Multi-stage intelligent adaptive control method for battery thermal isolation and heat dissipation integration

Abstract

This invention discloses a multi-stage intelligent adaptive control method integrating battery thermal isolation and heat dissipation, relating to the field of battery thermal regulation technology. The method includes: S1, periodically collecting battery heat dissipation data and preprocessing the data; S2, evaluating the battery's thermal management conditions, classifying the battery thermal management state, and activating a thermal state confirmation mechanism when a thermal risk state is continuously detected, and issuing a heat regulation execution signal after confirmation; S3, evaluating the battery's thermal risk level, implementing thermal regulation strategies in stages, and constructing a strategy response dataset; S4, evaluating the implementation deviation of the thermal regulation strategy, and determining whether to trigger a strategy adjustment mechanism based on the evaluation results. This solves the problem in existing technologies where heat dissipation components and thermal insulation structures are typically designed independently, leading to the rapid spread of high temperatures and flames through cell gaps when thermal runaway occurs.

Description

Technical Field

[0001] This invention relates to the field of battery thermal regulation technology, specifically a multi-stage intelligent adaptive control method integrating battery thermal isolation and heat dissipation. Background Technology

[0002] With the rapid development of new energy vehicles and energy storage systems, the power density and integration complexity of power battery systems are increasing, placing higher demands on battery thermal management performance. Localized hot spots in a single cell can easily lead to uneven temperature distribution. In existing technologies, heat dissipation components and thermal insulation structures are usually designed independently. When a cell experiences thermal runaway, high temperatures and flames can easily spread rapidly through the cell gaps, triggering a chain reaction and seriously threatening system safety. Therefore, how to construct efficient, adjustable, and multi-condition adaptable heat dissipation structures and systems has become an important research direction in current battery pack design.

[0003] For example, the invention with announcement number CN106299535B provides a battery pack heat dissipation system, including an outer frame. Several battery modules are arranged inside the outer frame. The battery modules can be pushed and pulled inside the outer frame. Two sets of fixing posts, A and B, are arranged on the ends of the battery modules that contact the back of the outer frame. The fixing posts are hollow structures and communicate with the inside of the battery modules. An air inlet pipe and an air outlet pipe are arranged on the back of the outer frame. The air inlet pipe has a slot for accommodating the fixing posts of set A, and the air outlet pipe has a slot for accommodating the fixing posts of set B. The air inlet pipe has one and only one open end, and an air inlet column is arranged at the open end of the air inlet pipe. The air outlet pipe has one and only one open end, and an air outlet column is arranged at the open end of the air outlet pipe. A locking device is provided between the battery modules and the outer frame.

[0004] For example, the invention disclosed in CN114725561B discloses a lithium battery heat dissipation structure, including a heat dissipation shell. The inner side of the heat dissipation shell has symmetrically distributed grooves, and a bridging top plate is movably connected to these grooves. The bridging top plate has uniformly distributed bridging clips fixedly connected inside. This lithium battery heat dissipation structure, by creating uniformly distributed cylindrical grooves on the inner side of the heat dissipation shell and movably connecting symmetrically distributed inner sleeves within these grooves, and by creating relatively distributed rectangular heat dissipation vents on the outer side of the inner sleeves, allows the lithium battery to be movably connected inside the inner sleeves and evenly separated. This ensures that the device can effectively separate the lithium battery by a distance, thus solving the problem in existing devices where the lithium battery is encased in insulating material and difficult to dissipate heat.

[0005] However, these methods generally lack a systematic understanding and dynamic control capability of battery thermal behavior, and have not yet constructed an integrated intelligent control mechanism that encompasses thermal risk perception, state identification, control strategy execution, effect evaluation, and adaptive adjustment. Especially when facing complex operating conditions such as long-term battery operation, load fluctuations, and changes in the external environment, existing systems exhibit lag in accurately identifying dynamic thermal risk states, and their control response mechanisms are too rigid. This makes it difficult to effectively implement differentiated, graded, and evolvable cooling control strategies, easily leading to energy waste, insufficient heat dissipation, or potential thermal runaway.

[0006] Therefore, in order to address the above problems, there is an urgent need for a multi-stage intelligent adaptive control method that integrates battery thermal isolation and heat dissipation. Summary of the Invention

[0007] Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides a multi-stage intelligent adaptive control method that integrates battery thermal isolation and heat dissipation. This method solves the problem that in existing technologies, heat dissipation components and thermal insulation structures are usually designed independently, which leads to the rapid spread of high temperatures and flames through the gaps between battery cells when thermal runaway occurs.

[0009] Technical solution

[0010] To achieve the above objectives, the present invention provides the following technical solution: a multi-stage intelligent adaptive control method integrating battery thermal isolation and heat dissipation, comprising: S1, periodically collecting battery heat dissipation data and performing time alignment, anomaly removal, standardization, and normalization processing on the battery heat dissipation data to obtain pre-processed battery heat dissipation data; S2, based on the pre-processed battery heat dissipation data, evaluating the battery's thermal management condition, classifying the battery thermal management state, and when a thermal risk state is continuously detected, initiating a thermal state confirmation mechanism and issuing a heat dissipation control execution signal after confirmation; S3, receiving the heat dissipation control execution signal, evaluating the battery's thermal risk level, and executing passive heat storage mode, active liquid cooling mode, and emergency protection mode in stages, and constructing a strategy response dataset after the strategy execution is completed; S4, based on the strategy response dataset, evaluating the implementation deviation of the thermal control strategy, and determining whether to trigger the strategy adjustment mechanism based on the evaluation results, forming a strategy adaptive update closed loop.

[0011] Furthermore, the specific steps for periodically collecting battery heat dissipation data are as follows: Set the sampling period of a fixed-length sliding window, collect battery heat dissipation data, including cell temperature, phase change material temperature, coolant inlet temperature, coolant outlet temperature, battery pack casing temperature, cell voltage, and cell current.

[0012] Further, the specific steps for time alignment, anomaly elimination, standardization, and normalization of battery heat dissipation data to obtain the preprocessed battery heat dissipation data are as follows: Automatically calibrate the battery heat dissipation data through a multi-source calibration and drift correction strategy to eliminate the baseline error caused by hardware differences and environmental changes between each acquisition channel; Denoise the battery heat dissipation data through a noise suppression method based on a time-domain sliding window and an adaptive filtering algorithm to filter out transient interference and high-frequency noise; Detect mutations in the battery heat dissipation data through an anomaly screening mechanism that combines physical evolution laws and dynamic interval thresholds, and eliminate data anomaly points that do not conform to the actual working conditions; Convert the format of the battery heat dissipation data through a numerical standardization method based on a unified standard physical unit to achieve consistent expression of data between different channels; Normalize the battery heat dissipation data through a numerical compression technology that combines typical operating intervals and adaptive normalization mapping.

[0013] Further, based on the preprocessed battery heat dissipation data, the specific steps for evaluating the thermal management working conditions of the battery are as follows: Set a fixed sliding time window as the evaluation period for the thermal management state. In each evaluation period, extract the cell temperature sequence, calculate the difference between the maximum cell temperature and the minimum cell temperature to obtain the maximum cell temperature difference, and calculate the average cell temperature; Divide the square of the maximum cell temperature difference by the sum of the average cell temperature and the minimum term to obtain the thermal imbalance factor; Subtract the coolant inlet temperature from the average cell temperature, divide by the difference between the coolant outlet temperature and the coolant inlet temperature plus the minimum term, add one to the obtained ratio, and take the natural logarithm to obtain the heat transfer efficiency factor; Subtract the lower limit of the phase change material temperature from the phase change material temperature, divide by the difference between the upper limit of the phase change material temperature and the lowest working temperature plus the minimum term, subtract the obtained ratio from one, and take the square to obtain the phase change heat storage capacity factor; Multiply the thermal imbalance factor, the heat transfer efficiency factor, and the phase change heat storage capacity factor to obtain the thermal management working condition evaluation value.

[0014] Further, the specific steps for dividing the battery thermal management state are as follows: Compare the thermal management working condition evaluation value H with the multi-level working condition thresholds H1 and H2 in real time to divide the battery thermal management state: When H ≤ H1, it is in a thermal stable state and no processing is required; When H1 < H < H2, it is in a thermal rising state and the sampling frequency is increased; When H ≥ H2, it is in a thermal risk state and a warning signal is sent to the control unit.

[0015] Furthermore, when the thermal risk state is continuously detected, the thermal state confirmation mechanism is activated, and the specific steps for sending the thermal regulation execution signal after confirmation are as follows: When the control unit continuously receives a fixed number of warning signals, the thermal state confirmation mechanism is activated: Extract the cell temperature sequence within the evaluation period, calculate the cell temperature change rate. When the cell temperature change rate is always positive and greater than the temperature rise threshold, it is determined that there is a continuous thermal risk trend, and the thermal regulation execution signal is sent; otherwise, clear the warning cumulative count and maintain the current thermal management state unchanged.

[0016] Furthermore, the specific steps for receiving the thermal regulation execution signal and evaluating the thermal risk level of the battery are as follows: The execution unit receives the thermal regulation execution signal and extracts the real-time battery heat dissipation data, thermal management working condition evaluation value, and the average cell temperature of the current evaluation period; Divide the square of the cell current by the sum of the square of the cell rated current and a minimum term, and add one to the obtained result to get the current load factor; Subtract the coolant outlet temperature from the battery pack housing temperature, divide by the sum of the average cell temperature and a minimum term, and add one to the obtained result to get the heat dissipation blockage factor; Multiply the thermal management working condition evaluation value, current load factor, and heat dissipation blockage factor in sequence to get the thermal risk level evaluation value.

[0017] Furthermore, the specific steps for hierarchically executing the passive heat storage mode, active liquid cooling mode, and emergency protection mode, and constructing a policy response data set after the strategy execution are as follows: Compare the thermal risk level evaluation value with the multi-level control strategy thresholds C1 and C2, and hierarchically execute the thermal regulation strategy: When C ≤ C1, it is determined that there is a first-level risk, and the passive heat storage mode is triggered: Keep the liquid cooling system closed and rely on the phase change material to absorb heat; When C1 < C < C2, it is determined that there is a second-level risk, and the active liquid cooling mode is triggered: Send a start-stop control instruction to the thermal management execution unit, start the liquid cooling pump and the condenser fan, and set the target speeds of the liquid cooling pump and the condenser fan respectively according to the linear mapping result between the thermal risk level evaluation value and the control strategy threshold interval; When C ≥ C2, it is determined that there is a third-level risk, and the emergency protection mode is triggered: Adjust both the liquid cooling pump and the condenser fan to the highest speed, activate the emergency heat dissipation ventilation circuit, cut off the charge and discharge circuits of the corresponding cells, and send an alarm signal to the main control system; After the thermal regulation strategy is executed, structurally store the battery heat dissipation data, thermal management working condition evaluation value, thermal risk level evaluation value, target speed of the liquid cooling pump, and target speed of the condenser fan before and after the strategy execution to construct a policy response data set.

[0018] Furthermore, based on the strategy response dataset, the specific steps for evaluating the implementation deviation of the thermal control strategy are as follows: After each thermal control strategy is executed, the corresponding strategy response dataset is extracted, and the real-time speed of the condenser fan and the real-time speed of the liquid cooling pump are collected; the real-time speed of the condenser fan is divided by the sum of the target speed of the condenser fan and the minimum term to obtain the fan execution ratio; the real-time speed of the liquid cooling pump is divided by the sum of the target speed of the liquid cooling pump and the minimum term to obtain the liquid cooling pump execution ratio; the fan execution ratio and the liquid cooling pump execution ratio are added together and multiplied by the thermal risk level assessment value to obtain the cooling execution item; the square of the cell current is divided by the square of the cell rated current and the sum of the minimum term to obtain the current load item; the cell temperature change rate is subtracted from the cooling execution item and then added to the current load item, and the absolute value is taken to obtain the thermal response deviation assessment value.

[0019] Furthermore, based on the evaluation results, the specific steps for determining whether to trigger the strategy adjustment mechanism and forming a closed loop of adaptive strategy update are as follows: Real-time comparison of the thermal response deviation evaluation value and the response deviation threshold. When the thermal response deviation evaluation value is less than the response deviation threshold, the current thermal control strategy remains unchanged. When the thermal response deviation evaluation value is greater than or equal to the response deviation threshold, the strategy adjustment mechanism is triggered: historical thermal response deviation evaluation values ​​and corresponding real-time speed data of the liquid cooling pump and condenser fan are extracted. The target speed mapping parameters are recalculated using a sliding window linear regression algorithm, and the corrected target speed mapping parameters are stored in the system configuration cache for loading and use in the next evaluation cycle.

[0020] Beneficial effects

[0021] The present invention has the following beneficial effects:

[0022] (1) The multi-stage intelligent adaptive control method integrating battery thermal isolation and heat dissipation accurately characterizes the current thermal state of the battery by constructing a thermal management condition assessment model. Combined with the thermal risk level calculation model, the thermal management condition assessment value, current load factor, and heat dissipation resistance factor are integrated to assess the thermal risk level, thereby realizing a quantitative and dynamic judgment of the thermal state. Compared with the existing control methods that use single-point temperature as the trigger condition, it can significantly improve the foresight and accuracy of the strategy response.

[0023] (2) The multi-stage intelligent adaptive control method integrating battery thermal isolation and heat dissipation divides the thermal regulation mode into three modes: passive heat storage, active liquid cooling, and emergency protection by setting multiple thresholds based on the thermal risk level. The cooling strategy is automatically switched according to different thermal risk levels. Under low thermal risk, the liquid cooling circuit is shut down and heat is absorbed by phase change materials. Under medium thermal risk, the speed of liquid cooling pump and fan is precisely adjusted. Under high thermal risk, emergency ventilation and power failure protection are activated, effectively balancing cooling energy consumption and thermal safety requirements.

[0024] (3) The multi-stage intelligent adaptive control method of the battery, which integrates thermal isolation and heat dissipation, combines active liquid cooling and phase change heat absorption technology to take into account both temperature control and thermal insulation protection during the thermal regulation process. When the temperature of a single cell rises abnormally, the heat source path can be quickly cut off, the heat propagation can be limited, the risk of high temperature being conducted to adjacent cells can be reduced, the chain reaction of thermal runaway can be avoided, and the overall thermal stability can be guaranteed.

[0025] (4) The multi-stage intelligent adaptive control method integrating battery thermal isolation and heat dissipation evaluates the implementation effect of the control strategy based on the thermal response deviation value after each round of thermal control execution, and automatically extracts historical response samples when the deviation exceeds the limit. The target speed mapping parameters are reconstructed through the sliding window linear regression algorithm to realize the adaptive update of the control strategy and continuously optimize the cooling control matching degree and actual operation effect. Attached Figure Description

[0026] Figure 1 Flowchart of a multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation;

[0027] Figure 2 This is a diagram for determining the thermal risk level based on the thermal risk level assessment value;

[0028] Figure 3 This is a schematic diagram of an integrated thermal isolation and heat dissipation structure for batteries.

[0029] In the diagram, 1 is the battery pack casing; 2 is the battery; 3 is the phase change material; 4 is the liquid cooling channel; 5 is the separator; 6 is the condenser; and 7 is the liquid cooling pump. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, 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.

[0031] Please see Figures 1-3This invention provides a technical solution: a multi-stage intelligent adaptive control method integrating battery thermal isolation and heat dissipation, comprising: S1, periodically collecting battery heat dissipation data, and performing time alignment, anomaly removal, standardization, and normalization processing on the battery heat dissipation data to obtain pre-processed battery heat dissipation data; S2, based on the pre-processed battery heat dissipation data, evaluating the battery's thermal management condition, classifying the battery thermal management state, and when a thermal risk state is continuously detected, initiating a thermal state confirmation mechanism, and issuing a heat dissipation control execution signal after confirmation; S3, receiving the heat dissipation control execution signal, evaluating the thermal risk level of battery 2, and executing passive heat storage mode, active liquid cooling mode, and emergency protection mode in stages, and constructing a strategy response dataset after the strategy execution is completed; S4, based on the strategy response dataset, evaluating the implementation deviation of the heat dissipation strategy, and determining whether to trigger the strategy adjustment mechanism based on the evaluation result, forming a strategy adaptive update closed loop.

[0032] Specifically, the steps for periodically collecting battery heat dissipation data are as follows: Based on a sampling mechanism with a fixed sliding time window, battery heat dissipation data is periodically collected. This data includes cell temperature, phase change material temperature, coolant inlet temperature, coolant outlet temperature, battery pack casing temperature, cell voltage, and cell current. Specifically, cell temperature is detected point-by-point by temperature sensors installed on the cell surface; phase change material temperature is periodically collected in real-time by thermocouples embedded within the phase change material 3; coolant inlet and outlet temperatures are synchronously measured and stored by temperature sensors installed at the inlet and outlet of the coolant flow path, respectively; battery pack casing temperature is collected by temperature sensors attached to the outer wall of the battery pack casing 1; cell voltage is periodically obtained by a high-precision digital voltmeter installed on the cell terminals; and cell current is detected in real-time by Hall current sensors installed in the cell circuit.

[0033] In this implementation scheme, battery heat dissipation data is periodically collected using a sampling mechanism based on a fixed sliding time window. This ensures that the data collection nodes are clearly defined, the detection methods are clear, and the data sources are reliable. This approach enables the simultaneous acquisition of multiple physical quantities, ensuring that cell temperature, cell voltage, and cell current remain consistent with the phase change material temperature, coolant temperature, and battery pack casing temperature over time. This guarantees high integrity and accuracy of the battery heat dissipation data, providing a solid data foundation for subsequent thermal management condition assessment, thermal risk level determination, and thermal control strategy implementation.

[0034] Specifically, the preprocessed battery heat dissipation data undergoes time alignment, anomaly removal, standardization, and normalization. The steps are as follows: Automatic calibration of the battery heat dissipation data is performed using multi-source calibration and drift correction strategies. The acquisition channels for cell voltage and cell current are compared using a reference voltage source and a constant current source. Simultaneously, the acquisition channels for cell temperature, phase change material temperature, coolant inlet temperature, coolant outlet temperature, and battery pack casing temperature are compared using a standard temperature source under constant temperature conditions, thereby eliminating baseline errors caused by hardware differences and environmental variations in each acquisition channel. Noise suppression of the battery heat dissipation data is performed using a noise suppression method based on a time-domain sliding window and adaptive filtering algorithm. Transient spike interference is filtered out from the cell voltage and cell current signals, and high-frequency noise is suppressed from the cell temperature, phase change material temperature, coolant temperature, and battery pack casing temperature signals. Finally, by combining physical evolution laws and dynamic range thresholds… The anomaly screening mechanism detects sudden changes in battery heat dissipation data. When cell voltage or cell current exhibits jumps that do not conform to electrochemical change patterns within a single cycle, or when cell temperature, phase change material temperature, coolant temperature, or battery pack casing temperature shows sudden increases or decreases that do not conform to thermal diffusion patterns within a single cycle, it is judged as an anomaly and the corresponding data points are removed. The battery heat dissipation data is formatted using a numerical standardization method based on unified standard physical units, so that cell voltage, cell current, phase change material temperature, coolant temperature, and battery pack casing temperature are consistently expressed under the same physical dimension. The battery heat dissipation data is normalized by combining typical operating ranges with adaptive normalization mapping, compressing cell voltage, cell current, phase change material temperature, coolant temperature, and battery pack casing temperature in different amplitude ranges to a unified numerical range, thereby ensuring that the pre-processed battery heat dissipation data has good comparability and stability.

[0035] In this implementation scheme, by performing time alignment, anomaly removal, standardization, and normalization on battery heat dissipation data, the acquisition accuracy and data consistency of cell temperature, phase change material temperature, coolant inlet temperature, coolant outlet temperature, battery pack casing temperature, cell voltage, and cell current can be significantly improved. Multi-source calibration and drift correction ensure that baseline errors of each acquisition channel are eliminated. Time-domain sliding windows and adaptive filtering algorithms ensure that cell voltage, cell current, and various temperature signals are protected from transient interference and high-frequency noise. An anomaly screening mechanism combining dynamic interval thresholds and physical evolution laws effectively removes data points that do not conform to operating conditions. The numerical standardization method using unified standard physical units achieves format consistency among different variables. Adaptive normalization mapping technology further ensures the comparability of data within a unified numerical range. This processing method not only improves the reliability and stability of preprocessed battery heat dissipation data but also lays a solid data foundation for thermal management condition assessment, thermal risk level determination, and accurate execution of subsequent thermal control strategies.

[0036] Specifically, based on the preprocessed battery heat dissipation data, the specific steps for evaluating the thermal management status of battery 2 are as follows: A fixed sliding time window is set as the evaluation period for the thermal management status. The length of the sliding time window used within the evaluation period is consistent with the sampling frequency to ensure the temporal integrity of various thermal management variables. Within each evaluation period, the cell temperature sequence is extracted, the difference between the maximum and minimum cell temperatures is calculated to obtain the maximum cell temperature difference, and the average cell temperature is calculated based on the same sequence. The square of the maximum cell temperature difference is divided by the sum of the average cell temperature and the minimum term to obtain the thermal imbalance factor, which reflects the degree of influence of the temperature difference between cells on the stability of the heat distribution. The minimum term is a very small positive constant used to avoid calculation anomalies where the denominator is zero in division operations, ensuring the stability and robustness of the calculation model. The heat transfer efficiency factor is obtained by subtracting the coolant inlet temperature collected by the coolant inlet temperature sensor from the average cell temperature, dividing by the difference between the coolant outlet temperature and the coolant inlet temperature plus a minimum term, adding one to the resulting ratio, and taking the natural logarithm. This ratio measures the actual heat transfer capacity of the coolant between the inlet and outlet. The phase change material temperature collected by the thermocouple is subtracted from the lower limit of the phase change material temperature, divided by the difference between the upper limit of the phase change material temperature and the minimum operating temperature plus a minimum term, subtracting one from the resulting ratio, and taking the square. This ratio reflects the remaining heat absorption capacity of the current phase change material. The thermal imbalance factor, the heat transfer efficiency factor, and the phase change heat storage capacity factor are multiplied together to obtain the thermal management condition assessment value, which serves as a quantitative indicator of the current thermal management status of battery 2.

[0037] The specific formula for calculating the thermal management condition assessment value is as follows:

[0038] ;

[0039] In the formula, This indicates the thermal management condition assessment value. Indicates the maximum cell temperature difference. This represents the average cell temperature. Indicates the coolant inlet temperature. Indicates the coolant outlet temperature. Indicates the temperature of the phase change material. Indicates the lower limit of the temperature of the phase change material. This indicates the upper limit of the temperature range for phase change materials. Indicates a minus term.

[0040] In this implementation scheme, a thermal management condition assessment method based on multi-source data such as cell temperature, coolant temperature, and phase change material temperature can quantitatively assess the thermal management status of battery 2 from three key dimensions: thermal imbalance, heat transfer efficiency, and phase change material heat storage capacity. This method introduces three structured indicators—thermal imbalance factor, heat transfer efficiency factor, and phase change material heat storage capacity factor—into the calculation, effectively avoiding the limitations of judging thermal status based on a single point temperature or a single heat index. Simultaneously, the inclusion of minima avoids numerical instability issues in division operations, significantly improving the stability and adaptability of the thermal management condition assessment model, and providing a reliable basis for subsequent thermal risk level determination and precise triggering of graded control strategies.

[0041] Specifically, the steps for classifying the battery thermal management status are as follows: compare the thermal management condition evaluation value H with the multi-level condition thresholds H1 and H2 in real time, and classify the battery thermal management status based on the comparison results. Specifically: When the thermal management condition assessment value H is less than or equal to the first-level condition threshold H1, battery 2 is determined to be in a thermally stable state. At this time, the cell temperature fluctuation is small, the coolant temperature difference is stable, and the phase change material temperature is within a reasonable heat storage range. No thermal control operation is required, and the current operating state is maintained. When the thermal management condition assessment value H is greater than the first-level condition threshold H1 and less than the second-level condition threshold H2, battery 2 is determined to be in a thermal rise state. At this time, the cell temperature fluctuation is enhanced, the coolant temperature difference is widened, and the phase change material temperature gradually approaches the upper limit of heat storage. In order to detect potential thermal risk trends in advance, the sampling frequency is increased to obtain data with higher timeliness and accuracy to support subsequent judgments. When the thermal management condition assessment value H is greater than or equal to the second-level condition threshold H2, battery 2 is determined to be in a thermal risk state, indicating that the degree of cell thermal imbalance has increased significantly, the heat exchange efficiency has decreased, and the heat storage capacity of phase change material 3 is close to the limit. An early warning signal is immediately sent to the control unit to provide a decision basis for the preparation of thermal control execution. The control unit serves as the computation and instruction center for the battery thermal management logic. It has functions such as data reception, threshold judgment, status recognition, and control signal output. It can analyze multi-dimensional heat dissipation data and evaluation results in real time, and issue early warning signals and thermal regulation execution signals according to the control strategy requirements, driving the execution unit to implement the corresponding thermal regulation strategy.

[0042] In this implementation plan, dynamic classification of battery thermal management status is achieved through real-time comparison of thermal management condition assessment values ​​with multi-level condition thresholds. By dividing the battery into thermally stable, thermally rising, and thermally risky states, thermal imbalance trends can be identified in a timely manner at the early stages of cell temperature changes, coolant temperature difference widening, and phase change material temperature rise. Furthermore, by increasing data sampling frequency and issuing early warning signals, the sensitivity and timeliness of thermal status perception are enhanced, providing a reliable basis for the accurate triggering of subsequent thermal regulation execution signals and significantly improving the initiative and accuracy of thermal management strategies.

[0043] Specifically, when a thermal risk state is continuously detected, the thermal state confirmation mechanism is activated, and the following steps are taken to issue a heating control execution signal after confirmation: When the control unit receives a fixed number of warning signals continuously, and each warning signal is triggered by the thermal management condition assessment value being greater than the thermal risk threshold H2, the thermal state confirmation mechanism is activated; the cell temperature sequence with the sliding window as the boundary is extracted within the current assessment period, and the cell temperature change rate at continuous time points is calculated using the front-to-back difference algorithm, and point-by-point judgment is performed throughout the entire period; when the cell temperature change rate corresponding to all sampling points is positive, and each change rate value is greater than the temperature rise threshold, it is determined that the cell temperature is in a continuous upward trend, with significant thermal risk evolution characteristics, and the continuous thermal risk trend is confirmed, and a heating control execution signal is immediately issued; otherwise, the cumulative warning count is cleared, the current assessment period state is reset to the previous interval state, and the current thermal management state remains unchanged to prevent the control mechanism from being mistakenly triggered by occasional disturbances.

[0044] In this implementation, a thermal status confirmation mechanism is activated after receiving multiple consecutive warning signals. The rate of temperature change is calculated using the cell temperature sequence within a sliding window. If the rate of change remains positive and exceeds the temperature rise threshold, a sustained thermal risk trend is confirmed, avoiding misjudgments caused by occasional temperature fluctuations. This method improves the robustness of cell thermal risk identification and the accuracy of control triggering, ensuring that the generation of thermal control execution signals is based on reliable dynamic trends, thereby enhancing the stability and safety of battery thermal management status determination.

[0045] Specifically, the steps for receiving the thermal regulation execution signal and assessing the thermal risk level of battery 2 are as follows: The execution unit receives the thermal regulation execution signal and extracts the real-time battery heat dissipation data, the calculated thermal management condition assessment value, and the average cell temperature within the current assessment period. Using the collected cell current and rated current, the current load factor is calculated, which is the square of the cell current divided by the sum of the square of the cell rated current and the minimum term, and the result is incremented by one to characterize the impact of the cell's current load on the thermal risk. At the same time, the battery pack casing temperature is subtracted from the coolant outlet temperature, and the difference is divided by the sum of the average cell temperature and the minimum term, and the ratio is incremented by one to obtain the heat dissipation hindrance factor, which is used to characterize the heat exchange efficiency between the casing and the heat dissipation path. Finally, the thermal management condition assessment value, the current load factor, and the heat dissipation hindrance factor are multiplied sequentially to calculate the thermal risk level assessment value, which serves as the core criterion for guiding the subsequent graded response of the thermal regulation strategy.

[0046] The specific formula for calculating the thermal risk level assessment value is as follows:

[0047] ;

[0048] In the formula, This indicates the thermal risk level assessment value. This indicates the thermal management condition assessment value. Indicates the cell current. Indicates the rated current of the battery cell. Indicates the battery pack casing temperature. Indicates the coolant outlet temperature. This represents the average cell temperature. Indicates a minus term.

[0049] In this embodiment, Table 1 is a data table of thermal risk level assessment values, listing the key operating parameters and corresponding thermal risk level assessment values ​​for five assessment cycles. Key operating parameters include thermal management condition assessment value, cell current, cell rated current, battery pack casing temperature, coolant outlet temperature, and average cell temperature. Specific data is explained as follows: In the first assessment cycle, the thermal management condition assessment value is 0.95, the cell current is 45, the cell rated current is 60, the battery pack casing temperature is 42, the coolant outlet temperature is 35, the average cell temperature is 38, and the corresponding thermal risk level assessment value is 1.76; In the second assessment cycle, the thermal management condition assessment value is 1.10, the cell current is 48, the cell rated current is 60, the battery pack casing temperature is 45, the coolant outlet temperature is 36, the average cell temperature is 40, and the corresponding thermal risk level assessment value is 2.21; In the third assessment cycle, the thermal management condition assessment value is 1.05, the cell current is 50, and the cell rated current is 60. In the first assessment cycle, the battery pack casing temperature was 47°C, the coolant outlet temperature was 36°C, and the average cell temperature was 42°C, resulting in a thermal risk level assessment value of 2.25. In the second assessment cycle, the thermal management condition assessment value was 1.20, the cell current was 55 kW, the cell rated current was 60 kW, the battery pack casing temperature was 49°C, the coolant outlet temperature was 37°C, and the average cell temperature was 44°C, resulting in a thermal risk level assessment value of 2.81. In the third assessment cycle, the thermal management condition assessment value was 1.30, the cell current was 60 kW, the cell rated current was 60 kW, the battery pack casing temperature was 52°C, the coolant outlet temperature was 38°C, and the average cell temperature was 46°C, resulting in a thermal risk level assessment value of 3.39.

[0050] Table 1. Data Table of Thermal Risk Level Assessment Values

[0051]

[0052] like Figure 2The figure shows the trend of the thermal risk level assessment value of Battery 2 and the corresponding thermal risk level over five consecutive thermal management assessment cycles. In the figure, the blue dashed line represents the first-level risk threshold; the red dashed line represents the second-level risk threshold; the green broken line dots represent the first-level risk; the orange broken line dots represent the second-level risk; and the red broken line dots represent the third-level risk. It can be seen from the figure that assessment cycle 1 is at the first-level risk, assessment cycles 2, 3, and 4 are at the second-level risk, and assessment cycle 5 is at the third-level risk. Figure 2 It intuitively reflects the correspondence between the quantitative assessment results of battery thermal state and the graded control strategy, and can serve as an important criterion for driving the switching of thermal regulation strategies.

[0053] In this implementation scheme, after receiving the thermal regulation execution signal, a quantitative calculation model for the thermal risk level assessment value is constructed by integrating the cell current, cell rated current, average cell temperature, battery pack casing temperature, coolant outlet temperature, and thermal management condition assessment value. This not only enhances the accuracy of thermal risk level judgment but also achieves a comprehensive consideration of the degree of heat dissipation obstruction and the impact of current load under different operating conditions, effectively improving the accuracy of battery thermal state assessment and the rationality of thermal regulation strategy triggering.

[0054] Specifically, the passive thermal storage mode, active liquid cooling mode, and emergency protection mode are executed in stages. The specific steps for constructing the strategy response dataset after strategy execution are as follows: The thermal risk level assessment value is compared in real time with the multi-level control strategy thresholds C1 and C2. The thermal regulation strategy is executed in stages to address the thermal management needs corresponding to different thermal risk levels. When the thermal risk level assessment value C is less than or equal to the control strategy threshold C1, it is determined that the current risk level is Level 1, triggering the passive thermal storage mode: by keeping the liquid cooling pump 7 and the condenser 6 fan off, the battery 2 relies entirely on the phase change material 3 covering the outside of the cell to absorb the heat released by the cell, thereby delaying the release of thermal energy and maintaining thermal balance. When the thermal risk level assessment value C is greater than the control strategy threshold C1 but less than the control strategy threshold C2, the current risk level is determined to be level two, triggering the active liquid cooling mode: A start / stop control command is sent to the execution unit to start the liquid cooling pump 7 and the condenser 6 fan. Based on the linear mapping relationship between the thermal risk level assessment value and the control strategy threshold range, the target speeds of the liquid cooling pump 7 and the condenser 6 fan are set respectively, achieving fine-tuning of cooling flow rate and heat dissipation intensity. The execution unit is an intelligent control device deployed in the thermal management channel, including a programmable logic controller, a liquid cooling pump driver, and a condenser fan control module, possessing functions such as command reception, signal parsing, and equipment start / stop linkage. When the thermal risk level assessment value C is greater than or equal to the control strategy threshold C2, the current risk level is determined to be level three, triggering the emergency protection mode: The liquid cooling pump 7 and the condenser 6 fan are simultaneously adjusted to their highest speeds to build extreme heat dissipation capacity. Simultaneously, a preset emergency heat dissipation ventilation circuit is activated, and the control unit forcibly cuts off the charging and discharging circuit of the corresponding battery cell, sending an alarm signal to the upper-level main control system to ensure that heat diffusion under fault conditions is promptly suppressed. After the thermal control strategy is executed, key parameters such as battery heat dissipation data, thermal management condition assessment value, thermal risk level assessment value, target speed of liquid cooling pump and target speed of condenser fan before and after the strategy execution are extracted, stored in a structured manner, and a strategy response dataset for control closed-loop feedback is constructed to provide basic data support for subsequent deviation assessment and adaptive adjustment of the strategy.

[0055] like Figure 3 As shown, an integrated battery thermal isolation and heat dissipation structure is illustrated, comprising multiple batteries 2 arranged inside a battery pack housing 1. A phase change material 3 and a liquid cooling channel 4 are sequentially arranged on the outer side of each battery 2. The phase change material 3 absorbs latent heat during battery temperature rise, slowing the temperature increase rate and achieving passive heat storage regulation. The liquid cooling channel 4, together with a liquid cooling pump 7 and a condenser 6, forms a circulating cooling path, activating an active heat dissipation mechanism to rapidly reduce battery temperature upon detection of a thermal risk state. A partition 5 is installed between the multiple batteries 2 to block lateral heat conduction paths, improve thermal isolation capabilities, reduce the risk of thermal runaway propagation, and effectively ensure the thermal safety performance of the batteries 2 under complex operating conditions.

[0056] In this implementation scheme, by comparing the thermal risk level assessment value with the control strategy thresholds C1 and C2 in real time, passive thermal storage mode, active liquid cooling mode, and emergency protection mode are executed in stages to ensure that the thermal control strategy can accurately match the control method according to different thermal risk levels. After the control is completed, the battery heat dissipation data, thermal management condition assessment value, thermal risk level assessment value, and thermal control command parameters are structured and stored to construct a strategy response dataset. This method improves the accuracy, flexibility, and safety of thermal control response, ensuring logical coherence and closed-loop execution throughout the entire thermal risk handling process.

[0057] Specifically, based on the strategy response dataset, the specific steps for evaluating the implementation deviation of the thermal control strategy are as follows: After each thermal control strategy is executed, the corresponding strategy response dataset is extracted; the real-time speed of the condenser fan and the real-time speed of the liquid cooling pump are collected synchronously to ensure timing consistency and parameter matching; the real-time speed of the condenser fan is divided by the sum of the target speed of the condenser fan and the minimum term to obtain the fan execution ratio; the real-time speed of the liquid cooling pump is divided by the sum of the target speed of the liquid cooling pump and the minimum term to obtain the liquid cooling pump execution ratio; the fan execution ratio and the liquid cooling pump execution ratio are added together to construct the execution capability response parameter, and multiplied by the thermal risk level assessment value to obtain the cooling execution item; the square of the cell current is divided by the square of the cell rated current and the sum of the minimum term to calculate the current load item, which serves as a quantitative indicator of load intensity; finally, the cell temperature change rate is subtracted from the cooling execution item and then the current load item is added, and the absolute value of the calculation result is taken to obtain the thermal response deviation assessment value, which is used to measure the response consistency and control deviation degree between the thermal control strategy and the actual operating state.

[0058] The specific formula for calculating the thermal response deviation assessment value is as follows:

[0059] ;

[0060] In the formula, This represents the thermal response deviation assessment value. Indicates the rate of temperature change of the battery cell. This indicates the thermal risk level assessment value. This indicates the real-time speed of the condenser fan. This indicates the real-time speed of the liquid cooling pump. Indicates the target speed of the condenser fan. Indicates the target speed of the liquid cooling pump. Indicates the cell current. Indicates the rated current of the battery cell. Indicates a minus term.

[0061] In this implementation plan, a strategy deviation quantification model is constructed by introducing a comprehensive calculation mechanism for cooling execution items and current load items, with the thermal response deviation assessment value as the core. This effectively solves the problem of deviation identification between thermal control strategies and actual operating conditions. Compared with traditional assessment methods based on a single temperature error, this method comprehensively considers the ratio of the real-time speed of the condenser fan to the target speed, the ratio of the real-time speed of the liquid cooling pump to the target speed, the ratio of the cell current to the rated current, the cell temperature change rate, and the thermal risk level assessment value during the assessment process. This comprehensively reflects the coupling relationship between cooling control response efficiency and load thermal disturbance, thereby improving the accuracy and adaptability of thermal control strategy effect assessment.

[0062] Specifically, the steps for determining whether to trigger the strategy adjustment mechanism based on the evaluation results and forming a closed loop of adaptive strategy updates are as follows: Real-time comparison of the thermal response deviation evaluation value calculated within the current evaluation cycle with the response deviation threshold is performed to determine whether there is a significant deviation between the current thermal control strategy and the actual heat dissipation effect. When the thermal response deviation evaluation value is less than the response deviation threshold, it indicates that the matching relationship between the real-time speed of the condenser fan, the real-time speed of the liquid cooling pump, the rate of change of the battery cell temperature, and the battery cell current intensity is reasonable, and the thermal management strategy response effect is good. The current thermal control strategy parameters remain unchanged, and the next evaluation cycle begins. When the thermal response deviation evaluation value is greater than or equal to the response deviation threshold, it indicates that there is a deviation between the thermal control response and the thermal risk state, triggering the strategy adjustment mechanism: The historical thermal response deviation evaluation value sequence stored within a fixed-length sliding window and the corresponding real-time speed data of the liquid cooling pump and the condenser fan are extracted. The mapping function between the target speed and the thermal risk level evaluation value is dynamically regressed using a sliding window linear regression algorithm. The target speed mapping parameters of the liquid cooling pump 7 and the condenser 6 fan are corrected in real time, and the corrected target speed mapping parameters are written into the strategy configuration cache for loading and execution in the next evaluation cycle, ensuring that the control parameters are adaptively updated with dynamic changes in thermal response. The strategy configuration cache is used to temporarily store the target speed mapping parameters of the liquid cooling pump 7 and condenser 6 fan generated after the thermal control strategy is adjusted, so that they can be directly called in the next evaluation cycle to realize the immediate application and closed-loop adaptive update of the strategy.

[0063] In this implementation scheme, the effectiveness and execution deviation of the thermal control strategy are dynamically identified by real-time comparison of the thermal response deviation assessment value and the response deviation threshold. If the strategy deviation exceeds the limit, the target speed mapping parameters are corrected online, and the corrected parameters are written to the strategy configuration cache in real time, providing an updated control basis for the next evaluation cycle. This method effectively enhances the adaptability of the strategy parameters to the actual thermal management state of battery 2 while ensuring the timeliness of thermal control response, improves the dynamic matching accuracy of the target speed setting, and ensures the closed-loop linkage between the thermal risk level assessment value and the control execution, thereby improving the stability, accuracy, and practicality of the thermal management strategy.

[0064] 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.

[0065] 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. Clearly, 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 multi-stage intelligent adaptive control method for battery thermal isolation and heat dissipation integration, characterized in that, Includes the following steps: S1 periodically collects battery heat dissipation data and performs time alignment, anomaly removal, standardization and normalization on the battery heat dissipation data to obtain preprocessed battery heat dissipation data. S2, based on the preprocessed battery heat dissipation data, evaluate the thermal management status of battery (2), classify the battery thermal management status, and when a thermal risk status is continuously detected, start the thermal status confirmation mechanism and send a heat dissipation control execution signal after confirmation. S3 receives the thermal regulation execution signal and assesses the thermal risk level of the battery (2), and executes the passive thermal storage mode, active liquid cooling mode and emergency protection mode in stages. After the strategy is executed, a strategy response dataset is constructed. S4, based on the strategy response dataset, evaluates the implementation deviation of the thermal control strategy, and determines whether to trigger the strategy adjustment mechanism based on the evaluation results, thus forming a closed loop of strategy adaptive update.

2. The battery thermal isolation, heat dissipation, integrated multi-stage intelligent adaptive control method of claim 1, wherein: The specific steps for periodically collecting battery heat dissipation data are as follows: Set a sampling period for a fixed-length sliding window to collect battery heat dissipation data, including cell temperature, phase change material temperature, coolant inlet temperature, coolant outlet temperature, battery pack casing temperature, cell voltage, and cell current.

3. The multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation according to claim 1, characterized in that: The specific steps for performing time alignment, anomaly removal, standardization, and normalization on the battery heat dissipation data to obtain preprocessed battery heat dissipation data are as follows: Automatic calibration of battery heat dissipation data is performed through multi-source calibration and drift correction strategies to eliminate baseline errors caused by hardware differences and environmental changes between acquisition channels. Noise suppression methods based on time-domain sliding windows and adaptive filtering algorithms are used to denoise the battery heat dissipation data, filtering out transient interference and high-frequency noise. Anomaly detection of battery heat dissipation data is performed through an anomaly screening mechanism combining physical evolution laws and dynamic interval thresholds to eliminate data anomalies that do not conform to actual operating conditions. Numerical standardization based on unified standard physical units is used to convert the format of battery heat dissipation data, achieving consistent data representation across different channels. Numerical compression techniques combining typical operating ranges and adaptive normalization mapping are used to normalize the battery heat dissipation data.

4. The multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation according to claim 1, characterized in that: The specific steps for evaluating the battery's thermal management condition based on the preprocessed battery heat dissipation data are as follows: A fixed sliding time window is set as the evaluation period for thermal management status. Within each evaluation period, the cell temperature sequence is extracted, the difference between the maximum and minimum cell temperature is calculated to obtain the maximum cell temperature difference, and the average cell temperature is calculated. The thermal imbalance factor is obtained by dividing the square of the maximum temperature difference of the battery cell by the sum of the average cell temperature and the minimum term. The heat transfer efficiency factor is obtained by subtracting the coolant inlet temperature from the average cell temperature, dividing by the difference between the coolant outlet temperature and the coolant inlet temperature, adding the minimum term, adding one to the resulting ratio, and taking the natural logarithm. The phase change material (PCM) temperature is then subtracted from the lower limit of the PCM temperature, divided by the difference between the upper limit of the PCM temperature and the minimum operating temperature, adding the minimum term, subtracting one from the resulting ratio, and taking the square. The phase change heat storage capacity factor is obtained by multiplying the thermal imbalance factor, the heat transfer efficiency factor, and the phase change heat storage capacity factor together.

5. The multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation according to claim 4, characterized in that: The specific steps for dividing the battery thermal management state are as follows: Compare the thermal management condition evaluation value H with the multi-level condition thresholds H1 and H2 in real time to divide the battery thermal management state: When H ≤ H1, it is in a thermal stable state and no treatment is required; When H1 < H < H2, it is in a thermal rising state and the sampling frequency is increased; When H ≥ H2, it is in a thermal risk state and a warning signal is sent to the control unit.

6. The multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation according to claim 5, characterized in that: The specific steps for starting the thermal state confirmation mechanism and sending a thermal regulation execution signal after confirmation when the thermal risk state is continuously detected are as follows: When the control unit continuously receives a fixed number of warning signals, start the thermal state confirmation mechanism: extract the cell temperature sequence within the evaluation period, calculate the cell temperature change rate. When the cell temperature change rate is always positive and greater than the temperature rise threshold, it is determined that there is a continuous thermal risk trend and a thermal regulation execution signal is sent; otherwise, clear the warning cumulative count and maintain the current thermal management state unchanged.

7. The multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation according to claim 1, characterized in that: The specific steps for receiving the thermal regulation execution signal and evaluating the thermal risk level of the battery (2) are as follows: The execution unit receives the thermal regulation execution signal and extracts the real-time battery heat dissipation data, thermal management condition evaluation value, and the average cell temperature of the current evaluation period; divide the square of the cell current by the square of the cell rated current and the sum of the minterms, add one to the obtained result to get the current load factor; subtract the coolant outlet temperature from the battery pack housing temperature, divide by the sum of the average cell temperature and the minterms, add one to the obtained result to get the heat dissipation blockage factor; multiply the thermal management condition evaluation value, current load factor, and heat dissipation blockage factor in sequence to get the thermal risk level evaluation value.

8. The multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation according to claim 7, characterized in that: The specific steps for hierarchically executing the passive heat storage mode, active liquid cooling mode, and emergency protection mode and constructing a policy response dataset after the strategy execution are as follows: thermal risk level assessment value The thermal regulation strategy is compared with the thresholds C1 and C2 of the multi-level control strategy, and the strategy is executed in stages: When C ≤ C1, it is determined that there is a first-level risk and the passive heat storage mode is triggered: keep the liquid cooling system closed and rely on the phase change material (3) to absorb heat; When C1 < C < C2, it is determined that there is a second-level risk and the active liquid cooling mode is triggered: send a start-stop control instruction to the thermal management execution unit, start the liquid cooling pump (7) and the condenser (6) fan, and set the target speeds of the liquid cooling pump (7) and the condenser (6) fan respectively according to the linear mapping result of the thermal risk level evaluation value and the control strategy threshold interval; When C ≥ C2, it is determined that there is a third-level risk and the emergency protection mode is triggered: adjust both the liquid cooling pump (7) and the condenser (6) fan to the highest speed, activate the emergency heat dissipation ventilation circuit, cut off the charge-discharge circuit of the corresponding cell, and send an alarm signal to the main control system; After the thermal regulation strategy execution is completed, structurally store the battery heat dissipation data, thermal management condition evaluation value, thermal risk level evaluation value, liquid cooling pump target speed, and condenser fan target speed before and after the strategy execution to construct a policy response dataset.

9. The multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation according to claim 1, characterized in that: The specific steps for evaluating the implementation deviation of the thermal regulation strategy based on the policy response dataset are as follows: After each thermal control strategy is executed, the corresponding strategy response dataset is extracted, and the real-time speeds of the condenser fan and liquid cooling pump are collected. The real-time speed of the condenser fan is divided by the sum of the target speed and the minimum term of the condenser fan to obtain the fan execution ratio. The real-time speed of the liquid cooling pump is divided by the sum of the target speed and the minimum term of the liquid cooling pump to obtain the liquid cooling pump execution ratio. The fan execution ratio and the liquid cooling pump execution ratio are added together and multiplied by the thermal risk level assessment value to obtain the cooling execution item. The square of the cell current is divided by the square of the cell rated current and the sum of the minimum term to obtain the current load item. The cell temperature change rate is subtracted from the cooling execution item and then added to the current load item. The absolute value is taken to obtain the thermal response deviation assessment value.

10. The multi-stage intelligent adaptive control method for integrated battery thermal isolation and heat dissipation according to claim 9, characterized in that: The specific steps for determining whether to trigger the policy adjustment mechanism based on the evaluation results, thus forming a closed loop of policy adaptive update, are as follows: The thermal response deviation assessment value and the response deviation threshold are compared in real time. When the thermal response deviation assessment value is less than the response deviation threshold, the current thermal control strategy remains unchanged. When the thermal response deviation assessment value is greater than or equal to the response deviation threshold, the strategy adjustment mechanism is triggered: historical thermal response deviation assessment values ​​and corresponding real-time speeds of liquid cooling pumps and condenser fans are extracted, the target speed mapping parameters are recalculated using a sliding window linear regression algorithm, and the corrected target speed mapping parameters are stored in the system configuration cache for loading and use in the next evaluation cycle.

Images