Battery power map construction method based on physical model and key point test
By constructing a battery power MAP based on physical models and key point testing, the problem of battery discharge capacity being affected by temperature under low SOC conditions is solved, achieving efficient and accurate power prediction and improved battery safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NAMEI NEW ENERGY TECH (LUOYANG) CO LTD
- Filing Date
- 2026-05-26
- Publication Date
- 2026-06-23
Smart Images
Figure CN122260140A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of battery power MAP construction technology, specifically to a battery power MAP construction method based on physical models and key point testing. Background Technology
[0002] In the development of hybrid vehicles, the battery management system requires a precise power map (MAP) to manage battery power output in real time, ensuring vehicle performance, safety, and battery life. Traditional methods for obtaining the power map require testing the battery's performance across a large matrix of temperature points (e.g., every 5°C from -30°C to 50°C) and state of charge (SOC) points (e.g., every 5% from 0% to 100%). This method is time-consuming, costly, and consumes a large number of cell samples, severely slowing down project development.
[0003] In low SOC conditions, battery discharge capacity is significantly affected by temperature. The nonlinear impact of temperature changes on battery discharge capacity is significantly amplified. At low SOC, concentration polarization within the battery intensifies, increasing polarization impedance and causing local deviations in the basic Arrhenius dynamics fitting. This results in a stronger nonlinear correlation between temperature and discharge current. Existing technologies primarily use the Arrhenius formula to fit currents at different temperatures. However, these technologies lack adaptability to the nonlinear electrical characteristics dominated by concentration polarization at low SOC conditions. This leads to significantly increased power prediction errors under extreme conditions such as low temperature and low SOC, affecting the battery management system's accurate judgment of peak power and posing risks of voltage collapse and battery life degradation. Summary of the Invention
[0004] To address the aforementioned technical problems, this application provides a battery power MAP construction method based on physical models and key point testing to solve existing issues.
[0005] The battery power MAP construction method based on physical model and key point testing in this application adopts the following technical solution: One embodiment of this application provides a method for constructing a battery power MAP based on a physical model and key point testing. The method includes the following steps: Within the target temperature range and target state of charge (SOC) range of the battery, key test points are selected according to preset principles; different rate pulses are applied to any key test point of the battery to form a working condition; the current sequence and voltage sequence of the battery at each acquisition time under each working condition are obtained; the maximum discharge current under each working condition in the temperature dimension is fitted to obtain the basic power value of the battery under each working condition. Using the filtered voltage and current sequences at each acquisition moment as input, the polarization impedance at each acquisition moment under each low SOC condition is calculated, and the concentration polarization resistance at each acquisition moment under each low SOC condition is calculated in combination with the reference polarization impedance; the low SOC condition is the condition where the SOC is less than or equal to a preset low threshold percentage. The concentration polarization resistance at the next acquisition time is predicted. All predicted values within a time window before each acquisition time are used as input to the genetic algorithm to obtain the optimal solution of the optimization sample. Based on the optimal solution, the low SOC power correction coefficient at each acquisition time under each low SOC condition is calculated. The optimization sample is a sequence composed of the concentration polarization resistance sequence at each acquisition time under each low SOC condition and the predicted value of the concentration polarization resistance at the next acquisition time. The base power value under low SOC conditions is corrected using a low SOC power correction factor; a two-dimensional power MAP of the battery is constructed based on the corrected base power values under all low SOC conditions and the base power values under non-low SOC conditions.
[0006] Preferably, fitting the maximum discharge current for each operating condition in the temperature dimension includes: For the maximum discharge current value at key test points at different temperatures under the same SOC, the reciprocal of the absolute temperature is used as the abscissa and the natural logarithm of the maximum discharge current value is used as the ordinate. After transforming it to the Arrhenius coordinate space, a linear fit is performed to establish the current-temperature relationship model under that SOC. The slope of the linear equation obtained by the linear fit is used to calculate the apparent activation energy.
[0007] Preferably, the key test point is a two-dimensional point consisting of a temperature point and a SOC point, and the selection principles for the temperature point and SOC point include: The temperature points include extreme low temperature points, standard test temperature points, room temperature reference points, and extreme high temperature points; SOC points include minimum cutoff SOC point, low SOC point, normal SOC point, high SOC point, and full charge SOC point.
[0008] Preferably, the current sequence and voltage sequence at each acquisition moment are respectively composed of current data and voltage data acquired within a time window set before each acquisition moment, arranged in chronological order of their acquisition time.
[0009] Preferably, the concentration polarization resistance is the ratio of the polarization impedance at each acquisition time under each low SOC condition to the reference polarization impedance at the corresponding acquisition time.
[0010] Preferably, the reference polarization impedance is calculated under the conditions of a temperature of 25°C, a discharge rate of 1C, and a state of charge (SOC) of 50%.
[0011] Preferably, the fitness function in the genetic algorithm is: Where min is the minimum value function, and I is the number of historical data collection moments. , These represent the measured value of the concentration polarization resistance at the i-th acquisition time within a sliding window prior to each acquisition time under each low SOC condition, and the optimized value of the concentration polarization resistance at the i-th acquisition time in the concentration polarization feature sequence at each acquisition time under each low SOC condition.
[0012] Preferably, the calculation method for the low SOC power correction coefficient is as follows: The mean of the sequence corresponding to the optimal solution is used as the optimal concentration polarization feature at each acquisition time under each low SOC condition. The reciprocal of the sum of the optimal concentration polarization characteristic and the value of 1 is calculated and used as the low SOC power correction coefficient for each acquisition time under each low SOC condition.
[0013] Preferably, the correction result of the base power value under the low SOC condition is the product of the low SOC power correction coefficient and the base power value under the low SOC condition.
[0014] Preferably, constructing the two-dimensional power MAP of the battery based on power values under all operating conditions includes: For each low SOC condition, the power value obtained by correcting it at different key temperature points is combined with the rated minimum operating voltage of the reference voltage platform under that SOC condition, and the equivalent maximum discharge current is calculated by back-calculating the ratio between the corrected power value and the rated minimum operating voltage. The corrected maximum discharge current data under low SOC conditions and the maximum discharge current under non-low SOC conditions are transformed into Arrhenius coordinate space for linear fitting and interpolation to generate a two-dimensional power MAP of the battery.
[0015] This application has at least the following beneficial effects: 1. To address the issue that the discharge capacity of batteries under low SOC conditions is significantly affected by the nonlinear effect of temperature, this application extracts the impedance degree of concentration polarization using a sliding window filtering algorithm and a recursive least squares method. This eliminates the interference of individual battery differences on the comparison of absolute impedance values, thereby improving the accuracy and comparability of polarization state assessment under low SOC conditions.
[0016] 2. To address the problem that the non-stationary characteristics of concentration polarization resistance under low SOC conditions make it difficult to directly correct the power under dynamic conditions, this application introduces the ARIMA model to predict its temporal evolution and combines it with a genetic algorithm to globally optimize the power correction coefficients, thereby eliminating the limitations of linear assumptions and local optima, and improving the dynamic adaptability and accuracy of power prediction under low SOC conditions.
[0017] 3. By applying the low SOC power correction factor B to the basic power MAP, targeted corrections are made to the power under operating conditions with SOC less than or equal to 20%. This application achieves improved accuracy of the battery power MAP under extreme operating conditions such as low temperature and low SOC, effectively reducing the risk of voltage collapse caused by overly aggressive power strategies, and ensuring the safety and service life of the battery system. Attached Figure Description
[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 A flowchart of the battery power MAP construction method based on physical model and key point testing provided in this application. Detailed Implementation
[0020] To further illustrate the technical means and effects adopted by this application to achieve the intended inventive objective, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the battery power MAP construction method based on physical models and key point testing proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.
[0022] The following description, in conjunction with the accompanying drawings, details the specific scheme of the battery power MAP construction method based on physical model and key point testing provided in this application.
[0023] One embodiment of this application provides a method for constructing a battery power MAP based on a physical model and key point testing.
[0024] Specifically, the following method for constructing a battery power MAP based on a physical model and key point testing is provided. Please refer to [link / reference]. Figure 1 The method includes the following steps: Step 1: Selection of key test points and modeling of temperature dimension.
[0025] Determine the target temperature range (set to -30°C to 50°C in this embodiment) and the target state of charge (SOC) range (set to 0% to 100% in this embodiment) for constructing the battery power MAP.
[0026] Within this scope, key test points are selected according to the following preset principles: The key test point is a two-dimensional point consisting of a temperature point and a SOC point. The selection principles for the temperature point and SOC point include: Temperature points: must include an extreme low temperature point (set to -30℃ in this embodiment), a standard test temperature point (such as -18℃ or -20℃, preferably -20℃ in this embodiment), a room temperature reference point (set to 25℃ in this embodiment), and an extreme high temperature point (set to 50℃ in this embodiment). Preferably, an additional point is added in the low-temperature performance drastic change zone (set to -10℃ in this embodiment).
[0027] SOC points: These must include the minimum cutoff SOC point (0% in this embodiment), low SOC point (20% in this embodiment), normal SOC point (60% in this embodiment), high SOC point (80% in this embodiment), and fully charged SOC point (100% in this embodiment). It should be noted that SOC refers to the battery's charge level. The higher the battery charge, the higher the voltage. A fully charged SOC is 100%, reaching the rated maximum voltage. When the SOC is 0%, the voltage is 0 (not considering battery degradation).
[0028] For each selected temperature and SOC combination, it is used as a key test point. Short pulses of different rates (increasing from 0.5C to 3C in 0.5C increments) are applied to each key test point in the battery (the duration of the short pulse in this embodiment is 10s). Each application of different rate pulses to any key test point in the battery represents a different operating condition. The Battery Management System records the current and voltage response data of the battery discharge at each key test point in real time during each short pulse application, thus acquiring the current and voltage data of the battery at each acquisition moment under each operating condition. In this embodiment, the data acquisition frequency is set to 1kHz. For each acquisition moment, a time window is set before each acquisition moment (the length of the time window is set to 3 seconds in this embodiment; if less than 3 seconds have passed before the current acquisition moment, all data from the start moment to the current moment is used as the time window). The current sequence and voltage sequence within the time window before each acquisition moment are acquired and recorded as the current sequence and voltage sequence for each acquisition moment. The elements in the sequence are arranged in chronological order of acquisition time.
[0029] Furthermore, the maximum discharge current under each operating condition is fitted using the Arrhenius equation to obtain the battery's base power value under each operating condition. .
[0030] The fitting of the maximum discharge current under each operating condition in the temperature dimension includes: for the maximum discharge current value of key test points at different temperatures under the same SOC, taking the reciprocal of the absolute temperature as the abscissa and the natural logarithm of the maximum discharge current value as the ordinate, transforming it to the Arrhenius coordinate space and then performing linear fitting to establish a current-temperature relationship model under that SOC; the slope of the linear equation obtained by the linear fitting is used to calculate the apparent activation energy.
[0031] The Arrhenius equation is a well-known technique, and its specific process will not be elaborated here.
[0032] Step 2: Data filtering and concentration polarization parameter extraction.
[0033] During battery discharge, the battery voltage and current signals are easily affected by electromagnetic interference and sensor noise, resulting in a low signal-to-noise ratio of the original battery voltage and current signal data. Directly using this data will cause a large calculation error, which will affect the accuracy of polarization impedance extraction and further affect the MAP accuracy of power under low SOC conditions.
[0034] Based on the above analysis, using the current and voltage sequences collected at each acquisition time as input, the sliding time window length is set to 1000 sampling points, and the sliding step size is 10 sampling points. A moving average filtering algorithm is used to output the filtered current and voltage sequences. The output current and voltage sequences represent the true changes in current and voltage after noise removal. The moving average filtering algorithm is a well-known technique and will not be described in detail here.
[0035] However, the parameters in the battery's equivalent circuit model change with temperature, SOC, and current. Traditional offline identification methods struggle to track and identify these changes, further leading to model inaccuracies under low SOC conditions. A low SOC condition is defined as an SOC less than or equal to a preset low threshold percentage. In this embodiment, the preset low threshold percentage is 20%, but the implementer can set the specific value according to actual conditions.
[0036] To address the aforementioned issues, this application introduces a recursive least squares method. Specifically, this embodiment employs the recursive least squares method, setting a forgetting factor of 0.98. Using the filtered voltage and current sequences at each acquisition moment as input, the polarization impedance R under each low SOC condition is calculated online based on a first-order RC equivalent circuit model. The output polarization impedance represents the impedance characteristics during the concentration polarization process within the battery under the current low SOC condition, expressing the degree of resistance experienced by lithium ions during solid-phase diffusion. The recursive least squares method is a well-known technique and will not be elaborated further.
[0037] Subsequently, under the reference conditions of 25°C, 1C discharge rate, and 50% SOC, a short-time pulse test was performed on the battery to collect current and voltage response data. The polarization impedance under this condition was calculated using the same recursive least squares method and used as the reference polarization impedance.
[0038] Based on the above analysis, the concentration polarization resistance A at each acquisition time under each low SOC condition is calculated. The specific calculation formula is as follows: Where R is the polarization impedance at each acquisition moment under each low SOC condition; The reference polarization impedance is the reference polarization impedance at each acquisition time under the reference operating conditions (25°C, 1C discharge, SOC 50%).
[0039] In the prior art, polarization impedance describes the resistance encountered by lithium ions during the solid-phase diffusion process of the electrode. The larger the value, the more difficult the solid-phase diffusion process of lithium ions is, and the stronger the concentration polarization effect during the discharge process, resulting in a decrease in power output capability.
[0040] Based on the above principles, in electrochemical impedance analysis, the polarization impedance value at each acquisition time under different low SOC conditions is obtained by fitting an equivalent circuit model to evaluate the state of the battery. However, due to individual differences in batteries, the calculated impedance value is difficult to directly compare the relative change in polarization. Based on this understanding, this application normalizes the polarization impedance obtained by fitting under each low SOC condition by comparing it with the reference impedance of the battery under standard reference conditions, and obtains the dimensionless concentration polarization resistance A. By eliminating individual differences through the ratio, the degree of polarization is transformed into a deterioration factor relative to the reference state. The larger the value of A, the more severe the concentration polarization, and the more significant the limitation on the battery's discharge capacity.
[0041] Step 3: Solve for the power correction coefficient based on time series prediction and genetic optimization.
[0042] As can be seen from the above analysis, under low SOC conditions, the calculated concentration polarization resistance A will exhibit complex non-stationary characteristics as the depth of discharge and temperature fluctuate. If the calculated concentration polarization resistance A is directly used to correct the power, the dynamic evolution of the battery power will be ignored, resulting in insufficient adaptability of the correction process under dynamic conditions, making it difficult to express the actual power of the battery and affecting the accuracy of MAP under transient conditions.
[0043] Based on the above analysis, the concentration polarization resistance A calculated at all acquisition times under each low SOC condition is sorted in chronological order to obtain the concentration polarization resistance sequence within the time window preceding each acquisition time. This sequence is recorded as the concentration polarization resistance sequence for each acquisition time under each low SOC condition. Using the concentration polarization resistance sequence for each acquisition time as input, an ARIMA (Autoregressive Integral Moving Average) model is used. In this embodiment, the autoregressive order p=2, the difference order d=1, and the moving average order q=2 are set to output the predicted value of the concentration polarization resistance at the next acquisition time under each low SOC condition. The output predicted value represents the degree of power attenuation predicted based on the historical evolution trend of A. The ARIMA model is a well-known technique and will not be elaborated further.
[0044] However, the ARIMA model used above assumes linearity and is difficult to characterize the strong nonlinear coupling relationship between concentration polarization resistance and temperature and SOC, resulting in large deviations in the predicted values output by the ARIMA model under extreme operating conditions.
[0045] The predicted values of concentration polarization resistance at all acquisition times within a sliding window preceding the current acquisition time under each low SOC condition, as output by the ARIMA model. The sequence arranged chronologically is denoted as the concentration polarization feature sequence at the current acquisition time. Using this sequence as input, each concentration polarization feature sequence is treated as a chromosome. A genetic algorithm is employed; in this embodiment, the population size is set to 50, crossover probability to 0.8, mutation probability to 0.1, and maximum generation count to 200. The concentration polarization resistance sequence at each acquisition time under each low SOC condition is compared with the predicted concentration polarization resistance value for the next acquisition time output by the ARIMA model. The resulting sequence is used as an optimization sample. A genetic algorithm is then used to search for an optimal concentration polarization feature sequence within a pre-defined correction space. This optimal concentration polarization feature sequence ensures that the corrected power value maximally covers the dynamic decay trajectory. The measured value of the concentration polarization resistance A is used as the basis for this search. and predicted value Set the fitness function to: Where min is the minimum value function, and I is the number of samples collected within a sliding window before each acquisition time. , These represent the measured value of the concentration polarization resistance at the i-th acquisition time within a sliding window prior to each acquisition time under each low SOC condition, and the optimized value of the concentration polarization resistance at the i-th acquisition time in the concentration polarization feature sequence at each acquisition time under each low SOC condition.
[0046] It should be noted that the above fitness function optimizes the concentration polarization feature sequence, minimizing the error between the measured value of concentration polarization resistance and the optimal predicted value of the dynamic power decay trajectory. This more effectively reflects the strong nonlinear coupling relationship between concentration polarization resistance and temperature and SOC. The genetic algorithm ultimately outputs the optimal solution, which is the most recent optimal concentration polarization feature sequence before the current acquisition time under each low SOC condition. This ensures that the corrected power value covers the dynamic decay trajectory to the greatest extent. Furthermore, the mean of the most recent optimal concentration polarization feature sequence before the current acquisition time under each low SOC condition is denoted as the optimal concentration polarization feature at the current acquisition time under each low SOC condition. This yields the optimal concentration polarization feature at each acquisition time under each low SOC condition. This allows for a more accurate characterization of the strong nonlinear coupling relationship between concentration polarization resistance and temperature and SOC.
[0047] Based on the above analysis, the low SOC power correction coefficient B is calculated for each acquisition time under each low SOC condition. The specific calculation formula is as follows: It should be noted that, A larger value indicates more severe concentration polarization and a weaker actual discharge capacity of the battery. To ensure that the corrected power decreases reasonably as polarization intensifies, B must be adjusted accordingly. The voltage decreases as the voltage increases, thus avoiding overestimation of power and preventing voltage collapse.
[0048] In existing technologies, the ARIMA (Autoregressive Integral Moving Average) model is used for time series analysis and is widely applied to the prediction of non-stationary data. ARIMA can effectively extract the linear trend of time series and periodically output the predicted values for future times. In existing technologies, the genetic algorithm is a global optimization algorithm that simulates natural evolution. It finds the optimal solution through selection, crossover and mutation operations and is often used to optimize parameters for nonlinear problems.
[0049] Based on the above analysis, this application fits the historical evolution of concentration polarization resistance A using the ARIMA model and outputs the predicted value of future concentration polarization resistance, thereby reflecting the power decay characteristics in time series. Furthermore, it uses a genetic algorithm to search for the optimal correction coefficient that satisfies the actual discharge constraints, combining nonlinear effects and global optimality to more accurately reflect the actual power capability under low SOC conditions.
[0050] Step 4: Improvement and construction of battery power MAP based on correction coefficient.
[0051] Through the analysis and calculation of the above steps, the low SOC power correction coefficient B was obtained under different combinations of low SOC operating conditions (temperature, SOC, current). This coefficient quantifies the degree to which concentration polarization suppresses battery discharge.
[0052] To improve the accuracy of the battery power MAP, this application improves the traditional power MAP based on the low SOC power correction coefficient B calculated above. The specific improvement method is as follows: First, the maximum discharge current under each operating condition is fitted using the Arrhenius equation to obtain the base power value for each condition. Further, in the power MAP construction process, the calculated low-SOC power correction coefficient B is applied to correct the power value for low-SOC conditions (SOC < 20%), resulting in the corrected power value. ;in, This represents the base power value for each operating condition.
[0053] For non-low SOC operating conditions (SOC ≥ 20%), the base power value remains unchanged.
[0054] The improved power MAP can significantly reduce power prediction errors under low SOC and low temperature conditions, thereby effectively avoiding the risk of voltage collapse caused by overly aggressive power strategies.
[0055] It should be noted that the power modification is only performed during the offline generation stage of the MAP table, and the generated P matrix is hard-coded and fixed in the BMS.
[0056] Furthermore, for each low SOC condition, the power value obtained by correcting the key test points at different temperatures through the above steps (the power value obtained after correcting the base power value) is combined with the rated minimum operating voltage of the reference voltage platform under that SOC condition. The ratio between the corrected power value and the rated minimum operating voltage is then used to calculate the equivalent maximum discharge current Imax(T). Subsequently, the corrected maximum discharge current data under these low SOC conditions and the maximum discharge current under non-low SOC conditions are transformed into Arrhenius coordinate space. That is, the reciprocal of the absolute temperature (1000 / T) is calculated as the horizontal axis, and the natural logarithm of the current ln(Imax) is calculated as the vertical axis.
[0057] Within this coordinate space, the data points exhibit a good linear relationship. A linear regression method is used to fit the equation of the straight line: .
[0058] Thus, the apparent activation energy under this SOC is obtained. (R is the gas constant) and pre-exponential factor Thus, the complete Arrhenius equation is established: .
[0059] Using this equation, the maximum discharge current in any temperature range under this SOC can be predicted.
[0060] The temperature fitting models for all the SOC points obtained above are then used to generate a continuous, high-precision two-dimensional power model covering the entire target temperature range and SOC range through bilinear interpolation or two-dimensional surface fitting methods. .
[0061] The generated power MAP is provided to the battery management system in the form of a lookup table to calculate the battery's peak discharge capacity in real time and implement corresponding power limiting strategies.
[0062] Thus, an invention of a battery power MAP construction method based on physical model and key point testing has been completed.
[0063] The above technical features constitute the preferred embodiment of this application, which has strong adaptability and the best implementation effect. Unnecessary technical features can be added or removed according to actual needs to meet the needs of different situations.
Claims
1. A method for constructing a battery power MAP based on a physical model and key point testing, characterized in that, The method includes the following steps: Within the target temperature range and target state of charge (SOC) range of the battery, key test points are selected according to preset principles; different rate pulses are applied to any key test point of the battery to form a working condition; the current sequence and voltage sequence of the battery at each acquisition time under each working condition are obtained; the maximum discharge current under each working condition in the temperature dimension is fitted to obtain the basic power value of the battery under each working condition. Using the filtered voltage and current sequences at each acquisition moment as input, the polarization impedance at each acquisition moment under each low SOC condition is calculated, and the concentration polarization resistance at each acquisition moment under each low SOC condition is calculated in combination with the reference polarization impedance; the low SOC condition is the condition where the SOC is less than or equal to a preset low threshold percentage. The concentration polarization resistance at the next acquisition time is predicted. All predicted values within a time window before each acquisition time are used as input to the genetic algorithm to obtain the optimal solution of the optimization sample. Based on the optimal solution, the low SOC power correction coefficient at each acquisition time under each low SOC condition is calculated. The optimization sample is a sequence composed of the concentration polarization resistance sequence at each acquisition time under each low SOC condition and the predicted value of the concentration polarization resistance at the next acquisition time. The base power value under low SOC conditions is corrected using a low SOC power correction factor; a two-dimensional power MAP of the battery is constructed based on the corrected base power values under all low SOC conditions and the base power values under non-low SOC conditions.
2. The battery power MAP construction method based on physical model and key point testing as described in claim 1, characterized in that, The fitting of the maximum discharge current under each operating condition in the temperature dimension includes: For the maximum discharge current value at key test points at different temperatures under the same SOC, the reciprocal of the absolute temperature is used as the abscissa and the natural logarithm of the maximum discharge current value is used as the ordinate. After transforming it to the Arrhenius coordinate space, a linear fit is performed to establish the current-temperature relationship model under that SOC. The slope of the linear equation obtained by the linear fit is used to calculate the apparent activation energy.
3. The battery power MAP construction method based on physical model and key point testing as described in claim 2, characterized in that, The key test point is a two-dimensional point consisting of a temperature point and a SOC point. The selection principles for the temperature point and SOC point include: The temperature points include extreme low temperature points, standard test temperature points, room temperature reference points, and extreme high temperature points; SOC points include minimum cutoff SOC point, low SOC point, normal SOC point, high SOC point, and full charge SOC point.
4. The battery power MAP construction method based on physical model and key point testing as described in claim 1, characterized in that, The current sequence and voltage sequence at each acquisition moment are respectively composed of current data and voltage data acquired within the time window set before each acquisition moment, arranged in chronological order of their acquisition time.
5. The battery power MAP construction method based on physical model and key point testing as described in claim 1, characterized in that, The concentration polarization resistance is the ratio of the polarization impedance at each acquisition moment under each low SOC condition to the reference polarization impedance at the corresponding acquisition moment.
6. The battery power MAP construction method based on physical model and key point testing as described in claim 5, characterized in that, The reference polarization impedance was calculated under the conditions of 25°C, 1C discharge rate, and 50% SOC.
7. The battery power MAP construction method based on physical model and key point testing as described in claim 1, characterized in that, The fitness function in the genetic algorithm is: Where min is the minimum value function, and I is the number of historical data collection moments. , These represent the measured value of the concentration polarization resistance at the i-th acquisition time within a sliding window prior to each acquisition time under each low SOC condition, and the optimized value of the concentration polarization resistance at the i-th acquisition time in the concentration polarization feature sequence at each acquisition time under each low SOC condition.
8. The battery power MAP construction method based on physical model and key point testing as described in claim 1, characterized in that, The calculation method for the low SOC power correction factor is as follows: The mean of the sequence corresponding to the optimal solution is used as the optimal concentration polarization feature at each acquisition time under each low SOC condition. The reciprocal of the sum of the optimal concentration polarization characteristic and the value of 1 is calculated and used as the low SOC power correction coefficient for each acquisition time under each low SOC condition.
9. The battery power MAP construction method based on physical model and key point testing as described in claim 8, characterized in that, The correction result of the base power value under the low SOC condition is: the product of the low SOC power correction coefficient and the base power value under the low SOC condition.
10. The battery power MAP construction method based on physical model and key point testing as described in claim 2, characterized in that, The construction of the battery's two-dimensional power MAP based on the corrected base power values under all low SOC conditions and the base power values under non-low conditions includes: For each low SOC condition, the power value obtained by correcting the key test points at different temperatures is combined with the rated minimum operating voltage of the reference voltage platform under that SOC condition, and the equivalent maximum discharge current is calculated by back-calculating the ratio between the corrected power value and the rated minimum operating voltage. The corrected maximum discharge current data under low SOC conditions and the maximum discharge current under non-low SOC conditions are transformed into Arrhenius coordinate space for linear fitting and interpolation to generate a two-dimensional power MAP of the battery.