Method and device for determining SOC (State of Charge) of battery and method and device for determining charging and discharging strategy of battery

Abstract

The invention provides a battery SOC determination method and device and a battery charging and discharging strategy determination method and device, and a specific implementation mode of the battery SOC determination method comprises the steps that battery parameters are acquired, and the battery parameters comprise at least one of voltage, current and temperature; according to the battery parameters, SOC estimation noise is determined, and the SOC estimation noise is related to the slope of a curve of the battery parameters changing along with charging and discharging time; and determining the SOC of the to-be-processed battery according to at least one of the voltage and the current of the to-be-processed battery and the SOC estimation noise. According to the method, the accuracy of the SOC of the to-be-processed battery can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery, in particular to a method for determining SOC of a battery, a method and device for determining battery charging and discharging strategy. BACKGROUND

[0002] A battery can provide power for electronic devices, and the SOC (State of charge, SOC for short) of the battery has a crucial influence on the performance and safety of the battery, so it is often necessary to determine the SOC of the battery.

[0003] In the related art, there are methods for determining SOC, such as ampere-hour integration method, open circuit voltage method, Kalman filtering algorithm, etc. However, the ampere-hour integration method leads to gradual growth of SOC error due to current error, thereby resulting in low accuracy of SOC. The open circuit voltage method needs to rely on the linear relationship between SOC and OCV (Open circuit voltage, OCV for short) and the battery static condition to determine the SOC, but there is a nonlinear relationship for batteries such as lithium iron phosphate battery and ternary lithium battery, and it is not easy to meet the battery static condition, thereby resulting in low accuracy of SOC. The Kalman filtering algorithm is based on pure algorithm mechanism to determine the SOC of the battery, which also leads to low accuracy of SOC.

[0004] Therefore, there is a problem of low accuracy of SOC in the related art. SUMMARY

[0005] Embodiments of the present application aim to provide a method for determining SOC of a battery, a method and device for determining battery charging and discharging strategy, so as to improve the accuracy of SOC of a battery to be processed.

[0006] In a first aspect, embodiments of the present application provide a method for determining SOC of a battery, the method comprising: obtaining battery parameters, the battery parameters comprising at least one of voltage, current, and temperature; determining SOC estimation noise according to the battery parameters, wherein the SOC estimation noise is related to the slope of the curve of the battery parameters changing with charging and discharging time; and determining the SOC of a battery to be processed according to at least one of the voltage and current of the battery to be processed and the SOC estimation noise. In this way, since the SOC estimation noise is related to the slope of the curve of the battery parameters changing with charging and discharging time, the SOC estimation noise can adaptively change with the battery parameters, thereby improving the accuracy of the SOC of the battery to be processed.

[0007] Optionally, the determining the SOC estimation noise according to the battery parameters comprises: determining an actual noise according to the battery parameters and a preset noise. In this way, the actual noise can adaptively change with the battery parameters, thereby improving the accuracy of the SOC of the battery.

[0008] Optionally, the actual noise is determined according to the battery parameter and a preset noise, including: the actual Kalman filter noise is determined according to the battery parameter and a preset Kalman filter noise; the actual Kalman filter noise includes actual observation noise and actual prediction noise of a circuit model; the actual observation noise represents an error existing in the observed SOC data of the battery; the actual prediction noise represents an error existing in the predicted SOC data of the battery; and the SOC of the battery to be processed is determined according to at least one of the voltage and the current of the battery to be processed and the SOC estimation noise, including: the actual SOC of the battery to be processed is determined by using a Kalman filter algorithm according to the voltage, the current of the battery to be processed and the actual Kalman filter noise. Here, the SOC of the battery is not determined based on pure algorithm mechanism, but is determined by combining the battery characteristics and the algorithm mechanism. In this way, since the actual Kalman filter noise is not a fixed constant, but adaptively changes with the battery parameter, the accuracy of the SOC of the battery can be improved.

[0009] Optionally, the actual Kalman filter noise is determined according to the battery parameter and a preset Kalman filter noise, including: a relationship coefficient between the actual prediction noise and the actual observation noise is determined according to the battery parameter; one of the actual Kalman filter noise is determined according to the battery parameter and the preset Kalman filter noise; and the other of the actual Kalman filter noise is determined according to the determined one of the actual Kalman filter noise and the relationship coefficient. In this way, after one of the actual observation noise and the actual prediction noise is determined, the other can be determined according to the relationship coefficient, so that the operation cost can be reduced and the operation rate can be accelerated.

[0010] Optionally, the actual prediction noise and the actual observation noise are negatively correlated, so that after one of them is determined, the other can be determined according to the negative correlation.

[0011] Optionally, the actual prediction noise is positively correlated with the slope of the curve of the battery parameter changing with the charging and discharging time, and the actual observation noise is negatively correlated with the slope of the curve of the battery parameter changing with the charging and discharging time. In this way, when the actual prediction noise is small, the actual prediction value at this moment is more reliable; when the actual observation noise is small, the actual observation value at this moment is more reliable.

[0012] Optionally, the actual prediction noise is a product of a first preset noise and a slope of the curve of the battery parameter changing with the charging and discharging time; and the actual observation noise is a quotient of a second preset noise and the slope of the curve of the battery parameter changing with the charging and discharging time. In this way, after the slope of the curve of the battery parameter changing with the charging and discharging time is obtained, the slope can be multiplied by the first preset noise to obtain the actual prediction noise, and the second preset noise can be divided by the slope to obtain the actual observation noise, thereby improving the operation speed.

[0013] Optionally, the two noises in the actual Kalman filtering noise are in proportional, exponential or logarithmic relationship with the slope of the curve of the battery parameter changing with the charging and discharging time. In this way, a corresponding relationship can be constructed according to the proportional, exponential or logarithmic relationship between the actual Kalman filtering noise and the slope of the curve of the battery parameter changing with the charging and discharging time, so that the noise value is more closely related to the slope and the noise changes more sensitively.

[0014] Optionally, the relationship coefficient includes at least one of a proportional coefficient, an exponential coefficient and a differential coefficient.

[0015] Optionally, the determining the actual Kalman filtering noise according to the battery parameter and a preset Kalman filtering noise includes: if the slope of the curve of the battery parameter changing with the charging and discharging time is in any preset interval, determining a preset Kalman filtering noise corresponding to the preset interval as the actual Kalman filtering noise. In this way, the actual Kalman filtering noise can be determined as the corresponding preset Kalman filtering noise according to the preset interval of the slope of the curve of the battery parameter changing with the charging and discharging time, so that the actual Kalman filtering noise is different in different preset intervals of the slope, the operation is more simple and convenient, and the accuracy of the battery SOC is improved to some extent.

[0016] Optionally, if the slope of the curve of the battery parameter changing with the charging and discharging time is in any preset interval, the preset Kalman filtering noise corresponding to the preset interval is determined as the actual Kalman filtering noise, including: if the voltage in the battery parameter is greater than a preset upper voltage limit value or the voltage in the battery parameter is less than a preset lower voltage limit value, a first preset Kalman filtering noise is determined as the actual Kalman filtering noise; and if the voltage in the battery parameter is between the preset lower voltage limit value and the preset upper voltage limit value, a second preset Kalman filtering noise is determined as the actual Kalman filtering noise. In this way, the first preset Kalman filtering noise or the second preset Kalman filtering noise can be adaptively determined as the actual Kalman filtering noise according to the voltage value, so that a more accurate actual Kalman filtering noise can be obtained, and the accuracy of the SOC is improved to some extent.

[0017] Optionally, if the slope of the curve of the battery parameter changing with the charging and discharging time is within any preset interval, the preset Kalman filter noise corresponding to the preset interval is determined as the actual Kalman filter noise, including: if the voltage in the battery parameter is greater than a preset upper limit of voltage, a third preset Kalman filter noise is determined as the actual Kalman filter noise; if the voltage in the battery parameter is less than a preset lower limit of voltage, a fourth preset Kalman filter noise is determined as the actual Kalman filter noise; if the voltage in the battery parameter is between the preset lower limit of voltage and the preset upper limit of voltage, a fifth preset Kalman filter noise is determined as the actual Kalman filter noise. In this way, a plurality of preset Kalman filter noises are set with smaller preset granularity, so that a more accurate actual Kalman filter noise can be determined, and the accuracy of the SOC is also improved to a certain extent.

[0018] Optionally, if the slope of the curve of the battery parameter changing with the charging and discharging time is within any preset interval, the preset Kalman filter noise corresponding to the preset interval is determined as the actual Kalman filter noise, including: if the rate of change of the voltage in the battery parameter with the SOC data is greater than a preset rate of change threshold, a sixth preset Kalman filter noise is determined as the actual Kalman filter noise; if the rate of change of the voltage in the battery parameter with the SOC data is less than a preset rate of change threshold, a seventh preset Kalman filter noise is determined as the actual Kalman filter noise. In this way, the corresponding preset Kalman filter noise can be determined as the actual Kalman filter noise by the preset interval in which the rate of change of the voltage with the SOC data is located, so that the actual Kalman filter noise adaptively changes with the change of the rate of change of the voltage with the SOC data, and the accuracy of the actual Kalman filter noise is improved to a certain extent, thereby improving the accuracy of the battery SOC.

[0019] Optionally, the SOC estimation noise is determined according to the battery parameter, including: the battery parameter is input into a preset model to obtain a predicted SOC; the SOC estimation noise is obtained according to the predicted SOC and an expected SOC; and the expected SOC is determined according to a measured historical SOC value. In this way, the SOC estimation noise can be obtained by the predicted SOC and the expected SOC, and the accuracy of the SOC estimation noise is improved to a certain extent due to the high accuracy of the expected SOC.

[0020] Optionally, the SOC estimation noise includes current noise, voltage noise and / or temperature noise. In this way, the SOC estimation noise includes a plurality of parameter influence noises, so that the SOC estimation is more accurate in an actual scenario including current, voltage and / or temperature.

[0021] Optionally, before determining the actual SOC of the battery to be processed using a Kalman filter algorithm based on the voltage, current, and actual Kalman filter noise of the battery to be processed, the method further includes: identifying the circuit model parameters.

[0022] Optionally, identifying the circuit model parameters includes: identifying the circuit model parameters based on the circuit model and the actual Kalman filter noise. This helps improve the accuracy of the SOC, thereby obtaining a more accurate SOC.

[0023] Optionally, identifying the circuit model parameters includes: identifying the circuit model parameters based on the circuit model and the recursive least squares method. This allows for a simpler and more convenient solution to the circuit model parameters, improving computational efficiency to some extent.

[0024] Optionally, after identifying the circuit model parameters, the method further includes: correcting the circuit model parameters according to correction rules, thereby improving the accuracy of the circuit model parameters.

[0025] Optionally, correcting the circuit model parameters according to the correction rules includes: if the circuit model parameters are greater than the upper limit of a preset identification range, then correcting the circuit model parameters to the upper limit of the preset identification range; if the circuit model parameters are less than the lower limit of the preset identification range, then correcting the circuit model parameters to the lower limit of the preset identification range. This limits the circuit model parameters to the preset identification range, thereby improving the accuracy of the circuit model parameters and preventing circuit model discrepancies.

[0026] Optionally, the step of correcting the circuit model parameters according to the correction rule includes: if the rate of change between the circuit model parameters identified in the current frame and the circuit model parameters identified in the previous frame is greater than the upper limit of the rate of change range, then the circuit model parameters identified in the current frame are corrected according to the upper limit of the rate of change range. Here, considering that circuit model parameters do not change abruptly in a short period of time, if a pure algorithm mechanism is used to identify the circuit model parameters, values ​​with large variations may be obtained. This does not conform to actual laws. Therefore, limiting the rate of change of the circuit model parameters identified in the current frame can obtain circuit model parameters that conform to actual laws, thereby improving the accuracy of the SOC of the battery to be processed to a certain extent.

[0027] Optionally, correcting the circuit model parameters according to the correction rules includes: weighting the circuit model parameters identified in the current frame and the circuit model parameters identified in the previous frame to correct the circuit model parameters identified in the current frame. This weights the circuit model parameters identified in adjacent frames, thereby limiting large abrupt changes in the circuit model parameters identified in adjacent frames and improving the accuracy of the SOC.

[0028] Optionally, before determining the actual SOC of the battery to be processed using the Kalman filter algorithm based on the voltage, current, and actual Kalman filter noise of the battery to be processed, the method further includes: if the voltage of the battery to be processed in the current frame is not within a preset voltage data range, then correcting the voltage to the voltage of the previous frame; and / or, if the current data of the battery to be processed in the current frame is not within a preset current data range, then correcting the current data to the preset current data. This allows for targeted correction of the collected voltage and current data, thereby reducing the errors caused by both and improving the accuracy of the SOC.

[0029] Optionally, the battery to be processed is the same type of battery used to determine the SOC estimation noise. In this way, after determining the SOC estimation noise corresponding to a type of battery, this SOC estimation noise can be used to determine the SOC of the same type of battery, thereby improving the accuracy of SOC while reducing computational costs.

[0030] Secondly, embodiments of this application provide a method for determining a battery charging and discharging strategy, comprising: acquiring battery parameters during the battery charging and discharging process, the battery parameters including at least one of voltage, current, and temperature; determining SOC estimation noise based on the battery parameters, wherein the SOC estimation noise is related to the slope of the curve of the battery parameters changing with charging and discharging time; determining the SOC of the battery to be processed based on at least one of voltage and current and the SOC estimation noise; and determining a charging and discharging strategy based on the SOC of the battery to be processed. In this way, a suitable charging and discharging strategy can be determined based on the SOC of the battery to be processed, thereby improving the service life of the battery to be processed to a certain extent.

[0031] Thirdly, embodiments of this application provide a device for determining the State of Charge (SOC) of a battery. The device includes: an acquisition module for acquiring battery parameters, the battery parameters including at least one of voltage, current, and temperature; a noise determination module for determining SOC estimation noise based on the battery parameters, wherein the SOC estimation noise is correlated with the slope of a curve showing the battery parameters changing with charge / discharge time; and a SOC determination module for determining the SOC of the battery to be processed based on at least one of the voltage and current of the battery to be processed and the SOC estimation noise. Thus, since the SOC estimation noise is correlated with the slope of the curve showing the battery parameters changing with charge / discharge time, the SOC estimation noise adapts to changes in battery parameters, thereby improving the accuracy of the SOC of the battery to be processed.

[0032] Fourthly, embodiments of this application provide an electronic device, including a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps of the method provided in the first aspect above are performed.

[0033] Fifthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method provided in the first aspect above.

[0034] In a sixth aspect, embodiments of this application provide a computer program product comprising a computer program or instructions that, when executed by a processor, perform the method described in the first aspect.

[0035] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 A flowchart illustrating a method for determining the state of charge (SOC) of a battery, as provided in an embodiment of this application;

[0038] Figure 2 A schematic diagram showing the voltage variation of a battery with SOC during charging and discharging processes, provided as an embodiment of this application.

[0039] Figure 3 This is a comparison chart showing the error of a battery-based SOC determination method provided by an embodiment of this application, and the SOC determined by the conventional extended Kalman filter algorithm.

[0040] Figure 4 A structural diagram of an equivalent circuit model provided in an embodiment of this application;

[0041] Figure 5 A schematic diagram illustrating the rate of change of battery voltage with SOC, provided for an embodiment of this application;

[0042] Figure 6A flowchart illustrating a method for determining a battery charging and discharging strategy provided in this application embodiment;

[0043] Figure 7 A structural block diagram of a battery SOC determination device provided in an embodiment of this application;

[0044] Figure 8 This is a schematic diagram of an electronic device used to perform a method for determining the SOC of a battery or a method for determining a battery charging and discharging strategy, as provided in an embodiment of this application. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0046] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0047] It should be noted that, unless otherwise specified, the embodiments or technical features in the embodiments of this application may be combined.

[0048] Furthermore, the defects in the solutions in the above-mentioned related technologies are all results obtained by the inventors after practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed by the embodiments of the present invention in the following text should all be contributions made by the inventors to the present invention.

[0049] It should be noted that this application can determine the battery's SOC through a server, server cluster, or cloud platform. In other application scenarios, the battery's SOC can also be determined through a battery management system, vehicle-mounted computer, etc., and this application does not impose any restrictions on this. For example, this application will hereinafter refer to the server.

[0050] It should be noted that in related technologies, there is a problem of low accuracy in SOC estimation. Therefore, this application determines the SOC estimation noise based on battery parameters, thereby enabling the determination of the SOC based on the SOC estimation noise and relevant parameters of the battery to be processed. Since the SOC estimation noise is related to the slope of the curve showing the change of battery parameters with charge / discharge time, the SOC estimation noise adapts to changes in battery parameters, thus improving the accuracy of the SOC of the battery to be processed.

[0051] For details, please refer to Figure 1 The diagram illustrates a flowchart of a method for determining the State of Charge (SOC) of a battery according to an embodiment of this application. Figure 1 As shown, the method for determining the SOC of the battery includes the following steps 101 to 103.

[0052] Step 101: Obtain battery parameters, including at least one of voltage, current, and temperature;

[0053] The battery parameters mentioned above can be regarded as parameters describing the state or performance of the battery. In addition to the battery's voltage, current, and temperature, they may also include, for example, battery capacity and charge / discharge rate.

[0054] It should be noted that the above battery parameters can be historical battery parameters. That is, the server can obtain historical battery parameters to determine the battery's SOC.

[0055] Step 102: Determine the SOC estimation noise based on the battery parameters, wherein the SOC estimation noise is related to the slope of the curve showing the change of battery parameters with charge and discharge time;

[0056] The aforementioned SOC estimation noise refers to the noise present in the process of determining the battery SOC.

[0057] In some application scenarios, the SOC estimation noise includes current noise, voltage noise, and / or temperature noise. Current noise can be determined, for example, based on the battery's internal resistance; the higher the battery's internal resistance, the greater the current noise. Temperature noise can be determined, for example, based on the battery's operating temperature range; for instance, the temperature noise is lower when the battery is at 25°C, and higher when it is below -10°C or above 55°C.

[0058] In these application scenarios, SOC estimation noise incorporates various parameters that affect noise, making SOC estimation more accurate in real-world scenarios that include current, voltage, and / or temperature.

[0059] Please see Figure 2It shows the battery voltage (i.e., charge ocv and discharge ocv) as a function of SOC during the charging and discharging processes (i.e., charge ocv fitting and discharge ocv fitting). It is evident that for the same SOC, the battery voltage differs during charging and discharging. In other words, battery parameters may change depending on the battery's state (e.g., charging or discharging).

[0060] Therefore, battery parameters include those occurring during charging or discharging, and these parameters change with charging / discharging time. Thus, a curve showing the change of battery parameters over charging / discharging time can be obtained, and the slope of this curve can be derived. It should be noted that the slope of this curve can be obtained by dividing the changed voltage value by the corresponding change in SOC (State of Charge).

[0061] In some application scenarios, the server can determine the state of the battery by the charge / discharge fields sent by the battery management system, and obtain the battery parameters in the corresponding state.

[0062] In other application scenarios, the server can also determine the battery's state by setting a time threshold T. Specifically, if the battery is initially in a discharging state, and the continuous charging time exceeds the time threshold T, then the battery is determined to be in a charging state; otherwise, it is determined to be in a discharging state. Conversely, if the battery is initially in a charging state, and the continuous discharging time exceeds the time threshold T, then the battery is determined to be in a discharging state; otherwise, it is determined to be in a charging state.

[0063] In other application scenarios, servers can also determine the battery's state using current data. Specifically, current data during charging is generally negative, while current data during discharging is generally positive; therefore, the battery's state can be determined based on the sign of the current data.

[0064] In other application scenarios, servers can also determine the charging and discharging status based on the on / off state of the charging and discharging relay. For example, if the normally open contact of the charging and discharging relay is closed, it can be determined that the battery is charging; if the normally open contact of the charging and discharging relay is open, it can be determined that the battery is discharging.

[0065] In other application scenarios, servers can also determine the battery's charging / discharging state by observing whether the battery voltage is continuously rising or falling. A continuous rise in voltage indicates the battery is charging, while a continuous fall indicates it is discharging.

[0066] Therefore, after determining the state of the battery, the server can obtain the slope of the curve showing how the battery parameters change over time in that state.

[0067] In some applications, the SOC estimation noise is positively correlated with this slope; in others, it is negatively correlated. This is described in detail below and will not be repeated here.

[0068] It should be noted that when the server executes step 102 above, it can determine the SOC estimation noise based on one or more of the battery parameters. Specifically, if the SOC estimation noise is determined based on multiple battery parameters, for example, the multiple battery parameters can be integrated using dot product or summation to obtain integrated data. This integrated data can then be used to determine the SOC estimation noise, thus taking into account the noise from multiple battery parameters and improving its accuracy.

[0069] In some alternative implementations, determining the SOC estimation noise based on the battery parameters in step 102 above may include: inputting the battery parameters into a preset model to obtain a predicted SOC; then, obtaining the SOC estimation noise based on the predicted SOC and the expected SOC; the expected SOC is determined based on the measured historical SOC values.

[0070] The aforementioned preset model can be, for example, a neural network model or a support vector machine. This preset model can be trained using one or more sample battery parameters. That is, one or more sample battery parameters are input into the initial preset model, and then the initial preset model is trained using the measured sample SOC as the desired output. When the error between the two is within a preset error range, it can be determined that the initial preset model has converged to the aforementioned preset model.

[0071] In some application scenarios, the server can input the acquired battery parameters into a preset model to obtain the predicted SOC, and the difference between the predicted SOC and the expected SOC can be determined as the SOC estimation noise.

[0072] It should be noted that the aforementioned expected SOC may be, for example, the historical SOC value obtained from actual battery measurements, or the average of historical SOC values ​​obtained from multiple actual measurements.

[0073] In this implementation, the SOC estimation noise can be obtained by predicting the SOC and the expected SOC. Since the expected SOC has high accuracy, the accuracy of the SOC estimation noise is also improved to a certain extent.

[0074] Step 103: Determine the SOC of the battery to be processed based on at least one of the voltage and current of the battery to be processed and the SOC-estimated noise.

[0075] The aforementioned batteries to be processed are also batteries whose SOC (State of Charge) needs to be determined.

[0076] In some application scenarios, the battery to be processed can be a battery used to determine the SOC estimation noise. In this case, the battery parameters obtained by the server can be historical battery parameters of the battery to be processed.

[0077] In other application scenarios, the battery to be processed is the same type as the battery used to determine the SOC estimation noise. In this case, the battery parameters obtained by the server can be historical battery parameters or current battery parameters of the same type. For example, if the battery used to determine the SOC estimation noise is a lithium iron phosphate battery, then the battery to be processed can be a lithium iron phosphate battery; if the battery used to determine the SOC estimation noise is a ternary lithium battery, then the battery to be processed can be a ternary lithium battery.

[0078] It should be noted that this application does not limit the type of battery applicable, which may include, for example, the aforementioned lithium iron phosphate battery, ternary lithium battery, or sodium-ion battery.

[0079] In this way, after determining the SOC estimation noise corresponding to a certain type of battery, the SOC of the same type of battery can be determined by using the same SOC estimation noise, thereby improving the accuracy of SOC while reducing the computational cost.

[0080] In some applications, a server can obtain at least one of the voltage and current of the battery to be processed, and then determine the other based on the resistance of the battery. Therefore, the SOC of the battery to be processed can be determined based on one of these values ​​and the SOC estimation noise.

[0081] In some applications, the State of Charge (SOC) of the battery can be determined using methods such as least squares or Kalman filtering. When determining the SOC using either algorithm, the noise involved can be replaced with the aforementioned SOC estimation noise during the calculation.

[0082] In this implementation, since the SOC estimation noise is related to the slope of the curve of battery parameters changing with charge and discharge time, the SOC estimation noise will change adaptively with the battery parameters, thereby improving the accuracy of the SOC of the battery to be processed.

[0083] For further details, please refer to Figure 3 The diagram illustrates an error comparison between the SOC determined based on this application and the conventional extended Kalman filter algorithm. Figure 3As shown, the accuracy of the SOC is highest when the error is 0. However, at the same time point, the error fluctuation of the SOC determined by the traditional extended Kalman filter algorithm is greater than that of the SOC determined by the method of this application. Therefore, the error of the SOC determined by the traditional extended Kalman filter algorithm is greater than that of the SOC determined by the method of this application. Thus, this application can improve the accuracy of the SOC of the battery to be processed.

[0084] It should be noted that, in Figure 3 In the error comparison chart shown, the vertical axis represents the SOC error, and the horizontal axis represents time (specifically, the charge and discharge time). The corresponding SOC error can then be determined by the relevant data at each time point.

[0085] In some alternative implementations, determining the SOC estimation noise based on the battery parameters in step 102 above includes: determining the actual noise based on the battery parameters and a preset noise level.

[0086] The aforementioned preset noise can be, for example, a preset empirical value. In some application scenarios, the actual noise can be determined using, for example, the least squares method or the Kalman filter algorithm. Specifically, if the actual noise is determined using the least squares method, the preset noise can be multiplied by the slope of the curve showing the battery parameters changing with charge / discharge time to obtain the aforementioned actual noise.

[0087] In this implementation, the actual noise can be made to adapt to changes in battery parameters, thereby improving the accuracy of battery SOC.

[0088] In related technologies, the Kalman filter algorithm can be used to determine the State of Charge (SOC). However, the Kalman filter algorithm determines the SOC based on a purely algorithmic mechanism, which leads to relatively low accuracy. Therefore, the accuracy of the SOC can be improved by leveraging the Kalman filter algorithm.

[0089] To facilitate understanding, we will first introduce the Kalman filter algorithm. Specifically, the Kalman filter algorithm is mainly used to estimate the state of a dynamic system. It uses a set of prediction equations to describe the evolution of the system's internal state over time, thus obtaining predicted values, and uses a set of observation equations to represent how to obtain observed values ​​from the true state. Then, the Kalman gain is updated based on the prediction noise corresponding to the predicted values ​​and the observation noise corresponding to the observed values. By weighting the observed and predicted values ​​with the Kalman gain, a more accurate true value is obtained.

[0090] Specifically, in some optional implementations, determining the actual noise based on the battery parameters and preset noise includes:

[0091] Sub-step 1021: Determine the actual Kalman filter noise based on the battery parameters and the preset Kalman filter noise; the actual Kalman filter noise includes the actual observation noise and the actual prediction noise of the circuit model; the actual observation noise characterizes the error in the observed SOC data of the battery; the actual prediction noise characterizes the error in the predicted SOC data of the battery.

[0092] The aforementioned preset Kalman filter noise can be, for example, a preset empirical value, which includes preset prediction noise and preset observation noise.

[0093] The circuit model described above can be regarded as the prediction model used to obtain the predicted value in the Kalman filter algorithm. It can include, for example, the Rint model, the first-order RC model, the second-order RC model, and other equivalent circuit models.

[0094] The Kalman filtering algorithm used in this implementation may include, for example, the classic Kalman filtering algorithm, the extended Kalman filtering algorithm, and the unscented Kalman filtering algorithm.

[0095] Therefore, the server can determine the actual Kalman filter noise present when calculating the SOC of the battery to be processed, by combining battery parameters and preset Kalman filter noise. The process of determining the actual Kalman filter noise will be described in detail later, and will not be repeated here.

[0096] Thus, the step 103 above, which describes determining the SOC of the battery to be processed based on at least one of the voltage and current of the battery to be processed and the SOC-estimated noise, includes: determining the actual SOC of the battery to be processed using a Kalman filter algorithm based on the voltage and current of the battery to be processed and the actual Kalman filter noise.

[0097] In some embodiments, when determining the state of charge (SOC) of a battery using a Kalman filter algorithm and an equivalent circuit model, several known parameters are required. These known parameters are the voltage U of the battery to be processed. OCV The data includes current data I, actual Kalman filter noise (including actual observed noise R and actual predicted noise Q), and circuit model parameters. In some application scenarios, if the circuit model is... Figure 4 The first-order RC circuit shown has the following circuit model parameters: first resistance parameter R0, second resistance parameter R1, and capacitance parameter C1.

[0098] Therefore, the calculation formula for the Kalman filter algorithm can be obtained:

[0099]

[0100] U k =G(X) k ,θ k ,Ik ) = OCV(SOC k )+U 1,k +I k R0+R (2)

[0101]

[0102] Wherein, calculation formula (1) represents the prediction equation in the Kalman filter algorithm, calculation formula (2) represents the observation equation in the Kalman filter algorithm, and calculation formula (3) represents the specific physical variables in the matrix; where Δt represents the time interval between the acquisition time and the previous acquisition time; Cap represents the battery capacity, k represents the time index for determining the SOC; U1 represents the voltage of the second resistor; X k+1 This is the estimated SOC value for the next time step, which is also the target value output by the Kalman filter algorithm.

[0103] In this implementation, the battery's State of Charge (SOC) is not determined based on pure algorithmic mechanisms, but rather by combining battery characteristics with the algorithmic mechanisms. This approach improves the accuracy of the battery's SOC because the actual Kalman filter noise is not a fixed constant but adapts to changes in battery parameters.

[0104] In some application scenarios, there is a correlation between the actual predicted noise and the actual observed noise. This correlation can be determined so that once one is determined, the other can be determined directly.

[0105] Therefore, in some optional implementations, the step 1021 above, which involves determining the actual Kalman filter noise based on the battery parameters and the preset Kalman filter noise, includes:

[0106] First, based on the battery parameters, determine the relationship coefficient between the actual predicted noise and the actual observed noise;

[0107] For example, the server can determine the relationship coefficient β between OCV and SOC based on the rate of change of OCV, thus obtaining the relationship Q = βR. In some optional implementations, the relationship coefficient includes at least one of the following: a proportional coefficient, an exponential coefficient, or a differential coefficient.

[0108] Then, based on the battery parameters and the preset Kalman filter noise, one noise item in the actual Kalman filter noise is determined;

[0109] It should be noted that the process of determining one noise item in the actual Kalman filter noise based on battery parameters and preset Kalman filter noise is similar to the process of determining the actual Kalman filter noise based on battery parameters and preset Kalman filter algorithm described above, and will not be repeated here.

[0110] Finally, based on one of the noises already identified in the actual Kalman filter noise and the relationship coefficient, another noise in the actual Kalman filter noise is determined.

[0111] For example, if the actual observed noise R is determined, the actual predicted noise Q can be determined according to the relationship Q = βR.

[0112] In this implementation, after determining one of the actual observed noise and the actual predicted noise, the other noise can be determined based on the relationship coefficient between the two, thereby reducing computational costs and accelerating computation speed.

[0113] In other application scenarios, the two noise items in the actual Kalman filter noise can also be determined separately, and this is not a limitation here. It should be noted that the process of determining the two noise items in the actual Kalman filter noise separately is similar to the process of determining the actual Kalman filter noise based on battery parameters and a preset Kalman filter algorithm, as described above, and will not be repeated here.

[0114] In some alternative implementations, the actual predicted noise is negatively correlated with the actual observed noise, so that after determining one type of noise, the other type of noise can be determined based on the negative correlation.

[0115] In some alternative implementations, the actual predicted noise is positively correlated with the slope of the curve showing the battery parameters changing with charge / discharge time, while the actual observed noise is negatively correlated with the slope of the curve showing the battery parameters changing with charge / discharge time.

[0116] In some applications, the actual predicted noise and the actual observed noise correspond to the slopes of the curves showing the changes in battery parameters over charge / discharge time, as described above. Thus, when the slope is small, the actual predicted noise decreases, making the predicted value at that moment more reliable; similarly, when the slope is large, the actual observed noise decreases, making the observed value at that moment more reliable.

[0117] In some alternative implementations, the actual predicted noise is the product of a first preset noise and the slope of the curve showing the battery parameters changing with charge / discharge time; the actual observed noise is the quotient of a second preset noise and the slope of the curve showing the battery parameters changing with charge / discharge time.

[0118] The aforementioned first preset noise is also the preset Kalman filter prediction noise;

[0119] The aforementioned second preset noise is also known as the preset Kalman filter observation noise.

[0120] Therefore, after the server learns the slope of the curve of battery parameters changing with charging and discharging time, it can multiply the slope by the first preset noise to obtain the actual predicted noise, and divide the second preset noise by the slope to obtain the actual observed noise, thereby improving the computing speed.

[0121] In some alternative implementations, the two noise items in the actual Kalman filter noise are respectively proportional to, exponentially related to, or logarithmically related to the slope of the curve of the battery parameters changing with charge and discharge time.

[0122] Therefore, in some application scenarios, the server can use a proportional relationship. The actual Kalman filter noise is determined, where Q represents the actual predicted noise, Q′ represents the preset predicted noise, λ represents the slope of the curve of battery parameters changing with charge and discharge time, R represents the actual observed noise, and R′ represents the preset observed noise.

[0123] In some application scenarios, servers can also use exponential relationships. Determine the actual Kalman filter noise, where γ1 and γ2 are exponential terms representing the slope of the curves of battery parameters changing with charge and discharge time.

[0124] In some application scenarios, servers can also use logarithmic relationships. Determine the actual Kalman filter noise, where a and b represent the base of the logarithm.

[0125] In this implementation, a corresponding relationship can be constructed based on the proportional, exponential, or logarithmic relationship between the actual Kalman filter noise and the slope of the curve showing the change of battery parameters with charge and discharge time. This makes the correlation between the noise value and the slope greater and the noise change more sensitive.

[0126] In some optional implementations, determining the actual Kalman filter noise based on the battery parameters and a preset Kalman filter noise includes:

[0127] If the slope of the curve showing the change of battery parameters with charge / discharge time is within any preset range, then the preset Kalman filter noise corresponding to that preset range is determined as the actual Kalman filter noise.

[0128] In some application scenarios, the server can pre-divide the slope of the curve showing the change of battery parameters with charge / discharge time into multiple intervals, and each interval can be pre-configured with a corresponding set of Kalman filter noise. In this way, the pre-set Kalman filter noise can be determined as the actual Kalman filter noise based on the interval where the slope of the curve showing the change of battery parameters with charge / discharge time falls.

[0129] In other application scenarios, the server can also set only two preset intervals, which are separated by an interval threshold. In this way, if the slope of the curve showing the change of battery parameters with charge / discharge time is greater than the interval threshold, the corresponding preset Kalman filter noise (e.g., preset prediction noise Q1 and preset observation noise R1) is determined as the actual Kalman filter noise; if the slope of the curve showing the change of battery parameters with charge / discharge time is less than the interval threshold, the corresponding preset Kalman filter noise (e.g., preset prediction noise Q2 and preset observation noise R2, where Q2 < Q1, R2 < R1) is determined as the actual Kalman filter noise.

[0130] In this implementation, the preset Kalman filter noise can be determined as the actual Kalman filter noise by the preset range in which the slope of the curve of battery parameters changes with charging and discharging time is located. This makes the actual Kalman filter noise different in different preset ranges of the slope, making the operation simpler and more convenient, and also improving the accuracy of battery SOC to a certain extent.

[0131] Please continue reading. Figure 2 As can be seen, there is a plateau period in the voltage. That is, within a certain period of time (charge and discharge time), the voltage does not change with the state of charge (SOC) or only changes slightly. At this time, the slope of the voltage-charge-discharge time curve can be considered constant. Therefore, the voltage-discharge time curve can be reflected by the voltage-SOC curve.

[0132] In some application scenarios, servers can also perform derivative or differential operations on the voltage value on the voltage-to-SOC curve to obtain, for example, Figure 5 The rate of change of voltage with SOC (i.e., the slope of the curve) is shown, thus more clearly demonstrating the change of the slope of the curve of voltage changing with the duration of energization.

[0133] Therefore, in some optional implementations, the statement that if the slope of the curve showing the change of battery parameters with charge / discharge time falls within any preset interval, then the preset Kalman filter noise corresponding to that preset interval is determined as the actual Kalman filter noise includes:

[0134] If the voltage in the battery parameters is greater than the preset upper voltage limit, or the voltage in the battery parameters is less than the preset lower voltage limit, then the first preset Kalman filter noise is determined as the actual Kalman filter noise.

[0135] If the voltage in the battery parameters is between the preset lower voltage limit and the preset upper voltage limit, then the second preset Kalman filter noise is determined as the actual Kalman filter noise.

[0136] It should be noted that the aforementioned preset upper voltage limit and preset lower voltage limit can be set according to the voltage plateau period, so that when the voltage is in a non-plateau period, the corresponding first preset Kalman filter noise is determined as the actual Kalman filter noise; and when the voltage is in a plateau period, the corresponding second preset Kalman filter noise is determined as the actual Kalman filter noise.

[0137] In this implementation, the first preset Kalman filter noise or the second preset Kalman filter noise can be adaptively determined as the actual Kalman filter noise based on the voltage value, thereby obtaining a more accurate actual Kalman filter noise and improving the accuracy of SOC to a certain extent.

[0138] In some optional implementations, if the slope of the curve showing the change of battery parameters with charge / discharge time falls within any preset interval, then the preset Kalman filter noise corresponding to that preset interval is determined as the actual Kalman filter noise, including:

[0139] If the voltage in the battery parameters is greater than the preset upper limit of voltage, then the third preset Kalman filter noise is determined as the actual Kalman filter noise;

[0140] If the voltage in the battery parameters is less than the preset lower voltage limit, then the fourth preset Kalman filter noise is determined as the actual Kalman filter noise;

[0141] If the voltage in the battery parameters is between the preset lower voltage limit and the preset upper voltage limit, then the fifth preset Kalman filter noise is determined as the actual Kalman filter noise.

[0142] Similarly, the aforementioned preset upper voltage limit and preset lower voltage limit can also be set according to the voltage plateau period to correspond to different preset Kalman filter noise when the voltage is in the plateau period or not.

[0143] Unlike the previous implementation, this implementation pre-sets three Kalman filter noises, corresponding to voltages greater than a preset upper voltage limit, voltages less than a preset upper voltage limit, and voltages between the preset upper voltage limit and the preset upper voltage limit, respectively. Therefore, this implementation sets multiple preset Kalman filter noises with a smaller preset granularity, thereby determining a more accurate actual Kalman filter noise and improving the accuracy of the SOC to some extent.

[0144] Furthermore, since the voltage also changes with the SOC, the actual Kalman filter noise can be determined based on the rate of change of the voltage with the SOC data.

[0145] Therefore, in some optional implementations, if the slope of the curve showing the change of battery parameters with charge / discharge time is within any preset interval, then the preset Kalman filter noise corresponding to that preset interval is determined as the actual Kalman filter noise, including:

[0146] If the rate of change of voltage in the battery parameters with SOC data is greater than a preset rate of change threshold, then the sixth preset Kalman filter noise is determined as the actual Kalman filter noise.

[0147] If the rate of change of the voltage in the battery parameters with the SOC data is less than a preset rate of change threshold, then the seventh preset Kalman filter noise is determined as the actual Kalman filter noise.

[0148] In other words, the server can set two preset intervals, which are separated by a preset rate of change threshold. This preset rate of change threshold could be, for example, 0.1, 0.2, or 0.3. Thus, if the rate of change of voltage with SOC data is greater than this preset rate of change threshold, the corresponding sixth preset Kalman filter noise (e.g., preset prediction noise Q1 and preset observation noise R1) is determined as the actual Kalman filter noise; if the rate of change of voltage with SOC data is less than this preset rate of change threshold, the corresponding seventh preset Kalman filter noise (e.g., preset prediction noise Q2 and preset observation noise R2, where Q2 < Q1, R2 > R1) is determined.

[0149] In this implementation, the preset Kalman filter noise can be determined as the actual Kalman filter noise by the preset range in which the rate of change of voltage with SOC data falls. This allows the actual Kalman filter noise to adaptively change with the rate of change of voltage with SOC data, thereby improving the accuracy of the actual Kalman filter noise to a certain extent and thus improving the accuracy of battery SOC.

[0150] In some alternative implementations, before determining the actual SOC of the battery to be processed using a Kalman filter algorithm based on the voltage, current, and actual Kalman filter noise of the battery to be processed, the method further includes: identifying the circuit model parameters.

[0151] In some application scenarios, if the circuit model is as follows Figure 4 For the first-order circuit model shown, the circuit model parameters that need to be identified are the first resistance parameter R0, the second resistance parameter R1, and the capacitance parameter C1. In these application scenarios, the circuit model parameters can be obtained through direct measurement.

[0152] In some alternative implementations, identifying the circuit model parameters includes: identifying the circuit model parameters based on the circuit model and the actual Kalman filter noise.

[0153] In other words, the server can use the Kalman filter algorithm to input the voltage and current data of the battery to be processed or the corresponding battery of the same type, as well as the actual Kalman filter noise, into the Kalman filter algorithm calculation formula to obtain the circuit model parameters.

[0154] That is, through the following calculation formula of the Kalman filtering algorithm:

[0155] θ k+1 =θ k +Q (1);

[0156] U k =G(χ) k ,θ k ,I k ) = OCV(SOC k )+U 1,k +I k R0+R (2);

[0157]

[0158] Thus, the parameters of each circuit model are obtained. Among them, OCV(SOC) k The parameter ) represents the OCV corresponding to the current SOC, and the meanings of the other parameters are the same as before. This helps to improve the accuracy of the SOC, thus obtaining a more accurate SOC.

[0159] In some alternative implementations, identifying the circuit model parameters includes: identifying the circuit model parameters based on the circuit model and the recursive least squares method.

[0160] In other words, the server can use the recursive least squares algorithm to substitute relevant parameters into its calculation formula to obtain the parameters of each circuit model. This makes it simpler and more convenient to solve for the circuit model parameters, thus improving computational efficiency to some extent.

[0161] In this implementation, different methods can be used to identify the circuit model parameters according to actual needs, so as to determine the SOC of the battery to be processed.

[0162] In some application scenarios, the identified circuit model parameters may be inaccurate. For example, the identified circuit model parameters may be outside a reasonable range (e.g., negative numbers) or have a large error compared to the actual circuit model parameters. Therefore, in some optional implementations, after identifying the circuit model parameters, the method further includes: correcting the circuit model parameters according to a correction rule.

[0163] In some optional implementations, if the circuit model parameter is greater than the upper limit of the preset identification range, the circuit model parameter is corrected to the upper limit of the preset identification range; if the circuit model parameter is less than the lower limit of the preset identification range, the circuit model parameter is corrected to the lower limit of the preset identification range.

[0164] For example, the preset recognition range of the first resistance parameter R0 is (0.1~1)mΩ. If the first resistance parameter R0 is greater than 1mΩ, it is corrected to 1mΩ; if the first resistance parameter R0 is less than 0.1mΩ, it is corrected to 0.1mΩ.

[0165] The preset recognition range of the second resistance parameter R1 is (0.1~1)mΩ. If the second resistance parameter R1 is greater than 1mΩ, it is corrected to 1mΩ; if the second resistance parameter R1 is less than 0.1mΩ, it is corrected to 0.1mΩ.

[0166] The preset recognition range of capacitor parameter C1 is (0.9~1.1)F. If capacitor parameter C1 is greater than 1.1F, it will be corrected to 1.1F; if capacitor parameter C1 is less than 0.9F, it will be corrected to 0.9F.

[0167] In this implementation, the circuit model parameters can be limited to a preset identification range, thereby improving the accuracy of the circuit model parameters and avoiding circuit model discrepancies.

[0168] In some optional implementations, if the rate of change between the circuit model parameters identified in the current frame and the circuit model parameters identified in the previous frame is greater than the upper limit of the rate of change range, then the circuit model parameters identified in the current frame are corrected according to the upper limit of the rate of change range.

[0169] The aforementioned rate of change could be, for example, the ratio of the circuit model parameters identified in the current frame to the circuit model parameters identified in the previous frame, or the quotient of the difference between the two and the circuit model parameters identified in the previous frame.

[0170] In some application scenarios, the rate of change of the first resistance parameter between two adjacent frames can range from (0.8 to 1.2) times. If the first resistance parameter identified in the current frame is R0′ k+1 R0′ k+1 For the previous frame R 0k If it is 1.5 times, then it can be limited to 1.2 times. That is, the first resistance parameter R of the current frame is... 0k+1 Revised to R 0k+1 =1.2×R 0k If the first resistor parameter R0′ identified in the current frame k+1 For the previous frame R 0k If it is 0.9 times, then no correction is needed. In this case, the first resistance parameter R0k+1 To identify R0′ k+1 .

[0171] In this implementation, considering that circuit model parameters do not change abruptly in a short period of time, if a pure algorithmic mechanism is used to identify the circuit model parameters, the values ​​may vary significantly. This does not conform to actual laws. Therefore, the rate of change of the circuit model parameters identified in the current frame is limited to obtain circuit model parameters that conform to actual laws, thereby improving the accuracy of the SOC of the battery to be processed to a certain extent.

[0172] In some optional implementations, the server may also weight the circuit model parameters identified in the current frame and the circuit model parameters identified in the previous frame to correct the circuit model parameters identified in the current frame.

[0173] For example, the first resistance parameter R 0k+1 The first resistance parameter R0′ identified in the current frame k+1 The first resistance parameter R identified in the previous frame 0k There is a relationship between them: in, The first resistance parameter R identified in the previous frame 0k The weighting coefficients can be, for example, empirical values.

[0174] In this implementation, the circuit model parameters identified in two adjacent frames can be weighted to prevent large abrupt changes in the circuit model parameters identified in two adjacent frames, thereby improving the accuracy of the SOC.

[0175] In some applications, sensors are typically used to collect battery current and voltage data. While it's possible to correct the voltage data after the battery has been left to rest for a period of time, this resting condition is often difficult to meet, and a linear relationship between the voltage data and the State of Charge (SOC) is required, whereas current data often doesn't undergo this correction. Therefore, using the raw collected data to determine the SOC has relatively low accuracy.

[0176] Therefore, in some optional implementations, before determining the actual SOC of the battery to be processed using a Kalman filter algorithm based on the voltage, current, and actual Kalman filter noise of the battery to be processed, the method further includes:

[0177] First, if the voltage of the battery to be processed in the current frame is not within the preset voltage data range, then the voltage is corrected to the voltage of the previous frame;

[0178] The aforementioned preset voltage data range may include, for example, (1 to 5)V. Therefore, if the voltage of the current frame is greater than 5V or less than 1V, it can be corrected to the voltage of the previous frame.

[0179] In this way, the battery to be processed does not need to be left to stand still, and its voltage does not need to have a linear relationship with the SOC, making the correction process relatively simple. Moreover, since the voltage often does not change abruptly, correcting to the voltage of the previous frame will not reduce the accuracy of the SOC.

[0180] Secondly, if the current data of the battery to be processed in the current frame is not within the preset current data range, the current data is corrected to the preset current data.

[0181] The aforementioned preset current data range may include, for example, (-200 to 200) A. Therefore, if the current data in the current frame is greater than 200 A or less than -200 A, it can be corrected to the preset current data. The aforementioned preset current data may, for example, be 0.

[0182] In this implementation, the collected voltage and current data can be specifically corrected, thereby reducing the errors caused by both and improving the accuracy of the SOC.

[0183] Those skilled in the art will understand that, in the methods described in the specific embodiments, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0184] Please see Figure 6 The diagram illustrates a flowchart of a battery charging and discharging strategy determination method provided in an embodiment of this application. Figure 6 As shown, the method includes:

[0185] Step 601: During the battery charging and discharging process, battery parameters are acquired, including at least one of voltage, current, and temperature;

[0186] Step 602: Determine the SOC estimation noise based on the battery parameters, wherein the SOC estimation noise is related to the slope of the curve showing the change of battery parameters with charge and discharge time;

[0187] Step 603: Determine the SOC of the battery to be processed based on at least one of the voltage and current of the battery to be processed and the SOC-estimated noise.

[0188] It should be noted that the implementation process and technical effects of steps 601 to 703 above may be the same as or similar to the implementation process and technical effects of steps 101 to 103 above, and will not be repeated here.

[0189] Step 604: Determine the charging and discharging strategy based on the SOC of the battery to be processed.

[0190] In some applications, if the state of charge (SOC) of the battery to be processed is low, the charging current can be reduced and the charge-discharge interval can be extended. Similarly, if the SOC of the battery to be processed is high, the charging current can be increased and the charge-discharge interval can be shortened.

[0191] In other applications, if there are multiple batteries to be processed, they can be charged and discharged in a balanced manner. For example, the battery with the lower SOC can be charged first, followed by the battery with the slightly higher SOC, to bring the SOC of all batteries closer together. Alternatively, during discharge, the battery with the higher SOC can be discharged first, followed by the battery with the slightly lower SOC. This can mitigate the damage to the battery with the lower SOC caused by over-discharge and also help to bring the SOC of all batteries closer together.

[0192] In this implementation, a suitable charging and discharging strategy can be determined based on the SOC of the battery to be processed, which improves the service life of the battery to be processed to a certain extent.

[0193] Please refer to Figure 7 This diagram illustrates a structural block diagram of a battery SOC determination device according to an embodiment of this application. The battery SOC determination device can be a module, program segment, or code on an electronic device. It should be understood that this device is similar to the one described above. Figure 1 The method implementation corresponds to this and can be executed. Figure 1 The various steps involved in the method implementation examples.

[0194] Optionally, the device for determining the SOC of the battery includes an acquisition module 701, a noise determination module 702, and a SOC determination module 703. The acquisition module 701 is used to acquire battery parameters, including at least one of voltage, current, and temperature. The noise determination module 702 is used to determine the SOC estimation noise based on the battery parameters, wherein the SOC estimation noise is related to the slope of the curve showing the battery parameters changing with charge / discharge time. The SOC determination module 703 is used to determine the SOC of the battery to be processed based on at least one of the voltage and current of the battery and the SOC estimation noise.

[0195] Optionally, the noise determination module 702 further determines the actual noise based on the battery parameters and the preset noise.

[0196] Optionally, the noise determination module 702 is further configured to: determine the actual Kalman filter noise based on the battery parameters and the preset Kalman filter noise; the actual Kalman filter noise includes the actual observed noise and the actual predicted noise of the circuit model; the actual observed noise characterizes the error in the observed SOC data of the battery; the actual predicted noise characterizes the error in the predicted SOC data of the battery; and the SOC determination module 703 is further configured to: determine the actual SOC of the battery to be processed using a Kalman filter algorithm based on the voltage and current of the battery to be processed and the actual Kalman filter noise.

[0197] Optionally, the noise determination module 702 is further configured to: determine the relationship coefficient between the actual predicted noise and the actual observed noise based on the battery parameters; determine one noise item in the actual Kalman filter noise based on the battery parameters and the preset Kalman filter noise; and determine another noise item in the actual Kalman filter noise based on the determined noise item in the actual Kalman filter noise and the relationship coefficient.

[0198] Optionally, the actual predicted noise is negatively correlated with the actual observed noise.

[0199] Optionally, the actual predicted noise is positively correlated with the slope of the curve showing the change of battery parameters with charge / discharge time, and the actual observed noise is negatively correlated with the slope of the curve showing the change of battery parameters with charge / discharge time.

[0200] Optionally, the actual predicted noise is the product of a first preset noise and the slope of the curve showing the change of battery parameters with charge / discharge time; the actual observed noise is the quotient of a second preset noise and the slope of the curve showing the change of battery parameters with charge / discharge time.

[0201] Optionally, the two noise items in the actual Kalman filter noise are respectively proportional, exponential, or logarithmic to the slope of the curve showing the change of battery parameters with charge and discharge time.

[0202] Optionally, the relationship coefficient includes at least one of the following: proportional coefficient, exponential coefficient, and differential coefficient.

[0203] Optionally, the noise determination module 702 is further configured to: if the slope of the curve of the battery parameters changing with charging and discharging time is within any preset range, then determine the preset Kalman filter noise corresponding to the preset range as the actual Kalman filter noise.

[0204] Optionally, the noise determination module 702 is further configured to: if the voltage in the battery parameters is greater than a preset upper voltage limit or less than a preset lower voltage limit, then determine the first preset Kalman filter noise as the actual Kalman filter noise; if the voltage in the battery parameters is between the preset lower voltage limit and the preset upper voltage limit, then determine the second preset Kalman filter noise as the actual Kalman filter noise.

[0205] Optionally, the noise determination module 702 is further configured to: if the voltage in the battery parameters is greater than a preset upper voltage limit, determine the third preset Kalman filter noise as the actual Kalman filter noise; if the voltage in the battery parameters is less than a preset lower voltage limit, determine the fourth preset Kalman filter noise as the actual Kalman filter noise; if the voltage in the battery parameters is between the preset lower voltage limit and the preset upper voltage limit, determine the fifth preset Kalman filter noise as the actual Kalman filter noise.

[0206] Optionally, the noise determination module 702 is further configured to: if the rate of change of the voltage in the battery parameters with the SOC data is greater than a preset rate of change threshold, then determine the sixth preset Kalman filter noise as the actual Kalman filter noise; if the rate of change of the voltage in the battery parameters with the SOC data is less than a preset rate of change threshold, then determine the seventh preset Kalman filter noise as the actual Kalman filter noise.

[0207] Optionally, the noise determination module 702 is further configured to: input the battery parameters into a preset model to obtain a predicted SOC; obtain an estimated SOC noise based on the predicted SOC and the expected SOC; and determine the expected SOC based on the measured historical SOC values.

[0208] Optionally, the SOC estimation noise includes current noise, voltage noise, and / or temperature noise.

[0209] Optionally, the device further includes an identification module, which is used to identify the circuit model parameters before determining the actual SOC of the battery to be processed using a Kalman filtering algorithm based on the voltage, current and actual Kalman filter noise of the battery to be processed.

[0210] Optionally, the identification module is further configured to: identify the circuit model parameters based on the circuit model and the actual Kalman filter noise.

[0211] Optionally, the identification module is further configured to: identify the circuit model parameters based on the circuit model and the recursive least squares method.

[0212] Optionally, the device further includes a first correction module, which is used to: correct the circuit model parameters according to correction rules after identifying the circuit model parameters.

[0213] Optionally, the first correction module is further configured to: if the circuit model parameter is greater than the upper limit of the preset identification range, then correct the circuit model parameter to the upper limit of the preset identification range; if the circuit model parameter is less than the lower limit of the preset identification range, then correct the circuit model parameter to the lower limit of the preset identification range.

[0214] Optionally, the first correction module is further configured to: if the rate of change between the circuit model parameters identified in the current frame and the circuit model parameters identified in the previous frame is greater than the upper limit of the range of the rate of change, then correct the circuit model parameters identified in the current frame according to the upper limit of the range of the rate of change.

[0215] Optionally, the first correction module is further configured to: weight the circuit model parameters identified in the current frame and the circuit model parameters identified in the previous frame to correct the circuit model parameters identified in the current frame.

[0216] Optionally, the device further includes a second correction module, which is configured to: before determining the actual SOC of the battery to be processed using a Kalman filter algorithm based on the voltage, current, and actual Kalman filter noise of the battery to be processed, if the voltage of the battery to be processed in the current frame is not within a preset voltage data range, then correct the voltage to the voltage of the previous frame; and / or, if the current data of the battery to be processed in the current frame is not within a preset current data range, then correct the current data to the preset current data.

[0217] Optionally, the battery to be processed is the same type of battery used to determine the SOC estimation noise.

[0218] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0219] Please refer to Figure 8 , Figure 8This is a schematic diagram of an electronic device for performing a method for determining the State of Charge (SOC) of a battery, provided in an embodiment of this application. The electronic device may include: at least one processor 801, such as a CPU; at least one communication interface 802; at least one memory 803; and at least one communication bus 804. The communication bus 804 is used to establish direct communication between these components. In this embodiment, the communication interface 802 is used for signaling or data communication with other node devices. The memory 803 may be a high-speed RAM or a non-volatile memory, such as at least one disk storage device. Optionally, the memory 803 may also be at least one storage device located remotely from the aforementioned processor. The memory 803 stores computer-readable instructions, which, when executed by the processor 801, enable the electronic device to perform the aforementioned... Figure 1 The method and process are shown.

[0220] Understandable. Figure 8 The structure shown is for illustrative purposes only; the electronic device may also include components that are more advanced than those shown. Figure 8 The more or fewer components shown, or having the same Figure 8 The different configurations shown. Figure 8 The components shown can be implemented using hardware, software, or a combination thereof.

[0221] This application provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it can perform actions such as... Figure 1 The method process executed by the electronic device in the illustrated method embodiment.

[0222] This application provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer can perform the methods provided in the above-described method embodiments. For example, the method may include: acquiring battery parameters, the battery parameters including at least one of voltage, current, and temperature; determining SOC estimation noise based on the battery parameters, wherein the SOC estimation noise is related to the slope of the curve of the battery parameters changing with charge and discharge time; and determining the SOC of the battery to be processed based on at least one of voltage and current of the battery to be processed and the SOC estimation noise.

[0223] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0224] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0225] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0226] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0227] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method of determining SOC of a battery, characterized by, The method comprises: acquiring a battery parameter, the battery parameter comprising at least one of voltage, current and temperature; determining a SOC estimation noise according to the battery parameter, wherein the SOC estimation noise is related to a slope of a curve of the battery parameter changing with charging and discharging time; determining a SOC of a battery to be processed according to at least one of voltage and current of the battery to be processed and the SOC estimation noise.

2. The method of claim 1, wherein, The determining of the SOC estimation noise according to the battery parameter comprises: determining an actual noise according to the battery parameter and a preset noise.

3. The method of claim 2, wherein, The determining of the actual noise according to the battery parameter and the preset noise comprises: determining an actual Kalman filter noise according to the battery parameter and a preset Kalman filter noise; the actual Kalman filter noise comprises an actual observation noise and an actual prediction noise of a circuit model; the actual observation noise represents an error existing in observed SOC data of the battery; the actual prediction noise represents an error existing in predicted SOC data of the battery; and The determining of the SOC of the battery to be processed according to at least one of voltage and current of the battery to be processed and the SOC estimation noise comprises: determining an actual SOC of the battery to be processed by using a Kalman filter algorithm according to voltage, current of the battery to be processed and the actual Kalman filter noise.

4. The method of claim 3, wherein, The determining of the actual Kalman filter noise according to the battery parameter and the preset Kalman filter noise comprises: determining a relationship coefficient between the actual prediction noise and the actual observation noise according to the battery parameter; determining one noise in the actual Kalman filter noise according to the battery parameter and the preset Kalman filter noise; and determining another noise in the actual Kalman filter noise according to the determined one noise in the actual Kalman filter noise and the relationship coefficient.

5. The method of claim 4, wherein, The actual prediction noise is negatively related to the actual observation noise.

6. The method of claim 5, wherein, The actual prediction noise is positively related to the slope of the curve of the battery parameter changing with charging and discharging time, and the actual observation noise is negatively related to the slope of the curve of the battery parameter changing with charging and discharging time.

7. The method of claim 6, wherein, The actual prediction noise is a product of a first preset noise and the slope of the curve of the battery parameter changing with charging and discharging time, and the actual observation noise is a quotient of a second preset noise and the slope of the curve of the battery parameter changing with charging and discharging time.

8. The method of claim 7, wherein, The two noises in the actual Kalman filter noise are in proportional, exponential or logarithmic relationship with the slope of the curve of the battery parameter changing with charging and discharging time.

9. The method of claim 4, wherein, The relationship coefficient at least comprises one of a proportional coefficient, an exponential coefficient and a differential coefficient.

10. The method according to any one of claims 3-9, characterized in that, The determining of the actual Kalman filter noise according to the battery parameter and the preset Kalman filter noise comprises: if the slope of the curve of the battery parameter changing with charging and discharging time is in any preset interval, then a preset Kalman filter noise corresponding to the preset interval is determined as the actual Kalman filter noise.

11. The method of claim 10, wherein, If the slope of the curve of the battery parameter changing with the charging and discharging time is within any preset interval, a preset Kalman filter noise corresponding to the preset interval is determined as the actual Kalman filter noise, including: If the voltage in the battery parameter is greater than a preset upper voltage limit value, or the voltage in the battery parameter is less than a preset lower voltage limit value, a first preset Kalman filter noise is determined as the actual Kalman filter noise. If the voltage in the battery parameter is between the preset lower voltage limit value and the preset upper voltage limit value, a second preset Kalman filter noise is determined as the actual Kalman filter noise.

12. The method of claim 10, wherein, If the slope of the curve of the battery parameter changing with the charging and discharging time is within any preset interval, a preset Kalman filter noise corresponding to the preset interval is determined as the actual Kalman filter noise, including: If the voltage in the battery parameter is greater than a preset upper voltage limit value, a third preset Kalman filter noise is determined as the actual Kalman filter noise. If the voltage in the battery parameter is less than a preset lower voltage limit value, a fourth preset Kalman filter noise is determined as the actual Kalman filter noise. If the voltage in the battery parameter is between the preset lower voltage limit value and the preset upper voltage limit value, a fifth preset Kalman filter noise is determined as the actual Kalman filter noise.

13. The method of claim 10, wherein, If the slope of the curve of the battery parameter changing with the charging and discharging time is within any preset interval, a preset Kalman filter noise corresponding to the preset interval is determined as the actual Kalman filter noise, including: If the rate of change of the voltage in the battery parameter changing with the SOC data is greater than a preset rate of change threshold, a sixth preset Kalman filter noise is determined as the actual Kalman filter noise. If the rate of change of the voltage in the battery parameter changing with the SOC data is less than a preset rate of change threshold, a seventh preset Kalman filter noise is determined as the actual Kalman filter noise.

14. The method of claim 1, wherein, The method further includes: Inputting the battery parameter into a preset model to obtain a predicted SOC; Obtaining an SOC estimation noise according to the predicted SOC and an expected SOC, wherein the expected SOC is determined according to a measured historical SOC value.

15. The method of claim 1, wherein, The SOC estimation noise includes a current noise, a voltage noise and / or a temperature noise.

16. The method of any one of claims 3-9, 11-13, wherein, The method further includes: Identifying the circuit model parameter.

17. The method of claim 16, wherein, The method further includes: Identifying the circuit model parameter according to the circuit model and the actual Kalman filter noise.

18. The method of claim 16, wherein, The method further includes: Identifying the circuit model parameter according to the circuit model and a recursive least square method.

19. The method according to any of claims 17-18, characterized by, The method further includes: Correcting the circuit model parameter according to a correction rule.

20. The method of claim 19, wherein, The method further includes: Correcting the circuit model parameter according to a correction rule. if the circuit model parameter is greater than an upper limit value of a preset identification range, the circuit model parameter is corrected to the upper limit value of the preset identification range; if the circuit model parameter is less than a lower limit value of a preset identification range, the circuit model parameter is corrected to the lower limit value of the preset identification range.

21. The method of claim 19, wherein, The correcting the circuit model parameter according to the correction rule comprises: if a change rate between the circuit model parameter identified in the current frame and the circuit model parameter identified in the last frame is greater than an upper limit value of a change rate range, the circuit model parameter identified in the current frame is corrected according to the upper limit value of the change rate range.

22. The method of claim 19, wherein, The correcting the circuit model parameter according to the correction rule comprises: weighting the circuit model parameter identified in the current frame and the circuit model parameter identified in the last frame to correct the circuit model parameter identified in the current frame.

23. The method of claim 3, wherein, Before the determining the actual SOC of the battery to be processed by using the Kalman filtering algorithm according to the voltage, the current of the battery to be processed and the actual Kalman filtering noise, the method further comprises: if the voltage of the current frame of the battery to be processed is not within a preset voltage data range, the voltage is corrected to the voltage of the last frame; and / or if the current data of the current frame of the battery to be processed is not within a preset current data range, the current data is corrected to preset current data.

24. The method of claim 1, wherein, The battery to be processed is of the same type as the battery used to determine the SOC estimation noise.

25. A method for determining a battery charging and discharging strategy, characterized in that, The method comprises: acquiring battery parameters in the process of charging and discharging of the battery, the battery parameters comprising at least one of voltage, current and temperature; determining a SOC estimation noise according to the battery parameters, wherein the SOC estimation noise is related to a slope of a curve of the battery parameters changing with charging and discharging time; determining a SOC of the battery to be processed according to at least one of voltage and current of the battery to be processed and the SOC estimation noise; determining a charging and discharging strategy according to the SOC of the battery to be processed.

26. A device for determining the SOC of a battery, characterized in that The method comprises: an acquiring module configured to acquire battery parameters, the battery parameters comprising at least one of voltage, current and temperature; a noise determining module configured to determine a SOC estimation noise according to the battery parameters, wherein the SOC estimation noise is related to a slope of a curve of the battery parameters changing with charging and discharging time; a SOC determining module configured to determine a SOC of the battery to be processed according to at least one of voltage and current of the battery to be processed and the SOC estimation noise.

27. An electronic device, comprising: The computer program is executed by the processor to run the method of any one of claims 1-25.

28. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to run the method of any one of claims 1-25.

29. A computer program product comprising computer programs or instructions, characterized in that, The computer program or instruction is executed by the processor to run the method of any one of claims 1-25.

Images