A circulating current suppression system, method, medium, and server based on an optimized parallel battery pack topology structure

Abstract

The application provides a circulating current suppression system, method, medium and server based on an optimized parallel battery pack topology structure. The application can actively adjust the circulating current suppression strategy according to the real-time state and historical change trend of the battery pack in the whole cycle of vehicle power-on, operation and high-voltage shutdown, and realize dynamic control of the inter-pack voltage difference and circulating current risk. It is especially suitable for the multi-pack parallel battery system of an electric mine truck under complex working conditions such as heavy load climbing, braking energy recovery and high power output, and can effectively reduce abnormal circulating current caused by large current working conditions and poor consistency of the battery pack, reduce invalid energy loss and local heating, delay the aging of the battery pack, and improve the operation safety, energy utilization efficiency and service life of the battery system.

Description

Technical Field

[0001] This application relates to the field of battery pack control technology, and in particular to a circulating current suppression system, method, medium, and server based on an optimized parallel battery pack topology. Background Technology

[0002] In the field of power battery technology, circulating current has always been a significant factor affecting battery pack performance, lifespan, and operational safety. Circulating current refers to the undesirable circulating current that forms within the battery system when multiple battery packs, battery clusters, or battery branches are connected in parallel. This current arises due to differences in terminal voltage, internal resistance, state of charge, or temperature between the parallel branches. Even battery cells produced in the same batch can gradually exhibit variations in capacity, internal resistance, open-circuit voltage, and dynamic response characteristics under long-term actual operation, influenced by factors such as operating temperature, charge / discharge rate, cycle count, load fluctuations, heat dissipation conditions, and manufacturing consistency deviations. When these differences accumulate to a certain extent, continuous or transient circulating currents may occur between the parallel branches.

[0003] The presence of circulating current not only reduces the effective output capacity and energy utilization efficiency of the battery system, but also subjects some battery branches to long-term additional charging and discharging stress, leading to increased local temperature rise, accelerated aging, and further deterioration of battery consistency. In severe cases, circulating current may also cause overheating problems in contactors, cables, busbars, or individual battery cells, increasing the probability of system failure and even causing safety hazards such as thermal runaway. Therefore, in large-capacity, multi-branch parallel power battery systems, accurately identifying and effectively suppressing circulating current is a crucial technical issue for ensuring reliable system operation.

[0004] As large, heavy-duty industrial vehicles, electric mining trucks typically require their battery systems to meet the demands of high power output, long driving range, and high reliability. Therefore, they generally employ a high-capacity architecture with multiple battery packs or clusters connected in parallel. Compared to ordinary electric vehicles, electric mining trucks have larger battery systems, more parallel branches, and higher operating currents. Furthermore, these vehicles operate under complex conditions such as heavy-load climbing, frequent braking regenerative braking, alternating high and low temperatures, and dust vibration and impact. In these application scenarios, differences in temperature distribution, load sharing, and aging levels among the battery packs are more likely to occur, leading to circulating current problems characterized by large amplitude, rapid changes, long duration, and complex influencing factors. Therefore, electric mining trucks place higher demands on the real-time performance, dynamics, and engineering adaptability of circulating current suppression solutions.

[0005] Currently, one of the more common approaches to suppressing circulating current is a passive balancing scheme based on the battery management system (BMS). This scheme typically involves a main battery management unit (BMU) within each battery pack, and slave control units and passive balancing circuits at the battery module or cell level. The passive balancing circuit generally consists of a resistor discharge branch connected in parallel with the cell. The BMS of each battery pack interacts with the vehicle controller or higher-level control unit via a CAN bus. The slave control unit collects parameters such as voltage and temperature of each cell and reports them to the main BMS. Based on the collected data, the main BMS identifies the cell with the higher voltage and controls the corresponding balancing resistor branch to conduct. This dissipates some of the cell's energy through resistor dissipation, making the voltage of each cell more consistent. This improves the internal consistency of the battery pack to some extent and indirectly reduces the voltage difference between different battery packs, thereby suppressing circulating current.

[0006] However, passive balancing solutions are primarily designed for cell-level or module-level consistency maintenance, with balancing current typically only in the hundreds of milliamperes range, resulting in low balancing power and regulation speed. For large power battery systems in electric mining trucks, with capacities reaching thousands of amperes and parallel branch currents potentially reaching hundreds of amperes or even higher, the regulation capabilities provided by passive balancing are clearly insufficient. It is difficult to quickly compensate for the large circulating currents caused by voltage or internal resistance differences between battery packs during actual vehicle operation. Therefore, this solution is more suitable for stationary states or low-power balancing scenarios, and cannot meet the circulating current suppression requirements of electric mining trucks under high-power, highly dynamic operating conditions.

[0007] In existing technologies, another common approach is to incorporate a pre-charge circuit in each battery pack or battery group branch to reduce the inrush current caused by voltage differences between branches during system power-on. This pre-charge circuit typically includes a pre-charge resistor and a pre-charge contactor. During system power-on, the pre-charge circuit is first closed, allowing each parallel branch to slowly equalize its voltage through the pre-charge resistor. The current in each battery group branch is detected by a current sensing element, or the completion of the pre-charge process is determined based on a preset pre-charge time. When the current in each branch meets the preset conditions, or the pre-charge time reaches the set value, the management system disconnects the pre-charge circuit and sequentially closes the main positive contactor and main circuit contactor on each battery group branch, allowing the parallel battery system to enter normal external power supply mode.

[0008] The aforementioned pre-charge resistor scheme can reduce the inrush current at the moment of connection to the parallel system during system startup, providing some protection during the power-on initialization process. However, this scheme typically operates only briefly during the power-on phase, representing a static, one-time method for suppressing circulating current or inrush current. In actual operation, the load power, regenerative braking power, battery temperature, internal resistance, and terminal voltage of electric mining trucks change in real time with varying operating conditions. The current distribution between parallel branches also fluctuates continuously, meaning the circulating current does not only occur at the moment of power-on but may persist throughout the entire vehicle operation. Traditional pre-charge resistor schemes cease operation after the main circuit is closed, failing to provide online monitoring, dynamic adjustment, and continuous suppression of circulating current during vehicle travel, hill climbing, regenerative braking, or prolonged high-power output.

[0009] In summary, existing passive balancing solutions suffer from low balancing current, slow response speed, and difficulty in handling high-amplitude circulating currents in large-capacity parallel systems. Existing pre-charge resistor-based solutions are primarily suitable for suppressing instantaneous inrush currents upon power-up, but cannot cover the dynamically changing circulating currents throughout vehicle operation. Therefore, considering the characteristics of electric mining trucks—multiple battery packs operating in parallel, high power, high capacity, and under complex conditions—there is an urgent need for a battery system circulating current management solution that can cover the entire process of power-up, operation, and shutdown, and can perform online monitoring, dynamic adjustment, and adaptive suppression based on real-time circulating current status. This would improve the energy utilization efficiency, lifespan consistency, and operational safety of parallel battery systems. Summary of the Invention

[0010] In view of the shortcomings of the prior art described above, the purpose of this application is to provide a circulating current suppression system, method, medium and server based on an optimized parallel battery pack topology. The technical problem to be solved is: how to monitor, dynamically adjust and adaptively suppress the continuously changing circulating current in the entire operation of the battery system in real time under the conditions of multiple battery packs in parallel, high power and complex dynamic conditions in electric mining trucks, so as to improve the energy utilization efficiency, lifespan consistency and operational safety of the battery system.

[0011] To achieve the above and other related objectives, a first aspect of this application provides a circulating current suppression system based on an optimized parallel battery pack topology, comprising: a vehicle controller, a battery management system, and a vehicle communication terminal on the vehicle side, and a cloud platform on the cloud side; the vehicle controller is communicatively connected to the battery management system and the vehicle communication terminal, and the vehicle communication terminal is communicatively connected to the cloud platform; the battery management system connects to and controls the vehicle battery system, the vehicle battery system comprising multiple battery packs connected in parallel via positive and negative buses, each battery pack being configured with a pack-level balancing branch for adjusting the terminal voltage or voltage difference of the corresponding battery pack; after high voltage is applied, the battery management system collects the current power-down state data of each battery pack and uploads it to the cloud platform; the cloud platform performs deep learning processing based on the current power-down state data and pre-stored historical operating data to generate a circulating current suppression control strategy for the pack-level balancing branch corresponding to each battery pack and distributes it to the vehicle side for storage, for use after the next high voltage application.

[0012] In some embodiments of the first aspect of this application, each battery pack is additionally provided with a pack-level balancing branch for adjusting the battery pack terminal voltage or voltage difference. The pack-level balancing branch is connected in series with a resistor and a relay and then in parallel to the positive and negative terminals of the corresponding battery pack.

[0013] In some embodiments of the first aspect of this application, the process by which the circulating current suppression system generates a circulating current suppression control strategy through a cloud platform includes: after receiving a high-voltage command from the vehicle controller, the battery management system first executes the closing of the main negative relay and simultaneously closes the pre-charge relay to start the pre-charge process; when preset conditions are met, the pre-charge relay is opened to exit the pre-charge process, and the main positive relay is closed to allow the battery system to enter a high-voltage ready state; after receiving a high-voltage de-energization command from the vehicle controller, the battery management system executes high-voltage power-off timing control, sequentially opening the main positive relay and the main negative relay to cut off the high-voltage circuit, and after power-off, collecting the current power-off state data of the battery pack and uploading it to the cloud platform; the cloud platform performs deep learning processing based on the current power-off state data of the battery pack and pre-stored historical operating data to generate a circulating current suppression control strategy and distribute it to the vehicle side.

[0014] In some embodiments of the first aspect of this application, the step of disconnecting the precharge relay to exit the precharge process when the preset conditions are met includes: when the battery management system detects that the precharge duration has reached a preset time, or detects that the difference between the voltage at the back end of the main positive relay and the total battery voltage is less than a preset threshold, and determines that the precharge is complete, the precharge relay can be disconnected to exit the precharge process.

[0015] In some embodiments of the first aspect of this application, the circulating current suppression control strategy includes a multidimensional probability vector and a relay opening duration corresponding to each battery pack; wherein, the multidimensional probability vector represents the probability value that each battery pack should preferentially start discharging after the next high voltage is applied.

[0016] In some embodiments of the first aspect of this application, the cloud platform performs deep learning processing based on the current power-down state data and pre-stored historical operating data using a pre-trained neural network. The processing includes: structuring the received current power-down state data to form multi-dimensional input features; preprocessing and normalizing the multi-dimensional input features; inputting the preprocessed features into a temporal feature extraction layer to extract the trend information of the battery pack state changing over time; processing the extracted trend information of the battery pack state changing over time using both spatial attention and temporal attention mechanisms to form fused features; using an Actor-Critic structure to perform decision calculations on the fused features, whereby the strategy output by the Actor network includes the probability distribution of priority discharge for each battery pack and the corresponding relay opening duration, and the Critic network evaluates the state value of the strategy output by the Actor network; and jointly optimizing based on the probability distribution of priority discharge for each battery pack, the relay opening duration, and the state value evaluation results of the strategy to obtain the optimal control strategy after the next high-voltage period.

[0017] In some embodiments of the first aspect of this application, after receiving the circulating current suppression control strategy, the vehicle-mounted side determines whether the battery state meets the preset activation conditions; if the battery state meets the preset activation conditions, after completing the high voltage reduction, it executes the previously generated optimal processing strategy stored in the vehicle controller and records the data; if the battery state does not meet the preset activation conditions, it records the data and uploads it to the cloud platform.

[0018] To achieve the above and other related objectives, a second aspect of this application provides a circulating current suppression method based on an optimized parallel battery pack topology. The method includes: receiving current power-down state data uploaded from the vehicle side; performing deep learning processing based on the current power-down state data and pre-stored historical operating data to generate a circulating current suppression control strategy for the pack-level equalization branch corresponding to each battery pack; and distributing the circulating current suppression control strategy to the vehicle side for use after the next high voltage is applied.

[0019] To achieve the above and other related objectives, a third aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the circulating current suppression method based on an optimized parallel battery pack topology.

[0020] To achieve the above and other related objectives, a fourth aspect of this application provides a computer program product comprising computer program code that, when executed on a computer, causes the computer to implement the circulating current suppression method based on an optimized parallel battery pack topology.

[0021] To achieve the above and other related objectives, a fifth aspect of this application provides a cloud server, including a memory, a processor, and a computer program stored in the memory; the processor executes the computer program to implement the circulating current suppression method based on an optimized parallel battery pack topology.

[0022] As described above, the circulating current suppression system, method, medium, and server based on optimized parallel battery pack topology of this application have the following beneficial effects: This invention constructs feature vectors from multi-dimensional data such as the terminal voltage, internal resistance, temperature, SOC, historical high-voltage state, and historical operating conditions of each battery pack, and inputs these vectors into a pre-trained neural network model. The model dynamically outputs the target battery pack selection result, the discharge relay opening duration, and the corresponding energy regulation strategy. Compared to traditional solutions that rely solely on pre-charging upon power-on or passive cell-level balancing, this invention can actively adjust the circulating current suppression strategy based on the real-time state and historical trends of the battery pack throughout the entire cycle of vehicle power-on, operation, and high-voltage reduction, achieving dynamic control of inter-pack voltage differences and circulating current risks. It is particularly suitable for multi-pack parallel battery systems in electric mining trucks under complex operating conditions such as heavy-load climbing, braking energy recovery, and high-power output. It can effectively reduce abnormal circulating currents caused by high-current operating conditions and differences in battery pack consistency, reduce ineffective energy loss and localized heating, delay battery pack aging, and improve the operational safety, energy utilization efficiency, and service life of the battery system. Attached Figure Description

[0023] Figure 1 The diagram shown is a topology diagram of a circulating current suppression system based on an optimized parallel battery pack topology in one embodiment of this application.

[0024] Figure 2 The diagram shown is a schematic of an optimized parallel battery topology in one embodiment of this application.

[0025] Figure 3 The diagram shown illustrates the process of implementing circulating current suppression using a circulating current suppression system based on an optimized parallel battery pack topology in one embodiment of this application.

[0026] Figure 4 The diagram shown illustrates the process of implementing circulating current suppression using a circulating current suppression system based on an optimized parallel battery pack topology in one embodiment of this application.

[0027] Figure 5The diagram shown is a flowchart of the optimal strategy for a neural network in one embodiment of this application.

[0028] Figure 6 The diagram shown is a topology diagram of a neural network in one embodiment of this application.

[0029] Figure 7 The diagram shown is a flowchart illustrating a circulating current suppression method based on an optimized parallel battery pack topology in one embodiment of this application.

[0030] Figure 8 The diagram shown is a schematic representation of the structure of a cloud server in one embodiment of this application. Detailed Implementation

[0031] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0032] In the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" do not necessarily imply that they are different.

[0033] It should be noted that, in the embodiments of this application, the words "exemplary" or "for example" indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0034] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0035] Before providing a further detailed description of the present invention, the nouns and terms used in the embodiments of the present invention are explained, and the nouns and terms used in the embodiments of the present invention are subject to the following interpretations:

[0036] <1> Battery pack circulating current: Battery pack circulating current refers to the undesirable circulating current formed between parallel branches when multiple battery packs are connected in parallel, due to differences in terminal voltage, internal resistance, state of charge, or temperature between the battery packs. This current typically does not perform effective work through an external load, resulting in additional energy loss, localized heating, and accelerated battery aging.

[0037] <2> State of Charge (SOC): This indicates the percentage of a battery's current remaining charge relative to its rated usable capacity, usually expressed as a percentage. A higher SOC indicates that the battery currently has more remaining release capacity.

[0038] <3> State of Health (SOH): This indicates the current health of the battery relative to its initial new battery state. It is typically assessed based on a combination of factors, including capacity decay, increased internal resistance, cycle count, and changes in charge / discharge performance.

[0039] <4> State of Power (SOP): This indicates the maximum power output or absorption capability of a battery under current voltage, temperature, SOC, internal resistance, and safety limitations. SOP can be used to determine the current discharge and rechargeable power of the battery system and the vehicle's power limiting strategy.

[0040] <5> Actor-Critic architecture: A reinforcement learning or intelligent decision-making model structure where the Actor outputs control strategies or actions, such as selecting a target battery pack and the relay's on / off duration, while the Critic evaluates the value or merit of the strategy in the current state. Through collaborative training of the Actor and Critic, the model can continuously optimize the control strategy.

[0041] <6> Maximum circulating current value: The maximum circulating current value refers to the peak value of the circulating current detected between the branches of each parallel battery pack during a monitoring cycle or a single high-voltage treatment process. This indicator can be used to characterize the degree of circulating current impact and the instantaneous safety risk of the system.

[0042] <7> Average circulating current value: The average circulating current value refers to the average value obtained by statistical calculation of the circulating current within a preset monitoring time range. This indicator can be used to reflect the overall level and continuous impact of the circulating current over a period of time.

[0043] <8> Circulating current duration: Circulating current duration refers to the length of time that the circulating current exceeds a preset circulating current threshold and persists. This metric can be used to measure the persistence of the circulating current phenomenon and assess its impact on battery heating, energy loss, and lifespan degradation.

[0044] <9> Circulating current energy loss: Circulating current energy loss refers to the energy consumed and converted into heat by circulating current in circuits such as battery packs, wiring harnesses, busbars, relays, or internal resistances. It can be estimated based on circulating current, equivalent circuit resistance, and duration, and is used to evaluate the degree of energy waste caused by circulating current.

[0045] <10> Capacity consistency: Capacity consistency refers to the degree of similarity between the available capacity or actual discharge capacity of parallel battery packs. The better the capacity consistency, the more balanced the load sharing and charge changes of each battery pack during charging and discharging, which is more conducive to improving system lifespan and available capacity.

[0046] <11> Internal resistance uniformity: Internal resistance uniformity refers to the degree of similarity between the DC internal resistance or equivalent internal resistance of each battery pack. When the internal resistance difference is large, it can easily lead to uneven current distribution among parallel battery packs and increase the risk of circulating current and localized heating.

[0047] <12> SOC Equalization: SOC equalization refers to the degree of similarity in the State of Charge (SOC) among multiple battery packs, reflecting the consistency of remaining charge in each pack. A higher SOC equalization indicates that the states of charge of each battery pack are closer, which is more beneficial for reducing voltage differences and circulating current risks between packs. In some embodiments of this application,

[0048] To facilitate understanding of the embodiments of this application, firstly, in conjunction with Figure 1 Detailed explanation. Figure 1 A schematic diagram of the topology of a circulating current suppression system based on an optimized parallel battery pack topology is shown in an embodiment of the present invention.

[0049] The circulating current suppression system based on an optimized parallel battery pack topology in this application includes: a vehicle controller, a battery management system, an on-board communication terminal, and a cloud platform. The vehicle controller and the battery management system are electrically connected via the vehicle's CAN bus or on-board Ethernet, exchanging high-voltage commands, low-voltage commands, battery status parameters, and fault information in real time. The vehicle controller establishes an electrical connection with the on-board communication terminal (T-Box) via the CAN bus or on-board Ethernet. The on-board communication terminal (T-Box) has a built-in cellular communication module responsible for wireless data interaction with the cloud platform via 4G or 5G networks. The battery management system can also independently send detailed data collected from the battery pack to the T-Box, which then uploads it to the cloud. In summary, the vehicle controller is responsible for vehicle-level control decisions, the battery management system is responsible for battery safety monitoring and high-voltage circuit control, and the T-Box is responsible for uploading vehicle or battery data to the cloud, enabling remote monitoring, fault diagnosis, and data management.

[0050] Specifically, the vehicle controller and the battery management system (BMS) establish an electrical connection via the vehicle's CAN bus or in-vehicle Ethernet, primarily for real-time interaction of control commands and status data. The vehicle controller can send control requests such as high-voltage activation and deactivation commands to the BMS. The BMS then determines whether to allow the corresponding operation based on the current state of the battery pack, fault conditions, and safety requirements, and feeds back battery status parameters and fault information to the vehicle controller. Thus, the vehicle controller can coordinate vehicle power output, high-voltage activation / deactivation, fault protection, and safety control based on the real-time operating status of the battery system.

[0051] The CAN bus and automotive Ethernet are the vehicle communication networks used for data communication between various controllers. The CAN bus features high reliability, good real-time performance, and strong anti-interference capabilities, and can be used for transmitting vehicle control commands and status information. Automotive Ethernet, on the other hand, has higher data transmission bandwidth, suitable for transmitting larger data volumes or higher frequency data. After establishing an electrical connection between the vehicle controller and the battery management system through these communication networks, bidirectional data interaction can occur. For example, the vehicle controller can send control commands to the battery management system, and the battery management system can provide feedback on battery status and fault information to the vehicle controller.

[0052] The high-voltage access command refers to the high-voltage system access request sent by the vehicle controller to the battery management system when the vehicle needs to enter a driveable, chargeable, or power-supplyable state. After receiving the high-voltage access command, the battery management system usually determines whether the battery voltage, insulation status, relay status, temperature status, fault status, etc., meet the safety conditions. If the conditions are met, it controls the pre-charge circuit and main positive and main negative high-voltage relays to operate according to a preset sequence, establishing an electrical connection between the power battery pack and the vehicle's high-voltage bus, thereby putting the vehicle's high-voltage system into operation.

[0053] The high-voltage disconnection command refers to the high-voltage disconnection request sent by the vehicle controller to the battery management system when the vehicle is stopped, under fault protection, undergoing maintenance, or when high-voltage power supply is no longer needed. Upon receiving the high-voltage disconnection command, the battery management system can control the relevant high-voltage relays to disconnect based on the current current, voltage, load status, and fault status, thereby disconnecting the power battery pack from the vehicle's high-voltage bus. This reduces the risk of high-voltage charging and ensures safety when the vehicle is stopped or in a fault condition.

[0054] The battery status parameters refer to relevant data that reflect the current operating status, health status, and safety status of the battery pack. These may include, but are not limited to, total battery pack voltage, total current, individual cell voltage, module voltage, cell temperature, module temperature, state of charge (SOC), state of health (SOH), state of power (SOP), insulation resistance, relay closing status, pre-charge status, equalization status, allowable charge / discharge current, charge / discharge power limits, and internal communication status of the battery pack. The vehicle controller can use these battery status parameters to determine whether the vehicle is permitted to drive, whether power limits are required, whether charge / discharge should be prohibited, or whether fault protection strategies need to be implemented.

[0055] The fault information refers to abnormal states or risk information identified by the battery management system during battery pack operation monitoring. This may include, but is not limited to, single-cell overvoltage, single-cell undervoltage, battery pack overvoltage, battery pack undervoltage, discharge overcurrent, charging overcurrent, excessively high temperature, excessively low temperature, excessive temperature difference, insulation fault, relay sticking, abnormal relay disconnection, pre-charge failure, sampling abnormality, communication failure, current sensor failure, voltage sampling failure, and thermal runaway warning. After the battery management system sends the above fault information to the vehicle controller, the vehicle controller can take measures such as alarms, power limiting, prohibition of charging and discharging, high voltage reduction, or vehicle shutdown protection based on the fault level.

[0056] In the embodiments of this application, to solve the problem of circulating current in multiple parallel battery packs, the topology of the parallel battery packs is optimized. Specific optimization measures include: adding a pack-level balancing branch for each parallel battery. This branch is connected in series with a resistor and a relay, and then in parallel to the positive and negative terminals of the corresponding battery pack. The optimized parallel battery pack topology, when needed, controls the relay to close, allowing the battery pack to release some energy through an external resistor, thereby reducing the terminal voltage of the battery pack or weakening the voltage difference between it and other battery packs. This reduces circulating current caused by voltage inconsistency, SOC inconsistency, or internal resistance differences when multiple battery packs are connected in parallel.

[0057] Specifically, in a multi-battery pack parallel system, if the terminal voltage of one battery pack is higher than that of the others, after parallel connection, the higher-voltage battery pack may "backflow" current into the lower-voltage battery pack, forming a circulating current. This circulating current does not perform useful work through the load but circulates internally between the battery packs, causing energy loss, localized heating, accelerated battery aging, and even affecting the safety of relays, wiring harnesses, and busbars. Therefore, by setting up a branch consisting of a resistor and a relay outside each battery pack, when a battery pack's voltage is detected to be too high or the risk of circulating current is high, pack-level discharge or buffer balancing can be performed on that battery pack, gradually bringing its terminal voltage closer to that of the other battery packs, thereby reducing the circulating current during parallel connection or operation.

[0058] It's worth noting that this optimization also improves the system's dynamic adjustment capability. Traditional BMS passive balancing mostly occurs at the cell level, with a relatively small balancing current, making it difficult to handle the pack-level voltage difference between large-capacity battery packs like those in mining trucks. However, the external parallel resistor branch can be configured with a higher-power resistor based on the actual design, providing stronger pack-level regulation capability than cell-level passive balancing. Therefore, this structure is more suitable for high-capacity electric mining truck power battery systems with multiple battery packs connected in parallel.

[0059] by Figure 2 Taking the optimized parallel battery topology shown as an example: In this topology, battery packs PACK1, PACK2, PACK3, and PACK4 are connected in parallel. Each parallel battery pack is equipped with a pack-level balancing branch. Resistor R1 and relay KA1 are connected in parallel across the positive and negative terminals of battery pack PACK1; resistor R2 and relay KA2 are connected in parallel across the positive and negative terminals of battery pack PACK2; resistor R3 and relay KA3 are connected in parallel across the positive and negative terminals of battery pack PACK3; and resistor R4 and relay KA4 are connected in parallel across the positive and negative terminals of battery pack PACK4.

[0060] In the embodiments of this application, after completing the high-voltage downvoltage operation, the battery management system collects the current power-down state data of the battery pack and uploads it to the cloud platform. The cloud platform performs deep learning processing based on the current power-down state data of the battery pack and pre-stored historical operating data to generate a circulating current suppression control strategy, which is then distributed to the vehicle-side storage as a circulating current suppression control strategy for the next high-voltage downvoltage operation. Specifically, the implementation process of the circulating current suppression system based on the optimized parallel battery pack topology to achieve circulating current suppression is as follows: Figure 3 and Figure 4 As shown, it includes the following steps:

[0061] Step S31: After receiving the high-voltage command from the vehicle controller, the battery management system first closes the main negative relay and simultaneously closes the pre-charge relay to start the pre-charge process; when the preset conditions are met, the pre-charge relay is disconnected to exit the pre-charge process, and the main positive relay is closed to make the battery system enter the high-voltage ready state.

[0062] In one embodiment of this application, after receiving a high-voltage power-on command from the vehicle controller, the battery management system enters the high-voltage power-on control process. The battery management system first performs a safety assessment of the battery system status, including but not limited to determining whether the battery total voltage, insulation status, relay status, battery temperature, and fault status meet the high-voltage power-on conditions. If the high-voltage power-on conditions are met, the system controls the main negative relay to close, establishing a connection between the battery negative terminal and the vehicle's high-voltage circuit.

[0063] After the main negative relay closes, the battery management system controls the pre-charge relay to close, enabling the battery system to charge the high-voltage bus or load-side capacitor through the pre-charge circuit with current limitation. Because a pre-charge resistor is installed in the pre-charge circuit, it can limit the inrush current at the moment of power-on, avoiding the risk of relay surge, sudden changes in bus current, or abnormal circulating current between parallel branches caused by direct closure when the voltage difference across the main positive relay is too large.

[0064] For example, after the precharge relay closes, precharge can continue for approximately 400ms to allow the voltage at the back end of the main positive relay to gradually rise and approach the total battery voltage. However, it should be noted that the 400ms precharge duration in this embodiment is not a fixed standard value, but rather an engineering parameter calculated and calibrated based on the precharge resistance value of the specific vehicle's high-voltage system, the high-voltage bus capacitor capacity, the total battery voltage, the target precharge voltage ratio, and the safety margin. The principle for determining the precharge duration is that the precharge resistance in the precharge circuit and the high-voltage bus capacitor form an RC charging circuit. The bus voltage will gradually approach the total battery voltage over time. When the voltage at the back end of the main positive relay reaches a certain proportion of the total battery voltage, such as approaching the total battery voltage or the voltage difference being less than a preset threshold, precharge is considered complete.

[0065] When the battery management system detects that the pre-charge duration has reached the preset time, or detects that the difference between the voltage at the rear end of the main positive relay and the total battery voltage is less than a preset threshold, it determines that pre-charge is complete and can disconnect the pre-charge relay to exit the pre-charge process. At this time, the battery management system controls the main positive relay to close, establishing a connection between the battery's positive terminal and the vehicle's high-voltage bus. After the main positive relay is reliably closed, the pre-charge relay is disconnected to exit the pre-charge process. Thus, the battery system completes the high-voltage connection process and enters the high-voltage ready state, capable of normally supplying power to the vehicle's high-voltage load.

[0066] It should be noted that the main positive relay is located between the positive terminal of the battery pack and the high-voltage positive bus of the vehicle, and is used to control the connection or disconnection of the high-voltage circuit on the positive side of the battery system. After it closes, the positive terminal of the battery establishes a main power connection with the high-voltage system of the vehicle, and it is an important actuator for the battery system to complete the high-voltage connection. The main negative relay is located between the negative terminal of the battery pack and the high-voltage negative bus of the vehicle, and is used to control the connection or disconnection of the high-voltage circuit on the negative side of the battery system. Usually, during the high-voltage connection process, the main negative relay is closed first to establish a connection on the negative side, and then the pre-charge relay and the main positive relay are used to complete the high-voltage connection. The pre-charge relay is located in the pre-charge circuit, and is usually connected in series with the pre-charge resistor and then in parallel across the main positive relay. It is used to perform current-limited charging of the high-voltage bus capacitor through the pre-charge resistor before the main positive relay closes. Its function is to reduce the voltage difference across the main positive relay and avoid the risk of excessive inrush current, arcing, or circulating current when the main positive relay is closed directly.

[0067] Step S32: After receiving the high-voltage reduction command from the vehicle controller, the battery management system executes high-voltage power-down timing control, sequentially disconnecting the main positive relay and the main negative relay to cut off the high-voltage circuit. After the power-down is completed, the current power-down status data of the battery pack is collected and uploaded to the cloud platform.

[0068] In one embodiment of this application, after receiving a high-voltage reduction command from the vehicle controller, the battery management system executes high-voltage power-down timing control. Specifically, the battery management system can first determine whether the current vehicle load state, high-voltage circuit current, and fault state meet the high-voltage reduction conditions; if the high-voltage reduction conditions are met, it sequentially controls the main positive relay and the main negative relay to disconnect, thereby severing the electrical connection between the power battery pack and the vehicle's high-voltage bus, and causing the battery system to exit the high-voltage operating state. By disconnecting the main positive relay and the main negative relay according to a preset timing sequence, the risks of arcing, inrush current, and abnormal backflow during the high-voltage circuit disconnection process can be reduced, improving the safety and controllability of the high-voltage reduction process.

[0069] After the high-voltage power-down is completed, the battery management system collects the current power-down status data of each battery pack and uploads this data to the cloud platform via the vehicle controller or onboard communication terminal. The cloud platform then records the data, analyzes the status, and optimizes subsequent circulating current suppression control strategies. This current power-down status data may include, but is not limited to, information such as battery pack voltage, battery pack current, battery pack temperature, internal resistance, ambient temperature, state of charge, relay disconnection status, fault information, and the time it took to complete the high-voltage power-down. Upon receiving this data, the cloud platform records the status of each battery pack at the moment of the high-voltage power-down, providing a data foundation for subsequently assessing battery pack consistency, analyzing circulating current risks, and generating circulating current suppression strategies for the next high-voltage power-down.

[0070] Step S33: The cloud platform performs deep learning processing based on the current power-down status data of the battery pack and the pre-stored historical operating data to generate a circulating current suppression control strategy and distribute it to the vehicle side; the circulating current suppression control strategy includes a multi-dimensional probability vector and the relay opening duration corresponding to each battery pack; wherein, the multi-dimensional probability vector represents the probability value that each battery pack should be preferentially started to discharge after the next high voltage is applied.

[0071] Specifically, the cloud platform is primarily used to process the current power-down status data of the battery packs uploaded by the vehicle, including data such as voltage differences between battery packs before and after high-voltage downtime, circulating current changes, battery pack temperatures, relay action records, and discharge results. Through deep learning processing of historical operating data, a circulating current suppression control strategy is generated based on a neural network model. This control strategy may include a multi-dimensional probability vector and the relay activation duration corresponding to each battery pack. The multi-dimensional probability vector represents the probability value that each battery pack should preferentially initiate discharge after the next high-voltage downtime; for example, [P1, P2, P3, P4…] correspond to the preferential discharge probabilities of different battery packs. Based on this probability value and historical execution results, the cloud platform determines the target battery pack for preferential discharge and its discharge control duration, thereby forming the optimal control strategy for the next high-voltage downtime condition.

[0072] After the cloud platform generates the control strategy, it distributes the strategy to the vehicle controller or battery management system for local storage. When the vehicle applies the high-voltage downvoltage, it does not execute the current strategy calculated in real-time by the cloud, but rather the control strategy already stored on the vehicle side. Simultaneously, data such as voltage changes, discharge current, temperature changes, relay action time, and equalization results during the high-voltage downvoltage process are continuously uploaded to the cloud platform as input data for training and correcting the next control strategy. This forms a closed-loop optimization mechanism of "executing the previous strategy and generating the next strategy from the current data."

[0073] Furthermore, combined Figure 5 The flowchart of the optimal policy for the neural network is shown, and Figure 6 The neural network structure diagram shown is explained in detail.

[0074] After receiving the current power-down status data of the battery pack uploaded by the vehicle, the cloud platform first summarizes and structures the data to form multi-dimensional input features for the neural network model. These multi-dimensional input features include real-time voltage features, temperature features, internal resistance features, historical features, and operating condition features. For example, real-time voltage features may include the current voltage of each battery pack; temperature features may include the temperature of each battery pack and the ambient temperature; internal resistance features may include the DC internal resistance of each battery pack; historical features may include the voltage difference, discharge duration, relay operation results, and equalization effect during previous high-voltage periods; and operating condition features may include the current of each battery pack, load status, or vehicle operating status. These features characterize the circulating current risk of the battery system from the perspectives of current status, thermal state, aging state, historical trends, and operating conditions.

[0075] The cloud platform preprocesses and normalizes the multi-dimensional input features, enabling data with different dimensions and value ranges to be converted into a unified feature representation suitable for neural network processing. For example, continuous data such as voltage, temperature, internal resistance, and current can be normalized, historical high-voltage data can be arranged into a time-series feature sequence, and different categories of features can be initially assigned importance according to preset weights.

[0076] For example, as shown in Table 1 below, the neural network input features cover multiple dimensions such as current voltage, temperature, internal resistance, historical data, and current conditions. Different weights are assigned to each feature to reflect its influence on the generation of the optimal strategy. The neural network input features mainly include five categories: real-time voltage features, temperature features, internal resistance features, historical features, and operating condition features. Real-time voltage features include four parameters, specifically the current voltage of each battery pack, with a weight of 0.25, the highest among all input features. This indicates that the current voltage of each battery pack is a crucial basis for judging differences between packs and formulating discharge strategies. Temperature features include five parameters, specifically the temperature of each battery pack and the ambient temperature, with a weight of 0.18, reflecting the current thermal state of the battery pack and the impact of the external environment on the safety of discharge control. Internal resistance features include four parameters, specifically the DC internal resistance of each battery pack, with a weight of 0.15, reflecting differences in aging, conductivity, and current distribution among different battery packs. Historical features include 20 features, specifically statistical data from the previous 10 high-voltage periods. Their importance weight is 0.22, second only to real-time voltage features, indicating that the model not only considers the current state but also references voltage changes, processing results, and state evolution patterns during historical high-voltage periods. Operating condition features include 4 features, specifically the current of each battery pack, with an importance weight of 0.20, used to reflect the current actual current state and load distribution of each battery pack. Through the above neural network input features, it can be seen that the cloud platform can consider the current voltage difference, temperature safety boundary, internal resistance difference, historical equalization effect, and the impact of current operating conditions on the circulating current suppression strategy during the model input stage.

[0077] Table 1: Input Features of Neural Networks

[0078] Feature categories Number of features Specific features Importance weight Real-time voltage characteristics 4 Current voltage of each package 0.25 Temperature characteristics 5 Temperature of each package + ambient temperature 0.18 Internal resistance characteristics 4 DC internal resistance of each package 0.15 Historical characteristics 20 Statistics of the first 10 high-pressure data 0.22 Operating characteristics 4 Current of each package 0.20

[0079] Within the neural network, preprocessed features first enter the temporal feature extraction layer to extract trend information about the battery pack's state over time. For example, the model can identify, based on previous high-voltage data, whether a battery pack has been in a high-voltage state for a long time, whether it frequently becomes a priority discharge target, whether the voltage drop effect after discharge is stable, and the variation pattern of the battery pack voltage difference under different ambient temperatures or load conditions. Subsequently, the model further introduces spatial attention and temporal attention mechanisms. The spatial attention mechanism is used to identify the relative differences and correlations between different battery packs, such as which battery pack is more likely to form circulating current compared to other battery packs. The temporal attention mechanism is used to identify key time segments in historical data that are more valuable for the current strategy, such as the voltage recovery status after the most recent high-voltage drops or the relay execution effect. After processing by the above attention mechanisms, the model sends the spatial and temporal enhanced features to the feature fusion layer to form a fused feature that comprehensively represents the battery pack's circulating current risk and discharge control requirements.

[0080] During the strategy generation phase, the cloud platform employs an Actor-Critic structure to perform decision calculations on the fused features. The Actor acts as a policy network, outputting specific control actions, including the probability distribution of each battery pack's priority discharge and corresponding continuous action parameters. For example, the Actor can output a probability vector [P1, P2, P3, P4…], representing the probability score of each battery pack as a priority discharge target; simultaneously, the Actor can also output relay opening duration to determine the closing duration of the discharge relay corresponding to the target battery pack. The Critic acts as a value network, evaluating the state value of the strategy output by the Actor, i.e., determining the comprehensive effect of the current strategy in reducing inter-pack voltage difference, suppressing circulating current, reducing energy waste, avoiding overheating, and improving safety.

[0081] Finally, the cloud platform performs joint optimization based on the probability distribution of the Actor's output, the relay's on-time, and the state value assessment results of the Critic's output to obtain the optimal control strategy after the next high-voltage reduction. The loss function for joint optimization is as follows:

[0082] ;Formula (1)

[0083] in, Represents the total loss function of the joint optimization; Represents the policy loss function; Represents the action regression loss function; Represents the value loss function; , , The weights of each loss function are represented. The strategy loss function measures whether the battery pack discharge priority output by the Actor is beneficial to reducing the voltage difference between packs and suppressing circulating current; the action regression loss measures the deviation between continuous control parameters such as relay opening time and the desired control effect; and the value loss measures the accuracy of the Critic's assessment of the value of the current state. The cloud platform weights and sums the above strategy loss, action regression loss, and value loss to obtain the total loss. By minimizing the total loss, the model parameters are continuously adjusted to make the output discharge target selection, relay opening time, and safety constraints closer to the optimal control result, ultimately generating the optimal circulating current suppression strategy for the next high-voltage period.

[0084] Furthermore, the optimal control strategy can include priority discharge battery pack numbers, discharge priorities for each battery pack, discharge relay activation duration, safety constraints, and strategy termination conditions. The cloud platform distributes the generated optimal control strategy to the vehicle controller or battery management system for local storage, enabling the vehicle to perform pack-level discharge adjustment according to the strategy during the next high-voltage period. Simultaneously, the battery state changes and strategy effects after this execution are continuously uploaded to the cloud platform for subsequent model training and strategy iteration, thus forming a closed-loop optimization process of data acquisition, model calculation, strategy distribution, vehicle execution, and result feedback.

[0085] Furthermore, after receiving the circulating current suppression control strategy, the vehicle-mounted side determines whether the battery status meets the preset activation conditions. If the battery status meets the preset activation conditions, the previously generated optimal processing strategy stored in the vehicle controller is executed, the data is recorded, and the high voltage is applied. If the battery status does not meet the preset activation conditions, the data is recorded and uploaded to the cloud platform.

[0086] It should be noted that the battery status meeting the preset activation conditions means that before executing the circulating current suppression control strategy, the vehicle side determines that the current battery system has both the triggering conditions for requiring pack-level discharge adjustment and the protection conditions for allowing safe discharge. Specifically, the triggering conditions may include the voltage difference between battery packs being greater than a preset voltage difference threshold, the voltage of a certain battery pack being significantly higher than that of other battery packs, the detection of circulating current or the existence of circulating current risk, etc.; the safety protection conditions may include the target battery pack voltage being within the allowable discharge voltage range, the maximum temperature of the battery pack being lower than a preset temperature threshold, the discharge relay and main circuit relay being in normal condition, the insulation being in normal condition, and there being no faults prohibiting execution such as overvoltage, undervoltage, overtemperature, or communication abnormalities, and the current vehicle being in a state where high-voltage discharge adjustment is allowed. Only when both the above triggering conditions and safety protection conditions are met is the battery status considered to meet the preset activation conditions, and the vehicle side can execute the last optimal processing strategy stored in the vehicle controller; otherwise, the discharge relay is not activated, and only the current battery status data is recorded and uploaded to the cloud platform.

[0087] The foregoing has provided a detailed explanation of a circulating current suppression system based on an optimized parallel battery pack topology, as provided in the embodiments of this application. The following will further illustrate this with a specific example, which includes the following steps.

[0088] Step A1: Pre-charge stage (avoid power-on circulating current).

[0089] After receiving the high-voltage command from the vehicle controller, the battery management system (BMS) first enters the pre-charge control phase to avoid the risk of inrush current or circulating current at the moment of power-on, caused by the high-voltage bus capacitor not being charged or differences in the terminal voltages of the parallel battery packs. Specifically, the BMS first controls the main negative relay to close, and after a preset delay, such as 100ms, controls the pre-charge relay to close, allowing the battery pack to perform current-limited charging on the vehicle's high-voltage bus or the capacitor at the rear of the main positive relay through the pre-charge resistor. The pre-charge process lasts for a preset time, such as 400ms, or until the voltage at the rear of the main positive relay reaches a preset proportion of the total battery voltage, such as 90%, at which point the pre-charge is considered complete. Subsequently, the BMS controls the main positive relay to close and the pre-charge relay to open, allowing the battery system to complete the high-voltage connection and enter normal high-voltage operation.

[0090] Step A2: Lowering the high-voltage intelligent control process.

[0091] After the vehicle receives the high-voltage reduction command from the vehicle controller, the battery management system executes the high-voltage reduction timing control and enters the high-voltage reduction intelligent control process, which mainly includes the following stages:

[0092] Phase 1: Data Acquisition and Upload. The battery management system sequentially disconnects the main positive relay and the main negative relay to sever the main power connection between the battery system and the vehicle's high-voltage bus. After the high-voltage period is completed, the system collects the current status information of each battery pack, including but not limited to battery pack voltage, battery pack temperature, battery pack internal resistance, instantaneous current before the high-voltage period, and ambient temperature. This status information is uploaded to the cloud platform via the vehicle communication terminal or other vehicle network to record the high-voltage status and serve as the data basis for subsequent intelligent strategy generation.

[0093] Phase Two: Cloud-based Intelligent Decision Making. After receiving the current high-voltage data, the cloud platform inputs the battery state data during this high-voltage period, historical high-voltage data, historical discharge control effects, and long-term battery pack health trend data into a deep learning model to generate the optimal circulating current suppression control strategy for the next high-voltage period. For example, the deep learning model can be trained based on 1250 sets of high-voltage process data, with a batch size of 32 and a learning rate of 0.001, using the Adam optimizer for parameter updates; the training cycle can be set to 300, and the model validation set accuracy can reach 89.3%. The model's loss function comprehensively considers strategy effectiveness and safety constraints, and its optimization objectives include minimizing inter-pack circulating current after high-voltage periods, balancing losses across battery packs, and ensuring that the battery pack temperature remains within a safe range.

[0094] Phase Three: Strategy Execution and Verification. After generating the control strategy, the cloud platform distributes it to the vehicle controller or battery management system for storage. The control strategy executed by the vehicle after this high-voltage downtime can be the optimal processing strategy generated and distributed by the cloud platform after the previous high-voltage downtime. The data collected and execution results during this high-voltage downtime are then uploaded to the cloud platform to generate the control strategy for the next high-voltage downtime, thus forming a closed-loop control process of "executing the previous strategy and optimizing the next strategy based on the current data."

[0095] As a specific implementation example, the control strategy issued by the cloud platform can include a probability score vector [P1, P2, P3, P4] for prioritizing the discharge of each battery pack, along with the corresponding discharge relay opening duration. For instance, when the probability score vector is [0.15, 0.60, 0.20, 0.05], it indicates that the second battery pack, Bat2, has the highest probability of priority discharge. When the preset opening conditions are met, the vehicle controller controls the corresponding discharge relay of Bat2 to close, allowing Bat2 to discharge in a controlled manner through a 100Ω discharge resistor. During the discharge process, the vehicle controller continuously monitors the battery pack voltage and temperature, limiting the discharge to a safe range where the battery pack voltage is between 550V and 700V and the maximum temperature does not exceed 85℃. Simultaneously, the maximum duration of a single discharge can be set to 60 seconds to prevent prolonged discharge. When the difference between the highest and lowest battery pack voltages is less than or equal to 0.1V, or when the relay opening duration set by the cloud control strategy is reached, the vehicle controller controls the corresponding discharge relay to open, ending the current discharge adjustment process and recording the execution result.

[0096] Finally, to highlight the superior performance of the embodiments of the present invention, a comparison was made between the traditional discharge scheme and the cloud-based intelligent decision-making scheme of this application on several performance indicators, as shown in Tables 2 and 3 below. As can be seen from the tables, the cloud-based intelligent decision-making scheme of this application shows significant improvements over the traditional discharge scheme in dimensions such as maximum circulating current value, average circulating current value, circulating current duration, and circulating current energy loss. Furthermore, it also shows significant improvements in battery capacity consistency, battery internal resistance consistency, and SOC balance.

[0097] Table 2: Comparison of Performance Test Results

[0098] Performance indicators Traditional discharge scheme Cloud-based intelligent decision making Improvement range Maximum circulation value 128.5A 4.8A 96.3% Average circulation value 85.2A 1.9A 97.8% Circulation duration Persistent <30s Circulation energy loss Unable to calculate <0.5kWh

[0099] Table 3: Results of Battery Consistency Improvement

[0100] Performance indicators Traditional discharge scheme Cloud-based intelligent decision making Improvement range Capacity Consistency 87.5% 96.2% 9.9% Internal resistance consistency 18.3% 6.7% 63.4% SOC balance 76.8% 94.5% 23.0%

[0101] Figure 7 A flowchart illustrating a circulating current suppression method based on an optimized parallel battery pack topology according to an embodiment of this application is shown. The method is applied to the circulating current suppression system based on the optimized parallel battery pack topology described in the above embodiment; the method includes:

[0102] Step S71: Receive the power-down status data uploaded by the vehicle side.

[0103] Step S72: Based on the current power-down state data and the pre-stored historical operating data, perform deep learning processing to generate the circulating current suppression control strategy for the pack-level equalization branch corresponding to each battery pack.

[0104] Step S73: Send the circulating current suppression control strategy to the vehicle side for use after the next high voltage is applied.

[0105] It should be noted that the circulating current suppression method based on the optimized parallel battery pack topology provided in this application is similar in implementation and process to the circulating current suppression system based on the optimized parallel battery pack topology described above, and will not be repeated here.

[0106] Figure 8 This is a schematic block diagram of a cloud server provided in an embodiment of this application. Figure 8 As shown, the cloud server 800 includes at least one processor 801, a memory 802, at least one network interface 803, and a user interface 805. The various components in the cloud server 800 are coupled together via a bus system 804. It is understood that the bus system 804 is used to implement communication between these components. In addition to a data bus, the bus system 804 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in… Figure 8The general will label all buses as bus systems.

[0107] The user interface 805 may include a monitor, keyboard, mouse, trackball, clicker, button, touchpad, or touch screen.

[0108] It is understood that memory 802 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM) or programmable read-only memory (PROM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM). The memories described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable categories of memory.

[0109] In this embodiment of the invention, the memory 802 is used to store various categories of data to support the operation of the cloud server 800. Examples of this data include: any executable program for operation on the cloud server 800, such as the operating system 8021 and application programs 8022; the operating system 8021 includes various system programs, such as the framework layer, core library layer, driver layer, etc., for implementing various basic services and handling hardware-based tasks. The application program 8022 may include various applications, such as a media player, browser, etc., for implementing various application services. The methods provided in this embodiment of the invention can be included in the application program 8022.

[0110] The methods disclosed in the above embodiments of the present invention can be applied to or implemented by processor 801. Processor 801 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 801 or by instructions in software form. The processor 801 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 801 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. General-purpose processor 801 may be a microprocessor or any conventional processor, etc. The steps of the accessory optimization method provided in the embodiments of the present invention can be directly reflected as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, which is located in memory. The processor reads the information in the memory and combines it with its hardware to complete the steps of the aforementioned method.

[0111] In an exemplary embodiment, the cloud server 800 may be used by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs) to perform the aforementioned method.

[0112] According to the method provided in the embodiments of this application, this application also provides a computer program product, which includes: computer program code, which, when run on a computer, causes the computer to perform the method of any of the embodiments described above.

[0113] According to the method provided in the embodiments of this application, this application also provides a computer-readable storage medium storing program code, which, when run on a computer, causes the computer to perform the method of any of the embodiments described above.

[0114] As used in this specification, the terms "component," "module," "system," etc., are used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As illustrated, applications running on computing devices and computing devices can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed among two or more computers. Furthermore, these components can be executed from various computer-readable media on which various data structures are stored. Components can communicate, for example, via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0115] Those skilled in the art will recognize that the various illustrative logical blocks and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0116] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0117] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0118] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0119] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0120] In the above embodiments, the functions of each functional unit can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. A computer program product includes one or more computer instructions (programs). When the computer program instructions (programs) are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., high-density digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0121] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0122] In summary, this application provides a circulating current suppression system, method, medium, and server based on an optimized parallel battery pack topology. This application effectively overcomes the various shortcomings of the prior art and has high industrial application value.

[0123] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A circulating current suppression system based on an optimized parallel battery pack topology, characterized by, include: The vehicle includes a vehicle controller, a battery management system, and a vehicle communication terminal on the vehicle side, as well as a cloud platform on the cloud side; the vehicle controller is communicatively connected to the battery management system and the vehicle communication terminal, and the vehicle communication terminal is communicatively connected to the cloud platform. The battery management system connects to and controls the vehicle battery system, which includes multiple battery packs connected in parallel via positive and negative buses. Each battery pack is equipped with a pack-level balancing branch for adjusting the terminal voltage or voltage difference of the corresponding battery pack. After the high voltage is applied, the battery management system collects the current power-down status data of each battery pack and uploads it to the cloud platform. The cloud platform performs deep learning processing based on the current power-down status data and pre-stored historical operating data to generate the circulating current suppression control strategy of the pack-level equalization branch corresponding to each battery pack and sends it to the vehicle side for storage, so that it can be used after the next high voltage application.

2. The system for circulating current mitigation based on optimized parallel battery pack topology of claim 1, wherein, Each battery pack is additionally provided with a pack-level balancing branch for adjusting the battery pack terminal voltage or voltage difference. The pack-level balancing branch is formed by connecting a resistor and a relay in series and then in parallel to the positive and negative terminals of the corresponding battery pack.

3. The circulating current suppression system based on an optimized parallel battery pack topology according to claim 1 or 2, characterized in that, The process by which the circulation suppression system generates a circulation suppression control strategy through a cloud platform includes: After receiving the high-voltage command from the vehicle controller, the battery management system first closes the main negative relay and simultaneously closes the pre-charge relay to start the pre-charge process. When the preset conditions are met, the pre-charge relay is opened to exit the pre-charge process, and the main positive relay is closed to put the battery system into the high-voltage ready state. After receiving the high-voltage reduction command from the vehicle controller, the battery management system executes high-voltage power-down timing control, sequentially disconnecting the main positive relay and the main negative relay to cut off the high-voltage circuit. After the power-down is completed, it collects the current power-down status data of the battery pack and uploads it to the cloud platform. The cloud platform performs deep learning processing based on the current power-off status data of the battery pack and the pre-stored historical operating data to generate a circulating current suppression control strategy and distribute it to the vehicle side.

4. The circulating current suppression system based on the optimized parallel battery pack topology according to claim 3, characterized in that, The step of disconnecting the precharge relay to exit the precharge process when the preset conditions are met includes: when the battery management system detects that the precharge duration has reached the preset time, or detects that the difference between the voltage at the back end of the main positive relay and the total battery voltage is less than the preset threshold, it determines that the precharge is complete and can disconnect the precharge relay to exit the precharge process.

5. The circulating current suppression system based on an optimized parallel battery pack topology according to claim 3, characterized in that, The circulating current suppression control strategy includes a multidimensional probability vector and the relay opening duration corresponding to each battery pack; wherein, the multidimensional probability vector represents the probability value that each battery pack should preferentially start discharging after the next high voltage is applied.

6. The circulating current suppression system based on an optimized parallel battery pack topology according to claim 3, characterized in that, The cloud platform uses a pre-trained neural network to perform deep learning processing based on the current power-off state data and pre-stored historical operating data. The processing includes: The received current power-down state data is structured to form multi-dimensional input features; The multi-dimensional input features are preprocessed and normalized; The preprocessed features are input into the temporal feature extraction layer to extract the trend information of the battery pack state changing over time; the extracted trend information of the battery pack state changing over time is processed based on spatial attention mechanism and temporal attention mechanism respectively to form fused features; The Actor-Critic structure is used to perform decision calculations on the fused features. The strategy output by the Actor network includes the probability distribution of each battery pack to start discharging first and the corresponding relay opening time. The Critic network evaluates the state value of the strategy output by the Actor network. Based on the probability distribution of priority discharge of each battery pack, the relay opening time, and the state value evaluation results of the strategy, the optimal control strategy after the next high voltage is applied is obtained through joint optimization.

7. The circulating current suppression system based on an optimized parallel battery pack topology according to claim 3, characterized in that, After receiving the circulating current suppression control strategy, the vehicle-mounted side determines whether the battery status meets the preset activation conditions. If the battery status meets the preset activation conditions, after completing the high voltage reduction, it executes the previously generated optimal processing strategy stored in the vehicle controller and records the data. If the battery status does not meet the preset activation conditions, it records the data and uploads it to the cloud platform.

8. A circulating current suppression method based on optimized parallel battery pack topology, characterized in that, The method is applied to the circulating current suppression system based on the optimized parallel battery pack topology as described in claim 1; the method includes: Receives front power-down status data uploaded from the vehicle side; Based on the current power-down state data and the pre-stored historical operating data, deep learning processing is performed to generate the circulating current suppression control strategy for the pack-level equalization branch corresponding to each battery pack. The circulating current suppression control strategy is sent to the vehicle side for use after the next high voltage is applied.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the circulating current suppression method based on the optimized parallel battery pack topology as described in claim 8.

10. A cloud server, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the circulating current suppression method based on the optimized parallel battery pack topology as described in claim 8.

Images