Intelligent temperature control management method and system for lithium iron phosphate battery

Abstract

The invention discloses an intelligent temperature control management method and system for a lithium iron phosphate battery, and relates to the technical field of battery temperature control, and the method comprises the steps: obtaining a battery sample data set which comprises sample working condition data and calibration health state data; constructing and training a health state evaluation model based on the sample data set, wherein the health state evaluation model is used for establishing a mapping relation between the working condition data and the health state; collecting ground state health data and historical working condition data of a target battery, inputting the ground state health data and the historical working condition data into the health state evaluation model for health state correction, and outputting real-time health state data; based on the real-time health state data and the real-time operation condition information of the target battery, a target state heat production model is established, and the predicted heat production rate of the target battery is calculated; and making a temperature control decision based on the predicted heat production rate, correspondingly generating a look-ahead temperature control adjustment strategy, and issuing the look-ahead temperature control adjustment strategy to a heat management assembly to execute temperature control management. The technical problem of poor battery temperature control effect in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of battery temperature control, in particular to an intelligent temperature control management method and system for lithium iron phosphate batteries. BACKGROUND

[0002] Lithium iron phosphate batteries have become the core power source of electric vehicles and energy storage systems due to their high safety, long cycle life and cost advantage. Thermal management technology is crucial to guarantee the performance, life and safety of the battery. Existing temperature control schemes mainly rely on temperature sensors to monitor in real time and trigger cooling strategies combined with preset thresholds. However, such methods have significant limitations: on the one hand, traditional static temperature control models cannot adapt to changes in battery health status, resulting in temperature control methods deviating from actual working conditions; on the other hand, existing schemes are difficult to cope with the risk of thermal runaway caused by sudden scenarios. Especially in actual operating scenarios, incorrect temperature control strategies can easily cause local overheating, accelerate life decay and even cause thermal failure. SUMMARY

[0003] The application provides an intelligent temperature control management method and system for lithium iron phosphate batteries, which is used to solve the technical problem that the temperature control strategy for lithium iron phosphate batteries in the prior art is inaccurate, resulting in poor temperature control effect.

[0004] In view of the above problems, the application provides an intelligent temperature control management method and system for lithium iron phosphate batteries.

[0005] In the first aspect, the application provides an intelligent temperature control management method for lithium iron phosphate batteries, which comprises: obtaining a battery sample data set, wherein the sample data set comprises sample working condition data and calibrated health status data; constructing and training a health status evaluation model based on the sample data set, wherein the health status evaluation model is used to establish a mapping relationship between working condition data and health status; collecting the base health data and historical working condition data of a target battery, inputting the health status evaluation model for health status correction, and outputting real-time health status data; based on the real-time health status data and real-time operating condition information of the target battery, establishing a target state heat generation model, and calculating the predicted heat generation rate of the target battery according to the target state heat generation model; based on the predicted heat generation rate, making a temperature control decision, corresponding to generating a forward-looking temperature control adjustment strategy, and issuing it to a thermal management component to perform temperature control management.

[0006] In the second aspect, the application provides an intelligent temperature control management system for lithium iron phosphate batteries, which comprises: a data acquisition module for acquiring a battery sample data set, wherein the sample data set comprises sample working condition data and calibrated health status data; a health model construction module configured to construct and train a health state evaluation model based on the sample data set, the health state evaluation model being configured to establish a mapping relationship between working condition data and a health state; a health state correction module configured to collect basal health data and historical working condition data of the target battery, input the health state evaluation model for health state correction, and output real-time health state data; a heat production model construction module configured to establish a target-state heat production model based on the real-time health state data and real-time running working condition information of the target battery, and calculate a predicted heat production rate of the target battery according to the target-state heat production model; a temperature control decision module configured to make a temperature control decision based on the predicted heat production rate, generate a forward-looking temperature control adjustment strategy in response, and issue the temperature control adjustment strategy to a thermal management component for temperature control management.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The present application provides an intelligent temperature control management method and system for lithium iron phosphate batteries. By dynamically fusing real-time health state evaluation of the battery and working condition driven heat production prediction mechanism, the dynamic adaptability and forward-looking ability of the lithium iron phosphate temperature control system are significantly improved. Compared with the traditional method, the technical solution provided by the present application significantly overcomes the inherent defects that the static method does not match the actual situation of the battery, realizes precise thermal management based on the individual characteristics of the battery, and achieves the technical effects of improving the temperature control effect of the lithium iron phosphate battery and reducing the comprehensive energy consumption of the temperature control system. BRIEF DESCRIPTION OF DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0009] Figure 1 A flowchart of an intelligent temperature control management method for lithium iron phosphate batteries provided by an embodiment of the present application.

[0010] Figure 2 A structure diagram of an intelligent temperature control management system for lithium iron phosphate batteries provided by an embodiment of the present application.

[0011] In the drawings, the components represented by the numbers are described as follows: The data acquisition module 100, the health model construction module 200, the health state correction module 300, the heat production model construction module 400, and the temperature control decision module 500. DETAILED DESCRIPTION

[0012] The application provides a lithium iron phosphate battery intelligent temperature control management method and system, which is used for solving the technical problem that the temperature control effect is poor due to the inaccurate temperature control strategy of the lithium iron phosphate battery in the prior art.

[0013] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0014] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server comprising a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to the process, method, product or device.

[0015] Embodiment one, as shown in the application provides a lithium iron phosphate battery intelligent temperature control management method, wherein the method comprises: Figure 1 S10: obtaining a battery sample data set, wherein the sample data set comprises sample working condition data and calibrated health state data.

[0016] Traditional battery health state evaluation relies on single laboratory test data or simplified working condition simulation, and it is difficult to construct a sample system covering real complex scenes. The sample working condition data lacks multi-dimensional and long-period actual operation characteristics, resulting in that the obtained battery health state evaluation is disconnected with the actual application scene.

[0017] The step S10 in the method provided in the embodiments of the application comprises: extracting the sample working condition data based on the battery operation log, wherein the sample working condition data at least comprises a current value sequence, a voltage value sequence, a temperature value sequence and a cycle number record; obtaining the calibrated health state data through an offline test platform, wherein the calibrated health state data at least comprises an internal resistance measurement value sequence and a residual capacity measurement value sequence.

[0018] In the embodiments of the application, the sample working condition data is extracted based on the battery operation log. The working condition data is the working condition data in the battery operation process, and the extracted sample working condition data at least comprises a current value sequence, a voltage value sequence, a temperature value sequence and a cycle number record.

[0019] ​The calibration health state data is obtained through an offline test platform, such as an electrochemical workstation including a constant current charge and discharge tester. The calibration health state data at least includes a sequence of internal resistance measurement values in units of mΩ and a sequence of remaining capacity measurement values in units of %.

[0020] By synchronously obtaining the multi-dimensional sample working condition data in the actual operation log and the calibration health state data, a spatio-temporally aligned battery sample data set is constructed. The data set can solve the mapping distortion problem caused by the fragmentation of data sources in traditional methods. Obtaining the battery sample data set can cover different battery health conditions and typical working conditions, and significantly improve the generalization ability and reliability basis of the subsequent health state evaluation model.

[0021] S20: constructing and training a health state evaluation model based on the sample data set, the health state evaluation model being used to establish a mapping relationship between working condition data and health state.

[0022] Existing health state evaluation models mostly use a single indicator such as capacity attenuation or a fixed formula for estimation, which cannot simultaneously capture the nonlinear changes in internal resistance and the gradual change characteristics of capacity, and often ignores the coupling effects of working condition changes on battery health.

[0023] The step S20 in the method provided in the embodiments of the present application includes: The sample working condition data is taken as input, and the sequence of internal resistance measurement values is taken as supervision, and a first mathematical fitting channel is constructed and trained accordingly, the first mathematical fitting channel being used to establish a mapping relationship between working condition data and internal resistance; The sample working condition data is taken as input, and the sequence of remaining capacity measurement values is taken as supervision, and a second machine learning channel is constructed and trained accordingly, the second machine learning channel being used to establish a mapping relationship between working condition data and remaining capacity; The first mathematical fitting channel and the second machine learning channel are fused in parallel to generate the health state evaluation model.

[0024] In the embodiments of the present application, a linear fitting method is used to construct the first mathematical fitting channel, which is used to establish a mapping relationship between working condition data and internal resistance. The first mathematical fitting channel is trained with the sample working condition as input and the sequence of internal resistance measurement values as supervision until convergence, for example, the error of the internal resistance value output by the input working condition data is within ±0.1 mΩ, which means that the training of the first mathematical fitting channel is completed.

[0025] The second machine learning channel is constructed by using a machine learning method, and is used to establish a mapping relationship between the working condition data and the remaining capacity. An example of a 3-layer structure is used, in which the input layer is used to receive the working condition data, the hidden layer uses 32 nodes and uses a ReLU function for activation, and the output layer is used to output the predicted remaining capacity. The second machine learning channel is trained with sample working condition data as input and a sequence of measured remaining capacity as supervision until convergence, for example, the error value of the output remaining capacity is within ±1% of the input working condition data, that is, the second machine learning channel training is completed.

[0026] The first mathematical fitting channel and the second machine learning channel are connected in parallel to generate a health state evaluation model. The working condition data is input into the health state evaluation model, and the internal resistance value evaluated by the first mathematical fitting channel and the remaining capacity evaluated by the second machine learning channel are output.

[0027] By constructing a parallel dual-channel health state evaluation model: the first mathematical fitting channel focuses on the physical mapping of working condition data and internal resistance, and retains the internal resistance characteristics; the second machine learning channel learns the complex nonlinear relationship between working conditions and remaining capacity. The parallel dual-channel fusion can accurately establish a dual mapping relationship between the working condition data and the health state, avoid the limitations of a single model, provide a high-precision basis for simultaneous correction of internal resistance and capacity, and ensure reliable parameters for subsequent use.

[0028] S30: Collecting the base state health data and the historical working condition data of the target battery, inputting the health state evaluation model for health state correction, and outputting real-time health state data.

[0029] The general health state model does not combine the individual base state characteristics and historical running cumulative effects of the target battery, and direct application will lead to evaluation deviation. The evaluation results of the traditional method are easy to deviate from the true situation.

[0030] The step S30 in the method provided in the embodiments of the present application includes: Calling a factory detection report of the target battery to extract base state health data of the target battery; Extracting features based on a pre-operation log of the target battery to obtain the historical working condition data, wherein the historical working condition data includes a plurality of working condition index sequences; Aligning the historical working condition data, inputting the aligned historical working condition data and the base state health data into the health state evaluation model, and obtaining the real-time health state data, wherein the real-time health state data includes a real-time internal resistance evaluation value and a remaining capacity evaluation value.

[0031] In the embodiments of the present application, a factory detection report of the target battery is called to extract the base state health data of the target battery. The base state health data includes an internal resistance value and a remaining capacity.

[0032] Feature extraction is performed based on the pre-operation log of the target battery. The pre-operation log refers to the operation log of the battery in the previous historical time, and the historical working condition data is obtained. The historical working condition data includes a plurality of working condition index sequences, such as a current value sequence, a voltage value sequence, a temperature value sequence, and a cycle number record.

[0033] According to the time stamps of the plurality of working condition index sequences, the working condition index data with the same time stamp is aligned until the historical working condition data is aligned. The aligned historical working condition data and the ground state health data are input into a health state evaluation model to obtain real-time health state data, wherein the real-time health state data includes a real-time internal resistance evaluation value and a remaining capacity evaluation value.

[0034] The application calls the factory ground state health data of the target battery, extracts the historical working condition data in the pre-operation log, and inputs the time series after alignment into the health state evaluation model. The ground state data is used as the degradation starting point calibration benchmark, and the historical working condition is used to quantify the cumulative aging effect, which realizes the improvement from the general model to the individual evaluation. The output internal resistance and remaining capacity evaluation values dynamically reflect the actual degradation degree of the target battery, and provide accurate parameter input for heat production prediction.

[0035] S40: Based on the real-time health state data and the real-time operation working condition information of the target battery, a target state heat production model is established, and a predicted heat production rate of the target battery is calculated according to the target state heat production model.

[0036] The static heat production model cannot adapt to the internal resistance increase and the intensified characteristics of the side reaction caused by the change of the battery health, resulting in inaccurate prediction of the heat production rate. The traditional method has fixed parameters, and cannot dynamically update the calculation rules according to the real-time health state, and the error is intensified in the high rate or low temperature scene. If the linkage mechanism of the health state and the heat production model is not established, the prediction result is difficult to reflect the real heat of the lithium iron phosphate battery.

[0037] The method provided in the embodiment of the application comprises the following steps: The residual capacity measurement value sequence and the corresponding side reaction heat measurement value sequence are combined to establish a side reaction heat calculation model through nonlinear fitting; The internal resistance heat production calculation model is constructed by calling the lithium iron phosphate material physical property knowledge base; The side reaction heat calculation model and the internal resistance heat production calculation model are fused to form a ground state heat production model, and the ground state heat production model is calibrated in reverse.

[0038] Based on the internal resistance evaluation value and the remaining capacity evaluation value, the internal resistance heat production calculation model parameter and the model parameter of the side reaction heat calculation model in the ground state heat production model are updated respectively, and the target state heat production model is obtained. inputting the real-time health state data and the real-time operation condition information into a target state heat production model for iterative calculation; When the deviation of the heat production rate calculated for a continuous preset number of times is less than a preset convergence threshold, outputting a final iteration value as the predicted heat production rate.

[0039] In the embodiments of the present application, the residual capacity measurement value sequence and the corresponding side reaction heat measurement value sequence are combined to establish a side reaction heat calculation model through nonlinear fitting. The side reaction heat measurement value sequence refers to the heat value generated by the side reaction existing in the change of the chemical potential energy of the battery during operation, which is obtained in the past laboratory determination and calculation. Exemplarily, the curve_fit function is used for nonlinear fitting until a nonlinear mathematical relationship representing the change rule between the residual capacity and the side reaction heat is found, and the side reaction heat calculation model is obtained.

[0040] A lithium iron phosphate material property knowledge base is called to construct an internal resistance heat production calculation model. The lithium iron phosphate material property knowledge base is a publicly accessible material property knowledge base, which contains the reaction equation of lithium iron phosphate, thermodynamic laws such as Joule's law, etc. By calling the lithium iron phosphate material property knowledge base, the lithium iron phosphate internal resistance heat production can be calculated. The internal resistance heat production calculation model is constructed, and the basic parameters of the lithium iron phosphate battery such as the current parameter are input to output the internal resistance heat production of the lithium iron phosphate battery through calculation.

[0041] The side reaction heat calculation model and the internal resistance heat production calculation model are fused to form a base state heat production model, and the base state heat production model is calibrated in reverse. Specifically, the total reaction heat of the lithium iron phosphate battery is measured by using a thermal sensor such as a thermometer, and the internal resistance heat production is calculated according to the internal resistance heat production calculation model, and the total reaction heat minus the internal resistance heat production is the side reaction heat. The base state heat production model is calibrated in reverse through this mathematical relationship, and the base state heat production model is calibrated until the output internal resistance heat production + side reaction heat = total reaction heat.

[0042] Based on the internal resistance evaluation value and the residual capacity evaluation value, the internal resistance heat production calculation model parameters in the base state heat production model and the model parameters of the side reaction heat calculation model are updated respectively, and the internal resistance heat production calculation model parameters and the side reaction heat calculation model whose model parameters are updated are integrated into a target state heat production model. Specifically, the internal resistance evaluation value and the residual capacity evaluation value are input into the internal resistance heat production calculation model and the side reaction heat calculation model, so that the model is initialized, the internal resistance heat production and the side reaction heat production are calculated, and the target state heat production model whose parameters are updated is obtained.

[0043] The real-time health state data and the real-time operation condition information are input into the target state heat production model for iterative calculation.

[0044] When the deviations of the calculated heat generation rates are all less than the preset convergence threshold value for a preset number of times, output the final iteration value as the predicted heat generation rate. The preset number of times can be set to 10 times, for example. The greater the preset number of times, the more stable the calculated heat generation rate, and the greater the possibility of obtaining an accurate heat generation rate. The preset convergence threshold value is a preset heat generation rate deviation threshold value. When the deviations of the calculated results are all less than the threshold value, it can be considered that the calculation result has converged, and the obtained final iteration value can be output as an optimal value. Preferably, the preset convergence threshold value can be set to a range of less than 2% of the total value of the calculated heat generation rates. When the deviations of the calculated results are all less than 2% of the predicted heat generation rate, output the final iteration value as the predicted heat generation rate, with a unit of J / s.

[0045] The parameters of the base-state heat generation model are dynamically updated based on real-time health state data, and a target-state heat generation model that adapts to the current aging state of the target battery is generated through iterative calculation until convergence, to accurately output the predicted heat generation rate. The method provided in the application can realize real-time coupling of the heat generation model and the health state, accurately predict the heat generation, and provide a reliable basis for subsequent temperature control decisions.

[0046] S50: making a temperature control decision based on the predicted heat generation rate, generating a forward-looking temperature control adjustment strategy accordingly, and delivering the strategy to a thermal management component to perform temperature control management.

[0047] Traditional temperature control relies on passive response of current temperature feedback and cannot predict heat generation trends, resulting in lagging regulation and control. The fixed threshold strategy does not consider the thermal capacity changes and heat generation acceleration characteristics of batteries in different health states, and is prone to cause local overheating or excessive cooling under sudden working conditions.

[0048] The step S50 in the method provided in the embodiments of the application includes: deducing a temperature change trend based on the predicted heat generation rate and a preset time window; controlling and judging the temperature change trend according to a preset temperature control constraint, wherein the temperature control constraint includes a process temperature control constraint and an end-point temperature control constraint; if the control and judgment result is not satisfied, calculating a temperature control power according to the predicted heat generation rate and generating the forward-looking temperature control adjustment strategy accordingly; programming the forward-looking temperature control adjustment strategy and delivering the strategy to the thermal management component to perform temperature control management.

[0049] In the embodiments of the present application, based on the predicted heat generation rate and the preset time window, the temperature change trend is deduced. For example, the preset time window is set to 30 minutes in the future, the real-time health state data and the real-time running condition are input into the target state heat generation model to obtain the predicted heat generation rate, which is in units of J / s. According to the predicted heat generation rate, the temperature change rate is obtained by linear prediction, which is in units of ℃ / s. Specifically, the time step is divided into 10s, and according to the heat generation rate and the battery heat capacity value (i.e. the heat required for the battery to rise by 1℃, in units of J / ℃, which is obtained through battery production data), a plurality of temperature values are obtained by linear prediction, and the temperature value = predicted heat generation rate x time step ÷ battery heat capacity value. The plurality of temperature values are integrated according to the time step to obtain the temperature change rate.

[0050] According to the preset temperature control constraint, the temperature change trend is controlled and discriminated. The temperature control constraint includes a process temperature control constraint and an end temperature control constraint. That is, the temperature control constraint threshold in the temperature change process and the temperature control constraint threshold at the end of the temperature change. For example, the process temperature control constraint threshold is set to a temperature rise speed less than 0.5℃ / s, and the key temperature control constraint is set to 50℃.

[0051] If the control discrimination result is not satisfied, that is, any one of the temperature control constraints is not satisfied or both are not satisfied, the temperature control power is calculated according to the predicted heat generation rate. For example, according to the predicted heat generation rate, the time step, and the battery heat capacity value, the heat generation rate that needs to be reduced to reach the temperature control threshold is calculated as the temperature control power. The temperature control power = [temperature control threshold - (predicted heat generation rate x time step ÷ battery heat capacity value)] ÷ time step. According to the temperature control power, a forward-looking temperature control adjustment strategy is generated, for example, the temperature control system is started, and the temperature control system power = temperature control power.

[0052] The program language forward-looking temperature control adjustment strategy, such as using C language to write the temperature control adjustment strategy, is issued to the thermal management component to execute temperature control management.

[0053] Based on the predicted heat generation rate, the temperature change trend is deduced, and the process and end temperature control constraints are dynamically discriminated. When the predicted trend does not meet the constraints, a temperature control strategy that adapts to the heat generation rate is automatically generated and issued to the thermal management component for execution. The method provided by the present application can change passive response to active defense, suppress the risk of thermal runaway, reduce temperature control energy consumption redundancy through precise power matching, and realize the synergistic optimization of temperature control effect and temperature control energy efficiency.

[0054] Embodiment two, as shown in Figure 2 based on the same inventive concept of the lithium iron phosphate battery intelligent temperature control management method provided in embodiment one, the present application embodiment further provides a lithium iron phosphate battery intelligent temperature control management system, comprising: The data acquisition module 100 is configured to acquire a battery sample data set, wherein the sample data set comprises sample working condition data and calibrated health state data. The health model construction module 200 is configured to construct and train a health state evaluation model based on the sample data set, wherein the health state evaluation model is used to establish a mapping relationship between working condition data and health state. The health state correction module 300 is configured to acquire basal health data and historical working condition data of a target battery, input the health state evaluation model for health state correction, and output real-time health state data. The heat production model construction module 400 is configured to establish a target state heat production model based on the real-time health state data and real-time running working condition information of the target battery, and calculate a predicted heat production rate of the target battery according to the target state heat production model. The temperature control decision module 500 is configured to make a temperature control decision based on the predicted heat production rate, generate a forward-looking temperature control adjustment strategy in correspondence, and issue the temperature control adjustment strategy to a thermal management component for temperature control management.

[0055] In one embodiment, the data acquisition module 100 is further configured to: extract the sample working condition data based on a battery running log, wherein the sample working condition data at least includes a current value sequence, a voltage value sequence, a temperature value sequence, and a cycle number record; acquire the calibrated health state data through an offline test platform, wherein the calibrated health state data at least includes an internal resistance measurement value sequence and a remaining capacity measurement value sequence.

[0056] In one embodiment, the health model construction module 200 is further configured to: construct and train a first mathematical fitting channel with the sample working condition data as input and the internal resistance measurement value sequence as supervision, wherein the first mathematical fitting channel is used to establish a mapping relationship between working condition data and internal resistance; construct and train a second machine learning channel with the sample working condition data as input and the remaining capacity measurement value sequence as supervision, wherein the second machine learning channel is used to establish a mapping relationship between working condition data and remaining capacity; parallelly fuse the first mathematical fitting channel and the second machine learning channel to generate the health state evaluation model.

[0057] In one embodiment, the health state correction module 300 is further configured to: call a factory detection report of the target battery to extract basal health data of the target battery; extract features based on a pre-order running log of the target battery to acquire the historical working condition data, wherein the historical working condition data comprises a plurality of working condition index sequences; aligning the historical operating condition data, and inputting the aligned historical operating condition data and the base state health data into the health state evaluation model to obtain the real-time health state data, wherein the real-time health state data includes a real-time internal resistance evaluation value and a remaining capacity evaluation value.

[0058] In one embodiment, the heat generation model construction module 400 is further configured to: combine the sequence of the remaining capacity measurement values and the corresponding sequence of the side reaction heat measurement values, and establish a side reaction heat calculation model through nonlinear fitting; invoke a lithium iron phosphate material physical property knowledge base to construct an internal resistance heat generation calculation model; fuse the side reaction heat calculation model and the internal resistance heat generation calculation model to form a base state heat generation model, and perform reverse parameter calibration on the base state heat generation model.

[0059] based on the internal resistance evaluation value and the remaining capacity evaluation value, update the internal resistance heat generation calculation model parameter and the model parameter of the side reaction heat calculation model in the base state heat generation model, respectively, to obtain the target state heat generation model; input the real-time health state data and the real-time operating condition information into the target state heat generation model for iterative calculation; when the deviation of the heat generation rate obtained by continuous preset times of calculation is less than a preset convergence threshold, output the final iteration value as the predicted heat generation rate.

[0060] In one embodiment, the temperature control decision module 500 is further configured to: based on the predicted heat generation rate and a preset time window, deduce a temperature change trend; control and discriminate the temperature change trend according to a preset temperature control constraint, wherein the temperature control constraint includes a process temperature control constraint and an end point temperature control constraint; if the control and discrimination result is not satisfied, calculate a temperature control power according to the predicted heat generation rate, and correspondingly generate the forward-looking temperature control adjustment strategy; program the forward-looking temperature control adjustment strategy and issue it to the thermal management component to perform temperature control management.

[0061] In summary, the embodiments of the present application have at least the following technical effects: The application provides a lithium iron phosphate battery intelligent temperature control management method and system. Through dynamic fusion of real-time health state evaluation of the battery and working condition driven heat generation prediction mechanism, the dynamic adaptability and foresight ability of the lithium iron phosphate temperature control system are significantly improved. Specifically, based on the real-time correction of the internal resistance and capacity of the battery by the health state evaluation model, the target state heat generation model can dynamically capture the nonlinear evolution of the battery state, avoiding the heat generation prediction deviation caused by parameter solidification. At the same time, through the forward deduction and constraint discrimination mechanism of the predicted heat generation rate, the system can identify the potential temperature rise risk in advance and generate a forward-looking temperature control strategy, changing the traditional passive response mode into an active defense mode, effectively inhibiting the overheating phenomenon under sudden working conditions. In addition, the temperature control decision process takes into account the multidimensional constraints of the process and the endpoint, optimizing the energy consumption distribution under the premise of ensuring the safety threshold, and avoiding excessive cooling or heating redundancy caused by response lag. Compared with the traditional method, the technical scheme provided by the application significantly overcomes the inherent defects that the static method does not match the actual situation of the battery, realizes accurate thermal management based on the individual characteristics of the battery, and achieves the technical effects of improving the temperature control effect of the lithium iron phosphate battery and reducing the comprehensive energy consumption of the temperature control system.

[0062] It should be noted that the above sequence of the embodiments of the application is only for description, and does not represent the advantages and disadvantages of the embodiments. Moreover, the above describes specific embodiments of the present application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or can be advantageous.

[0063] The above only describes the preferred embodiments of the application and does not limit the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall be included in the protection scope of the application.

[0064] The present application and the drawings are only exemplary descriptions of the application, and are considered to cover any and all modifications, changes, combinations or equivalents within the scope of the application. Obviously, those skilled in the art can make various modifications and changes to the application without departing from the scope of the application. Thus, if these modifications and changes of the application belong to the scope of the application and its equivalent technology, the application intends to include these modifications and changes.

Claims

1. A method for intelligent temperature control management of a lithium iron phosphate battery, characterized in that, The method comprises the following steps: acquiring a battery sample data set, the sample data set comprising sample working condition data and calibrated health state data; constructing and training a health state evaluation model based on the sample data set, the health state evaluation model being used to establish a mapping relationship between working condition data and health state; collecting base state health data and historical working condition data of a target battery, inputting the health state evaluation model for health state correction, and outputting real-time health state data; based on the real-time health state data and real-time running working condition information of the target battery, establishing a target state heat generation model, and calculating a predicted heat generation rate of the target battery according to the target state heat generation model; based on the predicted heat generation rate, making a temperature control decision, corresponding to generating a forward-looking temperature control adjustment strategy, and issuing the temperature control adjustment strategy to a thermal management component for temperature control management.

2. The intelligent temperature control management method for lithium iron phosphate battery according to claim 1, characterized in that, acquiring a battery sample data set, the sample data set comprising sample working condition data and calibrated health state data, comprising: extracting the sample working condition data based on a battery running log, the sample working condition data at least containing a current value sequence, a voltage value sequence, a temperature value sequence and a cycle number record; acquiring the calibrated health state data through an offline test platform, the calibrated health state data at least containing an internal resistance measurement value sequence and a remaining capacity measurement value sequence.

3. The intelligent temperature control management method for lithium iron phosphate battery according to claim 2, characterized in that, constructing and training a health state evaluation model based on the sample data set, the health state evaluation model being used to establish a mapping relationship between working condition data and health state, comprising: taking the sample working condition data as input and the internal resistance measurement value sequence as supervision, corresponding to constructing and training a first mathematical fitting channel, the first mathematical fitting channel being used to establish a mapping relationship between working condition data and internal resistance; taking the sample working condition data as input and the remaining capacity measurement value sequence as supervision, corresponding to constructing and training a second machine learning channel, the second machine learning channel being used to establish a mapping relationship between working condition data and remaining capacity; parallelly fusing the first mathematical fitting channel and the second machine learning channel to generate the health state evaluation model.

4. The intelligent temperature control management method for lithium iron phosphate battery according to claim 3, characterized in that, collecting base state health data and historical working condition data of a target battery, inputting the health state evaluation model for health state correction, and outputting real-time health state data, comprising: calling a factory detection report of the target battery to extract base state health data of the target battery; performing feature extraction based on a pre-order running log of the target battery to obtain the historical working condition data, wherein the historical working condition data comprises a plurality of working condition index sequences; aligning the historical working condition data, and inputting the aligned historical working condition data and the base state health data into the health state evaluation model to obtain the real-time health state data, wherein the real-time health state data comprises a real-time internal resistance evaluation value and a remaining capacity evaluation value.

5. The intelligent temperature control management method for lithium iron phosphate battery according to claim 4, characterized in that, based on the real-time health state data and real-time running working condition information of the target battery, establishing a target state heat generation model, and calculating a predicted heat generation rate of the target battery according to the target state heat generation model, previously comprising: establishing a side reaction heat calculation model through nonlinear fitting in combination with the remaining capacity measurement value sequence and a corresponding side reaction heat measurement value sequence; calling a lithium iron phosphate material property knowledge base to construct an internal resistance heat generation calculation model; The base-state heat generation model is formed by fusing the side reaction heat calculation model and the internal resistance heat generation calculation model, and the base-state heat generation model is calibrated in reverse.

6. The intelligent temperature control management method for lithium iron phosphate battery according to claim 5, characterized in that, Based on the real-time health state data and real-time operation condition information of the target battery, a target-state heat generation model is established, and a predicted heat generation rate of the target battery is calculated according to the target-state heat generation model, including: Based on the internal resistance evaluation value and the remaining capacity evaluation value, the model parameters of the internal resistance heat generation calculation model and the side reaction heat calculation model in the base-state heat generation model are updated respectively, and the target-state heat generation model is obtained; The real-time health state data and the real-time operation condition information are input into the target-state heat generation model for iterative calculation; When the deviation of the heat generation rate calculated for a continuous preset number of times is less than a preset convergence threshold, a final iteration value is output as the predicted heat generation rate.

7. The intelligent temperature control management method for lithium iron phosphate battery according to claim 1, characterized in that, Based on the predicted heat generation rate, a temperature control decision is made, a forward-looking temperature control adjustment strategy is generated accordingly, and the temperature control management is executed by the thermal management component, including: Based on the predicted heat generation rate and a preset time window, a temperature change trend is deduced; The temperature change trend is controlled and discriminated according to a preset temperature control constraint, wherein the temperature control constraint includes a process temperature control constraint and an end-point temperature control constraint; If the control discrimination result is not satisfied, a temperature control power is calculated according to the predicted heat generation rate, and the forward-looking temperature control adjustment strategy is generated accordingly; The forward-looking temperature control adjustment strategy is programmed and delivered to the thermal management component for temperature control management.

8. A lithium iron phosphate battery intelligent temperature control management system, characterized in that, A lithium iron phosphate battery intelligent temperature control management method for implementing any one of claims 1-7, the system comprising: A data acquisition module for acquiring a battery sample data set, the sample data set including sample condition data and calibrated health state data; A health model construction module for constructing and training a health state evaluation model based on the sample data set, the health state evaluation model being used to establish a mapping relationship between condition data and health state; A health state correction module for collecting base-state health data and historical condition data of a target battery, inputting the health state evaluation model for health state correction, and outputting real-time health state data; A heat generation model construction module for establishing a target-state heat generation model based on the real-time health state data and real-time operation condition information of the target battery, and calculating a predicted heat generation rate of the target battery according to the target-state heat generation model; A temperature control decision module for making a temperature control decision based on the predicted heat generation rate, generating a forward-looking temperature control adjustment strategy accordingly, and delivering the temperature control adjustment strategy to a thermal management component for temperature control management.

Images