and battery bonding status data is as follows: Real-time acquisition of cold plate and battery bonding status data, specifically in the following ways and with the following data: Pressure, abscissa deformation, ordinate deformation, and bonding layer thickness from the inner surface of the flexible bonding layer of the cold plate to the outer surface of the battery cell are acquired through pressure sensors and a micro piezoresistive strain sensor array distributed inside the flexible cold plate. The three bonding surfaces include the bottom surface, left side surface, and right side surface. Based on a sliding time window, the corresponding pressure change rate is obtained by dividing the difference between adjacent time values ​​of the pressure of the three continuously acquired bonding surfaces by the window length. Simultaneously, temperature sensors, flow meters, and battery heat flow sensors are used to acquire the cell surface temperature and cold plate inlet temperature. The system collects data on the temperature, cold plate outlet temperature, phase change thermal storage material temperature, coolant flow rate, and cell heating power. It also acquires the specific heat capacity of the coolant, the specific heat capacity of the phase change thermal storage material, the fixed melting temperature of the phase change thermal storage material, the fixed mass of the phase change thermal storage material, and the fixed area of ​​the bonding surface. Digital filtering of the cold plate-battery bonding status data is performed using mean and median filtering. Data collected from all different types of sensors is synchronized and aligned using a unified clock. The cold plate-battery bonding status data is standardized and normalized. An outlier detection algorithm is used to identify and mark abnormal values, signal loss, and sensor malfunctions. For detected short-term data gaps and abnormal discontinuities, a moving average completion algorithm is used to repair missing segments. A cold plate-battery operating condition database is established, and the cold plate-battery bonding status data is written into this database.

[0011] Furthermore, based on the pre-processed data on the bonding status between the cold plate and the battery, the specific process for determining whether the battery is bulging is as follows: Obtain the current pressure, corresponding pressure change rate, abscissa deformation, and ordinate deformation of the three bonding surfaces between the battery and the cold plate; for each bonding surface, calculate the sum of the squares of the abscissa deformation and the square root of the ordinate deformation to obtain the spatial composite deformation; simultaneously, multiply the pressure change rate by a constant value of time and add a constant to obtain the pressure dynamic factor; multiply the spatial composite deformation by the pressure dynamic factor to obtain the single-sided bulging activity value; sum the single-sided bulging activity values ​​of the three bonding surfaces and calculate the average to obtain the comprehensive bulging activity value.

[0012] Furthermore, the specific process for identifying bulging abnormal events is as follows: The overall bulging activity value is calculated in real time and compared with the activity threshold; when the overall bulging activity value is less than the activity threshold, it is determined to be a normal change, and only the overall bulging activity value is continuously monitored to maintain conventional liquid cooling; when the overall bulging activity value is greater than or equal to the activity threshold, it is determined to be a warning state and marked as a bulging abnormal event. The cold plate and battery bonding status data corresponding to the bulging abnormal event and the overall bulging activity value are uploaded to the central controller and proceed to the next process; all overall bulging activity values ​​and bulging abnormal events are written into the cold plate battery operating condition database, and the activity threshold is optimized using cluster analysis.

[0013] Furthermore, regarding the abnormal swelling event, the specific process of adaptive flexible control of the battery swelling condition is as follows: receiving the abnormal swelling event, the corresponding cold plate and battery bonding status data, and the comprehensive swelling activity value, and driving the flexible structure thermal optimization control: adjusting the current of the magnetically coupled electromagnet to increase the bonding pressure of the three bonding surfaces of the cold plate; driving the elastic hinge to divide the sub-cold plate, while the structured flexible layer of the flexible cold plate responds with a gradient modulus design, that is, the hard edge area maintains the frame stability, the soft middle area expands with the swelling, and the wavy cross section naturally extends along the deformation direction of the bonding surface; controlling the continuous connection of the graphene fiber thermal conductive network embedded in the flexible cold plate, and the coolant phase change cycle in the cold plate runs at the normal rate; if the comprehensive swelling activity value is detected to be continuously increasing for a period of time greater than the safety allowable threshold, thermal runaway interception is performed in real time, and the comprehensive swelling activity value is connected to the battery management system to trigger the charging and discharging power limitation: during the charging stage, the current is forcibly reduced; during the discharging stage, the circuit where the bulging cell is located is cut off, and the load is transferred to the healthy cell.

[0014] Furthermore, based on the adaptive flexible adjustment data of the cold plate and battery bonding state, the specific process for evaluating the liquid cooling thermal conductivity is as follows: After the flexible structure's thermal conductivity is optimized and adjusted, the cell surface temperature, cold plate inlet temperature, cold plate outlet temperature, coolant flow rate, and coolant specific heat capacity are obtained; the difference between the current cold plate outlet temperature and the current cold plate inlet temperature is calculated, and multiplied by the coolant flow rate and coolant specific heat capacity to obtain the cold plate heat dissipation flow rate; the cold plate heat dissipation flow rate is divided by the difference between the current cell surface temperature and the current cold plate inlet temperature to obtain the cold plate unit temperature difference heat dissipation intensity value; simultaneously, the pressure on the three bonding surfaces is obtained. The parameters are: spatial composite deformation, fixed area of ​​the bonding surface, and bonding layer thickness. When the pressure is greater than the contact threshold and the spatial composite deformation is greater than the deformation threshold, it is determined to be an effective bonding area. The three bonding surfaces are then compared to determine if they are effective bonding areas. The fixed areas of the bonding surfaces in the effective bonding areas are added together to obtain the effective thermally conductive contact area. The bonding layer thicknesses of the three bonding surfaces are summed and averaged to obtain the equivalent thickness of the bonding layer. The equivalent geometric factor of the thermal conduction path is obtained by dividing the effective thermally conductive contact area by the equivalent thickness of the bonding layer. The equivalent thermal resistance is then multiplied by the heat dissipation intensity per unit temperature difference of the cold plate and the equivalent geometric factor of the thermal conduction path to obtain the equivalent thermal resistance assessment value.

[0015] Furthermore, the specific process of optimizing and adjusting the local thermal conductivity performance based on the liquid cooling thermal conductivity efficiency is as follows: The equivalent thermal resistance assessment value is compared with the thermal conductivity threshold in real time. When the equivalent thermal resistance assessment value is greater than or equal to the thermal conductivity threshold, it is determined that the bonding and thermal conductivity of the flexible structure have been well restored after thermal optimization and adjustment, and the cold plate enters the normal liquid cooling heat dissipation working mode. When the equivalent thermal resistance assessment value is less than the thermal conductivity threshold, the bonding status data of the cold plate and battery on the three bonding surfaces is obtained, abnormal bonding areas are located, and local thermal conductivity performance is optimized and adjusted for these abnormal bonding areas: the local bonding pressure on the three bonding surfaces is increased, the magnetic coupling distribution is adjusted, the pump speed is increased to adjust the coolant flow rate, and the coolant flow velocity is increased, simultaneously linking the phase change heat storage material embedded in the cold plate and battery bonding structure to absorb and store heat. The equivalent thermal resistance assessment value, abnormal bonding areas, and corresponding thermal bottleneck repair measures are uploaded to the cold plate battery operating condition database and proceed to the next process.

[0016] Furthermore, after the local thermal conductivity optimization and adjustment are completed, the specific process of evaluating the collaborative heat dissipation margin using the cold plate and battery bonding state data is as follows: Obtain the phase change thermal storage material temperature, specific heat capacity, fixed melting temperature, fixed mass, cold plate heat dissipation heat flow, and cell heating power; calculate the difference between the fixed melting temperature and the current phase change thermal storage material temperature, and multiply it by the specific heat capacity and fixed mass to obtain the phase change thermal storage material heat dissipation heat flow; based on a sliding time window, statistically analyze the cold plate heat dissipation heat flow and the phase change thermal storage material heat dissipation heat flow within the current window, and calculate their standard deviations respectively; sum the standard deviations of the cold plate heat dissipation heat flow and the phase change thermal storage material heat dissipation heat flow to obtain the total heat dissipation heat flow fluctuation value; add the phase change thermal storage material heat dissipation heat flow to the cold plate heat dissipation heat flow, and subtract the cell heating power to obtain the multi-channel net heat dissipation redundancy value; divide the multi-channel net heat dissipation redundancy value by the sum of the total heat dissipation heat flow fluctuation value and the minimum constant value to obtain the collaborative heat dissipation redundancy margin value.

[0017] Furthermore, the specific process of providing early warning and implementing intervention measures for liquid cooling thermal runaway based on the collaborative heat dissipation capacity margin is as follows: Real-time comparison of the collaborative heat dissipation redundancy margin value with the thermal runaway threshold; when the collaborative heat dissipation redundancy margin value is greater than the thermal runaway threshold, maintaining conventional liquid cooling heat dissipation regulation and thermal energy storage allocation; when the collaborative heat dissipation redundancy margin value is less than or equal to the thermal runaway threshold, increasing the coolant flow rate and activating the standby pump group; adjusting the heat release rate of the phase change thermal storage material to release stored thermal energy in advance; optimizing the heat dissipation capacity allocation of each channel using a multi-objective programming algorithm; and maintaining the status quo when the collaborative heat dissipation redundancy margin value is less than or equal to the thermal runaway threshold. If the time exceeds the safety tolerance threshold, a thermal runaway warning signal is issued, triggering the battery management system to limit the charge and discharge rate and cut off some abnormal circuits; the cold plate is forced to maintain the maximum allowable coolant flow rate and the coolant circulation rate is increased, triggering external air cooling emergency response; warning signals are pushed to the operation and maintenance platform in real time, and all collaborative heat dissipation redundancy margin values, cell temperatures, warning signals and thermal runaway intervention measures are recorded. The correlation between collaborative heat dissipation redundancy margin values, cell temperatures, thermal runaway intervention measures and actual cell temperature changes under different operating conditions is analyzed regularly, and the thermal runaway threshold and thermal runaway intervention measures are corrected using a sliding window adaptive clustering algorithm.

[0018] The second aspect of this invention provides a battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning system, comprising: a multi-source state data acquisition and preprocessing module for real-time acquisition of cold plate and battery bonding state data, and performing data preprocessing on the cold plate and battery bonding state data; a battery bulging comprehensive discrimination module for determining whether the battery is bulging based on the preprocessed cold plate and battery bonding state data, and identifying bulging abnormal events; a structural adaptation and thermal conduction path repair module for adaptive flexible control of the battery bulging condition in response to bulging abnormal events; evaluating liquid cooling thermal conductivity based on the cold plate and battery bonding state data after adaptive flexible control, and optimizing local thermal conductivity performance based on the liquid cooling thermal conductivity efficiency; and a collaborative heat dissipation assessment and thermal runaway early warning module for evaluating the collaborative heat dissipation capacity margin using the cold plate and battery bonding state data after the local thermal conductivity performance optimization adjustment is completed, and performing liquid cooling thermal runaway early warning and implementing thermal runaway intervention measures based on the collaborative heat dissipation capacity margin.

[0019] Beneficial effects

[0020] The present invention has the following beneficial effects: (1) The present invention, by collecting multiple types of sensors such as pressure, deformation, thickness and temperature, realizes all-round and real-time perception of the bonding state of the cold plate and the battery on three sides. Combined with the comprehensive bulging activity value, it realizes high sensitivity, early and abnormal identification of battery bulging, laying a data foundation for subsequent precise intervention.

[0021] (2) In response to abnormal bulging events, this invention can adaptively adjust the bonding pressure, structural deformation mode and heat conduction network. Combined with magnetic coupling and gradient partitioning of flexible structure, it can achieve active slow release of local bulging and multi-face synchronous bonding, dynamically repair the bottleneck of heat conduction path, and effectively improve the overall heat conduction efficiency and safety redundancy of the bonding surface.

[0022] (3) This invention integrates cold plate liquid cooling and phase change thermal storage materials, calculates the collaborative heat dissipation redundancy margin value, dynamically quantifies the real-time balance between multi-channel heat dissipation capacity and cell heat load, realizes intelligent allocation of heat dissipation resources and multi-channel collaborative optimization, and enhances the ability to cope with large heat flow and extreme working conditions.

[0023] (4) This invention, through big data and self-learning algorithms, can record and analyze key criteria, early warning signals and intervention measures throughout the entire process, correct thermal runaway threshold and response logic, realize early intelligent warning of thermal runaway, multi-level safety intervention and active flow interruption and limiting, and improve safety and long-term operational stability through continuous adaptive optimization.

[0024] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0025] Figure 1 is a flowchart of the early warning method for adaptive flexible two-phase liquid cooling thermal runaway of battery bulging.

[0026] Figure 2 is a block diagram of the battery bulging adaptive flexible two-phase liquid cooling thermal runaway early warning system;

[0027] Figure 3 is a schematic diagram of the integrated structure of the three-sided flexible corrugated cold plate and the battery.

[0028] Figure 4 shows the dynamic change of the redundancy margin of the battery liquid cooling multi-channel coordinated heat dissipation.

[0029] Figure 5 is a flowchart of the multi-criteria intervention process for adaptive flexible two-phase liquid cooling thermal runaway of battery bulging.

[0030] In the diagram, 1 is the wavy cross-section; 2 is the coolant tank base; and 3 is the battery. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. As those skilled in the art will understand, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Please refer to Figures 1-5. This invention provides a technical solution: a method and system for early warning of battery bulging adaptive flexible two-phase liquid cooling thermal runaway. As shown in Figure 1, it includes the following steps: S1, real-time acquisition of cold plate and battery bonding state data, and data preprocessing of the cold plate and battery bonding state data; S2, based on the preprocessed cold plate and battery bonding state data, determining whether battery 3 is bulging, and identifying bulging abnormal events; S3, for bulging abnormal events, adaptive flexible control of the battery bulging condition; evaluating liquid cooling thermal conductivity based on the cold plate and battery bonding state data after adaptive flexible control, and optimizing local thermal conductivity performance based on the liquid cooling thermal conductivity efficiency; S4, after the local thermal conductivity performance optimization is completed, evaluating the margin of collaborative heat dissipation capacity using the cold plate and battery bonding state data, and issuing early warning of liquid cooling thermal runaway and implementing thermal runaway intervention measures based on the margin of collaborative heat dissipation capacity.

[0033] Specifically, the process of real-time acquisition of cold plate and battery bonding status data and data preprocessing of this data is as follows: Real-time acquisition of cold plate and battery bonding status data is conducted through pressure sensors and a micro piezoresistive strain sensor array distributed within the flexible cold plate. This data includes the pressure, abscissa deformation, ordinate deformation, and bonding layer thickness from the inner surface of the flexible bonding layer of the cold plate to the outer surface of the battery cell. The three bonding surfaces are the bottom, left, and right sides. The pressure sensors are high-precision thin-film pressure sensors, capable of accurately monitoring real-time pressure distribution changes at the battery interface. The micro piezoresistive strain sensors are arranged in an array to achieve multi-sensory acquisition of the bonding status data. The spatial resolution of point deformation and strain is improved to capture localized minor bulges and depressions in the bonding layer. The bonding layer thickness, representing the shortest physical distance between the inner surface of the flexible bonding layer of the cold plate and the outer surface of the battery cell, is a crucial parameter for characterizing interface tightness and bonding uniformity. Based on a sliding time window, the pressure change rate is obtained by dividing the difference between adjacent time values ​​of the pressure on three continuously acquired bonding surfaces by the window length. This pressure change rate measures the dynamic change rate of pressure at the bonding interface and serves as a sensitive criterion for early anomaly identification. Simultaneously, temperature sensors, flow meters, and battery heat flow sensors are used to collect data on cell surface temperature, cold plate inlet temperature, cold plate outlet temperature, phase change thermal storage material temperature, coolant flow rate, and cell heating power. The phase change thermal storage material (PCS) temperature directly reflects the phase change heat absorption and release process and is an important input for evaluating the synergistic heat dissipation capability. The specific heat capacity of the coolant, the specific heat capacity of the PCS, the fixed melting temperature of the PCS, the fixed mass of the PCS, and the fixed area of ​​the bonding surface are obtained. The specific heat capacity of the coolant, the specific heat capacity of the PCS, and the fixed melting temperature of the PCS are directly given from the material parameter library according to the material type. The fixed mass of the PCS is obtained through precise weighing and filling during initial assembly. The fixed area of ​​the bonding surface is set once according to the design dimensions of the cell and the cold plate. The bonding state data of the cold plate and the battery are digitally filtered using mean filtering and median filtering. Mean filtering effectively suppresses high-frequency noise interference, while median filtering... Isolated outliers have excellent rejection capabilities, and the combination of these two features improves the overall signal quality. Data collected by all different types of sensors are synchronized and aligned according to a unified clock. All sensor signals are collected using the standard timestamp issued by the main controller. A synchronous sampling algorithm is used to achieve integrated processing of multi-channel data, ensuring that subsequent multi-source feature extraction is completed under the same time reference, thus improving the accuracy of multi-modal information fusion. Furthermore, the data on the bonding status of the cold plate and battery are standardized and normalized. An outlier detection algorithm is used to identify and mark outliers, signal loss, and sensor malfunctions. For detected short-term data loss and abnormal discontinuities, a moving average completion algorithm is used to repair missing segments, greatly reducing the impact of short-term acquisition loss on feature calculation.A cold-plate battery operating condition database is established, incorporating the bonding status data between the cold plate and the battery. This database supports time-series queries, tag retrieval, and large-scale high-concurrency writes, providing underlying data support for subsequent fault tracing, algorithm optimization, criterion training, and status visualization. This ensures that the operating conditions throughout the entire battery lifecycle are traceable, verifiable, and optimizable.

[0034] In this implementation scheme, a multi-physics distributed sensor array is used to achieve full real-time data acquisition and high-precision preprocessing of the bonding state of the cold plate and battery 3 on three sides. This comprehensively reflects the dynamic changes of key physical quantities such as interface pressure, deformation, and thickness. Combined with a series of digital signal processing steps, the reliability of the data and the ability to identify early bulging, bonding abnormalities, and sensor malfunctions are significantly improved. The establishment of a cold plate-battery operating condition database enables closed-loop management and traceability of the entire bonding process, providing a solid data foundation and strong technical support for subsequent intelligent criterion calculations, adaptive control, and collaborative heat dissipation optimization.

[0035] Specifically, based on the pre-processed data on the bonding state between the cold plate and the battery, the process for determining whether battery 3 has bulged is as follows: The current pressure, corresponding pressure change rate, abscissa deformation, and ordinate deformation of the three bonding surfaces of battery 3 and the cold plate are obtained. Pressure reflects the tightness of the bonding, and the pressure change rate reflects the speed of bulging. The abscissa and ordinate deformations correspond to the physical displacement of each bonding surface along the principal axis and vertical axis, respectively, enabling sensitive detection of local bulging deformation. For each bonding surface, the sum of the squares of the abscissa and ordinate deformations is calculated, and the square root is taken to obtain the spatial composite deformation. The spatial composite deformation is the three-dimensional vector composite displacement, comprehensively... The absolute amplitude of the bulge on the bonding surface is the fundamental physical quantity for determining the bulge activity value. Simultaneously, the pressure change rate is multiplied by a time constant and then added to obtain a pressure dynamic factor. The time constant is 0.2, used to adjust the weight of the pressure change rate, while the constant ensures the lower limit of the pressure dynamic factor is not less than 1. The pressure dynamic factor reflects the dynamic weighting effect of pressure changes at the bonding interface on bulge activity, avoiding misjudgments caused by simple displacement. Multiplying the spatial composite deformation variable by the pressure dynamic factor yields a single-sided bulge activity value, coupling the spatial bulge amplitude with dynamic pressure fluctuations, which comprehensively reflects the abnormal activity level of the bulge. Finally, the single-sided bulge activity values ​​of the three bonding surfaces are summed and averaged to obtain the comprehensive bulge activity value.

[0036] The specific formula for the overall drum activity value is as follows:

[0037] ;

[0038] in, It represents the comprehensive bulging activity value, which quantitatively reflects the severity and dynamic activity of bulging on the three bonding surfaces of battery 3 and cold plate at a certain moment. It takes into account both the spatial deformation amplitude and the pressure change rate, and can promptly identify high-risk working conditions such as rapid bulging and sudden deformation, thereby improving the sensitivity and accuracy of safety control. The horizontal direction deformation of the i-th bonding surface reflects the current horizontal bulge of the i-th bonding surface; The vertical direction deformation of the i-th bonding surface reflects the current vertical bulge of the i-th bonding surface; This represents the rate of pressure change at the i-th bonding surface, reflecting the force acting on the bonding surface. This represents a constant time value, with a value of 0.2. Represents the spatial composite shape variable of the i-th bonding surface, which measures the actual absolute magnitude of the bulge; This represents the pressure dynamic factor of the i-th contact surface. By weighting the pressure change trend into the comprehensive bulge activity value, the weight of rapidly developing bulges is increased. This represents the single-sided bulge activity value of the i-th bonding surface, which quantitatively reflects the severity and dynamic activity of the bulge at a certain moment on the i-th bonding surface. The value takes into account both the size of the bulge and the rate of change.

[0039] In this implementation scheme, by integrating multiple physical quantities such as three-sided bonding pressure, deformation, and pressure change rate, a spatial composite deformation and pressure dynamic factor is constructed, which can highly sensitively and comprehensively quantify and identify the battery bulging state. The comprehensive bulging activity value not only improves the accuracy of early anomaly detection but also provides a solid data foundation for subsequent graded control and adaptive repair, thereby enhancing the intelligent identification and response capability to bulging risks.

[0040] Specifically, the process for identifying bulging abnormal events is as follows: The overall bulging activity value is calculated in real time and compared with an activity threshold. When the overall bulging activity value is less than the activity threshold, it is considered a normal change, and only the overall bulging activity value is continuously monitored while maintaining regular liquid cooling. When the overall bulging activity value is greater than or equal to the activity threshold, it is considered a warning state and marked as a bulging abnormal event. The corresponding cold plate and battery bonding status data and the overall bulging activity value are uploaded to the central controller, and the process proceeds to the next step. All overall bulging activity values ​​and bulging abnormal events are written into the cold plate battery operating condition database, and the activity threshold is optimized using cluster analysis: Utilizing K-means and density clustering unsupervised learning algorithms, the optimal classification boundary is identified by calling the distribution characteristics of historical cold plate and battery bonding status data and overall bulging activity values ​​from the cold plate battery operating condition database. This enables online dynamic adaptive adjustment of the activity threshold, improving the applicability and accuracy of the warning criteria under different battery types, environments, and operating conditions, and ensuring that the warning sensitivity and false alarm rate remain within the optimal range.

[0041] In this implementation plan, intelligent identification and precise hierarchical early warning of bulging anomalies are achieved by utilizing dynamic threshold comparison. By optimizing the activity threshold using cluster analysis, the warning sensitivity can be adaptively adjusted, effectively reducing the risk of false alarms and missed alarms. End-to-end data synchronization to the central controller and the cold plate battery operating condition database not only enhances the real-time performance and automation of anomaly response but also provides solid data support for subsequent control decisions and safety management, improving the intelligent monitoring capability for thermal safety.

[0042] Specifically, the adaptive flexible control process for battery bulging abnormal events is as follows: Receiving the bulging abnormal event, the corresponding cold plate and battery bonding status data, and the overall bulging activity value, the flexible structure's thermal optimization control is driven: adjusting the current of the magnetically coupled electromagnet to increase the bonding pressure of the three bonding surfaces of the cold plate, achieving real-time dynamic distribution of local bonding pressure on each bonding surface, ensuring that the cold plate and battery 3 maintain efficient physical contact and unobstructed thermal conduction pathways even when bulging abnormalities occur; driving the elastic hinge to divide the sub-cold plate, while the structured flexible layer of the flexible cold plate responds with a gradient modulus design, i.e., the hard edge region maintains frame stability, the soft middle region expands with the bulging, and the wavy section 1 naturally extends along the deformation direction of the bonding surface; the elastic hinge adopts a partitioned controllable structure, which can independently adjust the cold plate partitions according to the size and position of the bulging area, enhancing the elasticity of the middle region and strengthening the rigidity of the edge region, achieving a spatially differentiated adaptive response. The wavy section 1 design can naturally deform with the bulging direction of the bonding surface, greatly improving bonding flexibility and thermal uniformity. Gradient modulus refers to the mechanical properties of a cold plate structure that gradually transitions from rigidity to flexibility from the edge to the center through material composite and layering processes, achieving compatibility between overall stability and local self-adaptation. The graphene fiber thermal conductive network embedded within the flexible cold plate is kept continuously connected, and the coolant phase change cycle within the cold plate operates at a conventional rate. The graphene fiber thermal conductive network, made of highly thermally conductive carbon-based fiber material, forms multi-dimensional thermal channels through its continuous fiber structure embedded between the layers of the flexible cold plate. This effectively ensures that heat can still be rapidly transferred to the liquid cooling channels of the cold plate under local abnormalities such as bulging and warping, improving thermal conductivity. If the overall bulging activity value is detected to be continuously increasing for a period exceeding the safety tolerance threshold, thermal runaway is intercepted in real time. The overall bulging activity value is connected to the battery management system, triggering charge and discharge power limits: during the charging phase, the current is forcibly reduced; during the discharging phase, the circuit containing the bulging cell is disconnected, and the load is transferred to the healthy cell. The safety tolerance threshold is the tolerable limit determined based on historical fault statistics. Real-time thermal runaway interception forces the battery BMS safety control process through high-priority logic, adjusts the charging and discharging strategy as needed, and realizes the linkage between abnormal self-healing and safety protection to avoid the bulging from worsening.

[0043] Figure 3 shows a schematic diagram of the integrated structure of a three-sided flexible corrugated cold plate and a battery. The corrugated section 1 is a flexible thermally conductive layer in the three-sided bonding area between the cold plate and battery 3. It employs a structured corrugated design, allowing it to naturally extend along the bonding surface direction as the battery bulges and deforms. The flexible layer not only achieves adaptive bonding with hard edges and a soft center through a gradient modulus design, but also incorporates a graphene fiber thermally conductive network, ensuring efficient heat transfer to the cold plate channels even during battery bulging deformation, effectively preventing heat dissipation bottlenecks and localized overheating. The coolant tank base 2 is a liquid cooling circulation channel with multiple component flow channels inside. The coolant can flow around the bonding structure, working synergistically with the phase change thermal storage material to continuously remove the heat generated by battery 3, achieving safe heat dissipation under high heat flux density. Battery 3 is the cell module to be temperature-controlled, with all three sides tightly bonded to the flexible cold plate. The bonding status of battery 3 and the three sides of the cold plate, including pressure, deformation, and thickness changes, can be sensed in real time using pressure sensors and a micro piezoresistive sensor array.

[0044] In this implementation scheme, by integrating magnetic coupling electromagnetic voltage regulation, elastic hinge partitioning adaptive design, gradient modulus structure design, and graphene thermal conductive network, dynamic adaptive control of the contact pressure and structural deformation of the cold plate and battery 3 on three sides during abnormal bulging is achieved, ensuring the continuity and uniformity of the efficient heat conduction path. Simultaneously, the battery management system implements real-time current limiting and current interruption safety intervention measures when the abnormality intensifies, effectively improving the adaptive bulging mitigation capability and overall thermal safety protection level of battery 3.

[0045] Specifically, based on the adaptive flexible adjustment data of the cold plate and battery bonding state, the process for evaluating the liquid cooling thermal conductivity efficiency is as follows: After the flexible structure's thermal conductivity is optimized and adjusted, the cell surface temperature, cold plate inlet temperature, cold plate outlet temperature, coolant flow rate, and coolant specific heat capacity are obtained; the difference between the current cold plate outlet temperature and the current cold plate inlet temperature is calculated and multiplied by the coolant flow rate and coolant specific heat capacity to obtain the cold plate heat dissipation flow rate, which reflects the total heat carried away by the liquid cooling channel per unit time and is the core engineering parameter of heat dissipation capacity; the cold plate heat dissipation flow rate is divided by the difference between the current cell surface temperature and the current cold plate inlet temperature to obtain the cold plate heat dissipation intensity value per unit temperature difference, which is used to quantify the heat flow output capacity per unit temperature difference. The larger the value, the smoother the heat conduction path and the higher the interface heat dissipation efficiency; at the same time, the pressure, spatial composite deformation, fixed area of ​​the bonding surface, and bonding layer thickness of the three bonding surfaces are obtained; when the pressure is greater than the contact threshold... When the spatial composite deformation exceeds the deformation threshold, it is determined to be an effective bonding area. To determine whether the three bonding surfaces are effective bonding areas, the fixed areas of the bonding surfaces in the effective bonding areas are added together to obtain the effective thermal conductive contact area. The effective thermal conductive contact area directly reflects the total effective area of ​​the main heat flow channel at the interface. The bonding layer thicknesses of the three bonding surfaces are accumulated and averaged to obtain the equivalent thickness of the bonding layer. The equivalent thickness is the weighted average of the physical thicknesses of the three surfaces and is a structural parameter characterizing the overall thermal resistance of the interface. The effective thermal conductive contact area is divided by the equivalent thickness of the bonding layer to obtain the equivalent geometric factor of the heat conduction path, which reflects the effective channel size when the heat flow passes through the bonding interface. The larger the value, the better the thermal conductivity. The equivalent thermal resistance evaluation value is obtained by multiplying the heat dissipation intensity value per unit temperature difference of the cold plate by the equivalent geometric factor of the heat conduction path, which characterizes the thermal conductivity of the three interfaces of the cold plate battery. It is a key criterion for judging the health of the heat conduction path, locating thermal bottlenecks, and guiding adaptive repair.

[0046] The specific formula for evaluating the equivalent thermal resistance is as follows:

[0047] ;

[0048] In the formula, This represents the equivalent thermal resistance (ETR) assessment value, used to dynamically evaluate the ETR of the interface between the cold plate and the three sides of the battery. Specifically, it measures the efficiency of heat transfer from the cell to the cold plate flow channel under the current bulging and bonding conditions. A high ETR indicates good thermal conductivity, good bonding, and efficient heat transfer; a low ETR indicates poor bonding, severe bulging, and interface deterioration, resulting in obstructed heat flow and a tendency for localized overheating. This value guides the adaptive adjustment of the cold plate structure, determines whether bonding optimization and physical repair are needed. It represents the heat flow rate of the cold plate, which is the actual heat carried away by the coolant flow channel per unit time. It reflects the total heat dissipation capacity of the cold plate and is an important coupling parameter between the structure and thermal performance. This indicates the surface temperature of the battery cell, reflecting the actual operating temperature of the heat source in the bulging area; it is the temperature at the heat input end. This indicates the inlet temperature of the cold plate, reflecting the actual temperature of the coolant before it enters the cold plate; it is the temperature at the heat flow output end. It represents the effective thermally conductive contact area, reflecting the current ability of the bonding surface to cover space efficiently for heat transfer; It represents the equivalent thickness of the bonding layer. The actual physical distance between the cold plate and the cell surface is measured at multiple points, reflecting the length of the path required for heat conduction across the interface. It represents the heat dissipation intensity value per unit temperature difference of the cold plate, and the actual heat dissipation intensity of the cold plate under a unit temperature difference. It quantifies the relationship between the heat removal capacity of the cold plate and the temperature difference, and reflects the thermal conductivity efficiency of the synergy between structure and cooling. It represents the equivalent geometric factor of the heat conduction path, reflecting the spatial heat conduction conditions of the mating surface. A larger area or thinner thickness results in better heat conduction, while a smaller area or thicker thickness results in poorer heat conduction.

[0049] In this implementation plan, dynamic and quantitative assessment of the liquid cooling thermal conductivity efficiency at the interface between the cold plate and the battery is achieved through precise acquisition and calculation of multi-source physical quantities. By using indicators such as heat flow rate, heat dissipation intensity per unit temperature difference, effective thermal contact area, and equivalent thickness, the health status of the heat conduction path can be determined in real time, interface thermal bottlenecks can be quickly located, and a scientific basis can be provided for subsequent adaptive repair and heat dissipation optimization.

[0050] Specifically, the process of optimizing and adjusting local thermal conductivity based on liquid cooling thermal efficiency is as follows: The equivalent thermal resistance assessment value is compared with the thermal conductivity threshold in real time. When the equivalent thermal resistance assessment value is greater than or equal to the thermal conductivity threshold, it is determined that the bonding and thermal conductivity of the flexible structure have been well restored after thermal optimization and adjustment, and the cold plate enters the normal liquid cooling heat dissipation working mode. When the equivalent thermal resistance assessment value is less than the thermal conductivity threshold, the bonding status data of the cold plate and battery on the three bonding surfaces is obtained to locate abnormal bonding areas. For these abnormal bonding areas, local thermal conductivity optimization and adjustment are performed: the current of the magnetically coupled electromagnet is finely adjusted by the main control to increase the local bonding pressure on the three bonding surfaces, ensuring that the abnormal areas recover sufficient bonding force and thermal conductivity in a short time. Based on the real-time abnormal bonding area, the electromagnetic excitation is spatially differentiated and the magnetic coupling distribution is adjusted. The pump speed is increased to regulate the coolant flow rate and enhance the heat exchange capacity of the liquid cooling channel, shortening the thermal resistance bottleneck relief time. Simultaneously, the phase change thermal storage material embedded in the bonding structure of the cold plate and battery 3 absorbs and stores heat. When the temperature in the abnormal bonding area reaches the fixed melting temperature of the phase change thermal storage material, it is triggered. The phase change material efficiently absorbs battery heat during the solid-liquid phase change process, realizing hot spot heat absorption and temperature peak reduction, and improving thermal safety redundancy under extreme conditions. The equivalent thermal resistance assessment value, abnormal bonding area and corresponding thermal bottleneck repair measures are uploaded to the cold plate battery operating condition database and proceed to the next process.

[0051] In this implementation scheme, by assessing the health status of the heat conduction pathway in real time, regional thermal resistance bottlenecks at the interface between the cold plate and battery 3 can be accurately located. Dynamic measures such as localized bonding pressure adjustment, magnetic coupling optimization, coolant flow rate enhancement, and phase change thermal storage synergistic repair can be implemented to achieve rapid adaptive repair of localized thermal conduction anomalies and restoration of thermal management capabilities. The archiving of operating data and synchronization of execution results throughout the entire process provide a solid data foundation for intelligent optimization and thermal runaway prevention, enhancing the self-healing ability, safety margin, and operational reliability of the liquid cooling system.

[0052] Specifically, after the local thermal conductivity optimization and adjustment are completed, the process of evaluating the margin of collaborative heat dissipation capacity using the cold plate and battery bonding state data is as follows: Obtain the phase change thermal storage material temperature, specific heat capacity, fixed melting temperature, fixed mass, cold plate heat dissipation heat flow, and cell heating power; calculate the difference between the fixed melting temperature and the current temperature of the phase change thermal storage material, and multiply it by the specific heat capacity and fixed mass to obtain the heat dissipation heat flow of the phase change thermal storage material, which is the maximum heat that the material can collaboratively absorb per unit time under the current phase change process, directly reflecting the heat storage buffer capacity; based on a sliding time window... The heat flow rates of the cold plate and the phase change thermal storage material within the current window are statistically analyzed, and their standard deviations are calculated separately. The standard deviations of these two heat flow rates are then summed to obtain the total heat flow fluctuation value. The standard deviation measures the intensity of the heat flow fluctuation; a smaller fluctuation value indicates more stable heat dissipation. The heat flow rates of the phase change thermal storage material and the cold plate are added together, and the cell heating power is subtracted to obtain the multi-channel net heat dissipation redundancy value. This value represents the instantaneous net difference between the multi-channel heat dissipation capacity and the actual heating load. A positive value indicates that there is still a margin, while a negative value indicates thermal risk. The multi-channel net heat dissipation redundancy value is divided by the sum of the total heat flow fluctuation value and the minimum constant value to obtain the collaborative heat dissipation redundancy margin value. The minimum constant value is set to 0.5 to ensure that the denominator is not zero. The collaborative heat dissipation redundancy margin value comprehensively reflects the net heat dissipation margin and fluctuation stability of the multi-channel system, and is a core safety criterion guiding thermal runaway early warning, energy allocation optimization, and multi-level intervention decisions.

[0053] The specific formula for the collaborative heat dissipation redundancy margin value is as follows:

[0054] ;

[0055] In the formula, This represents the collaborative heat dissipation redundancy margin value, used to dynamically assess whether the net heat dissipation capacity margin at the current moment is sufficient to cover the actual heat generation power of the battery cell, while taking into account the fluctuation of heat dissipation capacity and safety margin. When the collaborative heat dissipation redundancy margin value is large, it indicates that the heat dissipation redundancy is sufficient and the safety margin is high. When the collaborative heat dissipation redundancy margin value is small, it is a warning that active heat dissipation adjustment and safety intervention are required. It represents the heat dissipation flow rate of the phase change thermal storage material, reflecting in real time the rate at which the phase change thermal storage material absorbs or releases heat from the battery. It represents the heat flow rate of the cold plate, which is the actual heat carried away by the coolant flow channel per unit time, reflecting the total heat dissipation capacity of the cold plate; It indicates the heating power of the battery cell, reflecting the total heat power that needs to be removed. This represents the total heat flow fluctuation value, reflecting the short-term fluctuation intensity of heat dissipation capacity; This represents a very small constant value, with a value of 0.5. This represents the net heat dissipation redundancy value of multiple channels, reflecting in real time the difference between the total heat removed by all current heat dissipation channels and the actual heat generation power of the battery cell. It is the net margin between the current total heat dissipation capacity and the heat load. A net heat dissipation redundancy value of multiple channels greater than zero indicates that the heat dissipation capacity is greater than the heat load and there is redundancy space; a net heat dissipation redundancy value of multiple channels less than zero indicates that the load exceeds the heat dissipation limit and there is a risk of overheating.

[0056] In this embodiment, Table 1 is a data table of collaborative heat dissipation redundancy margin values. The table details the heat dissipation flow of the phase change thermal storage material, the heat dissipation flow of the cold plate, the heating power of the battery cell, the total heat dissipation flow fluctuation value, and the collaborative heat dissipation redundancy margin value corresponding to five different time points. Specifically, at time 1, the heat dissipation flow of the phase change thermal storage material is 60, the heat dissipation flow of the cold plate is 150, the heating power of the battery cell is 180, the total heat dissipation flow fluctuation value is 12, and the collaborative heat dissipation redundancy margin value is 2.40; at time 2, the heat dissipation flow of the phase change thermal storage material is 80, the heat dissipation flow of the cold plate is 180, the heating power of the battery cell is 190, the total heat dissipation flow fluctuation value is 14, and the collaborative heat dissipation redundancy margin value is 4.83; at time 3... At time 4, the heat dissipation flow of the phase change thermal storage material is 100, the heat dissipation flow of the cold plate is 220, the heating power of the battery cell is 200, the total heat dissipation flow fluctuation value is 13, and the collaborative heat dissipation redundancy margin value is 8.89. At time 5, the heat dissipation flow of the phase change thermal storage material is 120, the heat dissipation flow of the cold plate is 240, the heating power of the battery cell is 230, the total heat dissipation flow fluctuation value is 20, and the collaborative heat dissipation redundancy margin value is 6.34.

[0057] Table 1. Data on Cooperative Heat Dissipation Redundancy Margin

[0058]

[0059] Figure 4 shows the dynamic change of the multi-channel coordinated heat dissipation redundancy margin of the battery liquid cooling system. The bar chart represents three thermal parameters at different times: heat flow rate of the phase change thermal storage material, heat flow rate of the cold plate, and heating power of the battery cell. The black solid line represents the dynamic change trend of the coordinated heat dissipation redundancy margin value over time. The left vertical axis represents the coordinated heat dissipation redundancy margin value, the right vertical axis represents the heat flow rate and power values, and the horizontal axis represents time. As can be seen from Table 1 and Figure 4, the coordinated heat dissipation redundancy margin value fluctuates with the changes of the three thermal parameters, with a significant difference between the maximum and minimum values, indicating that the thermal safety margin changes dynamically in actual operation. When the heat flow rate of the cold plate and the heat flow rate of the phase change thermal storage material increase, the coordinated heat dissipation redundancy margin value increases significantly, indicating enhanced heat dissipation capacity and increased safety margin. Conversely, when the heating power of the battery cell is high but the heat dissipation capacity fails to increase simultaneously, the coordinated heat dissipation redundancy margin value decreases, indicating a higher thermal risk state. Overall, by adjusting the multi-channel heat dissipation pathways and thermal energy storage, the synergistic heat dissipation redundancy margin is maintained within a certain range, reflecting the effective control capability of dynamic heat dissipation capacity and thermal runaway safety margin.

[0060] In this implementation plan, by real-time acquisition and collaborative analysis of the heat dissipation capacity of the cold plate liquid cooling and phase change thermal storage materials and the heating power of the battery cells, the net heat dissipation margin and fluctuation intensity are dynamically quantified, effectively determining the safety margin of the current thermal management system. The construction of the collaborative heat dissipation redundancy margin value not only improves the comprehensive assessment capability of extreme heat loads and thermal runaway risks, but also provides a scientific and reliable basis for subsequent intelligent allocation, early warning intervention, and full-process thermal safety closed-loop.

[0061] Specifically, the process of providing early warning and intervention measures for liquid cooling thermal runaway based on the collaborative heat dissipation capacity margin is as follows: Real-time comparison of the collaborative heat dissipation redundancy margin value with the thermal runaway threshold. When the collaborative heat dissipation redundancy margin value is greater than the thermal runaway threshold, normal liquid cooling heat dissipation regulation and thermal energy storage allocation are maintained. At this time, it is within the normal control range, and the coolant flow rate and the release rate of the phase change thermal storage material are adaptively adjusted according to real-time load demand to ensure a balance between energy efficiency and safety. When the collaborative heat dissipation redundancy margin value is less than or equal to the thermal runaway threshold, the coolant flow rate is increased, and the standby pump group is activated. The heat release rate of the phase change thermal storage material is adjusted to release the stored thermal energy in advance. A multi-objective programming algorithm is used to optimize the heat dissipation capacity allocation of each channel to reduce the thermal burden. The multi-objective programming algorithm dynamically optimizes the collaborative working parameters of liquid cooling and phase change thermal storage channels based on the current heat load distribution and the heat dissipation capacity of each channel. When the duration of the collaborative heat dissipation redundancy margin value being less than or equal to the thermal runaway threshold exceeds the safety tolerance threshold, a thermal runaway warning signal is issued, triggering the battery management system to limit the charge and discharge rates and disconnect some abnormal circuits. The cold plate is forced to maintain the maximum allowable coolant flow rate and the coolant circulation rate is increased, while external air cooling is activated to maximize heat transfer capacity. Warning signals are pushed to the operation and maintenance platform in real time, and all collaborative heat dissipation redundancy margin values, cell temperatures, warning signals, and thermal runaway intervention measures are recorded, enabling cloud data archiving, remote traceability, and big data-driven operation and maintenance decisions. The correlation between collaborative heat dissipation redundancy margin values, cell temperatures, thermal runaway intervention measures, and actual cell temperature changes under different operating conditions is analyzed periodically, and a sliding window adaptive clustering algorithm is used to correct the thermal runaway threshold and thermal runaway intervention measures. The sliding window adaptive clustering algorithm dynamically adjusts the thermal runaway threshold and intervention logic by periodically clustering and attributing historical data, enabling the criteria to adapt to different application scenarios and workloads, and continuously improving the intelligence and safety protection level of thermal management.

[0062] Figure 5 shows the flowchart of multi-criteria intervention for adaptive flexible two-phase liquid cooling thermal runaway of battery bulging. First, the contact status data between the cold plate and the battery is collected in real time by multiple types of sensors; after signal preprocessing and standardization, the comprehensive bulging activity criterion value is calculated to realize intelligent identification and classification of bulging abnormalities.

[0063] If the overall bulge activity criterion value is below the activity threshold, the conventional liquid cooling mode is maintained. If the overall bulge activity criterion value exceeds the activity threshold, the flexible structure adaptive optimization and control is triggered, dynamically increasing the bonding pressure and adjusting the structural zoning to actively suppress bulge development. Subsequently, the equivalent thermal resistance of the interface is evaluated in real time to determine the thermal conductivity. If the thermal resistance is good, conventional liquid cooling and bonding repair monitoring continue. If the thermal resistance is too high, the thermal bottleneck area is located and regional thermal conductivity repair is performed, including increasing local pressure, pump speed, and phase change thermal storage synergy. After all thermal conductivity assessments and repair measures are completed, the calculation and dynamic control of the synergistic heat dissipation redundancy margin value are initiated. If the margin is sufficient, conventional monitoring is performed. If the margin is insufficient, the process is upgraded to thermal runaway early warning and multi-level active intervention, including increasing flow rate, limiting power, linking the cloud platform, and self-learning optimization, forming an intelligent closed loop of data, criteria, intervention, and learning.

[0064] In this implementation plan, by constructing a collaborative heat dissipation redundancy margin value and a thermal runaway threshold, real-time early warning and graded intervention of liquid cooling and phase change multi-channel heat dissipation capabilities are achieved. When the thermal safety margin is insufficient, multi-level linkage measures such as flow rate increase, heat storage release, multi-objective optimization, and current interruption protection can be activated to dynamically ensure the operational safety of Battery 3 under extreme thermal risks. The closed-loop archiving and self-learning correction of the entire process criteria, execution, and temperature data enhance intelligent adaptability and extreme thermal runaway protection capabilities.

[0065] Referring to Figure 2, the second aspect of the present invention provides a battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning system, applied to the aforementioned battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method, comprising: a multi-source state data acquisition and preprocessing module, used to acquire real-time cold plate and battery bonding state data, and perform data preprocessing operations on the cold plate and battery bonding state data; a battery bulging comprehensive discrimination module, used to determine whether the battery 3 is bulging based on the preprocessed cold plate and battery bonding state data, and to identify bulging abnormal events; a structural adaptation and thermal conduction path repair module, used to adaptively and flexibly adjust the battery bulging condition in response to bulging abnormal events; evaluate the liquid cooling thermal conductivity based on the cold plate and battery bonding state data after adaptive flexible adjustment, and optimize and adjust the local thermal conductivity performance based on the liquid cooling thermal conductivity efficiency; and a collaborative heat dissipation assessment and thermal runaway early warning module, used to evaluate the collaborative heat dissipation capacity margin using the cold plate and battery bonding state data after the local thermal conductivity performance optimization and adjustment is completed, and to perform liquid cooling thermal runaway early warning and implement thermal runaway intervention measures based on the collaborative heat dissipation capacity margin.

[0066] In this implementation plan, high-precision real-time acquisition and preprocessing of multi-source state data enables intelligent comprehensive identification of battery bulging, adaptive flexible structure control, and bottleneck repair of heat conduction pathways, effectively ensuring efficient bonding between the cold plate and the three sides of the battery and the liquid cooling heat conduction efficiency. Combined with collaborative heat dissipation capacity assessment and thermal runaway graded early warning, it can dynamically link multi-channel heat dissipation resources to achieve proactive safety intervention and closed-loop optimization of data throughout the entire process under extreme operating conditions, significantly improving the intelligence level, operational safety margin, and long-term reliability of battery thermal management.

[0067] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0068] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. As those skilled in the art will understand, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for early warning of battery bulging adaptive flexible two-phase liquid cooling thermal runaway, characterized in that, Includes the following steps: S1, collects real-time data on the bonding status between the cold plate and the battery, and performs data preprocessing on the data on the bonding status between the cold plate and the battery. S2, based on the pre-processed cold plate and battery bonding state data, determine whether the battery (3) is bulging and identify bulging abnormal events; S3, for bulging abnormal events, perform adaptive flexible control on the battery bulging condition; based on the cold plate and battery bonding state data after adaptive flexible control, evaluate the liquid cooling thermal conductivity, and optimize and adjust the local thermal conductivity performance according to the liquid cooling thermal conductivity; S4, after the local thermal conductivity performance optimization and adjustment is completed, use the cold plate and battery bonding state data to evaluate the margin of collaborative heat dissipation capacity, and perform liquid cooling thermal runaway early warning and implement thermal runaway intervention measures according to the margin of collaborative heat dissipation capacity.

2. The battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method according to claim 1, characterized in that, The specific process of real-time acquisition of cold plate and battery bonding state data and data preprocessing of cold plate and battery bonding state data is as follows: real-time acquisition of cold plate and battery bonding state data, specifically the method and data are as follows: pressure, abscissa deformation, ordinate deformation and bonding layer thickness from the inner surface of the flexible bonding layer of the cold plate to the outer surface of the battery cell are acquired by pressure sensors and micro piezoresistive strain sensor arrays distributed inside the flexible cold plate. The three bonding surfaces include the bottom surface, left side surface and right side surface; based on the sliding time window, the corresponding pressure change rate is obtained by dividing the numerical difference of adjacent moments of the continuously acquired pressure of the three bonding surfaces by the window length; simultaneously, the surface temperature of the battery cell and the inlet temperature of the cold plate are acquired by temperature sensor, flow meter and battery heat flow sensor. The system collects data on the temperature, cold plate outlet temperature, phase change thermal storage material temperature, coolant flow rate, and cell heating power. It also acquires the specific heat capacity of the coolant, the specific heat capacity of the phase change thermal storage material, the fixed melting temperature of the phase change thermal storage material, the fixed mass of the phase change thermal storage material, and the fixed area of ​​the bonding surface. Digital filtering of the cold plate-battery bonding status data is performed using mean and median filtering. Data collected from all different types of sensors is synchronized and aligned using a unified clock. The cold plate-battery bonding status data is standardized and normalized. An outlier detection algorithm is used to identify and mark abnormal values, signal loss, and sensor malfunctions. For detected short-term data gaps and abnormal discontinuities, a moving average completion algorithm is used to repair missing segments. A cold plate-battery operating condition database is established, and the cold plate-battery bonding status data is written into this database.

3. The battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method according to claim 1, characterized in that, The specific process for determining whether the battery (3) is bulging based on the pre-processed data on the bonding state between the cold plate and the battery is as follows: obtain the current pressure, corresponding pressure change rate, abscissa deformation, and ordinate deformation of the three bonding surfaces of the battery (3) and the cold plate; for each bonding surface, calculate the sum of the square of the abscissa deformation and the square of the ordinate deformation, and take the square root to obtain the spatial composite deformation; at the same time, multiply the pressure change rate by the time constant and add a constant to obtain the pressure dynamic factor; multiply the spatial composite deformation by the pressure dynamic factor to obtain the single-sided bulging activity value; accumulate the single-sided bulging activity values ​​of the three bonding surfaces and calculate the average value to obtain the comprehensive bulging activity value.

4. The battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method according to claim 1, characterized in that, The specific process for identifying bulging abnormal events is as follows: The overall bulging activity value is calculated in real time and compared with an activity threshold. When the overall bulging activity value is less than the activity threshold, it is determined to be a normal change, and only the overall bulging activity value is continuously monitored to maintain normal liquid cooling. When the overall bulging activity value is greater than or equal to the activity threshold, it is determined to be a warning state and marked as a bulging abnormal event. The cold plate and battery bonding status data corresponding to the bulging abnormal event and the overall bulging activity value are uploaded to the central controller, and the process proceeds to the next step. All comprehensive bulging activity values ​​and bulging abnormal events are written into the cold plate battery operating condition database, and the activity threshold is optimized by cluster analysis.

5. The battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method according to claim 1, characterized in that, The specific process of adaptive flexible control of battery bulging condition in response to bulging abnormal events is as follows: receiving bulging abnormal events, corresponding cold plate and battery bonding status data and comprehensive bulging activity value, driving flexible structure thermal optimization control: adjusting the current of magnetic coupling electromagnet to increase the bonding pressure of the three bonding surfaces of the cold plate; Drive the elastic hinge to divide the cold plate, while the structured flexible layer of the flexible cold plate responds with gradient modulus design, that is, the hard area at the edge maintains the stability of the frame, the soft area in the middle stretches with the bulge, and the wavy section (1) naturally extends along the deformation direction of the mating surface. The graphene fiber thermal conductive network embedded in the flexible cold plate is kept continuously connected, and the coolant phase change cycle in the cold plate runs at a normal rate. If the overall bulging activity value is detected to be rising for a period of time longer than the safety allowable threshold, thermal runaway is intercepted in real time. The overall bulging activity value is connected to the battery management system to trigger the charging and discharging power limit: during the charging stage, the current is forcibly reduced. During the discharge phase, the circuit containing the bulging cell is disconnected, and the load is transferred to the healthy cell.

6. The battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method according to claim 1, characterized in that, The specific process for evaluating the liquid cooling thermal conductivity efficiency based on the adaptive flexible adjustment of the cold plate and battery bonding state data is as follows: After the flexible structure thermal conductivity optimization adjustment, the cell surface temperature, cold plate inlet temperature, cold plate outlet temperature, coolant flow rate, and coolant specific heat capacity are obtained; the difference between the current cold plate outlet temperature and the current cold plate inlet temperature is calculated, and multiplied by the coolant flow rate and coolant specific heat capacity to obtain the cold plate heat dissipation flow rate. The heat dissipation intensity per unit temperature difference of the cold plate is obtained by dividing the heat flow rate of the cold plate by the difference between the current surface temperature of the battery cell and the current inlet temperature of the cold plate; at the same time, the pressure, spatial composite deformation, fixed area of ​​the bonding surface and the thickness of the bonding layer of the three bonding surfaces are obtained; when the pressure is greater than the contact threshold and the spatial composite deformation is greater than the deformation threshold, it is determined to be an effective bonding area. The three bonding surfaces are then judged to be effective bonding areas, and the fixed areas of the bonding surfaces of the effective bonding areas are added together to obtain the effective thermal conductive contact area. The equivalent thickness of the bonding layer is obtained by summing the thicknesses of the bonding layers on the three bonding surfaces and averaging them; the equivalent geometric factor of the heat conduction path is obtained by dividing the effective thermal contact area by the equivalent geometric factor of the heat conduction path; the equivalent thermal resistance is obtained by multiplying the heat dissipation intensity per unit temperature difference of the cold plate by the equivalent geometric factor of the heat conduction path.

7. The battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method according to claim 1, characterized in that, The specific process of optimizing and adjusting the local thermal conductivity performance based on the liquid cooling thermal conductivity efficiency is as follows: compare the equivalent thermal resistance assessment value with the thermal conductivity threshold in real time. When the equivalent thermal resistance assessment value is greater than or equal to the thermal conductivity threshold, it is determined that the bonding and thermal conductivity of the flexible structure have been well restored after the thermal optimization and adjustment, and the cold plate enters the conventional liquid cooling heat dissipation working mode. When the equivalent thermal resistance assessment value is less than the thermal conductivity threshold, the bonding status data of the cold plate and battery on the three bonding surfaces are obtained, the abnormal bonding area is located, and the local thermal conductivity performance is optimized and adjusted for the abnormal bonding area: the local bonding pressure of the three bonding surfaces is increased, the magnetic coupling distribution is adjusted, the pump speed is increased to adjust the coolant flow rate, and the coolant flow rate is increased, and the phase change heat storage material embedded in the bonding structure of the cold plate and battery (3) absorbs and stores heat; the equivalent thermal resistance assessment value, the abnormal bonding area and the corresponding thermal bottleneck repair implementation measures are uploaded to the cold plate battery operating condition database and enter the next process.

8. The battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning method according to claim 1, characterized in that, The specific process of evaluating the margin of collaborative heat dissipation capability using the cold plate and battery bonding state data after the local thermal conductivity optimization and adjustment is as follows: obtain the phase change thermal storage material temperature, specific heat capacity of the phase change thermal storage material, fixed melting temperature of the phase change thermal storage material, fixed mass of the phase change thermal storage material, heat dissipation flow of the cold plate, and heating power of the battery cell; calculate the difference between the fixed melting temperature of the phase change thermal storage material and the current temperature of the phase change thermal storage material, and multiply it by the specific heat capacity and fixed mass of the phase change thermal storage material to obtain the heat dissipation flow of the phase change thermal storage material; Based on the sliding time window, the heat flow of the cold plate and the heat flow of the phase change thermal storage material within the current window are statistically analyzed, and their standard deviations are calculated separately. The standard deviations of the heat flow of the cold plate and the heat flow of the phase change thermal storage material are added together to obtain the total heat flow fluctuation value. The heat flow of the phase change thermal storage material is added to the heat flow of the cold plate, and the cell heating power is subtracted to obtain the multi-channel net heat dissipation redundancy value. The multi-channel net heat dissipation redundancy value is divided by the sum of the total heat flow fluctuation value and the minimum constant value to obtain the collaborative heat dissipation redundancy margin value.

9. The method for early warning of battery bulging adaptive flexible two-phase liquid cooling thermal runaway according to claim 1, characterized in that, The specific process of liquid cooling thermal runaway early warning and thermal runaway intervention measures based on the collaborative heat dissipation capacity margin is as follows: compare the collaborative heat dissipation redundancy margin value with the thermal runaway threshold in real time. When the collaborative heat dissipation redundancy margin value is greater than the thermal runaway threshold, maintain the normal liquid cooling heat dissipation regulation and thermal energy storage distribution. When the collaborative heat dissipation redundancy margin is less than or equal to the thermal runaway threshold, increase the coolant flow rate and activate the standby pump group; adjust the heat release rate of the phase change thermal storage material to release the stored thermal energy in advance; and use a multi-objective programming algorithm to optimize the heat dissipation capacity allocation of each channel. When the duration of the collaborative heat dissipation redundancy margin value being less than or equal to the thermal runaway threshold exceeds the safety tolerance threshold, a thermal runaway warning signal is issued, triggering the battery management system to limit the charge and discharge rates and disconnect some abnormal circuits; the cold plate is forced to maintain the maximum allowable coolant flow rate and the coolant circulation rate is increased, triggering external air cooling emergency response; warning signals are pushed to the operation and maintenance platform in real time, and all collaborative heat dissipation redundancy margin values, cell temperatures, warning signals, and thermal runaway intervention measures are recorded. The correlation between collaborative heat dissipation redundancy margin values, cell temperatures, thermal runaway intervention measures, and actual cell temperature changes under different operating conditions is analyzed periodically, and the thermal runaway threshold and thermal runaway intervention measures are corrected using a sliding window adaptive clustering algorithm.

10. A battery bulging adaptive flexible two-phase liquid-cooled thermal runaway early warning system, characterized in that, include: The multi-source status data acquisition and preprocessing module is used to acquire the bonding status data of the cold plate and the battery in real time, and to perform data preprocessing operations on the bonding status data of the cold plate and the battery. The battery bulging comprehensive judgment module is used to determine whether the battery (3) is bulging based on the pre-processed cold plate and battery bonding status data, and to identify bulging abnormal events; The structural self-adaptation and thermal conduction path repair module is used to adaptively and flexibly adjust the battery swelling condition in response to abnormal swelling events. Based on the data of the bonding state between the cold plate and the battery after adaptive flexible adjustment, the liquid cooling thermal conductivity is evaluated, and the local thermal conductivity is optimized and adjusted according to the liquid cooling thermal conductivity. The collaborative heat dissipation assessment and thermal runaway early warning module is used to assess the margin of collaborative heat dissipation capability by using the bonding state data between the cold plate and the battery after the local thermal conductivity optimization and adjustment is completed. Based on the margin of collaborative heat dissipation capability, liquid cooling thermal runaway early warning is given and thermal runaway intervention measures are implemented.

Images