CN112068004A

CN112068004A - Method and device for determining battery abnormity and battery charging remaining time

Info

Publication number: CN112068004A
Application number: CN202010977252.5A
Authority: CN
Inventors: 杨静
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2020-09-16
Filing date: 2020-09-16
Publication date: 2020-12-11

Abstract

The application relates to the technical field of vehicles, in particular to a method and a device for determining battery abnormity and battery charging remaining time. The method for determining the battery abnormity comprises the steps of obtaining vehicle state data of at least two adjacent moments including the current moment, inputting the obtained vehicle state data into a trained battery temperature parameter estimation model to obtain a battery predicted temperature parameter of the current moment, and further determining whether the battery is abnormal or not based on the battery predicted temperature parameter and a battery actual temperature parameter. By predicting the temperature parameter by using the vehicle state data of a plurality of adjacent moments which can affect the temperature of the battery, the accuracy of predicting the temperature parameter can be improved, and further, the reliability of judging whether the battery is abnormal can be improved.

Description

Method and device for determining battery abnormity and battery charging remaining time

Technical Field

The application relates to the technical field of vehicles, in particular to a method and a device for determining battery abnormity and battery charging remaining time.

Background

Batteries are a power source for electric vehicles and are an important component of electric vehicles. The battery temperature parameter is a key parameter for measuring whether the battery is abnormal or not.

At present, a prediction method of battery temperature parameters depends on internal thermal parameters of batteries, and only a battery factory generally has relevant parameter information; or simply and mechanically judging whether the temperature of the battery is abnormal or not only by judging whether the temperature difference of the battery exceeds a set threshold value or not, wherein the method is simple and rough, but the intelligent temperature inconsistency diagnosis cannot be carried out according to the real-time working condition of the battery; the current and the voltage of the battery are estimated by means of machine learning algorithms such as support vector machines, random forests and other methods and standard neural networks, but the data samples selected by the algorithms are relatively independent and randomly disordered to train a model, and the machine learning methods cannot effectively estimate battery temperature parameters which have time memory and have gentle change trend due to strong correlation between the temperature and the time.

Therefore, the method for determining whether the battery is abnormal is not accurate.

Disclosure of Invention

In view of this, the present disclosure provides at least a method and an apparatus for determining battery abnormality and remaining battery charging time, which can improve the accuracy of predicting temperature parameters and further improve the reliability of determining whether a battery is abnormal.

The application mainly comprises the following aspects:

in a first aspect, an embodiment of the present application provides a method for determining a battery abnormality, where the method for determining a battery abnormality includes:

acquiring vehicle state data of at least two adjacent moments including a current moment; the vehicle state data is data that affects a battery temperature;

inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery temperature parameter estimation model to obtain a battery predicted temperature parameter of the current moment;

and determining whether the battery is abnormal or not based on the battery predicted temperature parameter and the battery actual temperature parameter.

In one possible embodiment, the temperature parameters include a battery temperature range and a battery temperature; the battery actual temperature range is a difference value between the highest actual temperature and the lowest actual temperature of the battery, wherein the highest actual temperature is the highest temperature of the temperatures acquired at all the sampling points of the battery, and the lowest actual temperature is the lowest temperature of the temperatures acquired at all the sampling points of the battery.

In one possible embodiment, the vehicle state data comprises at least two of the following data:

ambient temperature, driving data, real-time performance data of the battery.

In one possible embodiment, the battery temperature parameter pre-estimation model is composed of at least two cycle layers connected in sequence; the step of inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery temperature parameter estimation model to obtain the battery predicted temperature parameter of the current moment comprises the following steps:

inputting the acquired vehicle state data at the first moment into the first circulation layer to obtain first intermediate state data; the first time is the earliest of the at least two adjacent times;

inputting the first intermediate state data and the vehicle state data at a second moment into a next circulation layer to obtain second intermediate state data; the second moment is the next moment of the first moment;

if the vehicle state data at more than two moments are obtained, taking the second intermediate state data as the first intermediate state data, and repeatedly executing the steps to input the first intermediate state data and the vehicle state data at a third moment into a next circulation layer to obtain the second intermediate state data; the third moment is the next moment of the second moment;

and if the vehicle state data at two moments are only obtained, outputting the battery predicted temperature parameter at the current moment.

In one possible embodiment, the battery temperature parameter estimation model is trained according to the following steps:

obtaining the vehicle state data corresponding to a plurality of moments of a sample vehicle, and taking the vehicle state data of the sample vehicle as sample data;

adding a sample label for each sample data; the sample label is a battery standard temperature parameter corresponding to each moment;

and training the initial temperature parameter estimation model according to the preset time steps, the sample data and the sample label corresponding to each sample data to obtain the battery temperature parameter estimation model.

In a possible implementation manner, the training an initial temperature parameter estimation model according to a preset time step, the sample data, and a sample label corresponding to each sample data to obtain the battery temperature parameter estimation model includes:

determining entropy loss between a battery standard temperature parameter and an output parameter corresponding to each model training;

adjusting model parameters of the initial temperature parameter estimation model according to the entropy loss;

if the entropy loss is larger than a first preset threshold value, continuing to train the initial temperature parameter estimation model;

and if the entropy loss is smaller than or equal to the first preset threshold value, stopping training the initial temperature parameter estimation model to obtain the battery temperature parameter estimation model.

In a possible embodiment, the determining whether the battery is abnormal based on the predicted battery temperature parameter and the actual battery temperature parameter includes:

judging whether the difference value between the battery predicted temperature parameter and the battery actual temperature parameter is larger than a second preset threshold value or not;

if so, determining that the battery is abnormal;

if not, determining that the battery is not abnormal.

counting the occurrence times of the difference value between the battery predicted temperature parameter and the battery actual temperature parameter corresponding to the previous time which is greater than a third preset threshold value;

and if the occurrence frequency is greater than or equal to the preset frequency, determining that the battery is abnormal.

In one possible embodiment, after determining that the abnormality occurs in the battery, the determination method further includes:

determining an abnormal grade according to a difference value between the battery predicted temperature parameter and the battery actual temperature parameter;

determining a disposal strategy for processing the battery according to the abnormity level.

In a second aspect, an embodiment of the present application further provides a method for determining a remaining battery charging time, where the method for determining includes:

acquiring vehicle state data of at least two adjacent moments including a current moment; the vehicle state data is data that affects a battery charging time;

and inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery charging remaining time estimation model to obtain the battery charging remaining time.

In one possible embodiment, the vehicle state data includes:

battery temperature, said battery real-time performance data.

In one possible embodiment, the battery charge remaining time estimation model is composed of at least two cycle layers connected in sequence; the step of inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery charging remaining time estimation model to obtain the battery charging remaining time of the current moment comprises the following steps:

if the vehicle-shaped y state data at more than two moments are acquired as the first intermediate state data, repeatedly executing the step, and inputting the first intermediate state data and the vehicle-shaped y state data at a third moment into a next circulation layer to acquire second intermediate state data; the third moment is the next moment of the second moment;

and if the vehicle state data at two moments are only acquired, outputting the battery charging remaining time at the current moment.

In one possible embodiment, the battery charge remaining time estimation model is trained according to the following steps:

adding a sample label for each sample data; the sample label is the standard charging remaining time of the battery corresponding to each moment;

and training the initial battery charging remaining time estimation model according to the preset time step number, the sample data and the sample label corresponding to each sample data to obtain the battery charging remaining time estimation model.

In a possible implementation manner, the training the initial charge remaining time estimation model according to a preset time step, the sample data, and a sample tag corresponding to each sample data to obtain the battery charge remaining time estimation model includes:

determining entropy loss between standard charging remaining time and output time of the battery corresponding to each model training;

adjusting model parameters of the initial charging remaining time estimation model according to the entropy loss;

if the entropy loss is larger than a first preset threshold value, continuing to train the initial charging remaining time estimation model;

and if the entropy loss is smaller than or equal to the first preset threshold, stopping training the initial charging remaining time estimation model to obtain the battery charging remaining time estimation model.

In a third aspect, an embodiment of the present application further provides a device for determining a battery abnormality, where the device for determining a battery abnormality includes:

the first acquisition module is used for acquiring vehicle state data of at least two adjacent moments including the current moment; the vehicle state data is data that affects a battery temperature;

the first input module is used for inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery temperature parameter estimation model to obtain a battery predicted temperature parameter of the current moment;

and the first determination module is used for determining whether the battery is abnormal or not based on the battery predicted temperature parameter and the battery actual temperature parameter.

In a fourth aspect, an embodiment of the present application further provides a device for determining a remaining battery charging time, where the device for determining a remaining battery charging time includes:

the second acquisition module is used for acquiring vehicle state data of at least two adjacent moments including the current moment; the vehicle state data is data that affects a battery charging time;

and the second input module is used for inputting the acquired vehicle state data of the at least two adjacent moments into the trained battery charging remaining time estimation model to obtain the battery charging remaining time.

In a fifth aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and the machine-readable instructions are executed by the processor to perform the steps of the method for determining battery abnormality according to any one of the possible embodiments of the first aspect or the first aspect, and/or to perform the steps of the method for determining remaining battery charging time according to any one of the possible embodiments of the second aspect or the second aspect.

In a sixth aspect, the present embodiments also provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to perform the steps of the method for determining battery abnormality in any one of the above-mentioned first aspect or any one of the above-mentioned possible embodiments of the first aspect, and/or to perform the steps of the method for determining remaining battery charging time in any one of the above-mentioned second aspect or any one of the above-mentioned possible embodiments of the second aspect.

The method and the device for determining the battery abnormity, provided by the embodiment of the application, adopt the vehicle state data of a plurality of adjacent moments which can influence the battery temperature to predict the temperature parameter, and compare with the method and the device which rely on the battery factory to have the internal parameter of the battery to predict the battery temperature parameter, or simply and mechanically judge whether the battery temperature is abnormal by judging whether the battery temperature difference exceeds the set threshold value, and can not carry out intelligent temperature inconsistency diagnosis according to the real-time working condition of the battery, and a machine learning method can not effectively estimate the battery temperature parameter which has time memory and has a gentle change trend.

Further, the method for determining the remaining battery charging time provided by the embodiment of the application obtains the remaining battery charging time of the target vehicle by obtaining the vehicle state data of at least two adjacent moments including the current moment, wherein the vehicle state data is data influencing the battery charging time, and inputting the obtained vehicle state data of the at least two adjacent moments into a trained estimated model of the remaining battery charging time, and compared with the method for calculating the remaining battery charging time based on the remaining battery charging time and the current in the prior art, the method for determining the remaining battery charging time has lower precision due to the factors that the charging current is unstable, the remaining battery charging time is attenuated along with the battery, the electric quantity is also attenuated along with the battery, and the like, and can improve the accuracy of predicting the remaining battery charging time.

In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

Fig. 1 is a flowchart illustrating a method for determining battery abnormality according to an embodiment of the present disclosure;

FIG. 2 is a flow chart illustrating a determination of remaining battery charge time provided by an embodiment of the present application;

fig. 3 is a functional block diagram of a battery abnormality determination apparatus according to an embodiment of the present application;

FIG. 4 is a functional block diagram of a first input block of the apparatus for determining battery abnormality of FIG. 3;

fig. 5 is a second functional block diagram of a battery abnormality determination apparatus according to an embodiment of the present application;

fig. 6 is a functional block diagram of a device for determining the remaining battery charge time according to an embodiment of the present disclosure;

FIG. 7 is a functional block diagram of a second input block of the apparatus for determining remaining battery charge time of FIG. 6;

fig. 8 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application.

Detailed Description

To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. Additionally, it should be understood that the schematic drawings are not necessarily drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and that steps without logical context may be performed in reverse order or concurrently. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.

In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

To enable those skilled in the art to use the present disclosure in conjunction with a particular application scenario "determine if a battery is abnormal", the following embodiments are presented, and it will be apparent to those skilled in the art that the general principles defined herein may be applied to other embodiments and application scenarios without departing from the spirit and scope of the present disclosure.

The method, the apparatus, the electronic device, or the computer-readable storage medium described in the embodiments of the present application may be applied to any scenario in which battery abnormality needs to be performed, and the embodiments of the present application do not limit a specific application scenario, and any scheme using the method and the apparatus for determining battery abnormality provided in the embodiments of the present application is within the scope of protection of the present application.

It is noted that, before the present application is proposed, in the prior art, a method for predicting battery temperature parameters relies on internal thermal parameters of the battery, and only a battery manufacturer generally has relevant parameter information; or simply and mechanically judging whether the temperature of the battery is abnormal or not only by judging whether the temperature difference of the battery exceeds a set threshold value or not, wherein the method is simple and rough, but the intelligent temperature inconsistency diagnosis cannot be carried out according to the real-time working condition of the battery; the current and the voltage of the battery are estimated by means of machine learning algorithms such as support vector machines, random forests and other methods and standard neural networks, but the data samples selected by the algorithms are relatively independent and randomly disordered to train a model, and the machine learning methods cannot effectively estimate battery temperature parameters which have time memory and have gentle change trend due to strong correlation between the temperature and the time.

In view of the above problems, in the embodiment of the present application, vehicle state data at least two adjacent moments including the current moment are obtained, and the obtained vehicle state data are input into a trained battery temperature parameter estimation model, so that a battery predicted temperature parameter at the current moment can be obtained, and further, based on the battery predicted temperature parameter and a battery actual temperature parameter, whether the battery is abnormal or not can be determined. By predicting the temperature parameter by using the vehicle state data of a plurality of adjacent moments which can affect the temperature of the battery, the accuracy of predicting the temperature parameter can be improved, and further, the reliability of judging whether the battery is abnormal can be improved.

For the convenience of understanding of the present application, the technical solutions provided in the present application will be described in detail below with reference to specific embodiments.

Referring to fig. 1, the device for executing the method for determining the battery abnormality may be a vehicle terminal, or may be a cloud platform for performing information transmission with a plurality of vehicle terminals. The method for determining battery abnormality provided in the embodiment of the present application is described below from the perspective that the execution subject is a cloud platform. The flow chart of the method for determining the battery abnormity provided by the embodiment of the application comprises the following steps:

s101: acquiring vehicle state data of at least two adjacent moments including a current moment; the vehicle state data is data that affects the battery temperature.

In an implementation, vehicle state data is obtained for at least two adjacent times of the target vehicle, where the at least two adjacent times include a current time. The vehicle state data is data that affects the battery temperature.

Here, the adjacent time refers to two adjacent times corresponding to the vehicle state data, and generally, the vehicle state data of the current time of the target vehicle may be collected in real time, and the vehicle state data corresponding to a time before the current time may be generally recorded and stored, so that when determining whether the battery is abnormal, the vehicle state data of at least two adjacent times including the current time may be directly obtained.

Here, the vehicle state data includes at least two of the following data: ambient temperature, driving data, real-time performance data of the battery. Wherein the driving data includes, but is not limited to, driving mileage and driving speed; the real-time performance data of the battery includes, but is not limited to, the real-time current of the battery, the real-time voltage of the battery, and the real-time state of charge of the battery. The state of charge (SOC) is a ratio of a remaining capacity of the battery to a rated capacity under the same condition at a certain discharge rate.

S102: and inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery temperature parameter estimation model to obtain the battery predicted temperature parameter of the current moment.

In specific implementation, the acquired vehicle state data of at least two adjacent moments are input into a trained battery temperature parameter estimation model, specifically, the acquired vehicle state data of at least two adjacent moments can be sequentially input into the trained battery temperature parameter estimation model according to the time sequence, a time label can also be added to the vehicle state data corresponding to each moment, and then the vehicle state data of at least two adjacent moments are input into the trained battery temperature parameter estimation model together, so that the battery predicted temperature parameter at the current moment can be obtained.

It should be noted that the temperature parameters of the battery at different times are not discrete and disordered but related to the temperature parameters at the historical times, so that the accuracy of predicting the temperature parameters can be improved by predicting the temperature parameters through the acquired vehicle state data of at least two adjacent times including the current time.

Here, the battery temperature parameter estimation model is a Recurrent Neural Network (RNN), which is a deep learning model that takes sequence data as input to perform modeling, and from the Network structure, the Recurrent Neural Network memorizes previous information and affects output of a following node using the previous information.

The temperature parameters comprise battery temperature range and battery temperature, and the battery temperature parameter estimation model comprises a battery temperature estimation model and a battery temperature range estimation model, so that the acquired vehicle state data of the at least two adjacent moments are input into the trained battery temperature estimation model, and the battery predicted temperature at the current moment can be obtained; and inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery temperature range estimation model to obtain the battery predicted temperature range of the current moment.

It should be noted that, as the difference between the battery usage, the battery life and the battery capacity significantly increases due to the difference between the production process, the internal material, the external conditions, and the like of the battery cell state, the battery inconsistency seriously affects the battery life and safety, and therefore, the battery inconsistency needs to be monitored and predicted in real time. The battery inconsistency is represented as temperature inconsistency, capacity inconsistency and voltage inconsistency, in the related technology, the battery temperature consistency diagnosis is mostly carried out by monitoring the battery temperature in real time, calculating the highest and lowest difference values of the battery and diagnosing whether the difference values are normal or not, the battery temperature consistency is diagnosed based on the battery real-time temperature extreme value difference, the judgment is carried out only according to real-time information, the diagnosis means is limited, and the diagnosis method is mechanical (whether the current real-time temperature difference value is in a reasonable range or not is diagnosed according to the battery state); there are also some methods such as support vector machine, random forest and so on and standard neural network to estimate the battery temperature by machine learning, because the samples selected by these algorithms are relatively independent, because the battery temperature is time-effective, and the change of the battery temperature is related to the temperature value and working condition at the historical moment, the temperature value at different moments is not discrete and unordered, but depends on the historical temperature change. When the temperature consistency is diagnosed, the effect is not ideal if the temperature consistency is estimated only by using an independent sample. The standard neural network algorithm cannot use historical temperature values for calculation, and model training samples also take isolated samples as input of each training, so that the consistency of the battery temperature cannot be accurately estimated and diagnosed. In contrast, in the vehicle running process or the vehicle charging process, the method obtains the vehicle networking data, arranges the vehicle networking data (comprising the environmental temperature, the vehicle running mileage, the vehicle running speed, the battery real-time current, the battery real-time voltage and the battery real-time charge state) in sequence according to the time sequence to obtain a time sequence, brings the time sequence into a trained battery temperature range estimation model to obtain the predicted battery temperature range, compares the predicted battery temperature range with the actual battery temperature range in the vehicle networking data, and can judge whether the current battery temperature range of the battery is abnormal or not by comparing the deviation of the two, and then the consistency of the battery temperature is diagnosed, so that the accuracy of judging the consistency of the battery temperature can be improved, and if the judgment is abnormal, an owner and a maintenance platform are reminded to pay attention to the vehicle and abnormal battery cores are checked in time.

Here, the battery actual temperature is very different by a difference between the highest actual temperature of the battery, which is the highest temperature among the temperatures collected at all sampling points of the battery, and the lowest actual temperature, which is the lowest temperature among the temperatures collected at all sampling points of the battery. A plurality of sampling points are arranged on one battery, and generally, the temperatures collected at different sampling points are different.

It should be further noted that, in the related art, most methods for estimating the battery temperature are battery thermal models and thermal models that simulate the battery heat transfer process, and model building and simulation need to be performed according to parameters such as thermal melting and thermal resistance of internal materials of the battery, so that the requirements on the accuracy of the parameters such as the thermal melting and thermal resistance of the internal materials of the battery are high, and the methods are mostly used for battery simulation of battery manufacturers who master the internal parameters of the battery. The temperature attribute of the battery is different from the current and the voltage, the current and the voltage of the battery are related to the load size in real time, the reaction real-time performance is strong, data samples at different moments are independent, and the independent samples at each moment can be input through algorithms such as machine learning or neural networks for model training and prediction, the temperature has time memory attribute due to the fact that the change trend is gentle, the response is asynchronous along with the change of real-time working conditions and load, the temperature slowly changes along with the lapse of time, and parameters such as the current and the voltage cannot be calculated by using a machine learning method for training multiple groups of independent samples. The change of the battery temperature is related to the temperature value at the historical moment and the working condition, and the temperature values at different moments are not discrete and disordered but depend on the historical temperature value, so that the historical temperature value cannot be used by machine learning and standard neural network algorithm calculation, and the error precision is further influenced. In contrast, in the vehicle running process or the vehicle charging process, vehicle networking data are obtained, vehicle networking data (including ambient temperature, vehicle running mileage, vehicle running speed, battery real-time current, battery real-time voltage and battery real-time charge state) are sequentially arranged according to time sequence to obtain a time sequence, the time sequence is brought into a trained battery temperature estimation model to obtain a battery predicted temperature, the battery predicted temperature is compared with the actual temperature of the battery in the vehicle networking data, and whether the battery temperature is abnormal or not can be judged by comparing the deviation of the battery predicted temperature and the actual temperature of the battery in the vehicle networking data, so that the accuracy of the battery predicted temperature can be improved, and if the battery is abnormal, a vehicle owner and a maintenance platform are reminded to pay attention to the vehicle and faults are timely checked.

Further, the battery temperature parameter pre-estimation model consists of at least two circulation layers which are connected in sequence; in step S102, inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery temperature parameter estimation model to obtain the battery predicted temperature parameter of the current moment, including the following steps:

step a 1: inputting the acquired vehicle state data at the first moment into the first circulation layer to obtain first intermediate state data; the first time instant is the earliest time instant of the at least two adjacent time instants.

Step a 2: inputting the first intermediate state data and the vehicle state data at a second moment into a next circulation layer to obtain second intermediate state data; the second time is the next time of the first time.

Step a 3: if the vehicle state data at more than two moments are obtained, taking the second intermediate state data as the first intermediate state data, and repeatedly executing the steps to input the first intermediate state data and the vehicle state data at a third moment into a next circulation layer to obtain the second intermediate state data; the third moment is the next moment of the second moment; and if the vehicle state data at two moments are only obtained, outputting the battery predicted temperature parameter at the current moment.

In a specific implementation, the number of the acquired time points in the vehicle state data at the adjacent time points is consistent with the number of the circulation layers in the battery temperature parameter estimation model, for example, if the number of the circulation layers in the battery temperature parameter estimation model is 3, the vehicle state data at 3 adjacent time points are acquired, wherein the battery temperature parameter estimation model is composed of at least two circulation layers which are sequentially connected.

Here, taking the number of the circulation layers as 3 as an example, the explanation is made on the battery predicted temperature parameter obtained at the current moment through the battery temperature parameter estimation model:

inputting the acquired vehicle state data at the first moment into a first circulation layer to obtain first intermediate state data; the first time is the earliest of the three adjacent times; inputting the first intermediate state data and the vehicle state data at the second moment into a second circulation layer to obtain second intermediate state data; the second moment is the next moment of the first moment; and inputting the second intermediate state data and the vehicle state data at the third moment (current moment) into a third circulation layer to obtain the battery predicted temperature parameter at the current moment.

The input of each intermediate cycle level is the intermediate state data output by the previous cycle level and the vehicle state data at the time corresponding to the cycle level of the level, and the output is the intermediate state data corresponding to the cycle level of the level. The input of the first circulation layer is only the vehicle state data corresponding to the earliest moment in the acquired at least two adjacent moments, and the output of the last circulation layer is the battery predicted temperature parameter corresponding to the current moment.

Further, training the battery temperature parameter estimation model according to the following steps:

step b 1: and acquiring the vehicle state data corresponding to a plurality of moments of a sample vehicle, and taking the vehicle state data of the sample vehicle as sample data.

Step b 2: adding a sample label for each sample data; the sample label is a battery standard temperature parameter corresponding to each moment.

Step b 3: and training the initial temperature parameter estimation model according to the preset time steps, the sample data and the sample label corresponding to each sample data to obtain the battery temperature parameter estimation model.

In the concrete implementation, vehicle state data corresponding to a plurality of moments of a sample vehicle are obtained through internet of vehicles data and charging pile data to serve as sample data, wherein the sample data comprises sample environment temperature, sample vehicle driving mileage, sample vehicle driving speed, sample battery real-time current, sample battery real-time voltage and sample battery real-time charge state, a sample label is added to each sample data, the sample labels are battery standard temperature parameters of corresponding moments, the sample data are sequentially arranged according to time sequence to obtain time sequence samples, the time sequence samples serve as training data, an initial temperature parameter estimation model is trained according to the training data, preset time steps and the sample label corresponding to each sample data to obtain a battery temperature parameter estimation model, and the set time steps are consistent with the number of layers of a circulation layer of the set battery temperature parameter estimation model, the number of time steps is preferably 3. The battery temperature parameter estimation model comprises a battery temperature estimation model and a battery temperature range estimation model, and for the battery temperature estimation model, a sample label is a battery standard temperature range corresponding to each moment, namely a difference value between the highest temperature and the lowest temperature of the battery at the moment; for the battery temperature range estimation model, a sample label is a battery standard temperature range corresponding to each moment, namely a difference value between the highest temperature and the lowest temperature of the battery at the moment; for the battery temperature estimation model, the sample label is the standard temperature of the battery corresponding to each moment, namely the actual temperature of the battery at the moment.

In one example, 1000 sample data are provided, the time step number is 3, the vehicle state data includes 6 dimensional data of a sample environment temperature, a sample vehicle mileage, a sample vehicle running speed, a sample battery real-time current, a sample battery real-time voltage, and a sample battery real-time state of charge, and the characteristic data is characterized as x.size ═ (x _ feature _ num, m, step) ═ 6, 1000, 3, and the initial temperature parameter estimation model can be trained through 1000 sample data to obtain a battery temperature parameter estimation model, where x _ feature _ num represents the data dimension of the vehicle state data, 1000 represents the number of the sample data, and step represents the time step.

Further, in the step b3, training the initial temperature parameter estimation model according to the preset time step number, the sample data and the sample label corresponding to each sample data to obtain the battery temperature parameter estimation model, which includes the following steps:

step b 31: and determining the entropy loss between the standard temperature parameter and the output parameter of the battery corresponding to each model training.

Step b 32: and adjusting the model parameters of the initial temperature parameter estimation model according to the entropy loss.

Step b 33: if the entropy loss is larger than a first preset threshold value, continuing to train the initial temperature parameter estimation model; and if the entropy loss is smaller than or equal to the first preset threshold value, stopping training the initial temperature parameter estimation model to obtain the battery temperature parameter estimation model.

In the specific implementation, in the process of training the initial temperature parameter estimation model, the entropy loss between the battery standard temperature parameter and the output parameter corresponding to each model training is determined, the model parameter of the initial temperature parameter estimation model is adjusted according to the obtained entropy loss each time, the entropy loss is compared with a first preset threshold value each time, if the entropy loss is larger than the first preset threshold value, the initial temperature parameter estimation model continues to be trained, and if the entropy loss is smaller than or equal to the first preset threshold value, the training of the initial temperature parameter estimation model is stopped, so that the battery temperature parameter estimation model is obtained. The first preset threshold value can be set according to actual precision requirements.

S103: and determining whether the battery is abnormal or not based on the battery predicted temperature parameter and the battery actual temperature parameter.

In particular implementation, after the predicted battery temperature parameter is determined through the predicted battery temperature parameter estimation model, whether the battery of the vehicle is abnormal or not can be determined according to the difference between the predicted battery temperature parameter and the actual battery temperature parameter. Specifically, it is possible to determine whether an abnormality occurs in the battery of the vehicle based on a difference between the predicted temperature of the battery and the actual temperature of the battery; it is also possible to determine whether an abnormality occurs in the battery of the vehicle based on a difference between the predicted temperature pole difference of the battery and the actual temperature pole difference of the battery.

It should be noted that the battery data is arranged into a time sequence which is arranged in sequence according to the time sequence instead of a general machine learning method, the battery data is an independent sample (irrelevant to time), a reasonable temperature value change of the battery under a normal working condition can be calculated by building a battery temperature parameter estimation model, if the battery is abnormal or a temperature sensor fails, and a calculated value of the battery temperature parameter estimation model has a large deviation with a battery temperature parameter in the vehicle network data, the vehicle battery can be judged to be abnormal, and therefore the accuracy of determining whether the battery is abnormal or not can be improved.

Further, in step S103, determining whether the battery is abnormal based on the predicted battery temperature parameter and the actual battery temperature parameter includes the following two ways:

the first method is as follows: judging whether the difference value between the battery predicted temperature parameter and the battery actual temperature parameter is larger than a second preset threshold value or not; if so, determining that the battery is abnormal; if not, determining that the battery is not abnormal.

In specific implementation, a second preset threshold value can be preset according to the battery characteristics, and after the battery predicted temperature parameter at the current moment is obtained through the battery temperature parameter estimation model, whether the difference value between the battery predicted temperature parameter and the actual battery temperature parameter is larger than the second preset threshold value or not is judged, and if yes, the battery is determined to be abnormal; if not, determining that the battery is not abnormal.

The second method comprises the following steps: counting the occurrence times of the difference value between the battery predicted temperature parameter and the battery actual temperature parameter corresponding to the previous time which is greater than a third preset threshold value; and if the occurrence frequency is greater than or equal to the preset frequency, determining that the battery is abnormal.

In specific implementation, in order to avoid that the battery is subjected to single-pass battery prediction temperature parameters and battery actual temperature parameters and determine whether the battery is abnormal or not, the number of times that the difference value between the battery prediction temperature parameters and the battery actual temperature parameters corresponding to the previous time is greater than a third preset threshold value is counted, and if the number of times that the difference value is greater than or equal to the preset number of times, the battery is determined to be abnormal, so that the accuracy of judging whether the battery is abnormal or not can be further improved. The third preset threshold may be set according to the battery characteristics, the third preset threshold may be the same as the second preset threshold, and the third preset threshold may also be smaller than the second preset threshold; the preset times can be set according to actual needs.

Further, after determining that the battery is abnormal, the method further comprises the following steps:

determining an abnormal grade according to a difference value between the battery predicted temperature parameter and the battery actual temperature parameter; determining a disposal strategy for processing the battery according to the abnormity level.

In a specific implementation, the abnormality level of the battery may be determined according to a difference between the predicted temperature parameter of the battery and the actual temperature parameter of the battery, where the larger the difference is, the higher the abnormality level is, which indicates that the battery is more serious in the case of abnormality, and for the abnormality levels of different batteries, different handling strategies need to be adopted for processing.

In one example, a1 ℃ < difference of 3 ℃ for 30 seconds, a class 1 fault is diagnosed (typically severe, requiring as soon as possible to go to service); 3 < difference of 5 ℃ for 30S, then a grade 2 fault is diagnosed (severe, need to stop immediately, contact 4S shop maintenance).

In the embodiment of the application, the vehicle state data of at least two adjacent moments including the current moment are obtained, the obtained vehicle state data are input into a trained battery temperature parameter estimation model according to the time sequence, the battery predicted temperature parameter of the current moment can be obtained, and then whether the battery is abnormal or not can be determined based on the battery predicted temperature parameter and the actual battery temperature parameter. By predicting the temperature parameter by using the vehicle state data of a plurality of adjacent moments which can affect the temperature of the battery, the accuracy of predicting the temperature parameter can be improved, and further, the reliability of judging whether the battery is abnormal can be improved.

Referring to fig. 2, the device for executing the method for determining the remaining battery charging time may be a vehicle terminal, or may be a cloud platform for performing information transmission with a plurality of vehicle terminals. The method for determining the remaining battery charging time provided in the embodiment of the present application is described below from the perspective that the execution subject is a cloud platform. The flow chart of the method for determining the remaining battery charging time provided by the embodiment of the application comprises the following steps:

s201: acquiring vehicle state data of at least two adjacent moments including a current moment; the vehicle state data is data that affects the battery charging time.

In an implementation, vehicle state data is obtained for at least two adjacent times of the target vehicle, where the at least two adjacent times include a current time. The vehicle state data is data that affects the battery charging time.

Here, the adjacent time refers to two adjacent times corresponding to the vehicle state data, and generally, the vehicle state data of the current time of the target vehicle may be collected in real time, and the vehicle state data corresponding to a time before the current time may be generally recorded and stored, so that the vehicle state data of at least two adjacent times including the current time may be directly obtained when the remaining battery charging time is determined.

Here, the vehicle state data includes: battery temperature, real-time performance data of the battery. The real-time performance data of the battery includes, but is not limited to, a real-time current of the battery, a real-time voltage of the battery, and a real-time state of charge of the battery. The state of charge (SOC) is a ratio of a remaining capacity of the battery to a rated capacity under the same condition at a certain discharge rate.

S202: and inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery charging remaining time estimation model to obtain the battery charging remaining time.

In specific implementation, the acquired vehicle state data of at least two adjacent moments are input into a trained battery charge remaining time estimation model, specifically, the acquired vehicle state data of at least two adjacent moments can be sequentially input into the trained battery charge remaining time estimation model according to the time sequence, a time label can also be added to the vehicle state data corresponding to each moment, and then the vehicle state data of at least two adjacent moments are input into the trained battery charge remaining time estimation model together, so that the battery charge remaining time of the current moment can be obtained.

It should be noted that, for the network car booking driver and the dispatching platform in the travel operation scene, the estimation of the battery charging remaining time is particularly important, the calculation precision is high, the reasonable time arrangement of the driver is facilitated, the dispatching platform is assisted to reasonably dispatch the order according to the transportation capacity rule and the like, the shorter the charging remaining time is, more time is saved for the private car owner and the network car booking driver, the higher the estimation precision of the charging remaining time is, and the user can reasonably plan the charging time of the user. For the network car booking business management platform, the charging remaining time of each vehicle type can be used as an important reference when a network car booking company purchases a vehicle. In the related art, the charging remaining time is calculated based on the remaining capacity and the current of the battery, and the method has low precision due to factors such as unstable charging current, attenuation of the charging remaining capacity along with the battery, and the attenuation of the capacity of the battery. Because the data samples selected by the algorithms are relatively independent and can be randomly disordered to train the model, and because the charging remaining time is strongly related to the time, any abnormal current and voltage jitter in the charging process can affect the calculation precision, the machine learning method cannot effectively estimate the charging remaining time with time memory and gentle change trend, the standard neural network algorithm cannot calculate and use the historical value, and the model training samples also take the isolated samples as the input of each training and cannot accurately estimate the charging remaining time. According to the method, the vehicle state data of at least two adjacent moments including the current moment are obtained, wherein the vehicle state data are data influencing the battery charging time, the obtained vehicle state data of the at least two adjacent moments are input into a trained battery charging remaining time estimation model, the battery charging remaining time of the target vehicle is obtained, and compared with the method for calculating the battery charging remaining time based on the battery remaining capacity and the current in the related art, the accuracy of predicting the battery charging remaining time can be improved due to the fact that charging current is unstable, the charging remaining capacity is attenuated along with the battery, the electric quantity is also attenuated along with the battery and the like, and the accuracy is low.

Here, the model for estimating the remaining battery charging time is a Recurrent Neural Network (RNN), which is a deep learning model that takes sequence data as input to perform modeling, and from the Network structure, the Recurrent Neural Network memorizes previous information and affects output of a following node using the previous information.

Further, the battery charging remaining time estimation model is composed of at least two circulation layers which are connected in sequence; in step S202, inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery charging remaining time estimation model to obtain the battery charging remaining time of the current moment, including the following steps:

step c 1: inputting the acquired vehicle state data at the first moment into the first circulation layer to obtain first intermediate state data; the first time instant is the earliest time instant of the at least two adjacent time instants.

Step c 2: inputting the first intermediate state data and the vehicle state data at a second moment into a next circulation layer to obtain second intermediate state data; the second time is the next time of the first time.

Step c 3: if the vehicle state data at more than two moments are obtained, taking the second intermediate state data as the first intermediate state data, and repeatedly executing the steps to input the first intermediate state data and the vehicle state data at a third moment into a next circulation layer to obtain the second intermediate state data; the third moment is the next moment of the second moment; and if the vehicle state data at two moments are only acquired, outputting the battery charging remaining time at the current moment.

In a specific implementation, the number of the acquired time points in the vehicle state data at the adjacent time points is consistent with the number of the circulation layers in the battery charge remaining time estimation model, for example, if the number of the circulation layers in the battery charge remaining time estimation model is 3, the vehicle state data at the 3 adjacent time points are acquired, wherein the battery charge remaining time estimation model is composed of at least two circulation layers which are sequentially connected.

Here, taking the number of the cycle layers as 3 as an example, the battery charge remaining time at the current time obtained by the battery charge remaining time estimation model is explained as follows:

inputting the acquired vehicle state data at the first moment into a first circulation layer to obtain first intermediate state data; the first time is the earliest of the three adjacent times; inputting the first intermediate state data and the vehicle state data at the second moment into a second circulation layer to obtain second intermediate state data; the second moment is the next moment of the first moment; and inputting the second intermediate state data and the vehicle state data at the third moment (current moment) into a third circulation layer to obtain the battery charging remaining time at the current moment.

The input of each intermediate cycle level is the intermediate state data output by the previous cycle level and the vehicle state data at the time corresponding to the cycle level of the level, and the output is the intermediate state data corresponding to the cycle level of the level. The input of the first circulation layer is only the vehicle state data corresponding to the earliest moment in the acquired at least two adjacent moments, and the output of the last circulation layer is the battery charging remaining time corresponding to the current moment.

Further, training the battery charge remaining time estimation model according to the following steps:

obtaining the vehicle state data corresponding to a plurality of moments of a sample vehicle, and taking the vehicle state data of the sample vehicle as sample data; adding a sample label for each sample data; the sample label is the standard charging remaining time of the battery corresponding to each moment; and training the initial battery charging remaining time estimation model according to the preset time step number, the sample data and the sample label corresponding to each sample data to obtain the battery charging remaining time estimation model.

In the specific implementation, vehicle state data corresponding to a plurality of moments of a sample vehicle are obtained through internet of vehicles data and charging pile data and serve as sample data, wherein the sample data comprises sample battery temperature, sample battery real-time current, sample battery real-time voltage and sample battery real-time charge state, a sample label is added to each sample data, the sample labels are battery standard charging remaining time of corresponding moments, the sample data are sequentially arranged according to time sequence to obtain time sequence samples, the time sequence samples serve as training data, an initial charging remaining time estimation model is trained according to the training data, preset time steps and the sample label corresponding to each sample data to obtain a battery charging remaining time estimation model, and the set time steps are consistent with the number of the circulating layers of the set battery charging remaining time estimation model, the number of time steps is preferably 3.

Further, the training of the initial charge remaining time estimation model according to the preset time steps, the sample data and the sample label corresponding to each sample data to obtain the battery charge remaining time estimation model comprises the following steps:

adjusting model parameters of the initial charging remaining time estimation model according to the entropy loss; if the entropy loss is larger than a first preset threshold value, continuing to train the initial charging remaining time estimation model; and if the entropy loss is smaller than or equal to the first preset threshold, stopping training the initial charging remaining time estimation model to obtain the battery charging remaining time estimation model.

In the specific implementation, in the process of training the initial charging remaining time estimation model, the entropy loss between the standard charging remaining time and the output time of the battery corresponding to each model training is determined, model parameters of the charging remaining time estimation model are adjusted according to the obtained entropy loss each time, the entropy loss is compared with a first preset threshold value each time, if the entropy loss is larger than the first preset threshold value, the initial charging remaining time estimation model continues to be trained, and if the entropy loss is smaller than or equal to the first preset threshold value, the training of the initial charging remaining time estimation model is stopped, so that the battery charging remaining time estimation model is obtained. The first preset threshold value can be set according to actual precision requirements.

Based on the same application concept, a device for determining battery abnormality corresponding to the method for determining battery abnormality provided in the foregoing embodiment is also provided in the embodiments of the present application.

Fig. 3 is a functional block diagram of a battery abnormality determination apparatus 300 according to an embodiment of the present application; fig. 4 is a functional block diagram illustrating the first input block 320 of the apparatus 300 for determining abnormality of a battery in fig. 3; fig. 5 shows a second functional block diagram of a battery abnormality determination apparatus 300 according to an embodiment of the present application.

As shown in fig. 3, the battery abnormality determination apparatus 300 includes: a first obtaining module 310, configured to obtain vehicle state data of at least two adjacent moments including a current moment; the vehicle state data is data that affects a battery temperature;

the first input module 320 is configured to input the acquired vehicle state data of the at least two adjacent moments into a trained battery temperature parameter estimation model to obtain a battery predicted temperature parameter of the current moment;

the first determining module 330 is configured to determine whether the battery is abnormal based on the predicted battery temperature parameter and the actual battery temperature parameter.

ambient temperature, driving data, real-time performance data of the battery.

In one possible embodiment, as shown in fig. 4, the battery temperature parameter pre-estimation model is composed of at least two cycle layers connected in sequence; the first input module 320 includes:

a first input unit 321, configured to input the acquired vehicle state data at the first time into the first circulation layer to obtain first intermediate state data; the first time is the earliest of the at least two adjacent times;

a second input unit 322, configured to input the first intermediate state data and the vehicle state data at a second time into a next cycle layer to obtain second intermediate state data; the second moment is the next moment of the first moment;

a third input unit 323, configured to, if vehicle state data at more than two times are obtained, take the second intermediate state data as the first intermediate state data, and repeatedly perform the step of inputting the first intermediate state data and the vehicle state data at a third time into a next cycle layer, so as to obtain the second intermediate state data; the third moment is the next moment of the second moment;

a first output unit 324, configured to output the predicted battery temperature parameter at the current time if the vehicle state data at only two times are acquired.

In one possible embodiment, as shown in fig. 3, the first input module 320 is configured to train the battery temperature parameter estimation model according to the following steps:

In a possible implementation manner, as shown in fig. 3, the first input module 320 is configured to train an initial temperature parameter estimation model to obtain the battery temperature parameter estimation model according to the following steps:

In one possible implementation, as shown in fig. 3, the first determining module 330 is configured to determine whether the battery is abnormal according to the following steps:

if so, determining that the battery is abnormal;

if not, determining that the battery is not abnormal.

In one possible embodiment, as shown in fig. 5, after determining that the battery is abnormal, the apparatus 300 for determining battery abnormality further includes a second determining module 340; the second determining module 340 is configured to:

Fig. 6 is a functional block diagram of a device 600 for determining the remaining battery charge time according to an embodiment of the present disclosure; fig. 7 is a functional block diagram of the second input block 620 of the apparatus 600 for determining the remaining battery charge time in fig. 6.

As shown in fig. 6, the device 600 for determining the remaining battery charging time includes:

a second obtaining module 610, configured to obtain vehicle state data of at least two adjacent moments including a current moment; the vehicle state data is data that affects a battery charging time;

and the second input module 620 is configured to input the acquired vehicle state data of the at least two adjacent moments into the trained battery charging remaining time estimation model to obtain the battery charging remaining time.

In one possible embodiment, the vehicle state data includes:

battery temperature, said battery real-time performance data.

In one possible embodiment, as shown in fig. 7, the estimation model of the remaining battery charge time is composed of at least two cycle layers connected in sequence; the second input module 620 includes:

a fourth input unit 621, configured to input the acquired vehicle state data at the first time into the first loop layer to obtain first intermediate state data; the first time is the earliest of the at least two adjacent times;

a fifth input unit 622, configured to input the first intermediate state data and the vehicle state data at the second time into the next circulation layer, so as to obtain second intermediate state data; the second moment is the next moment of the first moment;

a sixth input unit 623, configured to, if vehicle state data at more than two times are obtained, take the second intermediate state data as the first intermediate state data, and repeatedly perform the steps to input the first intermediate state data and the vehicle state data at a third time into a next cycle layer, so as to obtain the second intermediate state data; the third moment is the next moment of the second moment;

and a second output unit 624, configured to, if the vehicle state data at two times are only obtained, output the remaining battery charging time at the current time.

In one possible implementation, as shown in fig. 6, the second input module 620 is configured to train the battery charge remaining time estimation model according to the following steps:

In a possible implementation manner, as shown in fig. 6, the second input module 620 is configured to train an initial charge remaining time estimation model to obtain the battery charge remaining time estimation model according to the following steps:

In the embodiment of the application, the vehicle state data of at least two adjacent moments including the current moment are obtained, wherein the vehicle state data are data influencing the battery charging time, the obtained vehicle state data of the at least two adjacent moments are input into a trained battery charging remaining time estimation model to obtain the battery charging remaining time of the target vehicle, and compared with the method for calculating the battery charging remaining time based on the battery remaining capacity and the current in the prior art, the method has the advantages that the accuracy of predicting the battery charging remaining time is low due to the fact that the charging current is unstable, the charging remaining capacity is attenuated along with the battery, the electric quantity is also attenuated along with the battery and the like, and the accuracy is low.

Based on the same application concept, referring to fig. 8, a schematic structural diagram of an electronic device 800 provided in the embodiment of the present application includes: a processor 810, a memory 820 and a bus 830, wherein the memory 820 stores machine-readable instructions executable by the processor 810, the processor 810 and the memory 820 communicate via the bus 830 when the electronic device 800 is operating, and the machine-readable instructions are executed by the processor 810 to perform the steps of the method for determining battery abnormality as in any one of the above embodiments and/or to perform the steps of the method for determining battery charge remaining time as in any one of the above embodiments.

In particular, the machine readable instructions, when executed by the processor 810, may perform the following:

Based on the same application concept, embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to perform the steps of the method for determining battery abnormality and/or the steps of the method for determining remaining battery charging time provided by the above embodiments.

Specifically, the storage medium may be a general-purpose storage medium, such as a removable disk, a hard disk, or the like, and when the computer program on the storage medium is executed, the method for determining the abnormality of the battery may be executed, and by predicting the temperature parameter by using the vehicle state data at a plurality of adjacent times that may affect the temperature of the battery, the accuracy of predicting the temperature parameter may be improved, and further, the reliability of determining whether the battery is abnormal may be improved.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method of determining battery abnormality, characterized by comprising:

2. The determination method according to claim 1, wherein the temperature parameters include a battery temperature pole difference and a battery temperature; the battery actual temperature range is a difference value between the highest actual temperature and the lowest actual temperature of the battery, wherein the highest actual temperature is the highest temperature of the temperatures acquired at all the sampling points of the battery, and the lowest actual temperature is the lowest temperature of the temperatures acquired at all the sampling points of the battery.

3. The determination method according to claim 1, characterized in that the vehicle state data includes at least two of the following data:

ambient temperature, driving data, real-time performance data of the battery.

4. The determination method according to claim 1, wherein the battery temperature parameter pre-estimation model is composed of at least two cycle layers connected in sequence; the step of inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery temperature parameter estimation model to obtain the battery predicted temperature parameter of the current moment comprises the following steps:

5. The method of claim 1, wherein the battery temperature parameter estimation model is trained according to the following steps:

6. The method according to claim 5, wherein the training an initial temperature parameter estimation model according to a preset number of time steps, the sample data, and a sample label corresponding to each sample data to obtain the battery temperature parameter estimation model comprises:

7. The determination method according to claim 1, wherein the determining whether the battery is abnormal based on the predicted battery temperature parameter and the actual battery temperature parameter includes:

if so, determining that the battery is abnormal;

if not, determining that the battery is not abnormal.

8. The determination method according to claim 1, wherein the determining whether the battery is abnormal based on the predicted battery temperature parameter and the actual battery temperature parameter includes:

9. The determination method according to claim 1, wherein after determining that the abnormality occurs in the battery, the determination method further comprises:

10. A method for determining a remaining battery charge time, the method comprising:

11. The determination method according to claim 10, characterized in that the vehicle state data includes:

battery temperature, said battery real-time performance data.

12. The method of claim 10, wherein the model of estimating the remaining battery charge time is composed of at least two cycle layers connected in sequence; the step of inputting the acquired vehicle state data of the at least two adjacent moments into a trained battery charging remaining time estimation model to obtain the battery charging remaining time of the current moment comprises the following steps:

13. The method of claim 10, wherein the battery charge remaining time estimation model is trained according to the following steps:

14. The method according to claim 13, wherein the training an initial charge remaining time estimation model according to a preset number of time steps, the sample data, and a sample tag corresponding to each sample data to obtain the battery charge remaining time estimation model comprises:

15. A determination device of a battery abnormality, characterized by comprising:

16. The determination device of claim 15, wherein the temperature parameters include a battery temperature range and a battery temperature; the battery actual temperature range is a difference value between the highest actual temperature and the lowest actual temperature of the battery, wherein the highest actual temperature is the highest temperature of the temperatures acquired at all the sampling points of the battery, and the lowest actual temperature is the lowest temperature of the temperatures acquired at all the sampling points of the battery.

17. The determination device according to claim 15, characterized in that the vehicle state data comprises at least two of the following data:

ambient temperature, driving data, real-time performance data of the battery.

18. The apparatus according to claim 15, wherein the pre-estimation model of battery temperature parameters is composed of at least two cycle layers connected in sequence; the first input module includes:

the first input unit is used for inputting the acquired vehicle state data at the first moment into the first circulation layer to obtain first intermediate state data; the first time is the earliest of the at least two adjacent times;

the second input unit is used for inputting the first intermediate state data and the vehicle state data at a second moment into the next circulation layer to obtain second intermediate state data; the second moment is the next moment of the first moment;

a third input unit, configured to, if vehicle state data at more than two times are acquired, use the second intermediate state data as the first intermediate state data, and repeatedly perform the step of inputting the first intermediate state data and the vehicle state data at a third time into a next cycle layer to obtain the second intermediate state data; the third moment is the next moment of the second moment;

and the first output unit is used for outputting the battery predicted temperature parameter at the current moment if the vehicle state data at two moments are only acquired.

19. The apparatus of claim 15, wherein the first input module is configured to train the battery temperature parameter estimation model according to the following steps:

20. The apparatus of claim 19, wherein the first input module is configured to train an initial temperature parameter estimation model to obtain the battery temperature parameter estimation model according to the following steps:

21. The apparatus of claim 15, wherein the first determining module is configured to determine whether the battery is abnormal according to the following steps:

if so, determining that the battery is abnormal;

if not, determining that the battery is not abnormal.

22. The apparatus of claim 15, wherein the first determining module is configured to determine whether the battery is abnormal according to the following steps:

23. The apparatus according to claim 15, wherein after determining that the abnormality occurs in the battery, the apparatus further comprises a second determination module; the second determination module is to:

24. A device for determining a remaining time of battery charge, the device comprising:

25. The determination device of claim 24, wherein the vehicle state data comprises:

battery temperature, said battery real-time performance data.

26. The apparatus of claim 24, wherein the model of estimating the remaining battery charge time is comprised of at least two cyclic layers connected in sequence; the second input module includes:

the fourth input unit is used for inputting the acquired vehicle state data at the first moment into the first circulation layer to obtain first intermediate state data; the first time is the earliest of the at least two adjacent times;

a fifth input unit, configured to input the first intermediate state data and the vehicle state data at a second time into a next cycle layer to obtain second intermediate state data; the second moment is the next moment of the first moment;

a sixth input unit, configured to, if vehicle state data at more than two times are acquired, use the second intermediate state data as the first intermediate state data, and repeatedly perform the step of inputting the first intermediate state data and the vehicle state data at a third time into a next cycle layer, so as to obtain the second intermediate state data; the third moment is the next moment of the second moment;

and the second output unit is used for outputting the battery charging remaining time at the current moment if the vehicle state data at two moments are only acquired.

27. The apparatus of claim 24, wherein the second input module is configured to train the battery charge remaining time estimation model according to the following steps:

28. The apparatus of claim 27, wherein the second input module is configured to train an initial charge remaining time estimation model to obtain the battery charge remaining time estimation model according to the following steps:

29. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine readable instructions executable by the processor, the processor and the memory communicating over the bus when the electronic device is operating, the machine readable instructions when executed by the processor performing the steps of the method for determining battery abnormality according to any one of claims 1 to 9 and/or performing the steps of the method for determining battery charge remaining time according to any one of claims 10 to 14.

30. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for determining battery abnormality according to any one of claims 1 to 9 and/or the steps of the method for determining remaining battery charge time according to any one of claims 10 to 14.