Energy storage battery charging time prediction method, electronic device, and medium

Abstract

The embodiment of the present disclosure relates to the technical field of batteries, and provides a method for predicting the charging time of an energy storage battery, an electronic device and a medium, which can at least solve the problems of insufficient accuracy and low efficiency of the existing battery remaining charging time prediction model. The method comprises the following steps: according to a first data set composed of a plurality of frames of battery charging data of a first battery component in a first time period, using a trained double-depth prediction model to determine a second battery component with the highest voltage in the first battery component, a plurality of frames of predicted voltages in a second time period after the first time period, and then accurately determining the battery remaining charging time according to a first data frame greater than or equal to the charging cutoff voltage in the plurality of frames of predicted voltages.

Description

TECHNICAL FIELD

[0001] The embodiment of the present disclosure relates to the technical field of battery, in particular to a method for predicting charging time of energy storage battery, an electronic device and a medium. BACKGROUND

[0002] In the conventional method for predicting the remaining charging time of a battery, schemes based on piecewise lookup table, model and data driving are usually adopted. The scheme based on piecewise lookup table is specifically as follows: a mapping relationship among temperature section, state of charge (SOC) section and charging cumulative time of a battery system is established, and the charging time of the battery in different sections is calculated respectively, and the sum of the charging time is obtained to obtain the remaining charging time. The scheme based on model is specifically as follows: a thermal-electric coupling model of the battery is constructed, and the thermal-electric model is iteratively called from the initial state until the voltage cutoff or full charge, so as to estimate the remaining charging time. The scheme based on data driving is specifically as follows: features are extracted, and a mapping relationship between the features and the remaining charging time is established to predict the remaining charging time. SUMMARY

[0003] The embodiment of the present disclosure provides a method for predicting charging time of energy storage battery, an electronic device and a medium, which at least has advantages of solving the problems of insufficient accuracy and low efficiency of the existing battery remaining charging time prediction model.

[0004] According to some embodiments of the present disclosure, a first aspect of the embodiment of the present disclosure provides a method for predicting charging time of energy storage battery, comprising: obtaining a first data set, the first data set being a plurality of frames of battery charging data of a first battery assembly in a first time period, the first battery assembly comprising a plurality of battery components, a battery component with the highest voltage in the plurality of battery components constituting a second battery assembly, the battery component being one of: a battery pack, a battery module, or a battery cell;

[0005] processing the first data set based on a double-depth prediction model to obtain a plurality of frames of predicted voltages of the second battery assembly in a second time period, the second time period being a period of time after the first time period;

[0006] determining a first data frame, the first data frame being a data frame in which the first predicted voltage of the second battery assembly in the second time period is greater than or equal to a charging cutoff voltage;

[0007] determining a battery remaining charging time according to the first data frame;

[0008] The double-depth prediction model includes a first battery component prediction model and a second battery component prediction model. The first battery component prediction model is configured to determine, according to the first data set, a plurality of frames of predicted temperatures and a plurality of frames of predicted currents of the first battery component in the second time period. The second battery component prediction model is configured to determine, according to the plurality of frames of predicted temperatures and the plurality of frames of predicted currents of the first battery component in the second time period and a voltage of the second battery component, a plurality of frames of predicted voltages of the second battery component in the second time period.

[0009] According to some embodiments of the present disclosure, a second aspect of the embodiments of the present disclosure provides an electronic device, comprising:

[0010] at least one processor; and

[0011] a memory in communication with the at least one processor; wherein

[0012] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of the first aspect.

[0013] According to some embodiments of the present disclosure, a third aspect of the embodiments of the present disclosure provides a computer-readable storage medium, and the computer-readable storage medium stores computer instructions. When the computer instructions are executed by a processor, the method of the first aspect is implemented.

[0014] In some embodiments, the processing of the first data set based on the double-depth prediction model to obtain the plurality of frames of predicted voltages of the second battery component in the second time period comprises: inputting the first data set into the first battery component prediction model to obtain the plurality of frames of predicted temperatures and the plurality of frames of predicted currents of the first battery component in the second time period; and inputting the plurality of frames of predicted temperatures and the plurality of frames of predicted currents of the first battery component in the second time period and the voltage of the second battery component into the second battery component prediction model to obtain the plurality of frames of predicted voltages of the second battery component in the second time period.

[0015] In some embodiments, the first time period is continuous with the second time period.

[0016] In some embodiments, the battery remaining charging time is a difference between the time corresponding to the first data frame and the start time of the second time period.

[0017] In some embodiments, the frame intervals in the second time period are equal, the first data frame is the zth frame in the second time period, z is a natural number, 0≤z<b, and the battery remaining charging time satisfies: Treserved=zT; wherein Treserved is the battery remaining charging time, and T is the frame interval in the second time period.

[0018] In some embodiments, the method further comprises: obtaining a first training data set, the first training data set being a plurality of frames of battery charging data of the first battery assembly in a third time period; performing feature extraction on the first training data set to obtain a second training data set; and training the first battery assembly prediction model based on the second training data set.

[0019] In some embodiments, the first battery assembly prediction model comprises a first convolutional neural network and a first TRANSFORMER network, the first TRANSFORMER network comprising a first encoder and a first decoder; the performing feature extraction on the first training data set to obtain the second training data set comprises inputting the first training data set into the first convolutional neural network to perform feature extraction to obtain the second training data set; and the training the first battery assembly prediction model based on the second training data set comprises: dividing the third time period into a first sub-time period and a second sub-time period, and dividing the second training data set into a first training data subset and a second training data subset, the data in the first training data subset being battery charging data in the first sub-time period, the data in the second training data subset being battery charging data in the second sub-time period, the second sub-time period being after the first sub-time period; inputting the first training data subset into the first encoder, inputting a third training data set into the first decoder, and taking the second training data subset as a label to train the first battery assembly prediction model, the third training data set being a data set obtained by replacing data in the second training data subset in the second training data set with dummy data.

[0020] In some embodiments, the second sub-time period has a variable length.

[0021] In some embodiments, each frame of battery charging data in the first data set and / or the first training data set comprises at least one of: a maximum voltage of the first battery assembly, a minimum voltage of the first battery assembly, a current of the first battery assembly, a temperature of the first battery assembly, a heating power of the first battery assembly, and a cooling power of the first battery assembly.

[0022] In some embodiments, the method further comprises: obtaining a fourth training data set, the fourth training data set being a plurality of frames of battery charging data of a second battery assembly in the first battery assembly in a fourth time period, each frame of battery charging data in the fourth training data set comprising at least one of: a voltage, a current, and a temperature of the second battery assembly; performing feature extraction on the fourth training data set to obtain a fifth training data set; and training a second battery assembly prediction model based on the fifth training data set.

[0023] In some embodiments, the second battery assembly prediction model comprises a second convolutional neural network and a second TRANSFORMER network, the second TRANSFORMER network comprising a second encoder and a second decoder;

[0024] The fourth training data set is subjected to feature extraction to obtain a fifth training data set, including: inputting the fourth training data set into the second convolutional neural network for feature extraction to obtain the fifth training data set; correspondingly, training the second battery component prediction model based on the fifth training data set, including: dividing the fourth time period into a third sub-time period and a fourth sub-time period, and dividing the fifth training data set into a third training data subset and a fourth training data subset, the data in the third training data subset being battery charging data in the third sub-time period, the data in the fourth training data subset being battery charging data in the fourth sub-time period, the fourth sub-time period being after the third sub-time period; inputting the third training data subset into the second encoder, inputting the sixth training data set into the second decoder, and taking the fourth training data subset as a label, training the second battery component prediction model, the sixth training data set being a data set obtained by replacing the data in the fourth training data subset in the fifth training data set with invalid data.

[0025] In some embodiments, the length of time of the fourth sub-time period is variable.

[0026] In some embodiments, the fourth time period and the third time period are the same time period.

[0027] In some embodiments, the first battery component and the second battery component are one of the following combinations: the first battery component is a battery pack, and the second battery component is a battery module; the first battery component is a battery module, and the second battery component is a battery cell; or, the first battery component is a battery pack, and the second battery component is a battery cell.

[0028] The technical solutions provided by the embodiments of the present disclosure have at least the following advantages:

[0029] Based on the technical solutions provided by the embodiments of the present disclosure, the first data set composed of multiple frames of battery charging data of the first battery component in the first time period can be used to determine the second battery component with the highest voltage in the first battery component using the trained double-depth prediction model, and multiple frames of predicted voltages in the second time period after the first time period, and then the first data frame in which the first data frame is greater than or equal to the charging cutoff voltage in the multiple frames of predicted voltages can be used to accurately determine the battery remaining charging time, which can solve the problems of insufficient prediction accuracy and low efficiency of the existing battery remaining charging time, thereby improving the accuracy and efficiency.

[0030] In addition, the remaining charging time estimation problem can be converted into a future voltage curve prediction problem, and a variable-scale training mode can be used to solve the problems of complex control flow and low accuracy in the repeated iteration prediction scheme;

[0031] Further, a self-supervised learning and converter time series prediction large model is used to solve the prediction problem of the voltage to improve the generalization ability;

[0032] In addition, the remaining charging time prediction problem is decomposed into two non-end-to-end model problems: one is a first battery component prediction model that predicts future current and temperature from maximum voltage, minimum voltage, temperature, current, heating real-time power, cooling real-time power, etc.; the other is a second battery component prediction model that predicts future terminal voltage from cell temperature, current, and voltage, thereby converting the remaining charging time problem into a voltage prediction problem, simplifying the problem and improving prediction efficiency and accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0033] One or more embodiments are illustrated by way of example in the figures that form a part of this disclosure and which should not be construed as limiting the embodiments, unless otherwise explicitly stated. The figures in the drawings are not necessarily to scale, except if specifically noted. This disclosure will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:

[0034] Figure 1 A flowchart of a method for predicting charging time of an energy storage battery according to an embodiment of the present disclosure is shown in FIG. 1.

[0035] Figure 2 A structure diagram of a double-depth prediction model according to an embodiment of the present disclosure is shown in FIG. 2.

[0036] Figure 3 A flowchart of another method for predicting battery charging according to an embodiment of the present disclosure is shown in FIG. 3.

[0037] Figure 4 A scenario diagram for predicting remaining charging time of a battery according to an embodiment of the present disclosure is shown in FIG. 4.

[0038] Figure 5 A flowchart of a training method for a first battery component prediction model according to an embodiment of the present disclosure is shown in FIG. 5.

[0039] Figure 6 A flowchart of another training method for a first battery component prediction model according to an embodiment of the present disclosure is shown in FIG. 6.

[0040] Figure 7 An example diagram of a training data set for training a first battery component prediction model according to an embodiment of the present disclosure is shown in FIG. 7.

[0041] Figure 8 A flowchart of a training method for a second battery component prediction model according to an embodiment of the present disclosure is shown in FIG. 8.

[0042] Figure 9 A flowchart of another training method of a second battery component prediction model provided by an embodiment of the present disclosure is shown in FIG. 6.

[0043] Figure 10 An example diagram of a data set for training a second battery component prediction model provided by an embodiment of the present disclosure is shown in FIG. 7.

[0044] Figure 11 FIG. 8 is a structural diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0045] As can be known from the background, the method based on segmented lookup table cannot adapt to complex working conditions, and the accuracy of the calculation result is limited; the scheme based on the model needs to calibrate the parameters of the related thermoelectric coupling model, and once the working condition changes, the calibration needs to be re-performed, otherwise the accuracy is affected, and the generalization ability is poor; the scheme based on data driving generally directly establishes the mapping relationship between the features and the remaining charging time, and the accuracy is directly related to the number of label samples and the coverage of the working condition, and the process of constructing the label and the sample is time-consuming and resource-consuming, and the training efficiency is low, that is, the existing battery remaining charging time prediction model has the problems of insufficient accuracy and low efficiency.

[0046] To solve the problem, the present disclosure provides a method for predicting the remaining charging time of an energy storage battery, which can use a trained double-depth prediction model to determine a second battery component with the highest voltage in a first battery component according to a first data set composed of multiple frames of battery charging data of the first battery component in a first time period, and multiple frames of predicted voltages in a second time period after the first time period, and then accurately determine the battery remaining charging time according to the first data frame in which the first data frame is greater than or equal to the charging cutoff voltage in the multiple frames of predicted voltages, so as to solve the problems of insufficient accuracy and low efficiency of the existing battery remaining charging time prediction, thereby improving the accuracy and efficiency.

[0047] In the description of the embodiments of the present disclosure, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present disclosure, the meaning of "multiple" is two or more, unless otherwise specifically limited. Similarly, "multiple groups" means two or more groups (including two groups), and "multiple pieces" means two or more pieces (including two pieces).

[0048] Reference to“an embodiment” herein means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the disclosure. The appearances of the phrase“in an embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. It is expressly understood that any of the embodiments described herein can be incorporated into any other embodiments even though it can not be explicitly described within the other embodiments. It is expressly understood that combinations of the various embodiments and / or features of the embodiments can be made and still fall within the scope of the disclosure.

[0049] In the description of the embodiments of the disclosure, the term“and / or” only means an association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can mean that A exists, A and B exist at the same time, and B exists. In addition, the character“ / ” herein generally means that the front and rear associated objects are in an“or” relationship.

[0050] In the description of the embodiments of the disclosure, when a certain component“includes” another component, unless otherwise specified, other components are not excluded, and other components can also be further included. The first component is formed or provided above or on the first component, or the second component is formed or provided on the surface of the first component, or the second component is formed or provided on the side of the first component. It can include an embodiment in which the first component and the second component are in direct contact, and it can also include an embodiment in which additional components can be included between the first component and the second component, so that the first component and the second component can not be in direct contact. For the sake of simplicity and clarity, various components can be arbitrarily drawn in different proportions. In the drawings, some layers / components can be omitted for simplicity. Unless otherwise specified, the first component is formed or provided on the surface of the first component, which means that the first component is in direct contact with the second component. Among them, the“component” mentioned above can refer to a layer, a film, a region, a part, a structure, etc.

[0051] The terms used in the description of various embodiments described herein are only used to describe specific embodiments and are not intended to be limiting.

[0052] The embodiments of the disclosure will be described in detail below with reference to the accompanying drawings. However, those skilled in the art can understand that in the embodiments of the disclosure, many technical details are proposed in order to enable the reader to better understand the disclosure. However, the technical solutions claimed by the disclosure can be implemented even without these technical details and various changes and modifications based on the following embodiments.

[0053] Figure 1 A flowchart of a method for predicting the charging time of an energy storage battery according to an embodiment of the disclosure is shown in FIG. 1. As shown in the figure, the method comprises the following steps. Figure 1

[0054] S101, obtaining a first data set.

[0055] ​In some embodiments, the first data set is a plurality of frames of battery charging data of the first battery assembly in a first time period, the first battery assembly comprises a plurality of battery components, a battery component with the highest voltage among the plurality of battery components constitutes the second battery assembly, and the battery component is one of: a battery pack, a battery module, or a battery cell.

[0056] In some embodiments, the first battery assembly and the second battery assembly are both components of the energy storage battery.

[0057] In some embodiments, the second battery assembly can be a component of the first battery assembly, i.e., one first battery assembly can comprise a plurality of second battery assemblies. For example, the first battery assembly is a battery pack, the second battery assembly is a battery module, and one battery pack can comprise one or more battery modules. For another example, the first battery assembly is a battery module, the second battery assembly is a battery cell, and one battery module can comprise one or more battery cells. For yet another example, the first battery assembly is a battery pack, the second battery assembly is a battery cell, and one battery pack can comprise one or more battery cells.

[0058] In some other embodiments, the second battery assembly and the first battery assembly can also be battery components of the same type. In this case, the first battery assembly can be a set of a plurality of battery components of the same type, and the second battery assembly can be a battery component in the set that satisfies a certain condition (such as the highest voltage and / or the highest temperature, etc.), i.e., the second battery assembly can be a battery component selected from the set of the first battery assembly. For example, the first battery assembly is a set of a plurality of battery modules, such as a plurality of battery modules belonging to the same battery pack, and the second battery assembly is the battery module with the highest voltage in the set. For another example, the first battery assembly is a set of a plurality of battery cells, such as a plurality of battery cells belonging to the same battery module or battery pack, and the second battery assembly is the battery cell with the highest voltage in the set.

[0059] It should be noted that the battery module can also be referred to as a battery cluster, and the battery cell can also be referred to as a single battery, i.e., the names of the above-mentioned battery components are not limited in the embodiments of the present disclosure.

[0060] In some embodiments, the energy storage battery can comprise one or more battery packs, one battery pack can comprise one or more battery modules, and one battery module can comprise one or more battery cells. In some other embodiments, the energy storage battery can not comprise battery modules, such as a cell-to-pack (CTP) battery, the energy storage battery can comprise one or more battery packs, and one battery pack can comprise one or more battery cells. The specific structure of the battery is not limited in the embodiments of the present disclosure.

[0061] The first time period can be a period of time before a certain time in the charging process of the energy storage battery, for example, from the start time of charging to the current time. The first data set can be a plurality of frames of battery charging data collected in the first time period. Each frame of battery charging data can include at least one of the following: the maximum voltage of the first battery assembly, the minimum voltage of the first battery assembly, the current of the first battery assembly, the temperature of the first battery assembly, the heating power of the first battery assembly, and the cooling power of the first battery assembly.

[0062] In some embodiments, the state of the first battery assembly can be monitored periodically, and charging data of the first battery assembly can be collected. For example, data of the state of the first battery assembly can be collected every 15 seconds, 30 seconds, or 60 seconds, and each time the collected data forms a frame of battery charging data. In other words, the data collection interval between two adjacent frames of battery charging data is the inter-frame time slot between the two frames of battery charging data. Therefore, the amount of time related to charging, such as the time taken to charge from 0 to 80%, the remaining charging time, etc., can be calculated based on the frame number of the battery charging data and the inter-frame time slot.

[0063] It should be noted that the inter-frame time slot can be fixed, such as 30 seconds / time or 15 seconds / time, or it can be dynamically adjusted, such as adaptively adjusting the inter-frame time slot according to the monitored temperature, voltage, etc. of the first battery assembly. For example, when the voltage or temperature in the first battery assembly is greater than or equal to a certain threshold, such as voltage ≥ 4V or temperature ≥ 35℃, the inter-frame time slot can be reduced to increase the sampling frequency, such as from collecting one frame of data every 30 seconds to collecting one frame of data every 15 seconds, to collect more intensive battery charging data and more finely control the battery charging process based on the more intensive battery charging data, thereby improving the charging efficiency and safety.

[0064] The collected plurality of frames of battery charging data can be used as a data set, such as the first data set, which can be used to monitor and evaluate the battery charging process and state, such as predicting the remaining charging time of the battery, generating control instructions, etc. The control instructions are used to control the battery charging process, such as increasing or decreasing the charging power, increasing or decreasing the cooling power, etc.

[0065] S102, processing the first data set based on a double-depth prediction model to obtain a plurality of frames of predicted voltages of the second battery assembly in a second time period.

[0066] In some embodiments, the second time period is a period of time after the first time period.

[0067] The double-depth prediction model can be a pre-trained artificial intelligence neural network model. Optionally, Figure 2 A structural diagram of the double-depth prediction model provided by an embodiment of the present disclosure is shown in FIG. 1. As shown in FIG. 1, the double-depth prediction model can include an input layer, a first hidden layer, a second hidden layer, and an output layer.Figure 2 As shown, the double-depth prediction model includes a first battery component prediction model and a second battery component prediction model, the first battery component prediction model is used to determine a plurality of frames of predicted temperature and a plurality of frames of predicted current of the first battery component in the second time period according to the first data set, and the second battery component prediction model is used to determine a plurality of frames of predicted voltage of the second battery component in the second time period according to the plurality of frames of predicted temperature and the plurality of frames of predicted current of the first battery component in the second time period and the voltage of the second battery component.

[0068] S103, determining the first data frame.

[0069] In some embodiments, the first data frame is a data frame in which the first predicted voltage of the second battery component in the second time period is greater than or equal to the charging cutoff voltage.

[0070] S104, determining the battery remaining charging time according to the first data frame.

[0071] In some embodiments, the battery remaining charging time is a difference between a time corresponding to the first data frame and a starting time of the second time period.

[0072] Optionally, Figure 3 A flowchart of another battery charging prediction method provided by an embodiment of the present disclosure is shown. In combination with Figure 1 As shown, S102, processing the first data set based on a double-depth prediction model to obtain a plurality of frames of predicted voltage of the second battery component in the second time period, can include: Figure 3 S1021, inputting the first data set into the first battery component prediction model to obtain a plurality of frames of predicted temperature and a plurality of frames of predicted current of the first battery component in the second time period.

[0073] S1022, inputting the plurality of frames of predicted temperature and the plurality of frames of predicted current of the first battery component in the second time period, and the voltage of the second battery component into the second battery component prediction model to obtain a plurality of frames of predicted voltage of the second battery component in the second time period.

[0074] The specific implementation of S102-S104 will be further described below in combination with

[0075] Figure 4 The specific implementation of S102-S104 will be further described below in combination with

[0076] Optionally, Figure 4 A scene diagram of predicting the battery remaining charging time provided by an embodiment of the present disclosure is shown. As shown, Figure 4 ​As shown, t is the charging time, the first time period includes m charging data frames, the ending time is t1, the second time period includes b charging data frames, the starting time is t2, the time interval between data frames is T, m and b are positive integers greater than 1, b is much greater than m, for example, b is more than 5 times, more than 10 times, etc. of m, and the frame numbers are 1, 2, 3, …, b in turn, the starting times of the frames are t2, t2+T, t2+2T, t2+3T, …, t2+zT, …, t2+(b-1)T in turn, and the ending times of the frames are t2+T, t2+2T, t2+3T, …, t2+zT, …, t2+bT in turn, and then:

[0077] In S1021, m charging data frames of the first battery assembly in the first time period are input into the first battery assembly prediction model to obtain the predicted temperature and the multi-frame predicted current of b charging data frames of the first battery assembly in the second time period, and the battery component with the highest voltage in the first battery assembly is selected as the second battery assembly.

[0078] In S1022, the predicted temperature and the multi-frame predicted current of the b charging data frames and the voltage of the second battery assembly are input into the second battery assembly prediction model to obtain b frames of the predicted voltage of the second battery assembly in the second time period.

[0079] In S103, the first data frame in which the first predicted voltage in the b charging data frames is greater than or equal to the charging cutoff voltage is determined as the first data frame, for example, the zth data frame, that is, the first data frame is the first data frame in which the predicted voltage is greater than or equal to the charging cutoff voltage among the b charging data frames.

[0080] In S104, the battery remaining charging time Treserved is calculated according to the first data frame and the starting time t2 of the second time period, that is, Treserved=t3-t2, where t3 is the time corresponding to the first data frame.

[0081] Further, the time interval T between frames in the second time period is equal, for example, 30 seconds each, to simplify the operation and improve the prediction efficiency. Assuming that the first data frame is the zth frame in the second time period, z is a natural number, 0≤z<b, the battery remaining charging time satisfies Treserved=zT, where Treserved is the battery remaining charging time and T is the time interval between frames in the second time period.

[0082] In some embodiments, please continue to refer to Figure 4 The first time period and the second time period can be discontinuous, that is, t2>t1, or continuous, that is, t2=t1. In other words, the battery charging data as close as possible in time can be used to predict the battery charging remaining time, so as to fully utilize the correlation of the battery charging data in time and further improve the prediction accuracy of the battery charging remaining time.

[0083] Based on the technical solution provided in this disclosure, a trained dual-depth prediction model can be used to determine the second battery component with the highest voltage in the first battery component based on a first dataset composed of multiple frames of battery charging data of the first battery component within a first time period. The predicted voltage of the second battery component within a second time period after the first time period is then determined based on the first data frame (first data frame) of the predicted voltage of the multi-frames that is greater than or equal to the charging cutoff voltage. This can solve the problem that existing battery remaining charging time prediction schemes cannot simultaneously achieve both accuracy and efficiency.

[0084] Furthermore, the problem of estimating the remaining charging time can be transformed into a problem of predicting the future voltage curve, and the problem of complex control flow and low accuracy in iterative prediction schemes can be solved by using a variable-scale training mode.

[0085] Furthermore, self-supervised learning and a large-scale converter timing prediction model are used to solve the voltage prediction problem to improve generalization ability;

[0086] Furthermore, the remaining charging time prediction problem is deconstructed into two non-end-to-end model problems: one is the first battery module prediction model, which predicts future current and temperature based on the highest voltage, lowest voltage, temperature, current, real-time heating power, and real-time cooling power; the other is the second battery module prediction model, which predicts future terminal voltage based on cell temperature, current, and voltage. This transforms the remaining charging time problem into a voltage prediction problem, thereby simplifying the problem and improving prediction efficiency and accuracy.

[0087] The following is combined with Figures 5-8 This section details the training method for the dual-depth prediction model.

[0088] It should be noted that the first battery component prediction model and the second battery component prediction model in the dual-depth prediction model can be trained separately or in combination, and this disclosure does not limit this.

[0089] When the prediction model for the first battery module and the prediction model for the second battery module are trained separately, training datasets can be collected and training can be completed separately according to the training requirements of the two models, without affecting each other, thereby improving the flexibility and efficiency of training.

[0090] When the first battery component prediction model and the second battery component prediction model are trained together, they can be trained using the same training dataset. For example, the training dataset can be input into the first battery component prediction model for training, and the training results of the first battery component prediction model can be used to train the second battery component prediction model. Since the two models are trained using the same training dataset, they have a strong temporal correlation, which can improve the accuracy of the models.

[0091] In actual applications, the first battery component prediction model and the second battery component prediction model can also be trained respectively using respective training data sets, and then the first battery component prediction model and the second battery component prediction model in cascade are trained using the same training data set, so as to balance training accuracy and efficiency.

[0092] Figure 5 A flowchart of a training method of a first battery component prediction model according to an embodiment of the present disclosure is shown in FIG. 5A. As shown in FIG. 5A, the method comprises: Figure 5

[0093] S501: Obtain a first training data set.

[0094] The first training data set is a plurality of frames of battery charging data of the first battery component in a third time period.

[0095] In some embodiments, each frame of battery charging data in the first training data set comprises at least one of the following: a maximum voltage of the first battery component, a minimum voltage of the first battery component, a current of the first battery component, a temperature of the first battery component, a heating power of the first battery component, and a cooling power of the first battery component.

[0096] The third time period is a historical time period, and the first training data set comprises a plurality of frames of battery charging data of the first battery component collected in the third time period. The content and collection method of the first training data set can refer to the first data set, which will not be described here.

[0097] S502: Perform feature extraction on the first training data set to obtain a second training data set.

[0098] For example, the first training data set can be input into a convolutional neural network for feature extraction, such as extracting time correlation features of each frame of battery charging data in the first training data set to obtain the second training data set.

[0099] S503: Train the first battery component prediction model based on the second training data set.

[0100] In some embodiments, the first battery component prediction model comprises a first convolutional neural network and a first conversion network, and the first conversion network comprises a first encoder and a first decoder.

[0101] Optionally, Figure 6 A flowchart of another training method of a first battery component prediction model according to an embodiment of the present disclosure is shown in FIG. 5B. In combination with Figure 5 As shown in FIG. 5B, the method comprises: Figure 6

[0102] ​​In S502, feature extraction is performed on the first training dataset to obtain the second training dataset, which includes:

[0103] S5021, Input the first training dataset into the first convolutional neural network for feature extraction to obtain the second training dataset.

[0104] In S503, a prediction model for the first battery component is trained based on the second training dataset, including:

[0105] S5031, the third time period is divided into the first sub-time period and the second sub-time period, and the second training dataset is divided into the first training data subset and the second training data subset.

[0106] In some embodiments, the data in the first training data subset is battery charging data within a first sub-time period, and the data in the second training data subset is battery charging data within a second sub-time period, wherein the second sub-time period is after the first sub-time period.

[0107] S5032, the first training data subset is input into the first encoder, the third training dataset is input into the first decoder, and the second training data subset is used as the label to train the first battery component prediction model.

[0108] In some embodiments, the third training dataset is a dataset in which data from the second training data subset of the second training dataset is replaced with dummy data.

[0109] Optionally, Figure 7 This is an example diagram of a dataset provided for training a prediction model for a first battery component, according to an embodiment of this disclosure. (Combined with...) Figure 6 ,like Figure 7 As shown, m1+n1 data frames correspond to the aforementioned third time period, where m1 data frames correspond to the first sub-time period within the third time period, and n1 data frames correspond to the second sub-time period within the third time period. The first training dataset consists of m1+n1 original data frames collected during the third time period, and the second training dataset consists of m1+n1 feature data frames obtained after inputting the m1+n1 original data frames into the first convolutional neural network for feature extraction. Figure 7 The first training data subset, labeled A10, includes m1 feature data frames. Figure 7 The middle part is labeled A11), and the second training data subset consists of n1 feature data frames ( Figure 7The A12 is marked as A12). Then, A11 is input into the first encoder, the A12 part in A10 is replaced with dummy data input into the first decoder, A12 is used as a label to compare with the forward training results of the first encoder and the first decoder to generate output errors and gradients, and the weights of the first battery component prediction model are updated in the backward training process, so as to complete the training.

[0110] It should be noted that in the embodiments of the present disclosure, the "dummy data" can also be referred to as blank data, invalid data, or padding data, which refers to data filled or replaced to meet a specific data format, and the data itself does not participate in subsequent operations or has no effect on the operation results.

[0111] As can be seen, the training data includes at least 6 charging data such as the highest voltage, the lowest voltage, the current, the temperature, the real-time heating power, and the real-time cooling power, so that the first battery component prediction model can learn the information of the battery thermal management strategy and the charging strategy, and since the training data does not need to distinguish between charging, discharging, or standing scenes, all of which can be used as input and output of the model, the data amount is large, which can improve the training accuracy of the first battery component prediction model, thereby improving the prediction accuracy of the battery charging remaining time. For example, the training data set can be divided as follows: 98% for the training set, 1% for the validation set, and 1% for the test set.

[0112] In some embodiments, please continue to refer to Figure 7 The length of the second sub-time period is variable, such as n1 can be randomly or configurable between [a1, b1], where a1>0, b1 is much larger than m1, such as b1 is more than 5 times, more than 10 times of m1, etc., that is, the variable scale mode is used for training in the embodiments of the present disclosure, which solves the problems of complex process and low accuracy of repeated iteration prediction, to further improve the efficiency and accuracy.

[0113] Optionally, Figure 8 A flowchart of a training method of a second battery component prediction model according to an embodiment of the present disclosure is provided. As shown in Figure 8 The method comprises the following steps:

[0114] S801, obtaining a fourth training data set.

[0115] In some embodiments, the fourth training data set is a plurality of frames of battery charging data of a second battery component in the first battery component in a fourth time period.

[0116] In some embodiments, each frame of battery charging data in the fourth training data set includes at least one of the following: voltage, current, and temperature of the second battery component.

[0117] S802, performing feature extraction on the fourth training data set to obtain a fifth training data set.

[0118] For example, the fourth training data set can be input into a convolutional neural network for feature extraction, such as extracting time correlation features of each frame of battery charging data in the fourth training data set, to obtain a fifth training data set.

[0119] S803, training the second battery component prediction model based on the fifth training data set.

[0120] In some embodiments, the second battery component prediction model comprises a second convolutional neural network and a second conversion network, and the second conversion network comprises a second encoder and a second decoder.

[0121] Optionally, Figure 9 Another flowchart of a method for training a second battery component prediction model is provided for an embodiment of the present disclosure. In combination with Figure 8 As shown in the figure, the method comprises: Figure 9

[0122] In S802, the fourth training data set is subjected to feature extraction to obtain a fifth training data set, comprising:

[0123] S8021, inputting the fourth training data set into a second convolutional neural network for feature extraction to obtain the fifth training data set.

[0124] In S803, the second battery component prediction model is trained based on the fifth training data set, comprising:

[0125] S8031, dividing the fourth time period into a third sub-time period and a fourth sub-time period, and dividing the fifth training data set into a third training data subset and a fourth training data subset.

[0126] In some embodiments, the data in the third training data subset is battery charging data in the third sub-time period, the data in the fourth training data subset is battery charging data in the fourth sub-time period, and the fourth sub-time period is after the third sub-time period.

[0127] S8032, inputting the third training data subset into the second encoder, inputting the sixth training data set into the second decoder, and inputting the fourth training data subset as a label to train the second battery component prediction model.

[0128] In some embodiments, the sixth training data set is a data set obtained by replacing the data in the fourth training data subset in the fifth training data set with invalid data.

[0129] Optionally, Figure 10 An example diagram of a data set for training a second battery component prediction model is provided for an embodiment of the present disclosure. In combination with Figure 9 As shown in the figure, Figure 10 ​As shown, m² + n² data frames correspond to the fourth time period mentioned above, where m² data frames correspond to the third sub-time period within the fourth time period, and n² data frames correspond to the fourth sub-time period within the fourth time period. The fourth training dataset consists of m² + n² raw data frames collected during the fourth time period, and the fifth training dataset consists of m² + n² feature data frames obtained after inputting the m² + n² raw data frames into the second convolutional neural network for feature extraction. Figure 10 The third training data subset, labeled A20, includes m2 feature data frames. Figure 10 The fourth training data subset, marked as A21, consists of n2 feature data frames. Figure 10 (The data is labeled A22). Then, A21 is input into the second encoder, and the A22 portion in A20 is replaced with dummy data and input into the second decoder. A22 is used as a label to compare with the forward training results of the second encoder and the second decoder to generate output error and gradient, and to update the weights of the second battery component prediction model during the backward training process, thereby completing the training.

[0130] As can be seen, the training data includes at least three charging data points for the second battery module, such as voltage, current, and temperature. This allows the prediction model for the second battery module to learn information such as the battery's thermal management strategy and charging strategy. Furthermore, since the training data does not need to distinguish between charging, discharging, or resting scenarios, it can be used as both input and output to the model. The large amount of data can improve the training accuracy of the prediction model for the second battery module, thereby improving the accuracy of predicting the remaining battery charging time. For example, the training dataset can be divided as follows: 98% for the training set, 1% for the validation set, and 1% for the test set.

[0131] In some embodiments, please refer to Figure 10 The length of the fourth sub-time period is variable, such as n2, which can be random or configurable between [a2, b2], where a2>0 and b2 is much larger than m2, such as b2 being more than 5 times or 10 times greater than m2. That is, the present invention adopts a variable scale mode for training to solve the problems of complex process and low accuracy of repeated iterative prediction, so as to further improve efficiency and accuracy.

[0132] In some embodiments, the fourth time period and the third time period are the same time period, such as m1=m2, n1=n2, a1=a2, b1=b2, at this time A10=A20, A11=A21, A12=A22, that is, the same original data can be used when training the first battery component prediction model and the second battery component model, for example, the original charging data collected in the same time period, which can be used for training the first battery component prediction model and the second battery component model at the same time or in sequence, so as to improve the accuracy and efficiency of model training by using the time correlation of the original charging data collected in the same time period.

[0133] In some embodiments, the first battery component and the second battery component satisfy one of the following:

[0134] The first battery component can be a battery pack, and the second battery component can be a battery module; or,

[0135] The first battery component can be a battery module, and the second battery component can be a battery cell; or,

[0136] The first battery component can be a battery pack, and the second battery component can be a battery cell.

[0137] Figure 11 A structural schematic diagram of an electronic device is provided for an embodiment of the present disclosure. As shown in the figure, the electronic device includes at least one processor 1101, and a memory 1102 in communication connection with the at least one processor 1101; wherein the memory 1102 stores instructions executable by the at least one processor 1101, and the instructions are executed by the at least one processor 1101 to enable the at least one processor 1101 to perform the method described in the above method embodiment. Figure 11

[0138] The memory 1102 and the processor 1101 can be connected in a bus manner, the bus can include any number of interconnected buses and bridges, and the bus connects various circuits of the one or more processors 1101 and the memory 1102 together. The bus can also connect various other circuits such as peripheral devices, voltage stabilizers, and power management circuits together, which are well known in the art, and therefore, further description is not given herein. The bus interface can provide an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, which provide a unit for communicating with various other devices on a transmission medium. The data processed by the processor 1101 is transmitted on a wireless medium through an antenna, and further, the antenna also receives data and transmits the data to the processor 1101.

[0139] ​The processor 1101 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory 1102 can be used to store data used by the processor 1101 in performing operations.

[0140] In some embodiments, the electronic device can be an energy storage device, such as an energy storage cabinet, which can be a device in a larger energy storage system, or an independently operating energy storage device, such as a small household energy storage device, or a sub-device or energy storage battery component or module in other devices, such as a power battery assembly in an electric vehicle. In other embodiments, the electronic device can also be a charging chip or chip system arranged in various devices described above.

[0141] An embodiment of the present disclosure provides a computer readable storage medium, and the computer readable storage medium stores computer instructions. The computer instructions are executed by a processor to implement the method of the first aspect.

[0142] Those skilled in the art can understand that the above-mentioned embodiments are specific embodiments for implementing the present disclosure, and in actual application, various changes can be made in form and details without departing from the spirit and scope of the present disclosure. Any person skilled in the art can make various modifications and changes without departing from the spirit and scope of the present disclosure, and therefore the protection scope of the present disclosure should be limited by the scope defined by the claims.

Claims

1. A method for predicting the charging time of an energy storage battery, characterized in that, include: Obtain a first dataset, which is multi-frame battery charging data of a first battery component within a first time period. The first battery component includes multiple battery parts, and the battery part with the highest voltage among the multiple battery parts constitutes a second battery component. The battery part is one of the following: battery pack, battery module, or battery cell. The first dataset is processed based on a dual-depth prediction model to obtain the multi-frame predicted voltage of the second battery component in a second time period, which is a period of time after the first time period. A first data frame is determined, wherein the first data frame is the first data frame in which the predicted voltage of the second battery component is greater than or equal to the charging cutoff voltage during the second time period; The remaining battery charging time is determined based on the first data frame; The dual-depth prediction model includes a first battery component prediction model and a second battery component prediction model. The first battery component prediction model is used to determine the predicted temperature and predicted current of the first battery component in multiple frames within the second time period based on the first dataset. The second battery component prediction model is used to determine the predicted voltage of the second battery component in multiple frames within the second time period based on the predicted temperature and predicted current of the first battery component in multiple frames within the second time period, as well as the voltage of the second battery component.

2. The method according to claim 1, characterized in that, The process of processing the first dataset based on a dual-depth prediction model to obtain the multi-frame predicted voltage of the second battery component within a second time period includes: Input the first dataset into the first battery component prediction model to obtain the multi-frame predicted temperature and multi-frame predicted current of the first battery component in the second time period. The predicted temperature and current of the first battery module in the second time period and the voltage of the second battery module are input into the prediction model of the second battery module to obtain the predicted voltage of the second battery module in the second time period.

3. The method according to claim 1, characterized in that, The first time period and the second time period are consecutive.

4. The method according to claim 1, characterized in that, The remaining battery charging time is the difference between the time corresponding to the first data frame and the start time of the second time period.

5. The method according to claim 4, characterized in that, The inter-frame time slots within the second time period are equal, the first data frame is the z-th frame within the second time period, where z is a natural number, and the remaining battery charging time satisfies: Treserved = zT; Where Treserved represents the remaining charging time of the battery, and T represents the inter-frame time slot within the second time period.

6. The method according to claim 1, characterized in that, Also includes: Obtain a first training dataset, which is multi-frame battery charging data of the first battery component within a third time period; Feature extraction is performed on the first training dataset to obtain the second training dataset; The first battery component prediction model is trained based on the second training dataset.

7. The method according to claim 6, characterized in that, The first battery component prediction model includes a first convolutional neural network and a first conversion network, wherein the first conversion network includes a first encoder and a first decoder; The step of extracting features from the first training dataset to obtain the second training dataset includes: The first training dataset is input into the first convolutional neural network for feature extraction to obtain the second training dataset; The step of training the first battery component prediction model based on the second training dataset includes: The third time period is divided into a first sub-time period and a second sub-time period, and the second training dataset is divided into a first training data subset and a second training data subset. The data in the first training data subset is the battery charging data within the first sub-time period, and the data in the second training data subset is the battery charging data within the second sub-time period. The second sub-time period is after the first sub-time period. The first training data subset is input into the first encoder, the third training dataset is input into the first decoder, and the second training data subset is used as the label to train the first battery component prediction model. The third training dataset is the dataset after replacing the data in the second training data subset in the second training dataset with dummy data.

8. The method according to claim 7, characterized in that, The length of the second sub-time period is variable.

9. The method according to claim 6, characterized in that, In the first dataset and / or the first training dataset, each frame of battery charging data includes at least one of the following: the highest voltage of the first battery component, the lowest voltage of the first battery component, the current of the first battery component, the temperature of the first battery component, the heating power of the first battery component, and the cooling power of the first battery component.

10. The method according to claim 6, characterized in that, Also includes: Obtain a fourth training dataset, which is a multi-frame battery charging data of the second battery component in the first battery component within a fourth time period. Each frame of battery charging data in the fourth training dataset includes at least one of the following: the voltage, current and temperature of the second battery component. Feature extraction is performed on the fourth training dataset to obtain the fifth training dataset; The second battery component prediction model is trained based on the fifth training dataset.

11. The method according to claim 10, characterized in that, The second battery component prediction model includes a second convolutional neural network and a second conversion network, wherein the second conversion network includes a second encoder and a second decoder; The step of extracting features from the fourth training dataset to obtain the fifth training dataset includes: The fourth training dataset is input into the second convolutional neural network for feature extraction to obtain the fifth training dataset; The step of training the second battery component prediction model based on the fifth training dataset includes: The fourth time period is divided into a third sub-time period and a fourth sub-time period, and the fifth training dataset is divided into a third training data subset and a fourth training data subset. The data in the third training data subset is the battery charging data within the third sub-time period, and the data in the fourth training data subset is the battery charging data within the fourth sub-time period. The fourth sub-time period is after the third sub-time period. The third training data subset is input into the second encoder, the sixth training dataset is input into the second decoder, and the fourth training data subset is used as the label to train the second battery component prediction model. The sixth training dataset is the dataset after replacing the data in the fourth training data subset in the fifth training dataset with invalid data.

12. The method according to claim 11, characterized in that, The length of the fourth sub-time period is variable.

13. The method according to claim 10, characterized in that, The fourth time period and the third time period are the same time period.

14. The method according to any one of claims 1-12, characterized in that, The first battery assembly and the second battery assembly are one of the following combinations: The first battery assembly is a battery pack, and the second battery assembly is a battery module; or... The first battery assembly is a battery module, and the second battery assembly is a battery cell; or, The first battery component is a battery pack, and the second battery component is a battery cell.

15. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method as described in any one of claims 1-14.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method of any one of claims 1-14.

Images