A method and apparatus for temperature equalization control of a power battery using a battery management system (BMS).

Abstract

The present application relates to the field of BMS control technology, and discloses a method and device for controlling temperature balance of a power battery BMS, which includes: collecting temperature data and heat loss power of N battery cells in a power battery pack and analyzing thermodynamic characteristics to obtain dynamic characteristic parameters of battery pack temperature and predicted trajectory of temperature field distribution; estimating and compensating thermal coupling interference and temperature measurement noise between battery cells to obtain actual temperature data of each battery cell; calculating the mean value of battery pack temperature for the actual temperature data of each battery cell, generating a temperature balance target trajectory, and obtaining the temperature balance control quantity of each battery cell through multi-objective optimization solution; calculating the target heat dissipation power of each battery cell, and outputting control instructions for each battery cell radiator. The present application realizes the reasonable allocation of heat dissipation resources and improves heat dissipation efficiency and temperature balance effect through real-time adjustment of the heat dissipation power correction coefficient.

Description

Technical Field

[0001] The present application relates to the field of BMS control technology, and in particular to a method and device for controlling temperature balancing of a power battery BMS. Background Art

[0002] With the rapid development of the new energy vehicle industry, the safety and service life of power battery packs as core components are increasingly concerned. Due to the differences in the working conditions and environmental conditions of each single cell in the battery pack, it is easy to cause uneven temperature distribution, which in turn causes problems such as battery performance degradation and shortened life. Traditional temperature balance control methods are mainly based on switch control strategies, which cannot effectively deal with the thermal coupling effect and dynamic characteristics of the temperature field between battery cells, resulting in control accuracy and response speed that are difficult to meet actual needs.

[0003] At present, although some temperature balance control methods have introduced model predictive control strategies, the influence of time-varying transmission delay is often ignored in the temperature field modeling process, and the interference caused by temperature measurement noise and system uncertainty is not considered. This leads to insufficient accuracy of the temperature field model, and the control effect is easily affected by external environmental changes and measurement errors, and the stability and robustness of temperature balance control cannot be guaranteed. Summary of the invention

[0004] The present application provides a method and device for controlling temperature balancing of a power battery BMS. The present application achieves reasonable allocation of heat dissipation resources and improves heat dissipation efficiency and temperature balancing effect through real-time adjustment of the heat dissipation power correction coefficient.

[0005] In a first aspect, the present application provides a method for controlling temperature balancing of a power battery BMS, the method comprising:

[0006] The temperature data and heat loss power of N battery cells in the power battery pack are collected and the thermodynamic characteristics are analyzed to obtain the dynamic characteristic parameters of the battery pack temperature and the predicted trajectory of the temperature field distribution;

[0007] Based on the battery pack temperature dynamic characteristic parameters and the temperature field distribution prediction trajectory, thermal coupling interference and temperature measurement noise between battery cells are estimated and compensated to obtain actual temperature data of each battery cell;

[0008] Calculating the battery pack temperature average based on the actual temperature data of each battery cell, generating a temperature balance target trajectory taking into account the thermal coupling effect, and obtaining the temperature balance control amount of each battery cell through multi-objective optimization solution;

[0009] The target heat dissipation power of each battery cell is calculated according to the temperature balance control amount, the heat dissipation power is dynamically allocated in combination with the equivalent thermal resistance of each battery cell, and the control instructions of the radiators of each battery cell are output.

[0010] A second aspect of the present application provides a power battery BMS control temperature equalization device, the power battery BMS control temperature equalization device comprising:

[0011] The acquisition module is used to collect the temperature data and heat loss power of N battery cells in the power battery pack and analyze the thermodynamic characteristics to obtain the dynamic characteristic parameters of the battery pack temperature and the predicted trajectory of the temperature field distribution;

[0012] A compensation module, used to estimate and compensate for thermal coupling interference and temperature measurement noise between battery cells based on the battery pack temperature dynamic characteristic parameters and the temperature field distribution prediction trajectory, so as to obtain actual temperature data of each battery cell;

[0013] A solution module, used to calculate the battery pack temperature average based on the actual temperature data of each battery cell, generate a temperature balance target trajectory considering the thermal coupling effect, and obtain the temperature balance control amount of each battery cell through multi-objective optimization solution;

[0014] The output module is used to calculate the target heat dissipation power of each battery cell according to the temperature balance control amount, dynamically allocate the heat dissipation power in combination with the equivalent thermal resistance of each battery cell, and output the control instructions of each battery cell radiator.

[0015] Compared with the prior art, the present application has the following beneficial effects: by establishing a mathematical model of the temperature field including a time-varying transmission delay term and introducing a stochastic differential equation to describe the thermal coupling effect, the accuracy of temperature field modeling is improved, so that the control system can more accurately predict and respond to changes in the temperature field; the temperature field is decomposed into components by using the positive and negative sequence decomposition method, and the Kalman filter algorithm is combined to achieve accurate prediction of the temperature field, effectively reducing the temperature prediction error and improving the predictive performance of the control system; a temperature compensation mechanism based on an interference observer is designed, and through Q-type filters and generalized interference variable modeling, effective suppression of thermal coupling interference and measurement noise is achieved, enhancing The anti-interference ability of the control system is improved; an adaptive droop coefficient generation mechanism is constructed, and the adaptive update of the temperature balance target is realized through the dynamic adjustment of the temperature distribution characteristic parameters and the heat dissipation characteristic parameters, thereby improving the control system's adaptability to temperature changes; a multi-objective optimization method is used for temperature balance control, and through the coordinated optimization of temperature uniform component tracking and non-uniform component suppression, precise control of local temperature differences is achieved while ensuring overall temperature stability; a dynamic heat dissipation power allocation strategy based on equivalent thermal resistance is designed, and through real-time adjustment of the heat dissipation power correction coefficient, a reasonable allocation of heat dissipation resources is achieved, thereby improving heat dissipation efficiency and temperature balance effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0017] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0018] Figure 1 It is a flow chart of a method for controlling temperature balancing of a power battery BMS provided in an embodiment of the present invention;

[0019] Figure 2 It is a schematic block diagram of the structure of a power battery BMS control temperature equalization device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0022] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0023] It should be further understood that the term "and / or" used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. Figure 1In the embodiment of the present application, an embodiment of the method for controlling temperature balancing of a power battery BMS includes:

[0024] Step 100: collecting temperature data and heat loss power of N battery cells in the power battery pack and performing thermodynamic characteristic analysis to obtain the battery pack temperature dynamic characteristic parameters and temperature field distribution prediction trajectory;

[0025] It is understandable that the execution subject of the present application may be a power battery BMS control temperature equalization device, or a terminal or a server, which is not limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0026] Specifically, according to the preset collection cycle, temperature data and heat loss power are collected for all battery cells in the battery pack to obtain the initial data of the temperature distribution of the battery pack. The heat transfer characteristics between adjacent battery cells are considered. Since the heat transfer inside the battery pack has a certain time lag effect, the heat transfer delay between adjacent battery cells in the power battery pack is detected, and the heat transfer delay is introduced into the stochastic differential equation as a time-varying transmission delay term to establish the initial temperature field mathematical model. The mathematical model is expressed by the heat conduction equation, that is, based on Fourier's heat conduction law, the heat transfer process is described as a dynamic system affected by time-varying transmission delay, ensuring that the model reflects the actual thermal coupling relationship inside the battery pack. In order to improve this model, the collected initial data of the battery pack temperature distribution are processed, and these data are divided into data segments of multiple time intervals, and the least squares method is used to fit and analyze the data of each time interval to obtain the thermal conduction coefficient and control gain coefficient between battery cells. The thermal conduction coefficient is used to describe the thermal coupling strength between battery cells, while the control gain coefficient determines the response rate of the control strategy to temperature changes during the temperature balance control process. At the same time, the ambient temperature of the battery pack and the active heat dissipation power of each battery cell are collected in real time to accurately describe the heat dissipation process in subsequent calculations. The ambient temperature has an important influence on the thermal balance of the battery, and the active heat dissipation power is an important means for the BMS system to control the temperature. These two factors are monitored in real time. After obtaining the ambient temperature and heat dissipation power data, the equivalent thermal resistance of each battery cell is calculated according to the ratio of the temperature difference to the total heat dissipation power. The equivalent thermal resistance is a key parameter reflecting the heat dissipation capacity of the battery cell. It is affected by factors such as battery materials, structures, and heat dissipation methods. Therefore, it is dynamically calculated in combination with real-time data to ensure the accuracy of the subsequent temperature control strategy. The temperature field state equation is established based on the thermal conductivity coefficient, control gain coefficient, and equivalent thermal resistance, and the equation is time-discretized to obtain a discretized temperature field model. The temperature field state equation is a set of differential equations that describe the temperature change of the battery cell, taking into account factors such as heat conduction, heat dissipation, and ambient temperature. The time discretization process converts the continuous system into a discrete system suitable for digital control, so that the BMS can complete temperature prediction and control quantity calculation within a limited computing time. The discretized temperature field model is combined with the equivalent thermal resistance and time-varying transmission delay terms to obtain the temperature dynamic characteristic parameters of the power battery pack. Based on the temperature dynamic characteristic parameters of the battery pack, the temperature distribution of the power battery pack in the next H control cycles is predicted by using the state space method, Kalman filter or machine learning algorithm to obtain the temperature field distribution prediction trajectory.

[0027] The positive and negative sequence decomposition matrix is ​​constructed using the temperature dynamic characteristic parameters of the battery pack, and the temperature field of the battery pack is decomposed into positive and negative sequence components to obtain the uniform temperature component and the non-uniform temperature component. Among them, the uniform temperature component represents the average temperature change trend of the battery pack as a whole, while the non-uniform temperature component describes the temperature deviation between battery cells, thereby effectively separating the global effect and local difference in temperature distribution. The state observation equation is constructed according to the uniform temperature component and the non-uniform temperature component, and the state transfer matrix is ​​set in combination with the discretized temperature field model to establish the temperature field state observation model. The state observation equation is used to describe the dynamic process of temperature change. Among them, the state equation of the uniform temperature component mainly focuses on the evolution of the global temperature, while the state equation of the non-uniform temperature component focuses on analyzing the change pattern of the local temperature difference. The state transfer matrix is ​​used to characterize the relationship between the evolution of temperature over time, including factors such as heat conduction, heat dissipation, and environmental influence, and can ensure the continuity and accuracy of the prediction model on the time axis. The temperature field state observation model is input into the Kalman filter to configure the state noise covariance matrix and the observation noise covariance matrix using the temperature measurement data sequence to optimize the filtering parameters. The Kalman filter effectively suppresses the influence of measurement noise and improves the accuracy of temperature prediction by recursively calculating the optimal state estimate. The state noise covariance matrix and the observation noise covariance matrix are initialized to reasonably describe the temperature measurement error and model uncertainty. The filter parameters are continuously adjusted based on the historical temperature measurement data so that the filter can maintain a stable temperature estimation capability under noise interference. Based on the filter parameters and the temperature dynamic characteristic parameters of the battery pack, the optimal prediction value of the temperature field state is calculated through alternating iterative operations of time update and measurement update. In the time update step, the predicted value of the temperature is calculated according to the state transfer matrix, and the temperature distribution of the next time step is estimated using the prediction equation of the Kalman filter; in the measurement update step, the optimal state correction value is calculated in combination with the actual measured temperature data to ensure the accuracy of the prediction result. Through the iterative calculation method, while ensuring the calculation efficiency, a more accurate temperature prediction result is obtained, and the real temperature field evolution trend of the battery pack is gradually approached. The optimal prediction value of the temperature field state is iteratively recursively calculated according to the positive and negative sequence components, and the temperature evolution of the next H control cycles is forward recursively calculated based on the discretized temperature field model. Specifically, the positive-sequence temperature component is recursively extrapolated to predict the change trend of the overall temperature of the battery pack and analyze the influence of factors such as ambient temperature and heat dissipation control; the negative-sequence temperature component is recursively extrapolated to predict how the temperature difference between battery cells evolves and calculate the rate of change of the temperature difference between battery cells. By synthesizing the predicted data of the positive and negative-sequence temperature components, the predicted trajectory of the temperature field distribution is finally obtained.

[0028] Step 200: Based on the battery pack temperature dynamic characteristic parameters and the temperature field distribution prediction trajectory, estimate and compensate for the thermal coupling interference and temperature measurement noise between battery cells to obtain the actual temperature data of each battery cell;

[0029] Specifically, the disturbance observer state equation is constructed based on the temperature dynamic characteristic parameters of the battery pack, and the temperature field distribution prediction trajectory is set as the output reference value of the observer to obtain the disturbance observer system model. The disturbance observer is used to estimate the temperature deviation caused by factors such as the environment, the thermal coupling effect of adjacent battery cells, and the measurement error. By constructing a reasonable state equation, the observer can effectively track the temperature evolution and provide accurate disturbance estimation. A Q-type filter is designed for the disturbance observer system model to ensure that the measurement noise does not introduce additional errors during the compensation process. The Q-type filter determines the cutoff frequency of the filter according to the temperature measurement noise characteristics, thereby maximally suppressing the influence of high-frequency noise while retaining valuable low-frequency temperature change information. The filter design combines the characteristics of the battery temperature sensor and is dynamically adjusted according to factors such as the sampling period and the rate of environmental change, so that it can adapt to the noise interference under different working conditions. The interference signal filter is combined with the dynamic characteristic parameters of the battery pack temperature to construct a generalized interference variable to combine the uncertainty and external interference of the system and obtain the interference model. The interference model includes the influence of thermal coupling effect on temperature measurement, as well as temperature drift caused by battery aging, charge and discharge rate changes, and environmental factors. By constructing a generalized disturbance variable, the source of the temperature measurement error is described and effectively compensated in subsequent calculations. After the disturbance model is constructed, the deviation between the predicted trajectory of the temperature field distribution and the measured temperature data is input into the disturbance model, and the disturbance compensation is obtained through online iterative calculation. The core of online iterative calculation is to continuously optimize the disturbance estimation using the real-time updated measurement data and ensure that the calculation results can converge stably under different working conditions. The calculation method of the disturbance compensation adopts adaptive filtering or optimal estimation algorithm, so that it can be dynamically adjusted between different battery cells, thereby ensuring the improvement of compensation accuracy. Feedback correction is performed on the disturbance compensation to improve the compensation accuracy. The correction gain matrix is ​​designed according to the output error of the observer state equation, and the temperature correction value is calculated by the matrix. The design of the correction gain matrix refers to the adaptive control theory, and by optimizing the feedback coefficient, it can maintain good convergence and robustness under different working conditions. After the temperature correction value is calculated, it is compensated with the measured temperature data to obtain the actual temperature data of each battery cell. The compensation operation adopts the weighted correction method, so that the corrected temperature data can more truly reflect the actual thermal state of the battery cell and effectively reduce the error caused by measurement noise and thermal coupling effect.

[0030] Step 300, calculate the battery pack temperature average based on the actual temperature data of each battery cell, generate a temperature balance target trajectory considering the thermal coupling effect, and obtain the temperature balance control value of each battery cell through multi-objective optimization solution;

[0031] It should be noted that the actual temperature data of each battery cell is arithmetic averaged to calculate the temperature mean of the battery pack, and the standard deviation of the deviation between the temperature of each battery cell and the temperature mean is calculated to obtain the characteristic parameters of the temperature distribution. By calculating the characteristic parameters of the temperature distribution, the temperature distribution of the battery pack is described, including the uniformity and dispersion of the temperature. The temperature distribution characteristic parameters are weightedly combined with the power loss data of each battery cell to calculate the droop reference coefficient of each battery cell. By adjusting the reference value of temperature balance in combination with the power loss of the battery, the heat load of different batteries can be more accurately considered in the subsequent temperature adjustment process. The weighted combination method can reflect the importance of battery cells in heat loss and temperature balance, and then obtain the reference droop coefficient of each battery cell, which is used as the starting reference value for adjusting the temperature. Based on the actual temperature data of each battery cell, the temperature gradient between adjacent battery cells is calculated, and the heat dissipation capacity coefficient of each battery cell is evaluated in combination with the equivalent thermal resistance. The calculation of the temperature gradient helps to understand the thermal coupling effect between battery cells, while the equivalent thermal resistance reflects the resistance of the heat dissipation path. Through the combination of the two, the heat dissipation characteristic parameters of each battery cell are obtained. The heat dissipation characteristic parameters are combined with the reference droop coefficient to construct an exponential decay function to achieve dynamic adjustment and calculation of the droop coefficient. The exponential decay function adaptively adjusts the droop coefficient according to the change of battery pack temperature and heat dissipation characteristics, so that the thermal load change of the battery cell can be responded to in real time during the temperature equalization process. Through this dynamic adjustment, it can flexibly adapt to the changes in different battery working conditions and environmental conditions, thereby effectively improving the accuracy and response speed of temperature equalization. The mean temperature of the battery pack is corrected based on the adaptive droop coefficient, and the target temperature value of each battery cell is generated in combination with the real-time power data of each battery cell to obtain the temperature compensation sequence. The temperature compensation sequence is expanded according to the preset control cycle, and the trajectory is smoothed in combination with the temperature distribution characteristic parameters, so that the change of the temperature compensation sequence is more stable, avoiding overheating or overcooling of the battery pack caused by too fast temperature changes, and obtaining the temperature balance target trajectory. The temperature field distribution prediction trajectory and the temperature balance target trajectory are input into the model predictive controller to construct a multi-objective optimization function containing temperature uniformity constraints. The optimization function simultaneously considers multiple factors such as temperature uniformity, heat dissipation requirements and power loss to achieve optimal temperature balance control. When solving the optimization function, the controller continuously adjusts the temperature control amount of each battery cell according to the error between the temperature prediction trajectory and the target trajectory, and finally obtains the temperature balance control amount.

[0032] The trajectory deviation between the temperature field distribution prediction trajectory and the temperature equilibrium target trajectory is calculated to analyze the degree of deviation of the overall temperature distribution of the battery pack, and a temperature uniformity component tracking sub-objective function is constructed to ensure that the equilibrium trend of the temperature field distribution can be reasonably constrained during the optimization process. By constraining the temperature uniformity component in the temperature field distribution prediction trajectory, the first optimization sub-objective is obtained, which is used to minimize the error between the temperature uniformity component and the temperature equilibrium target. On the basis of constructing the first optimization sub-objective, a temperature uniformity sub-objective function is constructed based on the temperature non-uniform component in the temperature field distribution prediction trajectory, and the temperature non-uniform component is penalized to reduce the temperature difference between battery cells. Define a reasonable temperature uniformity index and apply appropriate penalty terms during the optimization process to avoid local overheating or cooling problems caused by excessive temperature deviation. By constructing the second optimization sub-objective, the variance of the temperature distribution is effectively compressed during the optimization calculation process, so that the temperature between battery cells is more uniform, and the overall thermal stability of the battery pack is improved. While optimizing the temperature balance, the change in the heat dissipation power of the battery cell is limited to avoid excessive power fluctuations during the temperature control process, and the control amount change sub-objective function is constructed in combination with the power change rate constraint to obtain the third optimization sub-objective. Since the heat dissipation regulation of the battery management system is limited by factors such as the radiator response speed, battery thermal capacity, and environmental conditions, the change rate of the heat dissipation power is ensured not to exceed the physically allowed range during the optimization process. By constructing the control amount change sub-objective function, the drastic fluctuation of the heat dissipation power is effectively reduced, the temperature balance control process is made more stable, and the control oscillation phenomenon caused by too fast adjustment is reduced. The first optimization sub-objective, the second optimization sub-objective, and the third optimization sub-objective are weightedly combined to construct a multi-objective optimization function and form a temperature balance optimization model. The upper and lower limits of the heat dissipation power and the power change rate constraints are imposed on the temperature balance optimization model to ensure that the optimization results meet the physical limitations of the battery system. In terms of the upper and lower limits of the heat dissipation power, the power range is set according to the maximum heat dissipation capacity of the battery cooling system, and the results of the optimization calculation are ensured not to exceed the actual achievable heat dissipation power range. In terms of power change rate constraints, appropriate dynamic change limits are set to avoid excessive temperature fluctuations inside the battery pack due to too fast heat dissipation power adjustment, thereby improving the stability of temperature balance control. On the basis of satisfying all constraints, the quadratic programming algorithm is used to solve the temperature balance optimization model to obtain the temperature balance control value of each battery cell.

[0033] Step 400 , calculate the target heat dissipation power of each battery cell according to the temperature balance control amount, dynamically allocate the heat dissipation power in combination with the equivalent thermal resistance of each battery cell, and output the control instructions of each battery cell radiator.

[0034] Specifically, the temperature balance control amount is multiplied by the heat dissipation efficiency coefficient of each battery cell to calculate the target heat dissipation power of each battery cell. The heat dissipation efficiency coefficient is a parameter that measures the heat release capacity of the battery cell under different working conditions. It is affected by factors such as battery materials, cooling methods, and ambient temperature. When calculating the target heat dissipation power, this coefficient needs to be fully considered so that the calculation result accurately reflects the actual heat dissipation requirements of the battery cell. The equivalent thermal resistance of each battery cell is calculated by ratio with the preset reference thermal resistance to reflect the deviation of the heat dissipation capacity of the current battery from the ideal reference value. If the ratio is large, it means that the thermal resistance of the battery is high and the heat dissipation capacity is relatively poor; conversely, if the ratio is small, it means that the heat dissipation of the battery is relatively sufficient. After calculating the ratio, the ratio is exponentially calculated to obtain the heat dissipation power correction coefficient. The role of the exponential calculation is to enhance the influence of the ratio on the heat dissipation power distribution, so that the battery with high thermal resistance obtains higher heat dissipation power, while the battery with low thermal resistance is allocated less heat dissipation power. Based on the target heat dissipation power and the correction coefficient, a correction calculation is performed to obtain the corrected heat dissipation power of each battery cell. The process of correction calculation is to dynamically adjust the target heat dissipation power to better match the heat dissipation capacity and actual heat dissipation requirements of the battery. The corrected heat dissipation power is judged to be zero value to determine whether the radiator needs to be activated, and the radiator duty cycle sequence is generated accordingly. The duty cycle sequence is used to characterize the working state of each battery cell radiator. For example, when the corrected heat dissipation power of a battery is lower than a certain threshold, the corresponding radiator does not need to be started, thereby reducing unnecessary energy consumption and improving the overall efficiency of the heat dissipation system. After generating the radiator duty cycle sequence, the radiators of each battery cell are grouped and controlled according to the sequence, and a phase-interleaved pulse width modulation (PWM) waveform is constructed to achieve precise control of the radiator. The phase-interleaved PWM waveform can effectively reduce the switching loss of the radiator and reduce the current pulsation when the heat dissipation system is running, making the operation of the heat dissipation system more stable. By grouping the radiators of different battery cells, the working mode of the radiator is optimized so that it can reduce the impact on the overall power consumption of the battery system while ensuring the heat dissipation capacity. While constructing the PWM waveform, the radiator control signal is encoded and processed according to the preset communication protocol to ensure that the signal can be accurately transmitted to each radiator. The encoded radiator control signal is output to the radiator of each battery cell, completing the entire process of heat dissipation regulation.

[0035] In the embodiment of the present application, by establishing a mathematical model of the temperature field including a time-varying transmission delay term and introducing a stochastic differential equation to describe the thermal coupling effect, the accuracy of temperature field modeling is improved, so that the control system can more accurately predict and respond to changes in the temperature field; the temperature field is decomposed into components by using a positive and negative sequence decomposition method, and the temperature field is accurately predicted in combination with the Kalman filter algorithm, which effectively reduces the temperature prediction error and improves the predictive performance of the control system; a temperature compensation mechanism based on an interference observer is designed, and effective suppression of thermal coupling interference and measurement noise is achieved through Q-type filters and generalized interference variable modeling, thereby enhancing the anti-interference ability of the control system; an adaptive droop coefficient generation mechanism is constructed, and adaptive update of the temperature balance target is achieved through dynamic adjustment of temperature distribution characteristic parameters and heat dissipation characteristic parameters, thereby improving the control system's ability to adapt to temperature changes; a multi-objective optimization method is used for temperature balance control, and through coordinated optimization of temperature uniform component tracking and non-uniform component suppression, precise control of local temperature differences is achieved while ensuring overall temperature stability; a dynamic heat dissipation power allocation strategy based on equivalent thermal resistance is designed, and through real-time adjustment of the heat dissipation power correction coefficient, reasonable allocation of heat dissipation resources is achieved, thereby improving heat dissipation efficiency and temperature balance effect.

[0036] In a specific embodiment, the process of executing step 100 may specifically include the following steps:

[0037] According to the preset collection cycle, the temperature data and heat loss power of N battery cells in the power battery pack are collected to obtain the initial data of the battery pack temperature distribution;

[0038] Detect the heat transfer delay between adjacent battery cells in the power battery pack, introduce the heat transfer delay as a time-varying transmission delay term into the stochastic differential equation, and obtain the mathematical model of the initial temperature field;

[0039] The initial data of the battery pack temperature distribution is divided into data segments of multiple time intervals, and each data segment is fitted and analyzed by the least square method to obtain the thermal conductivity coefficient and control gain coefficient between battery cells;

[0040] The ambient temperature of the battery pack and the active heat dissipation power of each battery cell are collected in real time, and the equivalent thermal resistance of each battery cell is calculated based on the ratio of the temperature difference to the total heat dissipation power;

[0041] The temperature field state equation is established based on the heat transfer coefficient and the control gain coefficient, and the temperature field state equation is time discretized to obtain a discretized temperature field model;

[0042] The discretized temperature field model is combined with the equivalent thermal resistance and time-varying transmission delay terms to obtain the dynamic characteristic parameters of the battery pack temperature.

[0043] Based on the dynamic characteristic parameters of the battery pack temperature, the temperature distribution of the power battery pack in the future H control cycles is predicted to obtain the temperature field distribution prediction trajectory.

[0044] Specifically, the battery pack is collected according to the preset collection cycle. The temperature data of each battery cell is collected and its heat loss power is measured to obtain the initial temperature distribution data of the battery pack. The heat loss power is measured by a high-precision temperature sensor. It is mainly calculated by Joule heat loss, that is:

[0045]

[0046] in, It's the battery At the moment The current, is its equivalent internal resistance. Since the battery cells in the power battery pack not only generate heat themselves, but also exchange heat with adjacent batteries through heat conduction, the heat transfer delay between adjacent battery cells is detected. Depends on the distance between battery cells, thermal conductivity and heat capacity and , expressed as:

[0047]

[0048] in, Represents battery With battery The thermal conductivity between and is the specific heat capacity of the battery cell. In order to more accurately describe the temperature change process, the heat transfer delay is introduced into the stochastic differential equation as a time-varying transmission delay term, and the mathematical model of the initial temperature field is established. The mathematical model is expressed by the following partial differential equation:

[0049]

[0050] in, Representatives and batteries A collection of adjacent battery cells, Represents external environmental disturbances and measurement errors. After establishing the initial temperature field mathematical model, the initial data of the battery pack temperature distribution is divided into multiple time intervals, and the least squares method is used to fit and analyze the data of each time interval to obtain the thermal conductivity coefficient between battery cells. and control gain coefficient This process is solved by constructing the following optimization problem:

[0051]

[0052] in, is the sampling time interval, is the historical data window length, Represents the gain coefficient of temperature control and determines the response speed of temperature regulation. These parameters are estimated by the least squares method to accurately characterize the thermal characteristics of the battery cell. At the same time, the temperature of the environment in which the battery pack is located is And the heat dissipation power of each battery cell Perform real-time measurements and calculate the equivalent thermal resistance of each battery cell based on the ratio of temperature difference to heat dissipation power , which is calculated as follows:

[0053]

[0054] in, Represents battery cells The temperature difference with the external environment, Represents the heat dissipation power of the battery cell. Equivalent thermal resistance The calculation helps to optimize the temperature control strategy and make the heat dissipation adjustment more accurate. The temperature field state equation is established based on the thermal conductivity, control gain coefficient and equivalent thermal resistance, and the time discretization is performed to obtain the discretized temperature field model. The discretized state equation is:

[0055]

[0056] This discretization process enables the BMS system to digitally calculate the temperature evolution trend, thereby improving the accuracy and response speed of temperature control. The discretized temperature field model is combined with the equivalent thermal resistance and time-varying transmission delay terms to obtain the dynamic characteristic parameters of the battery pack temperature, including the thermal conductivity matrix , equivalent thermal resistance vector And the temperature state transfer matrix Based on these parameters, the state space method is used to predict the future performance of the battery pack. The temperature distribution within a control cycle, the prediction model is as follows:

[0057]

[0058] in, represents the predicted temperature state vector, is the heat input matrix, is the heat dissipation control matrix. The temperature state is iteratively updated through the Kalman filter method to obtain the future temperature evolution trajectory and provide data support for temperature balance control.

[0059] In a specific embodiment, the execution step predicts the temperature distribution of the power battery pack in the future H control cycles based on the battery pack temperature dynamic characteristic parameters, and the process of obtaining the temperature field distribution prediction trajectory may specifically include the following steps:

[0060] Based on the temperature dynamic characteristic parameters of the battery pack, a positive and negative sequence decomposition matrix is ​​constructed to decompose the positive and negative sequence components of the battery pack temperature field to obtain the uniform temperature component and the non-uniform temperature component.

[0061] The state observation equations are constructed according to the temperature uniform component and the temperature non-uniform component respectively, and the state transfer matrix is ​​set according to the discretized temperature field model to obtain the temperature field state observation model;

[0062] The temperature field state observation model is input into the Kalman filter, and the state noise covariance matrix and the observation noise covariance matrix are configured according to the temperature measurement data sequence to obtain the filtering parameters;

[0063] Based on the filtering parameters and the dynamic characteristic parameters of the battery pack temperature, the optimal prediction value of the temperature field state is obtained through alternating iterative calculations of time update and measurement update;

[0064] The optimal prediction value of the temperature field state is iteratively recursively calculated according to the positive and negative sequence components. The temperature evolution of the next H control cycles is forward recursively calculated based on the discretized temperature field model to obtain the prediction data of the positive and negative sequence temperature components. The prediction data of the positive and negative sequence temperature components are synthesized to obtain the prediction trajectory of the temperature field distribution.

[0065] Specifically, a positive and negative sequence decomposition matrix is ​​constructed based on the dynamic characteristic parameters of the battery pack temperature, so as to decompose the positive and negative sequence components of the temperature field of the battery pack to obtain the uniform temperature component and the non-uniform temperature component. Inside the battery pack, the change of the temperature field is not only affected by the heating characteristics of the individual batteries, but also by the heat dissipation environment and the thermal coupling of adjacent battery cells. Therefore, the temperature distribution is represented as two main components, namely the uniform component and the non-uniform component. The uniform component describes the overall temperature level of the entire battery pack, while the non-uniform component represents the temperature deviation between each battery cell. Using the positive and negative sequence decomposition matrix Perform a linear transformation to transform the original temperature state vector Decompose into uniform components and uneven components :

[0066]

[0067] in, is a matrix constructed from the dynamic characteristic parameters of the battery pack temperature, so that the uniform component Represents the average temperature of the battery pack, while the non-uniform component It is used to characterize the temperature difference between each battery cell. The state observation equation is constructed according to the temperature uniform component and the temperature non-uniform component, and the state transfer matrix is ​​set based on the discretized temperature field model to obtain the temperature field state observation model. Consider the state equation of the discretized temperature field:

[0068]

[0069] in, is the temperature state transfer matrix, which describes the change of battery pack temperature over time. Represents the heat input matrix, including the heat loss of the battery itself , It is the heat dissipation control matrix, which determines the active heat dissipation The effect of temperature on is external noise interference. Since the temperature uniformity component and the temperature non-uniformity component are affected by different physical mechanisms, the state transfer equation is further decomposed into:

[0070]

[0071]

[0072] in, and Describe the evolution laws of the temperature uniform component and the temperature non-uniform component respectively. and represents the effect of heat input on different components, and The temperature field state observation model is input into the Kalman filter to use the temperature measurement data sequence to calculate the state noise covariance matrix. and the observation noise covariance matrix Perform optimization configuration to obtain the filtering parameters. The state update equation of the Kalman filter is as follows:

[0073]

[0074]

[0075] in, is the predicted value of the temperature state, is the forecast error covariance matrix, Represents the system noise covariance. Update the state estimate using the measured data:

[0076]

[0077]

[0078] in, is the Kalman gain, To measure temperature data, is the observation matrix. Through iterative updating, the Kalman filter provides the optimal prediction value of the temperature field state. After obtaining the optimal prediction value, iterative recursive calculation is performed according to the positive and negative sequence components, and the future is predicted based on the discretized temperature field model. The temperature evolution of the control cycle is forward recursive. and uneven components , respectively, make predictions:

[0079]

[0080]

[0081] By synthesizing the predicted data of the positive and negative sequence temperature components, the complete temperature field distribution prediction trajectory is obtained:

[0082]

[0083] This embodiment also includes: using a distributed double-layer optimization method to collaboratively optimize and correct the temperature dynamic characteristic parameters of the battery pack to obtain optimized temperature field model parameters, including: dividing the battery pack into a number of temperature field sub-areas, and configuring a local optimization agent for each sub-area, establishing a communication topology network based on the thermal coupling relationship between adjacent battery cells, and obtaining a distributed collaborative structure; constructing a double-layer optimization model for each local optimization agent, setting the temperature field model accuracy as an outer layer target, and setting the heat conduction parameter constraint as an inner layer constraint to obtain a hierarchical optimization structure; constructing a temperature field model error function based on the measurement data of the local optimization agent, and designing coupling constraints in combination with the temperature data of adjacent agents to obtain a local optimization target; for the local optimization agent A hybrid optimization algorithm is configured, the first-order gradient method is used for outer target optimization, and the second-order Newton method is used for inner constraint solving to obtain an optimization iteration sequence; gradient information and model parameters are exchanged between adjacent optimization agents through a communication topology network, and consistency constraints are processed on local optimization results to obtain collaborative optimization results; convergence judgment is made on the collaborative optimization results, and the optimization parameters are adaptively adjusted based on the real-time communication status to obtain an optimization mechanism with communication fault tolerance; the optimization mechanism with communication fault tolerance is combined with the temperature field modeling process, and the optimal estimate of the model parameters is obtained through iterative calculation of the hybrid algorithm; based on the optimal estimate of the model parameters, the dynamic characteristic parameters of the battery pack temperature are corrected and calibrated to obtain the optimized temperature field model parameters.

[0084] In a specific embodiment, the process of executing step 200 may specifically include the following steps:

[0085] The disturbance observer state equation is constructed based on the dynamic characteristic parameters of the battery pack temperature, and the temperature field distribution prediction trajectory is set as the output reference value of the observer to obtain the disturbance observer system model;

[0086] Design a Q-type filter for the disturbance observer system model, determine the filter cutoff frequency according to the temperature measurement noise characteristics, and obtain the interference signal filter;

[0087] The interference signal filter is combined with the dynamic characteristic parameters of the battery pack temperature to construct a generalized interference variable, and the system uncertainty and external interference are combined to model the interference model;

[0088] The deviation between the predicted trajectory of temperature field distribution and the measured temperature data is input into the interference model, and the interference compensation amount is obtained through online iterative calculation;

[0089] Feedback correction is performed on the interference compensation amount, and a correction gain matrix is ​​designed according to the output error of the observer state equation to obtain a temperature correction value. The temperature correction value and the measured temperature data are compensated to obtain the actual temperature data of each battery cell.

[0090] Specifically, the disturbance observer state equation is constructed based on the temperature dynamic characteristic parameters of the battery pack, and the temperature field distribution prediction trajectory is set as the output reference value of the observer to obtain the disturbance observer system model. The temperature state of each battery cell is described by the following state equation:

[0091]

[0092] in, It's time Battery Cell The temperature, is the temperature state transfer matrix, which represents the internal transfer relationship of temperature changing with time. Represents the heat input matrix, including the heat generation power inside the battery , For the heat dissipation control matrix, determine the heat dissipation power The effect of temperature, Represents external environmental disturbances, such as ambient temperature fluctuations and measurement noise. Based on the above temperature state equation, a disturbance observer system model is constructed to estimate the temperature deviation due to system uncertainty and measurement error. Define the disturbance state variable To characterize the unmodeled external disturbance, the observer state equation is expressed as:

[0093]

[0094] in, is the interference transfer matrix, describing the time evolution of the interference, and is an unknown noise term. In order to ensure the accuracy of the disturbance observer, the temperature field distribution prediction trajectory is set As the output reference value of the observer, and construct the observation equation:

[0095]

[0096] The disturbance observer system model is The filter is designed, and the cutoff frequency of the filter is determined according to the temperature measurement noise characteristics to obtain the interference signal filter. Considering that the temperature measurement data contains high-frequency noise, a low-pass filter is used to smooth the measurement signal. Assume that the transfer function of the filter is:

[0097]

[0098] in, is the filter cutoff frequency, which is determined by the noise power spectrum density of the temperature measurement system, and s is the Laplace transform variable, which is used to represent the dynamic characteristics of the system in the complex frequency domain (Laplace domain). In order to ensure the filtering effect, Adjustments are made based on the statistical characteristics of the measurement noise so that high-frequency noise is effectively suppressed while retaining the main trend of temperature change. After obtaining the interference signal filter, it is combined with the battery pack temperature dynamic characteristic parameters to construct a generalized interference variable, and the system uncertainty and external interference are combined to model the complete interference model. Generalized interference variable It is determined by the temperature measurement error sensor deviation and external environmental factors, and is expressed as:

[0099]

[0100] At this point, the complete disturbance observation equation is expressed in the following matrix form:

[0101]

[0102] After constructing the interference model, the temperature field distribution prediction trajectory Compared with the measured temperature data The deviation between is input into the interference model, and the interference compensation is obtained through online iterative calculation. Assuming that the temperature measurement error follows a Gaussian distribution, the Kalman filter equation is used to update the interference state estimate :

[0103]

[0104] in, is the Kalman gain matrix, which is used to dynamically adjust the disturbance compensation. In order to improve the compensation accuracy, the disturbance compensation is corrected by feedback, and the gain matrix is ​​designed according to the output error of the observer state equation. , and thus calculate the temperature correction value:

[0105]

[0106] The temperature correction value Compared with the measured temperature data Perform compensation calculation to obtain the actual temperature data of each battery cell:

[0107]

[0108] In a specific embodiment, the process of executing step 300 may specifically include the following steps:

[0109] Perform arithmetic averaging on the actual temperature data of each battery cell, calculate the mean temperature of the battery pack, and calculate the standard deviation of the deviation between the temperature of each battery cell and the mean temperature to obtain the temperature distribution characteristic parameters;

[0110] The temperature distribution characteristic parameters are weightedly combined with the power loss data of each battery cell, and the droop reference coefficient of each battery cell is initially assigned to obtain a reference droop coefficient;

[0111] Based on the actual temperature data of the battery cells, the temperature gradient between adjacent battery cells is calculated, and the heat dissipation capacity coefficient of each battery cell is calculated in combination with the equivalent thermal resistance to obtain the heat dissipation characteristic parameters;

[0112] The heat dissipation characteristic parameters are combined with the reference droop coefficient to construct an exponential decay function, and the droop coefficient is dynamically adjusted and calculated to obtain an adaptive droop coefficient.

[0113] The mean temperature of the battery pack is corrected based on the adaptive droop coefficient, and the target temperature value of each battery cell is generated by combining the real-time power data of each battery cell to obtain a temperature compensation sequence;

[0114] The temperature compensation sequence is expanded according to the preset control cycle, and the trajectory is smoothed in combination with the temperature distribution characteristic parameters to obtain the temperature balance target trajectory;

[0115] The temperature field distribution prediction trajectory and the temperature balance target trajectory are input into the model predictive controller, and a multi-objective optimization function including temperature uniformity constraints is constructed to solve the temperature balance control quantity of each battery cell.

[0116] Specifically, define the temperature set of the battery pack ,in Indicates the battery cell number. Indicates battery cells The temperature, Represents the total number of batteries in the battery pack. The average temperature of the battery pack The calculation formula is as follows:

[0117]

[0118] in, In order to measure the dispersion of the temperature distribution inside the battery pack, the standard deviation is calculated. , which is expressed as follows:

[0119]

[0120] Standard Deviation Reflects the degree of deviation between the battery cell temperature and the average temperature, that is, the uniformity of the temperature field. The larger the standard deviation, the more uneven the temperature distribution and the larger the temperature difference, which causes some batteries to be too high or too low, thus affecting the overall safety and performance of the battery pack. The temperature distribution characteristic parameters are compared with the power loss data of each battery cell. Weighted combination to calculate the droop reference factor for each battery cell . Power loss Calculated by Joule heat formula:

[0121]

[0122] in, For battery cells The working current, Is the equivalent internal resistance of the battery cell. The calculation method is:

[0123]

[0124] in, and is a weight coefficient used to balance the effects of temperature distribution and power loss on the droop coefficient. In order to evaluate the heat dissipation capacity of the battery cell, the temperature gradient between adjacent battery cells is calculated. , defined as follows:

[0125]

[0126] in, Represents adjacent battery cells. Combined with equivalent thermal resistance Calculate the heat dissipation coefficient , the calculation method of equivalent thermal resistance is:

[0127]

[0128] in, represents the ambient temperature, Represents the heat dissipation power. Heat dissipation capacity coefficient The definition is as follows:

[0129]

[0130] The larger the heat dissipation capacity coefficient, the higher the heat dissipation efficiency of the battery cell, and the more effectively the temperature is reduced. The reference droop coefficient and the heat dissipation characteristic parameters are combined, and an exponential decay function is constructed to dynamically adjust and calculate the droop coefficient to obtain an adaptive droop coefficient. :

[0131]

[0132] in, is the adjustment coefficient, which controls the effect of heat dissipation capacity on the droop coefficient. The role of this exponential decay function is to make the droop coefficient of battery cells with higher heat dissipation capacity lower, reducing its adjustment range, while battery cells with lower heat dissipation capacity have a larger droop coefficient to participate more actively in temperature balance control. Based on the adaptive droop coefficient, the mean temperature of the battery pack is corrected, and combined with the real-time power data of each battery cell , generate the target temperature value of each monomer and form a temperature compensation sequence :

[0133]

[0134] The physical meaning of this formula is: for battery cells with large power loss, lower their target temperature to actively dissipate heat, while for battery cells with small power loss, the target temperature is higher to avoid the temperature balance strategy affecting the overall working state of the battery. After obtaining the temperature compensation sequence, expand it according to the preset control cycle and combine it with the temperature distribution characteristic parameters Perform trajectory smoothing to obtain the temperature balanced target trajectory :

[0135]

[0136] in, is a smoothing factor, which is used to suppress the drastic changes in the temperature control trajectory and ensure the stability of temperature regulation. and temperature equilibrium target trajectory Input the model predictive controller to build a multi-objective optimization function including temperature uniformity constraints to solve the temperature balance control quantity of each battery cell :

[0137]

[0138] in, In order to smooth the control parameters and prevent the temperature control amount from drastically changing, the optimal temperature balance control amount of each battery cell is obtained by solving the optimization problem, thereby ensuring that the battery pack is temperature balanced while improving safety and extending life.

[0139] In a specific embodiment, the execution step inputs the temperature field distribution prediction trajectory and the temperature balance target trajectory into the model predictive controller, constructs a multi-objective optimization function including temperature uniformity constraints, and solves the process of obtaining the temperature balance control amount of each battery cell, which may specifically include the following steps:

[0140] The trajectory deviation between the temperature field distribution prediction trajectory and the temperature equilibrium target trajectory is calculated, the temperature uniform component tracking sub-objective function is constructed, the temperature uniform component in the temperature field distribution prediction trajectory is constrained, and the first optimization sub-objective is obtained;

[0141] Based on the temperature non-uniform component in the temperature field distribution prediction trajectory, a temperature uniformity sub-objective function is constructed, and the temperature non-uniform component is penalized to obtain the second optimization sub-objective;

[0142] The change in the heat dissipation power of the battery cell is processed by limiting the constraint, and the control variable change sub-objective function is constructed in combination with the power change rate constraint to obtain the third optimization sub-objective;

[0143] The first optimization sub-goal, the second optimization sub-goal and the third optimization sub-goal are weightedly combined to construct a multi-objective optimization function and obtain a temperature balance optimization model;

[0144] The upper and lower limits of heat dissipation power and power change rate constraints are imposed on the temperature balance optimization model, and the quadratic programming algorithm is used to solve the problem and obtain the temperature balance control value of each battery cell.

[0145] Specifically, the trajectory deviation between the temperature field distribution prediction trajectory and the temperature equilibrium target trajectory is calculated to construct the temperature uniformity component tracking sub-objective function, and the temperature uniformity component in the temperature field distribution prediction trajectory is constrained to obtain the first optimization sub-objective. Suppose the predicted temperature trajectory of the battery cell is , the temperature equilibrium target trajectory is The average temperature of the battery pack is , then the trajectory deviation The definition is as follows:

[0146]

[0147] in, Reflects the battery monomer The goal is to minimize the trajectory deviation of all battery cells to make them close to the temperature balance target. Construct the temperature uniformity component tracking sub-objective function As the first optimization sub-goal:

[0148]

[0149] in, It is used to ensure that the temperature balance target trajectory can effectively track the uniform component of the temperature field distribution prediction trajectory, so that the average temperature of the entire battery pack is kept within a reasonable range. On this basis, the temperature uniformity sub-objective function is constructed based on the temperature non-uniform component in the temperature field distribution prediction trajectory, and the temperature non-uniform component is penalized to obtain the second optimization sub-objective. The temperature non-uniform component is measured by calculating the variance between the cell temperature and the temperature mean, which is defined as follows:

[0150]

[0151] in, Represents the temperature uniformity index of the battery pack. In order to suppress the temperature deviation between battery cells, the second optimization sub-goal is constructed :

[0152]

[0153] The purpose of this objective function is to minimize the temperature non-uniformity of the battery pack and make the temperature of all battery cells as close to the mean as possible, thereby reducing the phenomenon of local overheating or overcooling and improving the overall safety and service life of the battery pack. At the same time, the change in the heat dissipation power of the battery cell is processed by limiting constraints, and combined with the power change rate constraint to construct the control variable change sub-objective function to obtain the third optimization sub-objective. Assume that the battery cell At the moment The heat dissipation power is , and its rate of change is defined as follows:

[0154]

[0155] In order to prevent the drastic change of heat dissipation power from affecting the stability of the system, the power change is limited, and the maximum change rate is set to , then the constraint condition for the change of heat dissipation power is:

[0156]

[0157] On this basis, the third optimization sub-goal is constructed :

[0158]

[0159] in, It is used to minimize the change of heat dissipation power, make temperature control more stable, reduce the control load of the heat dissipation system, and avoid the sudden change of heat dissipation power causing the cooling system to overwork or become unstable. The three optimization sub-objectives are weighted and combined according to appropriate weights to construct a complete multi-objective optimization function and form a temperature balance optimization model. , and are the weight coefficients of the first, second and third optimization sub-objectives respectively, then the final optimization objective function It is expressed as:

[0160]

[0161] in, Ensure that the temperature balance target can effectively track the temperature prediction trajectory, Constrain the temperature uniformity of battery cells, Constrain the change in heat dissipation power so that the optimization calculation can ensure both temperature balance and the stability of the control process. The upper and lower limits of heat dissipation power and the power change rate constraints are imposed on the temperature balance optimization model to ensure the feasibility of the optimization solution. The maximum and minimum values ​​of heat dissipation power are assumed to be and , then the constraints are:

[0162]

[0163] In addition, considering the feasible range of temperature control, the maximum allowable value of temperature adjustment is , then it is necessary to meet:

[0164]

[0165] In order to solve the temperature balance control quantity, the quadratic programming method is used for optimization calculation. The standard form of quadratic programming is as follows:

[0166]

[0167] in, Represents the optimization variable (i.e., heat dissipation power and temperature adjustment), is the coefficient matrix of the quadratic objective function, Represents the linear target vector. By solving the quadratic programming problem, the temperature balance control value of each battery cell is obtained , and ultimately used to regulate the cooling system.

[0168] In a specific embodiment, the process of executing step 400 may specifically include the following steps:

[0169] The temperature balance control amount is multiplied by the heat dissipation efficiency coefficient of each battery cell to obtain the target heat dissipation power of each battery cell;

[0170] Calculate the ratio of the equivalent thermal resistance of each battery cell to the preset reference thermal resistance to obtain a ratio calculation result, and perform exponential operation on the ratio calculation result to obtain a heat dissipation power correction coefficient;

[0171] Based on the target heat dissipation power and the heat dissipation power correction coefficient, a correction calculation is performed to obtain the corrected heat dissipation power of each battery cell, and a zero value judgment is performed on the corrected heat dissipation power to obtain a radiator duty cycle sequence;

[0172] The battery cell radiators are grouped and controlled according to the radiator duty cycle sequence, a phase-interleaved pulse width modulation waveform is constructed, a radiator control signal is obtained, and the radiator control signal is encoded and processed according to a preset communication protocol to output the control instructions for each battery cell radiator.

[0173] Specifically, define the temperature balance control quantity ,in Represents the battery cell number, It is obtained by solving the temperature balance optimization model, which means that in the control cycle The intensity of heat dissipation regulation that should be applied internally. Heat dissipation efficiency coefficient Reflect battery cells The heat dissipation capacity is defined as follows:

[0174]

[0175] in, The equivalent thermal resistance of the battery cell indicates the difficulty of heat dissipation of the battery cell. The larger the thermal resistance, the lower the heat dissipation efficiency. Calculated by the following formula:

[0176]

[0177] The physical meaning of this formula is to calculate the required heat dissipation power based on the heat dissipation efficiency of the battery cell and the temperature balance control amount, so that the heat dissipation adjustment meets the temperature balance requirements. After obtaining the target heat dissipation power, the equivalent thermal resistance of each battery cell is calculated. With the preset reference thermal resistance Perform a ratio calculation to evaluate the degree of deviation of the current thermal resistance from the standard value. Ratio calculation results It is given by the following formula:

[0178]

[0179] This ratio represents the increase or decrease of the thermal resistance of the current battery cell relative to the reference thermal resistance, and is used for subsequent heat dissipation power correction. In order to optimize the heat dissipation control strategy, the ratio calculation result is exponentially calculated to obtain the heat dissipation power correction coefficient. , and its calculation formula is as follows:

[0180]

[0181] in, is the adjustment coefficient, which is used to control the influence of thermal resistance on the adjustment of heat dissipation power. The role of exponential operation is that when Much greater than When Rapidly reduce to reduce the heat dissipation power and prevent excessive heat dissipation of battery cells with large thermal resistance; When it is small, the correction coefficient approaches 1, ensuring the normal heat dissipation control effect. Based on the target heat dissipation power and the heat dissipation power correction coefficient, a correction calculation is performed to obtain the corrected heat dissipation power of each battery cell. :

[0182]

[0183] The purpose of this calculation is to adjust the heat dissipation power according to the change of thermal resistance, so that the radiator works within a reasonable range and avoids heat dissipation imbalance caused by excessively high or low thermal resistance of individual batteries. The corrected heat dissipation power is judged to zero value to determine whether the radiator needs to work, and thus the radiator duty cycle sequence is obtained. :

[0184]

[0185] in, is the maximum power of the radiator, Indicates the duty cycle of the radiator. , the radiator of the battery cell does not work; otherwise, the duty cycle is calculated according to the relative size of the heat dissipation power. The radiators of each battery cell are grouped and controlled according to the radiator duty cycle sequence, and a phase-interleaved pulse width modulation (PWM) waveform is constructed to obtain the radiator control signal. Assume that the control period is , the duty cycle of the PWM signal Determined by the following formula:

[0186]

[0187] In order to optimize the heat dissipation control effect, phase interleaving technology is used between multiple battery cells to stagger the PWM waveforms of different battery cells in time to reduce current fluctuations and system interference. The PWM phase offset of each battery cell is The calculation is as follows:

[0188]

[0189] Ensure that the radiators of different battery cells are turned on at different time points, so as to balance the system load and reduce current pulsation. Encode the radiator control signal according to the preset communication protocol and output the encoded signal to the radiator of each battery cell. Assume that the communication protocol adopts the standard serial data format, and the radiator control signal It consists of the following packets:

[0190] ;

[0191] Among them, the start bit is used to mark the beginning of the data packet; and They represent PWM duty cycle and phase offset respectively; the check bit is used to detect data transmission errors; the end bit marks the end of the data packet. The BMS controller sends the data packet to the cooling control unit through the CAN bus or UART serial port to execute the cooling adjustment instruction.

[0192] The above describes the power battery BMS control temperature equalization method in the embodiment of the present application. The following describes the power battery BMS control temperature equalization device 10 in the embodiment of the present application. Figure 2 In the embodiment of the present application, one embodiment of the power battery BMS control temperature equalization device 10 includes:

[0193] The acquisition module 11 is used to collect the temperature data and heat loss power of N battery cells in the power battery pack and perform thermodynamic characteristic analysis to obtain the battery pack temperature dynamic characteristic parameters and temperature field distribution prediction trajectory;

[0194] The compensation module 12 is used to estimate and compensate the thermal coupling interference and temperature measurement noise between battery cells based on the dynamic characteristic parameters of the battery pack temperature and the temperature field distribution prediction trajectory, and obtain the actual temperature data of each battery cell;

[0195] A solution module 13 is used to calculate the battery pack temperature average based on the actual temperature data of each battery cell, generate a temperature balance target trajectory considering the thermal coupling effect, and obtain the temperature balance control value of each battery cell through multi-objective optimization solution;

[0196] The output module 14 is used to calculate the target heat dissipation power of each battery cell according to the temperature balance control amount, dynamically allocate the heat dissipation power in combination with the equivalent thermal resistance of each battery cell, and output the control instructions of each battery cell radiator.

[0197] Through the synergy of the above components, by establishing a mathematical model of the temperature field including a time-varying transmission delay term and introducing a stochastic differential equation to describe the thermal coupling effect, the accuracy of the temperature field modeling is improved, so that the control system can more accurately predict and respond to changes in the temperature field; the temperature field is decomposed into components by using the positive and negative sequence decomposition method, and the temperature field is accurately predicted in combination with the Kalman filter algorithm, which effectively reduces the temperature prediction error and improves the predictive performance of the control system; a temperature compensation mechanism based on a disturbance observer is designed, and the effective suppression of thermal coupling interference and measurement noise is achieved through Q-type filters and generalized disturbance variable modeling, thereby enhancing the anti-interference ability of the control system; an adaptive droop coefficient generation mechanism is constructed, and the adaptive update of the temperature balance target is achieved through the dynamic adjustment of the temperature distribution characteristic parameters and the heat dissipation characteristic parameters, thereby improving the control system's ability to adapt to temperature changes; a multi-objective optimization method is used for temperature balance control, and the coordinated optimization of temperature uniform component tracking and non-uniform component suppression is used to achieve precise control of local temperature differences while ensuring overall temperature stability; a dynamic heat dissipation power allocation strategy based on equivalent thermal resistance is designed, and the reasonable allocation of heat dissipation resources is achieved through real-time adjustment of the heat dissipation power correction coefficient, thereby improving the heat dissipation efficiency and temperature balance effect.

[0198] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0199] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable an electronic device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0200] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for controlling temperature balancing of a power battery BMS, characterized in that: The method comprises: The temperature data and heat loss power of N battery cells in the power battery pack are collected and the thermodynamic characteristics are analyzed to obtain the dynamic characteristic parameters of the battery pack temperature and the predicted trajectory of the temperature field distribution; specifically, the temperature data and heat loss power of N battery cells in the power battery pack are collected according to a preset collection period to obtain the initial data of the battery pack temperature distribution; the heat transfer delay between adjacent battery cells in the power battery pack is detected, and the heat transfer delay is introduced into the stochastic differential equation as a time-varying transmission delay term to obtain the initial temperature field mathematical model; the initial data of the battery pack temperature distribution is divided into data segments of multiple time intervals, and each data segment is fitted and analyzed by the least squares method to obtain the battery cell temperature distribution prediction trajectory. The thermal conductivity coefficient and control gain coefficient between the bodies are obtained; the ambient temperature of the battery pack and the active heat dissipation power of each battery cell are collected in real time, and the equivalent thermal resistance of each battery cell is calculated according to the ratio of the temperature difference to the total heat dissipation power; a temperature field state equation is established based on the thermal conductivity coefficient and the control gain coefficient, and the temperature field state equation is time discretized to obtain a discretized temperature field model; the discretized temperature field model is combined with the equivalent thermal resistance and the time-varying transmission delay term to obtain the dynamic characteristic parameters of the battery pack temperature; based on the dynamic characteristic parameters of the battery pack temperature, the temperature distribution of the power battery pack in the future H control cycles is predicted to obtain a temperature field distribution prediction trajectory; Based on the battery pack temperature dynamic characteristic parameters and the temperature field distribution prediction trajectory, thermal coupling interference and temperature measurement noise between battery cells are estimated and compensated to obtain actual temperature data of each battery cell; Calculating the battery pack temperature average based on the actual temperature data of each battery cell, generating a temperature balance target trajectory taking into account the thermal coupling effect, and obtaining the temperature balance control amount of each battery cell through multi-objective optimization solution; The target heat dissipation power of each battery cell is calculated according to the temperature balance control amount, the heat dissipation power is dynamically allocated in combination with the equivalent thermal resistance of each battery cell, and the control instructions of the radiators of each battery cell are output.

2. The power battery BMS temperature balancing control method according to claim 1, characterized in that: The method of predicting the temperature distribution of the power battery pack in the future H control cycles based on the battery pack temperature dynamic characteristic parameters to obtain a temperature field distribution prediction trajectory includes: Based on the battery pack temperature dynamic characteristic parameters, a positive-negative sequence decomposition matrix is ​​constructed to decompose the battery pack temperature field into positive-negative sequence components to obtain a uniform temperature component and an inhomogeneous temperature component; Constructing state observation equations according to the uniform temperature component and the non-uniform temperature component respectively, and setting a state transfer matrix according to a discretized temperature field model to obtain a temperature field state observation model; The temperature field state observation model is input into the Kalman filter, and the state noise covariance matrix and the observation noise covariance matrix are configured according to the temperature measurement data sequence to obtain the filtering parameters; Based on the filtering parameters and the battery pack temperature dynamic characteristic parameters, an optimal prediction value of the temperature field state is obtained through alternating iterative calculations of time update and measurement update; The optimal prediction value of the temperature field state is iteratively recursively calculated according to the positive and negative sequence components, and the temperature evolution of the future H control cycles is forward recursively calculated based on the discretized temperature field model to obtain the prediction data of the positive and negative sequence temperature components, and the prediction data of the positive and negative sequence temperature components are synthesized to obtain the temperature field distribution prediction trajectory.

3. The power battery BMS temperature balancing control method according to claim 1, characterized in that: The method estimates and compensates for the thermal coupling interference and temperature measurement noise between battery cells based on the battery pack temperature dynamic characteristic parameters and the temperature field distribution prediction trajectory to obtain actual temperature data of each battery cell, including: Constructing a disturbance observer state equation based on the battery pack temperature dynamic characteristic parameters, setting the temperature field distribution prediction trajectory as an output reference value of the observer, and obtaining a disturbance observer system model; Designing a Q-type filter for the interference observer system model, determining the filter cutoff frequency according to the temperature measurement noise characteristics, and obtaining an interference signal filter; The interference signal filter is combined with the battery pack temperature dynamic characteristic parameter to construct a generalized interference variable, and system uncertainty and external interference are combined for modeling to obtain an interference model; The deviation between the predicted trajectory of the temperature field distribution and the measured temperature data is input into the interference model, and the interference compensation amount is obtained through online iterative calculation; Feedback correction is performed on the interference compensation amount, and a correction gain matrix is ​​designed according to the output error of the observer state equation to obtain a temperature correction value. Compensation operation is performed on the temperature correction value and the measured temperature data to obtain actual temperature data of each battery cell.

4. The power battery BMS control temperature balancing method according to claim 1, characterized in that: The battery pack temperature average calculation is performed on the actual temperature data of each battery cell, a temperature balance target trajectory is generated considering the thermal coupling effect, and the temperature balance control amount of each battery cell is obtained by multi-objective optimization solution, including: Performing arithmetic averaging on the actual temperature data of each battery cell, calculating the mean temperature of the battery pack, and performing standard deviation calculation on the deviation between the temperature of each battery cell and the mean temperature, to obtain temperature distribution characteristic parameters; The temperature distribution characteristic parameters are weightedly combined with the power loss data of each battery cell, and the droop reference coefficient of each battery cell is initially assigned to obtain a reference droop coefficient; Based on the actual temperature data of the battery cells, the temperature gradient between adjacent battery cells is calculated, and the heat dissipation capacity coefficient of each battery cell is calculated in combination with the equivalent thermal resistance to obtain heat dissipation characteristic parameters; The heat dissipation characteristic parameter is combined with the reference droop coefficient to construct an exponential decay function, and the droop coefficient is dynamically adjusted and calculated to obtain an adaptive droop coefficient; The battery pack temperature mean value is corrected based on the adaptive droop coefficient, and a target temperature value of each battery cell is generated in combination with the real-time power data of each battery cell to obtain a temperature compensation sequence; Expanding the temperature compensation sequence according to a preset control period, and performing trajectory smoothing processing in combination with the temperature distribution characteristic parameters to obtain a temperature equilibrium target trajectory; The temperature field distribution prediction trajectory and the temperature balance target trajectory are input into a model predictive controller, a multi-objective optimization function including temperature uniformity constraints is constructed, and the temperature balance control quantity of each battery cell is obtained by solving.

5. The power battery BMS control temperature equalization method according to claim 4, characterized in that: The temperature field distribution prediction trajectory and the temperature balance target trajectory are input into a model predictive controller, a multi-objective optimization function including temperature uniformity constraints is constructed, and the temperature balance control quantity of each battery cell is obtained by solving the problem, including: Calculate the trajectory deviation between the temperature field distribution prediction trajectory and the temperature equilibrium target trajectory, construct a temperature uniform component tracking sub-objective function, perform constraint processing on the temperature uniform component in the temperature field distribution prediction trajectory, and obtain a first optimization sub-objective; Constructing a temperature uniformity sub-objective function based on the temperature non-uniform component in the temperature field distribution prediction trajectory, and performing penalty processing on the temperature non-uniform component to obtain a second optimization sub-objective; The change in the heat dissipation power of the battery cell is processed by limiting the constraint, and the control variable change sub-objective function is constructed in combination with the power change rate constraint to obtain the third optimization sub-objective; The first optimization sub-objective, the second optimization sub-objective and the third optimization sub-objective are weightedly combined to construct a multi-objective optimization function to obtain a temperature balance optimization model; The upper and lower limit constraints of heat dissipation power and the power change rate constraint are imposed on the temperature balance optimization model, and a quadratic programming algorithm is used to perform a solution operation to obtain the temperature balance control amount of each battery cell.

6. The power battery BMS temperature balancing control method according to claim 1, characterized in that: The method of calculating the target heat dissipation power of each battery cell according to the temperature balance control amount, dynamically allocating the heat dissipation power in combination with the equivalent thermal resistance of each battery cell, and outputting the control instructions of each battery cell radiator includes: Performing a product operation on the temperature balance control amount and the heat dissipation efficiency coefficient of each battery cell to obtain a target heat dissipation power of each battery cell; Calculating the ratio of the equivalent thermal resistance of each battery cell to a preset reference thermal resistance to obtain a ratio calculation result, and performing an exponential operation on the ratio calculation result to obtain a heat dissipation power correction coefficient; Performing correction calculation based on the target heat dissipation power and the heat dissipation power correction coefficient to obtain a corrected heat dissipation power of each battery cell, and performing a zero value judgment on the corrected heat dissipation power to obtain a radiator duty cycle sequence; The battery cell radiators are grouped and controlled according to the radiator duty cycle sequence, a phase-interleaved pulse width modulation waveform is constructed to obtain a radiator control signal, and the radiator control signal is encoded and processed according to a preset communication protocol to output a control instruction for each battery cell radiator.

7. A power battery BMS control temperature equalization device, characterized in that: Used to execute the power battery BMS control temperature equalization method according to claim 1, the power battery BMS control temperature equalization device comprises: The acquisition module is used to collect the temperature data and heat loss power of N battery cells in the power battery pack and analyze the thermodynamic characteristics to obtain the dynamic characteristic parameters of the battery pack temperature and the predicted trajectory of the temperature field distribution; A compensation module, used to estimate and compensate for thermal coupling interference and temperature measurement noise between battery cells based on the battery pack temperature dynamic characteristic parameters and the temperature field distribution prediction trajectory, so as to obtain actual temperature data of each battery cell; A solution module, used to calculate the battery pack temperature average based on the actual temperature data of each battery cell, generate a temperature balance target trajectory considering the thermal coupling effect, and obtain the temperature balance control amount of each battery cell through multi-objective optimization solution; The output module is used to calculate the target heat dissipation power of each battery cell according to the temperature balance control amount, dynamically allocate the heat dissipation power in combination with the equivalent thermal resistance of each battery cell, and output the control instructions of each battery cell radiator.

Images