Residual electric quantity estimation method and device, electronic equipment and storage medium

Abstract

The invention provides a residual electric quantity estimation method and device, electronic equipment and a storage medium, and relates to the technical field of power system energy storage control. The method comprises the following steps: determining a Kalman filter model according to target model parameters corresponding to a target battery; according to the innovation of a Kalman filter in the Kalman filter model, determining a corresponding relation among a target cross-correlation function, a target self-correlation function and an open-circuit voltage-residual electric quantity curve deviation; and determining the residual electric quantity of the target battery according to the corresponding relation. According to the method, accurate SOC estimation can be carried out on the battery under the condition that the open-circuit voltage-residual electric quantity curve has errors, so that the workload of continuously updating the open-circuit voltage-residual electric quantity curve by the battery management system is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power system energy storage control, and in particular to a method, device, electronic equipment and storage medium for estimating residual power. Background Art

[0002] Lithium-ion batteries are widely used in electric vehicles and energy storage systems, among which lithium iron phosphate batteries have gradually become the first choice for battery systems due to their high safety, long cycle life and low cost. The accuracy of battery remaining capacity (State of Charge, SOC) estimation will directly affect the evaluation of battery power and remaining energy. Accurate SOC estimation can also effectively avoid overcharging / over-discharging of batteries. At present, the relevant research on SOC estimation is mainly focused on model-based methods. Model-based methods comprehensively consider the ampere-hour integration method and the open circuit voltage method, and have low requirements on the calculation and storage capabilities of the battery management system BMS.

[0003] However, the model-based SOC estimation algorithm is heavily dependent on an accurate open circuit voltage-remaining charge OCV-SOC curve. When the battery open circuit voltage-remaining charge curve changes with battery aging and operating temperature fluctuations, the remaining charge estimation will be inaccurate, and constantly updating the open circuit voltage-remaining charge curve will result in a large workload. Summary of the invention

[0004] The present invention provides a method, device, electronic device and storage medium for estimating remaining power, so as to solve the defect in the prior art that the SOC estimation algorithm is heavily dependent on the accurate open circuit voltage-remaining power OCV-SOC curve, resulting in a large workload. The Kalman filter model is determined according to the target model parameters corresponding to the target battery, and the remaining power is determined according to the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation of the Kalman filter model. The SOC of the battery can be accurately estimated in the case of an open circuit voltage-remaining power curve error, thereby reducing the workload of continuously updating the open circuit voltage-remaining power curve.

[0005] The present invention provides a method for estimating remaining power, comprising the following steps: Determine a Kalman filter model according to target model parameters corresponding to a target battery, wherein the target model parameters corresponding to the target battery are obtained by performing model parameter identification on a circuit equivalent model corresponding to the target battery, and the target model parameters corresponding to the target battery are parameters of a Kalman filter in the Kalman filter model; Determine the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation according to the new information of the Kalman filter in the Kalman filter model, wherein the new information of the Kalman filter is used to characterize the difference between the actual measurement value and the predicted measurement value corresponding to the Kalman filter; The remaining capacity of the target battery is determined according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation.

[0006] According to a method for estimating remaining power provided by the present invention, the step of determining a Kalman filter model according to target model parameters corresponding to a target battery includes: Determine an equivalent circuit model corresponding to the target battery, wherein the equivalent circuit model includes a corresponding relationship between an ohmic internal resistance, a polarization resistance, a polarization capacitance, a terminal voltage, an open circuit voltage, and a current; Inputting the target measurement value of the target battery into the equivalent circuit model to obtain the target model parameters, wherein the target measurement value includes a target terminal voltage and a target current, and the target model parameters include a target ohmic internal resistance, a target polarization resistance, and a target polarization capacitance; The target model parameters corresponding to the target battery are used as input quantities of the Kalman filter model to perform modeling, thereby obtaining the Kalman filter model.

[0007] The present invention provides a method for estimating remaining power, wherein the target measurement value of the target battery is input into the equivalent circuit model to obtain the target model parameter, including: Discretizing the continuous-time state equation corresponding to the equivalent circuit model to obtain a discrete state equation of the target polarization voltage; According to the discrete state equation of the target polarization voltage, a discrete variation equation of the target terminal voltage is obtained; Simplifying the discrete variation equation of the target terminal voltage according to the discrete variation equation of the target open circuit voltage to obtain a measurement equation of the equivalent circuit model, wherein the measurement equation includes parameters to be estimated, and the parameters to be estimated are used to characterize the time-varying condition of the target model parameters; The adaptive recursive least square method including a forgetting factor is used to estimate the value of the parameter to be estimated in the measurement equation to obtain the target model parameter.

[0008] The present invention provides a method for estimating remaining power, wherein the method uses an adaptive recursive least square method containing a forgetting factor to estimate the value of the parameter to be estimated in the measurement equation to obtain the target model parameter, including: Determine the initial covariance of the parameters to be estimated and the covariance at the target time; According to the initial covariance of the parameter to be estimated, the covariance at the target time, and the gain of the equivalent circuit model at the target time, the value of the parameter to be estimated in the measurement equation is estimated by adaptively adjusting the forgetting factor to obtain the value of the parameter to be estimated; The target model parameters are obtained according to the values ​​of the parameters to be estimated.

[0009] The present invention provides a method for estimating a remaining capacity, wherein the method determines the corresponding relationship between a target cross-correlation function, a target autocorrelation function and an open circuit voltage-remaining capacity curve deviation according to the new information of the Kalman filter in the Kalman filter model, including: By analyzing the new information of the Kalman filter in real time, the relationship between the new information covariance and the currently used open circuit voltage-remaining power curve and the real open circuit voltage-remaining power curve is obtained; According to the relationship between the new information covariance and the currently used open circuit voltage-remaining power curve and the real open circuit voltage-remaining power curve, the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation is determined.

[0010] The present invention provides a method for estimating a remaining capacity, wherein the remaining capacity of the target battery is determined according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation, comprising: Setting multiple Kalman filters, each filter having a different measurement equation; Dynamically adjusting the parameters of the multiple Kalman filters according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation; Based on the Bayesian analysis results corresponding to the multiple Kalman filters, a Kalman filter whose confidence meets a preset confidence requirement is selected as an optimal filter, and an output value of the optimal filter is the remaining power of the target battery.

[0011] The present invention provides a method for estimating remaining power, wherein based on the Bayesian analysis results corresponding to the multiple Kalman filters, a Kalman filter whose confidence meets a preset confidence requirement is selected as the optimal filter, comprising: After each sampling period, the parameter probability of each Kalman filter is updated, and after the estimation period, the Kalman filter with the highest parameter probability is determined as the optimal filter.

[0012] The present invention also provides a device for estimating remaining power, comprising the following modules: A model determination module, used to determine a Kalman filter model according to target model parameters corresponding to a target battery, wherein the target model parameters corresponding to the target battery are obtained by performing model parameter identification on a circuit equivalent model corresponding to the target battery, and the target model parameters corresponding to the target battery are parameters of a Kalman filter in the Kalman filter model; A relationship determination module, used to determine the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation according to the new information of the Kalman filter in the Kalman filter model, wherein the new information of the Kalman filter is used to characterize the difference between the actual measurement value and the predicted measurement value corresponding to the Kalman filter; The power determination module is used to determine the remaining power of the target battery according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the remaining power estimation method as described in any one of the above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the remaining power estimation method as described in any one of the above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the remaining power estimation method as described above is implemented.

[0016] The remaining power estimation method, device, electronic device and storage medium provided by the present invention determine the Kalman filter model according to the target model parameters corresponding to the target battery, and determine the remaining power according to the correspondence between the target cross-correlation function, target autocorrelation function and open circuit voltage-remaining power curve deviation of the Kalman filter model. It can accurately estimate the SOC of the battery in the case of an open circuit voltage-remaining power curve error, thereby reducing the workload of constantly updating the open circuit voltage-remaining power curve. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flow chart of the remaining power estimation method provided by the present invention.

[0019] Figure 2 It is a flow chart of the method for determining the Kalman filter model provided by the present invention.

[0020] Figure 3 It is a schematic diagram of the equivalent circuit model provided by the present invention.

[0021] Figure 4 It is a flowchart of the method for obtaining the corresponding relationship provided by the present invention.

[0022] Figure 5 It is a schematic diagram of a flow chart of determining the remaining power of a target battery provided by the present invention.

[0023] Figure 6 It is a schematic diagram of the overall flow of the remaining power estimation method provided by the present invention.

[0024] Figure 7 It is a structural schematic diagram of the remaining power estimation device provided by the present invention.

[0025] Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Lithium-ion batteries are widely used in electric vehicles and energy storage systems, among which lithium iron phosphate batteries have gradually become the first choice for battery systems due to their high safety, long cycle life, and low cost. In a battery system, SOC is a key indicator that determines its safety and reliability. Accurate SOC estimation results are conducive to the battery management system to make correct charge and discharge management strategies to avoid battery overcharge / over-discharge in the system. At the same time, the battery energy storage energy management system also makes scheduling and power allocation strategies based on the SOC estimation results. Accurate SOC results will help solve the inconsistency of system power and avoid the emergence of the "short board effect". The accuracy of the battery remaining power (State of Charge, SOC) estimation will directly affect the evaluation of battery power and remaining energy. Accurate SOC estimation can also effectively avoid battery overcharge / over-discharge.

[0028] Among the SOC estimation methods for lithium-ion batteries, the ampere-hour integration method is the simplest, but it cannot correct the initial estimation error and will bring cumulative errors. The open circuit voltage method realizes SOC estimation by finding the open circuit voltage-remaining power meter, but requires the battery to be stationary for a long time, which is not suitable for online estimation. The data-driven method requires a large amount of sample data under various operating conditions for a specific battery. When the sample data is insufficient, its accuracy is low and for lithium iron phosphate batteries, the SOC estimation results generally fluctuate greatly. At the same time, it is difficult to embed in the battery management system BMS. At present, the research on SOC estimation mainly focuses on model-based methods and methods that integrate data-driven and model. The model-based method comprehensively considers the ampere-hour integration method and the open circuit voltage method, and has low requirements for BMS computing and storage capabilities. However, the OCV-SOC curve of lithium iron phosphate batteries has a long plateau period, and the OCV-SOC curve error, initial error, and measurement equipment noise will have a significant impact on the estimation results. Model-based methods are usually evaluated from the perspectives of estimation accuracy, convergence speed, and robustness.

[0029] Model-based SOC estimation algorithms rely heavily on accurate OCV-SOC curves, whether obtained through multiple experiments or fast estimation methods. However, this approach imposes a huge workload on battery systems containing a large number of cells. Therefore, it is very necessary to study a method to accurately estimate the SOC of lithium iron phosphate batteries under the deviation of the open circuit voltage-remaining capacity curve.

[0030] In view of this, an embodiment of the present invention provides a method for estimating remaining power, which determines a Kalman filter model according to target model parameters corresponding to a target battery; determines the corresponding relationship between a target cross-correlation function, a target autocorrelation function and an open circuit voltage-remaining power curve deviation according to the new information of the Kalman filter in the Kalman filter model; determines the remaining power of the target battery according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation. This method can accurately estimate the SOC of the battery even when there is an error in the open circuit voltage-remaining power curve, thereby reducing the workload of the battery management system to continuously update the open circuit voltage-remaining power curve.

[0031] The technical solutions in the embodiments of the present invention will be described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0032] Figure 1: is a flow chart of the remaining power estimation method provided by the present invention. The remaining power estimation method can be applied to electronic devices, which can be various types of devices with information processing capabilities during implementation. For example, the electronic device can include a personal computer, a laptop computer, a PDA or a server, etc.; the electronic device can also be a mobile terminal, for example, the mobile terminal can include a mobile phone, a car computer, a tablet computer or a projector, etc. Figure 1 As shown, the method may include the following steps 101 to 103: Step 101: determining a Kalman filter model according to target model parameters corresponding to a target battery, wherein the target model parameters corresponding to the target battery are obtained by performing model parameter identification on a circuit equivalent model corresponding to the target battery, and the target model parameters corresponding to the target battery are parameters of a Kalman filter in the Kalman filter model; It should be noted that there are many ways to determine the Kalman filter model according to the target model parameters corresponding to the target battery. For example, the target model parameters can be used as input to model the model, or a discrete Kalman filter can be established. The present invention does not limit the method for determining the Kalman filter model according to the target model parameters corresponding to the target battery.

[0033] Among them, the target model parameters corresponding to the target battery can be obtained by identifying the model parameters of the circuit equivalent model corresponding to the target battery. The circuit equivalent model corresponding to the target battery can be a Thevenin circuit equivalent model, etc. The present invention does not limit the type of the circuit equivalent model corresponding to the target battery and the method of identifying the model parameters.

[0034] Step 102: Determine the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation based on the new information of the Kalman filter in the Kalman filter model, wherein the new information of the Kalman filter is used to characterize the difference between the actual measurement value and the predicted measurement value corresponding to the Kalman filter.

[0035] It should be noted that there are many ways to determine the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation based on the new information of the Kalman filter. For example, the historical open circuit voltage-remaining power curve and the current correct open circuit voltage-remaining power curve can be obtained, and the relationship with the new information is established by obtaining the partial differential matrix. The present invention does not limit the method for determining the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation based on the new information of the Kalman filter in the Kalman filter model.

[0036] Among them, the new information of the Kalman filter can be the information obtained before each evaluation, and the corresponding relationship between the latest target cross-correlation function, target autocorrelation function and open circuit voltage-remaining power curve deviation can be determined through the new information.

[0037] Step 103: Determine the remaining capacity of the target battery according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation.

[0038] It should be noted that the input value of this step is the Kalman filter model obtained in the previous step, as well as the variation of the new information cross-correlation function and the autocorrelation function, and the output value is the SOC estimation result. After obtaining the corresponding relationship between the target cross-correlation function and the open circuit voltage-remaining power curve deviation, the remaining power can be evaluated by establishing an adaptive multi-model Kalman filter or other methods. The present invention does not limit the method of determining the remaining power of the target battery according to the corresponding relationship between the target cross-correlation function and the open circuit voltage-remaining power curve deviation.

[0039] It can be understood that the present invention determines the Kalman filter model according to the target model parameters corresponding to the target battery, and determines the remaining power of the target battery based on the analysis of the cross-correlation function CCM and the autocorrelation function ACM of the Kalman filter model. This method can accurately estimate the SOC of the target battery even if there is an error in the OCV-SOC curve, reducing the dependence on the real-time OCV-SOC curve, thereby reducing the workload of constantly updating the open circuit voltage-remaining power curve.

[0040] Figure 2 Schematic diagram of the method for determining the Kalman filter model provided by the present invention. Figure 2 As shown, determining the Kalman filter model according to the target model parameters corresponding to the target battery may include: Step 201: Determine an equivalent circuit model corresponding to the target battery, wherein the equivalent circuit model includes a correspondence between ohmic internal resistance, polarization resistance, polarization capacitance, terminal voltage, open circuit voltage, and current.

[0041] Figure 3 Schematic diagram of the equivalent circuit model provided by the present invention. Figure 3 As shown, a battery equivalent circuit model is established, where The ohmic internal resistance of the battery, reflecting the terminal voltage at the moment of switching between charge and discharge states. mutations, and Used to describe the battery polarization effect and reflect the terminal voltage The gradual process of is the open circuit voltage of the battery. Under any given operating conditions, It is positively correlated with the battery SOC and can therefore reflect the static characteristics of the battery. Represents the polarization voltage on the RC link. For the battery charge / discharge current, the charging current direction is selected as the positive direction.

[0042] Step 202: Input the target measurement value of the target battery into the equivalent circuit model to obtain the target model parameters, wherein the target measurement value includes a target terminal voltage and a target current, and the target model parameters include a target ohmic internal resistance, a target polarization resistance, and a target polarization capacitance.

[0043] It should be noted that the terminal voltage and current is the input quantity of the equivalent circuit model, that is, the measured value. , , is the output value of the equivalent circuit model, that is, the value to be estimated. Exemplarily, the equivalent circuit model can determine the target model parameters by discretizing and analyzing the continuous-time state equation of the Thevenin equivalent circuit model of the battery.

[0044] Step 203: Modeling is performed by taking the target model parameters corresponding to the target battery as input quantities of the Kalman filter model to obtain the Kalman filter model.

[0045] It should be noted that the target model parameters can be used as the input of the Kalman filter to model, for example, a discrete Kalman filter is established, and the Kalman filter model is determined according to the value at time k. The Kalman filter model serves the subsequent steps.

[0046] For example, a discrete Kalman filter can be written as: (17), where represents the SOC at the initial moment, It represents the result of ampere-hour integration, which represents the change of SOC. The meanings of the parameters are represents the charge utilization coefficient, is the charge integration operation, Represents the battery capacity.

[0047] (18), where represents the state quantity to be estimated at time k, represents the state transfer matrix, represents the state quantity to be estimated at time k-1, represents the k-1 moment excitation matrix, represents the current at time k-1, represents the process noise at k-1 time.

[0048] (19), among which, represents the terminal voltage in the measurement model at time k, represents the difference between the open circuit voltage and the polarized capacitor voltage at time k, represents the ohmic internal resistance at time k, represents the measurement noise at time k.

[0049] (20), where represents the state transfer matrix, represents an exponential operation, represents the polarization internal resistance at time k-1, represents the polarization voltage at time k-1, represents the sampling interval, Represents the sampling interval divided by the battery capacity.

[0050] (21), where represents the difference between the open circuit voltage and the polarized capacitor voltage at time k, represents the open circuit voltage at time k, represents the polarization capacitor voltage at time k, It represents the opposite number of the ohmic internal resistance at time k. It is expressed as the opposite number of the ohmic internal resistance at time k.

[0051] In the formula , represents the state quantity to be estimated at time k, represents the SOC value at time k, , represents the current at time k, for The value at time k, and is white noise, , for The variance of Represents a mathematical operator. hour, ,otherwise . , for The variance of ,when hour, ,otherwise The open circuit voltage is described as , Indicates the charge and discharge status.

[0052] Furthermore, the step 202 inputs the target measurement value of the target battery into the equivalent circuit model to obtain the target model parameters, which may include: discretizing the continuous-time state equation corresponding to the equivalent circuit model to obtain a discrete state equation of the target polarization voltage; obtaining a discrete variation equation of the target terminal voltage based on the discrete state equation of the target polarization voltage; simplifying the discrete variation equation of the target terminal voltage based on the discrete variation equation of the target open circuit voltage to obtain a measurement equation of the equivalent circuit model, the measurement equation including parameters to be estimated, and the parameters to be estimated are used to characterize the time-varying state of the target model parameters; and estimating the values ​​of the parameters to be estimated in the measurement equation using an adaptive recursive least squares method containing a forgetting factor to obtain the target model parameters.

[0053] It should be noted that the terminal voltage and current is the input quantity, i.e. the measured value. , , are the output values, i.e. the values ​​to be estimated. These output values ​​will be used in the Kalman filter model in step three and are the parameters of the Kalman filter.

[0054] Furthermore, the use of the adaptive recursive least squares method containing a forgetting factor to estimate the value of the parameter to be estimated in the measurement equation to obtain the target model parameters may include: determining the initial covariance of the parameter to be estimated and the covariance at the target moment; estimating the value of the parameter to be estimated in the measurement equation by adaptively adjusting the forgetting factor based on the initial covariance of the parameter to be estimated, the covariance at the target moment, and the gain of the equivalent circuit model at the target moment to obtain the value of the parameter to be estimated; obtaining the target model parameters based on the value of the parameter to be estimated.

[0055] It should be noted that the following is the battery model parameter online identification algorithm: According to Kirchhoff's law, the continuous-time state equation of the battery Thevenin equivalent circuit model can be obtained as follows: (1), In the formula , is the time constant of the first-order RC link, Represents the rate of change of polarization capacitor voltage. Represents the ratio of current to polarization capacitance. By discretizing equation (1), we can obtain equations (2) and (3): The discrete state equation is: (2), in, Represents a mathematical operation. Represents a mathematical operation.

[0056] (3), where Represents the ohmic internal resistance voltage at time k.

[0057] in is the sampling frequency, which is set to 1s in the present invention. , , , , are their respective values ​​at the kth sampling point, , are the values ​​of the k-1th sampling point. All the quantities with (k), (k-1), and (k-2) below indicate the values ​​of the physical quantity at the corresponding moment.

[0058] Substituting equation (3) into equation (2), we can eliminate get (4), where , the same can be obtained The expression of is subtracted from (4) to obtain (5), (5) in, , , , Battery open circuit voltage The dynamic change of is mainly affected by battery aging and current operating conditions, so it can be described as the remaining power , Ambient temperature , Current Capacity And the charge and discharge status Function expression , so taking the charging period as an example, The rate of change at any time is: (6) For the SOC range [20%, 80%] where lithium iron phosphate usually operates, the OCV-SOC curve is relatively flat. The value is small. At the same time, if an efficient algorithm is used to effectively shorten the convergence time of the parameter estimation results, the change in SOC can be ignored, that is, The battery temperature changes at a low rate in a non-faulty operating state. , the battery aging state hardly changes in a short period of time, so , and the hysteresis effect has little effect on Uoc. And because the sampling time is short enough, , so we can get: (7), The measurement noise is taken into account in the formula , represents a Gaussian distribution, Describes the noise variance, and the noise of each step of the measurement equipment is independent of each other, equation (5) can be simplified to (8) , (9) is the value of the noise at time k. Parameters to be estimated is time-varying because the circuit model parameters , , Considering the influence of temperature T, SOC, and capacity Q It is time-varying, but its changing rate is relatively slow. The present invention adopts the adaptive recursive least squares method (ARLS) with forgetting factor to estimate , forgetting factor It is used to prevent data saturation and make the parameter estimation results more biased towards real-time data and closer to the current parameter values. The specific algorithm flow of ARLS is shown in Table 1. It is worth noting that in the low SOC interval (SOC<0.1), the circuit parameters change more drastically. A smaller forgetting factor can be selected, or the sampling frequency can be increased to reduce the time range of the measured data.

[0059] Table 1 Flow of the adaptive identification algorithm for battery equivalent circuit model parameters in, for The initial covariance of for The covariance at time k is, is the gain of the model at time k. middle and are all white noise with a mean of 0. is the impulse function, middle is the forgetting factor at time k.

[0060] Through ARLS, a stable After the value is obtained, the ECM parameters of the equivalent circuit model to be calculated can be indirectly calculated by the following formula: (16).

[0061] It is understandable that by establishing an equivalent circuit model of a lithium iron phosphate battery and adaptively adjusting the forgetting factor, rapid and accurate identification of model parameters can be achieved. The adaptive identification algorithm of the parameters of the lithium iron phosphate battery equivalent circuit model proposed in the present invention can rapidly and accurately identify the latest parameter results and ensure the accuracy of the filter state equation.

[0062] Figure 4 FIG. 1 is a flow chart of a method for obtaining a corresponding relationship provided by the present invention. Figure 4 As shown, the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation is determined according to the new information of the Kalman filter in the Kalman filter model, including: Step 401: By analyzing the new information of the Kalman filter in real time, the relationship between the new information covariance and the currently used open circuit voltage-remaining power curve and the real open circuit voltage-remaining power curve is obtained.

[0063] Step 402: Determine the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation according to the relationship between the new information covariance and the currently used open circuit voltage-remaining power curve and the real open circuit voltage-remaining power curve.

[0064] It should be noted that the input of this step is the model established in step 101, and the output is the variation of the cross-correlation function and autocorrelation function of O new information.

[0065] Among them, the Kalman filter information cross-correlation function and autocorrelation function analysis. Here, by analyzing the new information of the filter in real time, the relationship between the new information covariance and the current OCV-SOC curve and the real curve is obtained, providing a basis for the next step.

[0066] Kalman filter , Defined as new information, In the current operating condition The correct OCV-SOC curve is: is the exact partial differential matrix of the measurement equation, and This is the only OCV-SOC curve we have currently measured during the battery's historical operation. Obviously, this curve has a certain error. is the partial differential matrix of the measurement equation currently used, let .

[0067] (twenty two) in , is the prior estimation error, and the autocorrelation matrix of the new information is shown below. For convenience, assume .

[0068] (twenty three) In the formula, and are all white noise with a mean of 0, and when k>=i and are independent, so the above formula can be simplified to: (twenty four), When k=i, the same formula can be used to obtain The autocorrelation matrix is (25).

[0069] So far, the variation formulas of the cross-correlation function and autocorrelation function of the information when there is a deviation in the OCV-SOC curve are obtained.

[0070] It can be understood that by characterizing the qualitative relationship between the Kalman filter information cross-correlation function (CCM), the autocorrelation function (ACM) and the OCV-SOC curve deviation, the more accurate direction of the filter measurement model can be clarified.

[0071] Figure 5 FIG. 1 is a flow chart of determining the remaining power of a target battery provided by the present invention. Figure 5 As shown, determining the remaining capacity of the target battery according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation may include: Step 501: setting a plurality of Kalman filters, each filter having a different measurement equation; Step 502: dynamically adjusting the parameters of the multiple Kalman filters according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation; Step 503: Based on the Bayesian analysis results corresponding to the multiple Kalman filters, a Kalman filter whose confidence meets a preset confidence requirement is selected as the optimal filter, and the output value of the optimal filter is the remaining power of the target battery.

[0072] It should be noted that the SOC estimation is completed based on the adaptive multi-model Kalman filter. According to the real-time estimation results of the new information covariance and the real-time results between the new information covariance and the OCV-SOC curve deviation in the previous step, a set of adaptive filters is generated in the next estimation cycle.

[0073] Will , Set as the common initial state of the multi-model Kalman filter, which is the output result of the Kalman filter with the best performance in the previous estimation cycle at the end of the estimation cycle. The length of the estimation cycle is L sampling points, and the length is In the new estimation cycle In the example, n Kalman filters are set up, and the measurement equation is The value of Estimated period Contains several filter sampling cycles to filter out the most likely measurement equation parameters , of course, the estimated cycle duration is shorter, can be considered as constant. ,Right now is the estimated value of the terminal voltage based on the a priori estimation result. At any time during the estimation period, is unknown due to measurement noise The uncertainty is obtained based on the Jacobian matrix of the incorrect OCV-SOC curve It may also be closer to , that is, the wrong new information New information than the accurate model The absolute value is smaller. However, as the number of sampling points increases, the estimated measurement value of the correct measurement model is generally more consistent with the actual measurement value, so it is necessary to study the given measurement value Under this condition, the probability of each filter measuring the equation parameters is According to Bayes' theorem: (26).

[0074] In the above formula, For measurement data Parameters before passing in The probability value is taken into account The probability update result after all previous data, so . for Estimated measurement value under parameter conditions, due to the measurement is continuous, not discrete, so any filter (even the right one) will output Probability is approximately 0, so here we use the fact that the probability of an event is proportional to its probability density function to rewrite the above formula as (27).

[0075] Since the measured value at time k will depend on and ,like If it holds, it means that the filter has a better estimation result, that is, , then we can get Due to the current recognition , then according to formula (22), the prior estimate is unbiased, that is ,right At the beginning of the estimation period Performing Taylor expansion and retaining the first differential term yields: (28).

[0076] In all Kalman filters constructed in the present invention, and are all Gaussian, so if Established, obtained is also a linear combination of Gaussian random variables, so its distribution is also Gaussian, with a mean of Approximately ,variance ,therefore The probability density function of The values ​​are: (29).

[0077] In some embodiments, the selection of a Kalman filter whose confidence level meets preset confidence requirements as the optimal filter based on the Bayesian analysis results corresponding to the multiple Kalman filters may include: updating the parameter probabilities of each Kalman filter after each sampling cycle, and after the estimation cycle ends, determining the Kalman filter with the highest parameter probability as the optimal filter.

[0078] It should be noted that each parameter can be updated after each sampling cycle. Probability , in the estimation cycle After completion, according to Choose the best As the parameter value of the measurement equation for this estimation cycle, and the last a posteriori estimate and its covariance of the filter in this estimation cycle are used as the parameter value of the next estimation cycle The output value of this filter at each moment in this estimation cycle is the remaining power estimation result finally output by this method.

[0079] In order to solve the problem of inaccurate remaining power estimation caused by the battery OCV-SOC curve changing with battery aging and operating temperature fluctuations in the battery energy storage system, the present invention adopts an adaptive multi-model Kalman filter to correct the estimation error, so that the remaining power estimation result takes into account both accuracy and robustness. Among them, an adaptive multi-model filter is established, and a Kalman filter containing different measurement equations is dynamically generated according to the information CCM and ACM analysis results. Based on the Bayesian analysis results, the filter with the highest confidence is selected as the optimal filter, and its output value is the final remaining power estimate. It is possible to obtain accurate SOC estimation results even when there is a deviation in the existing OCV-SOC curve, avoiding frequent measurement of the OCV-SOC curves of a large number of batteries.

[0080] In addition, the method of the present invention is a model-based method with a small amount of calculation and low requirements on software and hardware, and can be easily embedded in a battery management system.

[0081] The purpose of the present invention is to propose a method for accurately estimating the remaining power of a lithium iron phosphate battery when there is a deviation in the existing open circuit voltage-remaining power curve. The method performs external characteristic modeling and parameter identification on lithium iron phosphate to obtain a Kalman filter state model. Based on the qualitative relationship analysis results between the filter information cross-correlation function, the autocorrelation function and the curve deviation, a set of multi-model Kalman filters is adaptively generated, and the optimal filter is selected according to the Bayesian analysis results, and its output result is used as the final remaining power estimation result. This method can be applied to large-scale battery energy storage systems to solve the problem of inaccurate remaining power estimation caused by the open circuit voltage-remaining power curve deviation.

[0082] The following describes an exemplary application of an embodiment of the present invention in a practical application scenario.

[0083] Figure 6 FIG. 1 is a schematic diagram of the overall flow of the remaining power estimation method provided by the present invention. Figure 6 As shown, the method includes the following steps 601 to 606: Step 601: Establishing an equivalent circuit model of a lithium iron phosphate battery; Step 602: Model parameter identification based on ARLS; Step 603: Establishing a SOC estimation Kalman filter model; Step 604: Analyze the new information cross-correlation and autocorrelation functions; Step 605: Establish a multi-model Kalman filter and adjust measurement model parameters; Step 606: Select the optimal filter based on Bayesian analysis to obtain an estimated SOC value.

[0084] In the present invention, the lithium iron phosphate battery modeling is first carried out, the model parameters are identified, and the state equation information of the remaining power estimation Kalman filter is obtained. Then, the mapping relationship between the Kalman filter information cross-correlation function, the autocorrelation function and the OCV-SOC curve deviation is analyzed, and the measurement model of each filter in the adaptive multi-model Kalman filter is adjusted according to the analysis conclusion. Finally, according to the Bayesian analysis results of the output values ​​of each filter, the optimal filter is selected, and its remaining power estimation result is the final result. The present invention helps to solve the problem that it is difficult to accurately measure the OCV-SOC of lithium iron phosphate batteries in large-scale energy storage systems, and the problem that the remaining power estimation result has a large error due to untimely updating. The accurate estimation result is conducive to battery safety management and control and the reasonable arrangement of battery energy storage scheduling plans.

[0085] Based on the foregoing embodiments, an embodiment of the present invention provides a device for estimating remaining power. The modules included in the device and the units included in each module can be implemented by a processor; of course, they can also be implemented by a specific logic circuit; in the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP) or a field programmable gate array (FPGA), etc.

[0086] The remaining power estimation device provided by the present invention is described below. The remaining power estimation device described below and the remaining power estimation method described above can be referred to each other.

[0087] Figure 7 Schematic diagram of the structure of the remaining power estimation device provided by the present invention. Figure 7 As shown, the device 700 includes a model determination module 701, a relationship determination module 702 and an electric quantity determination module 703, wherein: A model determination module 701 is used to determine a Kalman filter model according to target model parameters corresponding to a target battery, wherein the target model parameters corresponding to the target battery are obtained by performing model parameter identification on a circuit equivalent model corresponding to the target battery, and the target model parameters corresponding to the target battery are parameters of a Kalman filter in the Kalman filter model; A relationship determination module 702 is used to determine the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation according to the new information of the Kalman filter in the Kalman filter model, wherein the new information of the Kalman filter is used to characterize the difference between the actual measurement value and the predicted measurement value corresponding to the Kalman filter; The power determination module 703 is used to determine the remaining power of the target battery according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation.

[0088] In some embodiments, the model determination module 701 includes a model determination unit, a parameter determination unit and a model generation unit, wherein: A model determination unit, used to determine an equivalent circuit model corresponding to the target battery, wherein the equivalent circuit model includes a corresponding relationship between an ohmic internal resistance, a polarization resistance, a polarization capacitance, a terminal voltage, an open circuit voltage, and a current; a parameter determination unit, configured to input a target measurement value of the target battery into the equivalent circuit model to obtain the target model parameters, wherein the target measurement value includes a target terminal voltage and a target current, and the target model parameters include a target ohmic internal resistance, a target polarization resistance, and a target polarization capacitance; The model generating unit is used to model the target model parameters corresponding to the target battery as input quantities of the Kalman filter model to obtain the Kalman filter model.

[0089] In some embodiments, the parameter determination unit is specifically used to: discretize the continuous-time state equation corresponding to the equivalent circuit model to obtain a discrete state equation of the target polarization voltage; According to the discrete state equation of the target polarization voltage, a discrete variation equation of the target terminal voltage is obtained; Simplifying the discrete variation equation of the target terminal voltage according to the discrete variation equation of the target open circuit voltage to obtain a measurement equation of the equivalent circuit model, wherein the measurement equation includes parameters to be estimated, and the parameters to be estimated are used to characterize the time-varying condition of the target model parameters; The adaptive recursive least square method including a forgetting factor is used to estimate the value of the parameter to be estimated in the measurement equation to obtain the target model parameter.

[0090] In some embodiments, the model determination unit is further specifically used to: determine the initial covariance of the parameter to be estimated and the covariance at the target moment; According to the initial covariance of the parameter to be estimated, the covariance at the target time, and the gain of the equivalent circuit model at the target time, the value of the parameter to be estimated in the measurement equation is estimated by adaptively adjusting the forgetting factor to obtain the value of the parameter to be estimated; The target model parameters are obtained according to the values ​​of the parameters to be estimated.

[0091] In some embodiments, the relationship determination module 702 includes an information acquisition unit and a relationship acquisition unit, wherein: The new information acquisition unit is used to obtain the relationship between the new information covariance and the currently used open circuit voltage-remaining power curve and the real open circuit voltage-remaining power curve by analyzing the new information of the Kalman filter in real time; The relationship acquisition unit is used to determine the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation based on the relationship between the new information covariance and the currently used open circuit voltage-remaining power curve and the actual open circuit voltage-remaining power curve.

[0092] In some embodiments, the power determination module 703 includes a filter setting unit, a parameter adjustment unit and a result acquisition unit, wherein: The filter setting unit is used to set a plurality of Kalman filters, each filter having a different measurement equation; The parameter adjustment unit is used to dynamically adjust the parameters of the multiple Kalman filters according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation; The result acquisition unit is used to select a Kalman filter whose confidence meets a preset confidence requirement as an optimal filter based on the Bayesian analysis results corresponding to the multiple Kalman filters, and the output value of the optimal filter is the remaining power of the target battery.

[0093] In some embodiments, the result acquisition unit is specifically used to: update the parameter probability of each Kalman filter after each sampling period ends, and after the estimation period ends, determine the Kalman filter with the highest parameter probability as the optimal filter.

[0094] In an embodiment of the present invention, a Kalman filter model can be determined according to target model parameters corresponding to a target battery, and the remaining power can be determined according to the correspondence between a target cross-correlation function, a target autocorrelation function and an open circuit voltage-remaining power curve deviation of the Kalman filter model. The SOC of the battery can be accurately estimated in the event of an open circuit voltage-remaining power curve error, thereby reducing the workload of continuously updating the open circuit voltage-remaining power curve.

[0095] Figure 8 Schematic diagram of the physical structure of the electronic device provided by the present invention. Figure 8As shown, the electronic device may include: a processor (processor) 810, a communication interface (Communications Interface) 820, a memory (memory) 830 and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 can call the logic instructions in the memory 830 to execute the remaining power estimation method, which includes: determining the Kalman filter model according to the target model parameters corresponding to the target battery, the target model parameters corresponding to the target battery are obtained by model parameter identification of the circuit equivalent model corresponding to the target battery, and the target model parameters corresponding to the target battery are the parameters of the Kalman filter in the Kalman filter model; determining the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation according to the new information of the Kalman filter in the Kalman filter model, the new information of the Kalman filter is used to characterize the difference between the actual measurement value and the predicted measurement value corresponding to the Kalman filter; determining the remaining power of the target battery according to the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation.

[0096] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0097] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the remaining power estimation method provided by the above-mentioned methods, and the method includes: determining a Kalman filter model according to target model parameters corresponding to a target battery, wherein the target model parameters corresponding to the target battery are obtained by performing model parameter identification on a circuit equivalent model corresponding to the target battery, and the target model parameters corresponding to the target battery are parameters of a Kalman filter in the Kalman filter model; determining a correspondence between a target cross-correlation function, a target autocorrelation function and an open circuit voltage-remaining power curve deviation according to new information of the Kalman filter in the Kalman filter model, wherein the new information of the Kalman filter is used to characterize the difference between an actual measurement value and a predicted measurement value corresponding to the Kalman filter; determining the remaining power of the target battery according to the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation.

[0098] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive Solid State Disk (SSD)), etc.

[0099] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the remaining power estimation method provided by the above-mentioned methods, the method comprising: determining a Kalman filter model according to target model parameters corresponding to a target battery, the target model parameters corresponding to the target battery being obtained by model parameter identification of a circuit equivalent model corresponding to the target battery, the target model parameters corresponding to the target battery being parameters of a Kalman filter in the Kalman filter model; determining a correspondence between a target cross-correlation function, a target autocorrelation function and an open circuit voltage-remaining power curve deviation according to new information of the Kalman filter in the Kalman filter model, the new information of the Kalman filter being used to characterize the difference between an actual measurement value and a predicted measurement value corresponding to the Kalman filter; determining the remaining power of the target battery according to the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation.

[0100] The above-mentioned computer-readable storage medium can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, - but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM) or flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0101] Computer-readable signal media may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0102] The program code embodied on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, radio frequency (RF), etc., or any suitable combination of the foregoing.

[0103] Computer program code for performing the operations of the present specification may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0104] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0105] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating remaining power, characterized in that: include: Determine a Kalman filter model according to target model parameters corresponding to a target battery, wherein the target model parameters corresponding to the target battery are obtained by performing model parameter identification on a circuit equivalent model corresponding to the target battery, and the target model parameters corresponding to the target battery are parameters of a Kalman filter in the Kalman filter model; Determine the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation according to the new information of the Kalman filter in the Kalman filter model, wherein the new information of the Kalman filter is used to characterize the difference between the actual measurement value and the predicted measurement value corresponding to the Kalman filter; The remaining capacity of the target battery is determined according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation.

2. The method for estimating remaining power according to claim 1, characterized in that: The step of determining the Kalman filter model according to the target model parameters corresponding to the target battery includes: Determine an equivalent circuit model corresponding to the target battery, wherein the equivalent circuit model includes a corresponding relationship between an ohmic internal resistance, a polarization resistance, a polarization capacitance, a terminal voltage, an open circuit voltage, and a current; Inputting the target measurement value of the target battery into the equivalent circuit model to obtain the target model parameters, wherein the target measurement value includes a target terminal voltage and a target current, and the target model parameters include a target ohmic internal resistance, a target polarization resistance, and a target polarization capacitance; The target model parameters corresponding to the target battery are used as input quantities of the Kalman filter model to perform modeling, thereby obtaining the Kalman filter model.

3. The method for estimating remaining power according to claim 2, characterized in that: The step of inputting the target measurement value of the target battery into the equivalent circuit model to obtain the target model parameter comprises: Discretizing the continuous-time state equation corresponding to the equivalent circuit model to obtain a discrete state equation of the target polarization voltage; According to the discrete state equation of the target polarization voltage, a discrete variation equation of the target terminal voltage is obtained; Simplifying the discrete variation equation of the target terminal voltage according to the discrete variation equation of the target open circuit voltage to obtain a measurement equation of the equivalent circuit model, wherein the measurement equation includes parameters to be estimated, and the parameters to be estimated are used to characterize the time-varying condition of the target model parameters; The adaptive recursive least square method including a forgetting factor is used to estimate the value of the parameter to be estimated in the measurement equation to obtain the target model parameter.

4. The method according to claim 3, characterized in that The method of estimating the value of the parameter to be estimated in the measurement equation by using the adaptive recursive least square method including a forgetting factor to obtain the target model parameter includes: Determine the initial covariance of the parameters to be estimated and the covariance at the target time; According to the initial covariance of the parameter to be estimated, the covariance at the target time, and the gain of the equivalent circuit model at the target time, the value of the parameter to be estimated in the measurement equation is estimated by adaptively adjusting the forgetting factor to obtain the value of the parameter to be estimated; The target model parameters are obtained according to the values ​​of the parameters to be estimated.

5. The method for estimating remaining power according to claim 1, characterized in that: The determining, according to the new information of the Kalman filter in the Kalman filter model, the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation comprises: By analyzing the new information of the Kalman filter in real time, the relationship between the new information covariance and the currently used open circuit voltage-remaining power curve and the real open circuit voltage-remaining power curve is obtained; According to the relationship between the new information covariance and the currently used open circuit voltage-remaining power curve and the real open circuit voltage-remaining power curve, the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation is determined.

6. The method for estimating remaining power according to claim 1, characterized in that: The determining the remaining capacity of the target battery according to the correspondence between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation includes: Setting multiple Kalman filters, each filter having a different measurement equation; Dynamically adjusting the parameters of the multiple Kalman filters according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining capacity curve deviation; Based on the Bayesian analysis results corresponding to the multiple Kalman filters, a Kalman filter whose confidence meets a preset confidence requirement is selected as an optimal filter, and an output value of the optimal filter is the remaining power of the target battery.

7. The method for estimating remaining power according to claim 6, characterized in that: The selecting, based on the Bayesian analysis results corresponding to the multiple Kalman filters, a Kalman filter whose confidence meets a preset confidence requirement as the optimal filter comprises: After each sampling period, the parameter probability of each Kalman filter is updated, and after the estimation period, the Kalman filter with the highest parameter probability is determined as the optimal filter.

8. A device for estimating remaining power, characterized in that: include: A model determination module, used to determine a Kalman filter model according to target model parameters corresponding to a target battery, wherein the target model parameters corresponding to the target battery are obtained by performing model parameter identification on a circuit equivalent model corresponding to the target battery, and the target model parameters corresponding to the target battery are parameters of a Kalman filter in the Kalman filter model; A relationship determination module, used to determine the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation according to the new information of the Kalman filter in the Kalman filter model, wherein the new information of the Kalman filter is used to characterize the difference between the actual measurement value and the predicted measurement value corresponding to the Kalman filter; The power determination module is used to determine the remaining power of the target battery according to the corresponding relationship between the target cross-correlation function, the target autocorrelation function and the open circuit voltage-remaining power curve deviation.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the remaining power estimation method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for estimating the remaining power as claimed in any one of claims 1 to 7 is implemented.

Images