A BMS-based fuzzy PID adaptive liquid cooling control method based on full-dimensional thermal model prediction, electronic devices, and storage media.
By using a full-dimensional thermal model for prediction and fuzzy PID adaptive control, the problems of lag and poor adaptability in the liquid cooling control of power batteries and energy storage batteries have been solved, achieving high-precision and adaptive battery thermal management, improving battery consistency and lifespan, and reducing energy consumption.
Patent Information
- Application Number
- CN202611469573.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-21
- Publication Date
- 2026-11-17
AI Technical Summary
Existing liquid cooling control technologies for power batteries and energy storage batteries suffer from problems such as lag, singleness, and fixed parameters, which cannot meet the requirements for high-precision, adaptive, and high-safety thermal management. This results in low battery temperature control accuracy, poor thermal uniformity, high energy consumption, and failure to effectively address issues such as battery aging and excessive temperature differences.
A BMS fuzzy PID adaptive liquid cooling control method based on full-dimensional thermal model prediction is adopted. By constructing a multi-parameter coupled battery thermal model, an adaptive dynamic duration thermal prediction mechanism is designed, and combined with fuzzy PID adaptive control strategy and pipeline pressure and flow correction, the battery liquid cooling system can achieve advanced predictive temperature control and full-condition adaptation, and prioritize the suppression of temperature difference expansion.
It achieves improved battery thermal safety redundancy, improved battery cell consistency, reduced energy consumption for energy storage and vehicle operation, extended battery cycle life, and improved temperature control accuracy, solving the problems of lag and poor adaptability to operating conditions in traditional liquid cooling control.
Smart Images

Figure CN123113296A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of liquid cooling control technology for power batteries or energy storage devices, and in particular to a BMS fuzzy PID adaptive liquid cooling control method based on full-dimensional thermal model prediction, electronic equipment, and storage medium. Background Technology
[0002] Currently, liquid cooling solutions are commonly used in new energy power batteries, industrial and commercial energy storage, grid-side energy storage, and outdoor container energy storage systems. The quality of the liquid cooling BMS control strategy directly determines the battery's temperature control accuracy, thermal balance, service life, operational safety, and system energy consumption.
[0003] Existing liquid cooling thermal management technologies for power batteries and energy storage battery packs generally suffer from several industry-wide technical deficiencies: Liquid cooling control often employs fixed temperature threshold start / stop and constant flow operation, relying solely on the battery's current highest temperature for lagging closed-loop control. This fails to identify battery heat accumulation trends and temperature rise rates, resulting in lag in the mechanical and heat exchange responses of the liquid cooling system. Under fast charging and high-rate discharging conditions, it is highly susceptible to temperature overshoot, temperature oscillations, and uncontrolled temperature rise. Furthermore, the control algorithms are mostly fixed-parameter PID or basic fuzzy PID, failing to adapt to varying operating conditions such as battery aging and degradation, SOC polarization differences, uneven battery pack temperature distribution, ambient temperature fluctuations, and liquid cooling pipeline pressure losses. This leads to poor compatibility between new and old batteries, and long-term operation results in a continuous deterioration in the consistency of individual battery cells.
[0004] The battery thermal model has a single modeling dimension, relying solely on current and temperature for heat generation calculations. It fails to integrate parameters from multiple dimensions, including battery temperature difference, state of health (SOH), charge / discharge rate, liquid cooling pipeline pressure, and inlet / outlet water temperature. This results in low accuracy in thermal state calculations, failing to accurately represent the battery's true heat generation, dissipation, and heat storage status. Fixed-duration temperature prediction is commonly used, lacking an adaptive prediction mechanism for different operating conditions. This makes it unsuitable for varying temperature rise rates, such as steady-state low-speed operation, normal operation, and high-rate fast charging. Insufficient prediction leads to delayed heat dissipation, while excessive prediction time results in large prediction errors and wasted computing power. Temperature control and thermal equilibrium are disconnected. Focusing only on achieving the overall battery temperature target, it neglects to prioritize and control the temperature difference between individual battery cells. Prolonged excessive temperature differences lead to accelerated localized battery degradation, reducing the overall battery pack lifespan and thermal safety redundancy. The model does not consider the actual flow rate reduction caused by liquid cooling pipeline aging, pressure fluctuations, and minor blockages. This results in a large discrepancy between theoretical and actual heat dissipation flow rates, leading to a continuous decline in temperature control accuracy.
[0005] In summary, most current mainstream BMS liquid cooling control technologies employ fixed threshold start / stop and constant flow operation logic, combined with traditional fixed-parameter PID algorithms to complete temperature closed-loop control. This type of technology only collects the battery's real-time highest temperature as the control basis, without modeling and analyzing the battery's heat generation mechanism, heat accumulation trends, pipeline heat dissipation losses, and battery aging characteristics, thus exhibiting significant technical shortcomings.
[0006] At present, some improvement solutions introduce fuzzy PID algorithms to optimize temperature control, but still rely on real-time temperature deviation for passive closed-loop control. They have not established a multi-dimensional thermal state model of the battery, lack heat prediction feedforward capability, and cannot solve core problems such as system lag, poor adaptability to operating conditions, insufficient temperature difference suppression, and inaccurate pipeline flow.
[0007] As the energy storage and new energy vehicle industries continue to increase their requirements for battery thermal safety, consistency, lifespan, and energy consumption, traditional lagging, single-parameter, and fixed-parameter liquid cooling control strategies can no longer meet the demands for high-precision, adaptive, and high-safety thermal management. Therefore, there is an urgent need for an adaptive intelligent liquid cooling control technology based on multi-dimensional thermal model prediction. Summary of the Invention
[0008] The purpose of this invention is to overcome the many shortcomings of existing liquid cooling control technologies for power batteries and energy storage batteries, and to provide a BMS fuzzy PID adaptive liquid cooling control method, electronic device, and storage medium based on a full-dimensional thermal model prediction. By constructing a multi-parameter coupled accurate battery thermal model, designing an adaptive dynamic duration thermal prediction mechanism, establishing a thermal parameter-driven fuzzy PID adaptive control strategy, and adding battery temperature difference-priority hard constraints and pipeline pressure and flow correction logic, the technical goals of advanced predictive temperature control, full-condition adaptation, high-precision temperature control, thermal balance priority, and low-energy operation of the battery liquid cooling system are achieved. Ultimately, this improves battery thermal safety redundancy, enhances battery cell consistency, delays battery degradation, and reduces energy consumption for energy storage and vehicle operation.
[0009] According to a first aspect of this application, a BMS fuzzy PID adaptive liquid cooling control method based on full-dimensional thermal model prediction is provided, comprising:
[0010] S1.BMS collects battery status parameters, environmental parameters, and liquid cooling circuit parameters in real time and in all dimensions, including the battery's highest temperature, lowest temperature, average temperature, real-time maximum temperature difference, SOC, SOH, charge / discharge current, voltage, charge / discharge rate, ambient temperature, liquid cooling inlet temperature, liquid cooling outlet temperature, and real-time pressure of the liquid cooling pipeline.
[0011] S2. Based on the constructed multi-coupled battery thermal model, the basic heat generation power of the battery is calculated based on the collected parameters. The real-time heat generation of the battery is obtained by combining the SOH aging correction coefficient, SOC polarization correction coefficient, temperature difference unevenness correction coefficient, and charge / discharge rate correction coefficient. The net heat storage and temperature rise rate of the battery cell are calculated. The effective flow rate of liquid cooling is corrected by combining the pipeline pressure. Based on the corrected effective flow rate of liquid cooling and the temperature difference between the inlet and outlet water of liquid cooling, the real-time heat dissipation of liquid cooling is calculated. The natural heat dissipation of the environment is superimposed to calculate the net heat storage and temperature rise rate of the battery pack. The maximum value of the net heat storage and temperature rise rate of the battery cell and the battery pack is taken as the real-time net heat storage and real-time temperature rise rate.
[0012] S3. Based on the real-time temperature rise rate of the battery, dynamically match the prediction time. If the real-time temperature rise rate is greater than the set threshold, output the multi-dimensional predicted thermal parameters of the battery within the prediction time corresponding to the current temperature rise rate based on the weighted time series prediction algorithm. These parameters include at least the predicted average temperature, the predicted maximum temperature difference, the predicted net heat storage, and the predicted temperature rise rate.
[0013] S4. Based on the battery's baseline optimal temperature, real-time temperature rise rate, predicted maximum temperature difference, battery aging degree, and charge / discharge rate, dynamically determine the optimal temperature control target value for the battery.
[0014] S5. The deviation between the predicted average temperature and the battery's optimal temperature control target value and the deviation rate of the fused net heat storage change rate are used as inputs to the fuzzy controller. The PID parameters are adaptively corrected through preset fuzzy rules, and the output of the continuous control quantity is controlled in the first PID control mode to regulate the flow of the liquid cooling system.
[0015] S6. If the real-time maximum temperature difference / predicted maximum temperature difference of the battery exceeds the temperature difference threshold, the second PID control mode is used to control the output of continuous control quantity to regulate the flow of the liquid cooling system and suppress the expansion of the battery temperature difference.
[0016] Preferably, in step S2, the real-time heat generation of the battery is determined by the following calculation formula:
[0017] ;
[0018] ;
[0019] in, For the basic heat generation power of the battery, This is the SOH aging correction factor. This is the SOC polarization correction factor. This is the correction factor for uneven temperature distribution. This is the charge / discharge rate correction factor; This is the charging and discharging current. Open circuit voltage, For voltage, This indicates the absolute temperature of the battery.
[0020] Preferably, the calculation formula for the effective liquid cooling flow rate, which is calculated in real time by correcting the pipeline pressure, is as follows:
[0021] ;
[0022] In the formula, Indicates the effective flow rate of liquid cooling. P is the rated flow rate of the liquid cooling pump, and P is the real-time pipeline pressure. The standard operating pressure for the pipeline.
[0023] Preferably, the step of dynamically matching and predicting the duration based on the real-time temperature rise rate of the battery is performed based on a preset operating condition matching rule, which includes:
[0024] ;
[0025] in, This indicates the predicted duration, and dT / dt is the real-time temperature rise rate of the battery.
[0026] Preferably, the formula for calculating the optimal temperature control target value of the battery includes:
[0027] ;
[0028] In the formula, This indicates the optimal target temperature for battery control. The optimal reference temperature for the battery is set between 25℃ and 30℃. To predict the maximum temperature difference, , , , This is the corresponding preset working condition correction factor. This indicates the charge / discharge rate.
[0029] Preferably, the deviation rate of the temperature deviation and the deviation rate of the net heat storage change rate are calculated and determined using the following formula:
[0030] ; ;
[0031] In the formula, Indicates temperature deviation. The deviation rate of change of the net heat storage rate. To predict the average temperature, The heat storage weight correction factor is jointly calibrated by SOC and ambient temperature; It represents the real-time net heat storage of the battery, and outputs the correction values of the PID proportional parameter Kp, integral parameter Ki, and derivative parameter Kd through fuzzy rules, and updates the PID parameters in real time.
[0032] Preferably, if the real-time maximum temperature difference / predicted maximum temperature difference of the battery does not exceed the temperature difference threshold, or if the real-time temperature rise rate does not exceed the set threshold, the first PID control mode is used to control the output of a continuous control quantity to regulate the flow of the liquid cooling system.
[0033] Preferably, when the predicted temperature rise rate, predicted average temperature, predicted net heat storage, and predicted maximum temperature exhibit a preset trend or state, liquid cooling pre-action is performed according to the pre-start strategy based on the predicted thermal parameters, including:
[0034] Level 1 Prediction and Pre-start: If the predicted temperature rise rate is increasing in the next 3 seconds and the predicted net heat storage is accumulating, the liquid cooling pump will be started in advance to run at a preset low speed in standby mode.
[0035] Secondary advance frequency increase: The predicted average temperature in the next 5 seconds is close to the average temperature threshold, and the predicted maximum temperature difference shows an increasing trend. The PID parameters are corrected in advance to increase the liquid cooling flow rate.
[0036] Level 3 Emergency Enhancement: If the predicted average temperature for the next 8 seconds exceeds the average temperature threshold and the rate of temperature rise exceeds the maximum rate of temperature rise threshold, the throttle valve will be fully opened and the liquid cooling pump will be increased to maximum power operation.
[0037] According to a second aspect of this application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the BMS fuzzy PID adaptive liquid cooling control method when executing the computer program.
[0038] According to a third aspect of this application, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the BMS fuzzy PID adaptive liquid cooling control method.
[0039] This invention employs a core technical architecture that integrates multi-dimensional parameter acquisition, real-time calculation using multi-coupled thermal models, dynamic duration-based thermal prediction, dynamic target tuning, fuzzy PID adaptive control, temperature difference constraints and pipeline correction, and a closed-loop liquid cooling execution mechanism to achieve effective thermal management of battery systems. In particular, it prioritizes battery temperature difference hard constraint logic, suppressing thermal unevenness when the temperature difference exceeds the limit; combined with pipeline flow correction errors, stepless adjustment of the liquid cooling pump PWM speed and electronic throttle valve opening from 0-100%, and a three-level advanced start-up strategy, it achieves precise liquid cooling temperature control and thermal equilibrium control under all operating conditions. Attached Figure Description
[0040] Figure 1 A schematic diagram of the closed-loop process of fuzzy PID adaptive liquid cooling control provided in this embodiment of the invention. Detailed Implementation
[0041] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
[0042] See Figure 1 As shown, embodiments of this application provide a BMS fuzzy PID adaptive liquid cooling control method based on full-dimensional thermal model prediction, including:
[0043] S1. Collect battery status parameters, environmental parameters, and liquid cooling circuit parameters:
[0044] The BMS collects real-time, multi-dimensional battery status parameters, environmental parameters, and liquid cooling circuit parameters, including the battery's highest temperature, lowest temperature, average temperature, real-time maximum temperature difference, SOC, SOH, charge / discharge current, voltage, charge / discharge rate, ambient temperature, liquid cooling inlet temperature, liquid cooling outlet temperature, and real-time pressure of the liquid cooling pipeline.
[0045] S2. Determine the real-time net heat storage capacity and real-time temperature rise rate of the battery based on the multi-coupled battery thermal model;
[0046] Based on the constructed multi-coupled battery thermal model, the basic heat generation power of the battery is calculated based on the collected parameters. The real-time heat generation of the battery is obtained by combining the SOH aging correction coefficient, SOC polarization correction coefficient, temperature difference unevenness correction coefficient, and charge / discharge rate correction coefficient. The net heat storage and temperature rise rate of the battery cell are calculated. The effective flow rate of liquid cooling is corrected by combining the pipeline pressure. Based on the corrected effective flow rate of liquid cooling and the temperature difference between the inlet and outlet water of liquid cooling, the real-time heat dissipation of liquid cooling is calculated. The natural heat dissipation of the environment is superimposed to calculate the net heat storage and temperature rise rate of the battery pack. The maximum value of the net heat storage and temperature rise rate of the battery cell and the battery pack is taken as the real-time net heat storage and real-time temperature rise rate.
[0047] S3. Dynamic matching prediction duration; predicting multi-dimensional thermal parameters of the battery when the real-time temperature rise rate exceeds the threshold:
[0048] Based on the real-time temperature rise rate of the battery, the prediction time is dynamically matched. If the real-time temperature rise rate is found to be greater than the set threshold, the multi-dimensional predicted thermal parameters of the battery within the prediction time corresponding to the current temperature rise rate are output based on the weighted time series prediction algorithm. These parameters include at least the predicted average temperature, the predicted maximum temperature difference, the predicted net heat storage, and the predicted temperature rise rate.
[0049] S4. Based on multi-dimensional predicted thermal parameters of the battery, dynamically determine the optimal temperature control target value of the battery:
[0050] Based on the battery's baseline optimal temperature, real-time temperature rise rate, predicted maximum temperature difference, battery aging degree, and charge / discharge rate, the optimal temperature control target value for the battery is dynamically determined.
[0051] S5. Adjust the flow rate of the liquid cooling system in the first PID control mode according to the optimal temperature control target value of the battery;
[0052] The deviation between the predicted average temperature and the battery's optimal temperature control target value, and the deviation rate of the fused net heat storage change rate are used as inputs to the fuzzy controller. The PID parameters are adaptively corrected through preset fuzzy rules, and the output continuous control quantity is controlled in the first PID control mode to regulate the flow rate of the liquid cooling system.
[0053] S6. When the temperature difference exceeds the first PID control mode, the flow regulation of the liquid cooling system is used to suppress the expansion of the battery temperature difference.
[0054] If the real-time maximum temperature difference / predicted maximum temperature difference of the battery exceeds the temperature difference threshold, the second PID control mode is used to control the output of a continuous control quantity to regulate the flow of the liquid cooling system and suppress the expansion of the battery temperature difference.
[0055] In this application, considering the hard constraint of battery pack temperature difference, priority is given to suppressing the expansion of battery temperature difference. This is achieved by steplessly adjusting the PWM speed of the liquid cooling pump and the opening of the electronic throttle valve, thus realizing feedforward prediction, closed-loop precise temperature control, and battery pack thermal balance control of the liquid cooling system. Specifically, the PID integral parameter Ki can be locked, and the proportional parameter Kp can be fine-tuned. The liquid cooling flow rate is prioritized to suppress the expansion of battery pack temperature difference, and then the overall temperature deviation is corrected, achieving a thermal management strategy that prioritizes thermal balance and uses temperature control as a secondary measure.
[0056] This application acquires four-dimensional parameters (T) of the battery through the collection of parameters across all dimensions, such as the BMS simultaneously acquiring four-dimensional parameters of battery temperature with a high precision period of 100ms. max T min T avg ΔT), battery electrical performance parameters (SOC, SOH, I, U, C-rate), and ambient temperature parameters (T). env ), liquid cooling circuit parameters (T) in T out The data (including pipeline pressure P) covers the entire process of battery heat generation, environmental heat exchange, and liquid cooling, providing a data foundation for accurately characterizing the battery's thermal state.
[0057] In this application, in step S2, a parallel computing link is designed, which is divided into two computing links: cell heat generation and liquid cooling heat dissipation. These links calculate the net heat storage and temperature rise rate of the cell and the net heat storage and temperature rise rate of the battery pack, respectively. The cell heat generation link is calculated sequentially by Bernardi basic heat generation calculation, and then by four corrections: SOH, SOC, temperature difference, and rate. The net heat storage and temperature rise rate of the cell are solved by thermal balance one. The liquid cooling heat dissipation link is calculated by correcting the effective flow rate of liquid cooling by pipeline pressure, liquid cooling forced heat dissipation and ambient natural heat dissipation. The net heat storage and temperature rise rate of the entire battery pack are solved by thermal balance two. The results of the two thermal balance calculations are summarized to the output local thermal parameter node, which outputs global high-precision thermal state parameters. The maximum value of these parameters is taken as the real-time net heat storage and real-time temperature rise rate for subsequent calculations.
[0058] During real-time control, the thermal prediction cycle is first adaptively matched based on the temperature rise rate. Then, it is determined whether the temperature rise rate exceeds the limit. If it does not exceed the limit, the continuous control output is used based on the first PID control mode (conventional PID control mode) to regulate the flow of the liquid cooling system. If it exceeds the limit, the temperature difference is determined again based on the thermal parameter prediction results. If the temperature difference exceeds the limit, the second PID control mode (enhanced PID control mode) is used to control the continuous control output to regulate the flow of the liquid cooling system and suppress the expansion of the battery temperature difference. In the PID control mode, the fuzzy PID algorithm solves for the optimal heat dissipation command, which is then sent to the liquid cooling actuator and terminated. The operating data feedback of the liquid cooling actuator refreshes the local thermal parameter input, and the thermal parameter prediction results refresh the prediction time benchmark, forming a closed-loop control.
[0059] In the case of fuzzy PID control, the optimal temperature control target value is dynamically tuned based on real-time and predicted thermal parameters; the deviation rate between temperature deviation and the fused net heat storage change rate is calculated, and the PID proportional, integral, and derivative parameters are adaptively corrected through fuzzy rules; it is determined whether the real-time and predicted maximum temperature difference of the battery exceeds the standard. If it does, the integral parameter Ki is locked, and the battery pack temperature difference is balanced first. If it does not exceed the standard, the PWM control quantity is directly output to adjust the opening of the liquid cooling pump and the electronic throttle valve; combined with a three-level advanced start-up strategy, active heat dissipation regulation is achieved, and the parameter feedback thermal model is continuously iterated and optimized to complete closed-loop control.
[0060] In particular, based on fuzzy PID adaptive control, with the predicted temperature deviation and the deviation change rate of the fused net heat storage change rate as dual inputs, the PID proportional, integral, and derivative parameters are adaptively corrected in real time through fuzzy rules, which can achieve fast response to large deviations, steady-state fine-tuning of small deviations, and adaptive adaptation to changing operating conditions.
[0061] In an optional implementation, in step S2, the real-time heat generation of the battery is determined by the following calculation formula:
[0062] ;
[0063] ;
[0064] in, For the basic heat generation power of the battery, This is the SOH aging correction factor. This is the SOC polarization correction factor. This is the correction factor for uneven temperature distribution. This is the charge / discharge rate correction factor; This is the charging and discharging current. Open circuit voltage, For voltage, This indicates the absolute temperature of the battery.
[0065] In one optional implementation, the calculation formula for the effective liquid cooling flow rate, which is calculated by real-time correction of pipeline pressure, is as follows:
[0066] ;
[0067] In the formula, Indicates the effective flow rate of liquid cooling. P is the rated flow rate of the liquid cooling pump, and P is the real-time pipeline pressure. The standard operating pressure for the pipeline.
[0068] This application employs a multi-coupled battery thermal model, based on a modified Bernardi heat generation formula, and superimposed with SOH aging correction, SOC polarization correction, temperature difference unevenness correction, and rate nonlinearity correction to accurately calculate the real-time heat generation of the battery. The actual effective flow rate of liquid cooling is corrected by pipeline pressure, and the liquid cooling heat dissipation is calculated by combining the inlet and outlet water temperature difference. The ambient natural heat dissipation is superimposed, and the net heat storage of the battery and the real-time temperature rise rate are solved by the heat balance equation, accurately characterizing the instantaneous thermal state of the battery and providing an effective foundation for subsequent liquid cooling control.
[0069] In one optional implementation, the step of dynamically matching the predicted duration based on the real-time temperature rise rate of the battery is performed based on a preset operating condition matching rule, which includes:
[0070] ;
[0071] in, This indicates the predicted duration, and dT / dt is the real-time temperature rise rate of the battery.
[0072] In this application, the design matches the prediction time according to the real-time battery operating conditions: a 3-second prediction time for low-speed steady-state conditions, a 5-second baseline prediction time for normal driving / charging conditions, and an 8-second prediction time for high-power fast charging / high-rate discharging high-temperature rise conditions. Based on a weighted time-series prediction algorithm, it outputs multi-dimensional predicted thermal parameters, including the predicted future temperature, predicted maximum temperature difference, predicted net heat storage, and predicted temperature rise rate. Based on the predicted thermal parameters and battery state parameters, the optimal battery temperature control target value T is dynamically tuned. set It is adapted to the heat dissipation needs of batteries with different aging levels, different charge / discharge rates, and different temperature differences.
[0073] In this application, the operating conditions are determined based on the real-time temperature rise rate of the battery. For example, the prediction duration is adaptively matched to 3s (low-speed steady state), 5s (normal operating conditions), and 8s (high-rate fast charging / high-load operating conditions). The predicted temperature, predicted temperature difference, predicted net heat storage, and predicted temperature rise rate are multi-dimensional thermal parameters within the future prediction cycle, so as to realize differentiated advance prediction and provide conditions for the liquid cooling system to start in advance or ahead of time.
[0074] In one embodiment, the formula for calculating the optimal temperature control target value of the battery includes:
[0075] ;
[0076] In the formula, This indicates the optimal temperature control target for the battery, ranging from 25℃ to 30℃; The optimal reference temperature for the battery is set between 25℃ and 30℃. To predict the maximum temperature difference, , , , This is the corresponding preset working condition correction factor. This indicates the charge / discharge rate.
[0077] In this application, the optimal temperature control target value is dynamically calculated by combining the battery reference optimal temperature, temperature rise rate, predicted temperature difference, battery aging degree, and charge / discharge rate. The heat dissipation margin is automatically reserved for aging batteries, high-rate operating conditions, and conditions with increased temperature difference, providing a target for subsequent PID control.
[0078] In one optional implementation, the deviation rate of the temperature deviation and the deviation rate of the net heat storage change rate are calculated and determined using the following formula:
[0079] ; ;
[0080] In the formula, Indicates temperature deviation. The deviation rate of change of the net heat storage rate. To predict the average temperature, The heat storage weight correction factor is jointly calibrated by SOC and ambient temperature; It represents the real-time net heat storage of the battery, and outputs the correction values of the PID proportional parameter Kp, integral parameter Ki, and derivative parameter Kd through fuzzy rules, and updates the PID parameters in real time.
[0081] In one implementation, if the real-time maximum temperature difference / predicted maximum temperature difference of the battery does not exceed the temperature difference threshold, or if the real-time temperature rise rate does not exceed the minimum temperature rise rate threshold, the system controls the output of a continuous control quantity in a first PID control mode to regulate the flow rate of the liquid cooling system.
[0082] In one optional implementation, when the predicted temperature rise rate, predicted average temperature, predicted net heat storage, and predicted maximum temperature exhibit a preset trend or state, liquid cooling pre-action is performed according to a pre-start strategy based on the predicted thermal parameters, including:
[0083] Level 1 Prediction and Pre-start: If the predicted temperature rise rate is increasing in the next 3 seconds and the predicted net heat storage is accumulating, the liquid cooling pump will be started in advance to run at a preset low speed in standby mode.
[0084] Secondary advance frequency increase: The predicted average temperature in the next 5 seconds is close to the average temperature threshold, and the predicted maximum temperature difference shows an increasing trend. The PID parameters are corrected in advance to increase the liquid cooling flow rate.
[0085] Level 3 Emergency Enhancement: If the predicted average temperature for the next 8 seconds exceeds the average temperature threshold and the rate of temperature rise exceeds the maximum rate of temperature rise threshold, the throttle valve will be fully opened and the liquid cooling pump will be increased to maximum power operation.
[0086] The proposed solution relies on thermal model heat prediction to achieve feedforward control, which completely solves the problems of temperature overshoot, oscillation, and heat dissipation lag in traditional pure time-delay closed-loop control. It enables the battery steady-state temperature control error to be controlled within ±0.8℃, and features high temperature control accuracy and no time-delay overshoot.
[0087] The solution proposed in this application has strong adaptability under all working conditions. It can adapt to all working conditions, from brand new batteries to aged batteries, low temperature, normal temperature, high temperature environment, static, driving, slow charging, high power fast charging, high rate discharge, etc., solving the problem of poor working condition adaptability of traditional algorithms.
[0088] This application features a unique temperature difference priority constraint control logic that prioritizes suppressing the expansion of temperature difference between individual battery cells, resulting in excellent battery thermal uniformity. It controls the steady-state temperature difference of the battery pack to within 2°C, significantly improving the consistency of individual battery cells and increasing battery cycle life by 8% to 15%.
[0089] This application integrates pipeline pressure to correct the actual liquid cooling flow rate in real time, offsetting heat dissipation errors caused by pipeline aging, pressure difference fluctuations, and minor blockages. The heat dissipation calculation is accurate and reliable, and the long-term temperature control stability is greatly improved.
[0090] This application shortens the prediction time and reduces the liquid cooling power in low-speed steady-state conditions, enhances heat dissipation in advance in high-risk conditions, matches heat dissipation power as needed, and reduces liquid cooling power consumption by 10% to 20% in low-load conditions, resulting in outstanding energy economy.
[0091] This application uses a dynamic graded prediction mechanism of 3s, 5s, and 8s and a three-level liquid cooling advanced start-up mechanism to identify heat accumulation and temperature rise trends in advance, effectively avoiding the risk of thermal runaway under fast charging and high-rate conditions, and providing higher thermal safety redundancy.
[0092] This application constructs a full-dimensional coupled battery thermal model that integrates SOH, SOC, temperature difference, rate, and pipeline pressure to achieve high-precision real-time calculation of battery heat generation, heat dissipation, and heat storage; based on a dynamic duration prediction strategy for temperature rise rate, it distinguishes between 3s, 5s, and 8s differentiated prediction cycles to adapt to the advanced temperature control requirements under all operating conditions.
[0093] This application adopts a quadruple-coupled control architecture, which includes thermal model feedforward prediction, fuzzy PID adaptive closed loop, temperature difference hard constraint priority, and pipeline pressure correction. Unlike the traditional single temperature closed-loop control, it innovatively places the PID control algorithm on the BMS side, which is beneficial for energy saving and precise temperature control.
[0094] This application uses the fusion of predicted temperature deviation and net heat storage change rate as inputs to the fuzzy controller to identify heat accumulation trends in advance and solve the control lag problem.
[0095] This application solution is compatible with all types of outdoor container energy storage, indoor integrated energy storage, industrial and commercial energy storage, and grid-side energy storage, and is compatible with various power batteries such as lithium iron phosphate and ternary lithium.
[0096] An embodiment of this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the BMS fuzzy PID adaptive liquid cooling control method.
[0097] An embodiment of this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the BMS fuzzy PID adaptive liquid cooling control method.
[0098] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention.
[0099] Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of the equivalents of the claims be included within the invention.
[0100] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art.
Claims
1. A BMS fuzzy PID adaptive liquid cooling control method based on full-dimensional thermal model prediction, characterized in that, include: S1.BMS collects battery status parameters, environmental parameters, and liquid cooling circuit parameters in real time and in all dimensions, including the battery's highest temperature, lowest temperature, average temperature, real-time maximum temperature difference, SOC, SOH, charge / discharge current, voltage, charge / discharge rate, ambient temperature, liquid cooling inlet temperature, liquid cooling outlet temperature, and real-time pressure of the liquid cooling pipeline. S2. Based on the constructed multi-coupled battery thermal model, the basic heat generation power of the battery is calculated based on the collected parameters. The real-time heat generation of the battery is obtained by combining the SOH aging correction coefficient, SOC polarization correction coefficient, temperature difference unevenness correction coefficient, and charge / discharge rate correction coefficient. The net heat storage and temperature rise rate of the cell are then calculated. The effective flow rate of liquid cooling is corrected by combining pipeline pressure. Based on the corrected effective flow rate of liquid cooling and the temperature difference between the inlet and outlet water of liquid cooling, the real-time heat dissipation of liquid cooling is calculated. The natural heat dissipation of the environment is added to calculate the net heat storage and temperature rise rate of the battery pack. The maximum value of the net heat storage and temperature rise rate of the battery cell and the battery pack is taken as the real-time net heat storage and real-time temperature rise rate. S3. Based on the real-time temperature rise rate of the battery, dynamically match the prediction time. If the real-time temperature rise rate is greater than the set threshold, output the multi-dimensional predicted thermal parameters of the battery within the prediction time corresponding to the current temperature rise rate based on the weighted time series prediction algorithm. These parameters include at least the predicted average temperature, the predicted maximum temperature difference, the predicted net heat storage, and the predicted temperature rise rate. S4. Based on the battery's baseline optimal temperature, real-time temperature rise rate, predicted maximum temperature difference, battery aging degree, and charge / discharge rate, dynamically determine the optimal temperature control target value for the battery. S5. The deviation between the predicted average temperature and the battery's optimal temperature control target value and the deviation rate of the fused net heat storage change rate are used as inputs to the fuzzy controller. The PID parameters are adaptively corrected through preset fuzzy rules, and the output of the continuous control quantity is controlled in the first PID control mode to regulate the flow of the liquid cooling system. S6. If the real-time maximum temperature difference / predicted maximum temperature difference of the battery exceeds the temperature difference threshold, the second PID control mode is used to control the output of continuous control quantity to regulate the flow of the liquid cooling system and suppress the expansion of the battery temperature difference.
2. The BMS fuzzy PID adaptive liquid cooling control method according to claim 1, characterized in that, In step S2, the real-time heat generation of the battery is determined by the following calculation formula: ; ; in, For the basic heat generation power of the battery, This is the SOH aging correction factor. This is the SOC polarization correction factor. This is the correction factor for uneven temperature distribution. This is the charge / discharge rate correction factor; This is the charging and discharging current. Open circuit voltage, For voltage, This indicates the absolute temperature of the battery.
3. The BMS fuzzy PID adaptive liquid cooling control method according to claim 2, characterized in that... The calculation formula for the effective liquid cooling flow rate, which is calculated in real time by correcting the pipeline pressure, is as follows: ; In the formula, Indicates the effective flow rate of liquid cooling. P is the rated flow rate of the liquid cooling pump, and P is the real-time pipeline pressure. The standard operating pressure for the pipeline.
4. The BMS fuzzy PID adaptive liquid cooling control method according to claim 2, characterized in that, The dynamic matching and prediction of duration based on the real-time temperature rise rate of the battery is performed based on preset operating condition matching rules, which include: ; in, This indicates the predicted duration, and dT / dt is the real-time temperature rise rate of the battery.
5. The BMS fuzzy PID adaptive liquid cooling control method according to claim 1, characterized in that, The formula for calculating the optimal temperature control target value of the battery includes: ; In the formula, This indicates the optimal target temperature for battery control. The optimal reference temperature for the battery is set between 25℃ and 30℃. To predict the maximum temperature difference, , , , This is the corresponding preset working condition correction factor. This indicates the charge / discharge rate.
6. The BMS fuzzy PID adaptive liquid cooling control method according to claim 5, characterized in that, The deviation rates of temperature and net heat storage are calculated and determined using the following formula: ; ; In the formula, Indicates temperature deviation. The deviation rate of change of the net heat storage rate. To predict the average temperature, The heat storage weight correction factor is jointly calibrated by SOC and ambient temperature; It represents the real-time net heat storage of the battery, and outputs the correction values of the PID proportional parameter Kp, integral parameter Ki, and derivative parameter Kd through fuzzy rules, and updates the PID parameters in real time.
7. The BMS fuzzy PID adaptive liquid cooling control method according to claim 6, characterized in that, If the real-time maximum temperature difference / predicted maximum temperature difference of the battery does not exceed the temperature difference threshold, or if the real-time temperature rise rate does not exceed the set threshold, the system will use the first PID control mode to control the output of a continuous control quantity to regulate the flow of the liquid cooling system.
8. The BMS fuzzy PID adaptive liquid cooling control method according to claim 1, characterized in that, When the predicted temperature rise rate, average temperature, net heat storage, and maximum temperature meet preset conditions or states, liquid cooling pre-actions are performed according to the pre-start strategy based on the predicted thermal parameters, including: Level 1 Prediction and Pre-start: If the predicted temperature rise rate is increasing in the next 3 seconds and the predicted net heat storage is accumulating, the liquid cooling pump will be started in advance to run at a preset low speed in standby mode. Secondary advance frequency increase: The predicted average temperature in the next 5 seconds is close to the average temperature threshold, and the predicted maximum temperature difference shows an increasing trend. The PID parameters are corrected in advance to increase the liquid cooling flow rate. Level 3 Emergency Enhancement: If the predicted average temperature for the next 8 seconds exceeds the average temperature threshold and the rate of temperature rise exceeds the maximum rate of temperature rise threshold, the throttle valve will be fully opened and the liquid cooling pump will be increased to maximum power operation.
9. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the BMS fuzzy PID adaptive liquid cooling control method as described in any one of claims 1 to 8.
10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the BMS fuzzy PID adaptive liquid cooling control method as described in any one of claims 1 to 8.
