Power battery fast charging control method, device and equipment and storage medium

Abstract

Embodiments of the present disclosure provide a power battery fast charging control method, device, equipment and storage medium. The power battery fast charging control method comprises: acquiring real-time characteristic parameters of a power battery, the real-time characteristic parameters comprising at least two of a real-time battery cell temperature and a real-time state of charge, a real-time working characteristic parameter of a thermal management component, and a real-time environmental characteristic parameter of an environment in which the power battery is located; processing the real-time characteristic parameters using a pre-trained reinforcement learning model to generate a first fast charging control instruction, the first fast charging control instruction comprising a first charging current control instruction for controlling a fast charging current and / or a first thermal management control instruction for controlling the thermal management component. Because the reinforcement learning model is trained according to multiple sets of sample data and corresponding fast charging reward value scores, the power battery control method provided by the embodiments of the present disclosure does not need to construct a control strategy lookup table in order to achieve accurate control of battery fast charging, thereby reducing the workload in the design and testing stages of the power battery.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of power batteries, and in particular to a power battery fast charging control method, device, equipment and storage medium. BACKGROUND

[0002] In the process of fast charging of a power battery, the charging current should be as large as possible, the temperature of the battery cell should be maintained within a reasonable range, and the power consumption caused by thermal management should be reduced as much as possible.

[0003] In related technologies, the fast charging control method for a power battery is mainly a rule-based control method. With the rule-based control method, a pre-set control strategy lookup table is searched according to the state of charge of the power battery, the temperature of the battery cell, the working capacity of the thermal management system and the environmental temperature, and a reasonable charging control strategy is determined. The charging control strategy includes the charging current size and the thermal management control strategy.

[0004] However, with the rule-based power battery fast charging control method, in order to realize accurate control of battery fast charging, a large amount of testing work needs to be carried out during the design and testing of the power battery, and a control strategy lookup table needs to be constructed. Moreover, as the performance of the battery cell and the thermal management system degrades, the aforementioned control strategy lookup table may not be well matched with the actual characteristics of the power battery, and in this case, the cost of reasonably modifying the control strategy lookup table is high. SUMMARY

[0005] To solve the above technical problems, the present disclosure provides a power battery fast charging control method, device, equipment and storage medium.

[0006] In a first aspect, the embodiments of the present disclosure provide a power battery fast charging control method, comprising:

[0007] obtaining real-time characteristic parameters of a power battery, the real-time characteristic parameters including at least two of a battery cell temperature and a real-time state of charge, a real-time working characteristic parameter of a thermal management component, and a real-time environmental characteristic parameter of an environment in which the power battery is located;

[0008] processing the real-time characteristic parameters by using a pre-trained reinforcement learning model to generate a first fast charging control instruction, the first fast charging control instruction including a first charging current control instruction for controlling a fast charging current and / or a first thermal management control instruction for controlling the thermal management component;

[0009] The reinforcement learning model is trained based on multiple sets of sample data and corresponding fast charging reward value scores of the sample data, and each set of the sample data includes at least two of a sample charging current, a sample battery temperature, a sample state of charge, a sample working characteristic parameter of a thermal management component, and a sample environmental characteristic parameter of an environment in which a fast charging process is located.

[0010] Optionally, the real-time characteristic parameter includes a real-time battery temperature.

[0011] Before the real-time characteristic parameter is processed by using the pre-trained reinforcement learning model, the method further includes:

[0012] determining whether the real-time battery temperature is within a preset safety threshold;

[0013] In a case where the real-time battery temperature is within the preset safety threshold, the operation of processing the real-time characteristic parameter by using the pre-trained reinforcement learning model to generate a first fast charging control instruction is performed.

[0014] Optionally, the sample data includes simulation data obtained by simulation based on a pre-constructed battery fast charging physical model, the battery fast charging physical model includes a battery temperature sub-model, a state of charge sub-model and an update strategy sub-model, the battery temperature sub-model, the state of charge sub-model and the update strategy sub-model are constructed based on multiple sets of historical data, and the historical data includes at least two of a historical charging current, a historical battery temperature, a historical state of charge, a historical working characteristic parameter of a thermal management component and a historical environmental characteristic parameter of an environment in which a fast charging process is located.

[0015] Optionally, the method further includes:

[0016] calculating a fast charging reward value score after fast charging control based on the first fast charging control instruction;

[0017] In a case where the fast charging reward value scores corresponding to a preset number of first fast charging control instructions are less than a preset score, the reinforcement learning model is retrained.

[0018] In a second aspect, the embodiments of the present disclosure provide a reinforcement learning model training method for fast charging of a power battery, including:

[0019] obtaining sample data and obtaining a fast charging reward value score corresponding to the sample data, the sample data including at least two of a sample charging current, a sample battery temperature, a sample state of charge, a sample working characteristic parameter of a thermal management component and a sample environmental characteristic parameter of an environment in which a fast charging process is located;

[0020] training a reinforcement learning model by using the sample data and the corresponding fast charging reward value scores.

[0021] Optionally, the obtaining the sample data comprises:

[0022] constructing a battery fast charging physical model according to a plurality of sets of historical data, wherein the battery fast charging physical model comprises a battery cell temperature sub-model, a state of charge sub-model, and an update strategy sub-model; the historical data comprises at least two of historical charging currents, historical battery cell temperatures, historical states of charge, historical working characteristic parameters of thermal management components, and historical environmental characteristic parameters of environments in which fast charging processes are located;

[0023] performing data simulation by using the battery fast charging physical model to generate a plurality of sets of simulation data, wherein the simulation data comprises at least one of a simulation-before battery cell temperature, a simulation-before state of charge, a simulation charging current, a simulation working characteristic parameter of a thermal management component, a simulation environmental characteristic parameter, a simulation-after battery cell temperature, and a simulation-after state of charge;

[0024] using the simulation data as the sample data.

[0025] Optionally, the fast charging return value score corresponding to the sample data comprises:

[0026] calculating a state of charge change value according to the simulation-before state of charge and the simulation-after state of charge, and calculating a simulation energy consumption value of the thermal management component according to the simulation working characteristic parameter of the thermal management component;

[0027] determining value scores corresponding to the state of charge change value, the simulation energy consumption value, and the simulation-after battery cell temperature respectively based on a pre-set scoring rule;

[0028] calculating a sum of the value scores corresponding to the state of charge change value, the simulation energy consumption value, and the simulation-after battery cell temperature to obtain the fast charging return value score.

[0029] Optionally, the obtaining the sample data comprises:

[0030] obtaining a plurality of sets of historical data, wherein the historical data comprises at least two of historical charging currents, historical battery cell temperatures, historical states of charge, historical working characteristic parameters of thermal management components, and historical environmental characteristic parameters of environments in which fast charging processes are located;

[0031] using the historical data as the sample data.

[0032] In a third aspect, the embodiments of the present disclosure provide a power battery fast charging control device, comprising:

[0033] a real-time state acquisition unit configured to acquire real-time characteristic parameters of the power battery, the real-time characteristic parameters including at least two of a real-time cell temperature and a real-time state of charge, a real-time working characteristic parameter of the thermal management component, and a real-time environmental characteristic parameter of an environment in which the power battery is located;

[0034] a control instruction generation unit configured to process the real-time characteristic parameters by using a pre-trained reinforcement learning model to generate a first fast-charging control instruction, the first fast-charging control instruction including a first charging current control instruction for controlling a fast-charging current and / or a first thermal management control instruction for controlling the thermal management component;

[0035] The reinforcement learning model is trained based on a plurality of sets of sample data and corresponding fast-charging reward value scores of the sample data. Each set of the sample data includes at least two of a sample charging current, a sample cell temperature, a sample state of charge, a sample working characteristic parameter of the thermal management component, and a sample environmental characteristic parameter of an environment in which a fast-charging process is located.

[0036] In a fourth aspect, an embodiment of the present disclosure provides a reinforcement learning model training apparatus for fast charging of a power battery, including:

[0037] a sample acquisition unit configured to acquire sample data and corresponding fast-charging reward value scores of the sample data, the sample data including at least two of a sample charging current, a sample cell temperature, a sample state of charge, a sample working characteristic parameter of the thermal management component, and a sample environmental characteristic parameter of an environment in which a fast-charging process is located;

[0038] a model training unit configured to train a reinforcement learning model by using the sample data and the corresponding fast-charging reward value scores.

[0039] In a fifth aspect, an embodiment of the present disclosure provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the computer program implements the power battery fast-charging control method or the reinforcement learning model training method for fast charging of a power battery.

[0040] In a fifth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor,

[0041] the power battery fast-charging control method or the reinforcement learning model training method for fast charging of a power battery.

[0042] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0043] The solution provided in this disclosure uses a pre-trained reinforcement learning model to process at least two of the acquired real-time cell temperature, real-time state of charge, real-time operating characteristic parameters, and real-time environmental characteristic parameters to generate a first fast-charging control command. Because the reinforcement learning model is trained based on multiple sets of sample data and corresponding fast-charging reward value scores, the power battery control method provided in this disclosure eliminates the need to construct a control strategy lookup table for precise fast-charging control, reducing the workload in the power battery design and testing phases. Furthermore, the fast-charging control method provided in this disclosure allows for retraining the reinforcement learning model using newer sample data when the performance of the power battery cells and thermal management system degrades, ensuring that the control strategy for the power battery matches its actual characteristics. Moreover, the cost of retraining the reinforcement learning model is relatively low. Attached Figure Description

[0044] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0045] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without any creative effort, wherein:

[0046] Figure 1 This is a flowchart of a power battery fast charging control method provided in an embodiment of this disclosure;

[0047] Figure 2 This is a flowchart of a power battery fast charging control method provided in some other embodiments of this disclosure;

[0048] Figure 3 This is a flowchart of a reinforcement learning model training method for fast charging of power batteries provided in an embodiment of this disclosure;

[0049] Figure 4 This is a schematic diagram of the structure of the power battery fast charging control device provided in the embodiments of this disclosure;

[0050] Figure 5 This is a schematic diagram of the reinforcement learning model training device for fast charging of power batteries provided in this embodiment of the disclosure;

[0051] Figure 6 This is a schematic diagram of the structure of a computing device provided in an embodiment of this disclosure. Detailed Implementation

[0052] Embodiments of the present disclosure will be described herein below with reference to the drawings. Although certain embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be implemented in various forms and should not be interpreted as being limited to the embodiments set forth herein, but rather these embodiments are provided so as to more thoroughly and completely understand the present disclosure. It is understood that the drawings and embodiments of the present disclosure are merely for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0053] The term "comprising" and variations thereof as used herein are open-ended, that is "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment". The term "another embodiment" means "at least one additional embodiment". The term "some embodiments" means "at least some embodiments". Related terms are defined as follows. It is to be noted that the terms "first", "second", and the like used in the present disclosure merely distinguish different apparatuses, modules, or units, and do not limit the order or interdependence of the functions performed by these apparatuses, modules, or units.

[0054] It is to be noted that the modification of "one" or "multiple" mentioned in the present disclosure is illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise explicitly stated in the context, it should be understood as "one or more".

[0055] The embodiments of the present disclosure provide a power battery fast charging control method for controlling the fast charging of a power battery to achieve a faster fast charging speed and ensure that the temperature of the power battery is within a safe threshold range. Wherein, the power battery is charged with a large current.

[0056] In the embodiments of the present disclosure, the power battery is a battery with large energy storage capacity and large volume as a power source. For example, the power battery can be a battery configured in an electric vehicle to provide electrical energy driving for vehicle travel, or a storage battery configured in a data center to avoid sudden power failure.

[0057] During the charging process with a large current, the real-time characteristics of the power battery will affect its charging speed and charging safety, and the power battery needs to be heated or cooled. For example, in the case of too low temperature of the power battery, the activity of the chemical substances inside the power battery is low, and the energy storage speed cannot be too fast. In this case, the power battery needs to be heated to improve the activity of the chemical substances in the power battery. For another example, in the case of too high temperature of the power battery, the activity of the chemical substances inside the power battery is very high, and it may have a risk of spontaneous combustion. In this case, the power battery needs to be cooled to reduce the activity of the chemical substances in the power battery.

[0058] The power battery fast charging control method provided by the embodiments of the present disclosure controls the charging current of the power battery and controls the heat management component of the power battery to make the charging speed as fast as possible and ensure that the temperature of the power battery is within a reasonable temperature range.

[0059] It should be noted that the power battery fast charging control method provided by the embodiments of the present disclosure can be executed by a computing device. In specific implementations, the computing device can exist in various possible forms. For example, when the power battery is an electric vehicle battery, the computing device can be a vehicle infotainment system of the electric vehicle; for another example, when the power battery is an energy storage battery in a data center, the computing device can be a server used for device management in the data center.

[0060] Figure 1 is a flowchart of the power battery fast charging control method provided by the embodiments of the present disclosure. As shown in Figure 1 The power battery fast charging control method provided by the embodiments of the present disclosure includes steps S110-S120.

[0061] Step S110: Obtain real-time characteristic parameters of the power battery in real time, the real-time characteristic parameters including at least two of the cell temperature and the real-time state of charge, the real-time working characteristic parameters of the heat management component, and the real-time environmental characteristic parameters of the environment in which the power battery is located.

[0062] In some embodiments of the present disclosure, a temperature sensor for measuring the cell temperature and an electrical characteristic sensor for measuring the cell resistance, current and / or voltage state are arranged in the power battery. The temperature sensor is arranged close to the cell of the power battery, which can measure the real-time cell temperature of the power battery and send the real-time cell temperature to the computing device. The electrical characteristic sensor measures the electrical characteristic state data such as the resistance, current and / or voltage of the power battery, and sends the aforementioned electrical characteristic state data to the computing device, which can estimate the real-time state of charge of the power battery according to the model of the pre-device.

[0063] In some embodiments of the present disclosure, the power battery is configured with a heat management component. The heat management component includes a refrigeration component and a heating component. In some specific applications, the heat management component can be a heat pump air conditioning system, which can heat the battery when the cell temperature is low and can cool the battery when the cell temperature is high. In some other specific applications, the heat management component can include a PTC thermistor, a circulating water pump and a cooling fan. The PTC thermistor can heat the power battery when the power cell temperature is low, and the circulating water pump and the cooling fan can cool the power battery when the power cell temperature is high.

[0064] In some embodiments of the present disclosure, the corresponding sensors are configured in the thermal management component. The corresponding sensors can measure the real-time working characteristic parameters of the thermal management component and send the real-time working characteristic parameters to the computing device. In the embodiments of the present disclosure, the real-time working characteristic parameters of the thermal management component can include the heating power, the heat dissipation power or the refrigerant compression rate of the thermal management component, etc. Through the real-time working characteristic parameters, the real-time working state of the thermal management component can be determined. In addition, in the embodiments of the present disclosure, the computing device can also store the refrigeration capacity or the heating capacity of the thermal management component under various environmental temperature conditions to determine the maximum working capacity range of the thermal management component.

[0065] In some embodiments of the present disclosure, the power battery can also be configured with sensors for measuring environmental features. The sensors for measuring environmental features at least include an environmental temperature sensor, and can also include an environmental humidity sensor. After measuring the real-time environmental feature parameters, the aforementioned sensors for measuring environmental features can send the real-time environmental feature parameters to the computing device.

[0066] It should be noted that the aforementioned real-time cell temperature, real-time state of charge, real-time working characteristic parameters and real-time environmental feature parameters refer to the parameters during the fast charging of the power battery, and the aforementioned “real-time” refers to sampling at a small sampling period during the fast charging of the power battery, and the sampling period is much smaller than the time required for the fast charging of the power battery.

[0067] More preferably, in some embodiments of the present disclosure, the real-time feature parameters can include all of the cell temperature and the real-time state of charge, the real-time working characteristic parameters of the thermal management component, and the real-time environmental feature parameters of the environment in which the power battery is located; in some other embodiments, the real-time feature parameters can only include the cell temperature and the real-time state of charge; in some other embodiments, the real-time working characteristic parameters of the thermal management component can only include the cell temperature and the real-time state of charge, and the real-time working characteristic parameters of the thermal management component.

[0068] Step S120: processing the real-time feature parameters by using the pre-trained reinforcement learning model to generate a first fast charging control instruction.

[0069] In the embodiments of the present disclosure, the computing device stores a pre-trained reinforcement learning model. The reinforcement learning model is a model trained based on a plurality of groups of sample data and sample data corresponding fast charging reward value scores, and each group of sample data includes at least two of a sample charging current, a sample cell temperature, a sample state of charge, sample working characteristic parameters of a thermal management component, and sample environmental feature parameters of an environment in which the fast charging process is located.

[0070] In a specific application of the present disclosure, the pre-trained reinforcement learning model can be a Q-learning based model, a Deep Q-Learning Network (DQN) based model, or a Deep Deterministic Policy Gradient (DDPG) based model, or other types of reinforcement learning models, which are not particularly limited by embodiments of the present disclosure. However, considering that a large amount of tables are obtained after training of the Q-Learning model, the tables occupy too much memory and the overhead of table retrieval is large, and states that do not occur during training of the Q-Learning model may occur in actual control process, causing control failure, the DQN model or the DDPG model is preferred in the specific application of the present disclosure.

[0071] In one specific application of the present disclosure, the pre-trained reinforcement learning model is a DDPG model. The DDPG model includes a Q network and a learning policy network. Through comprehensive training of the Q network and the learning policy network, a final reinforcement learning model can be obtained.

[0072] It should be noted that the pre-trained reinforcement learning model is a model that can generate relatively reasonable control instructions based on at least two of the real-time battery cell temperature, the real-time state of charge, the real-time operating characteristic parameter, and the real-time environmental characteristic parameter. The relatively reasonable control instructions refer to a model that can quickly increase the charge of the power battery, ensure that the battery cell temperature is within a safe threshold range, and reduce the thermal management energy consumption.

[0073] In some embodiments of the present disclosure, the first fast charging control instruction includes a first charging current control instruction and / or a first thermal management control instruction. The first charging current control instruction is an instruction for controlling the fast charging current, and the first thermal management control instruction is a control instruction for controlling the thermal management component. After the first charging current control instruction and the first thermal management control instruction are generated, the computing device sends the first charging current control instruction to the charging control device and sends the first thermal management control instruction to the thermal management component to control the charging current and the thermal management, thereby controlling the fast charging of the power battery.

[0074] By using the power battery fast charging control method provided by the embodiments of the present disclosure, the pre-trained reinforcement learning model is used to process the obtained real-time characteristic parameters to generate the first fast charging control instruction. Because the reinforcement learning model is trained according to sample data, the power battery control method provided by the embodiments of the present disclosure does not need to construct a control strategy lookup table to achieve accurate control of battery fast charging, thereby reducing the workload in the design and testing stages of the power battery.

[0075] In addition, the power battery fast charging control method provided by the embodiment of the present disclosure can retrain and use the reinforcement learning model by using new sample data in the case of performance degradation of the power battery cell and the thermal management system, so that the control strategy of the power battery is consistent with the actual characteristics of the power battery. Moreover, the cost of retraining the reinforcement learning model is low.

[0076] In the case of setting a high precision of the reinforcement model, the power battery fast charging control method provided by the embodiment of the present disclosure can make the temperature of the power battery in an optimal range, improve the fast charging speed, and reduce the energy consumption of the thermal management component.

[0077] Please continue to see Figure 2 . Figure 2 is a flowchart of the power battery fast charging control method provided by another embodiment of the present disclosure. As Figure 2 shown, in some embodiments of the present disclosure, the power battery control method includes steps S210-S230.

[0078] Step S210: Obtain real-time characteristic parameters of the power battery.

[0079] In the embodiment of the present disclosure, the real-time characteristic parameters include at least two of the real-time cell temperature and the real-time state of charge, the real-time working characteristic parameters of the thermal management component, and the real-time environmental characteristic parameters of the environment in which the power battery is located, wherein the real-time cell temperature is included.

[0080] Step S220: Determine whether the real-time cell temperature is within a preset safety threshold; if yes, execute step S230.

[0081] Step S230: Process the real-time characteristic parameters by using a pre-trained reinforcement learning model to generate a first fast charging control instruction.

[0082] In the embodiment of the present disclosure, step S210 is the same as step S110 of the previous embodiment, and step S230 is the same as step S120 of the previous embodiment. Different from the previous embodiment, before executing step S230, it is first determined whether the cell temperature is within the preset safety threshold.

[0083] Since the reinforcement learning model is a model trained by using multiple sample data, due to the problems of the sample data, the reinforcement learning model may have weak generalization ability. The multiple first thermal management control instructions output by the reinforcement learning model may cause the cell temperature to continuously exceed the preset safety threshold when controlling the power battery fast charging, thereby increasing the probability of self-ignition of the battery.

[0084] To avoid the foregoing problems, in some embodiments of the present disclosure, after the real-time cell temperature of the power battery is acquired, it is first determined whether the real-time cell temperature is within a preset safety threshold. Only when the real-time cell temperature is within the preset safety threshold range, the first fast charging control instruction is generated by using the reinforcement learning model, thereby avoiding the problem of weak model generalization ability caused by model training problems, and improving the safety of the power battery fast charging.

[0085] Optionally, in some embodiments of the present disclosure, if the computing device determines that the real-time cell temperature is not within the preset safety threshold range when performing step S220, the computing device can send the acquired real-time cell temperature, real-time state of charge, real-time working characteristic parameters of the thermal management component, and real-time environmental characteristic parameters to a remote server, and generate a fast charging control instruction by the remote server to control the fast charging current size and the working characteristics of the thermal management component.

[0086] Optionally, in some embodiments of the present disclosure, if the computing device determines that the real-time cell temperature is not within the preset safety threshold range when performing step S220, step S240 can also be performed to control the power battery fast charging in a local control manner.

[0087] Step S240: generating a second fast charging control instruction according to the implementation characteristic parameters by using a rule-based lookup table control strategy.

[0088] If the real-time cell temperature is not within the safety threshold range, in order to ensure the safety of the power battery, the computing device generates a second fast charging control instruction by using a rule-based lookup table control strategy. The second fast charging control instruction includes a second charging current control instruction for controlling the fast charging current and a second thermal management control instruction for controlling the thermal management component.

[0089] The rule-based lookup table control strategy determines the second thermal management control instruction according to at least two of the real-time cell temperature, the real-time state of charge, the real-time working characteristic parameters of the thermal management component, and the real-time environmental characteristic parameters, and looks up a pre-calibrated control instruction lookup table.

[0090] Because the lookup table is artificially calibrated and verified by a large amount of measured data in the design and test stage of the power battery, its reliability is high, so determining the second fast charging control instruction based on the rule-based lookup table can better ensure the safety of the power battery.

[0091] In some embodiments of the present disclosure, the sample data used to train the reinforcement learning model can include simulation data obtained by simulation using a pre-constructed battery fast charging physical model, the battery fast charging physical model including a battery cell temperature sub-model, a state of charge sub-model, and an update strategy sub-model, the battery cell temperature sub-model, the state of charge sub-model, and the update strategy sub-model being constructed based on a plurality of sets of historical data, the historical data including at least two of historical charging current, historical battery cell temperature, historical state of charge, historical working characteristic parameters of thermal management components, and historical environmental characteristic parameters of an environment in which the fast charging process is performed.

[0092] In some other embodiments of the present disclosure, the sample data used to train the reinforcement learning model can include historical data, the historical data including at least two of historical charging current, historical battery cell temperature, historical state of charge, historical working characteristic parameters of thermal management components, and historical environmental characteristic parameters of an environment in which the fast charging process is performed.

[0093] In some embodiments of the present disclosure, the computing device can further perform steps S130-S140 after performing the aforementioned steps S110-S120 or steps S210-S240.

[0094] Step S130: Calculate a fast charging reward value score based on fast charging control performed according to the first fast charging control instruction.

[0095] In the embodiments of the present disclosure, the fast charging reward value score based on fast charging control performed according to the first fast charging control instruction can be calculated by the following steps. First, obtain the change in state of charge of the power battery, the battery cell temperature, and the energy consumption value of the thermal management components based on the fast charging control performed according to the first fast charging control instruction. Then, calculate the value scores corresponding to the change in state of charge, the battery cell temperature, and the energy consumption value of the thermal management components by using a method similar to the aforementioned steps S332-S333, and calculate the fast charging reward value score by using the three value scores.

[0096] Step S140: If the fast charging reward value scores corresponding to a preset number of first fast charging control instructions are less than a preset score value, retrain the reinforcement learning model.

[0097] In the embodiments of the present disclosure, the computing device compares the fast charging reward value scores corresponding to the first fast charging control instructions executed in sequence with the preset score value, and counts the sizes of the fast charging reward value scores and the preset score value. If the fast charging reward value scores corresponding to the preset number of first fast charging control instructions are less than the preset score value, it is determined that the reinforcement learning model no longer matches the actual characteristics of the power battery, and thus the reinforcement learning model needs to be retrained. It should be noted that the reinforcement learning model is retrained based on newer sample data of the preset number.

[0098] In addition to the foregoing power battery fast charging control method, the embodiment of the present disclosure also provides a reinforcement learning model training method for power battery fast charging. Figure 3 The reinforcement learning model training method for power battery fast charging provided by the embodiment of the present disclosure is shown in the flowchart of Figure 3 As shown in the flowchart, the reinforcement learning model training method comprises steps S310-S320.

[0099] Step S310: obtaining sample data and obtaining a fast charging reward value score corresponding to the sample data, the sample data comprising at least two of a sample charging current, a sample battery temperature, a sample state of charge, a sample working characteristic parameter of a thermal management component, and a sample environmental characteristic parameter of an environment in which the fast charging process is located;

[0100] Step S320: training the reinforcement learning model using the sample data and the corresponding fast charging reward value score.

[0101] In specific implementation, the process of training the reinforcement learning model is different according to different types of reinforcement learning model. In some embodiments of the present disclosure, the reinforcement learning model is a model based on the DDPG algorithm, and the process of memory training and learning of the model is: outputting a control instruction using a policy network, and estimating a future possible reward value score of the output control instruction using a Q network; at this time, the policy network can update network parameters according to the returned possible reward value score; the Q network updates network parameters according to feedback and reward value scores of the environment, and maximizes the future possible reward value score. The method of optimizing the Q network is to construct a loss function using the difference between the real reward value score of each step and the possible reward value score of the next step, and to minimize the loss function by optimizing the network parameters, wherein the loss function can be expressed by the mean square error of the difference between the real reward value score of each step and the possible reward value score of the next step. Because the reinforcement learning model is trained according to the sample data, the fast charging control of the power battery based on the reinforcement learning model does not need to use a control strategy lookup table, thereby reducing the workload in the design and testing stages of the power battery.

[0102] Optionally, in some embodiments of the present disclosure, the step of obtaining sample data in step S310 can comprise steps S311-S313.

[0103] Step 311: constructing a battery fast charging physical model according to a plurality of sets of historical data, wherein the battery fast charging physical model comprises a battery temperature sub-model, a state of charge sub-model, and a policy update sub-model.

[0104] The battery fast charging physical model in the embodiments of the present disclosure is a model for simulating the fast charging of the power battery. Specifically, the state of charge sub-model can be used to simulate the change of the state of charge of the power battery under various charging currents; the cell temperature sub-model is used to simulate the temperature change characteristics of the battery under various thermal management strategies provided by the thermal management component and various charging currents; the strategy updating sub-model can be used to simulate the update of the fast charging current sub-strategy of the power battery under various states of charge and temperature states.

[0105] In the embodiments of the present disclosure, the state of charge sub-model and the cell temperature sub-model can be simulated according to at least two of the real-time cell temperature, the real-time state of charge, the real-time working characteristic parameters of the thermal management component and the real-time environmental characteristic parameters in the historical data. Specifically, the state of charge sub-model and the cell temperature sub-model are sub-models constructed by using state space equations, transfer functions or differential equations, and the parameters in the two sub-models are trained and identified by using the least square method, the gradient descent algorithm or the genetic algorithm. The strategy updating sub-model can be specified according to the charging control strategy developed and designed for the power battery. After the parameter training and model identification, the battery fast charging physical model can accurately simulate the state of charge change and the temperature change of the battery under various conditions, and determine the control parameters of the thermal management component required for the battery to be in a specific temperature range.

[0106] It should be noted that the various sub-models described above are mutually related, that is, the parameters in each sub-model need to be identified simultaneously in the parameter identification process. Specifically, the parameters in the state of charge sub-model and the cell temperature sub-model need to be identified simultaneously.

[0107] Step S312: data simulation is performed by using the battery fast charging physical model to generate a plurality of groups of simulation data.

[0108] In the embodiments of the present disclosure, each group of simulation data includes at least two of the simulation pre-cell temperature, the simulation state of charge, the simulation charging power, the simulation working characteristic parameters of the thermal management component, the simulation environmental characteristic parameters, the simulation post-cell temperature and the simulation post-cell temperature.

[0109] Specifically, the data simulation by using the battery fast charging physical model is as follows: the simulation pre-cell temperature, the simulation pre-state of charge, the simulation environmental characteristic parameters, the selected simulation charging current and the simulation working characteristic parameters of the thermal management component are input into the battery fast charging physical model to obtain the simulation post-cell temperature and the simulation post-state of charge.

[0110] Step S313: the simulation data is used as sample data.

[0111] Optionally, in some embodiments of the present disclosure, the fast charging reward value score corresponding to the sample data in step S310 can include steps S314-S316.

[0112] Step S314: Calculate the state of charge change value according to the pre-simulation state of charge and the post-simulation state of charge, and calculate the simulation energy consumption value of the thermal management component according to the simulation working characteristic parameters of the thermal management component.

[0113] In the embodiments of the present disclosure, the state of charge change value can be calculated by subtracting the post-simulation state of charge from the pre-simulation state of charge. In specific applications, the state of charge change value is a positive value or zero.

[0114] In the embodiments of the present disclosure, the simulation energy consumption value of the thermal management component can be calculated according to the simulation working characteristic parameters of the thermal management component. The simulation energy consumption value can be determined according to the corresponding relationship between the historical energy consumption of the thermal management component, the pre-set energy consumption and the working characteristic parameters. The simulation energy consumption value represents the power consumption of the thermal management component when working, and the power consumption can be a positive value or zero.

[0115] Step S315: Determine the value score corresponding to the state of charge change value, the simulation energy consumption value and the post-simulation battery cell temperature respectively based on the pre-set scoring rules.

[0116] According to the purpose and constraint conditions of the fast charging of the power battery, the target of the fast charging includes: (1) the charging rate is as large as possible; (2) the battery cell temperature does not exceed the pre-set safety threshold; (3) the energy consumption of the thermal management component is as small as possible during the fast charging process. The scoring rules can be pre-set according to the foregoing charging targets.

[0117] In the embodiments of the present disclosure, the scoring rules can include the charging rate score, the thermal management component energy consumption score and the battery cell temperature score.

[0118] The charging rate value score is in a positive proportional relationship with the state of charge change value. The larger the state of charge change value is, the larger the charging rate value score is. If the state of charge change value is zero, the charging rate value score can be set to a small negative score.

[0119] The thermal management component energy consumption value score is in an inverse proportional relationship with the energy consumption of the thermal management component. The smaller the thermal management energy consumption is, the larger the energy consumption value score is. If the thermal management energy consumption value is zero, the energy consumption value score can be set to a large positive score.

[0120] The battery cell temperature score is in an inverse proportional relationship with the difference between the post-simulation battery cell temperature and the median value of the ideal battery cell temperature. The larger the difference between the post-simulation battery cell temperature and the median value of the ideal battery cell temperature is, the smaller the value score corresponding to the post-simulation battery cell temperature is. If the battery cell temperature exceeds the pre-set safety threshold, the value score corresponding to the post-simulation battery cell temperature is a very large negative score.

[0121] Step S316: Calculate the sum of the value score corresponding to the state of charge change value, the simulation energy consumption value and the simulated battery cell temperature, to obtain the fast charging reward value score.

[0122] After obtaining the value score of the state of charge change value, the value score of the simulation energy consumption value and the value score corresponding to the simulated battery cell temperature, the sum of the aforementioned three value scores can obtain the fast charging reward value score obtained by using the control strategy in the simulation data.

[0123] Of course, in other embodiments of the present disclosure, other ways can also be used to calculate the fast charging reward value score corresponding to each group of simulation data. For example, in another embodiment of the present disclosure, the artificial overall evaluation score can be obtained according to the state of charge change value, the thermal management component energy consumption value and the battery cell temperature value corresponding to each group of sample data in advance, and the sample data score and the corresponding artificial overall evaluation score are fitted or model trained to obtain the fast charging reward value score model. When calculating the fast charging reward value score corresponding to each group of simulation data, the state of charge change value, the simulation energy consumption value and the simulated battery cell temperature are input into the aforementioned model to obtain the fast charging reward value score.

[0124] In actual application, in the method with less historical data groups, more groups of simulation data can be generated by using the aforementioned steps S311-S316, the simulation data is used as sample data, and the sample data and the corresponding fast charging reward value score are used to train the reinforcement learning model, so that the reinforcement learning model has better generalization ability and calculation precision.

[0125] In other embodiments of the present disclosure, step S310 of obtaining sample data can include steps S317-S318.

[0126] Step S317: Obtain a plurality of groups of historical data, the historical data including at least two of historical charging current, historical battery cell temperature, historical state of charge, historical working characteristic parameter of thermal management component and historical environmental characteristic parameter of environment in historical fast charging process.

[0127] Step S318: Use the historical data as sample data.

[0128] Similar to the calculation of the fast charging reward value of the foregoing simulation data, in the case that the sample data is historical data, the state of charge change value, the thermal management component energy consumption value and the battery temperature after charging corresponding to each group of historical data can be calculated first. The state of charge change value can be obtained by subtracting the state of charge at the previous time from the state of charge at the later time, the thermal management component energy consumption value can be calculated according to the working characteristic parameters of the thermal management component and the pre-set lookup table, and the battery temperature after charging is the battery temperature in the sample data. After obtaining the foregoing historical data, the fast charging reward value score can be calculated by using the foregoing steps S314-S316.

[0129] In addition to providing the foregoing power battery fast charging control method, the embodiments of the present disclosure also provide a power battery fast charging control device. Figure 4 FIG. 1 is a structural schematic diagram of a power battery fast charging control device provided by the embodiments of the present disclosure. As shown in FIG. 1, the power battery fast charging control device 400 provided by the embodiments of the present disclosure includes a real-time state acquisition unit 401 and a control instruction generation unit 402. Figure 4 The real-time state acquisition unit 401 is configured to acquire real-time characteristic parameters of the power battery, and the real-time characteristic parameters include at least two of a real-time battery temperature and a real-time state of charge, a real-time working characteristic parameter of a thermal management component, and a real-time environmental characteristic parameter of an environment in which the power battery is located.

[0130] The control instruction generation unit 402 is configured to process the real-time characteristic parameters by using a pre-trained reinforcement learning model to generate a first fast charging control instruction, and the first fast charging control instruction includes a first charging current control instruction for controlling a fast charging current and / or a first thermal management control instruction for controlling the thermal management component.

[0131] The reinforcement learning model is trained based on a plurality of groups of sample data and a fast charging reward value score corresponding to the sample data, and each group of sample data includes at least two of a sample charging current, a sample battery temperature, a sample state of charge, a sample working characteristic parameter of a thermal management component, and a sample environmental characteristic parameter of an environment in which a fast charging process is located.

[0132] Optionally, in some embodiments of the present disclosure, the power battery fast charging control device 400 can further include a judgment unit. The judgment unit is configured to judge whether the real-time battery temperature is within a pre-set safety threshold. Correspondingly, in the case that the judgment unit determines that the real-time battery temperature is within the pre-set safety threshold, the control instruction generation unit 402 performs the operation of processing the real-time battery temperature, the real-time state of charge, the real-time working characteristic parameter and the real-time environmental characteristic parameter by using the pre-trained reinforcement learning model to generate the first fast charging control instruction.

[0133] Optionally, in some embodiments of the present disclosure, the power battery fast charging control device 400 can further include a judgment unit. The judgment unit is configured to judge whether the real-time battery temperature is within a pre-set safety threshold. Correspondingly, in the case that the judgment unit determines that the real-time battery temperature is within the pre-set safety threshold, the control instruction generation unit 402 performs the operation of processing the real-time battery temperature, the real-time state of charge, the real-time working characteristic parameter and the real-time environmental characteristic parameter by using the pre-trained reinforcement learning model to generate the first fast charging control instruction.

[0134] Optionally, in a case where the judging unit determines that the battery cell temperature is not within the preset safety threshold, the control instruction generation unit 402 adopts a control strategy based on a rule lookup table to generate a second fast charging control instruction according to the real-time battery cell temperature, the real-time state of charge, the real-time working characteristic parameter of the thermal management component, and the real-time environmental characteristic parameter. The second fast charging control instruction includes a second charging current control instruction for controlling the fast charging current and a second thermal management control instruction for controlling the thermal management component.

[0135] In some embodiments of the present disclosure, the sample data includes simulation data obtained by simulation using a pre-constructed battery fast charging physical model. The battery fast charging physical model includes a battery cell temperature sub-model, a state of charge sub-model, and an update strategy sub-model. The battery cell temperature sub-model, the state of charge sub-model, and the update strategy sub-model are constructed based on a plurality of sets of historical data, and the historical data includes at least two of historical charging current, historical battery cell temperature, historical state of charge, historical working characteristic parameter of the thermal management component, and historical environmental characteristic parameter of the environment in which the fast charging process is located.

[0136] In some embodiments of the present disclosure, the power battery fast charging control device 400 can further include a first return score calculation unit. The first return score calculation unit is configured to calculate a fast charging return value score after fast charging control based on the first fast charging control instruction. Correspondingly, the model training unit re-trains the reinforcement learning model in a case where the fast charging return value scores corresponding to the preset number of first fast charging control instructions are less than a preset score.

[0137] Figure 5 FIG. 5 is a structural schematic diagram of a reinforcement learning model training device for power battery fast charging provided by an embodiment of the present disclosure. As shown in FIG. 5, the reinforcement learning model training device 500 includes a sample acquisition unit 501 and a model training unit 502. Figure 5

[0138] The sample acquisition unit 501 is configured to acquire sample data and acquire a fast charging return value score corresponding to the sample data. The sample data includes at least two of sample charging current, sample battery cell temperature, sample state of charge, sample working characteristic parameter of the thermal management component, and sample environmental characteristic parameter of the environment in which the fast charging process is located.

[0139] The model training unit 502 is configured to train the reinforcement learning model using the sample data and the corresponding fast charging return value score.

[0140] Optionally, in some embodiments of the present disclosure, the sample acquisition unit 501 includes a physical model construction sub-unit, a simulation sub-unit, and a first sample determination sub-unit.

[0141] ​The physical model construction subunit is used to construct a battery fast charging physical model based on multiple sets of historical data. The battery fast charging physical model includes a cell temperature submodel, a state of charge submodel, and an update strategy submodel. The historical data includes at least two of the following: historical charging current, historical cell temperature, historical state of charge, historical operating characteristic parameters of thermal management components, and historical environmental characteristic parameters of the environment in which the fast charging process takes place.

[0142] The simulation subunit is used to perform data simulation using the battery fast charging physical model, generating multiple sets of simulation data. The simulation data includes at least two of the following: cell temperature before simulation, state of charge before simulation, simulation charging current, simulation operating characteristic parameters of thermal management components, simulation environmental characteristic parameters, cell temperature after simulation, and state of charge after simulation.

[0143] The first sample determination subunit is used to use simulation data as sample data.

[0144] Optionally, in some embodiments of this disclosure, the sample acquisition unit 501 further includes a simulation energy consumption calculation subunit, a sub-score calculation subunit, and a return value score calculation subunit.

[0145] The simulation energy consumption calculation subunit is used to calculate the change in state of charge based on the state of charge before and after the simulation, and to calculate the simulation energy consumption of the thermal management component based on the simulation operating characteristic parameters of the thermal management component.

[0146] The sub-score calculation sub-unit is used to determine the value scores corresponding to the change in state of charge, the simulated energy consumption, and the cell temperature after simulation, based on the pre-set scoring rules.

[0147] The reward value score calculation subunit is used to calculate the sum of the value scores corresponding to the change in state of charge, the simulated energy consumption, and the cell temperature after simulation, to obtain the fast charging reward value score.

[0148] Optionally, in some embodiments of this disclosure, the sample acquisition unit includes a historical data acquisition subunit and a second sample determination subunit. The historical data acquisition subunit is used to acquire multiple sets of historical data, including at least two of the following: historical charging current, historical cell temperature, historical state of charge, historical operating characteristic parameters of thermal management components, and historical environmental characteristic parameters of the environment in which the historical fast charging process took place. The second sample determination subunit is used to use the historical data as sample data.

[0149] Figure 6 This is a schematic diagram of a computing device provided in an embodiment of this disclosure. See below for details. Figure 6 It shows a schematic diagram of a structure suitable for implementing the computing device 600 in the embodiments of this disclosure. Figure 6The computing device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0150] like Figure 6 As shown, the computing device 600 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory ROM 602 or a program loaded from a storage device 608 into a random access memory RAM 603. The RAM 603 also stores various programs and data required for the operation of the computing device 600. The processing unit 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0151] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, a touchscreen, touchpad, camera, microphone, accelerometer, gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 608 including, for example, magnetic tape, hard disk, etc.; and communication devices 609. Communication device 609 allows computing device 600 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 A computing device 600 with various devices is shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.

[0152] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, it performs the functions defined in the methods of embodiments of this disclosure.

[0153] It is noted that the aforementioned computer-readable medium of the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium can be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any suitable combination of the foregoing. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program used by or in connection with an instruction execution system, apparatus or device. In the present disclosure, the computer-readable signal medium can include a data signal propagated in baseband or propagated as a carrier wave in a propagated data signal, in which the computer-readable program code is contained. Such a propagated data signal can take any of a variety of forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium that is not a storage medium and that can communicate, propagate or transport a program for use by or in connection with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wire, cable, RF, infrared, or any suitable combination thereof.

[0154] In some embodiments, the client, server, or both can communicate using any current known or future developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet, and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any current known or future developed networks.

[0155] The aforementioned computer-readable medium can be contained in the aforementioned computing device; or can exist separately from the computing device, not installed in the computing device.

[0156] The computer readable medium described above carries one or more programs, when the one or more programs are executed by the computing device, cause the computing device to: acquire real-time characteristic parameters of the power battery, the real-time characteristic parameters including at least two of a real-time cell temperature and a real-time state of charge, a real-time working characteristic parameter of the thermal management component, and a real-time environmental characteristic parameter of an environment in which the power battery is located; process the real-time characteristic parameters using a pre-trained reinforcement learning model to generate a first fast-charging control instruction, the first fast-charging control instruction including a first charging current control instruction for controlling a fast-charging current and / or a first thermal management control instruction for controlling the thermal management component; wherein the reinforcement learning model is trained based on a plurality of sets of sample data and sample data corresponding fast-charging reward value scores, each set of sample data including at least two of a sample charging current, a sample cell temperature, a sample state of charge, a sample working characteristic parameter of the thermal management component, and a sample environmental characteristic parameter of an environment in which a fast-charging process is located.

[0157] acquire a real-time cell temperature and a real-time state of charge of the power battery, a real-time working characteristic parameter of the thermal management component, and a real-time environmental characteristic parameter of an environment in which the power battery is located; process the real-time cell temperature, the real-time state of charge, the real-time working characteristic parameter, and the real-time environmental characteristic parameter using a pre-trained reinforcement learning model to generate a first fast-charging control instruction, the first fast-charging control instruction including a first charging current control instruction for controlling a fast-charging current and a first thermal management control instruction for controlling the thermal management component.

[0158] Computer program code for carrying out operations of the present disclosure can be written in any of one or more programming languages or combinations of languages including object or visual programming languages such as Java, Smalltalk, C++ or conventional procedural programming languages such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0159] The computer program product of the first aspect can include one or more non-transitory computer-readable media storing instructions that, when executed, cause one or more processors to perform the operations of the first aspect. The computer program product of the first aspect can include a non-transitory computer-readable medium storing code that, when executed, causes a computer to perform operations for the first aspect.

[0160] The units described in the embodiments of the present disclosure can be implemented by software, or by hardware, or by a combination of software and hardware. In some cases, the names of the units do not constitute a limitation on the units themselves.

[0161] The functions described in this document can be implemented in hardware, software, or any combination thereof. In some embodiments, the functions described in this document can be implemented in one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system on a chip (SOCs), complex programmable logic devices (CPLDs), etc.

[0162] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium will include one or more of an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0163] The embodiments of the present disclosure further provide a computer readable storage medium, wherein the storage medium stores a computer program. When the computer program is executed by a processor, the method of any one of the above method embodiments can be implemented, and the execution manner and beneficial effects are similar, which will not be described here again.

[0164] It should be noted that, in this document, relational terms such as "first" and "second", and the like, are used solely to distinguish one entity or action from another entity or action, without necessarily requiring or implying any actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without more limitations, an element defined by the statement "comprising a... " does not exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0165] The above description is merely one specific implementation of the present disclosure, and persons skilled in the art can understand or implement the present disclosure from the above description. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to these embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fast-charging control method for a power battery, characterized in that, include: The real-time characteristic parameters of the power battery are obtained, including real-time cell temperature and real-time state of charge, real-time operating characteristic parameters of thermal management components, and real-time environmental characteristic parameters of the environment in which the power battery is located. The real-time feature parameters are processed using a pre-trained reinforcement learning model to generate a first fast charging control command. The first fast charging control command includes a first charging current control command for controlling the fast charging current and a first thermal management control command for controlling the thermal management component. The reinforcement learning model is trained based on multiple sets of sample data and the fast charging reward value scores corresponding to the sample data. Each set of sample data includes sample charging current, sample cell temperature, sample state of charge, sample operating characteristic parameters of thermal management components, and sample environmental characteristic parameters of the environment in which the fast charging process takes place. The sample data includes simulation data obtained by simulating a pre-built battery fast charging physical model. The simulation data includes cell temperature before simulation, state of charge before simulation, simulation charging current, simulation operating characteristic parameters of thermal management components, simulation environmental characteristic parameters, cell temperature after simulation, and state of charge after simulation. The method further includes: Calculate the fast charging reward value score after fast charging control is performed based on the first fast charging control command; If the fast charging reward value score corresponding to a preset number of first fast charging control commands is less than a preset score, the reinforcement learning model is retrained. The calculation steps for the fast charging reward value score include: The change in state of charge is calculated based on the pre-simulation state of charge and the post-simulation state of charge, and the simulated energy consumption of the thermal management component is calculated based on the simulated operating characteristic parameters of the thermal management component. Based on the pre-set scoring rules, the value scores corresponding to the change in state of charge, the simulated energy consumption, and the simulated cell temperature are determined respectively. The sum of the value scores corresponding to the change in state of charge, the simulated energy consumption, and the simulated cell temperature is calculated to obtain the fast charging reward value score.

2. The method according to claim 1, characterized in that, The real-time characteristic parameters include real-time cell temperature; Before processing the real-time feature parameters using a pre-trained reinforcement learning model, the method further includes: Determine whether the real-time cell temperature is within a preset safety threshold; When the real-time cell temperature is within a preset safety threshold, the operation of using a pre-trained reinforcement learning model to process the real-time feature parameters and generate a first fast charging control command is executed.

3. The method according to any one of claims 1 or 2, characterized in that, The battery fast charging physical model includes a cell temperature sub-model, a state of charge sub-model, and an update strategy sub-model. The cell temperature sub-model, the state of charge model, and the update strategy sub-model are constructed based on multiple sets of historical data. The historical data includes historical charging current, historical cell temperature, historical state of charge, historical operating characteristic parameters of thermal management components, and historical environmental characteristic parameters of the environment in which the fast charging process takes place.

4. A reinforcement learning model training method for fast charging of power batteries, characterized in that, include: Acquire sample data and the corresponding fast charging return value score. The sample data includes sample charging current, sample cell temperature, sample state of charge, sample operating characteristic parameters of thermal management components, and sample environmental characteristic parameters of the environment in which the fast charging process takes place. The sample data also includes simulation data obtained by simulating a pre-built battery fast charging physical model. The simulation data includes cell temperature before simulation, state of charge before simulation, simulation charging current, simulation operating characteristic parameters of thermal management components, simulation environmental characteristic parameters, cell temperature after simulation, and state of charge after simulation. The reinforcement learning model is trained using the sample data and the corresponding fast charging reward value score; The method further includes: Calculate the fast charging reward value score after fast charging control is performed based on the first fast charging control command; If the fast charging reward value score corresponding to a preset number of first fast charging control commands is less than a preset score, the reinforcement learning model is retrained. The obtained fast charging return value score corresponding to the sample data includes: The change in state of charge is calculated based on the pre-simulation state of charge and the post-simulation state of charge, and the simulated energy consumption of the thermal management component is calculated based on the simulated operating characteristic parameters of the thermal management component. Based on the pre-set scoring rules, the value scores corresponding to the change in state of charge, the simulated energy consumption, and the simulated cell temperature are determined respectively. The sum of the value scores corresponding to the change in state of charge, the simulated energy consumption, and the simulated cell temperature is calculated to obtain the fast charging reward value score.

5. The method according to claim 4, characterized in that, The acquisition of sample data includes: A fast-charging physical model for batteries is constructed based on multiple sets of historical data. The fast-charging physical model for batteries includes a cell temperature sub-model, a state of charge sub-model, and an update strategy sub-model. The historical data includes historical charging current, historical cell temperature, historical state of charge, historical operating characteristic parameters of thermal management components, and historical environmental characteristic parameters of the environment in which the fast-charging process takes place. Data simulation was performed using the aforementioned battery fast charging physical model to generate multiple sets of simulation data; The simulation data is used as the sample data.

6. The method according to claim 4, characterized in that, The acquisition of sample data includes: Acquire multiple sets of historical data, including historical charging current, historical cell temperature, historical state of charge, historical operating characteristic parameters of thermal management components, and historical environmental characteristic parameters of the environment in which the historical fast charging process took place; The historical data is used as the sample data.

7. A fast-charging control device for a power battery, characterized in that, include: The real-time status acquisition unit is used to acquire real-time characteristic parameters of the power battery. The real-time characteristic parameters include real-time cell temperature and real-time state of charge, real-time operating characteristic parameters of thermal management components, and real-time environmental characteristic parameters of the environment in which the power battery is located. The control command generation unit is used to process the real-time feature parameters using a pre-trained reinforcement learning model to generate a first fast charging control command. The first fast charging control command includes a first charging current control command for controlling the fast charging current and a first thermal management control command for controlling the thermal management component. The reinforcement learning model is trained based on multiple sets of sample data and the fast charging reward value scores corresponding to the sample data. Each set of sample data includes sample charging current, sample cell temperature, sample state of charge, sample operating characteristic parameters of thermal management components, and sample environmental characteristic parameters of the environment in which the fast charging process takes place. The sample data includes simulation data obtained by simulating a pre-built battery fast charging physical model. The simulation data includes cell temperature before simulation, state of charge before simulation, simulation charging current, simulation operating characteristic parameters of thermal management components, simulation environmental characteristic parameters, cell temperature after simulation, and state of charge after simulation. The device further includes: a first reward score calculation unit; The first reward score calculation unit is used to calculate the fast charging reward value score after fast charging control is performed based on the first fast charging control command; if the fast charging reward value score corresponding to a preset number of first fast charging control commands is less than a preset score, the reinforcement learning model is retrained. The calculation steps for the fast charging reward value score include: The change in state of charge is calculated based on the pre-simulation state of charge and the post-simulation state of charge, and the simulated energy consumption of the thermal management component is calculated based on the simulated operating characteristic parameters of the thermal management component. Based on the pre-set scoring rules, the value scores corresponding to the change in state of charge, the simulated energy consumption, and the simulated cell temperature are determined respectively. The sum of the value scores corresponding to the change in state of charge, the simulated energy consumption, and the simulated cell temperature is calculated to obtain the fast charging reward value score.

8. A reinforcement learning model training device for fast charging of power batteries, characterized in that, include: The sample acquisition unit is used to acquire sample data and the fast charging reward value score corresponding to the sample data. The sample data includes sample charging current, sample cell temperature, sample state of charge, sample operating characteristic parameters of thermal management components, and sample environmental characteristic parameters of the environment in which the fast charging process takes place. The sample data also includes simulation data obtained by simulating a pre-built battery fast charging physical model. The simulation data includes cell temperature before simulation, state of charge before simulation, simulation charging current, simulation operating characteristic parameters of thermal management components, simulation environmental characteristic parameters, cell temperature after simulation, and state of charge after simulation. The model training unit is used to train the reinforcement learning model using the sample data and the corresponding fast charging reward value score, and is also used to calculate the fast charging reward value score after fast charging control is performed based on the first fast charging control command. If the fast charging reward value score corresponding to a preset number of first fast charging control commands is less than a preset score, the reinforcement learning model is retrained. The sample acquisition unit also includes a simulation energy consumption calculation subunit, a sub-score calculation subunit, and a return value score calculation subunit; The simulation energy consumption calculation subunit is used to calculate the change value of the state of charge based on the state of charge before simulation and the state of charge after simulation, and to calculate the simulation energy consumption value of the thermal management component based on the simulation operating characteristic parameters of the thermal management component. The sub-score calculation sub-unit is used to determine the value scores corresponding to the change in state of charge, the simulated energy consumption, and the simulated cell temperature based on the pre-set scoring rules. The reward value score calculation subunit is used to calculate the sum of the value scores corresponding to the change in state of charge, the simulated energy consumption value, and the simulated cell temperature, to obtain the fast charging reward value score.

9. A computer device, characterized in that, include: A memory and a processor, wherein the memory stores a computer program that, when executed by the processor, implements the power battery fast charging control method as described in any one of claims 1-3 or the reinforcement learning model training method for power battery fast charging as described in any one of claims 4-6.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which is executed by a processor. Implement the power battery fast charging control method as described in any one of claims 1-3 or the reinforcement learning model training method for power battery fast charging as described in any one of claims 4-6.

Images