Multi-position battery swap management system, control method and device, equipment and medium

Abstract

The application discloses a multi-position battery replacement management system, a control method and device, equipment and a medium. The system comprises a multi-position battery replacement cabinet, the multi-position battery replacement cabinet comprises a plurality of charging positions, storage positions, power taking positions and power returning positions; a carrying mechanism is used for carrying batteries; a controller is connected to the control end of the carrying mechanism, and an automatic scheduling model is arranged in the controller. The automatic scheduling model is used for controlling the carrying mechanism to carry the batteries between the charging positions, the storage positions, the power taking positions and the power returning positions. The automatic scheduling model balances the use of the plurality of charging positions according to a frequency balancing algorithm, and selects batteries for the battery supply of the power taking positions according to a comprehensive scoring algorithm. The application can improve the intelligent degree of the battery replacement cabinet and improve the utilization rate of the positions.

Description

Technical Field

[0001] This application relates to the field of battery charging and swapping, and in particular to a multi-compartment battery swapping management system, control method, device, equipment and medium. Background Technology

[0002] With the rapid popularization of two-wheeled electric vehicles, the demand for battery charging and swapping services is increasing daily. Battery swapping, due to its speed and convenience, has become an important way to replenish energy for two-wheeled electric vehicles. However, existing battery swapping stations still have shortcomings in the following aspects: Currently, battery swapping cabinets in related technologies can only achieve simple charging and discharging control and compartment door opening and closing management. In large battery swapping cabinets with a large number of compartments, there is a lack of effective intelligent management methods for issues such as battery scheduling between compartments, balanced use of charging compartments, and real-time replenishment of power supply compartments, resulting in low compartment utilization.

[0003] In summary, the battery swapping cabinets in related technologies lack sufficient intelligence and suffer from low utilization of storage space. Summary of the Invention

[0004] This application aims to propose a multi-compartment battery swapping management system, control method, device, equipment, and medium, which can improve the intelligence level of the battery swapping cabinet and increase the utilization rate of the compartments.

[0005] In a first aspect, embodiments of this application provide a multi-bay battery swapping management system, including: A multi-compartment battery swapping cabinet, comprising multiple charging compartments, storage compartments, power extraction compartments, and power return compartments; A transport mechanism for transporting batteries between the charging compartment, the storage compartment, the power taking compartment, and the power returning compartment; A controller is connected to the control terminal of the transport mechanism. The controller is equipped with an automatic scheduling model. The automatic scheduling model is used to control the transport mechanism to transport batteries between the charging compartment, the storage compartment, the power taking compartment, and the power returning compartment. The automatic scheduling model balances the use of multiple charging compartments according to a frequency balancing algorithm. The automatic scheduling model selects batteries for power taking compartment replenishment according to a comprehensive scoring algorithm.

[0006] According to some embodiments of this application, the controller is further provided with a micro machine learning module. The micro machine learning module acquires multi-dimensional time-series data accumulated during the operation of the multi-compartment battery swapping cabinet, and performs battery health prediction based on the battery health prediction model, performs charging anomaly detection based on the charging anomaly detection model, and performs compartment demand prediction based on the compartment demand prediction model.

[0007] Secondly, embodiments of this application provide a control method for a multi-bay battery swapping management system, applied to the controller of the multi-bay battery swapping management system as described in the first aspect embodiment, the method comprising: Acquire multi-dimensional time-series data, which is used to indicate the operation process data of the multi-compartment battery swapping cabinet; The multi-dimensional time-series data is input into the battery health prediction model to obtain the battery health prediction result; The multi-dimensional time-series data is input into the charging anomaly detection model to obtain the charging anomaly detection results; A charging adjustment command is generated based on the battery health prediction result and the charging anomaly detection result.

[0008] According to some embodiments of this application, in the step of inputting the multi-dimensional time-series data into the battery health prediction model to obtain the battery health prediction result, the multi-dimensional time-series data includes charging voltage curve, charging current curve, charging time, battery temperature change rate, charge-discharge cycle count, and historical SOH decay trend; the battery health prediction result includes the predicted future SOH value and the predicted remaining cycle life value of the battery.

[0009] According to some embodiments of this application, the step of inputting the multi-dimensional time-series data into the charging anomaly detection model to obtain the charging anomaly detection result includes: The multi-dimensional time-series data is input into the charging anomaly detection model; The charging anomaly detection model calculates a charging anomaly score for each battery based on the multi-dimensional time-series data. The formula for calculating the charging anomaly score is: ; Where score is the charging anomaly rating, D is the input dimension, and x i For the first The original input for each sampling point, For the autoencoder targeting the first Reconstructed output of each sampling point; The charging anomaly detection result is obtained based on the charging anomaly score and the preset dynamic threshold. The expression for the dynamic threshold is as follows: ; in, For dynamic thresholds, The mean, The standard deviation represents the charging anomaly detection results, which include mild anomalies, moderate anomalies, and severe anomalies. like If so, it is considered a mild abnormality; like If so, it is considered moderately abnormal; like If so, it is considered a severe abnormality.

[0010] According to some embodiments of this application, the charging adjustment command includes a first charging adjustment command, a second charging adjustment command, and a third charging adjustment command. Generating the charging adjustment command based on the battery health prediction result and the charging anomaly detection result includes: If the charging anomaly detection result indicates that the battery is slightly abnormal, a first charging adjustment command is generated. The first charging adjustment command is used to instruct the current charging parameters of the corresponding battery to be maintained. If the charging anomaly detection result indicates that the battery is in a moderate abnormality, a second charging adjustment command is generated. The second charging adjustment command is used to instruct the charging current of the corresponding battery to be reduced to 80% of the rated current. If the charging anomaly detection result indicates that the battery is severely abnormal, a third charging adjustment command is generated, which is used to instruct the charging of the corresponding battery to be stopped.

[0011] According to some embodiments of this application, after acquiring multi-dimensional time-series data, the method further includes: The multi-dimensional time-series data is input into the position demand prediction model to obtain the position demand prediction results; A charging adjustment command is generated based on the battery health prediction result, the charging anomaly detection result, and the warehouse demand prediction result.

[0012] Thirdly, embodiments of this application provide a control device for a multi-bay battery swapping management system, applied to the multi-bay battery swapping management system described in the first aspect embodiment, the control device comprising: The data acquisition module is used to acquire multi-dimensional time-series data, which is used to indicate the operation process data of the multi-compartment battery swapping cabinet. The battery health prediction module is used to input the multi-dimensional time-series data into the battery health prediction model to obtain the battery health prediction result. The charging anomaly detection module is used to input the multi-dimensional time-series data into the charging anomaly detection model to obtain the charging anomaly detection result; The charging adjustment module is used to generate charging adjustment commands based on the battery health prediction results and the charging anomaly detection results.

[0013] Fourthly, embodiments of this application provide an electronic device, the device comprising: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the control method of the multi-compartment battery swapping management system as described in the second aspect.

[0014] Fifthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the control method of the multi-bay battery swapping management system as described in the second aspect.

[0015] The multi-bay battery swapping management system, control method, device, equipment, and medium of this application have at least the following beneficial effects: In this embodiment, the multi-compartment battery swapping cabinet is divided into charging compartments, storage compartments, battery collection compartments, and battery return compartments. The controller uses an automatic scheduling model to control the transport mechanism to move batteries between these compartments. The automatic scheduling model uses a frequency balancing algorithm to ensure balanced use of the multiple charging compartments and a comprehensive scoring algorithm to select batteries for battery replenishment in the battery collection compartments. This application achieves autonomous battery scheduling in the battery swapping cabinet through the automatic scheduling model, improving the cabinet's intelligence level. Furthermore, the frequency balancing algorithm for balanced use of the multiple charging compartments and the comprehensive scoring algorithm for battery replenishment in the battery collection compartments significantly improve compartment utilization.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] The present application will be further described below with reference to the accompanying drawings and embodiments, wherein: Figure 1 Architecture diagram of the multi-compartment battery swapping management system provided in this application; Figure 2 A flowchart of the control method for the multi-compartment battery swapping management system provided in this application; Figure 3 A schematic diagram of the control device for the multi-compartment battery swapping management system provided in this application; Figure 4 A schematic diagram of the structure of the electronic device provided in this application. Detailed Implementation

[0018] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0019] In this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, without necessarily requiring or implying any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0020] To address the problems of the prior art, embodiments of this application provide a multi-compartment battery swapping management system, control method, apparatus, equipment, and medium. The multi-compartment battery swapping management system provided in this application embodiment will be described below first.

[0021] refer to Figure 1 As shown, a multi-bay battery swapping management system includes: Multi-compartment battery swapping cabinet, which includes multiple charging compartments, storage compartments, power collection compartments and power return compartments; The handling mechanism is used to move batteries. The controller is connected to the control terminal of the handling mechanism. The controller is equipped with an automatic scheduling model, which is used to control the handling mechanism to move batteries between the charging compartment, storage compartment, power taking compartment and power returning compartment. The automatic scheduling model uses a frequency balancing algorithm to balance the use of multiple charging compartments, and selects batteries for power taking compartment battery replenishment based on a comprehensive scoring algorithm.

[0022] In this embodiment, the multi-compartment battery swapping cabinet is divided into charging compartments, storage compartments, battery collection compartments, and battery return compartments. The controller uses an automatic scheduling model to control the transport mechanism to move batteries between these compartments. The automatic scheduling model uses a frequency balancing algorithm to ensure balanced use of the multiple charging compartments and a comprehensive scoring algorithm to select batteries for battery replenishment in the battery collection compartments. This application achieves autonomous battery scheduling in the battery swapping cabinet through the automatic scheduling model, improving the cabinet's intelligence level. Furthermore, the frequency balancing algorithm for balanced use of the multiple charging compartments and the comprehensive scoring algorithm for battery replenishment in the battery collection compartments significantly improve compartment utilization.

[0023] Specifically, the multi-compartment battery swapping cabinet has N battery compartments arranged in a matrix in two rows, with a movement channel for the handling mechanism between the two rows; the battery compartments are divided into four types, as follows: Charging case: Used to charge the battery; Storage compartment; used to store fully charged batteries for later retrieval; Power collection compartment: serves as the exit for users to remove fully charged batteries; Battery return compartment: serves as the entry point for users to return batteries.

[0024] Specifically, the handling mechanism is driven by a controller to move three motors in the aisle between the front and rear rows of the cabinet, responsible for moving the batteries from one compartment to another. The three motors are the X-axis horizontal movement motor, the Y-axis vertical lifting motor, and the Z-axis forward and backward telescopic motor. Specifically, in this embodiment, the controller uses an embedded processor that supports floating-point operation acceleration and runs a real-time operating system (RTOS). As the central control and decision-making node of the entire system, the controller can implement multi-stage adaptive charging strategies, intelligent battery swapping strategies, etc.

[0025] The multi-stage adaptive charging strategy maintains an independent charging state machine for each charging compartment, including the following states and transition logic: 1) Idle state: No charging activity. When a valid battery connection is detected or an external blind charging command is received, the system switches to blind charging state.

[0026] 2) Blind charging state: For batteries that cannot communicate with the BMS (such as damaged BMS or incompatible communication protocols), charging is performed with low current for a limited time. If valid battery BMS information is successfully received during blind charging, the system switches to normal charging state; if battery information is not obtained after the timeout, the battery is marked as faulty, transported to the storage compartment by the handling mechanism, and returned to idle state.

[0027] 3) Normal charging state: Charging parameters are automatically matched according to the number of battery cells: 15-16 cells in series (48V battery): charging voltage 54.8V, charging current 15A; 20 cells in series (60V battery): charging voltage 73.0V, charging current 15A; 24-cell battery (72V battery): charging voltage 84.0V, charging current 15A.

[0028] The battery status is continuously monitored during charging. If a battery fault is detected (BMS disconnection, abnormal temperature, abnormal voltage), the fault is marked and the transport is triggered. When the SOC reaches 100%, it enters the full charge protection state.

[0029] 4) Fully charged protection state: After the SOC reaches 100%, continue charging at the current parameters for 1 minute, then turn off the charger and fan, and enter the waiting-for-handling state.

[0030] 5) Waiting for transport status: Waiting for the transport mechanism to transport the fully charged battery to the storage compartment. After successful transport, it returns to the idle status.

[0031] The intelligent battery swapping strategy enables full-state machine management of three operations: battery rental, battery swapping, and battery return. The operation process includes electromagnetic lock control, door status detection, RFID authentication, and battery routing. After the operation is completed, the status is reported to the cloud server. Battery routing refers to the process of directly placing the returned battery into the storage compartment if the SOC is >85%, otherwise placing it into the charging compartment.

[0032] It should be noted that the automatic scheduling model can achieve autonomous battery scheduling within the battery swapping station, including battery transfer between the return and charging compartments, battery transfer between the charging and storage compartments, isolation of faulty batteries, and battery replenishment in the power extraction compartment. Specifically, during battery transfer between the return and charging compartments, the automatic scheduling model uses a frequency balancing algorithm to ensure balanced use of multiple charging compartments; during battery replenishment in the power extraction compartment, the automatic scheduling model selects batteries for replenishment based on a comprehensive scoring algorithm, as detailed below: Battery transfer between the return and charging compartments: When a returned battery is detected in the return compartment, an idle charging compartment is automatically selected and the transport mechanism is dispatched to move the battery. The automatic scheduling model uses a frequency balancing algorithm to balance the use of multiple charging compartments. The automatic scheduling model maintains a counter for the cumulative number of uses of each charging compartment and selects the charging compartment with the fewest uses each time to balance the usage frequency of each charging compartment and extend the life of each compartment.

[0033] Battery transfer between charging compartment and storage compartment: When a fully charged battery (SOC≥99%) is detected in the charging compartment, the charger is turned off and the transport mechanism is dispatched to move the battery to an empty storage compartment.

[0034] Faulty battery isolation: When a battery with a fault mark is detected in the charging compartment, it is automatically moved to the storage compartment and the relevant mark is cleared, thus achieving automatic isolation of the faulty battery.

[0035] Battery replenishment in the charging compartment: When no battery is detected in the charging compartment, the optimal battery is selected from the storage compartment for replenishment. Battery selection uses a comprehensive scoring algorithm: Score = SOC × 2 + SOH. Priority is given to the battery with the highest remaining State of Charge (SOC). If SOCs are the same, the battery with the higher State of Health (SOH) is selected. This comprehensive scoring algorithm implements a first-in, first-out (FIFO) strategy, as the earliest fully charged battery naturally has the highest SOC maintenance value.

[0036] It should be noted that in addition to the automatic scheduling tasks in the background, users can also manually issue operation tasks. The background scheduling tasks and user operation tasks are mutually exclusive through binary semaphores. The background scheduling is paused after the user operation locks the semaphore, and the scheduling is resumed after the operation is completed by releasing the semaphore, thus preventing concurrent conflicts.

[0037] In some implementations, the controller is also equipped with a micro machine learning module. The micro machine learning module acquires multi-dimensional time-series data accumulated during the operation of the multi-compartment battery swapping cabinet, and performs battery health prediction based on the battery health prediction model, performs charging anomaly detection based on the charging anomaly detection model, and performs compartment demand prediction based on the compartment demand prediction model.

[0038] In this embodiment, a micro machine learning module is also set up in the controller to perform local intelligent analysis using multi-dimensional time-series data accumulated during device operation. This enables charging anomaly detection, battery health prediction, and warehouse demand prediction without relying on cloud computing resources.

[0039] Specifically, the Tiny Machine Learning (TinyML) module is an edge intelligent inference module deployed on the embedded processor of the controller. The TinyML module can perform local intelligent analysis using multi-dimensional time-series data accumulated during device operation, and can achieve battery health prediction, charging anomaly detection and warehouse demand prediction without relying on cloud computing resources.

[0040] In this embodiment, the charging anomaly detection model employs a lightweight sequence model with time-series modeling capabilities, such as LSTM, GRU, bidirectional LSTM, or a lightweight Transformer. The input features of the charging anomaly detection model include: charging voltage curve, charging current curve, charging duration, battery temperature change rate, charge / discharge cycle count, and historical state of equilibrium (SOH) decay trend. The model outputs the predicted future SOH value and the estimated remaining cycle life.

[0041] The charging anomaly detection model runs on an embedded processor in fixed-point quantization (INT8) format, with a single inference memory usage of no more than 32KB and an inference latency of less than 50ms. The system triggers one inference after each charging cycle, attaching the prediction result to the battery data and reporting it to the cloud. When the predicted SOH (State of Harm) is lower than a set threshold, the system actively marks the battery in an observation list for priority charging monitoring.

[0042] In this embodiment, the charging anomaly detection model employs a lightweight unsupervised anomaly detection model, such as an autoencoder, variational autoencoder (VAE), isolation forest, or local anomaly factor (LOF). During model training, multi-dimensional time-series data from the normal charging process are used to establish a baseline representation of the normal charging pattern, including voltage, current, temperature, and charging duration. During deployment, the charging process is monitored in real time, and the current charging data is input into the model to calculate anomaly scores. When the error exceeds a dynamic threshold, it is considered a charging anomaly. Charging anomalies can include one or more types, such as being categorized as mild, moderate, and severe anomalies, with different strategies employed based on the anomaly type.

[0043] In this embodiment, the battery swapping demand forecasting model employs a lightweight time-series model with sequence prediction capabilities, such as TCN, LSTM, GRU, or lightweight WaveNet. The model predicts future battery swapping demand based on historical battery swapping service data. Input features include: historical battery swapping frequency for each time period, current time, number of currently available fully charged batteries, and number of batteries charging. The model output is the predicted battery swapping demand for the next 1 to 4 hours.

[0044] Based on demand forecasting results, the micro-machine learning module dynamically adjusts the charging strategy: it increases charging power and the number of parallel charging operations in advance before the predicted peak period to ensure a sufficient supply of fully charged batteries during the peak period; and it reduces charging power during off-peak periods to save electricity costs and extend equipment life.

[0045] Figure 2 A flowchart illustrating a control method for a multi-bay battery swapping management system provided in an embodiment of this application is shown. This method is applied to the controller of the above embodiment. A control method for a multi-bay battery swapping management system includes: S101. Obtain multi-dimensional time-series data, which is used to indicate the operation process data of multi-compartment battery swapping cabinets; S102. Input multi-dimensional time-series data into the battery health prediction model to obtain the battery health prediction results; S103. Input the multi-dimensional time series data into the charging anomaly detection model to obtain the charging anomaly detection results; S104. Generate a charging adjustment command based on the battery health prediction results and charging anomaly detection results.

[0046] In this embodiment, multi-dimensional time-series data is first acquired, then input into a battery health prediction model to obtain battery health prediction results. The multi-dimensional time-series data is then input into a charging anomaly detection model to obtain charging anomaly detection results. Finally, a charging adjustment command is generated based on the battery health prediction results and the charging anomaly detection results. This allows for dynamic adjustment of the charging process based on the battery health prediction results and the charging anomaly detection results, improving the system's response speed and autonomous decision-making capability.

[0047] The multi-dimensional time-series data in step S101 above refers to various real-time and historical data during the battery charging process in the charging compartment, such as voltage, current, temperature, charging time, battery cycle count, and SOH.

[0048] In step S102 above, the multi-dimensional time-series data is input into the battery health prediction model to obtain the battery health prediction result. The multi-dimensional time-series data includes the charging voltage curve, charging current curve, charging time, battery temperature change rate, charge-discharge cycle count, and historical SOH decay trend. The battery health prediction result includes the predicted future SOH value and the predicted remaining cycle life value.

[0049] It should be noted that when the predicted SOH is lower than the set threshold, the system will actively mark the battery into the observation list and prioritize charging monitoring.

[0050] In step S103 above, inputting multi-dimensional time-series data into the charging anomaly detection model to obtain the charging anomaly detection result refers to inputting multi-dimensional time-series data into the charging anomaly detection model. The charging anomaly detection model can calculate a charging anomaly score, and when the charging anomaly score exceeds a dynamic threshold, it is determined to be a charging anomaly. Alternatively, it can directly identify the anomaly type through different time-series data. For example, a micro-short circuit anomaly is manifested as a slow rise in charging voltage and a high current, which the model identifies through the deviation of the voltage-current relationship; a cell consistency degradation anomaly is manifested as a deformation of the charging curve relative to the historical baseline, which is identified through comparison of time-series features; and a charger performance degradation anomaly is manifested as a continuous deviation between the actual output power and the set value, which is identified through power efficiency monitoring.

[0051] In step S104 above, generating a charging adjustment command based on the battery health prediction result and the charging anomaly detection result refers to making targeted adjustments to the battery charging process based on the battery's future SOH prediction value, remaining cycle life prediction value, and charging anomaly type. This includes, for example, changing the charging strategy, adjusting the voltage and current, shielding or issuing warnings for batteries with specific anomalies, and other processing.

[0052] In some implementations, inputting multi-dimensional time-series data into a charging anomaly detection model to obtain charging anomaly detection results may include: Input multi-dimensional time-series data into the charging anomaly detection model; The charging anomaly detection model calculates a charging anomaly score for each battery based on multi-dimensional time-series data. The formula for calculating the charging anomaly score is: ; Where score is the charging anomaly rating, D is the input dimension, and x i For the first The original input for each sampling point, For the autoencoder targeting the first Reconstructed output of each sampling point; The charging anomaly detection result is obtained based on the charging anomaly score and the preset dynamic threshold. The expression for the dynamic threshold is as follows: ; in, For dynamic thresholds, The mean, The standard deviation represents the charging anomaly detection results, which include mild anomalies, moderate anomalies, and severe anomalies. like If so, it is considered a mild abnormality; like If so, it is considered moderately abnormal; like If so, it is considered a severe abnormality.

[0053] In this embodiment, the charging anomaly detection model calculates the charging anomaly score for each battery based on multi-dimensional time-series data, and then obtains the charging anomaly detection result based on the charging anomaly score and a preset dynamic threshold. After detecting an anomaly, it automatically executes a graded response. The charging anomaly detection model has higher sensitivity and lower false alarm rate compared to traditional threshold detection by using a three-level threshold mechanism for reconstruction error.

[0054] For example, in the formula for calculating the charging anomaly score, the input dimension D=32, corresponding to 32 sampling points within a sliding window with a sampling interval of 10 seconds, meaning a single window covers approximately 5 minutes of charging process data. The original input x at each sampling pointi The input feature vector of the autoencoder contains four charging state dimensions: charging voltage, charging current, battery temperature, and charging time percentage. The autoencoder targets the first... Reconstructed output of each sampling point This refers to the model applying the learned normal charging patterns to... The restored feature vector.

[0055] During normal charging, the input data conforms to the learned pattern, and the reconstruction error is extremely small (score approaches 0). When charging anomalies occur, the current features deviate from the normal pattern, the autoencoder cannot accurately restore them, and the reconstruction error increases significantly, thus enabling the detection of charging anomalies.

[0056] Upon detecting an anomaly, the system automatically executes a tiered response based on dynamic thresholds. Based on this, the specific judgment rules are as follows: If three consecutive sliding windows are triggered If the above conditions are met, it is considered a mild abnormality. Typical examples include: slight periodic fluctuations in charging current (within ±5%), and occasional slight deviations in the voltage rise curve, but the overall trend is normal.

[0057] If two consecutive sliding windows are triggered If the voltage rise curve deviates significantly from the historical baseline, or the charger output power deviates from the set value by more than 10% continuously.

[0058] If a single sliding window triggers If the charging current suddenly drops to 0, it is considered a severe abnormality. Typical examples include: a temperature rise of more than 10°C within 5 minutes is considered a possible precursor to thermal runaway; and an abnormal reversal of the charging voltage.

[0059] It should be noted that dynamic threshold This is calculated by maintaining the scoring statistics of the 100 most recent normal charging processes.

[0060] In some implementations, the charging adjustment command includes a first charging adjustment command, a second charging adjustment command, and a third charging adjustment command. The charging adjustment command is generated based on the battery health prediction result and the charging anomaly detection result, including: If the charging anomaly detection result indicates that the battery is slightly abnormal, a first charging adjustment command is generated. The first charging adjustment command is used to instruct to maintain the current charging parameters of the corresponding battery. If the charging anomaly detection result indicates that the battery is in a moderate abnormality, a second charging adjustment command is generated. The second charging adjustment command is used to instruct the charging current of the corresponding battery to be reduced to 80% of the rated current. If the charging anomaly detection result indicates that the battery is severely abnormal, a third charging adjustment command is generated, which is used to instruct the charging of the corresponding battery to be stopped.

[0061] In this embodiment, the corresponding charging adjustment command is executed according to the different abnormality levels of the charging abnormality detection results, so as to realize the graded response to charging abnormalities, which is highly targeted and flexible.

[0062] Specifically, in the event of a minor anomaly, a first charging adjustment command is generated to maintain the current charging parameters of the corresponding battery, record the anomaly log, and report it to the cloud. In the event of a moderate anomaly, a second charging adjustment command is generated to reduce the charging current to 80% of the rated value, improve the temperature alarm sensitivity, and report it to the cloud in a timely manner. In the event of a severe anomaly, a third charging adjustment command is generated to immediately stop charging, send a fault marking command to the backend, and simultaneously send a high-priority alarm to the cloud.

[0063] In some implementations, after acquiring multi-dimensional time-series data, the process further includes: Input multi-dimensional time series data into the position demand forecasting model to obtain position demand forecasting results; The charging adjustment command is generated based on the battery health prediction results, charging anomaly detection results, and warehouse demand prediction results.

[0064] In this implementation, the demand forecasting results are obtained through a warehouse demand forecasting model. Then, charging adjustment commands are generated based on the battery health forecasting results, charging anomaly detection results, and warehouse demand forecasting results. Based on the demand forecasting results, the charging strategy can be dynamically adjusted. Charging power and the number of parallel charging operations can be increased in advance before the predicted peak period to ensure a sufficient supply of fully charged batteries during peak periods; charging power can be reduced during off-peak periods to save electricity costs and extend equipment life.

[0065] For example, a demand forecast for battery swapping bays is performed every hour, outputting the predicted battery swapping demand for the next 4 hours. The predicted demand is used to adjust the charging scheduling strategy. When the predicted demand after N hours is greater than 0.8 times the number of currently available fully charged batteries, the system increases the charging priority, increases the number of parallel charging bays, and increases the charging current to the maximum allowable value.

[0066] In summary, this application achieves three main functions—battery health prediction, charging anomaly detection, and battery compartment demand prediction—on the embedded device by incorporating a micro-machine learning module within the controller. This eliminates reliance on cloud computing, reducing communication latency and server load, and improving system response speed and autonomous decision-making capabilities. The charging anomaly detection model, through a three-level threshold mechanism for reconstruction error, exhibits higher sensitivity and a lower false alarm rate compared to traditional threshold detection. The battery compartment demand prediction model can adjust charging power and parallel charging frequency in advance based on predicted peak battery swapping times, achieving dynamic charging control optimization. An autonomous scheduling algorithm enables fully automated operation of battery return, charging, storage, and charging extraction, balancing the number of times charging compartments are used and extending equipment lifespan. Battery comprehensive scoring ensures users receive the optimal battery, and timely replenishment of charging compartments reduces user waiting time, resulting in high scheduling efficiency.

[0067] Based on the control method of the bay battery swapping management system provided in the above embodiments, this application also provides a specific implementation of the control device for the multi-bay battery swapping management system.

[0068] like Figure 3 As shown, the control device 200 of the multi-bay battery swapping management system provided in this application embodiment may include: The data acquisition module 201 is used to acquire multi-dimensional time-series data, which is used to indicate the operation process data of the multi-compartment battery swapping cabinet. The battery health prediction module 202 is used to input the multi-dimensional time-series data into the battery health prediction model to obtain the battery health prediction result. The charging anomaly detection module 203 is used to input the multi-dimensional time-series data into the charging anomaly detection model to obtain the charging anomaly detection result; The charging adjustment module 204 is used to generate a charging adjustment command based on the battery health prediction result and the charging anomaly detection result.

[0069] The control device 200 of the multi-bay battery swapping management system in this application embodiment is used to execute the control method of the multi-bay battery swapping management system in the above embodiment. Its specific processing is the same as the control method of the multi-bay battery swapping management system in the above embodiment, and will not be described in detail here.

[0070] Figure 3 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0071] The electronic device may include a processor 301 and a memory 302 storing computer program instructions.

[0072] Specifically, the processor 301 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0073] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 302 may include removable or non-removable (or fixed) media. Where appropriate, memory 302 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 302 is non-volatile solid-state memory.

[0074] In some embodiments, memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Thus, generally, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this disclosure.

[0075] The processor 301 reads and executes computer program instructions stored in the memory 302 to implement any of the control methods of the multi-bay battery swapping management system in the above embodiments.

[0076] In one example, the electronic device may also include a communication interface 303 and a bus 310. For example, Figure 3 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 310 and complete communication with each other.

[0077] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0078] Bus 310 may include hardware, software, or both. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 310 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0079] Furthermore, in conjunction with the control methods of the multi-bay battery swapping management system in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the control methods of the multi-bay battery swapping management system in the above embodiments.

[0080] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0081] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0082] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0083] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0084] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A multi-bay battery swapping management system, characterized in that, include: A multi-compartment battery swapping cabinet, comprising multiple charging compartments, storage compartments, power extraction compartments, and power return compartments; The transport mechanism is used to transport batteries; A controller is connected to the control terminal of the transport mechanism. The controller is equipped with an automatic scheduling model. The automatic scheduling model is used to control the transport mechanism to transport batteries between the charging compartment, the storage compartment, the power taking compartment, and the power returning compartment. The automatic scheduling model balances the use of multiple charging compartments according to a frequency balancing algorithm. The automatic scheduling model selects batteries for power taking compartment replenishment according to a comprehensive scoring algorithm.

2. The multi-bay battery swapping management system according to claim 1, characterized in that, The controller is also equipped with a micro machine learning module, which acquires multi-dimensional time-series data accumulated during the operation of the multi-compartment battery swapping cabinet, and performs battery health prediction based on the battery health prediction model, performs charging anomaly detection based on the charging anomaly detection model, and performs compartment demand prediction based on the compartment demand prediction model.

3. A control method for a multi-bay battery swapping management system, characterized in that, The method, applied to the controller of the multi-bay battery swapping management system according to claim 2, comprises: Acquire multi-dimensional time-series data, which is used to indicate the operation process data of the multi-compartment battery swapping cabinet; The multi-dimensional time-series data is input into the battery health prediction model to obtain the battery health prediction result; The multi-dimensional time-series data is input into the charging anomaly detection model to obtain the charging anomaly detection results; A charging adjustment command is generated based on the battery health prediction result and the charging anomaly detection result.

4. The control method for the multi-bay battery swapping management system according to claim 3, characterized in that, In the step of inputting the multi-dimensional time-series data into the battery health prediction model to obtain the battery health prediction result, the multi-dimensional time-series data includes the charging voltage curve, charging current curve, charging time, battery temperature change rate, charge-discharge cycle count, and historical SOH decay trend; the battery health prediction result includes the predicted future SOH value and the predicted remaining cycle life value of the battery.

5. The control method for the multi-bay battery swapping management system according to claim 3, characterized in that, The step of inputting the multi-dimensional time-series data into the charging anomaly detection model to obtain the charging anomaly detection result includes: The multi-dimensional time-series data is input into the charging anomaly detection model; The charging anomaly detection model calculates a charging anomaly score for each battery based on the multi-dimensional time-series data. The formula for calculating the charging anomaly score is: ； Where score is the charging anomaly rating, D is the input dimension, and x i For the first The original input for each sampling point, For the autoencoder targeting the first Reconstructed output of each sampling point; The charging anomaly detection result is obtained based on the charging anomaly score and the preset dynamic threshold. The expression for the dynamic threshold is as follows: ； in, For dynamic thresholds, The mean, The standard deviation represents the charging anomaly detection results, which include mild anomalies, moderate anomalies, and severe anomalies. like If so, it is considered a mild abnormality; like If so, it is considered moderately abnormal; like If so, it is considered a severe abnormality.

6. The control method for the multi-bay battery swapping management system according to claim 5, characterized in that, The charging adjustment command includes a first charging adjustment command, a second charging adjustment command, and a third charging adjustment command. Generating the charging adjustment command based on the battery health prediction result and the charging anomaly detection result includes: If the charging anomaly detection result indicates that the battery is slightly abnormal, a first charging adjustment command is generated. The first charging adjustment command is used to instruct the current charging parameters of the corresponding battery to be maintained. If the charging anomaly detection result indicates that the battery is in a moderate abnormality, a second charging adjustment command is generated. The second charging adjustment command is used to instruct the charging current of the corresponding battery to be reduced to 80% of the rated current. If the charging anomaly detection result indicates that the battery is severely abnormal, a third charging adjustment command is generated, which is used to instruct the charging of the corresponding battery to be stopped.

7. The control method for the multi-bay battery swapping management system according to claim 3, characterized in that, After acquiring the multi-dimensional time-series data, the process also includes: The multi-dimensional time-series data is input into the position demand prediction model to obtain the position demand prediction results; A charging adjustment command is generated based on the battery health prediction result, the charging anomaly detection result, and the warehouse demand prediction result.

8. A control device for a multi-bay battery swapping management system, characterized in that, The control device, applied to the multi-bay battery swapping management system of claim 2, comprises: The data acquisition module is used to acquire multi-dimensional time-series data, which is used to indicate the operation process data of the multi-compartment battery swapping cabinet. The battery health prediction module is used to input the multi-dimensional time-series data into the battery health prediction model to obtain the battery health prediction result. The charging anomaly detection module is used to input the multi-dimensional time-series data into the charging anomaly detection model to obtain the charging anomaly detection result; The charging adjustment module is used to generate charging adjustment commands based on the battery health prediction results and the charging anomaly detection results.

9. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the control method of the multi-bay battery swapping management system as described in any one of claims 3-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the control method of the multi-compartment battery swapping management system as described in any one of claims 3-7.

Images