Vehicle charging remaining time determination method and apparatus, and vehicle-side control device

Abstract

The application relates to a vehicle charging remaining time determination method and device, a vehicle-side control device, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring an initial charging remaining time of a vehicle at a current charging time in a charging process and a charging remaining time influence feature of the vehicle in the charging process; and correcting the initial charging remaining time by using the charging remaining time influence feature to obtain a target charging remaining time of the vehicle at the current charging time. By using the method, the acquired charging remaining time influence feature of the vehicle in the charging process can be used to more accurately evaluate the change factors in the charging process, so that the initial charging remaining time is corrected to obtain a more accurate target charging remaining time. Moreover, the accurate target charging remaining time, i.e. the accurate charging remaining time prediction, can help a user to better arrange a charging plan time, reduce a waiting time and improve charging management efficiency of the user.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery charging, in particular to a vehicle charging remaining time determination method and device, a vehicle end control device, a computer readable storage medium and a computer program product. BACKGROUND

[0002] In the use of the current new energy vehicles, the user of the new energy vehicle hopes to charge the vehicle as much as possible within the planned time, and the user usually formulates the planned time according to the estimated charging remaining time displayed by the vehicle. However, in the actual charging process, the real charging remaining time of the vehicle will change due to various factors, and there will be a deviation between the real charging remaining time and the estimated charging remaining time, resulting in that the actual charging condition of the vehicle does not match the user's expectation.

[0003] At present, the estimation method of the charging remaining time is mainly obtained by predicting the state of charge (SOC) of the battery and the current parameter, and the prediction scheme is to fit the SOC and the current parameter into a two-dimensional curve. During real-time operation, the current charging remaining time is queried according to the current SOC and current value. However, the above scheme has the problem that the charging remaining time is greatly different from the actual charging time in actual use. SUMMARY

[0004] Therefore, it is necessary to provide a vehicle charging remaining time determination method, device, vehicle end control device, computer readable storage medium and computer program product capable of improving the accuracy of the charging remaining time in view of the above technical problems.

[0005] In a first aspect, the present application provides a vehicle charging remaining time determination method, comprising:

[0006] obtaining an initial charging remaining time of a vehicle at a current charging time in a charging process and a charging remaining time influencing feature of the vehicle in the charging process;

[0007] correcting the initial charging remaining time by using the charging remaining time influencing feature to obtain a target charging remaining time of the vehicle at the current charging time.

[0008] In one of the embodiments, the charging remaining time influencing feature includes a global degradation feature and a local fluctuation feature of the vehicle in the charging process; the global degradation feature is battery data associated with the battery charging and discharging performance degradation of the vehicle battery at the current charging time; the local fluctuation feature includes charging state data associated with the charging remaining time fluctuation in the charging process; the correction of the initial charging remaining time by using the charging remaining time influencing feature to obtain the target charging remaining time of the vehicle at the current charging time includes:

[0009] The initial charging remaining time is corrected by using at least one of the global degradation feature and the local fluctuation feature, to obtain the target charging remaining time of the vehicle at the current charging time.

[0010] In one of the embodiments, the charging remaining time influencing feature includes a global degradation feature and a local fluctuation feature of the vehicle during the charging process; the global degradation feature is battery data associated with the battery performance degradation of the vehicle battery at the current charging time; and the local fluctuation feature includes state of charge data associated with the charging remaining time fluctuation during the charging process.

[0011] The initial charging remaining time is corrected by using at least one of the global degradation feature and the local fluctuation feature, to obtain the target charging remaining time of the vehicle at the current charging time.

[0012] The initial charging remaining time is first corrected according to the global degradation feature, to obtain a corrected initial charging remaining time.

[0013] A charging remaining time correction amount of the vehicle at the current charging time is obtained according to the local fluctuation feature.

[0014] The corrected initial charging remaining time is second corrected by using the charging remaining time correction amount, to obtain the target charging remaining time of the vehicle at the current charging time.

[0015] In one of the embodiments, the initial charging remaining time is first corrected according to the global degradation feature, to obtain a corrected initial charging remaining time, including:

[0016] The global degradation feature is input into a time correction model, and the initial charging remaining time is first corrected by using the time correction model and the state of charge data at the current charging time, to obtain a corrected initial charging remaining time.

[0017] In one of the embodiments, the charging remaining time correction amount of the vehicle at the current charging time is obtained according to the local fluctuation feature, including:

[0018] The local fluctuation feature is input into a pre-constructed local fluctuation feature influencing time function, to obtain a charging remaining time correction amount corresponding to the local fluctuation feature; the local fluctuation feature influencing time function is obtained based on historical state of charge data of the vehicle battery in a current life cycle and historical charging remaining time corresponding to the historical state of charge data.

[0019] In one of the embodiments, the step of obtaining the local fluctuation feature influencing time function includes:

[0020] obtaining historical charging state data in a current life cycle of a vehicle battery and historical charging remaining time corresponding to the historical charging state data;

[0021] sampling the historical charging remaining time by using a preset charging time sampling interval to obtain a plurality of historical charging remaining time change values, and in a case where the historical charging remaining time change value is greater than a preset time change threshold, taking the charging state data corresponding to the historical charging remaining time change value as a local fluctuation feature, and taking each of the local fluctuation features as a local fluctuation feature sequence;

[0022] tracking and predicting the local fluctuation feature sequence by using a particle filtering algorithm to obtain a predicted local fluctuation feature sequence, and fitting the predicted local fluctuation feature sequence by using a polynomial regression to obtain a local fluctuation feature influence time function.

[0023] In one of the embodiments, the tracking and predicting the local fluctuation feature sequence by using a particle filtering algorithm to obtain a predicted local fluctuation feature sequence, and fitting the predicted local fluctuation feature sequence by using a polynomial regression to obtain a local fluctuation feature influence time function, comprises:

[0024] initializing by using an initial probability density filtering function to generate a particle swarm;

[0025] updating and normalizing a weight of each particle to obtain a weight of each particle;

[0026] resampling each particle to obtain a new particle set and a weight;

[0027] predicting the local fluctuation feature sequence by using a state equation to obtain a predicted local fluctuation feature sequence, and fitting the predicted local fluctuation feature sequence by using a polynomial regression to obtain a monomial polynomial regression function;

[0028] solving parameters of the monomial polynomial regression function by using a least square method to obtain a coefficient matrix;

[0029] obtaining the local fluctuation feature influence time function according to the coefficient matrix and the monomial polynomial regression function.

[0030] In one of the embodiments, the obtaining of the local fluctuation feature of the vehicle in the charging process comprises:

[0031] obtaining historical charging remaining time of the vehicle from a charging start time to a current charging time in a current charging process;

[0032] According to the historical charging remaining time and a preset charging time sampling interval, a plurality of historical charging remaining time change values are obtained.

[0033] In a case where the historical charging remaining time change value is greater than a preset time change threshold value, charging state data corresponding to a charging time interval associated with the historical charging remaining time change value is taken as a local fluctuation feature.

[0034] In a second aspect, the present application further provides a vehicle charging remaining time determination device, comprising:

[0035] A feature extraction module is configured to obtain an initial charging remaining time of a vehicle at a current charging time in a charging process and charging remaining time influence features of the vehicle in the charging process.

[0036] A remaining time correction module is configured to correct the initial charging remaining time by using a charging remaining time correction amount to obtain a target charging remaining time of the vehicle at the current charging time.

[0037] In a third aspect, the present application further provides a vehicle-side control device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in the first aspect when executing the computer program.

[0038] In a fourth aspect, the present application further provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the method in the first aspect when executed by a processor.

[0039] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, and the computer program implements the steps of the method in the first aspect when executed by a processor.

[0040] The vehicle charging remaining time determination method, device, vehicle-side control device, computer readable storage medium and computer program product can first obtain an initial charging remaining time of a vehicle at a current charging time in a charging process and charging remaining time influence features of the vehicle in the charging process, then correct the initial charging remaining time by using the charging remaining time influence features, and finally obtain a target charging remaining time of the vehicle at the current charging time. In the present application, the charging remaining time influence features of the vehicle in the charging process can be used to more accurately evaluate the change factors in the charging process, so as to correct the initial charging remaining time and obtain a more accurate target charging remaining time. Moreover, an accurate target charging remaining time, i.e., an accurate charging remaining time prediction, can help users better arrange charging plan time, reduce waiting time, and improve the charging management efficiency of the users. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to make the technical solutions in the embodiments of the present application or the related art clearer, the accompanying drawings needed in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the accompanying drawings in the following description only only some embodiments of the present application, and for those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart of a method for determining the charging remaining time of a vehicle in an embodiment;

[0043] Figure 2 A flowchart of a method for correcting the initial charging remaining time by using the charging remaining time influence feature to obtain the target charging remaining time of the vehicle at the current charging moment in an embodiment;

[0044] Figure 3 A flowchart of a method for obtaining the initial charging remaining time in an embodiment;

[0045] Figure 4 A flowchart of a method for training the MAP table update model in an embodiment;

[0046] Figure 5 A flowchart of a method for obtaining the local fluctuation feature in an embodiment;

[0047] Figure 6 A block diagram of a device for determining the charging remaining time of a vehicle in an embodiment;

[0048] Figure 7 An internal structure diagram of a vehicle-side control device in an embodiment. DETAILED DESCRIPTION

[0049] In order to make the technical solutions in the embodiments of the present application or the related art clearer, the accompanying drawings needed in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the accompanying drawings in the following description only only some embodiments of the present application, and for those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0050] The method for determining the charging remaining time of a vehicle provided by the embodiments of the present application can be applied in a vehicle battery management system (BMS, Battery Management System). The vehicle battery management system obtains the initial charging remaining time of the vehicle at the current charging moment in the charging process and the charging remaining time influence feature of the vehicle in the charging process from the data recorded in the vehicle management system. The vehicle battery management system corrects the initial charging remaining time by using the charging remaining time influence feature to obtain the target charging remaining time of the vehicle at the current charging moment.

[0051] In an exemplary embodiment, asFigure 1 As shown, a vehicle charging remaining time determination method is provided, comprising the following steps S102 to S104. Among them:

[0052] Step S102, obtaining the initial charging remaining time of the vehicle at the current charging time in the charging process and the charging remaining time influencing feature of the vehicle in the charging process.

[0053] Among them, the initial charging remaining time can refer to the remaining time required to fully charge the vehicle from the current time in the charging process.

[0054] Specifically, during the charging process of the vehicle, the vehicle battery management system will first record the initial charging remaining time at the current time, which refers to the remaining time required to fully charge the vehicle from the current time. This remaining time is usually an original charging remaining time estimate value calculated based on the current state of charge of the battery, the power of the charging pile and the charging curve of the battery, etc.; during the charging process of the vehicle, the change of the charging remaining time may be affected by various factors, and the vehicle battery management system obtains these influencing factors and forms the charging remaining time influencing feature, and then corrects the initial charging remaining time according to the charging remaining time influencing feature.

[0055] Step S104, correcting the initial charging remaining time using the charging remaining time influencing feature to obtain the target charging remaining time of the vehicle at the current charging time.

[0056] Specifically, the vehicle battery management system can use a preset algorithm to correct the initial charging remaining time using the charging remaining time influencing feature, for example, the preset algorithm can be an empirical formula, a machine learning model or other mathematical model, etc., and then obtain the target charging remaining time of the vehicle at the current charging time.

[0057] In the above vehicle charging remaining time determination method, by obtaining the charging remaining time influencing feature of the vehicle in the charging process, the change factors in the charging process can be more accurately evaluated, so as to correct the initial charging remaining time and obtain more accurate target charging remaining time. And accurate target charging remaining time, i.e. accurate charging remaining time prediction, can help users better arrange charging plan time and reduce waiting time, which can improve the charging management efficiency of users.

[0058] In an exemplary embodiment, the initial charging remaining time is corrected using the charging remaining time influencing feature to obtain the target charging remaining time of the vehicle at the current charging time, comprising:

[0059] The initial charging remaining time is corrected by using at least one of the global degradation feature and the local fluctuation feature, to obtain the target charging remaining time of the vehicle at the current charging time.

[0060] The charging remaining time influence feature includes the global degradation feature and the local fluctuation feature of the vehicle in the charging process; the global degradation feature is battery data associated with the battery charging and discharging performance attenuation of the vehicle battery at the current charging time; and the local fluctuation feature includes charging state data associated with the charging remaining time fluctuation in the charging process.

[0061] With the use of the vehicle battery, the battery has corresponding charging and discharging performance attenuation, that is, the battery has the phenomenon corresponding to the global degradation feature, but the battery may not have the phenomenon corresponding to the local fluctuation feature in the use process. Therefore, the correction of the initial charging remaining time can include:

[0062] Example 1: Only the global degradation feature of the battery is considered; that is, the vehicle battery management system can correct the initial charging remaining time by using the global degradation feature, to obtain the target charging remaining time of the vehicle at the current charging time.

[0063] Example 2: Only the local fluctuation feature of the battery in the use process is considered; the vehicle battery management system can correct the initial charging remaining time by using the local fluctuation feature, to obtain the target charging remaining time of the vehicle at the current charging time.

[0064] Example 2: Both the global degradation feature and the local fluctuation feature of the battery in the use process are considered; the vehicle battery management system can correct the initial charging remaining time twice by using the global degradation feature and the local fluctuation feature, to obtain the target charging remaining time of the vehicle at the current charging time.

[0065] In this embodiment, the initial charging remaining time is corrected by using at least one of the global degradation feature and the local fluctuation feature, to obtain the charging remaining time correction amount of the vehicle at the current charging time. The initial charging remaining time can be corrected according to the global degradation feature and / or the local fluctuation feature. Different features can be selected according to different scenarios, and then corrected. The prediction accuracy of the charging remaining time can be effectively improved.

[0066] In an exemplary embodiment, the charging remaining time influence feature includes the global degradation feature and the local fluctuation feature of the vehicle in the charging process; the global degradation feature is battery data associated with the battery charging and discharging performance attenuation of the vehicle battery at the current charging time; and the local fluctuation feature includes charging state data associated with the charging remaining time fluctuation in the charging process; as Figure 2 The above step S104 can specifically include:

[0067] Step S202, the initial charging remaining time is first corrected according to the global degradation feature, and a corrected initial charging remaining time is obtained.

[0068] The global degradation feature can be a feature of the overall performance degradation of the battery, which can be evaluated by historical data and the current battery state, and reflects the health of the battery.

[0069] For example, the vehicle battery management system can first correct the initial charging remaining time according to the global degradation feature, and obtain the corrected initial charging remaining time.

[0070] Step S204, the charging remaining time correction amount of the vehicle at the current charging time is obtained according to the local fluctuation feature.

[0071] The local fluctuation feature includes charging state data associated with charging remaining time fluctuation; the charging state data can be data used to represent the charging state of the battery, such as SOC, voltage, current, temperature, etc. of the battery; specifically, the local fluctuation feature can be a feature of the change of the charging state data, such as the change of SOC, current, voltage, temperature, etc. during the charging process, which can cause the fluctuation of the charging remaining time.

[0072] For example, at time T1, the vehicle battery voltage is 400V, the current is 50A, the temperature is 25℃, and the SOC is 90%. The vehicle battery management system calculates the initial charging remaining time as 20 minutes through the above parameters; the charging remaining time change during the charging process is obtained, and if there is an abnormal fluctuation of the charging remaining time, the charging state data of the battery voltage, current, temperature and SOC corresponding to the abnormal fluctuation is obtained, for example, the current jumps from 50A to 40A.

[0073] The charging remaining time correction amount can be a correction value of the initial charging remaining time calculated according to the local fluctuation feature, which is used to improve the accuracy of the initial charging remaining time estimation.

[0074] For example, the vehicle battery management system obtains the charging remaining time correction amount of the vehicle at the current charging time according to the local fluctuation feature, such as the current jumping from current A to current B, and the temperature jumping from M to N, through current A, current B, temperature M and temperature N.

[0075] Step S206, the corrected initial charging remaining time is second corrected by using the charging remaining time correction amount, and the target charging remaining time of the vehicle at the current charging time is obtained.

[0076] The target charging remaining time can be the initial charging remaining time after correction, representing a more accurate time required from the current charging time to the completion of charging.

[0077] For example, the initial charging remaining time is A, and the charging efficiency is reduced due to the current fluctuation, resulting in a charging remaining time correction amount of -B. The vehicle battery management system corrects the initial charging remaining time A by the charging remaining time correction amount -B to obtain the target charging remaining time A-B of the vehicle at the current charging time. The target charging remaining time A-B can be displayed on the vehicle-mounted system or the user terminal application for the user to refer.

[0078] In this embodiment, the initial charging remaining time is first corrected by the global degradation feature associated with the battery data of the vehicle battery at the charging time and the battery charge and discharge performance degradation, to obtain the initial charging remaining time after correction. Then, the charging remaining time correction amount of the vehicle at the current charging time is obtained by the local fluctuation feature associated with the charging state data related to the charging remaining time fluctuation. Finally, the initial charging remaining time is corrected by the charging remaining time correction amount. Based on the initial charging remaining time after correction by the global degradation feature, and in combination with the charging remaining time correction amount obtained by the local fluctuation feature, the target charging remaining time of the vehicle can be more accurately determined, and the accuracy of the charging remaining time prediction is improved. In addition, the more accurate target charging remaining time is displayed to the user, which can enable the user to reasonably plan the time and reduce the waiting time of the user.

[0079] In an exemplary embodiment, the above step S102 can specifically include:

[0080] In step S6, the global degradation feature is input into the time correction model, and the initial charging remaining time is first corrected by the time correction model and the charging state data at the current charging time to obtain the initial charging remaining time after correction.

[0081] The time correction model can be a model obtained based on an empirical formula, a machine learning model, or other mathematical models.

[0082] Specifically, the vehicle battery management system can input the global degradation feature into the time correction model, and the initial charging remaining time is first corrected by the correction parameter output by the time correction model in combination with the charging state data at the current charging time to obtain the initial charging remaining time after correction.

[0083] In the embodiment, the global degradation feature is converted into a direct influence on the charging remaining time by the time correction model, and the initial charging remaining time is corrected for the first time by combining the charging state data at the current charging time, so that the correction accuracy of the initial charging remaining time can be improved.

[0084] In an exemplary embodiment, the time correction model is a pre-trained MAP table updating model, as shown in Figure 3 The step S6 can specifically include:

[0085] In step S302, the global degradation feature is input into the pre-trained MAP table updating model to obtain a corresponding charging remaining time interpolation estimation MAP table at the current charging time.

[0086] The global degradation feature can refer to the global degradation feature of the vehicle battery corresponding to the current entire charging process. For example, if the current charging process is the Nth cycle charging, the battery data associated with the battery charging and discharging performance degradation after the (N-1)th cycle charging can be obtained as the global degradation feature, and the battery data after the (N-1)th cycle discharging can also be obtained as the global degradation feature. In another example, the battery data associated with the battery charging and discharging performance degradation in the current charging process can also be used as the global degradation feature; the battery data associated with the battery charging and discharging performance degradation in the current charging process and the battery data associated with the battery charging and discharging performance degradation in the N-1 cycle charging process before the current charging process can also be combined as the global degradation feature.

[0087] The MAP (Multi-dimensional Array Map) table can refer to a multi-dimensional data structure for storing and querying the relationship between multiple variables. In the embodiment, the charging remaining time interpolation estimation MAP table can be used to represent the relationship between the battery state (such as SOC, voltage, current, temperature, etc.) and the charging remaining time. Through the charging remaining time interpolation estimation MAP table, the charging remaining time under a specific battery state can be quickly found. The pre-trained MAP table updating model can be a machine learning model. The global degradation feature of the battery, such as the number of charging and discharging cycles, charging efficiency, internal resistance, battery health, and other battery charging and discharging performance battery data, is input into the MAP table updating model. The corresponding charging remaining time interpolation estimation MAP table at the current charging time can be updated according to the global degradation feature of the battery through the model output.

[0088] Exemplarily, after obtaining the global degradation feature, the vehicle battery management system inputs the global degradation feature into the pre-trained MAP table updating model, and updates the MAP table by interpolating the historical charging remaining time to obtain a corresponding charging remaining time interpolation estimation MAP table at the current charging time.

[0089] At step S304, the charging remaining time matching the charging state data at the current charging time is obtained from the charging remaining time interpolation estimation MAP table as the initial charging remaining time.

[0090] Exemplarily, the vehicle battery management system obtains the charging state data at the current charging time, and then obtains the charging remaining time matching the charging state data at the current charging time from the charging remaining time interpolation estimation MAP table, and takes the charging remaining time as the initial charging remaining time.

[0091] Exemplarily, the historical charging remaining time interpolation estimation MAP table is a first MAP table, the global degradation feature is input into the pre-trained MAP table updating model to obtain a corresponding charging remaining time interpolation estimation MAP table at the current charging time, that is, the first MAP table is updated by the MAP table updating model to obtain a second MAP table corresponding to the charging remaining time interpolation estimation MAP table at the current charging time, and the charging remaining time A matching the charging state data at the current charging time is obtained from the second MAP table as the initial charging remaining time; if the charging remaining time B matching the charging state data at the current charging time is obtained from the first MAP table, the charging remaining time A is not the same as the charging remaining time B, because the second MAP table contains the influence of the current global feature on the basis of the first MAP, and therefore the charging remaining time obtained by different MAP tables is not the same under the same charging state data.

[0092] In this embodiment, by inputting the global degradation feature into the pre-trained MAP table updating model, the latest charging remaining time interpolation estimation MAP table affected by the global degradation feature can be obtained, so that the charging remaining time matching the charging state data at the current charging time is obtained from the latest charging remaining time interpolation estimation MAP table as the initial charging remaining time, and therefore the accuracy of the estimation of the initial charging remaining time corresponding to the current charging time can be improved, that is, the vehicle battery management system can provide more accurate initial charging remaining time estimation according to the current state of the battery, and further provide reliable basic data for the subsequent time correction process through the accurate initial charging remaining time.

[0093] In one exemplary embodiment, as shown in Figure 4 the training step of the above-mentioned MAP table updating model comprises:

[0094] In step S402, the historical charging state data in the historical charging process in the current life cycle of the vehicle battery, the historical charging remaining time corresponding to the historical charging state data, the initial charging remaining time interpolation estimation MAP table, and the charging cycle number corresponding to the historical charging process are obtained.

[0095] The historical charging state data in the historical charging process in the current life cycle of the vehicle battery can be various state information recorded by the vehicle in the current life cycle, such as current, voltage, temperature, SOC, etc. The historical charging remaining time can be a time prediction corresponding to the historical charging state data, indicating the charging remaining time predicted at that time. The initial charging remaining time interpolation estimation MAP table can be a table established by using known charging state and charging remaining time data through interpolation method, which is used to estimate future charging conditions. The charging cycle number can be the total number of charging and discharging cycles experienced by the battery.

[0096] Exemplarily, the vehicle battery management system obtains the historical charging state data in the historical charging process in the current life cycle of the vehicle battery, the historical charging remaining time corresponding to the historical charging state data, the initial charging remaining time interpolation estimation MAP table, and the charging cycle number corresponding to the historical charging process from the recorded historical data.

[0097] In step S404, the historical charging remaining time, the historical charging state data, the initial charging remaining time interpolation estimation MAP table, and the charging cycle number are input into the MAP table updating model to be trained to obtain a predicted charging remaining time interpolation estimation MAP table corresponding to the historical charging process.

[0098] The charging remaining time interpolation estimation MAP table and the MAP table updating model can be deployed in the vehicle battery management system. The vehicle battery management system inputs the historical charging remaining time, the historical charging state data, the initial charging remaining time interpolation estimation MAP table, and the charging cycle number into the MAP table updating model to be trained to obtain a predicted charging remaining time interpolation estimation MAP table corresponding to the historical charging process.

[0099] In step S406, according to the historical charging remaining time and the historical charging state data, the charging state data representing local fluctuation characteristics is removed to obtain target historical charging state data and corresponding target historical charging remaining time.

[0100] The target historical charging state data and the corresponding target historical charging remaining time can be obtained by removing the local fluctuation characteristics from the historical charging state data and the corresponding historical charging remaining time.

[0101] In step S408, an association relationship curve of the target historical charging remaining time and the target historical charging state data is obtained, a loss value is obtained according to a difference between the association relationship curve and a corresponding relationship curve of the MAP table, and the MAP table update model is trained by using the loss value.

[0102] Exemplarily, the vehicle battery management system analyzes the relationship between the target historical charging state data and the target historical charging remaining time, constructs an association relationship curve, compares the association relationship curve with a corresponding relationship curve generated by the MAP table, calculates a difference between the two relationship curves, obtains a loss value, and further trains the model by using the loss value to optimize the model parameters of the MAP table update model, so that the generated prediction charging remaining time interpolation estimation MAP table is more accurate.

[0103] In the embodiment, the loss value is further obtained by removing the charging state data with local fluctuation characteristics, and the model is trained, so that the model can more accurately capture the stable charging state trend, which helps to reduce the error caused by abnormal data noise, thereby improving the generation accuracy of the prediction charging remaining time interpolation estimation MAP table.

[0104] In one exemplary embodiment, as shown in Figure 5 The acquisition of the local fluctuation characteristics of the vehicle in the charging process includes:

[0105] In step S502, the historical charging remaining time of the vehicle from the start of charging to the current charging time in the current charging process is obtained.

[0106] The historical charging remaining time of the vehicle from the start of charging to the current charging time in the current charging process can refer to the charging remaining time data of the battery from the current charging time back to the start of charging in the charging process, which is updated in real time as the charging time changes.

[0107] In step S504, a plurality of historical charging remaining time change values are obtained according to the historical charging remaining time and a preset charging time sampling interval.

[0108] The preset charging time sampling interval can refer to a time interval for sampling data in a specific time period during the charging process. By periodically collecting data in the time interval, a plurality of historical charging remaining time change values in the charging process can be obtained.

[0109] For example, the time interval can be 1 minute, 3 minutes, 5 minutes, or other specific time intervals. Here, only examples are given, and no specific limitations are made. The actual needs can be set as required. Taking the time interval of 1 minute as an example, the historical charging remaining time is sampled every minute from the time when the charging starts, and the change value of the historical charging remaining time in each minute from the time when the charging starts is obtained.

[0110] In step S506, in a case where the historical charging remaining time change value is greater than the preset time change threshold, the charging state data corresponding to the historical charging remaining time change value associated with the charging time interval is taken as the local fluctuation feature.

[0111] The preset time change threshold is used to determine whether the historical charging remaining time change value exceeds the expectation.

[0112] For example, the preset time change threshold is threshold M, the preset charging time sampling interval is 3 minutes, and a plurality of historical charging remaining time change values are obtained, including change value A, change value B, change value C, change value D, and the like. Change values A and C are greater than threshold M, and therefore, the charging state data corresponding to the 3-minute charging time interval associated with change values A and C is taken as the local fluctuation feature.

[0113] In this embodiment, the change value of the historical charging remaining time is obtained through the preset charging time sampling interval, and the local fluctuation feature in the charging process is effectively identified and obtained through the preset time change threshold. The local fluctuation feature is associated with the charging state data, thereby providing important reference data for subsequent correction of the charging remaining time.

[0114] In an example embodiment, according to the local fluctuation feature, a charging remaining time correction amount of the vehicle at the current charging time is obtained, including: inputting the local fluctuation feature into a pre-constructed local fluctuation feature influence time function to obtain a charging remaining time correction amount corresponding to the local fluctuation feature.

[0115] The local fluctuation feature influence time function is based on historical charging state data of the vehicle battery in the current life cycle and historical charging remaining time corresponding to the historical charging state data.

[0116] The pre-constructed local fluctuation feature influence time function can be used to output a time change amount of the influence of the local fluctuation feature on the charging remaining time according to the input local fluctuation feature.

[0117] Exemplarily, the local fluctuation features include feature A and feature B, the vehicle battery management system inputs the feature A and the feature B into the pre-constructed local fluctuation feature influence function, and calculates a charging remaining time correction amount corresponding to the local fluctuation feature through the local fluctuation feature influence function, that is, the feature A outputs a time correction amount a, and the feature B outputs a time correction amount b, and then the time correction amount a and the time correction amount b can be corrected.

[0118] In the embodiment, the charging remaining time correction amount is calculated according to the local fluctuation feature and the local fluctuation feature influence time function, which can be used to adjust the initial charging remaining time, and the actual charging remaining time can be more accurately predicted.

[0119] In an exemplary embodiment, the step of obtaining the local fluctuation feature influence time function includes:

[0120] In step S11, historical charging state data in a current life cycle of a vehicle battery and historical charging remaining time corresponding to the historical charging state data are obtained.

[0121] The historical charging state data in the current life cycle of the vehicle battery can refer to various data recorded during the use of the vehicle battery so far, such as SOC, voltage, current, temperature, charging time, etc. The historical charging remaining time corresponding to the historical charging state data can refer to the remaining time required for the vehicle to complete charging corresponding to each charging time in the historical charging state data.

[0122] In step S12, the historical charging remaining time is sampled by using a preset charging time sampling interval to obtain a plurality of historical charging remaining time change values, and in the case that the historical charging remaining time change value is greater than a preset time change threshold, the charging state data corresponding to the historical charging remaining time change value is taken as a local fluctuation feature, and each local fluctuation feature is taken to form a local fluctuation feature sequence.

[0123] The preset charging time sampling interval can refer to a time interval for selecting a specific time period for data sampling during the charging process. By periodically collecting data within the time interval, a plurality of historical charging remaining time change values during the charging process can be obtained. The preset time change threshold is used to determine whether the historical charging remaining time change value exceeds the expectation. The local fluctuation feature sequence can refer to a sequence composed of a plurality of local fluctuation features.

[0124] In step S13, the particle filtering algorithm is used to track and predict the local fluctuation feature sequence to obtain a predicted local fluctuation feature sequence, and the polynomial regression is used to fit the predicted local fluctuation feature sequence to obtain the local fluctuation feature influence time function.

[0125] wherein the particle filter algorithm is a recursive estimation method based on Bayesian theory, which estimates the state of the system by weighting and resampling a set of particles, and is suitable for systems with nonlinear and non-Gaussian noise. Polynomial regression is a regression analysis method that uses a polynomial function to fit data to capture the nonlinear relationship in the data. The local fluctuation feature influence time function can be a function obtained by analyzing and modeling the local fluctuation feature, which describes the influence of the local fluctuation feature on the remaining charging time.

[0126] Illustratively, the vehicle management system uses the particle filter algorithm to track and predict the local fluctuation feature sequence to obtain a predicted local fluctuation feature sequence, and performs polynomial regression fitting on the predicted local fluctuation feature sequence to obtain the local fluctuation feature influence time function.

[0127] In this embodiment, the particle filter algorithm is used to track and predict the local fluctuation feature sequence, and the predicted local fluctuation feature sequence is fitted using polynomial regression, which can more accurately extract the correlation between the local fluctuation feature and the remaining charging time, making the obtained local fluctuation feature influence time function more accurate.

[0128] In one exemplary embodiment, the above step S13 can specifically include: initializing using an initial probability density filter function to generate a particle swarm; updating the weight of each particle and normalizing to obtain the weight of each particle; resampling each particle to obtain a new particle set and weight; predicting the local fluctuation feature sequence using a state equation to obtain a predicted local fluctuation feature sequence, fitting the predicted local fluctuation feature sequence using polynomial regression to obtain a monomial polynomial regression function; solving the parameters of the monomial polynomial regression function using the least squares method to obtain a coefficient matrix; and obtaining the local fluctuation feature influence time function according to the coefficient matrix and the monomial polynomial regression function.

[0129] Specifically, the above step S13 can specifically include:

[0130] Step 1: Initialize using an initial probability density filter function to generate a particle swarm. Initialize model parameters x 0 =[ a 0 , b 0 , c 0 , d 0 , e 0 , f 0 ] from the prior probability P( ) generating a particle set , N is the total number of particles, and the weight of all particles is .

[0131] Step 2: When the measured value is observed, the state equation can be updated as follows:

[0132]

[0133] Step 3: The particle is calculated in the kth cycle as shown in the following equation:

[0134]

[0135] Step 4: Update the weight of each particle and normalize to obtain the weight of each particle; update the particle weight at time k and normalize:

[0136]

[0137] where is the weight of the ith particle at time k, is the observation at time k, is the ith particle corresponding to time k, and then the mean square error estimate of the particle is solved by the following formula:

[0138]

[0139] Step 5: Expand each state variable using the Taylor formula, which can be used as the state value with a fluctuation time of 0, and the Taylor expansion formula can be represented by the following formula:

[0140]

[0141] Because the formula can be optimized to obtain the following one-dimensional polynomial regression function:

[0142] Step 6: Solve the model parameters for the above one-dimensional polynomial regression function by the least square method, and the formula is as follows:

[0143]

[0144] The expanded formula of the above formula is as follows:

[0145]

[0146] And the equation after the expansion is decomposed into matrix form, and the Vandermonde decomposition is carried out to obtain the coefficient matrix, and the obtained coefficient matrix is substituted into the above one-dimensional polynomial regression function to finally obtain the local fluctuation characteristic influence time function :

[0147]

[0148] In this embodiment, the particle filter can process nonlinear and non-Gaussian noise, and can reflect the changes of the local fluctuation characteristics in real time by generating and updating the particle group. This method performs well in dynamic systems and can provide more accurate state estimation. By performing polynomial regression on the predicted local fluctuation characteristic sequence, the nonlinear relationship in the data can be captured, further improving the accuracy of the prediction. The combination of particle filtering and polynomial regression can ensure accuracy while maintaining high computational efficiency, making it suitable for real-time applications.

[0149] In one specific embodiment, a vehicle charging remaining time determination method is provided, specifically comprising:

[0150] Obtaining the initial charging remaining time of the vehicle at the current charging time in the charging process and the local fluctuation characteristics of the vehicle in the charging process; the local fluctuation characteristics include charging state data associated with the fluctuation of the charging remaining time; the initial charging remaining time is obtained based on the global degradation characteristics of the vehicle battery; the global degradation characteristics are battery data associated with the performance decay of the vehicle battery at the current charging time. It should be noted that the extraction of global degradation characteristics is mainly aimed at the case that the battery performance will decay regularly as the battery usage time increases and the number of charge and discharge cycles accumulates, resulting in corresponding time following changes in the MAP table for interpolation estimation of the charging remaining time. Generally, the battery SOH state is only updated according to the number of cycles, and the overall performance decay of the battery is not directly compensated for the charging remaining time. Therefore, we also need to collect data of global degradation characteristics. The global degradation characteristics phenomenon exists, but the local abnormal fluctuation characteristics are not necessarily an event. Therefore, the global degradation characteristics phenomenon is a phenomenon that exists throughout the service life, so the estimation of the charging remaining time should be one of the compensation conditions.

[0151] The step of obtaining the initial charging remaining time of the vehicle at the current charging time in the charging process includes: inputting the global degradation characteristics into the pre-trained MAP table update model to obtain the corresponding charging remaining time interpolation estimation MAP table at the current charging time; obtaining the charging remaining time matched with the charging state data at the current charging time from the charging remaining time interpolation estimation MAP table as the initial charging remaining time.

[0152] The training step of the MAP table updating model comprises: obtaining historical charging state data in a historical charging process in a current life cycle of a vehicle battery, historical charging residual time corresponding to the historical charging state data, an initial charging residual time interpolation estimation MAP table, and a charging cycle number corresponding to the historical charging process; inputting the historical charging residual time, the historical charging state data, the initial charging residual time interpolation estimation MAP table, and the charging cycle number into the MAP table updating model to be trained to obtain a predicted charging residual time interpolation estimation MAP table corresponding to the historical charging process; removing charging state data representing local fluctuation characteristics according to the historical charging residual time and the historical charging state data to obtain target historical charging state data and corresponding target historical charging residual time; obtaining a correlation relationship curve of the target historical charging residual time and the target historical charging state data, and obtaining a loss value according to a difference between the correlation relationship curve and a corresponding relationship curve of the predicted charging residual time interpolation estimation MAP table, and training the MAP table updating model by using the loss value.

[0153] The step of obtaining the local fluctuation characteristics of the vehicle in the charging process comprises: obtaining historical charging residual time of the vehicle from the start of charging to the current charging time in the current charging process; obtaining a plurality of historical charging residual time change values according to the historical charging residual time and a preset charging time sampling interval; in the case that the historical charging residual time change value is greater than a preset time change threshold, charging state data corresponding to the historical charging residual time change value and the charging time interval are taken as the local fluctuation characteristics.

[0154] The local fluctuation characteristics are input into a pre-constructed local fluctuation characteristic influence time function to obtain a charging residual time correction amount corresponding to the local fluctuation characteristics; the local fluctuation characteristic influence time function is obtained based on historical charging state data in a current life cycle of a vehicle battery and historical charging residual time corresponding to the historical charging state data. The step of obtaining the local fluctuation characteristic influence time function comprises: obtaining the historical charging state data in the current life cycle of the vehicle battery and the historical charging residual time corresponding to the historical charging state data; sampling the historical charging residual time by using a preset charging time sampling interval to obtain a plurality of historical charging residual time change values; in the case that the historical charging residual time change value is greater than a preset time change threshold, charging state data corresponding to the historical charging residual time change value are taken as the local fluctuation characteristics, and the local fluctuation characteristics are grouped into a local fluctuation characteristic sequence.

[0155] The local fluctuation characteristics refer to instantaneous time changes caused by other factors, and the extraction of the local fluctuation characteristics can be extraction of each charging as a sampling sample, and extraction of relatively abnormal fluctuations therefrom, for example, a time range The local feature is extracted when the deviation of the charging remaining time from the preset charging time sampling interval is greater than M (a preset time change threshold, which can be determined by experiment within a threshold range) minutes. A total of N local abnormal fluctuation samples are taken. Each time a local feature abnormal change occurs, there will be a corresponding factor causing it, such as a sudden change in voltage, a sudden change in temperature, a jump in SOC, and an abnormal fluctuation in current, all of which will cause a large change in the estimation of the current charging remaining time. Therefore, a function relationship equation is established between each local abnormal fluctuation and the current, temperature, SOC, and voltage, and each local feature is composed into a corresponding array sequence. The formula is as follows:

[0156]

[0157] The initial probability density filter function is used for initialization to generate a particle swarm. The weight of each particle is updated and normalized to obtain the weight of each particle. Each particle is resampled to obtain a new particle set and weight. The state equation is used to predict the local fluctuation feature sequence to obtain a predicted local fluctuation feature sequence. A polynomial regression is used to fit the predicted local fluctuation feature sequence to obtain a one-dimensional polynomial regression function. The least squares method is used to solve the parameters of the one-dimensional polynomial regression function to obtain a coefficient matrix. According to the coefficient matrix and the one-dimensional polynomial regression function, a local fluctuation feature influence time function is obtained.

[0158] Step 1: The initial probability density filter function is used for initialization to generate a particle swarm. The model parameters are initialized x 0 =[ a 0 , b 0 , c 0 , d 0 , e 0 , f 0 ] The particle set is generated from the prior probability P( ). N is the total number of particles, and the weight of all particles is .

[0159] Step 2: When the measurement value is observed, the state equation can be updated as follows:

[0160]

[0161] Step 3: The particle is calculated in the kth cycle as follows:

[0162]

[0163] Step 4: Update and normalize the weight of each particle to obtain the weight of each particle; update the particle weight at time k and normalize:

[0164]

[0165] wherein is the weight of the i-th particle at time k, is the observation at time k, is the i-th particle corresponding to time k, and then the mean square error estimate of the particle is solved by the following formula:

[0166]

[0167] Step 5: expand each state variable using Taylor's formula, which can be used as the state value with a fluctuation time of 0, and the Taylor expansion formula can be represented by the following formula:

[0168]

[0169] Because The formula can be optimized to obtain the following one-dimensional polynomial regression function:

[0170] Step 6: Solve the model parameters for the above one-dimensional polynomial regression function by least squares method, and the formula is as follows:

[0171]

[0172] The expanded formula of the above formula is as follows:

[0173]

[0174] And the expanded equation is decomposed into a matrix form, and the Vandermonde decomposition is performed to obtain the coefficient matrix, and the obtained coefficient matrix is substituted into the above one-dimensional polynomial regression function to finally obtain the local fluctuation characteristic influence time function :

[0175]

[0176] The initial charging remaining time is corrected by the charging remaining time correction amount to obtain the target charging remaining time of the vehicle at the current charging time, which can be specifically: the target charging remaining time can be obtained by subtracting the time change caused by the characteristic fluctuation from the initial charging remaining time obtained by the BMS through the interpolation method, and the specific formula can be simplified as the following equation:

[0177]

[0178] wherein i represents current, v represents voltage, and T represents temperature.

[0179] In this embodiment, it can help to accurately estimate the remaining charging time, improve the estimation accuracy of the remaining time; it can help the whole vehicle and the battery parts to more accurately understand the service life and safety level; it is helpful to the standardization of industry algorithm software function, and it is easier to upgrade and iterate.

[0180] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0181] Based on the same inventive concept, the embodiments of the present application also provide a vehicle charging remaining time determination device for implementing the vehicle charging remaining time determination method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more vehicle charging remaining time determination device embodiments provided below can refer to the limitations of the vehicle charging remaining time determination method in the above, which will not be repeated here.

[0182] In one exemplary embodiment, as shown in Figure 6 A vehicle charging remaining time determination device 600 is provided, comprising a feature extraction module 601 and a remaining time correction module 602, wherein:

[0183] The feature extraction module 601 is configured to obtain an initial charging remaining time of the vehicle at a current charging time during charging and a charging remaining time influence feature of the vehicle during charging.

[0184] The remaining time correction module 602 is configured to correct the initial charging remaining time using the charging remaining time influence feature to obtain a target charging remaining time of the vehicle at the current charging time.

[0185] In an exemplary embodiment, the charging remaining time influencing features include global degradation features and local fluctuation features of the vehicle during the charging process; the global degradation features are battery data associated with the battery degradation of the vehicle battery at the current charging time; the local fluctuation features include charging state data associated with the fluctuation of the charging remaining time during the charging process; the above-mentioned remaining time correction module 602 is further configured to correct the initial charging remaining time by using at least one of the global degradation features and the local fluctuation features, to obtain the target charging remaining time of the vehicle at the current charging time.

[0186] In an exemplary embodiment, the charging remaining time influencing features include global degradation features and local fluctuation features of the vehicle during the charging process; the global degradation features are battery data associated with the battery degradation of the vehicle battery at the current charging time; the local fluctuation features include charging state data associated with the fluctuation of the charging remaining time during the charging process; the above-mentioned remaining time correction module 602 is further configured to correct the initial charging remaining time according to the global degradation features to obtain the corrected initial charging remaining time; obtain a charging remaining time correction amount of the vehicle at the current charging time according to the local fluctuation features; correct the corrected initial charging remaining time by using the charging remaining time correction amount to obtain the target charging remaining time of the vehicle at the current charging time.

[0187] In an exemplary embodiment, the above-mentioned remaining time correction module 602 is further configured to input the global degradation features into a time correction model, and correct the initial charging remaining time by using the time correction model and the charging state data at the current charging time to obtain the corrected initial charging remaining time.

[0188] In an exemplary embodiment, the above-mentioned time correction model is a pre-trained MAP table updating model; the above-mentioned remaining time correction module 602 is further configured to input the global degradation features into the pre-trained MAP table updating model to obtain a corresponding charging remaining time interpolation estimation MAP table at the current charging time; and obtain the charging remaining time matched with the charging state data at the current charging time from the charging remaining time interpolation estimation MAP table as the initial charging remaining time.

[0189] In an example embodiment, the remaining time correction module 602 is further configured to obtain historical charging state data in a historical charging process in a current life cycle of the vehicle battery, historical charging remaining time corresponding to the historical charging state data, an initial charging remaining time interpolation estimation MAP table, and a charging cycle number corresponding to the historical charging process; input the historical charging remaining time, the historical charging state data, the initial charging remaining time interpolation estimation MAP table, and the charging cycle number into a MAP table updating model to be trained to obtain a predicted charging remaining time interpolation estimation MAP table corresponding to the historical charging process; remove charging state data representing local fluctuation characteristics from the historical charging state data according to the historical charging remaining time and the historical charging state data to obtain target historical charging state data and corresponding target historical charging remaining time; obtain a correlation relationship curve of the target historical charging remaining time and the target historical charging state data, and obtain a loss value according to a difference between the correlation relationship curve and a corresponding relationship curve of the predicted charging remaining time interpolation estimation MAP table, and train the MAP table updating model using the loss value.

[0190] In an example embodiment, the remaining time correction module 602 is further configured to input the local fluctuation characteristics into a pre-constructed local fluctuation characteristic influence time function to obtain a charging remaining time correction amount corresponding to the local fluctuation characteristics; and the local fluctuation characteristic influence time function is obtained based on historical charging state data in a current life cycle of the vehicle battery and historical charging remaining time corresponding to the historical charging state data.

[0191] In an example embodiment, the remaining time correction module 602 is further configured to obtain historical charging state data in a current life cycle of the vehicle battery and historical charging remaining time corresponding to the historical charging state data; sample the historical charging remaining time using a pre-set charging time sampling interval to obtain a plurality of historical charging remaining time change values, and in a case where the historical charging remaining time change value is greater than a pre-set time change threshold, charging state data corresponding to the historical charging remaining time change value is taken as a local fluctuation characteristic, and each local fluctuation characteristic is combined to form a local fluctuation characteristic sequence; track and predict the local fluctuation characteristic sequence using a particle filtering algorithm to obtain a predicted local fluctuation characteristic sequence, and fit the predicted local fluctuation characteristic sequence using a polynomial regression to obtain a local fluctuation characteristic influence time function.

[0192] In an example embodiment, the remaining time correction module 602 is further configured to initialize a particle swarm using an initial probability density filter function, update and normalize a weight of each particle to obtain a weight of each particle, resample each particle to obtain a new particle set and a weight, predict a local fluctuation feature sequence using a state equation to obtain a predicted local fluctuation feature sequence, fit the predicted local fluctuation feature sequence using a polynomial regression to obtain a one-dimensional polynomial regression function, solve parameters of the one-dimensional polynomial regression function using a least square method to obtain a coefficient matrix, and obtain a local fluctuation feature influence time function based on the coefficient matrix and the one-dimensional polynomial regression function.

[0193] In an example embodiment, the feature extraction module 601 is further configured to obtain a historical charging remaining time of the vehicle from a start of charging to a current charging time in a current charging process, obtain a plurality of historical charging remaining time change values based on the historical charging remaining time and a preset charging time sampling interval, and obtain charging state data corresponding to a local fluctuation feature in a case where the historical charging remaining time change value is greater than a preset time change threshold.

[0194] The modules in the vehicle charging remaining time determination apparatus can be implemented in whole or in part by software, hardware, or a combination thereof. The modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform operations corresponding to the modules.

[0195] In an example embodiment, a vehicle-side control device is provided, an internal structure diagram of which can be as shown in FIG. 6, including a memory and a processor, and the memory stores a computer program, and the processor implements steps in the above method embodiments when executing the computer program. Figure 7

[0196] In an example embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement steps in the above method embodiments.

[0197] In an example embodiment, a computer program product is provided, which includes a computer program, and the computer program is executed by a processor to implement steps in the above method embodiments.

[0198] ​Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0199] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0200] The above-described embodiments are merely illustrative of several embodiments of the present application, which are described in more detail and in a specific manner, but should not be construed as limiting the scope of the patent of the present application. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for determining the remaining charging time of a vehicle, characterized in that, The method includes: The system acquires the initial remaining charging time at the current charging moment during the vehicle's charging process, as well as the characteristics affecting the remaining charging time during the charging process. The remaining charging time impact characteristics include global degradation characteristics and local fluctuation characteristics of the vehicle during the charging process. The global degradation characteristics are battery data related to the degradation of the vehicle's battery charging and discharging performance at the current charging moment. The local fluctuation characteristics include charging state data related to fluctuations in the remaining charging time during the charging process. The initial remaining charging time is first corrected based on the global degradation characteristics to obtain the corrected initial remaining charging time. The local fluctuation feature is input into a pre-constructed local fluctuation feature influence time function to obtain the charging remaining time correction amount corresponding to the local fluctuation feature; the local fluctuation feature influence time function is obtained based on the historical charging state data within the current life cycle of the vehicle battery and the historical charging remaining time corresponding to the historical charging state data; The initial remaining charging time is corrected a second time using the remaining charging time correction amount to obtain the target remaining charging time of the vehicle at the current charging time.

2. The method according to claim 1, characterized in that, The step of performing a first correction on the initial remaining charging time based on the global degradation characteristics to obtain the corrected initial remaining charging time includes: The global degradation feature is input into the time correction model, and the initial remaining charging time is corrected by the time correction model and the charging status data at the current charging time to obtain the corrected initial remaining charging time.

3. The method according to claim 1, characterized in that, The steps for obtaining the time function influenced by the local fluctuation characteristics include: Obtain historical charging status data within the current lifecycle of the vehicle battery and the historical remaining charging time corresponding to the historical charging status data; The historical charging remaining time is sampled using a preset charging time sampling interval to obtain multiple historical charging remaining time change values. When the historical charging remaining time change value is greater than a preset time change threshold, the charging status data corresponding to the historical charging remaining time change value is used as a local fluctuation feature, and the local fluctuation features are combined into a local fluctuation feature sequence. The local fluctuation feature sequence is tracked and predicted using a particle filter algorithm to obtain a predicted local fluctuation feature sequence. The predicted local fluctuation feature sequence is then fitted using multinomial regression to obtain the time function of the local fluctuation feature influence.

4. The method according to claim 3, characterized in that, The process involves using a particle filter algorithm to track and predict the local fluctuation feature sequence to obtain a predicted local fluctuation feature sequence, and then using multinomial regression to fit the predicted local fluctuation feature sequence to obtain a time function that influences the local fluctuation features, including: The initial probability density filtering function is used for initialization to generate a particle swarm. The weights of each particle are updated and normalized to obtain the weight of each particle. Each particle is resampled to obtain a new set of particles and weights; The predicted local fluctuation feature sequence is obtained by using the state equation to predict the local fluctuation feature sequence, and the predicted local fluctuation feature sequence is fitted by polynomial regression to obtain a univariate polynomial regression function. The coefficient matrix is ​​obtained by solving for the parameters of the univariate polynomial regression function using the least squares method. Based on the coefficient matrix and the univariate polynomial regression function, the time function of the influence of the local fluctuation characteristics is obtained.

5. The method according to claim 1, characterized in that, The step of obtaining the local fluctuation characteristics of the vehicle during the charging process includes: Obtain the historical remaining charging time of the vehicle from the start of charging to the current charging moment; Based on the historical remaining charging time and the preset charging time sampling interval, multiple historical remaining charging time change values ​​are obtained; If the change value of the remaining historical charging time is greater than a preset time change threshold, the charging status data corresponding to the charging time interval associated with the change value of the remaining historical charging time is used as a local fluctuation feature.

6. A device for determining the remaining charging time of a vehicle, characterized in that, The device includes: The feature extraction module is used to obtain the initial remaining charging time of the vehicle at the current charging moment during the charging process and the features affecting the remaining charging time of the vehicle during the charging process; the features affecting the remaining charging time include global degradation features and local fluctuation features of the vehicle during the charging process; the global degradation features are battery data related to the degradation of the vehicle battery's charge and discharge performance at the current charging moment; the local fluctuation features include charging state data related to the fluctuation of the remaining charging time during the charging process; The remaining time correction module is used to perform a first correction on the initial remaining charging time based on the global degradation characteristics to obtain a corrected initial remaining charging time; input the local fluctuation characteristics into a pre-constructed local fluctuation characteristic influence time function to obtain a remaining charging time correction amount corresponding to the local fluctuation characteristics; the local fluctuation characteristic influence time function is obtained based on historical charging state data within the current life cycle of the vehicle battery and the historical remaining charging time corresponding to the historical charging state data; and use the remaining charging time correction amount to perform a second correction on the corrected initial remaining charging time to obtain the target remaining charging time of the vehicle at the current charging moment.

7. A vehicle-end control device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images