Method for correcting remaining power, vehicle and storage medium

Abstract

The present application relates to the fields of lithium battery technology and artificial intelligence technology, and specifically to a method, a vehicle, and a storage medium for correcting remaining power. The method for correcting remaining power includes: obtaining an estimated remaining power of a battery, and obtaining a remaining power correction value of the battery, wherein the estimated remaining power is determined based on the collected remaining power-related information of the battery and / or the ampere-hour integration method; correcting the estimated remaining power using the remaining power correction value; wherein the remaining power correction value is a correction value obtained after a trained power correction model processes input data, wherein the input data includes remaining power-related information collected in a specific power interval, and the training data of the power correction model uses the remaining power correction value corresponding to the remaining power-related information as label information. The solution provided by the present application can improve the accuracy of the remaining power of the battery.

Description

Technical Field

[0001] The present application relates to the fields of lithium battery technology and artificial intelligence technology, and in particular to a method for correcting remaining power, a vehicle, and a storage medium. Background Art

[0002] Society is paying more and more attention to environmental pollution. Reducing automobile exhaust emissions is an effective way to reduce environmental pollution. Compared with the internal combustion engine of fuel vehicles, the electric motor of electric vehicles does not need to emit exhaust gas, which can effectively reduce air pollution. However, electric vehicles also have the disadvantage of short driving range after a single charge, which causes range anxiety for electric vehicle passengers.

[0003] In order to improve passengers' range anxiety, the related technology can provide users with more accurate remaining power (also called state of charge, SOC for short) information so that users can charge in time according to SOC and travel distance, reducing the risk of being stranded due to power shortage during driving. However, the related technology can only perform remaining power correction when the battery is low or fully charged, and it is not easy to perform power correction on batteries in the commonly used remaining power range, resulting in the user experience to be improved. Summary of the invention

[0004] In order to solve or partially solve the problems existing in the related art, the present application provides a method, a vehicle and a storage medium for correcting the remaining power, which can realize the power correction of the battery in the commonly used remaining power range, and effectively improve the user experience.

[0005] On the one hand, the present application provides a method for correcting remaining power, comprising: obtaining an estimated remaining power of a battery, and obtaining a remaining power correction value of the battery, wherein the estimated remaining power is determined based on collected remaining power-related information of the battery and / or the ampere-hour integration method; and correcting the estimated remaining power using the remaining power correction value; wherein the remaining power correction value is a correction value obtained after a trained power correction model processes input data, the input data includes remaining power-related information collected in a specific power interval, and the training data of the power correction model uses the remaining power correction value corresponding to the remaining power-related information as label information.

[0006] In one embodiment, the above method also includes: obtaining the training data; the obtaining of the training data includes: obtaining the remaining power related information collected in a specific power interval of the estimated remaining power; obtaining the difference between the estimated remaining power when fully charged and the actual remaining power; and using the difference as label information of the remaining power related information.

[0007] In one embodiment, the method of obtaining the difference between the estimated remaining power when fully charged and the actual remaining power includes: if the estimated remaining power reaches 100% and the battery is in a charging state, obtaining the charging power from the current moment until the battery cell with the highest voltage in the battery reaches the charging cut-off voltage, and using the charging power as the difference between the estimated remaining power and the actual remaining power; if the estimated remaining power does not reach 100% and the battery is in a charging completion state, using the difference between the full charge and the estimated remaining power as the difference between the estimated remaining power and the actual remaining power.

[0008] In one embodiment, the power correction model includes: an input layer, at least one convolution pair and at least one fully connected layer connected in series, each convolution pair includes a convolution layer and a pooling layer connected in series; the input data of the input layer includes: charged power, estimated remaining power, maximum single cell voltage, voltage difference, voltage-to-power ratio of the highest voltage battery cell, voltage-to-power ratio of the lowest voltage battery cell, or at least one of the consistency of voltage-to-power ratios of the battery cells.

[0009] In one embodiment, the method of correcting the estimated remaining power by using the remaining power correction value includes: during charging the battery, if the estimated remaining power reaches a remaining power characteristic point, correcting the estimated remaining power by using the remaining power correction value, and the remaining power characteristic point is determined based on a voltage-to-power ratio characteristic point.

[0010] In one embodiment, the above method also includes: obtaining voltage-to-charge ratio data of the battery during the charging process; performing a differential operation on the voltage-to-charge ratio data to extract characteristic points of the voltage-to-charge ratio data; determining a current remaining power corresponding to the characteristic points of the voltage-to-charge ratio data, and using the current remaining power as the remaining power characteristic point.

[0011] In one embodiment, the battery includes iron and lithium elements; the value range of the remaining power characteristic point includes 60%×full power±10%×full power.

[0012] In one embodiment, the charging current of the battery is less than a specific charging current threshold, and the fluctuation range of the battery between a specific maximum value and a specific minimum value under the specific charging current threshold exceeds a preset fluctuation range; wherein the specific maximum value is the maximum value of the differential calculation result of the voltage-to-electricity ratio of the estimated remaining power within a specific power interval, and the specific minimum value is the minimum value of the differential calculation result of the voltage-to-electricity ratio of the estimated remaining power within the specific power interval.

[0013] In one embodiment, the method is applicable to a server side; the method also includes: receiving the input data from the vehicle side, processing the input data using a trained power correction model to obtain the remaining power correction value; sending the remaining power correction value to the vehicle side, so that the vehicle side uses the remaining power correction value to correct the estimated remaining power.

[0014] In one embodiment, the method is applicable to the vehicle side; the method also includes: at least receiving model topology information and / or model parameter information of a trained power correction model from the server side; constructing the power correction model based on the model topology information and / or model parameter information; obtaining the remaining power correction value of the battery includes: the vehicle side uses the power correction model to process the input data to obtain the remaining power correction value.

[0015] On the other hand, the present application provides a device for correcting the remaining power, including: an information acquisition module and a power correction module. The information acquisition module is used to obtain the estimated remaining power of the battery and obtain the remaining power correction value of the battery, wherein the estimated remaining power is determined based on the collected remaining power-related information of the battery and / or the ampere-hour integration method; the power correction module is used to correct the estimated remaining power using the remaining power correction value; wherein the remaining power correction value is a correction value obtained after the trained power correction model processes the input data, wherein the input data includes the remaining power-related information collected in a specific power interval, and the training data of the power correction model uses the remaining power correction value corresponding to the remaining power-related information as label information.

[0016] On the other hand, the present application provides a vehicle, comprising: a vehicle body; a battery, arranged on the vehicle body, for storing electrical energy; a sensing device, coupled to the battery, for collecting information related to the remaining power of the battery; a processor, coupled to the sensing device, for processing the information related to the remaining power; and a memory, on which executable code is stored, and when the executable code is executed by the processor, the processor executes the method as described above.

[0017] On the other hand, the present application provides a computer-readable storage medium having executable codes stored thereon. When the executable codes are executed by a processor of an electronic device, the processor executes the above method.

[0018] On the other hand, the present application provides a computer program product, including executable codes, which implement the above method when the executable codes are executed.

[0019] The technical solution provided by this application may have the following beneficial effects:

[0020] The technical solution of the embodiment of the present application, after obtaining the estimated remaining power of the battery, further determines the remaining power correction value based on the power correction model and the remaining power associated information collected in a specific power interval. In this way, the estimated remaining power can be corrected based on the remaining power correction value for a specific power interval, so that the user can obtain the accurate remaining power for, for example, a commonly used remaining power interval.

[0021] In certain embodiments of the present application, the difference between the estimated remaining power in the full charge stage and the actual remaining power is used as a label for training data, which improves the accuracy of the labels for the training data and helps to improve the accuracy of the remaining power correction value determined based on the trained power correction model.

[0022] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and other objects, features and advantages of the present application will become more apparent by describing in more detail exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0024] Figure 1 It is an exemplary system architecture of a method for correcting remaining power, a vehicle, and a storage medium shown in an embodiment of the present application;

[0025] Figure 2 is a schematic diagram of the structure of a vehicle and a battery shown in an embodiment of the present application;

[0026] Figure 3 is a flow chart of a method for correcting remaining power according to an embodiment of the present application;

[0027] Figure 4 is a schematic diagram of the corresponding relationship between the remaining power of a lithium iron phosphate battery and the battery voltage according to an embodiment of the present application;

[0028] Figure 5 is a schematic diagram of a characteristic curve of SOC / voltage shown in an embodiment of the present application;

[0029] Figure 6 This is an embodiment of the present application. Figure 5 Schematic diagram of the difference curve obtained by differentiating the middle curve;

[0030] Figure 7 is a schematic diagram of differential curves corresponding to different test temperatures shown in an embodiment of the present application;

[0031] Figure 8is a schematic diagram of the structure of a power calibration model shown in an embodiment of the present application;

[0032] Fig. 9 is a block diagram of a device for correcting remaining power according to an embodiment of the present application;

[0033] Fig.10 is a block diagram of a device for correcting remaining power according to another embodiment of the present application; and

[0034] Fig.11 It is a block diagram of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0035] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0036] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "an" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0037] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0038] Range anxiety refers to the mental pain or worry caused by passengers' fear of running out of power when riding in electric vehicles. Range anxiety is a factor restricting the development of electric vehicles. The fundamental reason is that the battery has a limited range and takes a long time to charge.

[0039] The remaining power refers to the ratio of the available power in the battery to the nominal capacity, which is an important monitoring data of the battery management system. The battery management system controls the working state of the battery according to the SOC value. The remaining power of the battery reflects the state of charge of the battery.

[0040] The driving range of electric vehicles depends on many factors, and is much more sensitive to weather, road conditions, driving habits, and the number of passengers than traditional vehicles. Therefore, there is a great deal of uncertainty about the actual driving range of electric vehicles after a single charge. Drivers are not used to doing research on various aspects such as road conditions before starting a journey, which increases the range anxiety of passengers.

[0041] Battery power management is one of the core contents of battery management. It is of great significance to the control of the entire battery status and the prediction and estimation of the driving range of electric vehicles. For example, passengers can adjust the driving route, charging plan, driving speed, air conditioning control strategy, etc. according to the remaining power to ensure that passengers can reach the destination or the nearest charging station.

[0042] Due to the nonlinearity of the power battery state of charge (SOC), or the unclear correspondence between the remaining power and the battery voltage, and the influence of various factors, the battery power estimation and prediction methods are complicated, and it is difficult to accurately estimate the SOC. For example, the characteristic of the iron-lithium battery is that the terminal voltage-SOC curve (v / q curve) is approximately a straight line during the period of 30% SOC to 80% SOC, making it difficult to achieve the correction of the remaining power.

[0043] Related technologies In order to correct the remaining power of the battery, considering that the SOC changes with the voltage in the high SOC range and the low SOC range, the SOC correction point used is also located below 30% SOC and close to the full charge stage. Due to mileage anxiety, battery over-discharge protection, and charging time limitations, the charging and discharging range of electric vehicles used by most users is between 30% SOC and 90% SOC, which makes it almost impossible for users to trigger the SOC correction strategy when using electric vehicles. As time goes by, the errors of the ampere-hour integration method will accumulate, causing the error of the estimated SOC to gradually increase, affecting the user experience.

[0044] For example, when the remaining power is sufficient, it is displayed as limited, causing the user to be afraid to go to the destination or to charge first. For example, when the remaining power is insufficient, it is displayed as relatively sufficient, causing the user to run out of power before reaching the destination.

[0045] The embodiment of the present application utilizes a trained power correction model to process the remaining power associated information collected in a specific power interval to obtain a remaining power correction value. The remaining power correction value can be a correction value for a commonly used remaining power interval, so that the user can calibrate the remaining power of the battery in the commonly used remaining power interval. The embodiment of the present application can accurately implement the remaining power correction without the user having to perform specific operations such as exhausting or fully charging the battery, thereby improving the user experience.

[0046] The following will be passed Figures 1 to 11 A method for correcting remaining power, a vehicle, and a storage medium in an embodiment of the present application are described in detail.

[0047] Figure 1 This is an exemplary system architecture of a method, a vehicle and a storage medium to which the remaining power correction method, a vehicle and a storage medium can be applied in one embodiment of the present application. It should be noted that: Figure 1 What is shown is merely an example of a system architecture to which the embodiments of the present application can be applied, in order to help those skilled in the art understand the technical content of the present application, but it does not mean that the embodiments of the present application cannot be used in other devices, systems, environments or scenarios.

[0048] See also Figure 1 According to this embodiment, the system architecture 100 may include mobile platforms 101, 102, 103, a network 104 and a cloud 105. The network 104 is used to provide a medium for communication links between the mobile platforms 101, 102, 103 and the cloud 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc. The mobile platforms 101 and 102 may be equipped with batteries and battery management systems to power the mobile platforms. In addition, the mobile platforms 101 and 102 may also be provided with sensing devices to sense the surrounding environment information of the mobile platforms 101 and 102, such as obstacle information and lane line information.

[0049] Users can use mobile platforms 101, 102, and 103 to interact with other mobile platforms and the cloud 105 through the network 104 to receive or send information, etc. Such as sending model training requests, model parameter download requests, and receiving trained model parameters. The trained model can process the remaining power-related information to obtain the remaining power correction value. In addition, mobile platforms 101 and 102 can also obtain obstacle information through image recognition, point cloud recognition, etc. Mobile platforms 101, 102, and 103 can be installed with various communication client applications, such as assisted driving applications, automatic driving applications, car applications, web browser applications, database applications, search applications, instant messaging tools, email clients, social platform software, and other applications.

[0050] Mobile platforms 101 , 102 , 103 include, but are not limited to, electronic devices such as automobiles, robots, aircraft, tablet computers, and laptop computers.

[0051] The cloud 105 can receive model training requests, receive model parameter download requests, adjust model parameters for model training, issue model topology structures, issue trained model parameters, etc., and can also send weather information, real-time traffic information, etc. to the mobile platforms 101, 102, 103. For example, the cloud 105 can be a background management server, a server cluster, a car network, etc.

[0052] It should be noted that the number of mobile platforms, networks and cloud servers is only illustrative, and any number of mobile platforms, networks and cloud servers may be provided as required.

[0053] Figure 2 It is a schematic diagram of the structure of a vehicle and a vehicle-mounted sensor shown in an embodiment of the present application.

[0054] See also Figure 2 , the vehicle 100 may include one or more of the following components: a vehicle body 110, and a battery 150. The battery 150 is used to provide electrical energy to the electrical equipment of the vehicle 100. The electric vehicle energy management system is a software and hardware system that can coordinate, distribute and control the working energy of the energy conversion device of the electric vehicle power system. The hardware of the energy management system consists of a series of sensors, control units ECU and actuators, etc. The function of the software system is mainly to analyze and process the signals of the sensors, optimize and analyze the working energy of the energy conversion device, and issue instructions to the actuators.

[0055] In addition, in order to improve the riding experience, assisted driving or automatic driving performance can also be provided. For example, the vehicle 100 can also be provided with a laser detection and ranging (Lidar) 120, an inertial measurement unit (IMU) 130, a camera 140, etc. The Lidar 120, the IMU 130, and the camera 140 are provided on the vehicle body 110.

[0056] For example, the laser radar 120 can be installed on both sides of the front bumper. In addition, the laser radar can also be installed on the roof or above the front window of the vehicle, etc., which is not limited here. The laser radar 120 can include pulse laser radar and continuous wave laser radar, and can use ultraviolet light, visible light or near infrared light as a light source. Since the laser radar 120 can accurately detect the distance between a specific target and the mobile platform, it can effectively eliminate some traffic accidents caused by misidentification of images.

[0057] For example, the shooting device 140 includes a camera, and an image of the light signal collected by the camera is formed on a photosensitive device and transmitted to an image processor, so that obstacles can be obtained by image recognition.

[0058] In addition, one or more ultrasonic sensors, millimeter wave radars, etc. may be provided on the vehicle 100 to improve the comprehensiveness and accuracy of the perceived vehicle environment information. For example, ultrasonic sensors may be provided inside the front and rear bumpers. For example, millimeter wave radars may be provided inside the front and rear bumpers.

[0059] All of the above devices require the battery 150 to provide power to achieve specific functions. In order to improve the utilization of power and prevent the battery from being overcharged and over-discharged, the battery 150 can be managed by a battery management system (Battery Management System, referred to as BMS). BMS is a subsystem of the electric vehicle energy management system, which can achieve the purpose of extending the service life of the battery and monitoring the battery status. For example, BMS can output the remaining power so that the user can understand the current charge state of the battery in order to better arrange travel strategies. However, the accuracy of the remaining power output by the relevant technology needs to be improved.

[0060] Taking lithium iron batteries as an example, the characteristic of lithium iron batteries is that the terminal voltage-SOC curve during the SOC30% to 80% period is approximately a straight line. Only in the high SOC range (such as the range close to the full power state) and the low SOC range (such as the range less than 30% SOC), the trend of SOC changing with voltage is more obvious. For lithium iron batteries, the SOC correction point used by the current vehicle-side strategy is also below 30% SOC and when fully charged. Due to mileage anxiety and charging time limitations, the charging and discharging range of most users using electric vehicles is between [30% SOC, 90% SOC], and the SOC correction strategy cannot be triggered. As time goes by, the error of the ampere-hour integration method will accumulate, causing the error of the estimated SOC to gradually increase, affecting the user experience.

[0061] Certain embodiments of the present application are based on battery slow charging data (such as charging current <20A), use a power correction model to identify the characteristic points of the terminal voltage-SOC curve change rate (dv / dq) at the characteristic points, and correct the current SOC of the battery, thereby achieving accurate correction of the remaining power of the battery in the common charging and discharging range of electric vehicles [30% SOC, 90% SOC].

[0062] Figure 3 This is an application scenario shown in the related art where traffic accidents are prone to occur.

[0063] See also Figure 3 , the method for correcting the remaining power may include operations S310 to S320.

[0064] In operation S310, an estimated remaining power of a battery is obtained, and a remaining power correction value of the battery is obtained, wherein the estimated remaining power is determined based on collected remaining power associated information of the battery and / or an ampere-hour integration method.

[0065] In this embodiment, the remaining power related information includes but is not limited to: at least one of voltage, current, charge and discharge rate, self-discharge rate, cell voltage uniformity, voltage-to-power ratio, voltage-to-power ratio curve change rate, etc.

[0066] For example, during the use of the battery, when the remaining battery power is different, the voltage displayed to the outside is different. Taking a certain ternary battery cell as an example, when the battery is fully charged, the voltage is 4.2v, and when the battery power is discharged, the voltage drops to about 2.7v. In this way, the remaining power of the battery can be characterized based on the open circuit voltage of the battery. Specifically, the remaining power of the battery can be approximately obtained by measuring the voltage at both ends of the battery. This method is simple but the accuracy needs to be improved. For example, during the discharge process of a lithium battery, the relationship between the remaining power and the voltage is not linear. For example, a ternary battery can work at 3.6v for a long time.

[0067] In some embodiments, the accuracy of estimating the remaining power can be improved by combining the ampere-hour integration. Among them, the ampere-hour integration is to integrate the current over time. An is the unit of current size, and its full name is ampere. According to Ohm's law, in a pure resistance circuit, the current size is equal to the ratio of voltage to resistance. The ampere-hour integration method uses a component to continuously measure the size of the loop current, and conducts real-time testing (such as measuring every few milliseconds). The current size at each moment is known. In this way, the power consumption and charged power during the entire use process can be calculated, and then the original remaining power can be added to get the current remaining power. The calculation formula can be: Current remaining power = original remaining power - power consumption + power charged. Among them, power consumption SOC 消耗 The calculation formula can be shown as formula (1).

[0068] SOC 消耗 =(I now ×t) / C max Formula (1)

[0069] Among them, I now is the current, t is the time, SOC 消耗 is the battery capacity (in Ah).

[0070] It should be noted that there are many factors that affect battery charging and discharging. Charging 20% ​​of the rated power does not mean that 20% of the rated power can be discharged. For example, one of the most obvious influencing factors is temperature. If the battery is fully charged at room temperature, the amount of power that can be discharged at low temperatures will be discounted. This is because the chemical substances are less active at low temperatures, and the chemical reactions proceed slowly and with low intensity. The above reasons lead to inaccurate estimates of the remaining power.

[0071] In some embodiments, the estimated remaining power can be obtained from the SOC system. As shown above, since the ampere-hour integration method still has problems, the estimated remaining power needs to be further corrected. For example, the total power consumption in the above-mentioned ampere-hour calculation process is not measured continuously, but at intervals of time. Although it is a millisecond interval, it will still accumulate a large error over time. One solution is to calibrate when the battery is exhausted or fully charged. That is, when the battery is fully charged, even if the SOC determines that the current remaining power is 95% according to the ampere-hour integration method, the power will be calibrated to 100%; or, when the battery is exhausted, even if the remaining power output by the ampere-hour integration method is 3%, the power will be calibrated to 0%.

[0072] For example, the SOC correction point used by the vehicle-side strategy in the related art is also below 30% SOC and when fully charged. Due to mileage anxiety, over-discharge protection (below 20% SOC may affect battery life) and charging time (DC fast charging takes about 30 minutes from 30% SOC to 80% SOC; in order to protect the battery, the charging current will be limited above 80% SOC, resulting in about 70 minutes from 30% SOC to 100% SOC, and some users are unwilling to wait until it is fully charged). Due to the above reasons, the charging and discharging intervals of electric vehicles used by most users are between 30% SOC and 90% SOC, and the SOC correction strategy cannot be triggered. As time goes by, the errors of the ampere-hour integration method will accumulate, causing the error in the estimated SOC to gradually increase, affecting the user experience.

[0073] In order to correct the estimated remaining power in a commonly used charging and discharging interval, a remaining power correction value of the battery is also obtained in this embodiment. The remaining power correction value is a correction value obtained after the trained power correction model processes the input data. The input data includes remaining power related information collected in a specific power interval. The training data of the power correction model uses the remaining power correction value corresponding to the remaining power related information as label information.

[0074] The types of power correction models include, but are not limited to, linear regression, logistic regression, decision trees, Naive Bayes, K-Nearest Neighbors, support vector machines, random decision forests or bagging, neural networks, etc.

[0075] Taking the neural network as a power correction model as an example, the accuracy of its output results is affected by the network topology, training data (input data type, remaining power range, label quality), loss function, device computing power, etc. Convolutional neural network (CNN) is an image feature extraction algorithm. As the network deepens, through the alternating use of convolutional layer-pooling layer, the algorithm can gradually extract low-dimensional features and global image features, thereby achieving the purpose of classification.

[0076] The core of the convolutional neural network lies in the convolution kernel. The convolution kernel can be compared to a filter. The biggest difference between it and the filter is that the parameters of the filter are known and are defined based on the experience of predecessors using methods such as first-order / second-order differences; while the parameters of the convolution kernel are unknown and are trained through the back-propagation algorithm of the neural network. The input data of the power correction model includes the remaining power association information collected in a specific power interval, and the remaining power association information has label information. The label information represents the remaining power correction value. In this way, the trained power correction model can output the corresponding remaining power correction value after processing the input remaining power association information.

[0077] In operation S320, the estimated remaining power is corrected using the remaining power correction value.

[0078] In this embodiment, after obtaining the estimated remaining power of the battery and the corresponding remaining power correction value, the estimated remaining power can be corrected using the remaining power correction value. For example, the estimated remaining power and the remaining power correction value (which can be a positive number, zero, or a negative number) can be added to obtain the corrected remaining power.

[0079] In this embodiment, since the remaining power correction value is a correction value for a specific power interval (such as a commonly used remaining power interval), it is possible to calibrate the remaining power within a specific power interval, rather than only being able to calibrate the remaining power when the power is exhausted, low on power, or fully charged. Users can obtain accurate remaining power within a specific power interval, thereby improving user experience.

[0080] In some embodiments, in order to improve the accuracy of the calibration result, a remaining power calibration is performed when a feature point is identified.

[0081] Specifically, the use of the remaining power correction value to correct the estimated remaining power may include the following operations: during the battery charging process, if the estimated remaining power reaches a remaining power characteristic point, the estimated remaining power is corrected using the remaining power correction value, and the remaining power characteristic point is determined based on the voltage-to-power ratio characteristic point.

[0082] Figure 4 It is a schematic diagram of the corresponding relationship between the remaining power of a lithium iron phosphate battery and the battery voltage shown in an embodiment of the present application.

[0083] See also Figure 4 Due to the characteristics of lithium iron phosphate batteries, the corresponding relationship between voltage and remaining power is not significant enough. For example, when the collected voltage fluctuates slightly, the remaining power of the battery fluctuates violently. In order to solve this problem, the voltage-to-power ratio is obtained in this embodiment.

[0084] In some embodiments, in order to improve the accuracy of the correction value output by the power correction model, the applicable scenarios of the model can be more carefully defined, such as being more suitable for slow charging scenarios. Specifically, the charging current of the battery is less than a specific charging current threshold, and the fluctuation range between the specific maximum value and the specific minimum value of the battery under the specific charging current threshold exceeds the preset fluctuation range; wherein the specific maximum value is the maximum value of the differential operation result of the voltage-to-electricity ratio of the estimated remaining power in a specific power interval, and the specific minimum value is the minimum value of the differential operation result of the voltage-to-electricity ratio of the estimated remaining power in a specific power interval. For example, the specific charging current threshold can be determined by calibration or the like. The larger the fluctuation range between the specific maximum value and the specific minimum value, the more obvious the feature is, which facilitates the extraction of more significant features. The preset fluctuation range can be determined based on expert experience, calibration, and the like, such as the ratio of the specific maximum value to the specific minimum value exceeds 30%, 40%, 60%, 80%, 100%, 150% or 300%, etc.

[0085] Figure 5 It is a schematic diagram of a characteristic curve of SOC / voltage shown in an embodiment of the present application.

[0086] See also Figure 5 , relative to Figure 4 The nearly horizontal curve in the commonly used remaining power range, Figure 5 The rate of change of the curve in the commonly used remaining power range has fluctuated significantly, which provides a good basis for improving the accuracy of the model prediction results.

[0087] In the SOC30%SOC~80%SOC period, the terminal voltage-SOC curve has a relatively obvious change area near 60%SOC, and the rest area is approximately a straight line. In addition, the trend of SOC changing with voltage is only obvious in the high SOC range and the low SOC range, while the low SOC range does not belong to the user's commonly used power range.

[0088] See also Figure 5 When charging the lithium iron phosphate battery (LFP) at a lower current rate, the SOC-voltage curve of the lithium iron battery is shown in the figure (discrete at every 0.5% SOC). At about 60% SOC, the voltage change rate changes significantly compared with the previous and the next. For example, both the highest voltage (voltmax) and the lowest voltage (voltmin) change significantly.

[0089] In order to make the characteristic point more obvious, in this embodiment, a differential operation may be performed on the SOC / voltage characteristic curve to determine the remaining power characteristic point.

[0090] Specifically, the above method may also include the following operations.

[0091] First, the voltage-to-electricity ratio data of the battery during the charging process is obtained.

[0092] Then, the voltage-to-electricity ratio data is differentially calculated to extract the characteristic points of the voltage-to-electricity ratio data. Specifically, the voltage-to-electricity ratio data can be represented in the form of a curve, and then the curve is differentially calculated: dv / dq=(v t -v t-1 ) / (q t -q t-1 ). See Figure 6 , the maximum battery voltage differential result (dv / dq_cellvoltmax) and the minimum battery voltage differential result (dv / dq_cellvoltmin) both show obvious characteristic peaks.

[0093] Next, the current remaining power corresponding to the characteristic point of the voltage-to-electricity ratio data is determined, and the current remaining power is used as the remaining power characteristic point.

[0094] In a specific embodiment, the battery includes iron and lithium elements. Accordingly, the value range of the remaining power characteristic point includes 60%×full power±10%×full power. Due to the characteristics of the iron-lithium battery, during the SOC30%SOC~80%SOC period, the terminal voltage-SOC curve has a relatively obvious change area except near 60%SOC. For example, the remaining power characteristic points are 55%SOC, 58%SOC, 60%SOC, 62%SOC, 65%SOC, etc. Figure 5 It is a schematic diagram of a characteristic curve of SOC / voltage shown in an embodiment of the present application.

[0095] Please also see Figure 5 and Figure 6 , through Figure 5 The curve shown in Figure 1 is differentiated by: dv / dq, and the result is Figure 6 The curve shown. Figure 5 It can be seen that there is a very obvious peak characteristic point at 60% SOC, and the area near 60% SOC can be used as the remaining power characteristic point. This is very helpful for improving the accuracy of the remaining power correction value.

[0096] It should be noted that this embodiment is well applicable to a variety of different temperature scenarios, that is, the remaining power characteristic point can be well determined at different temperatures. Figure 7 It is a schematic diagram of differential curves corresponding to different test temperatures shown in an embodiment of the present application.

[0097] See also Figure 7 Although the terminal voltage-SOC curves obtained by differential calculation do not overlap at different temperatures, the position of the remaining power characteristic point (such as 60% SOC) is very stable, so that a stable remaining power characteristic point can be found well when correcting the remaining power, which helps to improve the accuracy of the corrected remaining power.

[0098] In a specific embodiment, for lithium iron batteries, based on the battery slow charging data (charging current <20A), a convolutional neural network (CNN) is used to identify the point where the terminal voltage-SOC curve has a larger rate of change at 60% SOC, and the current SOC of the battery is corrected.

[0099] Compared with the SOC correction point used in the related art, which is also located below 30% SOC and close to full charge, the probability of being triggered is limited. This embodiment corrects the current SOC by identifying the characteristic changes of the dv / dq curve near 60% SOC during low-rate charging, greatly improving the coverage and effectively improving the correction accuracy.

[0100] In addition, compared with the fixed internal parameters and strategies of the BMS in the related art, this embodiment uses the historical data of the battery (pack) and, through big data modeling, can not only eliminate the jump caused by a single random error, but also correct the battery with inaccurate SOC.

[0101] In order to better understand the technical solution of the present application, the relevant information of the power calibration model (such as training data, training method, topological structure, etc.) is exemplarily described below.

[0102] In some embodiments, high-quality training data may be obtained in the following manner: Specifically, the above method further includes obtaining the training data.

[0103] Specifically, obtaining the training data may include the following operations. First, obtain the remaining power related information collected within a specific power interval of the estimated remaining power. For example, the specific power interval is 30% SOC ~ 90% SOC, then when the estimated remaining power is in this interval, collect the remaining power related information. Such as reading the battery voltage, current, voltage-to-power ratio, cell voltage uniformity and other information from the battery management system, such as collecting battery voltage, current and other information through sensors.

[0104] Then, the difference between the estimated remaining power when fully charged and the actual remaining power is obtained. For example, when the battery is fully charged, the difference between the estimated remaining power and the actual remaining power is obtained. Then, the difference is used as the tag information of the remaining power associated information.

[0105] The label information can be defined as follows: As a supervised learning algorithm, the neural network needs to be modeled with known inputs and outputs. The input of the model is voltage, current, temperature and other features, and the output (which can be used as a label) is the difference between the current observed SOC and the true SOC, as shown in formula (2).

[0106] label = |SOC BMS -SOC real | Formula (2)

[0107] Among them, the estimated remaining power (SOC BMS ) can be observed or recorded by the BMS. real ) cannot be directly observed or recorded by the BMS, but can be determined by calibration, detection, etc. For example, the remaining power at the end of automatic charging of the battery pack is recorded as 100% SOC.

[0108] In some embodiments, obtaining the difference between the estimated remaining power when fully charged and the actual remaining power may include the following operations.

[0109] For example, if the estimated remaining power reaches 100% and the battery is in charging state, the charging power from the current moment to the charging cut-off voltage of the battery cell with the highest voltage in the battery is obtained, and the charging power is taken as the difference between the estimated remaining power and the actual remaining power.

[0110] For example, if the estimated remaining power does not reach 100% and the battery is in a charging complete state, the difference between the full power and the estimated remaining power is used as the difference between the estimated remaining power and the actual remaining power.

[0111] In a specific embodiment, when the battery is fully charged, the SOC is undoubtedly the most accurate, so in this embodiment, the moment before full charging can be taken as the label.

[0112] Based on the current SOC system strategy, if SOC BMS With SOC real The difference is not big, then when it is close to full charge, SOC BMS Should rise to 100% at a nearly uniform rate, and if SOC BMS With SOC real If the difference is large, the following two situations may occur:

[0113] 1.SOC BMS Immediately jump from a lower SOC to 100%. In this case, it can be considered that the SOC sent by the BMS is falsely low, that is, the SOC BMS <SOC real , resulting in the highest cell reaching the cut-off voltage just before full charging, and charging must be terminated, SOC BMS Forced jump to 100% SOC.

[0114] 2.SOC BMS The SOC at 98.9% continues to charge for a long time before jumping to 100%. In this case, the SOC sent by the BMS is falsely high, that is, the SOC BMS >SOC real , resulting in SOC BMS Stay in the high SOC section and continue charging until the highest cell reaches the cut-off voltage.

[0115] Since the error caused by a single charge can be ignored without losing data, the algorithm aims to observe the SOC at the end of the charge for each full charge segment. BMS With SOC real The difference between the training data and the test data is used as the label. The label information (label information of training data and test data) is obtained in the above manner so that model training and model testing can be performed based on the label.

[0116] The following is an exemplary description of the model topology structure and input data.

[0117] In some embodiments, the power correction model can be a convolutional neural network, which can specifically include: an input layer, at least one convolution pair, and at least one fully connected layer connected in series, each convolution pair including a convolution layer and a pooling layer connected in series. As the convolutional neural network deepens, it can gradually extract low-dimensional features and global features through downsampling (specifically using alternating convolution layers-pooling layers), thereby achieving the purpose of classification. The core of the power correction model lies in the convolution kernel. The parameters of the convolution kernel are unknown and are trained through the back propagation algorithm of the neural network.

[0118] Figure 8 is a schematic diagram of the structure of a power calibration model shown in an embodiment of the present application. Figure 8 ,The power calibration model is a convolutional neural network including an input layer, three convolution pairs, two fully connected layers and an output layer. For example, the number of convolution kernels, the number of convolution layers, the number of fully connected layers, and the number of nodes in the fully connected layers are the main model parameters adjusted in model training.

[0119] For example, the input data of the input layer include: charged power, estimated remaining power, maximum cell voltage, voltage difference, voltage-to-power ratio of the highest voltage cell, voltage-to-power ratio of the lowest voltage cell, or at least one of the consistency of voltage-to-power ratio of the cell. For example, the input of the convolutional neural network is: charged power q, remaining power SOC of the battery, maximum cell voltage, voltage difference, dv / dq value of the highest voltage cell, dv / dq value of the lowest voltage cell. The above 6 variables are averaged every 0.5% SOC, and the SOC range can be [50%, 75%], [30%, 90%], [40%, 80%], etc. In this embodiment, considering that there are multiple cells in the battery pack, and the SOC calculation of the whole pack actually uses the weighted SOC of the highest cell and the lowest cell, the use of the highest and lowest dv / dq can consider the overall situation of the battery pack, which can effectively increase the accuracy of the algorithm. In addition, the dv / dq value of the highest voltage cell and the dv / dq value of the lowest voltage cell can also be expanded to calculate the cell consistency, such as calculating the SOC of the highest cell and the SOC of the lowest cell respectively to evaluate the consistency.

[0120] For example, charging data in the range of 50% SOC to 75% SOC can be used as input, with each 0.5% SOC being discrete. In terms of coverage, the charging segments covered by this embodiment include the user's commonly used remaining power range, which effectively improves the problem that some users will not discharge the battery power below 30%, resulting in the inability to activate the remaining power correction process. It is very common for the power to be between 50% SOC and 75% SOC, allowing users to accurately know the current remaining power in the commonly used remaining power range.

[0121] It should be noted that the technical solution of the present application can be applicable to the server side and / or the vehicle side.

[0122] Fig. 9 is a block diagram of a device for correcting remaining power according to an embodiment of the present application. Fig. 9 The apparatus 900 for correcting the remaining power may include a server end 910 and a vehicle end 920. The server end 910 and the vehicle end 920 are in communication connection with each other.

[0123] Taking the above method applied to the server 910 as an example, the server 910 may determine the remaining power correction value using the power correction model so that the vehicle 920 corrects the remaining power based on the received remaining power correction value. Specifically, the above method may also include the following operations.

[0124] First, the input data from the vehicle side is received, and the input data is processed by using the trained power correction model to obtain the remaining power correction value. The vehicle side can obtain the above input data from the BMS and send it to the server side. The training method of the power correction model can refer to the above-mentioned relevant parts.

[0125] Then, the remaining power correction value is sent to the vehicle end so that the vehicle end uses the remaining power correction value to correct the estimated remaining power. In this way, the vehicle end does not need to consume a lot of computing power to calculate the remaining power correction value, thereby reducing the hardware cost of the vehicle end.

[0126] Taking the above method applied to the vehicle side 920 as an example, the server side 910 can train the power correction model, and the vehicle side 920 downloads the model topology structure and model parameters from the server side 910 to calculate the remaining power correction value locally. Specifically, the above method can also include the following operations.

[0127] First, at least receive the model topology information and / or model parameter information of the trained power calibration model from the server. It should be noted that if the server does not update the model, it is not necessary to download the model topology information and / or model parameter information.

[0128] Then, the power correction model is constructed based on the model topology information and / or model parameter information.

[0129] Correspondingly, the obtaining of the remaining power correction value of the battery includes: the vehicle end processes the input data using the power correction model to obtain the remaining power correction value.

[0130] The method for correcting the remaining power provided by this embodiment can be deployed in a variety of ways, which helps to meet the diverse needs of the market.

[0131] In this embodiment, when correcting the remaining power of the battery, there is no need to over-discharge or over-charge the battery. For batteries, overcharging and over-discharging may cause permanent damage to the battery and seriously shorten the service life of the battery. The remaining power after correction is more accurate, which is convenient for the vehicle control strategy to control the SOC within a certain range (such as 20% power to 80% power), which prevents the battery from being overcharged or over-discharged, thereby ensuring the normal use of the battery and extending the battery life.

[0132] Another aspect of the present application also provides a block diagram of a device for correcting remaining power.

[0133] Fig.10 It is a block diagram of a device for correcting remaining power according to another embodiment of the present application.

[0134] See also Fig.10 The device 1000 for correcting the remaining power may include: an information obtaining module 1010 and a power correction module 1020.

[0135] The information acquisition module 1010 is used to obtain the estimated remaining power of the battery and obtain the remaining power correction value of the battery. The estimated remaining power is determined based on the collected remaining power associated information of the battery and / or the ampere-hour integration method.

[0136] The power correction module 1020 is used to correct the estimated remaining power using the remaining power correction value; wherein the remaining power correction value is a correction value obtained after the trained power correction model processes the input data, the input data includes remaining power associated information collected in a specific power interval, and the training data of the power correction model uses the remaining power correction value corresponding to the remaining power associated information as label information.

[0137] In addition, the device 1000 for correcting the remaining power may also include: a model training module, a label information acquisition module, etc., and specific reference is made to the relevant parts above.

[0138] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0139] Another aspect of the present application also provides a vehicle.

[0140] Fig.11 It is a block diagram of a vehicle according to an embodiment of the present application.

[0141] See also Fig.11 The vehicle 1100 includes a vehicle body, a memory 1110 , a processor 1120 , a sensing device 1130 , and a battery 1140 .

[0142] The memory 1110 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage. Among them, ROM may store static data or instructions required by the processor 1120 or other modules of the computer. The permanent storage may be a readable and writable storage device. The permanent storage may be a non-volatile storage device that does not lose the stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a large-capacity storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In other embodiments, the permanent storage device may be a removable storage device (such as a floppy disk, optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as a dynamic random access memory. The system memory may store some or all instructions and data required by the processor at run time. In addition, the memory 1110 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (such as DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 1110 may include a readable and / or writable removable storage device, such as a laser disc (CD), a read-only digital versatile disc (such as a DVD-ROM, a double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (such as an SD card, a mini SD card, a Micro-SD card, etc.), a magnetic floppy disk, etc. The computer-readable storage medium does not include carrier waves and transient electronic signals transmitted wirelessly or wired.

[0143] The processor 1120 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is coupled to at least two laser radars, and is at least used to receive point clouds from the laser radars.

[0144] The memory 1110 stores executable codes, and when the executable codes are processed by the processor 1120 , the processor 1120 can execute part or all of the methods described above.

[0145] The sensing device 1130 is coupled to the battery and is used to collect information related to the remaining power of the battery and environmental information. For example, the sensing device 1130 may include sensors, camera devices, etc., which are arranged inside, on the top, or at the front of the vehicle body. For example, the sensor includes a temperature sensor, a current sensor, a voltage sensor, etc. For example, the camera device may include a body, a lens, and a photosensitive material. In addition, it may also include systems such as aperture, shutter, distance measurement, framing, light metering, film transport, counting, Selfie, focus, zoom, etc.

[0146] In addition, the vehicle 1100 may further include a laser radar 1130. One or more laser radars 1130 may be provided. For example, at least two laser radars 1130 are respectively provided at at least two positions of the vehicle body, and are respectively used to collect point clouds. For example, one laser radar 1130 is provided on the left side of the front end of the vehicle body, and one laser radar 1130 is provided on the right side of the front end of the vehicle body. For example, one laser radar 1130 is provided on the left side of the front end of the vehicle body, one laser radar 1130 is provided on the right side of the front end of the vehicle body, and one laser radar 1130 is provided above the front windshield of the vehicle body, or is provided above the roof.

[0147] It is understandable that, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, or each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0148] If the integrated unit / module is implemented in the form of hardware, the hardware may be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. If not otherwise specified, the artificial intelligence processor may be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, and an ASIC, etc. If not otherwise specified, the storage module may be any appropriate magnetic storage medium or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random-Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), or Hybrid Memory Cube (HMC), etc.

[0149] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), mobile hard disk, disk or optical disk and other media that can store program codes.

[0150] In a possible implementation, a board is also disclosed, which includes a storage device, an interface device, a control device and the above-mentioned processor; wherein the processor is connected to the storage device, the control device and the interface device respectively; the storage device is used to store data; the interface device is used to realize data transmission between the processor and an external device; and the control device is used to monitor the status of the processor.

[0151] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.

[0152] Alternatively, the present application can also be implemented as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium) on which executable code (or computer program or computer instruction code) is stored. When the executable code (or computer program or computer instruction code) is executed by a processor of an electronic device (or server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.

[0153] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for correcting remaining power, characterized in that: include: Obtaining an estimated remaining power of a battery and obtaining a remaining power correction value of the battery, wherein the estimated remaining power is determined based on the collected remaining power association information of the battery and an ampere-hour integration method; Correcting the estimated remaining power by using the remaining power correction value, including: during the battery charging process, if the estimated remaining power reaches a remaining power characteristic point, correcting the estimated remaining power by using the remaining power correction value, the remaining power characteristic point is determined according to a characteristic point of a voltage-to-power ratio, the battery includes an iron element and a lithium element, and the value range of the remaining power characteristic point includes (60%±10%)×full power; Among them, the remaining power correction value is a correction value obtained after the trained power correction model processes the input data, the input data includes the remaining power related information collected in a specific power interval, and the training data of the power correction model uses the remaining power correction value corresponding to the remaining power related information as label information.

2. The method according to claim 1, characterized in that Also includes: obtaining the training data; The obtaining of the training data comprises: Obtaining the remaining power associated information collected when the estimated remaining power is in the specific power interval; Get the difference between the estimated remaining power when fully charged and the actual remaining power; The difference is used as label information of the remaining power associated information.

3. The method according to claim 2, characterized in that The obtaining of the difference between the estimated remaining power when fully charged and the actual remaining power includes: If the estimated remaining power reaches 100% and the battery is in a charging state, then the charging power from the current moment until the battery cell with the highest voltage in the battery reaches the charging cut-off voltage is obtained, and the charging power is used as the difference between the estimated remaining power and the actual remaining power; If the estimated remaining power does not reach 100% and the battery is in a charging complete state, the difference between the full power and the estimated remaining power is used as the difference between the estimated remaining power and the actual remaining power.

4. The method according to claim 1, characterized in that The power correction model includes: an input layer, at least one convolution pair and at least one fully connected layer connected in series, each convolution pair includes a convolution layer and a pooling layer connected in series; the input data of the input layer includes: charged power, estimated remaining power, maximum single cell voltage, voltage difference, voltage-to-power ratio of the highest voltage battery cell, voltage-to-power ratio of the lowest voltage battery cell or at least one of the consistency of voltage-to-power ratio of the battery cells.

5. The method according to claim 1, characterized in that Also includes: Obtaining voltage-to-electricity ratio data of the battery during charging; Performing a differential operation on the voltage-to-electricity ratio data to extract characteristic points of the voltage-to-electricity ratio data; The current remaining power corresponding to the characteristic point of the voltage-to-electricity ratio data is determined, and the current remaining power is used as the remaining power characteristic point.

6. The method according to any one of claims 1 to 5, characterized in that: The charging current of the battery is less than a specific charging current threshold, and the fluctuation range of the battery between a specific maximum value and a specific minimum value under the specific charging current threshold exceeds a preset fluctuation range; wherein the specific maximum value is the maximum value of the differential calculation result of the voltage-to-electricity ratio of the estimated remaining power within the specific power interval, and the specific minimum value is the minimum value of the differential calculation result of the voltage-to-electricity ratio of the estimated remaining power within the specific power interval.

7. The method according to any one of claims 1 to 5, characterized in that The method is applicable to the server side; The method further comprises: Receiving the input data from the vehicle end, and processing the input data using the trained power correction model to obtain the remaining power correction value; The remaining power correction value is sent to the vehicle end so that the vehicle end uses the remaining power correction value to correct the estimated remaining power.

8. The method according to any one of claims 1 to 5, characterized in that The method is applicable to the vehicle end; The method further comprises: At least receiving model topology information and / or model parameter information of a trained power correction model from a server; Constructing the power correction model based on the model topology information and / or model parameter information; The obtaining of the remaining power correction value of the battery includes: the vehicle end processes the input data using the power correction model to obtain the remaining power correction value.

9. A vehicle, characterized in that: include: The main body of the vehicle; A battery, disposed on the vehicle body, for storing electrical energy; a sensing device, coupled to the battery, for collecting information related to the remaining power of the battery; a processor, coupled to the sensing device, and configured to process the remaining power related information; A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 8.

Images