Sodium-ion battery soe estimation method and storage medium

Abstract

The application provides a sodium-ion battery SOE estimation method and a storage medium; the sodium-ion battery SOE estimation method comprises the following steps: collecting battery state parameters and environmental state parameters in real time; determining an SOE estimation model according to the battery state parameters and the environmental state parameters based on a multi-level working condition classification model; and estimating an SOE value by using an improved extended Kalman filtering algorithm based on the SOE estimation model. The method considers voltage intervals and composite working conditions, uses the improved extended Kalman filtering algorithm, realizes high-precision estimation of the SOE of the sodium-ion battery under the composite working conditions, and can be applied to SOE estimation of other new batteries with nonlinear thermal effects.

Description

Technical Field

[0001] This invention relates to the field of new energy technology, specifically to a method for estimating SOE (Sodium-ion Battery Energy) and a storage medium. Background Technology

[0002] In battery management systems (BMS), accurate estimation of the battery's state of energy (SOE) is crucial for optimizing battery performance, ensuring safe system operation, and extending battery life. Current SOE estimation techniques are primarily based on lithium-ion battery characteristics, and common methods include the open-circuit voltage method, the ampere-hour integration method, and Kalman filtering. However, sodium-ion batteries, due to their unique electrochemical characteristics, suffer from the following problems:

[0003] First, when the voltage of a sodium-ion battery is below 2.75V, internal polarization intensifies and heat generation increases significantly. Traditional models do not consider the nonlinear thermal coupling effect in this range, resulting in excessive deviations in SOE estimation.

[0004] Secondly, under complex operating conditions (such as high-rate charging and discharging, and sudden temperature changes), a single model cannot adapt to the dynamically changing battery behavior. Summary of the Invention

[0005] In view of this, the present invention provides a method for estimating SOE of sodium-ion batteries, thereby solving one or more problems mentioned in the background art above.

[0006] To address the above problems, this invention provides a method for estimating the SOE of a sodium-ion battery, comprising: real-time acquisition of battery state parameters and environmental state parameters;

[0007] Based on the multi-level operating condition classification model, the SOE estimation model is determined according to the battery state parameters and the environmental state parameters.

[0008] Based on the SOE estimation model, the SOE value is estimated using an improved extended Kalman filter algorithm.

[0009] In the method described above, optionally, the step of determining the SOE estimation model based on the multi-level operating condition classification model, according to the battery state parameters and the environmental state parameters, includes:

[0010] Based on the voltage-dominated switching model, the voltage-dominated switching probability is calculated according to the battery state parameters.

[0011] Based on the joint probability model, the joint probability of the current operating condition and the typical operating condition is calculated according to the battery state parameters, the environmental state parameters and the preset typical operating condition center point.

[0012] The switching probability of the SOE estimation model is calculated based on the voltage-dominated switching probability and the combined probability of the current operating condition and the typical operating condition.

[0013] Based on the comprehensive smooth switching model, the SOE estimation model is determined.

[0014] In the method described above, optionally, the step of estimating the SOE value based on the SOE estimation model using the improved extended Kalman filter algorithm includes:

[0015] Based on the SOE estimation model, the SOE state vector, SOE state equation, and SOE observation vector are selected.

[0016] Based on the SOE estimate from the previous time step, the SOE estimate for the current time step is calculated. The formula for calculating the SOE estimate for the current time step is: SOE k - =SOE k-1 -I k ·V k-1 ·Δt / Etotal, where SOE k-1 The SOE estimate is the value of SOE at the previous moment. k - For the current SOE estimate, I k The current value at the current moment, V k-1 Δt is the voltage value at the previous moment, Δt is the time difference between the current moment and the previous moment, and Etotal is the total energy of the battery.

[0017] Based on the current SOE estimate, calculate the current battery terminal voltage estimate using the following formula: V k - =OCV(SOE) k - )-I k ·R int , where I k R is the current value at the current moment. int OCV(SOE) is the internal resistance of the battery. k - ) is the OCV-SOE relational function, SOE k - V is the estimated SOE value at the current moment. k - This is the estimated terminal voltage at the current moment;

[0018] The formula for predicting covariance is: P k - =P k +Q, where P k Let P be the initial covariance, Q be the process noise, and P be the initial covariance. k - For covariance;

[0019] The observed Jacobian is calculated using the following formula: H k =dOCV / dSOE|SOE k - , which represents the partial derivative of the open-circuit voltage OCV with respect to SOE calculated at the SOE estimate at the current moment;

[0020] The Kalman gain is calculated using the following formula: K k =P k - ·H k / (H k ·P k - ·H k +R), where P k - For the covariance estimate, H k R is the observation Jacobian, and R is the observation noise covariance.

[0021] The residual is calculated using the formula: y k =V meas -V k - , where V meas The actual measured terminal voltage value, V k - The estimated terminal voltage at the current moment is y. k For residuals;

[0022] The corrected SOE value is calculated using the following formula: SOE k =SOE k - +K k ·(y k / H k ), where SOE k - K is the estimated SOE value at the current moment. k Let y be the Kalman gain. k For the residual, H k For the observed Jacobi;

[0023] The updated covariance is calculated using the formula: P k = (1-K) k ·H k )·P k - K k The Kalman gain, H k For the observed Jacobian, P k - Let be the covariance.

[0024] In the method described above, optionally, the calculation formula for the voltage-dominated switching model is as follows:

[0025]

[0026] Wherein, V is the battery terminal voltage. This is the boundary voltage; For extreme low voltage warning voltage, l is the length scale parameter, w is the weighting coefficient, σ compresses the output to [0, 1], P V (V) represents the voltage-dominated switching probability;

[0027] The calculation formula for the joint probability model is as follows:

[0028]

[0029] Among them, z other Let Cm be the current operating condition vector, and α be the center point of the typical operating condition vector. m The output is compressed to [0, 1], where m is the number of typical operating conditions, and P... other (z other ) represents the joint probability of the current operating condition and the typical operating condition;

[0030] The calculation formula for the SOE estimation model, based on the voltage-dominated switching probability and the joint probability of the current operating condition and the typical operating condition, is as follows:

[0031] P switch (z)=P V (V)·P other (z other )

[0032] Among them, P V (V) represents the voltage-dominated switching probability, P other (z other P represents the joint probability of the current operating condition and the typical operating condition. switch (z) represents the SOE estimation model switching probability;

[0033] The comprehensive smooth switching model includes:

[0034]

[0035] Among them, P switch (z) represents the switching probability of the SOE estimation model, and Model represents the SOE estimation model.

[0036] In the method described above, optionally, the current operating condition vector is a multi-dimensional feature vector, expressed as:

[0037]

[0038] Among them, I t Q is the current. nom For the rated energy of the battery, T amb,t For ambient temperature, N cycle This represents the number of battery cycles.

[0039] In the method described above, optionally, the SOE state vector, SOE state equation, and observation vector include:

[0040] SOE state vector in high-heat zone, SOE state equation in high-heat zone, observation vector in high-heat zone, SOE state vector in low-heat zone, SOE state equation in low-heat zone, and observation vector in low-heat zone;

[0041] The expression for the SOE state vector in the high-heat range is: Where T is the battery terminal temperature, U is the battery terminal voltage, and R is the battery terminal temperature. int The internal resistance of the battery, For temperature gradient;

[0042] The equation of state for the high-heat zone SOE is:

[0043]

[0044] The expression for the observation vector of the high-fever zone is: y = [V meas T surface Q diss ], where V meas For the terminal voltage, T surface Q is the surface temperature of the electrode post. diss For heat dissipation rate;

[0045] The SOE state vector expression for the low-heating range is: x SOE =[SOE, T, U, R] int ] T Where T is the battery terminal temperature, U is the battery terminal voltage, and R is the battery terminal voltage. int The internal resistance of the battery;

[0046] The equation of state for the low-heat zone SOE is:

[0047]

[0048] The expression for the observation vector of the low-heat zone is: y = [V meas T surface ], where V meas For terminal voltage, T surface The temperature is the surface temperature of the electrode.

[0049] In the method described above, optionally, the center point of the typical working condition vector includes:

[0050] The center point of the first typical working condition vector is: c1 = [0.5, 25, 1.32]. T Where 0.5 represents 0.5 times the battery's rated power, 25 represents an ambient temperature of 25°C, and 1.32 represents the number of cycles N. cycle The result is 20 times, calculated using the formula ln(1+N). cycle ) / 10 is calculated to obtain;

[0051] The center point of the second typical working condition vector is: c2 = [1.0, 35, 3.70]. T Where 1.0 represents 1.0 times the battery's rated power, 35 represents an ambient temperature of 35℃, and 3.70 represents the number of cycles N. cycle For 5000 iterations, use the formula ln(1+N) cycle ) / 10 is calculated to obtain;

[0052] The center point of the third typical working condition vector is: c3 = [0.7, 30, 2.30]. T Where 0.7 represents 0.7 times the battery's rated power, 30 represents an ambient temperature of 30℃, and 2.30 represents the number of cycles N. cycle For 200 iterations, use the formula ln(1+N) cycle ) / 10 is calculated.

[0053] In the method described above, optionally, the battery state parameters include at least battery terminal voltage, battery current, and battery temperature, and the environmental state parameters include at least ambient temperature.

[0054] In the method described above, optionally, the OCV-SOE relationship function is obtained by calibrating and fitting the measured values ​​of SOE and OCV obtained experimentally.

[0055] The present invention also discloses a computer storage medium, characterized in that the computer storage medium stores a computer executable program, which, when executed by a processor, implements the sodium-ion battery SOE estimation method as described above.

[0056] The SOE estimation method for sodium-ion batteries of the present invention reduces the SOE estimation error in the voltage range by integrating voltage range and combined operating conditions adaptively estimating SOE of sodium-ion batteries, supports seamless switching of combined operating conditions, and avoids battery failure caused by sudden changes in operating conditions in traditional methods. The present invention uses an improved extended Kalman filter algorithm for SOE estimation, which takes into account the nonlinear thermal coupling effect of the battery, thereby reducing the SOE estimation error. Attached Figure Description

[0057] The disclosure of this invention will become more apparent from the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings:

[0058] Figure 1 This is a schematic diagram of a sodium-ion battery SOE estimation method disclosed in an embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram of the multi-level working condition classification model disclosed in an embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram illustrating the process of estimating SOE values ​​using an improved extended Kalman filter algorithm as disclosed in an embodiment of the present invention. Detailed Implementation

[0061] Referring to the accompanying drawings and specific embodiments, the process, working principle, characteristics and advantages of the sodium-ion battery SOE estimation method and storage medium of the present invention will be described below by way of example. However, all descriptions should not be construed as limiting the present invention in any way.

[0062] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection. They can refer to a mechanical connection or an electrical connection. They can refer to a direct connection or an indirect connection through an intermediate medium, and they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0063] Furthermore, for any single technical feature described or implied in the embodiments described herein, or any single technical feature shown or implied in the accompanying drawings, the present invention still allows for any combination or deletion among these technical features (or their equivalents) without any technical obstacle. Therefore, these further embodiments according to the present invention should also be considered within the scope of this description.

[0064] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature.

[0065] Please see Figures 1 to 3 , Figure 1 This is a flowchart illustrating a method for estimating the SOE of a sodium-ion battery according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the multi-level working condition classification model disclosed in an embodiment of the present invention. Figure 3 This is a schematic diagram illustrating the process of estimating SOE values ​​using an improved extended Kalman filter algorithm, as disclosed in an embodiment of the present invention. A method for estimating SOE in a sodium-ion battery according to an embodiment of the present invention may include the following steps:

[0066] Step 100: Real-time acquisition of battery status parameters and environmental status parameters.

[0067] In this embodiment, the battery status parameters include at least the battery terminal voltage, battery current, and battery temperature, and the environmental status parameters include at least the ambient temperature.

[0068] Step 101: Based on the multi-level operating condition classification model, determine the SOE estimation model according to the battery state parameters and the environmental state parameters.

[0069] In this embodiment, the multi-level working condition classification model includes the following steps:

[0070] Step 200: Based on the voltage-dominated switching model, calculate the voltage-dominated switching probability according to the battery state parameters.

[0071] In this embodiment, the calculation formula for the voltage-dominated switching model is as follows:

[0072]

[0073] Where V is the battery terminal voltage. This is the boundary voltage; For extreme low voltage warning voltage, l is the length scale parameter, w is the weighting coefficient, σ compresses the output to [0, 1], P V (V) represents the voltage-dominated switching probability. σ can be a sigmoid compression function used to limit the output to values ​​within the range [0, 1].

[0074] Step 201: Based on the joint probability model, calculate the joint probability of the current operating condition and the typical operating condition according to the battery state parameters, environmental state parameters and the preset typical operating condition center point.

[0075] In this embodiment, the calculation formula for the joint probability model is:

[0076]

[0077] Among them, z other Let Cm be the current operating condition vector, and α be the center point of the typical operating condition vector. m The output is compressed to [0, 1], where m is the number of typical operating conditions, and P... other (z other ) represents the joint probability of the current operating condition and the typical operating condition.

[0078] Step 202: Calculate the switching probability of the SOE estimation model based on the voltage-dominated switching probability and the joint probability of the current operating condition and the typical operating condition.

[0079] In this embodiment, the calculation formula for the SOE estimation model is as follows:

[0080] P switch (z)=P V (V)·P other (z other )

[0081] Among them, P V (V) represents the voltage-dominated switching probability, P other (z other Let P be the joint probability of the current operating condition and the typical operating condition. switch (z) represents the SOE-estimated model switching probability. Typical operating condition vector center points include:

[0082] The center point of the first typical working condition vector is: c1 = [0.5, 25, 1.32]. T Where 0.5 represents 0.5 times the battery's rated power, 25 represents an ambient temperature of 25°C, and 1.32 represents the number of cycles N. cycle The result is 20 times, calculated using the formula ln(1+N). cycle ) / 10 is calculated to obtain;

[0083] The center point of the second typical working condition vector is: c2 = [1.0, 35, 3.70]. T Where 1.0 represents 1.0 times the battery's rated power, 35 represents an ambient temperature of 35℃, and 3.70 represents the number of cycles N. cycle For 5000 iterations, use the formula ln(1+N) cycle ) / 10 is calculated to obtain;

[0084] The center point of the vector for the third typical working condition is: c3 = [0.7, 30, 2.30]. T Where 0.7 represents 0.7 times the battery's rated power, 30 represents an ambient temperature of 30℃, and 2.30 represents the number of cycles N. cycle For 200 iterations, use the formula ln(1+N) cycle ) / 10 is calculated.

[0085] Step 203: Determine the SOE estimation model based on the comprehensive smooth switching model.

[0086] In this embodiment, the comprehensive smooth handover model includes:

[0087]

[0088] Among them, P switch (z) represents the switching probability of the SOE estimation model, and Model represents the SOE estimation model.

[0089] The SOE estimation method for sodium-ion batteries of the present invention reduces the SOE estimation error in the voltage range by integrating voltage range and combined operating conditions for adaptive estimation of sodium-ion battery SOE, supports seamless switching of combined operating conditions, and avoids battery failure caused by sudden changes in operating conditions in traditional methods.

[0090] Step 102: Based on the SOE estimation model, the SOE value is estimated using the improved extended Kalman filter algorithm.

[0091] In this embodiment, step 102 includes the following steps:

[0092] Step 300: Select the SOE state vector, SOE state equation, and SOE observation vector based on the SOE estimation model.

[0093] In this embodiment, the SOE state vector, SOE state equation, and observation vector for the high-heat range and the low-heat range are given.

[0094] The SOE state vector expression for the low-heat range in this embodiment is: x SOE =[SOE, T, U, R] int ] T Where T is the battery terminal temperature, U is the battery terminal voltage, and R is the battery terminal temperature. int This refers to the battery's internal resistance.

[0095] Modeling the rate of change of each state variable in the aforementioned low-heat SOE state vector yields the state equation for the low-heat SOE:

[0096]

[0097] in, E represents the rate of change of SOE. total I represents the total energy of the battery, and I·U represents the power. I is the rate of temperature change. 2 R int For Ohmic heat, |I|η is polarization heat, hA(TT) amb For heat dissipation, C p Where is the heat capacity, and w is the process noise; The rate of change of voltage. This represents the rate of change of internal resistance.

[0098] Define the observation vector for the low-heat range: y = [V meas T surface ], where Vmeas For the measured terminal voltage, T surface The observation vector for the low-heating range, representing the pole surface temperature, is used in Kalman filtering. By comparing it with the SOE prediction, the SOE state vector is corrected, thus achieving accurate estimation of SOE in the low-heating range.

[0099] The SOE state vector expression for the high-heat-generating region in this embodiment is: Where T is the battery terminal temperature, U is the battery terminal voltage, and R is the battery terminal temperature. int For battery internal resistance, This represents the temperature gradient. Compared to the SOE state vector in the low-heat region, the SOE state vector in the high-heat region incorporates the temperature gradient. This temperature gradient is used to characterize the internal heat distribution of the battery, such as the temperature difference between the battery's bottom side and the terminal posts.

[0100] The SOE state equation for the high-heat-generating region in this embodiment is:

[0101]

[0102] Compared with the SOE equation of state in the low-heat range, The rate of change characterizes the rate of change of the internal temperature of the battery, where τ is a time constant and Q is a constant. gen For the heat generation equation, k th The thermal conductivity coefficient,

[0103] The observation vector expression for the high-heat zone in this embodiment is: y = [V meas T surface Q diss ], where V meas For terminal voltage, T surface Q is the surface temperature of the electrode column. diss This refers to the heat dissipation rate.

[0104] The difference between the SOE state vector in the low-heating range and the SOE state vector in the high-heating range is that, since the temperature distribution inside the battery is approximately uniform in the low-heating range, there is no need to consider the temperature difference in the space inside the battery, thus omitting the temperature gradient factor, which simplifies the model and reduces computational complexity. In contrast, the temperature distribution inside the battery is non-uniform in the high-heating range, so the temperature gradient factor needs to be considered.

[0105] Step 301: Calculate the SOE estimate at the current time based on the SOE estimate at the previous time.

[0106] In this embodiment, the formula for calculating the SOE estimate at the current time is: SOE k - =SOE k-1 -I k ·V k-1·Δt / Etotal, where SOE k-1 For the previous moment SOE, SOE k - For the current SOE estimate, I k The current value at the current moment, V k-1 Let be the voltage value at the previous moment, Δt be the time difference between the current moment and the previous moment, and Etotal be the total energy of the battery. The first part of this formula represents the SOE value at the previous moment, and the second part, the numerator I... k ·V k-1 Δt represents the energy consumed by the battery from the previous moment to the current moment, and the denominator represents the total battery energy. Therefore, this formula generally predicts the SOE by subtracting the energy consumed by the battery at the current moment from the SOE at the previous moment. Where I... k A value greater than 0 indicates positive energy consumption, and SOE decreases. If I k If the value is less than 0, it means that the energy consumption is negative and SOE increases.

[0107] Step 302: Calculate the estimated battery terminal voltage at the current moment based on the estimated SOE value at the current moment.

[0108] In this embodiment, the formula for calculating the terminal voltage estimate is: V k - =OCV(SOE) k - )-I k ·R int , where I k R is the current value at the current moment. int OCV(SOE) is the internal resistance of the battery. k - The function represents the OCV-SOE relationship, where OCV is the battery open-circuit voltage and SOE is the open-circuit voltage. k - V is the estimated SOE value at the current moment. k - This is the estimated terminal voltage at the current moment. The calculation formula consists of two parts: OCV(SOE) k - ) is the battery open-circuit voltage OCV, I k ·R int It is the internal resistance voltage drop. In this embodiment, the OCV-SOE relationship function is obtained by calibrating and fitting the measured values ​​of SOE and OCV obtained experimentally.

[0109] Step 303: Predict covariance.

[0110] The calculation formula is: P k - =P k +Q, where Pk Let Q be the initial covariance and Q be the process noise. In the Kalman filter algorithm, process noise describes the uncertainty of model prediction. If Q is large (e.g., battery aging causes large parameter fluctuations), it means that the confidence in the predicted state is reduced, and subsequent measurement updates will rely more on sensor data.

[0111] Step 304: Calculate the observed Jacobian.

[0112] The calculation formula is: H k =dOCV / dSOE|SOE k - , represents the partial derivative of the open-circuit voltage OCV with respect to the SOE at the current SOE estimate, and describes the sensitivity of the open-circuit voltage OCV to the energy state SOE.

[0113] Step 305: Calculate the Kalman gain.

[0114] The calculation formula is: K k =P k - ·H k / (H k ·P k - ·H k +R), where P k - For the covariance estimate, H k Let R be the observation Jacobian, R be the observation noise covariance, and P be the observation noise covariance. k - The uncertainty of the predicted state is represented by R, and the uncertainty of the observation noise is represented by R. The uncertainty of the predicted state and the uncertainty of the observation noise are balanced by the Kalman gain, thereby determining the weight of the observation value on the state update.

[0115] Step 306: Calculate the residuals.

[0116] The calculation formula is: y k =V meas -V k - , where V meas The actual measured terminal voltage value, V k - This is the estimated terminal voltage at the current moment. The residual represents the difference between the actual measured terminal voltage and the estimated terminal voltage at the previous moment, and is used to correct the SOE estimate.

[0117] Step 307: Calculate the corrected SOE value.

[0118] The calculation formula is: SOE k =SOE k -+K k ·(y k / H k ), where SOE k - K is the estimated SOE value at the current moment. k For Kalman gain, y k For the residual, H k To observe the Jacobian. This formula is derived by observing the Jacobian H. k The residual y k Convert to SOE difference, then use Kalman gain K k By balancing the weights of the model predictions and observation data, optimal estimation can be achieved, making it suitable for nonlinear observation models of battery SOE.

[0119] Step 308: Update the covariance for prediction at the next time step.

[0120] The calculation formula is: P k = (1-K) k ·H k )·P k - K k The Kalman gain, H k For the observed Jacobian, P k - This is the estimated value for covariance.

[0121] This invention employs an improved extended Kalman filter algorithm that takes into account the nonlinear thermal coupling effect of the battery, thereby reducing the error in SOE estimation.

[0122] This invention also provides a computer storage medium storing a computer-executable program, which, when executed by a processor, implements the sodium-ion battery SOE estimation method of any embodiment of this invention.

[0123] The technical scope of this invention is not limited to the contents of the above specification. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the scope of this invention.

Claims

1. A method for estimating the SOE of a sodium-ion battery, characterized in that, The method includes at least the following steps: Real-time acquisition of battery status parameters and environmental status parameters; Based on the multi-level operating condition classification model, the SOE estimation model is determined according to the battery state parameters and the environmental state parameters. Based on the SOE estimation model, the SOE value is estimated using an improved extended Kalman filter algorithm.

2. The method as described in claim 1, characterized in that, The SOE estimation model, based on the multi-level operating condition classification model and according to the battery state parameters and the environmental state parameters, includes: Based on the voltage-dominated switching model, the voltage-dominated switching probability is calculated according to the battery state parameters. Based on the joint probability model, the joint probability of the current operating condition and the typical operating condition is calculated according to the battery state parameters, the environmental state parameters and the preset typical operating condition center point. The switching probability of the SOE estimation model is calculated based on the voltage-dominated switching probability and the combined probability of the current operating condition and the typical operating condition. Based on the comprehensive smooth switching model, the SOE estimation model is determined.

3. The method as described in claim 1, characterized in that, The estimation of SOE values ​​based on the SOE estimation model, using the improved extended Kalman filter algorithm, includes: Based on the SOE estimation model, the SOE state vector, SOE state equation, and SOE observation vector are selected. Based on the SOE estimate from the previous time step, the SOE estimate for the current time step is calculated. The formula for calculating the SOE estimate for the current time step is: SOE k - =SOE k-1 - I k ·V k-1 ·Δt / Etotal, where SOE k-1 The SOE estimate is the value of SOE at the previous moment. k - For the current SOE estimate, I k The current value at the current moment, V k-1 Δt is the voltage value at the previous moment, Δt is the time difference between the current moment and the previous moment, and Etotal is the total energy of the battery. Based on the current SOE estimate, calculate the current battery terminal voltage estimate using the following formula: V k - =OCV(SOE k - ) - I k ·R int , where I k R is the current value at the current moment. int OCV(SOE) is the internal resistance of the battery. k - ) is the OCV-SOE relational function, SOE k - V is the estimated SOE value at the current moment. k - This is the estimated terminal voltage at the current moment; The formula for predicting covariance is: P k - = P k + Q, where P k Let P be the initial covariance, Q be the process noise, and P be the initial covariance. k - For covariance; The observed Jacobian is calculated using the following formula: H k =dOCV / dSOE|SOE k - , which represents the partial derivative of the open-circuit voltage OCV with respect to SOE calculated at the SOE estimate at the current moment; The Kalman gain is calculated using the following formula: K k = P k - ·H k / (H k ·P k - ·H k + R), where P k - For the covariance estimate, H k R is the observation Jacobian, and R is the observation noise covariance. The residual is calculated using the formula: y k = V meas - V k - , where V meas The actual measured terminal voltage value, V k - The estimated terminal voltage at the current moment is y. k For residuals; The corrected SOE value is calculated using the following formula: SOE k = SOE k - + K k ·(y k / H k ), where SOE k - K is the estimated SOE value at the current moment. k Let y be the Kalman gain. k For the residual, H k For the observed Jacobi; The updated covariance is calculated using the formula: P k = (1 - K k · H k ) ·P k - K k The Kalman gain, H k For the observed Jacobian, P k - Let be the covariance.

4. The method as described in claim 2, characterized in that, The calculation formula for the voltage-dominated switching model is as follows: ， Wherein, V is the battery terminal voltage. This is the boundary voltage; The voltage is the extreme low voltage warning voltage, l is the length scale parameter, and w is the weighting coefficient. Compress the output to [0, 1]. Voltage-dominated switching probability; The calculation formula for the joint probability model is as follows: ， in, Let Cm be the current operating condition vector, and Cm be the center point of the typical operating condition vector. The output is compressed to [0, 1], where m is the number of typical operating conditions. The probability is the combined probability of the current operating condition and the typical operating condition. The calculation formula for the SOE estimation model, based on the voltage-dominated switching probability and the joint probability of the current operating condition and the typical operating condition, is as follows: ， in, The voltage-dominated switching probability. The probability is the joint probability of the current operating condition and the typical operating condition. Estimate the model switching probability for SOE; The comprehensive smooth switching model includes: ， in, The model switching probability is estimated for SOE, and Model is the SOE estimation model.

5. The method as described in claim 4, characterized in that, The current operating condition vector is a multi-dimensional feature vector, expressed as: ， in, For current, For the battery's rated energy, For ambient temperature, This represents the number of battery cycles.

6. The method as described in claim 3, characterized in that, The SOE state vector, SOE state equation, and observation vector include: SOE state vector in high-heat zone, SOE state equation in high-heat zone, observation vector in high-heat zone, SOE state vector in low-heat zone, SOE state equation in low-heat zone, and observation vector in low-heat zone; The SOE state vector expression for the high-heat zone is: x SOE =[SOE, T, U, R] int ,∇T] T Where T is the battery terminal temperature, U is the battery terminal voltage, and R is the battery terminal temperature. int Let ∇T be the internal resistance of the battery, and ∇T be the temperature gradient. The equation of state for the high-heat zone SOE is: ， The expression for the observation vector of the high-fever zone is: y=[V meas T surface Q diss ], where V meas For the terminal voltage, T surface Q is the surface temperature of the electrode post. diss For heat dissipation rate; The SOE state vector expression for the low-heating range is: x SOE =[SOE, T, U, R] int ] T Where T is the battery terminal temperature, U is the battery terminal voltage, and R is the battery terminal voltage. int The internal resistance of the battery; The equation of state for the low-heat zone SOE is: ， The observation vector expression for the low-heat range is: y=[V meas T surface ], where V meas For terminal voltage, T surface The temperature is the surface temperature of the electrode.

7. The method as described in claim 4, characterized in that, The typical working condition vector center point includes: The center point of the first typical working condition vector is: c1=[0.5,25,0.3]. T Where 0.5 represents 0.5 times the battery's rated power, 25 represents an ambient temperature of 25℃, and 0.6 represents the number of cycles. For 20 times, using the formula Calculated; The center point of the second typical working condition vector is: c2=[1.0,35,0.8] T Where 1.0 represents 1.0 times the battery's rated power, 35 represents an ambient temperature of 35℃, and 0.8 represents the number of cycles. For 5000 times, using the formula Calculated; The center point of the third typical working condition vector is: c3=[0.7,30,0.5] T Where 0.7 represents 0.7 times the battery's rated power, 30 represents an ambient temperature of 30℃, and 0.8 represents the number of cycles. For 200 times, using the formula Calculated.

8. The method as described in claim 1, characterized in that, The battery status parameters include at least battery terminal voltage, battery current, and battery temperature, and the environmental status parameters include at least ambient temperature.

9. The method as described in claim 3, characterized in that, The OCV-SOE relationship function was obtained by calibrating and fitting the measured values ​​of SOE and OCV obtained through experiments.

10. A computer storage medium, characterized in that, The computer storage medium stores a computer-executable program, which, when executed by a processor, implements the sodium-ion battery SOE estimation method according to any one of claims 1 to 9.

Images