Battery remaining capacity estimation method and device, control device and storage medium

Abstract

The invention relates to a battery remaining capacity estimation method and device, a control device and a storage medium, and the method comprises the steps: obtaining the battery information of a current vehicle, and the battery information comprises the current remaining capacity; obtaining a current cell output voltage according to the battery information; calculating by using unscented Kalman filtering according to the current output voltage of the battery cell and the current residual electric quantity to obtain the actual residual electric quantity; and correcting the current residual electric quantity according to the actual residual electric quantity. And obtaining the current cell output voltage of the predicted value based on the battery information, and obtaining the corrected actual residual electric quantity based on the current cell output voltage of the predicted value and the current residual electric quantity. The actual residual electric quantity obtained through the battery information and the unscented Kalman filtering analysis is combined with the current residual electric quantity and the battery information, and compared with the prior art, the precision is higher, the fitting performance is better, and the error is smaller.

Description

Technical Field

[0001] The invention relates to the field of automobile energy and automobile state monitoring, and specifically provides a method, a device, a control device and a storage medium for estimating the remaining power of a battery. Background Art

[0002] Accurately estimating the remaining capacity (SOC) of a battery helps prevent overcharging and overdischarging of the battery. However, the remaining capacity of a battery is an indirect quantity that cannot be directly measured. Currently, the remaining capacity of a battery can only be obtained through indirect estimation.

[0003] Currently, the commonly used method for estimating the remaining battery capacity is the ampere-hour integration method. However, the remaining battery capacity obtained by the ampere-hour integration method may be inaccurate due to error accumulation.

[0004] Accordingly, the art needs a new solution for calculating the remaining battery power to solve the above problem. Summary of the invention

[0005] In order to overcome the above defects, the present invention is proposed to provide a solution or at least partially solve the problem of inaccurate battery SOC estimation in the prior art.

[0006] In a first aspect, the present invention provides a method for estimating a remaining battery power, the method comprising: obtaining battery information of a current vehicle, the battery information including a current remaining power; calculating based on the battery information to obtain a current battery cell output voltage; calculating based on the current battery cell output voltage and the current remaining power and using an unscented Kalman filter to obtain an actual remaining power; and correcting the current remaining power based on the actual remaining power.

[0007] As an alternative or supplement to the above scheme, in a method according to an embodiment of the present invention, deriving the current battery cell output voltage based on the battery information includes: inputting the battery information into a trained battery cell voltage neural network model to obtain the current battery cell output voltage.

[0008] As an alternative or supplement to the above solution, in a method according to an embodiment of the present invention, the battery information further includes: current status information of the battery and historical status information of the battery.

[0009] As an alternative or supplement to the above solution, in a method according to an embodiment of the present invention, the current status information of the battery includes at least one current status information, and the historical status information of the battery includes at least two historical status information.

[0010] As an alternative or supplement to the above solution, in a method according to an embodiment of the present invention, the current state information of the battery includes: the current of the current step; the historical state information of the battery includes: the current one step ago, the voltage one step ago, and the voltage two steps ago.

[0011] As an alternative or supplement to the above solution, in a method according to an embodiment of the present invention, correcting the current remaining power according to the actual remaining power includes: obtaining a first difference between the current remaining power and the actual remaining power based on the current remaining power; if the first difference is greater than a preset first threshold, the current remaining power is inaccurate and the current remaining power is replaced by the actual remaining power.

[0012] As an alternative or supplement to the above scheme, in a method according to an embodiment of the present invention, the actual remaining power is obtained based on the current battery cell output voltage and the current remaining power and using unscented Kalman filtering, including: constructing a corresponding Sigma point set based on the current remaining power through insensitive transformation; obtaining the corresponding predicted voltage based on the Sigma point set; obtaining the Kalman gain at the current moment according to the predicted voltage and the Sigma point set; and correcting the current remaining power based on the Kalman gain to obtain the actual remaining power.

[0013] In a second aspect, a device for estimating a remaining battery power is provided, the device comprising: a vehicle information acquisition module for acquiring battery information of a current vehicle, the battery information including a current remaining power; a model voltage calculation module for obtaining a current battery cell output voltage according to the battery information; an actual power calculation module for calculating an actual remaining power according to the current battery cell output voltage and the current remaining power using an unscented Kalman filter; and a current power correction module for correcting the current remaining power according to the actual remaining power.

[0014] In a third aspect, a control device is provided, which includes a processor and a storage device, wherein the storage device is suitable for storing multiple computer programs, and the computer programs are suitable for being loaded and run by the processor to execute the method for estimating the remaining battery power described in any one of the technical solutions of the above-mentioned method for estimating the remaining battery power.

[0015] In a fourth aspect, a computer-readable storage medium is provided, which stores a plurality of computer programs, wherein the computer programs are suitable for being loaded and run by a processor to execute the method for estimating the remaining battery power described in any one of the technical solutions of the above-mentioned method for estimating the remaining battery power.

[0016] The above one or more technical solutions of the present invention have at least one or more of the following Beneficial effects:

[0017] In the technical solution of the present invention, the predicted value of the current cell output voltage is obtained based on the battery information, and the corrected actual remaining power is obtained based on the predicted value of the current cell output voltage and the current remaining power. The actual remaining power obtained by analyzing the battery information and the unscented Kalman filter combines the battery information with the current remaining power, and has higher accuracy, better fit, and smaller error than the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The disclosure of the present invention will become more easily understood with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present invention. In addition, similar numbers in the figures are used to represent similar components, among which:

[0019] Figure 1 is a schematic flow chart of main steps of a method for estimating remaining battery power according to an embodiment of the present invention;

[0020] Figure 2 is a partial Kalman filter flow chart of a method for estimating the remaining battery capacity according to an embodiment of the present invention;

[0021] Figure 3 is a secondary flow chart of a method for estimating the remaining battery power according to an embodiment of the present invention;

[0022] Figure 4 is a graph showing the relationship between the battery SOC and the open circuit voltage according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] Some embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0024] In the description of the present invention, "module" and "processor" may include hardware, software or a combination of the two. A module may include hardware circuits, various suitable sensors, communication ports, and memories, and may also include software parts, such as computer programs, or a combination of software and hardware. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, hardware, or a combination of the two. Non-temporary computer-readable storage media include any suitable medium that can store computer programs, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The term "at least one A or B" or "at least one of A and B" has a similar meaning to "A and / or B", and may include only A, only B, or A and B. The singular terms "one" and "the" may also include plural forms.

[0025] The directional terms used in this document, such as "front", "front side", "front part", "rear", "rear side" and "rear part", are based on the front-to-back direction of the vehicle after the component is installed on the vehicle. The "longitudinal", "longitudinal" and "longitudinal section" mentioned in this document are based on the front-to-back direction after the component is installed on the vehicle, while "transverse", "transverse" and "cross section" refer to the direction perpendicular to the longitudinal direction.

[0026] Here we first explain some terms involved in the present invention.

[0027] Remaining capacity: State of charge (abbreviated as: SOC), that is, the state of charge, is used to reflect the remaining capacity of the battery. Its value is defined as the ratio of the remaining capacity to the battery capacity, usually expressed as a percentage. Its value range is 0 to 1. When SOC = 0, it means that the battery is fully discharged, and when SOC = 1, it means that the battery is fully charged.

[0028] In addition to the method described in the background technology, the prior art also has a technical solution for estimating the remaining power of a vehicle by using a Kalman filter, as follows:

[0029] A Chinese patent document with document number CN 103616647A discloses a battery remaining capacity estimation method for an electric vehicle battery management system, wherein the method comprises: 1) initializing the starting battery remaining capacity of the battery; 2) predicting the remaining capacity at the next moment according to the ampere-hour current integration method, and obtaining the battery polarization state through the current polarization model; 3) estimating the open circuit voltage according to the open circuit voltage model; 4) estimating the battery voltage; 5) comparing the actual sampled voltage with the estimated battery voltage, and calculating the voltage error; 6) obtaining the corrected remaining capacity through the Kalman filter model; 7) introducing a temperature and battery life compensation strategy, and estimating the remaining capacity based on the corrected remaining capacity.

[0030] The above document uses the ampere-hour current integration method and the Kalman filter to estimate the remaining battery capacity. Compared with the ampere-hour integration method, it has some improvement, but it is only applicable to some relatively linear batteries. That is, the Kalman filter only fits well under linear conditions, and the error in estimating the remaining battery capacity is large.

[0031] Therefore, the technical solution in the present invention will solve the problem of large errors in the ampere-hour current integration method and Kalman filtering used in the above-mentioned literature.

[0032] See attached Figure 1 , Figure 1 FIG. 1 is a flow chart showing the main steps of a method for estimating the remaining battery power according to an embodiment of the present invention. Figure 1 As shown, the method for estimating the remaining battery power in the embodiment of the present invention mainly includes the following steps S10 to S30.

[0033] Step S10: Obtain the battery information of the current vehicle.

[0034] In this embodiment, the battery information is information reflecting the battery status, wherein the battery information can be direct information or indirect information. Direct information refers to numerical information that can be directly measured by an instrument, such as the voltage across the battery, the current flowing through the circuit in the working state, and the temperature of the battery. Indirect information is numerical information that is not easy or cannot be directly measured by current means. This information often needs to be directly derived from the direct information of the battery or obtained by estimation, such as the health of the battery and the remaining power of the battery.

[0035] In one embodiment, taking the lithium iron phosphate battery commonly used in automobiles as an example, the battery information includes: remaining power information, current information, and voltage information. Obtaining the battery information of the vehicle is to obtain the battery information of the vehicle from a cloud database. In this embodiment, the battery information of the vehicle is collected regularly, and the collected data is stored in the cloud. When calculating the SOC of the vehicle battery, the battery information of the vehicle is obtained from the cloud database.

[0036] It should be noted that the remaining power information in the cloud includes at least two types of remaining power information, at least one of which is the SOC displayed by the vehicle, and the other is the accurate SOC. In this embodiment, the SOC displayed by the vehicle at the current moment is called the current remaining power. The current remaining power is limited by objective reasons and is generally not the real SOC of the battery cell. In other words, in this embodiment, it cannot be guaranteed that the difference between the current remaining power and the real SOC of the battery cell is within an acceptable error range. In contrast, in this embodiment, the accurate SOC is called the accurate remaining power. The accurate remaining power is the SOC of the battery cell obtained by a certain method, which is characterized by being quite close to the real SOC of the battery cell. In other words, within an acceptable error range, the accurate remaining power is equal to the real SOC of the battery cell. However, relatively speaking, the steps and methods for calculating the accurate remaining power are generally relatively complicated and are not suitable for obtaining the SOC displayed by the vehicle in real time.

[0037] Step S20: Calculate according to the battery information to obtain the current cell output voltage.

[0038] In this embodiment, the current cell output voltage is the cell output voltage of the current step length predicted based on the battery information. The step length means the length of the kth step in the iterative search. Iterative search refers to the process of using a numerical method to solve a solution close to the optimal solution through cyclic iteration.

[0039] In one implementation, the current cell output voltage is obtained through step S201, specifically as follows:

[0040] Step S201: putting the battery information into the trained battery cell voltage neural network model to obtain the current battery cell output voltage.

[0041] In this embodiment, the cell voltage neural network model is a neural network model for predicting the cell voltage. After the battery information is put into the trained cell voltage neural network model, the cell voltage neural network model will output the current cell output voltage based on the battery information, wherein the current cell output voltage corresponds to the battery information.

[0042] In one embodiment, the battery information includes: the current remaining power, the current status information of the battery, and the historical status information of the battery. In this embodiment, the battery information is divided into three categories: the current remaining power, the current status information of the battery, and the historical status information of the battery. This division is based on a deep understanding of the behavior and properties of the battery, as well as a comprehensive consideration of the voltage prediction problem. Specifically, these three parts of information are designed as inputs to the input layer, and the current status information and historical status information of the battery are obtained to more accurately reflect and predict the output voltage of the battery cell.

[0043] It should be noted that the current remaining power is not necessarily the real SOC of the battery cell. When the current remaining power is close to the real SOC of the battery cell, the value of the current battery cell output voltage obtained from the battery information is close to the value of the actual battery cell voltage, wherein in this embodiment, the actual battery cell voltage is called the current battery cell voltage, and the value of the current battery cell voltage can be measured. Correspondingly, when the current remaining power is significantly different from the real SOC of the battery cell, the current battery cell output voltage obtained is significantly different from the current battery cell voltage. Therefore, by comparing the current battery cell output voltage obtained by the neural network with the measured current battery cell voltage, you can know the relationship between the current remaining power input into the neural network and the real SOC of the battery cell.

[0044] The current state information of the battery includes at least one current state information, and the historical state information of the battery includes at least two historical state information.

[0045] In this embodiment, preferably, the current status information of the battery includes: the current of the current step, and the historical status information of the battery includes: the current one step ago, the voltage one step ago, and the voltage two steps ago.

[0046] The current status information of the battery directly reflects the real-time working status of the battery. The current of the current step reflects the real-time load of the battery. The magnitude and direction of the current (charging or discharging) will directly affect the chemical reaction rate of the battery cell, and thus affect the output voltage.

[0047] The historical status information of the battery provides the dynamic change trend of the battery working status, which is convenient for predicting the future status and output voltage of the battery. For example, if the historical status information shows that the battery current has been increasing in the past two consecutive steps, it can be predicted that the battery may continue to bear high current load in the next step, the chemical reaction of the battery cell will be further accelerated, and the output voltage may drop.

[0048] Since the current cell output voltage is based on the battery information, it can be known that the more accurate the battery information is, the more consistent the current cell output voltage is with the actual value. In the battery information, the current of the current step, the current one step ago, the voltage one step ago, and the voltage two steps ago are all measurable and accurate values.

[0049] In this embodiment, preferably, the cell voltage neural network model is composed of a feedforward neural network.

[0050] The calculation method of the current cell output voltage is:

[0051]

[0052] Among them, v k is the current cell output voltage, i k is the current of the current step; i k-1 is the current before one step; v k-1 is the voltage before a step; v k-2 is the voltage two steps ago; SoC k - The current remaining power.

[0053] It should be noted that in this embodiment, the cell voltage neural network model is used to obtain the cell voltage. Compared with the traditional cell model, the neural network model built by the feedforward neural network is used in this solution, so that the result is more accurate than the traditional tabular cell model. At the same time, due to the characteristics of the neural network, the selected battery information may be different. Preferably, in this embodiment, the current remaining power, the current of the current step, the current before a step, the voltage before a step, and the voltage two steps ago are used as inputs of the neural network model.

[0054] In this embodiment, the battery information is used to train the cell voltage neural network model. In this embodiment, the battery information within at least three steps is first obtained, wherein the battery information includes the cell voltage, the cell current and the accurate remaining power. The accurate remaining power is the accurate value in the embodiment of step S10.

[0055] In this embodiment, a specific cell voltage neural network model is provided, wherein a multi-layer feedforward neural network is used, including an input layer, a plurality of hidden layers, and an output layer.

[0056] The number of neurons in the input layer is equal to the number of input features, that is, the number of battery information input into the cell voltage neural network model. In this embodiment, the input layer has 5 neurons, corresponding to five input parameters: the current remaining power (SoC k - ), the current step current (i k ), the current before one step (i k-1 ), the voltage before one step (v k-1 ), the voltage two steps ago (v k-2). In this embodiment, it is considered that there is more than one factor that affects the output voltage. The input parameters cover the current state of the battery (such as the current remaining power and the current step current) and the historical state (such as the current and voltage before the previous steps), providing comprehensive information to predict the cell voltage. These five input parameters represent five different features, which enables the neural network model to learn and understand the complex relationship between these features and the cell voltage without making the model too complex and causing overfitting. The number of neurons in the hidden layer can be adjusted according to actual needs, usually between 10 and 200. Each neuron is connected to all neurons in the input layer, and the connection has a weight. The input received by the neuron is the sum of the product of the output of the previous layer and the corresponding weight, and then output through an activation function (such as ReLU or sigmoid function).

[0057] In this embodiment, an activation function used in a cell voltage neural network model is provided. In this embodiment, ReLU (Rectified Linear Unit) is selected as the activation function of neurons in the hidden layer and the output layer. ReLU can provide nonlinear mapping and avoid the problem of gradient disappearance.

[0058] In this embodiment, the mean square error (MSE) is selected as the loss function.

[0059] The output layer is the output, that is, the current cell output voltage to be predicted (v k ). The output layer has one neuron, which is connected to all neurons in the hidden layer, and the connections have weights. The input received by the neuron is the sum of the output of the previous layer multiplied by the corresponding weight, and then it is directly output without going through the activation function.

[0060] In another possible implementation, the cell voltage neural network model can achieve a stronger nonlinear effect by increasing the number of hidden layers or using a more suitable activation function, such as using a deep neural network, in which each hidden layer can learn different levels of features of the input data, and multiple hidden layers can jointly represent very complex nonlinear relationships.

[0061] For example, in this embodiment, another cell voltage neural network model includes three hidden layers, each hidden layer has 50 neurons, and the cell voltage neural network model uses the ReLU activation function. The input layer and output layer are the same as the model described above. By using the deep neural network structure in this embodiment, the complex nonlinear relationship between the cell voltage and the input parameters can be better captured.

[0062] Step S30: Calculate the actual remaining power based on the current cell output voltage and the current remaining power using an unscented Kalman filter.

[0063] In this embodiment, the actual remaining power is the remaining power that is more consistent with the actual battery cell SOC after calculation.

[0064] It should be noted that Figure 4 It is the relationship between the battery cell SOC and the battery cell voltage, where curve A is a lithium battery and curve B is a lithium iron phosphate battery, and most of the batteries in cars are lithium iron phosphate batteries. Compared with other batteries, the RC characteristics of lithium iron phosphate batteries are not obvious, and the relationship between the battery cell SOC and the battery cell voltage is not linear, so it is not possible to collect them with high accuracy using traditional methods and models.

[0065] In one embodiment, the current cell voltage is first obtained, and then the actual remaining power is obtained based on the current cell output voltage and the current cell voltage. Specifically, step S301 is as follows:

[0066] Step S301: Based on the current cell output voltage and the current remaining power, the actual remaining power is obtained by using an adaptive unscented Kalman filter.

[0067] In this embodiment, the actual remaining power is a parameter variable obtained after correction. Unscented Kalman Filter (UKF) is a combination of lossless transform (UT transform, also known as insensitive transform) and standard Kalman filter system. Through lossless transform, nonlinear system equations are adapted to the standard Kalman system under linear assumption.

[0068] In one implementation, the actual remaining power is obtained through steps S3011-S3017, such as Figure 2 and 3 As shown, the details are as follows:

[0069] Step S3011: construct a corresponding Sigma point set based on the current remaining power through insensitive transformation.

[0070] In this embodiment, the Sigma points are distributed around the state mean and have a certain relationship with the sum of the standard deviations of each state size. The Sigma point set can represent the entire distribution.

[0071] In one embodiment, the SOC curve of the battery cell is a nonlinear system, where x k =f(x k-1 ,u k), where x is the Sigma point set of the SOC of the battery cell after insensitive transformation, x is the n-dimensional state vector, and its average value is known to be ō, and the estimated variance is P0. In this embodiment, the symmetric method is preferably used. After insensitive transformation, 2n+1 Sigma points are constructed. Then x is obtained k It should be noted that, in this embodiment, the construction of the Sigma point set uses the second error covariance obtained in the previous cycle.

[0072] Step S3012: obtaining the corresponding predicted voltage based on the Sigma point set.

[0073] In this embodiment, the first error covariance obtained by using the Sigma point set is P k - , where the first error covariance is the error covariance of the prior estimate.

[0074] In one implementation, the Sigma point is updated backward based on the Sigma point set and the state equation of the system. The prior estimation of the system and the weight coefficient W corresponding to the prior estimation are obtained according to the updated Sigma point. i , the error covariance of the prior estimate and the weight coefficient W of the error covariance of the prior estimate c . Where W i is the weight of the state quantity, W c is the weight of the variance P.

[0075] The predicted voltage is updated based on the Sigma point set. Specifically, the Sigma point set is input into the trained cell voltage neural network model to obtain the predicted voltage. It should be noted that the Sigma point set is obtained after the current remaining power is insensitively transformed, so the unit of the Sigma point set is the same as the unit of the current remaining power.

[0076] Step S3013: Obtain the Kalman gain at the current moment according to the predicted voltage and the Sigma point set.

[0077] In this embodiment, the Kalman gain K k It is a parameter used for correction, which is used to weigh the uncertainty between the predicted state and the observed data to obtain the optimal state estimate. The Kalman gain can be regarded as a weight factor used to perform a weighted average between the predicted state and the observed data.

[0078] In one implementation, the Kalman gain is obtained by calculating the prediction error covariance and the observation error covariance.

[0079] The prediction error covariance is based on the predicted voltage. In this embodiment, the predicted voltage is obtained by inputting the Sigma point set into the trained cell voltage neural network model. When the Sigma point is 2n+1, it is not difficult to understand that 2n+1 corresponding predicted voltages will be obtained. These predicted voltages are combined with the weight coefficient W corresponding to the prior estimate. i The predicted observation value of the predicted voltage will be obtained, which will be referred to as the predicted observation value in the following. The covariance between the predicted observation value and the predicted voltage will be calculated and the correction coefficient will be added to finally obtain the prediction error covariance.

[0080] The observed error covariance is obtained based on the predicted voltage and the Sigma point set, and the observed error covariance is obtained by calculating the covariance of the two.

[0081] After obtaining the observation error covariance, the Kalman gain is calculated based on the prediction error covariance and the observation error covariance. In this embodiment, the calculation formula of the Kalman gain is:

[0082] K k =R / P k h

[0083] Among them, P k h is the forecast error covariance, and R is the observation error covariance.

[0084] Step S3014: Correct the current remaining power based on the Kalman gain to obtain the actual remaining power.

[0085] In this embodiment, the actual remaining capacity is a state estimation correction obtained based on the Kalman gain, and in this embodiment, it is a correction of the battery cell SOC.

[0086] In one implementation, the Kalman filter gain is used to correct the cell SOC, that is, to correct the current remaining power. Specifically:

[0087] x k =x k - +K k (y k -y k - )

[0088] where x k is the actual remaining power, x k - K is the remaining power obtained by weighting the Sigma point set corresponding to the current remaining power with the weight coefficient corresponding to the prior estimate. k is the Kalman filter gain, where y k -yk - is the difference between the observed and predicted voltage values.

[0089] After the correction, the covariance of the prior estimation error is corrected as follows:

[0090] P k＝ P k - -K k P k h K k T ,

[0091] Where P k is the corrected covariance, that is, the second error covariance, P k - is the first error covariance, K k is the Kalman filter gain, P k h is the forecast error covariance.

[0092] The corrected covariance obtained will be used as the second error covariance used in constructing the Sigma point set in the next cycle calculation.

[0093] Step S40: Correcting the current remaining power according to the actual remaining power.

[0094] Based on the current remaining power and the actual remaining power, a first difference between the two is obtained.

[0095] If the first difference is greater than a preset first threshold, the current remaining power is inaccurate and the actual remaining power is used to replace the current remaining power.

[0096] If the first difference is greater than a preset first threshold, the current remaining power is accurate.

[0097] In this embodiment, another method is provided, which directly uses the actual remaining power to replace the current remaining power, which is more convenient and concise, and is beneficial to the calculation of the next calculation cycle.

[0098] In this embodiment, the use of unscented Kalman filtering is perfectly suitable for nonlinear situations compared to traditional Kalman filtering, is more suitable for the lithium iron phosphate battery in this embodiment, and has smaller errors than Kalman filtering.

[0099] In this embodiment, preferably, an adaptive unscented Kalman filter is used. Compared with the unscented Kalman filter, the adaptive unscented Kalman filter omits the step of setting parameters at the beginning, is simpler to use than the unscented Kalman filter, and also avoids the disadvantage of not converging when the parameters of the unscented Kalman filter are set incorrectly.

[0100] It should be noted that, in this embodiment, steps S10-S40 are steps of a calculation cycle. It is understandable that the above steps can be continuously operated in a cycle while the car is driving until the driving is finished.

[0101] Furthermore, the present invention also provides a device for estimating the remaining battery power.

[0102] The device mainly includes a vehicle information acquisition module, a model voltage calculation module, an actual power calculation module and a current power correction module.

[0103] A device for estimating remaining battery power, comprising:

[0104] Vehicle information acquisition module: obtains the battery information of the current vehicle, the battery information includes the current remaining power;

[0105] Model voltage calculation module: obtains the current cell output voltage according to the battery information;

[0106] Actual power calculation module: calculates the actual remaining power according to the current cell output voltage and the current remaining power using unscented Kalman filter;

[0107] Current power correction module: corrects the current remaining power according to the actual remaining power.

[0108] In one implementation, the description of the specific implementation functions can refer to steps S10-S30.

[0109] The above-mentioned battery remaining capacity estimation device is used to perform Figure 1 The embodiment of the method for estimating the remaining battery power shown in the figure has similar technical principles, technical problems solved and technical effects produced. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process and related instructions of the device for estimating the remaining battery power can refer to the contents described in the embodiment of the method for estimating the remaining battery power, which will not be repeated here.

[0110] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art can understand that in order to achieve the effects of the present invention, different steps do not have to be performed in such an order. They can be performed simultaneously (in parallel) or in other orders. These changes are within the scope of protection of the present invention.

[0111] It is understood by those skilled in the art that the present invention implements all or part of the processes in the method of the above embodiment, and can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of each of the above method embodiments when executed by the processor. Among them, the computer program includes a computer program, and the computer program can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium that can carry the computer program. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electric carrier signals and telecommunication signals.

[0112] Furthermore, the present invention also provides a control device. In one embodiment of the control device according to the present invention, the control device includes a processor and a storage device, the storage device can be configured to store a program for executing the method for estimating the remaining battery power of the above-mentioned method embodiment, and the processor can be configured to execute the program in the storage device, which includes but is not limited to the program for executing the method for estimating the remaining battery power of the above-mentioned method embodiment. For ease of explanation, only the parts related to the embodiment of the present invention are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present invention. The control device can be a control device device formed by various electronic devices.

[0113] Furthermore, the present invention also provides a computer-readable storage medium. In a computer-readable storage medium embodiment according to the present invention, the computer-readable storage medium can be configured to store a program for executing the method for estimating the remaining battery power of the above-mentioned method embodiment, and the program can be loaded and run by the processor to implement the above-mentioned method for estimating the remaining battery power. For ease of explanation, only the parts related to the embodiment of the present invention are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present invention. The computer-readable storage medium can be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present invention is a non-temporary computer-readable storage medium.

[0114] Further, it should be understood that since the setting of each module is only for illustrating the functional units of the device of the present invention, the physical devices corresponding to these modules may be the processor itself, or a part of the software in the processor, a part of the hardware, or a part of the combination of software and hardware. Therefore, the number of each module in the figure is only schematic.

[0115] Those skilled in the art will appreciate that the modules in the device can be adaptively split or merged. Such splitting or merging of specific modules will not cause the technical solution to deviate from the principle of the present invention, and therefore, the technical solutions after splitting or merging will fall within the protection scope of the present invention.

[0116] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A method for estimating the remaining power of a battery, characterized in that: include: Obtaining battery information of the current vehicle, wherein the battery information includes the current remaining power; Obtaining the current cell output voltage according to the battery information; Calculating the actual remaining power using an unscented Kalman filter according to the current output voltage of the battery cell and the current remaining power; The current remaining power is corrected according to the actual remaining power.

2. The method for estimating the remaining battery power according to claim 1, characterized in that: The obtaining the current cell output voltage according to the battery information includes: The battery information is input into a trained cell voltage neural network model to obtain the current cell output voltage.

3. The method for estimating the remaining battery power according to claim 1 or 2, characterized in that: The battery information also includes: current status information of the battery and historical status information of the battery.

4. The method for estimating the remaining battery power according to claim 3, characterized in that: The current state information of the battery includes at least one current state information, and the historical state information of the battery includes at least two historical state information.

5. The method for estimating the remaining battery power according to claim 3 or 4, characterized in that: The current state information of the battery includes: the current of the current step length; the historical state information of the battery includes: the current one step length ago, the voltage one step length ago, and the voltage two steps length ago.

6. The method for estimating the remaining battery power according to claim 1, characterized in that: Correcting the current remaining power according to the actual remaining power includes: Based on the current remaining power and the actual remaining power, obtaining a first difference between the two; If the first difference is greater than a preset first threshold, the current remaining power is inaccurate and the current remaining power is replaced by the actual remaining power.

7. The method for estimating the remaining battery power according to claim 5, characterized in that: The actual remaining power is obtained based on the current cell output voltage and the current remaining power and using unscented Kalman filtering, including: Based on the current remaining power, the corresponding Sigma point set is constructed through insensitive transformation; Based on the Sigma point set, the corresponding predicted voltage is obtained; The Kalman gain at the current moment is obtained according to the predicted voltage and the Sigma point set; The current remaining power is corrected based on the Kalman gain to obtain the actual remaining power.

8. A device for estimating the remaining battery power, characterized in that: include: Vehicle information acquisition module: obtains the battery information of the current vehicle, the battery information includes the current remaining power; Model voltage calculation module: obtains the current cell output voltage according to the battery information; Actual power calculation module: calculates the actual remaining power according to the current cell output voltage and the current remaining power using unscented Kalman filter; Current power correction module: corrects the current remaining power according to the actual remaining power.

9. A control device comprising a processor and a storage device, wherein the storage device is suitable for storing a computer program, characterized in that: The computer program is suitable for being loaded and run by the processor to execute the method for estimating the remaining battery capacity according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a plurality of computer programs, characterized in that: The computer program is suitable for being loaded and run by a processor to execute the method for estimating the remaining battery capacity according to any one of claims 1 to 7.

Images