Charging method, electronic equipment, battery pack and energy storage product

Abstract

The embodiment of the invention provides a charging method, electronic equipment, a battery pack and an energy storage product, and relates to the technical field of battery management, and the charging method comprises the steps: obtaining to-be-processed data of a target battery; processing the to-be-processed data by using a pre-trained first potential prediction model to obtain first predicted negative electrode potential values of the target battery at a plurality of future moments in a future time period after the current moment in the charging process; on the basis of the obtained multiple first predicted negative electrode potential values, whether lithium precipitation occurs in the target battery at multiple future moments or not according to the current charging condition is judged; in response to the situation that the target battery is charged according to the current charging condition, lithium precipitation occurs at multiple future moments, the charging condition is adjusted, and the target battery is charged according to the adjusted charging condition; the adjusting mode comprises the steps of reducing the charging rate and / or increasing the battery temperature so as to reduce the influence on the service life of the target battery and prolong the cruising ability of the target battery.

Description

Technical Field

[0001] The present application relates to the technical field of battery management, and in particular, to a charging method, an electronic device, a battery pack, and an energy storage product. Background Art

[0002] Due to advantages such as high energy density, long cycle life, and no memory effect, lithium-ion batteries are widely used in many fields such as new energy vehicles, two / three-wheel electric vehicles, drones, and energy storage products. With the continuous expansion of applications, users have higher and higher requirements for the battery's endurance, and improper charging will cause the battery's endurance to decline.

[0003] Therefore, how to charge the battery to reduce the impact on its life and extend its endurance has become an urgent problem to be solved. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to provide a charging method, an electronic device, a battery pack, and an energy storage product. The specific technical solutions are as follows:

[0005] In the first aspect of the present application, a charging method is provided. The method includes: obtaining the data to be processed of the target battery; wherein, the data to be processed includes: during the current charging process, the real state data of the target battery at multiple historical moments within the historical time period before the current moment, and the real state data includes: the full battery potential value and the charging current value; using the pre-trained first potential prediction model to process the data to be processed, and obtaining the negative electrode potential values of the target battery at multiple future moments within the future time period after the current moment during the current charging process as the first predicted negative electrode potential values; wherein, the first potential prediction model is constructed as: being trained based on the real state data of the sample battery at multiple first sample moments during the historical charging process, and the real potential values at multiple second sample moments after the multiple first sample moments; based on the obtained multiple first predicted negative electrode potential values, determining whether lithium plating will occur in the target battery at the multiple future moments according to the current charging conditions; in response to lithium plating occurring in the target battery at the multiple future moments when charging according to the current charging conditions, adjusting the charging conditions, and charging the target battery according to the adjusted charging conditions; wherein, the adjustment methods include: reducing the charging rate, and / or, increasing the battery temperature.

[0006] Since there is a mapping relationship between the negative electrode potential value of the target battery at a future time and the true state data at a historical time, the true state data at the historical time is used to predict the negative electrode potential value at the future time. In addition, during the use of the target battery, the negative electrode potential value characterizes whether lithium plating occurs in the target battery. Therefore, the first predicted negative electrode potential values at multiple future times are used to determine whether lithium plating will occur in the target battery at multiple future times if the target battery continues to be charged according to the current charging conditions. Correspondingly, if it is determined that lithium plating will occur in the target battery at multiple future times according to the current charging conditions, the charging rate is reduced, and / or the battery temperature is increased to reduce the risk of charging lithium plating in the target battery, so as to reduce the impact on the life of the target battery and extend its endurance ability.

[0007] In one or more embodiments of the present application, the adjusting the charging conditions includes: reducing the charging rate and / or increasing the battery temperature to obtain a standby charging condition; using the first electrochemical mechanism model of the target battery pre-constructed to calculate the simulated potential values of the target battery at the multiple future times when charged according to the standby charging condition; wherein, the simulated potential values include at least one of the following: simulated full battery potential value, simulated positive electrode potential value, and simulated negative electrode potential value; inputting the simulated potential values at the multiple future times into a pre-trained second potential prediction model to obtain the negative electrode potential values of the target battery at the multiple future times when charged according to the standby charging condition as the second predicted negative electrode potential values; wherein, the second potential prediction model is constructed based on the simulated potential values and the true negative electrode potential values of the sample battery at multiple third sample times, and the simulated potential values at the multiple third sample times are calculated using the second electrochemical mechanism model of the sample battery pre-constructed; based on the second predicted negative electrode potential values, determining whether lithium plating will occur in the target battery at the multiple future times when charged according to the standby charging condition; in response to lithium plating occurring in the target battery at the multiple future times when charged according to the standby charging condition, returning to execute the step of reducing the charging rate and / or increasing the battery temperature to obtain a standby charging condition until it is determined that lithium plating will not occur in the target battery at the multiple future times when charged according to the latest standby charging condition, and determining the latest standby charging condition as the adjusted charging condition.

[0008] By adding an electrochemical mechanism model to monitor the charging process of the target battery, when it is determined that lithium plating will occur in the target battery during the subsequent charging process according to the current charging conditions, the current charging conditions are adjusted to control the target battery to be charged according to the adjusted charging conditions, further reducing the risk of lithium plating occurring in the target battery during the entire charging process and completing the charging process of the target battery as soon as possible.

[0009] In one or more embodiments of the present application, processing the data to be processed by using the pre-trained first potential prediction model to obtain the negative electrode potential values at multiple future moments within a future time period after the current moment during the current charging process of the target battery as the first predicted negative electrode potential values includes: inputting the data to be processed into the pre-trained first potential prediction model to obtain the full battery potential values, the positive electrode potential values, and the negative electrode potential values at multiple future moments within a future time period after the current moment during the current charging process of the target battery, respectively, as the predicted full battery potential values, the predicted positive electrode potential values, and the third predicted negative electrode potential values; for each future moment, calculating the first predicted negative electrode potential value at this future moment by using a preset formula based on the predicted full battery potential value, the predicted positive electrode potential value, and the third predicted negative electrode potential value; where the preset formula is:

[0010]

[0011] SOLP represents the first predicted negative electrode potential value at this future moment, U CTP represents the predicted positive electrode potential value at this future moment, U BTP represents the predicted full battery potential value at this future moment, U ATP represents the third predicted negative electrode potential value at this future moment.

[0012] The negative electrode potential values, the positive electrode potential values, and the full battery potential values are predicted respectively. Since there are certain deviations in the prediction itself, the final negative electrode potential value is comprehensively calculated based on the predicted potential values according to the preset formula to reduce the above deviations and further improve the accuracy of the first predicted negative electrode potential value, providing a precise basis for determining whether lithium plating will occur in the target battery subsequently.

[0013] In one or more embodiments of the present application, the first potential prediction model includes a first sub-model, a second sub-model, and a third sub-model; inputting the data to be processed into the pre-trained first potential prediction model to obtain the full-cell potential values, the positive electrode potential values, and the negative electrode potential values at multiple future times within a future time period after the current time during the current charging process of the target battery, and respectively serving as the predicted full-cell potential values, the predicted positive electrode potential values, and the third predicted negative electrode potential values, including: inputting the data to be processed into the pre-trained first sub-model to obtain the negative electrode potential values at multiple future times within a future time period after the current time during the current charging process of the target battery, and serving as the third predicted negative electrode potential values, wherein the first sub-model is constructed by: training based on the true state data of the sample battery at multiple first sample times and the true negative electrode potential values at multiple second sample times during the historical charging process; inputting the data to be processed into the pre-trained second sub-model to obtain the positive electrode potential values at multiple future times within a future time period after the current time during the current charging process of the target battery, and serving as the predicted positive electrode potential values, wherein the second sub-model is constructed by: training based on the true state data of the sample battery at multiple first sample times and the true positive electrode potential values at multiple second sample times during the historical charging process; inputting the data to be processed into the pre-trained third sub-model to obtain the full-cell potential values at multiple future times within a future time period after the current time during the current charging process of the target battery, and serving as the predicted full-cell potential values, wherein the third sub-model is constructed by: training based on the true state data of the sample battery at multiple first sample times and the true full-cell potential values at multiple second sample times during the historical charging process.

[0014] The full-cell potential values, the positive electrode potential values, and the negative electrode potential values are predicted by three different models respectively to further improve the accuracy of the predicted first predicted negative electrode potential value and provide a precise basis for determining whether the target battery will undergo lithium deposition in the subsequent stage.

[0015] In one or more embodiments of the present application, based on the obtained multiple first predicted negative electrode potential values, determining whether the target battery will undergo lithium deposition at the multiple future times according to the current charging conditions includes: calculating the average value of the remaining predicted negative electrode potential values except the maximum value and the minimum value among the obtained multiple first predicted negative electrode potential values; in response to the average value being greater than the preset threshold, determining that the target battery will not undergo lithium deposition at the multiple future times according to the current charging conditions; in response to the average value not being greater than the preset threshold, determining that the target battery will undergo lithium deposition at the multiple future times according to the current charging conditions.

[0016] The average value of the remaining predicted negative electrode potential values among the obtained multiple first predicted negative electrode potential values, excluding the maximum value and the minimum value, accurately represents the negative electrode potential value within the future time period. Correspondingly, this average value is used to detect whether lithium plating will occur in the target battery at multiple future moments, so as to improve the accuracy of the lithium plating detection result.

[0017] In one or more embodiments of the present application, the first potential prediction model includes at least one of the following: convolutional neural network, densely connected network, recurrent neural network, long short-term memory neural network, model with attention mechanism, and Transformer network; and / or, the second potential prediction model includes at least one of the following: convolutional neural network, densely connected network, recurrent neural network, long short-term memory neural network, model with attention mechanism, and Transformer network.

[0018] Based on the above network, the accuracy of the first predicted negative electrode potential value is further improved, providing a precise basis for subsequent determination of whether lithium plating will occur in the target battery.

[0019] In one or more embodiments of the present application, the first electrochemical mechanism model is: single particle model, quasi-two-dimensional model, or multi-dimensional multi-field electrochemical model; the second electrochemical mechanism model is: single particle model, quasi-two-dimensional model, or multi-dimensional multi-field electrochemical model.

[0020] Based on the above electrochemical model, the accuracy of the obtained simulated potential value is improved, so as to further improve the accuracy of the first predicted negative electrode potential value, providing a precise basis for subsequent determination of whether lithium plating will occur in the target battery.

[0021] In one or more embodiments of the present application, the real state data further includes: battery temperature; and / or, the data to be processed further includes: the simulated potential values of the target battery at the multiple future moments according to the current charging conditions; the first potential prediction model is constructed by: training based on the real state data of the sample battery at multiple first sample moments during the historical charging process, and the real potential values and simulated potential values at multiple second sample moments after the multiple first sample moments; the simulated potential values at the multiple second sample moments are calculated by using the pre-constructed second electrochemical mechanism model of the sample battery.

[0022] Since the negative electrode potential value of the target battery is affected by temperature, therefore, the negative electrode potential value at future moments is predicted by combining the battery temperature, further improving the accuracy of the first predicted negative electrode potential value. Since the first electrochemical mechanism model predicts the real potential value of the target battery at future moments, therefore, the accuracy of the first predicted negative electrode potential value is further improved by combining the simulated potential value obtained from the first electrochemical mechanism model.

[0023] In one or more embodiments of the present application, the obtaining of the data to be processed of the target battery includes: when any one of a plurality of preset adjustment times is reached, obtaining the data to be processed of the target battery; wherein, the adjustment interval between every two adjacent adjustment times is not greater than the duration of the future time period.

[0024] After determining the charging conditions within a future time period each time, control the target battery to charge within this future time period according to the determined charging conditions. Subsequently, before reaching the last moment of this future time period, determine the charging conditions within a new future time period again according to this adjustment interval. In this way, the charging conditions for a subsequent period of time are determined in real time according to the latest state of the target battery, further reducing the impact on the life of the target battery and ensuring its endurance.

[0025] In one or more embodiments of the present application, the method further includes: in response to the target battery charging according to the current charging conditions and no lithium plating occurring at the plurality of future times, keeping the current charging conditions unchanged and continuing to charge the target battery.

[0026] Keep the current charging conditions unchanged and continue to charge the target battery to reduce the risk of lithium plating occurring in the target battery during the entire charging process and complete the charging process of the target battery as soon as possible.

[0027] The second aspect of the present application provides a charging device, the device includes: a data to be processed obtaining module configured to obtain the data to be processed of the target battery; wherein, the data to be processed includes: during the current charging process, the true state data of the target battery at a plurality of historical times within a historical time period before the current time, and the true state data includes: the full battery potential value and the charging current value; a prediction module configured to process the data to be processed by using a pre-trained first potential prediction model to obtain the negative electrode potential values of the target battery at a plurality of future times within a future time period after the current time during the current charging process as the first predicted negative electrode potential values; wherein, the first potential prediction model is constructed by: training based on the true state data of the sample battery at a plurality of first sample times during the historical charging process and the true potential values at a plurality of second sample times after the plurality of first sample times; a judgment module configured to judge whether lithium plating will occur in the target battery at the plurality of future times according to the current charging conditions based on the obtained plurality of first predicted negative electrode potential values; an adjustment module configured to adjust the charging conditions in response to the target battery charging according to the current charging conditions and lithium plating occurring at the plurality of future times; a first control module configured to charge the target battery according to the adjusted charging conditions; wherein, the adjustment method includes: reducing the charging rate, and / or, increasing the battery temperature.

[0028] Since there is a mapping relationship between the negative electrode potential value of the target battery at a future moment and the real state data at a historical moment, the real state data at the historical moment is used to predict the negative electrode potential value at the future moment. Additionally, during the use of the target battery, the negative electrode potential value indicates whether lithium plating occurs in the target battery. Therefore, the first predicted negative electrode potential values at multiple future moments are used to determine whether lithium plating will occur in the target battery at multiple future moments if the target battery continues to be charged according to the current charging conditions. Correspondingly, if it is determined that lithium plating will occur in the target battery at multiple future moments according to the current charging conditions, the charging rate is reduced, and / or the battery temperature is increased to reduce the risk of charging lithium plating in the target battery, so as to reduce the impact on the life of the target battery and extend its endurance capacity.

[0029] In one or more embodiments of the present application, the adjustment module is configured to reduce the charging rate and / or increase the battery temperature to obtain a standby charging condition; use the pre-constructed first electrochemical mechanism model of the target battery to calculate the simulated potential values of the target battery at the multiple future moments when charged according to the standby charging condition; wherein the simulated potential values include at least one of the following: simulated full battery potential value, simulated positive electrode potential value, and simulated negative electrode potential value; input the simulated potential values at the multiple future moments into a pre-trained second potential prediction model to obtain the negative electrode potential values of the target battery at the multiple future moments when charged according to the standby charging condition, as the second predicted negative electrode potential values; wherein the second potential prediction model is constructed based on the simulated potential values and the real negative electrode potential values of the sample battery at multiple third sample moments; the simulated potential values at the multiple third sample moments are calculated using the pre-constructed second electrochemical mechanism model of the sample battery; based on the second predicted negative electrode potential values, determine whether lithium plating will occur in the target battery at the multiple future moments when charged according to the standby charging condition; in response to lithium plating occurring in the target battery at the multiple future moments when charged according to the standby charging condition, return to execute the step of reducing the charging rate and / or increasing the battery temperature to obtain a standby charging condition until it is determined that lithium plating will not occur in the target battery at the multiple future moments when charged according to the latest standby charging condition, and determine the latest standby charging condition as the adjusted charging condition.

[0030] By adding an electrochemical mechanism model, the charging process of the target battery is monitored. When it is determined that lithium plating will occur in the target battery during the subsequent charging process according to the current charging conditions, the current charging conditions are adjusted to control the target battery to be charged according to the adjusted charging conditions, further reducing the risk of lithium plating occurring in the target battery during the entire charging process and completing the charging process of the target battery as soon as possible.

[0031] In one or more embodiments of the present application, the prediction module is configured to input the data to be processed into a pre-trained first potential prediction model to obtain the full-cell potential values, the positive electrode potential values, and the negative electrode potential values at multiple future times within a future time period after the current time during the current charging process of the target battery, and use them as the predicted full-cell potential values, the predicted positive electrode potential values, and the third predicted negative electrode potential values respectively; for each future time, based on the predicted full-cell potential values, the predicted positive electrode potential values, and the third predicted negative electrode potential values, use a preset formula to calculate the first predicted negative electrode potential value at this future time; where the preset formula is:

[0032]

[0033] SOLP represents the first predicted negative electrode potential value at this future time, U CTP represents the predicted positive electrode potential value at this future time, U BTP represents the predicted full-cell potential value at this future time, U ATP represents the third predicted negative electrode potential value at this future time.

[0034] Predict the negative electrode potential value, the positive electrode potential value, and the full-cell potential value respectively. Since there are certain deviations in the prediction itself, according to the preset formula, the final negative electrode potential value is comprehensively calculated based on the predicted potential values to reduce the above deviations and further improve the accuracy of the first predicted negative electrode potential value, providing a precise basis for determining whether lithium plating will occur in the target battery in the subsequent process.

[0035] In one or more embodiments of the present application, the first potential prediction model includes a first sub-model, a second sub-model, and a third sub-model; the prediction module is configured to input the data to be processed into the pre-trained first sub-model to obtain the negative electrode potential values at multiple future moments within a future time period after the current moment during the current charging process of the target battery, as the third predicted negative electrode potential values; wherein, the first sub-model is constructed by: being trained based on the true state data of the sample battery at multiple first sample moments and the true negative electrode potential values at multiple second sample moments during the historical charging process; inputting the data to be processed into the pre-trained second sub-model to obtain the positive electrode potential values at multiple future moments within a future time period after the current moment during the current charging process of the target battery, as the predicted positive electrode potential values; wherein, the second sub-model is constructed by: being trained based on the true state data of the sample battery at multiple first sample moments and the true positive electrode potential values at multiple second sample moments during the historical charging process; inputting the data to be processed into the pre-trained third sub-model to obtain the full battery potential values at multiple future moments within a future time period after the current moment during the current charging process of the target battery, as the predicted full battery potential values; wherein, the third sub-model is constructed by: being trained based on the true state data of the sample battery at multiple first sample moments and the true full battery potential values at multiple second sample moments during the historical charging process.

[0036] The full battery potential value, the positive electrode potential value, and the negative electrode potential value are predicted by three different models respectively to further improve the accuracy of the predicted first predicted negative electrode potential value, providing an accurate basis for determining whether the target battery will undergo lithium plating in the subsequent stage.

[0037] In one or more embodiments of the present application, the judgment module is configured to calculate the average value of the remaining predicted negative electrode potential values among the multiple first predicted negative electrode potential values obtained, excluding the maximum value and the minimum value; in response to the average value being greater than the preset threshold, it is determined that according to the current charging conditions, the target battery will not undergo lithium plating at the multiple future moments; in response to the average value not being greater than the preset threshold, it is determined that according to the current charging conditions, the target battery will undergo lithium plating at the multiple future moments.

[0038] The average value of the other predicted negative electrode potential values among the multiple first predicted negative electrode potential values obtained accurately represents the negative electrode potential value within the future time period. Correspondingly, this average value is used to detect whether the target battery will undergo lithium plating at multiple future moments to improve the accuracy of the lithium plating detection result.

[0039] In one or more embodiments of the present application, the first potential prediction model includes at least one of the following: convolutional neural network, densely connected network, recurrent neural network, long short-term memory neural network, model with attention mechanism, and Transformer network; and / or, the second potential prediction model includes at least one of the following: convolutional neural network, densely connected network, recurrent neural network, long short-term memory neural network, model with attention mechanism, and Transformer network.

[0040] Based on the above network, the accuracy of the first predicted negative electrode potential value is further improved, providing a precise basis for subsequent determination of whether lithium plating will occur in the target battery.

[0041] In one or more embodiments of the present application, the first electrochemical mechanism model is: single particle model, quasi-two-dimensional model, or multi-dimensional multi-field electrochemical model; and / or, the second electrochemical mechanism model is: single particle model, quasi-two-dimensional model, or multi-dimensional multi-field electrochemical model.

[0042] Based on the above electrochemical model, the accuracy of the obtained simulated potential value is improved, so as to further improve the accuracy of the first predicted negative electrode potential value, providing a precise basis for subsequent determination of whether lithium plating will occur in the target battery.

[0043] In one or more embodiments of the present application, the real state data further includes: battery temperature; and / or, the data to be processed further includes: simulated potential values of the target battery at the multiple future moments according to the current charging conditions; the first potential prediction model is constructed by: training based on the real state data of the sample battery at multiple first sample moments during the historical charging process, and the real potential values and simulated potential values at multiple second sample moments after the multiple first sample moments; the simulated potential values at the multiple second sample moments are calculated using the pre-constructed second electrochemical mechanism model of the sample battery.

[0044] Since the negative electrode potential value of the target battery is affected by temperature, therefore, the negative electrode potential value at future moments is predicted by combining the battery temperature, further improving the accuracy of the first predicted negative electrode potential value. Since the first electrochemical mechanism model predicts the real potential value of the target battery at future moments, therefore, by combining the simulated potential value obtained from the first electrochemical mechanism model, the accuracy of the first predicted negative electrode potential value is further improved.

[0045] In one or more embodiments of the present application, the data acquisition module to be processed is configured to acquire the data to be processed of the target battery when any one of a preset plurality of adjustment moments is reached; wherein, the adjustment interval between every two adjacent adjustment moments is not greater than the duration of the future time period.

[0046] After determining the charging conditions for a future time period each time, control the target battery to charge within the future time period according to the determined charging conditions. Subsequently, before reaching the last moment of the future time period, determine the charging conditions for a new future time period again at the adjustment interval. In this way, determine the charging conditions for a subsequent period of time in real time according to the latest state of the target battery, further reduce the impact on the life of the target battery, and ensure its endurance ability.

[0047] In one or more embodiments of the present application, the device further includes: a second control module configured to, in response to the target battery charging according to the current charging conditions and no lithium plating occurring at the multiple future moments, keep the current charging conditions unchanged and continue to charge the target battery.

[0048] Keep the current charging conditions unchanged and continue to charge the target battery to reduce the risk of lithium plating occurring in the target battery during the entire charging process, and at the same time complete the charging process of the target battery as soon as possible.

[0049] The third aspect of the present application provides an electronic device configured to execute the method described in any one of the first aspects above.

[0050] The fourth aspect of the present application provides a battery pack, which includes a battery module and the electronic device described in the third aspect above.

[0051] The fifth aspect of the present application provides an energy storage product, which includes a battery module and the electronic device described in the third aspect above.

[0052] The sixth aspect of the present application provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by a processor, it implements the method described in any one of the first aspects above.

[0053] The seventh aspect of the present application provides a computer program product containing instructions, which when running on a computer causes the computer to execute the method described in any one of the first aspects above.

[0054] Advantages of the embodiments of the present application:

[0055] A charging method provided by an embodiment of the present application includes obtaining data to be processed. The data to be processed includes true state data of a target battery at multiple historical moments within a historical time period before the current moment during the current charging process. The true state data includes a full-cell potential value and a charging current value. Then, the data to be processed is processed using a pre-trained first potential prediction model to obtain negative electrode potential values of the target battery at multiple future moments within a future time period after the current moment during the current charging process, which are used as first predicted negative electrode potential values. The first potential prediction model is constructed by training based on the true state data of a sample battery at multiple first sample moments during a historical charging process and the true potential values at multiple second sample moments after the multiple first sample moments. Furthermore, based on the obtained multiple first predicted negative electrode potential values, it is determined whether lithium plating will occur in the target battery at multiple future moments according to the current charging conditions. If it is determined that lithium plating will occur in the target battery at multiple future moments according to the current charging conditions, the charging conditions are adjusted, and the target battery is charged according to the adjusted charging conditions. Among them, the adjustment methods include: reducing the charging rate and / or increasing the battery temperature.

[0056] Since there is a mapping relationship between the negative electrode potential value of the target battery at a future moment and the true state data at a historical moment, the true state data at a historical moment is used to predict the negative electrode potential value at a future moment. In addition, during the use of the target battery, the negative electrode potential value indicates whether lithium plating has occurred in the target battery. Therefore, the first predicted negative electrode potential values at multiple future moments are used to determine whether lithium plating will occur in the target battery at multiple future moments if the target battery continues to be charged according to the current charging conditions. Correspondingly, if it is determined that lithium plating will occur in the target battery at multiple future moments according to the current charging conditions, the charging rate is reduced and / or the battery temperature is increased to reduce the risk of charging lithium plating in the target battery, so as to reduce the impact on the life of the target battery and extend its endurance ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application, and those of ordinary skill in the art can also obtain other embodiments based on these drawings.

[0058] Figure 1 It is a flowchart of a charging method provided by an embodiment of the present application;

[0059] Figure 2A It is a schematic diagram of the true negative electrode potential value of a battery (referred to as a battery to be utilized) provided by an embodiment of the present application;

[0060] Figure 2BSchematic diagram of the predicted negative electrode potential value of a battery to be utilized obtained by using a trained first sub-model in an embodiment of the present application;

[0061] Figure 2C For Figure 2B the difference between the predicted negative electrode potential value in Figure 2A and the true negative electrode potential value in

[0062] Figure 3A Schematic diagram of the true negative electrode potential value of a battery (referred to as a battery to be utilized) provided in an embodiment of the present application;

[0063] Figure 3B Schematic diagram of the predicted negative electrode potential value of a battery to be utilized obtained by using a trained second sub-model and third sub-model in an embodiment of the present application;

[0064] Figure 3C For Figure 3B the difference between the predicted negative electrode potential value in Figure 3A and the true negative electrode potential value in

[0065] Figure 4 Flowchart of another charging method provided in an embodiment of the present application;

[0066] Figure 5 Schematic flowchart of determining the charging conditions at a future moment in the charging method provided in an embodiment of the present application;

[0067] Figure 6 Structural diagram of a charging device provided in an embodiment of the present application. Detailed implementation manners

[0068] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art based on the present application belong to the scope of protection of the present application.

[0069] In order to reduce the impact on the battery life and extend the battery's endurance, embodiments of the present application provide a charging method, an electronic device, a battery pack, an energy storage product, a device, and a medium.

[0070] Next, the charging method provided by the embodiments of the present application will be introduced first. In some embodiments of the present application, this method is applied to an electronic device that provides computing services. For example, the electronic device is a battery management system (BMS) in a battery pack, or an energy management system (EMS) of an energy storage product. Or it is a server, including a local server and a cloud server. It can be understood that a battery pack is a product that includes a battery module and a battery management system electrically connected to the battery module, and the battery module includes at least one battery. When the battery module includes multiple batteries, the electrical connection methods between the batteries include: series connection, parallel connection, or hybrid connection. In the present application, hybrid connection means that the electrical connection between the batteries includes series connection and parallel connection.

[0071] As Figure 1 shown, the embodiments of the present application provide a flowchart of a charging method, and this method includes the following steps:

[0072] S101: Obtain the data to be processed of the target battery.

[0073] The data to be processed includes: during the current charging process, the true state data of the target battery at multiple historical moments within the historical time period before the current moment. The true state data includes: the full-cell potential value and the charging current value.

[0074] In some embodiments of the present application, the full-cell potential value of the battery represents the potential value of the positive electrode of the battery relative to the negative electrode, the positive electrode potential value represents the potential value of the positive electrode of the battery relative to the reference electrode, and the negative electrode potential value represents the potential value of the negative electrode of the battery relative to the reference electrode.

[0075] It can be understood that the target battery is the battery being currently charged, and the electronic device controls the charging process of the target battery based on the charging method provided by the embodiments of the present application. At this time, the electronic device is the battery management system of the battery pack including the target battery, or the energy management system in the energy storage product including the target battery.

[0076] In one or more embodiments of the present application, during a complete charging process, the electronic device executes step S101 - S104 one or more times. During the process of adjusting the charging conditions of the battery, after obtaining the above-mentioned true state data, the electronic device determines the charging conditions at multiple future moments based on the charging method provided by the embodiments of the present application to control the charging process.

[0077] In one or more embodiments of the present application, the duration between every two moments of obtaining the real state data is defined as the adjustment interval. As a specific example of the present application, in some actual usage scenarios, the electronic device periodically determines the charging conditions at multiple future moments according to a fixed adjustment interval. That is, the electronic device periodically controls the charging process of the target battery.

[0078] In one or more embodiments of the present application, the multiple historical moments in step S101 are determined by uniformly sampling the moments within the historical time period before the current moment. That is, the duration between every two adjacent historical moments is the same. As some specific examples of the present application, this duration is 1 second, 10 seconds, 100 seconds, 1000 seconds, etc. The present application does not make a special limitation on this duration.

[0079] It can be understood that the number of the multiple historical moments is consistent with the number of the first sample moments used for training the first potential prediction model in step S102. Correspondingly, the duration between every two adjacent first sample moments is the same, and this duration is consistent with the duration between two adjacent historical moments. In this way, the data input to the first potential prediction model when predicting the negative electrode potential value of the target battery at a future moment is consistent with the acquisition method of the input data when training the first potential prediction model, improving the accuracy of obtaining the negative electrode potential value of the target battery at a future moment using the first potential prediction model.

[0080] S102: Process the data to be processed using the pre-trained first potential prediction model to obtain the predicted negative electrode potential values (referred to as the first predicted negative electrode potential values) at multiple future moments within the future time period after the current moment during the current charging process of the target battery.

[0081] The first potential prediction model is constructed as follows: It is trained based on the real state data of the sample battery at multiple first sample moments during the historical charging process and the real potential values at multiple second sample moments after the multiple first sample moments. Therefore, the first potential prediction model reflects the mapping relationship between the state data of the battery at historical moments and the real potential values at future moments, and is used to predict the negative electrode potential value of the target battery at a future moment.

[0082] In order to further improve the robustness of the first potential prediction model to adapt to the state and working environment of the battery, in one or more embodiments of the present application, the real state data of the sample battery at the first sample moment and the real potential values at the second sample moment are obtained under a variety of different application scenarios. The state and working environment of the battery include: battery aging state, environmental temperature, environmental humidity, charging scheme, irradiation, etc., to obtain a variety of application scenarios. The charging scheme includes constant current charging, constant current constant voltage charging, multi-stage constant current charging, and pulse charging, etc.

[0083] In the process of predicting the negative electrode potential value of the target battery at future moments, the plurality of future moments are located after the above historical time period, and the earliest future moment is adjacent to the historical time period. In one or more embodiments of the present application, the plurality of future moments represent the moments determined by uniformly sampling the moments within the future time period after the current moment, that is, the time duration between every two adjacent future moments is the same. As some specific examples of the present application, the time duration is 1 second, 10 seconds, 100 seconds, or 1000 seconds, etc., and the present application does not make a special limitation on the time duration.

[0084] It can be understood that the number of the plurality of future moments is the same as the number of the second sample moments used for training the first potential prediction model. Correspondingly, the time duration between every two adjacent second sample moments is the same, and this time duration is the same as the time duration between two adjacent future moments.

[0085] In the process of predicting the negative electrode potential value of the target battery at future moments, the duration of the future time period after the current moment is the same as, or different from, the duration of the previous historical time period.

[0086] In one or more embodiments of the present application, in order to continuously adjust the charging conditions according to the state of the target battery during a complete charging process, the duration of the future time period is not less than the above adjustment interval. That is, after determining the charging conditions within a future time period each time, the target battery is controlled to be charged within this future time period according to the determined charging conditions. Subsequently, before reaching the last moment of this future time period, the charging conditions within a new future time period are determined again according to the adjustment interval. In this way, the charging conditions for the subsequent period of time are determined in real time according to the latest state of the target battery, further reducing the impact on the battery life and ensuring its endurance.

[0087] For example, in the process of charging the target battery, the durations of the future time period and the historical time period are both 1 minute, and the real state data of the target battery is collected once every second. That is, the plurality of historical moments are the moments corresponding to each second within the one minute before the current moment, so the number of historical moments is 60. Correspondingly, the plurality of future moments are the moments corresponding to each second within the one minute after the current moment, that is, the first predicted negative electrode potential values of 60 future moments are obtained each time.

[0088] If the duration of the future time period is equal to the above adjustment interval, that is, the adjustment interval is 1 minute, during the charging process of the target battery, when the charging duration reaches one minute, the first potential prediction model is used to process the true state data of the target battery every second within the first minute to obtain the first predicted negative electrode potential value within the second minute. Then, according to steps S103 and S104, the charging conditions of the target battery within the second minute are determined, and the target battery is charged according to the determined charging conditions.

[0089] Correspondingly, when the charging duration reaches two minutes, the first potential prediction model is used to process the true state data of the target battery every second within the second minute to obtain the first predicted negative electrode potential value within the third minute. Then, according to steps S103 and S104, the charging conditions of the target battery within the third minute are determined, and the target battery is charged according to the determined charging conditions.

[0090] And so on until the charging process of the target battery ends. For example, the target battery is charged to a fully charged state, or the charging process is interrupted.

[0091] In some other embodiments of the present application, if the duration of the future time period is greater than the above adjustment interval, for example, the adjustment interval is 30 seconds, during the charging process of the target battery, the charging conditions are not adjusted within the first 60 seconds of starting charging. When the charging duration reaches 60 seconds, the first potential prediction model is used to process the true state data of the target battery every second from the 60th second to the 120th second to obtain the first predicted negative electrode potential value from the 60th second to the 120th second. Then, according to steps S103 and S104, the charging conditions of the target battery from the 60th second to the 120th second are determined, and the target battery is charged according to the determined charging conditions.

[0092] Correspondingly, when the charging duration reaches 90 seconds, the first potential prediction model is used to process the true state data of the target battery every second from the 30th second to the 90th second to obtain the first predicted negative electrode potential value from the 90th second to the 150th second. Then, according to steps S103 and S104, the charging conditions of the target battery from the 90th second to the 150th second are determined, and the target battery is charged according to the determined charging conditions.

[0093] And so on until the charging process of the target battery ends. For example, the target battery is charged to a fully charged state, or the charging process is interrupted.

[0094] S103: Based on the obtained multiple first predicted negative electrode potential values, determine whether lithium plating will occur in the target battery at multiple future moments according to the current charging conditions.

[0095] Lithium plating refers to the phenomenon that during the charging process of a battery, lithium ions are deposited on the surface of the negative electrode to form metallic lithium. This phenomenon occurs when the battery temperature is relatively low or under other improper charging conditions. Lithium plating can lead to the formation of lithium dendrites inside the battery, which may pierce the separator and cause an internal short circuit, and even trigger thermal runaway and safety accidents. In addition, lithium plating also reduces the number of lithium ions, resulting in battery capacity attenuation, shortening the battery life, and further shortening the battery's endurance ability. Therefore, in order to reduce the impact on the battery life and extend its endurance ability, in the embodiments of the present application, it is determined whether the target battery will undergo lithium plating at multiple future moments according to the current charging conditions, and the subsequent charging conditions are adjusted according to the determination result.

[0096] It can be understood that when charging according to the current charging conditions (i.e., the charging conditions remain unchanged), the first predicted negative electrode potential values of the target battery at multiple future moments are obtained. In some embodiments of the present application, whether the battery undergoes lithium plating is reflected by the negative electrode potential value of the battery, and the first predicted negative electrode potential values at multiple future moments are used to determine whether the target battery will undergo lithium plating at multiple future moments if the target battery continues to be charged according to the current charging conditions.

[0097] In one or more embodiments of the present application, since there are first predicted negative electrode potential values at multiple future moments, therefore, the average level of these multiple first predicted negative electrode potential values is calculated. Furthermore, according to the magnitude relationship between the average level and a preset threshold, it is determined whether the target battery will undergo lithium plating at multiple future moments. For example, if the average level is greater than the preset threshold, it is determined that the target battery will not undergo lithium plating at multiple future moments; if the average level is not greater than the preset threshold, it is determined that the target battery will undergo lithium plating at multiple future moments. As some specific examples of the present application, the preset threshold is 0V, or, in order to improve the fault tolerance rate of the solution, the preset threshold is set to other values near 0V. For example, the preset threshold is 0.01V, 0.03V, 0.05V, or 0.1V.

[0098] In one or more embodiments of the present application, the maximum value and the minimum value among the multiple first predicted negative electrode potential values are removed, and the average value of the remaining first predicted negative electrode potential values is calculated to obtain the above-mentioned average level.

[0099] S104: In response to the target battery undergoing lithium plating at multiple future moments when charged according to the current charging conditions, adjust the charging conditions, and charge the target battery according to the adjusted charging conditions; the adjustment methods include: reducing the charging rate, and / or, increasing the battery temperature.

[0100] In some embodiments of the present application, the battery temperature is characterized by the ambient temperature, and the ambient temperature is increased to increase the battery temperature by heat transfer, so as to reduce the degree of subsequent lithium plating of the target battery.

[0101] In some other embodiments of the present application, an electrical device (such as an energy storage product, an electric vehicle) includes a heating component for heating the battery. In these electrical devices, the battery is heated by the heating component to raise the temperature of the battery, so as to reduce the degree of lithium plating of the target battery subsequently.

[0102] In addition, the degree of lithium plating of the subsequent target battery is reduced by reducing the charging rate. Therefore, when it is determined that the target battery will undergo lithium plating at multiple future moments if charging continues according to the current charging conditions, the electronic device adjusts the charging conditions by reducing the charging rate and / or raising the battery temperature, reduces the risk of lithium plating of the target battery, so as to reduce the impact on the life of the target battery and extend its endurance.

[0103] In one or more embodiments of the present application, in response to the fact that the target battery will not undergo lithium plating at multiple future moments according to the current charging conditions, the current charging conditions are kept unchanged and the target battery continues to be charged.

[0104] In the related art, the fast charging scheme for the battery includes a segmented charging scheme. For example, based on parameters such as a preset maximum charging duration and maximum charging rate, the charging process is divided into several state of charge intervals, and different charging rates are executed in each state of charge interval. This method needs to select the charging scheme with the best cycle life under the current charging duration condition as the fast charging method, so as to achieve a balance between the battery cycle life and the charging duration. However, this method requires a large number of tests on the battery, takes a lot of test time and resources, and the test time and the number of test batteries increase with the increase of the state of charge intervals, which increases the difficulty of finding a fast charging strategy that does not affect the battery life.

[0105] However, the charging method provided by the embodiments of the present application does not require multi-stage fast charging and does not need to perform a large number of tests on the battery, saving test time. Based on the existing two-electrode system battery, the electronic device predicts the lithium plating state of the battery in the next period of time by collecting battery data, thereby changing charging conditions such as the charging rate and battery temperature of the battery, and reducing the occurrence of lithium plating side reactions.

[0106] It can be understood that the initial charging rate of the target battery is set according to actual needs. In one or more embodiments of the present application, since multiple operations of reducing the charging rate may be performed according to the above steps S101-S104 during one charging process. Therefore, in order to shorten the charging duration, in some embodiments, the target battery starts charging at the maximum charging rate, and at the same time, in combination with the charging method of the present application, the charging conditions are adjusted to reduce the risk of lithium plating of the target battery.

[0107] For step S104, for electrical devices without a battery temperature control component, such as portable electrical devices like headphones, mobile phones, and tablet computers, the charging conditions are adjusted by reducing the charging rate; for some electrical devices with a battery temperature control component, such as energy storage products and electric vehicles, the charging conditions are adjusted by raising the battery temperature through the battery temperature control component, or by reducing the charging rate, or by raising the battery temperature and reducing the charging rate simultaneously.

[0108] In some implementation manners, the charging rate is reduced according to a preset gradient. For example, the charging rate is reduced to 95%, 90%, 85%, or 80% of the current charging rate. In other implementation manners, the battery temperature is raised according to a preset gradient. For example, the battery temperature is raised to 150%, 200%, or 250% of the current battery temperature.

[0109] In other implementation manners, a preset charging rate is reduced based on the current charging rate. For example, the charging rate is reduced by 0.2C, 0.5C, or 1C. In other implementation manners, a preset temperature is raised based on the current battery temperature. For example, the battery temperature is raised by 5°C, 8°C, 10°C, 15°C, or 20°C.

[0110] In one or more embodiments of the present application, by adding an electrochemical mechanism model to an electronic device, the charging process of a target battery is monitored. When it is determined that the target battery will experience lithium plating in the subsequent charging process according to the current charging conditions, the electronic device then adjusts the current charging conditions and controls the target battery to charge according to the adjusted charging conditions to improve or reduce the lithium plating phenomenon of the target battery.

[0111] In one or more embodiments of the present application, the adjusted charging conditions are determined through the following steps:

[0112] A: Reduce the charging rate and / or raise the ambient temperature or battery temperature during charging to obtain the adjusted charging conditions (referred to as standby charging conditions);

[0113] B: Use a pre-constructed electrochemical mechanism model of the target battery (referred to as the first electrochemical mechanism model) to calculate the simulated potential values of the target battery at multiple future times when charging according to the standby charging conditions. The simulated potential values include at least one of the following: the simulated full-cell potential value, the simulated positive electrode potential value, and the simulated negative electrode potential value;

[0114] C: Input the simulated potential values into a pre-trained second potential prediction model to calculate the predicted negative electrode potential values of the target battery at multiple future times when charging according to the standby charging conditions (referred to as the second predicted negative electrode potential values).

[0115] In actual working conditions, it is generally difficult to obtain the true negative electrode potential value of the target battery while collecting the true full battery potential value of the target battery. In the embodiments of the present application, the simulation potential value is used to predict the negative electrode potential value of the target battery, and then it is determined whether lithium plating occurs during the charging of the target battery under standby charging conditions.

[0116] The electrochemical mechanism model is understood as a simulation model of the battery.

[0117] Technicians construct a corresponding electrochemical mechanism model (i.e., the first electrochemical mechanism model) according to the specification parameters such as the size and material of the target battery. In some embodiments of the present application, after constructing the first electrochemical mechanism model, the current voltage, remaining power, and current charging conditions of the target battery are input into the first electrochemical mechanism model for simulation calculation to obtain the potential value of the target battery at a future moment (i.e., the simulation potential value).

[0118] The second potential prediction model is constructed based on the simulation potential values and true negative electrode potential values of the sample battery at multiple third sample moments. The third sample moment is the same moment as the above-mentioned second sample moment, or a different moment. It can be understood that the simulation potential value of the sample battery is calculated using the electrochemical mechanism model of the sample battery (referred to as the second electrochemical mechanism model). Technicians construct a corresponding electrochemical mechanism model (i.e., the second electrochemical mechanism model) according to the specification parameters such as the size and material of the sample battery. The calculation method of the simulation potential values of the sample battery at multiple third sample moments refers to the relevant introduction of calculating the simulation potential value of the target battery at a future moment in the above embodiments. The types of the simulation potential values of the sample battery at multiple third sample moments are the same as those of the simulation potential values of the target battery at multiple future moments. That is, if the simulation potential values of the sample battery at multiple third sample moments include the simulation full battery potential value, then the simulation potential values of the target battery at multiple future moments also include the simulation full battery potential value; if the simulation potential values of the sample battery at multiple third sample moments include the simulation negative electrode potential value, then the simulation potential values of the target battery at multiple future moments also include the simulation negative electrode potential value; if the simulation potential values of the sample battery at multiple third sample moments include the simulation positive electrode potential value, then the simulation potential values of the target battery at multiple future moments also include the simulation positive electrode potential value.

[0119] In an actual scenario, the target battery and the sample battery have the same specifications. Technicians match the target battery and select the same materials such as the positive electrode material, negative electrode material, electrolyte, and separator as the target battery to assemble a three-electrode battery to obtain the sample battery. Correspondingly, the second electrochemical mechanism model and the first electrochemical mechanism model are the same electrochemical mechanism model.

[0120] That is to say, the second potential prediction model reflects the mapping relationship between the simulated potential value of the battery and the true negative electrode potential value at the same moment. Correspondingly, the true negative electrode potential value at the same moment (i.e., the future time period) is predicted by using the simulated potential value in the future time period.

[0121] For example, the simulated potential values of the sample battery at multiple third sample moments are input into the second potential prediction model with the initial structure to obtain the predicted negative electrode potential values of the sample battery at multiple third sample moments. Furthermore, based on the difference between the obtained predicted negative electrode potential values and the true negative electrode potential values of the sample battery at multiple third sample moments, the loss value is calculated, and the model parameters of the second potential prediction model with the initial structure are adjusted based on the loss value until the convergence condition is reached, and the trained second potential prediction model is obtained.

[0122] In order to further improve the robustness of the second potential prediction model to adapt to the state and working environment of the battery, the true negative electrode potential values of the sample battery at multiple third sample moments in a variety of different application scenarios are obtained, and the simulated potential values of the sample battery in these various different application scenarios are calculated by using the second electrochemical mechanism model to train the second potential prediction model with the initial structure. The state and working environment of the battery include: battery aging state, environmental temperature, environmental humidity, charging scheme, irradiation, etc., to obtain a variety of different application scenarios. The charging scheme includes constant current charging, constant current constant voltage charging, multi-stage constant current charging, pulse charging, etc.

[0123] D: Based on the second predicted negative electrode potential value, determine whether lithium plating will occur at the target battery at multiple future moments when charging according to the standby charging condition;

[0124] This step is similar to step S103 above. Refer to the relevant introduction in step S103.

[0125] E: In response to lithium plating occurring at the target battery at multiple future moments when charging according to the standby charging condition, return to execute steps A - E until it is determined that lithium plating will not occur at the target battery at multiple future moments when charging according to the latest standby charging condition, and determine the latest standby charging condition as the adjusted charging condition.

[0126] Based on the above steps A - E, the second potential prediction model predicts the negative electrode potential values of the target battery at multiple future times when charging according to the standby charging conditions based on the simulated potential values at multiple future times, so that after each adjustment of the charging conditions, it is further determined whether lithium plating will still occur at multiple future times when the target battery charges according to the adjusted charging conditions; if lithium plating still occurs, the charging conditions are continuously adjusted, that is, the subsequent adjustment is a further adjustment based on the previous adjustment. For example, the magnitude of the charging current is further reduced, and / or the battery temperature is further increased until it is determined that lithium plating will not occur at multiple future times when the target battery charges according to the latest adjusted charging conditions, so as to reduce the risk of lithium plating occurring in the target battery during the entire charging process and complete the battery charging process as soon as possible.

[0127] For the above step S102, it is implemented in one of the following ways:

[0128] Method 1:

[0129] In this method, for each future time, the negative electrode potential value, positive electrode potential value, and full - cell potential value of the target battery at multiple future times are respectively predicted, and the first predicted negative electrode potential value at this future time is calculated.

[0130] In some embodiments, the first potential prediction model includes three models, which are respectively used to predict the negative electrode potential value, positive electrode potential value, and full - cell potential value. For the convenience of description, the model for predicting the negative electrode potential value is called the first sub - model, the model for predicting the positive electrode potential value is called the second sub - model, and the model for predicting the full - cell potential value is called the third sub - model. The input data of these three sub - models are the same (including the real - state data at multiple historical times), and the output data are different.

[0131] The first sub - model is constructed as follows: It is trained based on the real - state data of the sample battery at multiple first sample times and the real negative electrode potential values at multiple second sample times during the historical charging process. For example, the real - state data of the sample battery at multiple first sample times are input into the first sub - model with an initial structure to obtain the predicted negative electrode potential values of the sample battery at multiple second sample times. Then, based on the difference between the obtained predicted negative electrode potential values and the real negative electrode potential values of the sample battery at multiple second sample times, the loss value is calculated, and the model parameters of the first sub - model with the initial structure are adjusted based on the loss value until the convergence condition is reached, and the trained first sub - model is obtained.

[0132] See Figure 2A and Figure 2B , Figure 2A which is a schematic diagram of the real negative electrode potential value of a battery (referred to as the battery to be utilized) provided by an embodiment of the present application.Figure 2B Schematic diagram of the predicted negative electrode potential value of the battery to be utilized obtained by using the trained first sub-model. Figure 2A And Figure 2B In, the abscissa represents the charging duration, with the unit of second (S), and the ordinate represents the negative electrode potential value, with the unit of volt (V). See Figure 2C , Figure 2C Is Figure 2B The difference between the predicted negative electrode potential value in Figure 2A And the true negative electrode potential value in Figure 2C In, the abscissa represents the charging duration, with the unit of second (S), and the ordinate represents the prediction error, with the unit of volt (V), that is, the difference between the predicted negative electrode potential value and the true negative electrode potential value. It can be seen from Figure 2A , Figure 2B And Figure 2C That the difference between the predicted negative electrode potential value obtained by the first sub-model provided in the embodiment of the present application and the true negative electrode potential value is relatively small, and this difference is generally less than 0.01V. A relatively high-accuracy battery negative electrode potential value is predicted through the first sub-model.

[0133] The second sub-model is constructed as follows: Based on the true state data of the sample battery at multiple first sample moments and the true positive electrode potential values at multiple second sample moments during the historical charging process for training. For example, input the true state data of the sample battery at multiple first sample moments into the second sub-model with an initial structure to obtain the predicted positive electrode potential values of the sample battery at multiple second sample moments. Then, based on the difference between the obtained predicted positive electrode potential values and the true positive electrode potential values of the sample battery at multiple second sample moments, calculate the loss value, and adjust the model parameters of the second sub-model with the initial structure based on the loss value until the convergence condition is reached, and a trained second sub-model is obtained.

[0134] The third sub-model is constructed as follows: Based on the true state data of the sample battery at multiple first sample moments and the true full battery potential values at multiple second sample moments during the historical charging process for training. For example, input the true state data of the sample battery at multiple first sample moments into the third sub-model with an initial structure to obtain the predicted full battery potential values of the sample battery at multiple second sample moments. Then, based on the difference between the obtained predicted full battery potential values and the true full battery potential values of the sample battery at multiple second sample moments, calculate the loss value, and adjust the model parameters of the third sub-model with the initial structure based on the loss value until the convergence condition is reached, and a trained third sub-model is obtained.

[0135] See Figure 3A And Figure 3B , Figure 3AA schematic diagram of the true negative electrode potential value of a battery (referred to as the battery to be utilized) provided by an embodiment of the present application. Figure 3B A schematic diagram of the predicted negative electrode potential value of the battery to be utilized obtained by using the trained second sub-model and third sub-model. Figure 3A And Figure 3B In, the abscissa represents the charging duration, with the unit of second (S), and the ordinate represents the negative electrode potential value, with the unit of volt (V). Refer to Figure 3C , Figure 3C Is Figure 3B The difference between the predicted negative electrode potential value in and Figure 3A The true negative electrode potential value in. Figure 3C In, the abscissa represents the charging duration, with the unit of second (S), and the ordinate represents the prediction error, with the unit of volt (V), that is, the difference between the predicted negative electrode potential value and the true negative electrode potential value. The predicted negative electrode potential value in this embodiment is the difference between the predicted positive electrode potential value obtained by the second sub-model and the predicted full battery potential value obtained by the third sub-model. It can be seen from Figure 3A , Figure 3B And Figure 3C That the difference between the predicted negative electrode potential value obtained by using the second sub-model and the third sub-model provided by the embodiment of the present application and the true negative electrode potential value is small, and this difference is generally less than 0.01V. A relatively accurate positive electrode potential value is predicted by the second sub-model, and a relatively accurate full battery potential value is predicted by the third sub-model.

[0136] In some other embodiments of the present application, the first potential prediction model is a complete model, and the first potential prediction model simultaneously realizes the functions of the first sub-model, the second sub-model, and the third sub-model in the above examples, that is, a single model is used to predict the positive electrode potential value, the negative electrode potential value, and the full battery potential value simultaneously. Correspondingly, the first potential prediction model is constructed as: trained based on the true state data of the sample battery at multiple first sample moments, and the true full battery potential value, true positive electrode potential value, and true negative electrode potential value at multiple second sample moments during the historical charging process.

[0137] For example, the true state data of the sample battery at multiple first sample times is input into the first potential prediction model with the initial structure, and the predicted full battery potential value, predicted positive electrode potential value, and predicted negative electrode potential value (i.e., the third predicted negative electrode potential value) of the sample battery at multiple second sample times are obtained. Furthermore, based on the difference between the predicted full battery potential value and the true full battery potential value of the sample battery at multiple second sample times, the difference between the predicted positive electrode potential value and the true positive electrode potential value of the sample battery at multiple second sample times, and the difference between the predicted negative electrode potential value and the negative electrode potential value of the sample battery at multiple second sample times, the loss value is calculated, and the model parameters of the first potential prediction model with the initial structure are adjusted based on the loss value until the convergence condition is reached, and the trained first potential prediction model is obtained.

[0138] Based on the above embodiments, the negative electrode potential value, positive electrode potential value, and full battery potential value of the target battery at multiple future times are predicted. Furthermore, the first predicted negative electrode potential value at each future time is calculated according to the following formula:

[0139]

[0140] SOLP represents the first predicted negative electrode potential value at this future time, U CTP represents the positive electrode potential value predicted at this future time, U BTP represents the full battery potential value predicted at this future time, U ATP represents the negative electrode potential value predicted at this future time (i.e., the third predicted negative electrode potential value).

[0141] Based on the above processing, the negative electrode potential value, positive electrode potential value, and full battery potential value are predicted respectively. Due to certain deviations in the prediction, according to formula (1), the final negative electrode potential value is comprehensively calculated based on the predicted potential values to reduce the above deviations and further improve the accuracy of the first predicted negative electrode potential value, providing a precise basis for subsequent judgment of whether lithium plating occurs in the target battery.

[0142] Method 2: In this method, only the negative electrode potential value of the target battery at multiple future times is predicted. Correspondingly, for each future time, the negative electrode potential value of the target battery predicted at this future time is used as the first predicted negative electrode potential value at this future time. That is to say, only the first sub-model in the above embodiments is used in this method to obtain the first predicted negative electrode potential values at multiple future times.

[0143] For the above step S102, in one or more embodiments of the present application, the full-cell potential values and charging current values of the target battery at multiple historical moments are input into a pre-trained first potential prediction model to obtain the first predicted negative electrode potential values of the target battery at multiple future moments.

[0144] In one or more embodiments of the present application, in addition to the full-cell potential values and charging current values of the target battery at multiple historical moments, other information is combined to predict the negative electrode potential values of the target battery at multiple future moments, so as to further improve the accuracy of the obtained first predicted negative electrode potential values.

[0145] In some embodiments of the present application, the true state data of the target battery at each historical moment further includes: the temperature of the target battery at that historical moment. Since the negative electrode potential value of the target battery is affected by temperature, therefore, the temperature of the target battery is combined to predict the negative electrode potential value at a future moment, further improving the accuracy of the first predicted negative electrode potential value. It can be understood that when training the first potential prediction model, the temperatures of the sample batteries at multiple first sample moments are also used.

[0146] And / or,

[0147] Using the first electrochemical mechanism model, according to the current charging conditions, calculate the simulated potential values of the target battery at multiple future moments. The simulated potential values include at least one of the following: simulated full-cell potential value, simulated positive electrode potential value, and simulated negative electrode potential value. Input the true state data of the target battery at multiple historical moments and the calculated simulated potential values at multiple future moments into a pre-trained first potential prediction model to obtain the first predicted negative electrode potential values of the target battery at multiple future moments. That is, the data to be processed of the target battery further includes: the simulated potential values of the target battery at multiple future moments according to the current charging conditions.

[0148] It can be understood that for the above step S102, in one or more embodiments of the present application, when three sub-models are used to predict the negative electrode potential value, the positive electrode potential value, and the full-cell potential value respectively, the input data of each sub-model includes the simulated potential value at a future moment.

[0149] Since the first electrochemical mechanism model predicts the true potential value of the target battery at future moments, the simulation potential value obtained by combining the first electrochemical mechanism model is used to further improve the accuracy of the first predicted negative electrode potential value. It can be understood that during the training process of the first potential prediction model, the simulation potential values of the sample battery at multiple second sample moments are also used. The simulation potential values of the sample battery at multiple second sample moments are obtained by performing simulation calculations using the second electrochemical mechanism model, and the calculation method refers to the relevant introduction of calculating the simulation potential value of the target battery at future moments in the above embodiments. The types of the simulation potential values of the sample battery at multiple second sample moments and the simulation potential values of the target battery at multiple future moments are the same. That is, if the simulation potential values of the sample battery at multiple second sample moments include the simulation full-cell potential value, then the simulation potential values of the target battery at multiple future moments also include the simulation full-cell potential value; if the simulation potential values of the sample battery at multiple second sample moments include the simulation negative electrode potential value, then the simulation potential values of the target battery at multiple future moments also include the simulation negative electrode potential value; if the simulation potential values of the sample battery at multiple second sample moments include the simulation positive electrode potential value, then the simulation potential values of the target battery at multiple future moments also include the simulation positive electrode potential value.

[0150] In some embodiments of the present application, based on the true state data of the sample battery at multiple first sample moments during the historical charging process, as well as the true potential values and simulation potential values at multiple second sample moments, the first potential prediction model is trained. For example, for the first sub-model, second sub-model, and third sub-model introduced in the above embodiments, when training each sub-model, the simulation potential values of the sample battery at multiple second sample moments are also input as input data into the sub-model with the initial structure. In this way, the trained sub-model reflects the mapping relationship between the state data of the battery at historical moments, the simulation potential values at future moments, and the true potential values at future moments.

[0151] Similarly, in this case, in order to further improve the robustness of the first potential prediction model to adapt to the state and working environment of the battery, the first electrochemical mechanism model is used to calculate the simulation potential values of the sample battery in the above-mentioned various different application scenarios.

[0152] In one or more embodiments of the present application, the first potential prediction model (including the first sub-model, second sub-model, and third sub-model) and the second potential prediction model mentioned above are models based on deep learning algorithms. For example, the above potential prediction models include one or more of a convolutional neural network, a densely connected network, a recurrent neural network, a long short-term memory neural network, a model with an attention mechanism, or a Transformer network, etc.

[0153] In this way, through the deep learning algorithm model, a high-precision and strong-robust mapping relationship is established between the potential value, current, temperature of the two-electrode full cell, the simulated full cell potential value, the simulated cathode potential value, and the simulated anode potential value of the electrochemical mechanism model, and the lithium plating state of the battery in a future period of time, so as to predict the lithium plating situation of the battery.

[0154] In one or more embodiments of the present application, the first electrochemical mechanism model and the second electrochemical mechanism model mentioned above are electrochemical mechanism models including lithium plating side reaction and thermal field models. For example, they are single particle models, quasi-two-dimensional models, or multi-dimensional multi-field electrochemical models, etc.

[0155] In one or more embodiments of the present application, for a target battery that has been used in an actual scenario, after it has been used for a period of time, it is regarded as a sample battery, and relevant data is collected in the manner described in the above embodiments. And based on the collected data, incremental training is performed on the first potential prediction model (including the first sub-model, the second sub-model, and the third sub-model) and the second potential prediction model mentioned in the above embodiments. In this way, using the data of the battery under different actual working conditions, the models involved in the present application are updated, and the accuracy of model prediction is continuously improved.

[0156] See Figure 4 , Figure 4 which is a flowchart of another charging method provided by the embodiment of the present application.

[0157] This flowchart is divided into two major parts: an offline training part and an online application part.

[0158] The offline training part includes the following steps:

[0159] Step 1: Obtain three-electrode battery data. Corresponding to the above embodiments, obtain the real state data of the sample battery at multiple first sample times, and the real potential values at multiple second sample times. And obtain the real anode potential value of the sample battery at multiple third sample times.

[0160] Step 2: Construct an electrochemical mechanism model including lithium plating side reaction and thermal field model and generate simulation data. Corresponding to the above embodiments, construct the second electrochemical mechanism model, and calculate the simulated potential values of the sample battery at multiple second sample times and third sample times.

[0161] Step 3: Split the three - electrode battery data and simulation data into data segments. That is, split the data obtained in Step 1 and Step 2 according to the adjustment interval in the above - mentioned embodiment to obtain the input data and label data for training the first potential prediction model (including the first sub - model, the second sub - model, and the third sub - model) and the second potential prediction model. The true state data at the first sample time and the simulated potential value at the second sample time are used as input data, and the true potential value at the second sample time is used as the corresponding label data; the simulated potential value at the third sample time is used as input data, and the true negative electrode potential value at the third sample time is used as the corresponding label data.

[0162] Step 4: Deep - learning prediction of the future full - cell potential value, deep - learning prediction of the future negative - electrode potential value, deep - learning prediction of the future positive - electrode potential value, and deep - learning reconstruction of the current negative - electrode potential value.

[0163] Deep - learning prediction of the future full - cell potential value, that is, input the input data (i.e., the true state data at the first sample time and the simulated potential value at the second sample time) for training the third sub - model in the split data segments into the third sub - model with the initial structure to obtain the future full - cell potential value (i.e., the predicted full - cell potential value at the second sample time), and use the obtained future full - cell potential value and the label data (i.e., the true full - cell potential value at the second sample time) to adjust the model parameters of the third sub - model with the initial structure to obtain the trained third sub - model.

[0164] Deep - learning prediction of the future negative - electrode potential value, that is, input the input data (i.e., the true state data at the first sample time and the simulated potential value at the second sample time) for training the first sub - model in the split data segments into the first sub - model with the initial structure to obtain the future negative - electrode potential value (i.e., the predicted negative - electrode potential value at the second sample time), and use the obtained future negative - electrode potential value and the label data (i.e., the true negative - electrode potential value at the second sample time) to adjust the model parameters of the first sub - model with the initial structure to obtain the trained first sub - model.

[0165] Deep - learning prediction of the future positive - electrode potential value, that is, input the input data (i.e., the true state data at the first sample time and the simulated potential value at the second sample time) for training the second sub - model in the split data segments into the second sub - model with the initial structure to obtain the future positive - electrode potential value (i.e., the predicted positive - electrode potential value at the second sample time), and use the obtained future positive - electrode potential value and the label data (i.e., the true positive - electrode potential value at the second sample time) to adjust the model parameters of the second sub - model with the initial structure to obtain the trained second sub - model.

[0166] The deep learning reconstructs the current negative electrode potential value, that is, the input data for training the second potential prediction model in the segmented data segment (i.e., the simulated potential value at the third sample moment) is input into the second potential prediction model with the initial structure to obtain the negative electrode potential value (i.e., the predicted negative electrode potential value at the third sample moment), and the obtained negative electrode potential value and the label data (i.e., the true negative electrode potential value at the third sample moment) are used to adjust the model parameters of the second potential prediction model with the initial structure to obtain the trained second potential prediction model.

[0167] The online application part includes the following steps:

[0168] Step Five: Collect the actual battery operation data in the scenario and combine it with the simulation data. It can be understood that the battery here is the target battery in the above text.

[0169] Step Six: Predict the future positive electrode potential value, negative electrode potential value, and full battery potential value through the trained deep learning algorithm.

[0170] Corresponding to the data to be processed of the target battery in the above embodiment, using the first potential prediction model, the predicted negative electrode potential value, predicted positive electrode potential value, and predicted full battery potential value of the target battery at multiple future moments are obtained.

[0171] Step Seven: Judge the comprehensive lithium plating state in the future for a period of time. Corresponding to the above embodiment, based on the predicted negative electrode potential value, predicted positive electrode potential value, and predicted full battery potential value of the target battery at multiple future moments, calculate the first predicted negative electrode potential value of the target battery at multiple future moments, and calculate the average level of multiple first predicted negative electrode potential values to judge whether lithium plating will occur in the target battery at multiple future moments.

[0172] Step Eight: Optimize the charging rate and environmental conditions. Corresponding to the relevant introduction of Step A in the above embodiment.

[0173] Step Nine: Optimize the search. Corresponding to Steps B - E in the above embodiment.

[0174] Step Ten: Accumulate the actual operation data of the target battery under one or more scenarios.

[0175] Step Eleven: Update the deep learning algorithm online.

[0176] Steps Ten and Eleven correspond to the part of incremental training of the model in the above embodiment.

[0177] See Figure 5 , Figure 5 is the flow chart of determining the charging conditions at future moments in the charging method provided by the embodiment of the present application, including the following steps:

[0178] S501: Calculate the future comprehensive lithium plating state.

[0179] That is, in the above embodiments, the first predicted negative electrode potential value at each future moment is calculated according to formula (1), and the average level of multiple first predicted negative electrode potential values is calculated.

[0180] S502: Determine whether the comprehensive lithium plating state is greater than the system preset value. If it is greater, execute step S503; if it is not greater, execute step S504. That is, in the above embodiments, it is determined whether the average level is greater than the preset threshold.

[0181] S503: Continue to charge at the current rate and temperature conditions. That is, continue to charge the target battery while maintaining the current charging conditions and temperature conditions. It can be understood that the temperature conditions include the ambient temperature and / or the temperature of the target battery.

[0182] S504: Determine that lithium plating will occur in the future, and adjust the charging rate and temperature conditions. That is, step A in the above embodiments.

[0183] S505: Calculate the lithium plating state after adjusting the charging rate and temperature conditions. That is, steps B - D in the above embodiments.

[0184] S506: Determine whether the comprehensive lithium plating state is greater than the system preset value. If it is greater, execute step S507; if it is not greater, execute steps S504 and S505.

[0185] S507: Charge the target battery with the adjusted charging rate and temperature conditions.

[0186] Steps S506 and S507 correspond to step E in the above embodiments. That is, through multiple adjustments, the charging conditions that will not cause lithium plating at future moments determined based on the simulated potential values are found, and the target battery is charged according to the found charging conditions.

[0187] Based on the same inventive concept, the embodiments of the present application also provide a charging device. See Figure 6 , Figure 6The structural diagram of a charging device provided by an embodiment of this application. The device includes: a to-be-processed data acquisition module 601, which acquires the to-be-processed data of the target battery. The to-be-processed data includes: during the current charging process, the true state data of the target battery at multiple historical moments within the historical time period before the current moment. The true state data includes: the full battery potential value and the charging current value; a prediction module 602, which processes the to-be-processed data by using a pre-trained first potential prediction model to obtain the negative electrode potential values of the target battery at multiple future moments within the future time period after the current moment during the current charging process, as the first predicted negative electrode potential values. The first potential prediction model is constructed as follows: it is trained based on the true state data of the sample battery at multiple first sample moments during the historical charging process and the true potential values at multiple second sample moments after the multiple first sample moments; a judgment module 603, which judges whether lithium plating will occur in the target battery at multiple future moments according to the current charging conditions based on the obtained multiple first predicted negative electrode potential values; an adjustment module 604, which responds to the situation that lithium plating will occur in the target battery at multiple future moments when charging according to the current charging conditions, and adjusts the charging conditions; the adjustment methods include: reducing the charging rate, and / or raising the battery temperature; a first control module 605, which charges the target battery according to the adjusted charging conditions.

[0188] Since there is a mapping relationship between the negative electrode potential value of the target battery at future moments and the true state data at historical moments, the true state data at historical moments is used to predict the negative electrode potential value at future moments. In addition, during the use of the target battery, the negative electrode potential value indicates whether lithium plating has occurred in the battery. Therefore, the first predicted negative electrode potential values at multiple future moments are used to judge whether lithium plating will occur in the target battery at multiple future moments if the target battery continues to be charged according to the current charging conditions. Correspondingly, if it is determined that lithium plating will occur in the target battery at multiple future moments according to the current charging conditions, the charging rate is reduced, and / or the battery temperature is raised to reduce the risk of charging lithium plating in the target battery, so as to reduce the impact on the life of the target battery and extend its endurance ability.

[0189] In one or more embodiments of the present application, the above adjustment module 604 obtains a standby charging condition by reducing the charging rate and / or increasing the battery temperature. Using a pre-constructed first electrochemical mechanism model of the target battery, the simulated potential values of the target battery at multiple future times when charging according to the standby charging condition are calculated. The simulated potential values include at least one of the following: the simulated full-cell potential value, the simulated positive electrode potential value, and the simulated negative electrode potential value. Then, the calculated simulated potential values are input into a pre-trained second potential prediction model to obtain the negative electrode potential values of the target battery at multiple future times when charging according to the standby charging condition, which are used as the second predicted negative electrode potential values. The second potential prediction model is constructed based on the simulated potential values and the true negative electrode potential values of the sample battery at multiple third sample times. The simulated potential values at the third sample times are calculated using a pre-constructed second electrochemical mechanism model of the sample battery. Based on the obtained second predicted negative electrode potential values, it is determined whether lithium plating will occur at multiple future times when the target battery is charged according to the standby charging condition. In response to the occurrence of lithium plating at multiple future times when the target battery is charged according to the standby charging condition, the charging rate is further reduced and / or the battery temperature is increased to obtain a new standby charging condition until it is determined that lithium plating will not occur at multiple future times when the target battery is charged according to the latest standby charging condition, and the latest standby charging condition is determined as the adjusted charging condition.

[0190] By adding an electrochemical mechanism model, the charging process of the target battery is monitored. When it is determined that lithium plating will occur in the subsequent charging process when the target battery is charged according to the current charging condition, the current charging condition is adjusted to control the target battery to charge according to the adjusted charging condition, further reducing the risk of lithium plating occurring in the entire charging process of the target battery and completing the charging process of the target battery as soon as possible.

[0191] In one or more embodiments of the present application, the above prediction module 602 inputs the data to be processed into a pre-trained first potential prediction model to obtain the predicted full-cell potential values, predicted positive electrode potential values, and third predicted negative electrode potential values at multiple future times within a future time period after the current time during the current charging process of the target battery. For each future time, based on the obtained predicted full-cell potential values, predicted positive electrode potential values, and third predicted negative electrode potential values, a first predicted negative electrode potential value at this future time is calculated using a preset formula:

[0192]

[0193] SOLP represents the first predicted negative electrode potential value at this future time, U CTP represents the predicted positive electrode potential value at this future time, U BTP represents the predicted full-cell potential value at this future time, U ATPRepresents the third predicted negative electrode potential value at that future moment.

[0194] Predict the negative electrode potential value, the positive electrode potential value, and the full cell potential value respectively. Due to certain deviations in the prediction itself, according to a preset formula, based on the predicted potential values, the final negative electrode potential value is comprehensively calculated to reduce the above deviations and further improve the accuracy of the first predicted negative electrode potential value, providing a precise basis for determining whether the target battery will experience lithium plating in the subsequent process.

[0195] In one or more embodiments of the present application, the first potential prediction model includes a first sub-model, a second sub-model, and a third sub-model. The above prediction module 602 inputs the data to be processed into the pre-trained first sub-model to obtain the negative electrode potential values at multiple future moments within a future time period after the current moment during the current charging process of the target battery, as the third predicted negative electrode potential values. The first sub-model is constructed by training based on the real state data of the sample battery at multiple first sample moments and the real negative electrode potential values at multiple second sample moments during the historical charging process. Input the data to be processed into the pre-trained second sub-model to obtain the positive electrode potential values at multiple future moments within a future time period after the current moment during the current charging process of the target battery, as the predicted positive electrode potential values. The second sub-model is constructed by training based on the real state data of the sample battery at multiple first sample moments and the real positive electrode potential values at multiple second sample moments during the historical charging process. Input the data to be processed into the pre-trained third sub-model to obtain the full cell potential values at multiple future moments within a future time period after the current moment during the current charging process of the target battery, as the predicted full cell potential values. The third sub-model is constructed by training based on the real state data of the sample battery at multiple first sample moments and the real full cell potential values at multiple second sample moments during the historical charging process.

[0196] Predict the full cell potential value, the positive electrode potential value, and the negative electrode potential value respectively through three different models to further improve the accuracy of the predicted first predicted negative electrode potential value, providing a precise basis for determining whether the target battery will experience lithium plating in the subsequent process.

[0197] In one or more embodiments of the present application, the above judgment module 603 removes the maximum value and the minimum value from multiple first predicted negative electrode potential values and calculates the average value of the remaining predicted negative electrode potential values. If the calculated average value is greater than the preset threshold, it is determined that according to the current charging conditions, the target battery will not experience lithium plating at multiple future moments; if the calculated average value is not greater than the preset threshold, it is determined that according to the current charging conditions, the target battery will experience lithium plating at multiple future moments.

[0198] In one or more embodiments of the present application, the first potential prediction model includes at least one of the following: convolutional neural network, densely connected network, recurrent neural network, long short-term memory neural network, model with attention mechanism, and Transformer network; and / or, the second potential prediction model includes at least one of the following: convolutional neural network, densely connected network, recurrent neural network, long short-term memory neural network, model with attention mechanism, and Transformer network.

[0199] Based on the above networks, the accuracy of predicting the first negative electrode potential value is further improved, providing a precise basis for subsequent determination of whether lithium plating will occur in the target battery.

[0200] In one or more embodiments of the present application, the first electrochemical mechanism model is: single particle model, quasi-two-dimensional model, or multi-dimensional multi-field electrochemical model; and / or, the second electrochemical mechanism model is: single particle model, quasi-two-dimensional model, or multi-dimensional multi-field electrochemical model.

[0201] Based on the above electrochemical models, the accuracy of the obtained simulated potential value is improved, so as to further improve the accuracy of predicting the first negative electrode potential value, providing a precise basis for subsequent determination of whether lithium plating will occur in the target battery.

[0202] In one or more embodiments of the present application, the real state data of the target battery further includes: battery temperature; and / or, the data to be processed includes: simulated potential values of the target battery at multiple future moments according to the current charging conditions. The first potential prediction model is constructed by training based on the real state data of the sample battery at multiple first sample moments during the historical charging process, and the real potential values and simulated potential values at multiple second sample moments after the multiple first sample moments. The simulated potential values at the multiple second sample moments are calculated using the second electrochemical mechanism model of the pre-constructed sample battery.

[0203] Since the negative electrode potential value of the target battery is affected by temperature, therefore, the negative electrode potential value at future moments is predicted by combining the target battery temperature, further improving the accuracy of predicting the first negative electrode potential value. Since the first electrochemical mechanism model predicts the real potential value of the target battery at future moments, therefore, the accuracy of predicting the first negative electrode potential value is further improved by combining the simulated potential value obtained by the first electrochemical mechanism model.

[0204] In one or more embodiments of the present application, the above data to be processed acquisition module 601 acquires the data to be processed when any one of the preset multiple adjustment moments is reached. In some embodiments of the present application, the adjustment interval between every two adjacent adjustment moments is not greater than the duration of the future time period.

[0205] After determining the charging conditions for a future time period each time, control the target battery to charge according to the determined charging conditions within the future time period. Subsequently, before reaching the last moment of the future time period, determine the charging conditions for a new future time period again at the adjustment interval. In this way, determine the charging conditions for a subsequent period of time in real time according to the latest state of the target battery, further reducing the impact on the battery life and ensuring its endurance ability.

[0206] In one or more embodiments of the present application, the above charging device further includes: a second control module. If it is determined that the target battery will not undergo lithium plating at multiple future moments according to the current charging conditions, keep the current charging conditions unchanged and continue to charge the target battery.

[0207] Keep the current charging conditions unchanged and continue to charge the target battery to reduce the risk of lithium plating occurring in the target battery during the entire charging process, and at the same time complete the charging process of the target battery as soon as possible.

[0208] Based on the same inventive concept, an embodiment of the present application further provides an electronic device, and the electronic device executes the charging method described in any one of the above embodiments.

[0209] Based on the same inventive concept, an embodiment of the present application further provides a battery pack, including a battery module and the electronic device described in the above embodiment.

[0210] Based on the same inventive concept, an embodiment of the present application further provides an energy storage product, and the energy storage product includes a battery module and an electronic device, and the electronic device is used to execute the charging method described in any one of the above embodiments.

[0211] Based on the same inventive concept, in another embodiment provided by the present application, a computer-readable storage medium is further provided. A computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the steps of any of the above charging methods are implemented.

[0212] In another embodiment provided by the present application, a computer program product including instructions is further provided. When it runs on a computer, it causes the computer to execute any of the charging methods in the above embodiments.

[0213] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0214] It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including", or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements but also other elements that are not explicitly listed, or elements that are inherent to such process, method, article, or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article, or device that includes the element.

[0215] Each embodiment in this specification is described in a related manner. The same or similar parts among the embodiments can be referred to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for electronic devices, battery packs, charging devices, energy storage products, computer-readable storage media, and computer program products, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0216] The above are only the preferred embodiments of the present application and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present application are all included in the protection scope of the present application.

Claims

1. A charging method, characterized in that: The method comprises: Obtaining the data to be processed of the target battery; wherein the data to be processed includes: the real state data of the target battery at multiple historical moments in the historical time period before the current moment during the current charging process, and the real state data includes: the full battery potential value and the charging current value; The data to be processed is processed using a pre-trained first potential prediction model to obtain negative electrode potential values ​​of the target battery at multiple future moments in a future time period after the current moment during this charging process as first predicted negative electrode potential values; The first potential prediction model is constructed as follows: the model is trained based on the real state data of the sample battery at multiple first sample moments in the historical charging process and the real potential values ​​at multiple second sample moments after the multiple first sample moments; Based on the obtained multiple first predicted negative electrode potential values, determining whether lithium deposition will occur in the target battery at the multiple future moments according to the current charging condition; In response to the target battery being charged according to the current charging condition, lithium plating will occur at the multiple future moments, adjusting the charging condition, and charging the target battery according to the adjusted charging condition; The adjustment methods include: reducing the charging rate, and / or increasing the battery temperature.

2. The method according to claim 1, characterized in that The adjusting the charging condition comprises: Reduce the charge rate and / or increase the battery temperature to obtain standby charging conditions; Utilizing the pre-constructed first electrochemical mechanism model of the target battery, calculating the simulated potential values ​​of the target battery at the multiple future moments when it is charged according to the backup charging condition; wherein the simulated potential values ​​at the multiple future moments include at least one of the following: a simulated full battery potential value, a simulated positive electrode potential value, and a simulated negative electrode potential value; Inputting the simulated potential values ​​at the multiple future moments into a pre-trained second potential prediction model, obtaining the negative electrode potential values ​​of the target battery when charged according to the backup charging condition at the multiple future moments as second predicted negative electrode potential values; The second potential prediction model is constructed as follows: the second potential prediction model is trained based on the simulated potential values ​​and the real negative electrode potential values ​​of the sample battery at multiple third sample moments, and the simulated potential values ​​at the multiple third sample moments are calculated using the pre-constructed second electrochemical mechanism model of the sample battery; Based on the second predicted negative electrode potential value, determining whether lithium deposition will occur at the plurality of future moments when the target battery is charged according to the backup charging condition; In response to the target battery being charged according to the backup charging condition, lithium plating will occur at the multiple future moments, returning to execute the steps of reducing the charging rate and / or increasing the battery temperature to obtain the backup charging condition, until it is determined that the target battery is charged according to the latest backup charging condition, lithium plating will not occur at the multiple future moments, and the latest backup charging condition is determined as the adjusted charging condition.

3. The method according to claim 1, characterized in that The method of using the pre-trained first potential prediction model to process the data to be processed to obtain the negative electrode potential values ​​of the target battery at multiple future moments in a future time period after the current moment during the current charging process as the first predicted negative electrode potential value includes: Input the data to be processed into a pre-trained first potential prediction model to obtain the full battery potential value, the positive electrode potential value and the negative electrode potential value of the target battery at multiple future moments in a future time period after the current moment during this charging process, as the predicted full battery potential value, the predicted positive electrode potential value and the third predicted negative electrode potential value, respectively; For each future moment, based on the predicted full battery potential value, the predicted positive electrode potential value and the third predicted negative electrode potential value, a preset formula is used to calculate the first predicted negative electrode potential value at the future moment; wherein the preset formula is: SOLP represents the first predicted negative electrode potential value at the future time, U CTP represents the predicted positive electrode potential value at the future time, U BTP represents the predicted full battery potential value at that future moment, U ATP Indicates the third predicted negative electrode potential value at the future time.

4. The method according to claim 3, characterized in that The first potential prediction model includes a first sub-model, a second sub-model and a third sub-model; The data to be processed is input into a pre-trained first potential prediction model to obtain the full battery potential value, the positive electrode potential value and the negative electrode potential value of the target battery at multiple future moments in a future time period after the current moment in this charging process, which are used as the predicted full battery potential value, the predicted positive electrode potential value and the third predicted negative electrode potential value, respectively, including: Input the data to be processed into a pre-trained first sub-model, and obtain the negative electrode potential values ​​of the target battery at multiple future moments in a future time period after the current moment in this charging process as the third predicted negative electrode potential value, wherein the first sub-model is constructed as follows: based on the real state data of the sample battery at multiple first sample moments and the real negative electrode potential values ​​at multiple second sample moments in the historical charging process; Input the data to be processed into a pre-trained second sub-model, and obtain the positive electrode potential values ​​of the target battery at multiple future moments in a future time period after the current moment in this charging process as the predicted positive electrode potential value, wherein the second sub-model is constructed as follows: based on the real state data of the sample battery at multiple first sample moments and the real positive electrode potential values ​​at multiple second sample moments in the historical charging process; Inputting the data to be processed into a pre-trained third sub-model to obtain the full-battery potential values ​​of the target battery at multiple future moments in a future time period after the current moment during this charging process as predicted full-battery potential values; The third sub-model is constructed as follows: the model is trained based on the real state data of the sample battery at multiple first sample moments and the real full-battery potential values ​​at multiple second sample moments in the historical charging process.

5. The method according to claim 1, characterized in that The determining, based on the obtained plurality of first predicted negative electrode potential values, whether lithium deposition will occur in the target battery at the plurality of future moments according to the current charging condition, comprises: an average value of the remaining predicted negative electrode potential values ​​excluding the maximum value and the minimum value among the plurality of first predicted negative electrode potential values ​​calculated; In response to the average value being greater than a preset threshold, determining that according to the current charging condition, lithium plating will not occur in the target battery at the multiple future moments; In response to the average value being not greater than a preset threshold, it is determined that according to the current charging condition, lithium plating will occur in the target battery at the multiple future moments.

6. The method according to claim 2, characterized in that The first potential prediction model includes at least one of the following: a convolutional neural network, a densely connected network, a recurrent neural network, a long short-term memory neural network, a model of an attention mechanism, and a Transformer network; and / or, The second potential prediction model includes at least one of the following: a convolutional neural network, a densely connected network, a recurrent neural network, a long short-term memory neural network, a model of an attention mechanism, and a Transformer network.

7. The method according to claim 2, characterized in that The first electrochemical mechanism model is: a single particle model, a quasi-two-dimensional model, or a multi-dimensional multi-field electrochemical model; and / or, The second electrochemical mechanism model is: a single particle model, a quasi-two-dimensional model, or a multi-dimensional multi-field electrochemical model.

8. The method according to any one of claims 1 to 7, characterized in that: The real status data also includes: battery temperature; and / or, The data to be processed also includes: simulated potential values ​​of the target battery at the multiple future moments according to the current charging condition; The first potential prediction model is constructed as follows: the model is trained based on the real state data of the sample battery at multiple first sample moments in the historical charging process, and the real potential values ​​and the simulated potential values ​​at multiple second sample moments after the multiple first sample moments; The simulated potential values ​​at the plurality of second sample moments are calculated using a pre-constructed second electrochemical mechanism model of the sample battery.

9. The method according to any one of claims 1 to 7, characterized in that: The step of obtaining the data to be processed of the target battery includes: When any one of the preset multiple adjustment times is reached, the data to be processed of the target battery is obtained; The adjustment interval between every two adjacent adjustment moments is not greater than the duration of the future time period.

10. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: In response to the target battery being charged according to the current charging condition, lithium plating will not occur at the multiple future moments, the current charging condition is maintained unchanged, and the target battery continues to be charged.

11. An electronic device, characterized in that: The electronic device is used to execute the method according to any one of claims 1 to 10.

12. A battery pack, characterized in that: The battery pack comprises a battery module and the electronic device as claimed in claim 11.

13. An energy storage product, characterized in that: The energy storage product comprises a battery module and the electronic device as claimed in claim 11.

Images