High-voltage energy storage system protection control method and device, equipment, storage medium and optimization method thereof

Abstract

The application discloses a high-voltage energy storage system protection control method and device, equipment, a storage medium and an optimization method thereof, and relates to the technical field of battery management. The method comprises the following steps: acquiring battery cluster data of a high-voltage energy storage system; processing the battery cluster data based on a CUDA parallel computing architecture to obtain battery cluster parameters; diagnosing the fault of the battery cluster based on D-S evidence theory according to the battery cluster parameters, and recording the fault data of the fault battery cluster; repairing the fault of the fault battery cluster according to the fault data of the fault battery cluster, and restarting the fault battery cluster in multiple stages. In the process of protecting the high-voltage energy storage system, the application reduces the data processing delay, improves the working condition adaptability, improves the fault identification accuracy, and reduces the fault diffusion risk.

Description

Technical Field

[0001] This application relates to the field of battery management technology, specifically to a protection and control method, device, equipment, storage medium, and optimization method for a high-voltage energy storage system. Background Technology

[0002] With the advancement of the "dual carbon" goals, electrochemical energy storage, as a core supporting technology for new energy consumption and grid peak shaving, is ushering in large-scale development. 1500V high-voltage energy storage systems have become the mainstream technology for grid-side and user-side energy storage power stations due to their high energy density (approximately 30% higher than 1000V systems), low cable and equipment costs, and high charge and discharge efficiency (≥92%).

[0003] Such systems typically use 416 or more ternary lithium batteries or lithium iron phosphate batteries connected in series to form a single battery cluster, with a rated voltage of up to 1500V. They possess the following technical characteristics: First, a large number of cells: a single battery cluster contains 416 cells, requiring the simultaneous acquisition of 416 voltage and 416 temperature data points, resulting in a massive data volume. Second, high voltage level: 1500V high voltage places stringent requirements on insulation performance, fault isolation, and personal safety protection; the consequences of short circuits, leakage, and other faults are more severe. Third, high power density: the rated current of a single battery cluster in a large energy storage power station reaches 800A-1000A, and the overload current can reach 1200A, resulting in high risks of overcurrent and overheating. Fourth, rapid fault propagation: single-cell faults (such as overvoltage or thermal runaway) can easily propagate to the entire cluster through the series circuit, even triggering a cascading failure of the entire cabinet / station.

[0004] However, existing BMS technologies still have many bottlenecks and cannot meet the above requirements, specifically including the following defects.

[0005] High data processing latency: Traditional BMS systems mostly use MCUs with ARM Cortex-A53 / A72 architecture, which process massive amounts of data based on CPU serial computing. For voltage and temperature data of 416-cell batteries, the calculation time for extreme value extraction, mean value calculation, and temperature difference analysis is usually 50ms-100ms, far exceeding the ≤20ms response requirement specified in GB / T 42268-2022, and thus cannot achieve rapid protection against instantaneous faults such as short circuits and overvoltages.

[0006] Static protection thresholds: Existing BMS protection thresholds are mostly fixed values ​​(e.g., 4.2V overvoltage, 1.2 times rated overcurrent), without considering the impact of operating conditions on cell characteristics. For example, at low temperatures (≤0℃), the cell internal resistance increases and the risk of lithium deposition rises; using a 4.2V overvoltage threshold can easily lead to overcharging. At high temperatures (≥45℃), the risk of thermal runaway is aggravated; a fixed temperature rise rate threshold (5℃ / min) can easily lead to leakage triggering. When the power grid fluctuates, the insulation resistance is prone to falsely low values; a fixed anti-jitter window can easily lead to false triggering.

[0007] Insufficient fault propagation control: When a single battery cluster triggers a severe fault, traditional BMS only disconnects the main circuit of that cluster without taking restrictive measures for adjacent battery clusters. Under high-voltage conditions, voltage surges and temperature conduction in the faulty cluster can easily trigger cascading faults in adjacent clusters, leading to the shutdown of the entire cabinet or even the entire station.

[0008] Inadequate reliability design: The data acquisition process lacks redundancy, and failure of a single voltage acquisition channel or current sensor may lead to misjudgment or failure of protection; the software has not been optimized for real-time performance, and issues such as scheduling delays and memory paging in the Linux system can easily lead to delayed response of protection commands; insufficient suppression of electromagnetic interference under high-voltage conditions results in distorted data acquisition and high packet loss rate in communication, affecting the accuracy of protection logic.

[0009] Low fault identification accuracy: Existing BMS mostly use a single parameter criterion (such as judging overcharge only by voltage), without considering factors such as sensor error and electromagnetic interference, resulting in a high false trigger rate and affecting the continuity of system operation. Summary of the Invention

[0010] In view of this, the technical problem to be solved by this application is: how to reduce data processing delay, improve operating condition adaptability, improve fault identification accuracy, and reduce the risk of fault propagation in the process of protecting high-voltage energy storage systems.

[0011] To solve the above-mentioned technical problems, this application provides the following technical solution.

[0012] Firstly, a protection and control method for a high-voltage energy storage system includes:

[0013] Acquire battery cluster data from a high-voltage energy storage system;

[0014] The battery cluster data is processed using the CUDA parallel computing architecture to obtain battery cluster parameters.

[0015] Based on the battery cluster parameters, the faults of the battery clusters are diagnosed using DS evidence theory, and the fault data of the faulty battery clusters are recorded.

[0016] Based on the fault data of the faulty battery cluster, the faulty battery cluster is repaired and restarted in multiple stages.

[0017] As an alternative to the first aspect, the battery cluster parameters include at least one of voltage parameters, current parameters, temperature parameters, and insulation parameters.

[0018] As an optional solution to the first aspect, the step of diagnosing battery cluster faults based on DS evidence theory according to the battery cluster parameters and recording fault data of the faulty battery cluster includes:

[0019] Based on the battery cluster parameters, establish multiple corresponding battery cluster sub-criteria;

[0020] Based on the DS evidence theory, a basic probability allocation is set for each of the corresponding battery cluster sub-criteria.

[0021] By fusing the basic probability assignments of each of the battery cluster sub-criteria, the fusion confidence of the corresponding multiple battery cluster criteria is obtained;

[0022] Based on the fusion confidence of the multiple battery cluster sub-criteria, a corresponding multi-parameter fusion criterion is established;

[0023] The faults of battery clusters are diagnosed and the fault data of faulty battery clusters are recorded by using the multi-parameter fusion criteria.

[0024] As an alternative to the first aspect, the plurality of battery cluster sub-criteria includes at least one of a plurality of voltage sub-criteria, a plurality of current sub-criteria, and a plurality of temperature sub-criteria;

[0025] The plurality of voltage sub-criteria are established based on the voltage parameter, the plurality of current sub-criteria are established based on the current parameter, and the plurality of temperature sub-criteria are established based on the temperature parameter;

[0026] The multi-parameter fusion criteria include at least one of the following: overvoltage protection criteria, overcurrent protection criteria, and temperature rise protection criteria;

[0027] The overvoltage protection criterion is established based on the fusion confidence level of the multiple voltage sub-criterions, the overcurrent protection criterion is established based on the fusion confidence level of the current sub-criterion, and the temperature rise protection criterion is established based on the fusion confidence level of the multiple temperature sub-criterions.

[0028] As an alternative to the first aspect, the plurality of voltage sub-criteria include:

[0029] First overvoltage criterion: The maximum voltage of a single cell in a battery cluster is greater than or equal to the first voltage threshold.

[0030] Second overvoltage criterion: Total voltage of battery cluster ≥ second voltage threshold;

[0031] Third overvoltage criterion: Overvoltage duration of battery cluster ≥ third voltage threshold;

[0032] Wherein, the maximum voltage of a single cell string in the battery cluster, the total voltage of the battery cluster, and the overvoltage duration of the battery cluster are all voltage parameters;

[0033] The formula for calculating the fusion confidence of the multiple voltage sub-criteria is as follows:

[0034] ;

[0035] In the formula: This represents the fusion confidence level of the multiple voltage sub-criteria; This represents the fusion value of the basic probability allocation of the first overvoltage subcriterion and the second overvoltage subcriterion; This represents the basic probability allocation of the third overpressure criterion; This represents the conflict coefficient between the first and second overvoltage criteria and the third overvoltage criterion after their basic probability allocation fusion; " indicates the XOR operation;

[0036] The overvoltage protection criterion is established as follows: the fusion confidence level of the multiple voltage sub-criterions is greater than or equal to the first confidence threshold; if the overvoltage protection criterion is valid, then the fault of the battery cluster is an overvoltage fault.

[0037] As an alternative to the first aspect, the plurality of current sub-criteria include:

[0038] First overcurrent criterion: The charging and discharging current of the battery cluster is greater than or equal to the first current threshold.

[0039] Second current criterion: The rate of increase of the charging and discharging current of the battery cluster is greater than or equal to the second current threshold.

[0040] Third overcurrent criterion: Overcurrent duration of battery cluster ≥ third current threshold;

[0041] Wherein, the charging and discharging current of the battery cluster, the rate of increase of the charging and discharging current of the battery cluster, and the overcurrent duration of the battery cluster are all current parameters.

[0042] The formula for calculating the fusion confidence of the multiple current sub-criteria is as follows:

[0043] ;

[0044] In the formula: This represents the fusion confidence level of the multiple current sub-criteria; This represents the fusion value of the basic probability allocation of the first overcurrent sub-criteria and the second overcurrent sub-criteria; This represents the basic probability allocation of the third overcurrent sub-criterion; This represents the conflict coefficient between the first and second overcurrent sub-criteria after their basic probability allocation fusion and the third overcurrent criterion; " indicates the XOR operation;

[0045] The overcurrent protection criterion is established as follows: the fusion confidence level of the multiple current sub-criterions is greater than or equal to the second confidence threshold; if the overcurrent protection criterion is valid, then the fault of the battery cluster is an overcurrent fault.

[0046] As an alternative to the first aspect, the plurality of temperature sub-criteria include:

[0047] First temperature rise criterion: The maximum temperature of a single cell string in the battery cluster is ≥ the first temperature threshold.

[0048] Second temperature rise criterion: The rate of temperature rise of a single cell string within a battery cluster is greater than or equal to the second temperature threshold.

[0049] The third temperature rise criterion is: the temperature difference between three adjacent strings of cells within a battery cluster is greater than or equal to the third temperature threshold.

[0050] The temperature parameters are: the maximum temperature of a single cell string within the battery cluster, the temperature rise rate of a single cell string within the battery cluster, and the temperature difference between three adjacent cells strings within the battery cluster.

[0051] The formula for calculating the fusion confidence of the multiple temperature sub-criterions is as follows:

[0052] ;

[0053] In the formula: This represents the fusion confidence level of the multiple temperature sub-criterions; This represents the fusion value of the basic probability allocation of the first temperature riser criterion and the second temperature riser criterion; This represents the basic probability allocation of the third temperature riser criterion; This represents the conflict coefficient between the first and second temperature riser criteria and the third temperature riser criterion after their basic probability allocation is fused; " indicates the XOR operation;

[0054] The temperature rise protection criterion is established as follows: the fusion confidence level of the multiple temperature sub-criterions is greater than or equal to the third confidence threshold; if the temperature rise protection criterion is valid, then the fault of the battery cluster is a temperature rise fault.

[0055] As an alternative to the first aspect, the method further includes:

[0056] It provides Level 1 protection, Level 2 protection, and Level 3 protection, along with their respective triggering conditions and response time limits; the Level 1 protection has a higher priority than the Level 2 protection, and the Level 2 protection has a higher priority than the Level 3 protection.

[0057] Based on the fault data of the battery cluster, the triggering conditions of the first-level protection, second-level protection and third-level protection are determined;

[0058] Based on the judgment result, the first-level protection, the second-level protection, or the third-level protection are executed in descending order of priority.

[0059] As an alternative to the first aspect, the method further includes:

[0060] Acquire operational data from the high-voltage energy storage system;

[0061] Based on the operating data of the high-voltage energy storage system and the parameters of the battery cluster, the operating parameters are obtained by analysis using the CUDA parallel computing architecture.

[0062] Based on the battery cluster parameters and the operating condition parameters, a multi-dimensional operating condition evaluation model is established.

[0063] Based on the multi-dimensional operating condition evaluation model, the triggering conditions or response time limits of the first-level protection, the second-level protection, or the third-level protection are adjusted.

[0064] As an alternative to the first aspect, the primary protection includes: shutting off the main contactor and precharge contactor of the battery cluster and triggering the fire-fighting linkage device of the high-voltage energy storage system.

[0065] The triggering conditions for the first-level protection include at least one of the following: the maximum voltage of a single cell string in the battery cluster is ≥ the first protection threshold, the minimum voltage of a single cell string in the battery cluster is ≤ the second protection threshold, the charging and discharging current of the battery cluster is ≥ the third protection threshold, the temperature rise rate of a single cell string in the battery cluster is ≥ the fourth protection threshold, and the rate of change of the insulation resistance of the positive and negative electrodes of the battery cluster to ground is ≥ the fifth protection threshold.

[0066] The secondary protection includes: cutting off the charging circuit of the battery cluster, limiting the discharge power of the battery cluster to ≤20% of the rated power of the battery cluster, and sending a fault alarm signal to the BMS of the high-voltage energy storage system.

[0067] The triggering conditions for the secondary protection include at least one of the following: the total voltage of the battery cluster ≥ the sixth protection threshold, the total voltage of the battery cluster ≤ the seventh protection threshold, the charge / discharge current rate of the battery cluster ≥ the eighth protection threshold, and the overcurrent duration of the battery cluster ≥ the ninth protection threshold.

[0068] The three-level protection includes: triggering the audible and visual alarm device of the high-voltage energy storage system and limiting the discharge power of the battery cluster to ≤50% of the rated power of the battery cluster;

[0069] The triggering conditions for the three-level protection include at least one of the following: the deviation of the equalization current of the battery cluster is greater than or equal to the tenth protection threshold, the communication interruption time of the high-voltage energy storage system is greater than or equal to the eleventh protection threshold, and the cooling system of the high-voltage energy storage system fails.

[0070] The response time of the first-level protection is shorter than that of the second-level protection, and the response time of the second-level protection is shorter than that of the third-level protection.

[0071] in:

[0072] The maximum voltage of a single cell string within the battery cluster, the minimum voltage of a single cell string within the battery cluster, and the total voltage of the battery cluster are all the voltage parameters.

[0073] The charging and discharging current of the battery cluster, the charging and discharging current ratio of the battery cluster, the overcurrent duration of the battery cluster, and the equilibrium current deviation of the battery cluster are all current parameters.

[0074] The temperature rise rate of a single string of cells within the battery cluster is the temperature parameter.

[0075] The rate of change of the insulation resistance of the positive and negative electrodes of the battery cluster to ground is the insulation parameter;

[0076] The data on the communication interruption time of the high-voltage energy storage system and the data on the cooling system of the high-voltage energy storage system are both operational data of the high-voltage energy storage system.

[0077] As an alternative to the first aspect, the multi-dimensional working condition assessment model includes:

[0078] Low temperature condition: The minimum temperature of a single cell string within the battery cluster is ≤ the threshold of the first operating condition;

[0079] High-temperature operating conditions: The maximum temperature of a single cell string within the battery cluster is ≥ the threshold of the second operating condition;

[0080] Overload condition: The charge / discharge current rate of the battery cluster is ≥ the threshold of the third condition;

[0081] Low SOC condition: SOC of battery cluster ≤ fourth condition threshold;

[0082] High SOC condition: SOC of the battery cluster ≥ threshold of condition 5;

[0083] Power grid fluctuation condition: Power grid voltage fluctuation amplitude ≥ threshold of condition 6;

[0084] in:

[0085] The minimum temperature of a single cell string within the battery cluster and the maximum temperature of a single cell string within the battery cluster are the temperature parameters.

[0086] The SOC of the battery cluster and the voltage fluctuation of the grid are both operating parameters.

[0087] As an optional solution to the first aspect, adjusting the triggering conditions or response time limits of the primary protection, the secondary protection, or the tertiary protection based on the multi-dimensional operating condition evaluation model includes:

[0088] Based on the aforementioned low-temperature operating condition, when the low-temperature operating condition is established, the first protection threshold and the eighth protection threshold are lowered.

[0089] Based on the aforementioned high-temperature operating condition, when the high-temperature operating condition is established, the fourth protection threshold is lowered.

[0090] Based on the overload condition, when the overload condition is established, the response time limit of the sixth protection threshold, the ninth protection threshold, and the secondary protection is reduced.

[0091] Based on the low SOC condition, when the low SOC condition is established, the second protection threshold is increased.

[0092] Based on the high SOC condition, when the high SOC condition is established, the first protection threshold is lowered.

[0093] Based on the power grid fluctuation condition, when the power grid fluctuation condition is established, the anti-jitter window of the fifth protection threshold is extended.

[0094] As an alternative to the first aspect, the method further includes:

[0095] Based on the judgment result, if the triggering conditions of the first-level protection or the second-level protection are met, the connection between the faulty battery cluster and the power grid will be cut off by the PCS, and the charging and discharging permissions of the faulty battery cluster and its adjacent battery clusters will be locked.

[0096] As an optional solution to the first aspect, the step of repairing the faulty battery cluster based on the fault data of the faulty battery cluster and restarting the faulty battery cluster in multiple stages includes:

[0097] Based on the fault data of the faulty battery cluster, the faulty battery cluster is repaired by restarting the faulty battery cluster in three stages: pre-charge stage, low-power discharge stage, and rated power recovery stage.

[0098] in:

[0099] The pre-charge phase includes: closing the pre-charge circuit of the faulty battery cluster within 5 seconds, limiting the pre-charge current of the faulty battery cluster to ≤ the first restart threshold, until the total voltage change rate of the faulty battery cluster is ≤ the second restart threshold.

[0100] The low-power discharge stage includes: limiting the discharge current of the faulty battery cluster to the third restart threshold within 10 seconds, until the difference between the maximum and minimum voltages of the single-string cells in the faulty battery cluster is ≤ the fourth restart threshold and the temperature change rate of the single-string cells in the faulty battery cluster is ≤ the fifth restart threshold.

[0101] The rated power recovery phase includes: gradually increasing the charging and discharging power of the faulty battery cluster within 30 seconds; if the faulty battery cluster no longer malfunctions, then the normal operation of the faulty battery cluster is restored.

[0102] As an optional solution to the first aspect, the acquisition of battery cluster data of the high-voltage energy storage system includes:

[0103] Voltage data of each cell in the battery cluster is collected synchronously through multiple primary channels and multiple backup channels.

[0104] The charging and discharging current data of the battery cluster are collected synchronously through one second main channel and one second backup channel;

[0105] Temperature data of each cell in the battery cluster is collected synchronously through multiple third main channels and multiple third backup channels;

[0106] The insulation resistance data of the positive and negative terminals of the battery cluster to ground are collected simultaneously through two fourth channels.

[0107] in:

[0108] The first main channel of the Nth channel and the first backup channel of the Nth channel simultaneously collect the voltage data of the Nth string of cells in the battery cluster;

[0109] The second main channel and the second backup channel are connected in parallel to the charging and discharging current circuit of the battery cluster;

[0110] The third main channel and the third backup channel of the Nth path synchronously collect temperature data of the Nth string of cells in the battery cluster;

[0111] One of the fourth channels constructs a positive-to-ground balance bridge circuit for the battery cluster, and the other of the fourth channels constructs a negative-to-ground balance bridge circuit for the battery cluster.

[0112] As an optional solution to the first aspect, the acquisition of battery cluster data of the high-voltage energy storage system further includes:

[0113] The battery cluster data is preprocessed using a combination algorithm of sliding window filtering and Kalman filtering.

[0114] Secondly, this application provides a protection and control device for a high-voltage energy storage system, comprising:

[0115] The data acquisition unit is used to acquire battery cluster data of the high-voltage energy storage system;

[0116] The data processing unit is used to process the battery cluster data based on the CUDA parallel computing architecture to obtain battery cluster parameters;

[0117] The fault diagnosis unit is used to diagnose the faults of the battery clusters based on the battery cluster parameters and the DS evidence theory, and to record the fault data of the faulty battery clusters.

[0118] The fault repair unit is used to repair the fault of the faulty battery cluster based on the fault data of the faulty battery cluster, and restart the faulty battery cluster in multiple stages.

[0119] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method provided by the first aspect of this application or any implementation thereof.

[0120] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method provided by the first aspect of this application or any implementation thereof.

[0121] Fifthly, this application provides an optimization method, comprising:

[0122] Configure the software architecture; the software architecture is based on the Ubuntu 20.04 LTS operating system, loads the PREEMPT_RT real-time kernel, and includes:

[0123] Thread management: The first-level protected thread is bound to CPU core 0, the second-level protected thread is bound to CPU core 1, the third-level protected thread is bound to CPU core 2, and the computing tasks of the CUDA parallel computing architecture are bound to CPU core 3 and GPU.

[0124] Memory optimization: Lock all code and data pages using the mlockall(MCL_CURRENT\MCL_FUTURE) function to prevent Linux memory paging;

[0125] Process Guardian: Create a guardian unit to protect the logical process through systemd, configure automatic restart when the process crashes, and automatically load the most recently saved battery cluster parameters and operating condition parameters after restarting;

[0126] Fault Log: The storage method combines local SD card storage and cloud backup to record the battery cluster data, battery cluster parameters, status of the first-level protection thread, status of the second-level protection thread, status of the third-level protection thread, and operating parameters within 10 seconds before and after the fault. The log retains the most recent 100 entries and supports cyclic overwriting.

[0127] The present application adopts the above technical solution, and compared with the prior art, it mainly has the following technical effects:

[0128] For the massive data of hundreds of cells in a battery cluster, the CUDA parallel computing architecture divides the computing tasks into local operations within blocks and global summaries between blocks, improving data processing speed and ensuring computational accuracy, which meets the core requirement of millisecond-level protection for high-voltage energy storage systems.

[0129] By constructing a multi-parameter fusion criterion using DS evidence theory, voltage, current, and temperature parameters are used as multi-source evidence, and each is assigned a basic probability allocation for fault identification. The fault characteristics characterized by each battery cluster sub-criterion are integrated. Compared with the case of considering only a single battery cluster parameter, this approach suppresses false fault identification and improves the accuracy of fault identification.

[0130] Based on the severity of faults in high-voltage energy storage systems, protection measures are divided to form a tiered protection mechanism, which balances system safety and operational continuity through differentiated response actions.

[0131] An adaptive threshold adjustment mechanism is formed by a multi-dimensional operating condition evaluation model, which dynamically adapts to complex operating conditions such as low temperature, high temperature, overload, and power grid fluctuations, reducing the false trigger rate and missed trigger rate of protection measures at all levels, and solving the problem of poor adaptability of traditional static thresholds.

[0132] For primary and secondary protection, fault cascading suppression measures are established so that when a single battery cluster triggers primary or secondary protection, not only is its own circuit cut off, but also the charging and discharging permissions of its adjacent battery clusters are locked, thereby suppressing fault escalation and reducing the risk of the entire station shutting down due to fault propagation.

[0133] By implementing a tiered restart process from pre-charging to low-power discharging and then to rated power recovery, the fault repair results of the battery cluster are verified in a stepwise manner. This avoids secondary faults caused by the faulty battery clusters being directly operated at full load after repair, thereby improving the reliability of fault repair and restart.

[0134] By using a dual-channel approach to synchronously collect and redundantly back up battery cluster data, the efficiency and reliability of battery cluster data collection are improved.

[0135] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0136] To more clearly illustrate the technical solutions in this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0137] Figure 1This is a flowchart illustrating the method provided in an embodiment of this application;

[0138] Figure 2 This is a schematic diagram of the processing flow of the CUDA parallel computing framework in the method provided in the embodiments of this application;

[0139] Figure 3 yes Figure 2 The processing timeline of the CUDA parallel computing framework;

[0140] Figure 4 This is a flowchart illustrating step S300 of the method provided in the embodiments of this application;

[0141] Figure 5 This is a schematic diagram of the structure of the device provided in the embodiments of this application;

[0142] Figure 6 This is a schematic diagram of the hardware architecture provided in the embodiments of this application;

[0143] Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0144] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0145] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or modules is not limited to the steps or modules listed, but may optionally include steps or modules not listed, or may optionally include other steps or modules inherent to such process, method, product, or apparatus.

[0146] See Figure 1 To reduce data processing latency, improve operational adaptability, enhance fault identification accuracy, and reduce the risk of fault propagation during the protection of high-voltage energy storage systems, this application provides a protection and control method for high-voltage energy storage systems, comprising the following steps:

[0147] S100, acquires battery cluster data of high-voltage energy storage system;

[0148] S200 processes battery cluster data based on the CUDA parallel computing architecture to obtain battery cluster parameters;

[0149] S300 diagnoses battery cluster faults based on battery cluster parameters and DS evidence theory, and records fault data of faulty battery clusters.

[0150] The S400 repairs the faults of the faulty battery clusters based on their fault data and restarts them in multiple stages.

[0151] A battery cluster is the basic energy unit of an energy storage system, consisting of multiple series-connected cells and multiple parallel-connected cells. In the embodiments of this application, the battery cluster is preferably a single cluster composed of 416 series-connected and 1 parallel-connected cells, with a rated voltage of 1497.6V and a nominal voltage of 1500V.

[0152] Battery cluster data is preferably acquired through a BAU (Battery Acquisition Unit). The BAU is the front-end execution component of the BMS (Battery Management System), responsible for the high-precision acquisition of physical quantities such as cell voltage, current, and temperature, and the real-time execution of protection commands. It is the core node for data acquisition in high-voltage energy storage systems. According to GB / T 34131-2023 "Technical Requirements for Battery Management Systems in Electrochemical Energy Storage Power Stations," the BMS is the core system used for real-time monitoring of the cell status of the energy storage system, achieving fault protection and energy optimization control.

[0153] The aforementioned cell status includes at least voltage, temperature, SOC (State of Charge, percentage of remaining cell charge), and SOH (State of Health, cell health status); among which, SOC reflects the current energy state of the cell, and SOH reflects the rate of degradation of the cell's current performance compared to its nominal performance.

[0154] CUDA (Compute Unified Device Architecture) is a parallel computing platform and programming model launched by NVIDIA. It achieves parallel acceleration processing of massive amounts of data through the multi-core architecture of GPUs.

[0155] DS evidence theory, or Dempster-Shafer evidence theory, is a multi-source data fusion reasoning method for handling uncertain information. Its core is to quantify the uncertainty of different data sources (evidence) through basic probability assignment (BPA), and then use Dempster's combination rule to fuse the probability assignments of multiple independent pieces of evidence to obtain a unified decision result, thereby achieving accurate judgment of targets in fuzzy and uncertain scenarios.

[0156] See Figures 2-3 For a 1500V battery cluster consisting of 416 connected cells, the CUDA parallel computing architecture preferably adopts a thread partitioning method of 256 threads / block, with a grid size of ceil(416 / 256) = 2 blocks, covering the parallel allocation of 416 data paths of the battery cluster. The sub-computation tasks of each data path are carried by the corresponding thread, ensuring that all computation items can be executed in parallel at the same time, avoiding the delay of serial processing.

[0157] On the one hand, the architecture enables data interaction within blocks through shared memory. Each thread block loads the voltage data of 256 battery clusters into shared memory. Threads within the block exchange data through shared memory and perform reduction operations in parallel. For example, thread 0 compares values ​​with thread 128 and retains the maximum value. Therefore, there is no need to frequently access global memory, reducing global memory access latency.

[0158] On the other hand, the architecture achieves data aggregation between blocks through atomic operations. The first thread of each thread block calls the atomic operation (atomicMax / atomicMin / atomicAdd). For example, the local maximum value of thread block 0 is written to the global variable through atomicMax, and the local maximum value of thread block 1 is then compared and updated through the same instruction to ensure the accuracy of the global maximum value.

[0159] Taking the maximum and minimum voltage of each cell in the battery cluster as an example, the voltage data transmission takes 0.1ms, the block calculation takes 0.5ms, the atomic aggregation takes 0.2ms, and the calculation result feedback takes 0.1ms, with a total time of ≤1ms. Compared with the traditional ARM Cortex-A53 architecture-based CPU serial computing architecture under the same working conditions, the computing speed is improved by ≥58 times.

[0160] In this embodiment, for the massive data of hundreds of cells in a battery cluster, the CUDA parallel computing architecture is used to divide the computing task into local operations within blocks and global summaries between blocks, thereby improving data processing speed, ensuring computational accuracy, and meeting the core requirement of millisecond-level protection for high-voltage energy storage systems.

[0161] It should be noted that the battery cluster parameters specifically include at least one of the following: voltage parameters, current parameters, temperature parameters, and insulation parameters.

[0162] In some embodiments, see Figure 4 The S300 specifically includes:

[0163] S301, Based on the battery cluster parameters, establish multiple corresponding battery cluster sub-criteria;

[0164] S302, based on DS evidence theory, sets the basic probability allocation of each battery cluster sub-criteria;

[0165] S303, integrate the basic probability distributions of each battery cluster sub-criteria to obtain the fusion confidence of the corresponding multiple battery cluster criteria;

[0166] S304, establish a corresponding multi-parameter fusion criterion based on the fusion confidence of multiple battery cluster sub-criteria;

[0167] The S305 diagnoses battery cluster faults and records fault data of faulty battery clusters through multi-parameter fusion criteria.

[0168] In some embodiments, the multiple battery cluster sub-criteria specifically include at least one of multiple voltage sub-criteria, multiple current sub-criteria, and multiple temperature sub-criteria; the multiple voltage sub-criteria are established based on voltage parameters, the multiple current sub-criteria are established based on current parameters, and the multiple temperature sub-criteria are established based on the temperature parameters; the multi-parameter fusion criterion specifically includes at least one of overvoltage protection criterion, overcurrent protection criterion, and temperature rise protection criterion; the overvoltage protection criterion is established based on the fusion confidence level of the multiple voltage sub-criteria, the overcurrent protection criterion is established based on the fusion confidence level of the current sub-criteria, and the temperature rise protection criterion is established based on the fusion confidence level of the multiple temperature sub-criteria.

[0169] In this embodiment of the application, a multi-parameter fusion criterion is constructed through DS evidence theory, and voltage parameters, current parameters and temperature parameters are used as multi-source evidence, and each is assigned a basic probability allocation for fault identification.

[0170] In some embodiments, the multiple voltage sub-criteria specifically include: a first overvoltage sub-criteria: the maximum voltage of a single string of cells in the battery cluster is ≥ a first voltage threshold; a second overvoltage sub-criteria: the total voltage of the battery cluster is ≥ a second voltage threshold; and a third overvoltage sub-criteria: the overvoltage duration of the battery cluster is ≥ a third voltage threshold.

[0171] Among them, the maximum voltage of a single cell in the battery cluster, the total voltage of the battery cluster, and the overvoltage duration of the battery cluster are all voltage parameters; the overvoltage duration of the battery cluster is specifically the duration during which the total voltage of the battery cluster is greater than or equal to the second voltage threshold.

[0172] The formula for calculating the fusion confidence score of multiple voltage sub-criteria is as follows:

[0173] ;

[0174] In the formula: This represents the fusion confidence level of multiple voltage sub-criteria; This represents the fusion value of the basic probability assignments of the first and second overpressure criteria; This represents the basic probability allocation of the third overpressure criterion; This represents the conflict coefficient between the first and second overpressure criteria and the third overpressure criterion after their basic probability allocation is fused; " indicates the XOR operation.

[0175] The overvoltage protection criterion is established as follows: the fusion confidence of multiple voltage sub-criters is greater than or equal to the first confidence threshold; if the overvoltage protection criterion is valid, the fault of the battery cluster is an overvoltage fault.

[0176] The purpose of setting the basic probability allocation of the first, second, and third overvoltage sub-criteria is to quantify the probability of the above three sub-criteria being valid based on the voltage parameters.

[0177] Dempster's evidence theory calculates the fusion confidence of multiple pieces of evidence based on the following Dempster combination rule: When two pieces of evidence are fused, the expression for the fusion confidence is:

[0178] ;

[0179] In the formula: This indicates the fusion confidence level of evidence X and evidence Y. This represents the basic probability distribution of evidence X. This represents the basic probability distribution of evidence Y; The conflict coefficient represents the relationship between evidence X and evidence Y, used to measure the degree of contradiction between the two pieces of evidence; " is the logical NOT operation, and .

[0180] For example, the specific construction process of the above overvoltage protection criterion is as follows.

[0181] Based on the DS evidence theory, the basic probability allocation of the first overvoltage criterion is set according to the maximum voltage of a single string of cells within the battery cluster. Based on the total voltage of the battery cluster, the basic probability allocation of the second overvoltage criterion is set as follows: Based on the overvoltage duration of the battery cluster, the basic probability allocation of the third overvoltage sub-criterion is set as follows: .

[0182] Based on Dempster's combination rules, the first overpressure criterion, the second overpressure criterion, and the third overpressure criterion are gradually integrated below.

[0183] The first step is to fuse the basic probability assignments of the first and second overvoltage criteria to obtain the first fused confidence level:

[0184] ;

[0185] In the formula: This represents the conflict coefficient between the first overvoltage criterion and the second overvoltage criterion.

[0186] The second step is to fuse the first fusion confidence score and the basic probability assignment of the third overpressure criterion to obtain the second fusion confidence score:

[0187] ;

[0188] In the formula: This represents the conflict coefficient between the first and second overpressure criteria and the third overpressure criterion after their basic probability allocation is fused. ;

[0189] After obtaining the second fusion confidence score, the basic probability assignments of the first overpressure criterion, the second overpressure criterion, and the third overpressure criterion are fused. Therefore, the fusion confidence score of the three is:

[0190] ;

[0191] The third step is to... The overvoltage protection criterion is determined by comparing it with a first confidence threshold. For example, when the first confidence threshold is 0.7, because... The overvoltage protection criterion is met, and the battery cluster fault is an overvoltage fault.

[0192] The stepwise fusion process described above for the first, second, and third overpressure sub-criteria can be extended to cases with more than three overpressure sub-criteria: after obtaining the second fusion confidence, the other overpressure sub-criteria are fused stepwise according to the Dempster combination rule until the basic probability assignments of all overpressure sub-criteria are fused, and the fusion confidence of more than three overpressure sub-criteria is obtained.

[0193] In some embodiments, the multiple current sub-criteria specifically include: a first overcurrent sub-criteria: the charging and discharging current of the battery cluster ≥ a first current threshold; a second overcurrent sub-criteria: the rate of increase of the charging and discharging current of the battery cluster ≥ a second current threshold; and a third overcurrent sub-criteria: the overcurrent duration of the battery cluster ≥ a third current threshold.

[0194] Among them, the charging and discharging current of the battery cluster, the rate of increase of the charging and discharging current of the battery cluster, and the overcurrent duration of the battery cluster are all current parameters; the overcurrent duration of the battery cluster is specifically the duration during which the charging and discharging current of the battery cluster is greater than or equal to the first current threshold.

[0195] The formula for calculating the fusion confidence score of multiple current quantum criteria is as follows:

[0196] ;

[0197] In the formula: This represents the fusion confidence level of multiple current quantum criteria; This represents the fusion value of the basic probability assignments of the first and second overcurrent sub-criteria; This represents the basic probability assignment of the third overcurrent criterion; This represents the conflict coefficient between the first and second overcurrent criterion and the third overcurrent criterion after their basic probability allocation is fused; " indicates the XOR operation.

[0198] The overcurrent protection criterion is established as follows: the fusion confidence of multiple current sub-criters is greater than or equal to the second confidence threshold; if the overcurrent protection criterion is valid, the fault of the battery cluster is an overcurrent fault.

[0199] The construction process and application of the overcurrent protection criterion are similar to those of the overvoltage protection criterion, and will not be repeated here.

[0200] In some embodiments, the multiple temperature sub-criteria specifically include: a first temperature rise sub-criteria: the maximum temperature of a single string of cells in the battery cluster is ≥ a first temperature threshold; a second temperature rise sub-criteria: the rate of temperature rise of a single string of cells in the battery cluster is ≥ a second temperature threshold; and a third temperature rise sub-criteria: the temperature difference between three adjacent strings of cells in the battery cluster is ≥ a third temperature threshold.

[0201] Among them, the maximum temperature of a single cell string in a battery cluster, the rate of temperature rise of a single cell string in a battery cluster, and the temperature difference between three adjacent cells strings in a battery cluster are all temperature parameters.

[0202] The formula for calculating the fusion confidence of multiple temperature sub-criterions is as follows:

[0203] ;

[0204] In the formula: This represents the fusion confidence level of multiple temperature sub-criterions; This represents the fusion value of the basic probability assignments of the first thermoparticle criterion and the second thermoparticle criterion; This represents the basic probability allocation of the third thermoparticle criterion; This represents the conflict coefficient between the first and second thermoparticle criteria and the third thermoparticle criterion after their basic probability allocation is fused; " indicates the XOR operation.

[0205] The temperature rise protection criterion is established as follows: the fusion confidence of multiple temperature sub-criters is greater than or equal to the third confidence threshold; if the temperature rise protection criterion is valid, the fault of the battery cluster is a temperature rise fault.

[0206] The construction process and application of the temperature rise protection criterion are similar to those of the overvoltage protection criterion mentioned above, and will not be repeated here.

[0207] In this embodiment, the fault characteristics characterized by each battery cluster sub-criteria are fused by the Dempster combination rule. Compared with the case of considering only the parameters of a single battery cluster, this suppresses false fault judgment and improves the accuracy of fault identification.

[0208] In some embodiments, the method provided in this application further includes: providing primary protection, secondary protection, and tertiary protection, and their respective triggering conditions and response time limits; judging the triggering conditions of primary protection, secondary protection, and tertiary protection based on the fault data of the battery cluster; and executing primary protection, secondary protection, or tertiary protection in descending order of priority based on the judgment result.

[0209] Among them, Level 1 protection has a higher priority than Level 2 protection, and Level 2 protection has a higher priority than Level 3 protection. Specifically, Level 1 protection is for personal safety, Level 2 protection is for equipment protection, and Level 3 protection is for performance warning.

[0210] In some embodiments, the method provided in this application further includes: acquiring the working data of the high-voltage energy storage system; analyzing the working data of the high-voltage energy storage system and the battery cluster parameters based on the CUDA parallel computing architecture to obtain operating condition parameters; establishing a multi-dimensional operating condition evaluation model based on the battery cluster parameters and the operating condition parameters; and adjusting the triggering conditions or response time limits of the first-level protection, second-level protection, or third-level protection based on the multi-dimensional operating condition evaluation model.

[0211] High-voltage energy storage systems can communicate with devices such as BMS, PCS (Power Conversion System), and cooling systems via CAN bus or RS485 bus. The acquisition of their operating data utilizes existing technologies. For example, grid voltage is uploaded via PCS, communication interruption time is determined by the heartbeat packet loss time on the CAN bus or RS485 bus, and cooling system data is collected via sensors.

[0212] In some embodiments, the primary protection specifically includes: shutting off the main contactor and pre-charge contactor of the battery cluster, triggering the fire-fighting linkage device of the high-voltage energy storage system; the triggering conditions for the primary protection specifically include at least one of the following: the maximum voltage of a single string of cells in the battery cluster ≥ a first protection threshold, the minimum voltage of a single string of cells in the battery cluster ≤ a second protection threshold, the charging and discharging current of the battery cluster ≥ a third protection threshold, the temperature rise rate of a single string of cells in the battery cluster ≥ a fourth protection threshold, and the rate of change of the insulation resistance of the positive and negative electrodes of the battery cluster to ground ≥ a fifth protection threshold.

[0213] The secondary protection specifically includes: cutting off the charging circuit of the battery cluster, limiting the discharge power of the battery cluster to ≤20% of the rated power of the battery cluster, and sending a fault alarm signal to the BMS of the high-voltage energy storage system; the triggering conditions of the secondary protection specifically include at least one of the following: the total voltage of the battery cluster ≥ the sixth protection threshold, the total voltage of the battery cluster ≤ the seventh protection threshold, the charge / discharge current ratio of the battery cluster ≥ the eighth protection threshold, and the overcurrent duration of the battery cluster ≥ the ninth protection threshold; the charge / discharge current ratio of the battery cluster is specifically the ratio of the charge / discharge current of the battery cluster to the rated current.

[0214] The Level 3 protection specifically includes: triggering the audible and visual alarm device of the high-voltage energy storage system and limiting the discharge power of the battery cluster to ≤50% of the rated power of the battery cluster; the triggering conditions of Level 3 protection specifically include at least one of the following: the deviation of the battery cluster's equalization current ≥ the tenth protection threshold, the communication interruption time of the high-voltage energy storage system ≥ the eleventh protection threshold, and the cooling system of the high-voltage energy storage system malfunctioning (cooling fan shutdown or insufficient coolant).

[0215] Among them, the maximum voltage of a single cell string within the battery cluster, the minimum voltage of a single cell string within the battery cluster, and the total voltage of the battery cluster are all voltage parameters; the charging and discharging current of the battery cluster, the charging and discharging current rate of the battery cluster, the overcurrent duration of the battery cluster, and the equilibrium current deviation of the battery cluster are all current parameters; the temperature rise rate of a single cell string within the battery cluster is a temperature parameter; the rate of change of the insulation resistance between the positive and negative electrodes of the battery cluster and ground is an insulation parameter; the data on the communication interruption time of the high-voltage energy storage system and the data on the cooling system of the high-voltage energy storage system are all operating data of the high-voltage energy storage system.

[0216] For a 1500V battery cluster consisting of 416 cells connected in series, the first protection threshold is preferably 4.2V, the second protection threshold is preferably 2.5V, the third protection threshold is preferably 2000A, the fourth protection threshold is preferably 5℃ / min, the fifth protection threshold is preferably 500Ω / V and its anti-shake window is preferably 5ms, the sixth protection threshold is preferably 1550V, the seventh protection threshold is preferably 1300V, the eighth protection threshold is preferably 1.2, the ninth protection threshold is preferably 100ms, the tenth protection threshold is preferably 20%, and the tenth protection threshold is preferably 3s.

[0217] The response time of Level 1 protection is shorter than that of Level 2 protection, and the response time of Level 2 protection is shorter than that of Level 3 protection. The preferred response time for Level 1 protection is 10ms, for Level 2 protection it is 50ms, and for Level 3 protection it is 100ms.

[0218] In this embodiment, protection measures are divided based on the severity of faults in the high-voltage energy storage system to form a tiered protection mechanism, which balances system safety and operational continuity through differentiated response actions.

[0219] In some embodiments, the multi-dimensional operating condition evaluation model specifically includes: low temperature operating condition: the minimum temperature of a single string of cells in the battery cluster is ≤ the first operating condition threshold; high temperature operating condition: the maximum temperature of a single string of cells in the battery cluster is ≥ the second operating condition threshold; overload operating condition: the charge / discharge current ratio of the battery cluster is ≥ the third operating condition threshold; low SOC operating condition: the SOC of the battery cluster is ≤ the fourth operating condition threshold; high SOC operating condition: the SOC of the battery cluster is ≥ the fifth operating condition threshold; grid fluctuation operating condition: the grid voltage fluctuation amplitude is ≥ the sixth operating condition threshold.

[0220] Among them, the minimum temperature and maximum temperature of a single cell string within the battery cluster are temperature parameters; the charge / discharge current ratio of the battery cluster is a current parameter; and the state of charge (SOC) of the battery cluster and the grid voltage fluctuation amplitude are operating condition parameters. The specific calculation process for the temperature parameters, current parameters, and grid voltage fluctuation amplitude can be referred to the CUDA parallel computing architecture provided in the aforementioned embodiments, and will not be explained in detail here. The SOC of the battery cluster is preferably calculated based on the ampere-hour integration method and the open-circuit voltage calibration method, which is existing technology.

[0221] For a 1500V battery cluster consisting of 416 cells connected in series, the first operating condition threshold is preferably 0℃, the second operating condition threshold is preferably 45℃, the third operating condition threshold is preferably 1.1℃, the fourth operating condition threshold is preferably 10%, the fifth operating condition threshold is preferably 90%, and the sixth operating condition threshold is preferably 10%.

[0222] In some embodiments, the above-mentioned adjustment of the triggering conditions or response time limits of the first-level protection, second-level protection, or third-level protection based on the multi-dimensional operating condition evaluation model specifically includes: based on low-temperature operating conditions, lowering the first protection threshold and the eighth protection threshold when the low-temperature operating condition is established; based on high-temperature operating conditions, lowering the fourth protection threshold when the high-temperature operating condition is established; based on overload operating conditions, lowering the sixth protection threshold, the ninth protection threshold, and the response time limit of the second-level protection when the overload operating condition is established; based on low SOC operating conditions, raising the second protection threshold when the low SOC operating condition is established; based on high SOC operating conditions, lowering the first protection threshold when the high SOC operating condition is established; and based on grid fluctuation operating conditions, extending the anti-jitter window of the fifth protection threshold when the grid fluctuation operating condition is established.

[0223] For a 1500V battery cluster consisting of 416 series cells, when the low-temperature condition is established, it is preferable to lower the first protection threshold from 4.2V to 4.1V and the eighth protection threshold from 1.2V to 1.1V; when the high-temperature condition is established, it is preferable to lower the fourth protection threshold from 5℃ / min to 3℃ / min; when the overload condition is established, it is preferable to lower the sixth protection threshold from 1550V to 1530V, the ninth protection threshold from 100ms to 50ms, and the response time of the secondary protection from 50ms to 20ms; when the low SOC condition is established, it is preferable to raise the second protection threshold from 2.5V to 2.7V; when the high SOC condition is established, it is preferable to lower the first protection threshold from 4.2V to 4.15V; when the grid fluctuation condition is established, it is preferable to extend the anti-jitter window of the fifth protection threshold from 5ms to 10ms.

[0224] The low SOC and high SOC operating conditions are adapted to the undervoltage and overvoltage conditions of the battery cluster respectively, which is beneficial to maintaining the cycle life of the battery cells; the insulation data is processed by moving average filtering (window length preferably 5), which helps to avoid false triggering caused by power grid interference.

[0225] In practical applications, when the low-temperature operating condition is met and the SOC of the battery cluster is ≥80%, the charging current of the battery cluster can be further limited to 0.5 times the rated current to adapt to the characteristics of increased cell internal resistance and increased risk of lithium deposition at low temperatures; when the high-temperature operating condition is met and the SOC of the battery cluster is ≤20%, the discharge current of the battery cluster can be limited to not exceed the rated current to suppress the risk of thermal runaway.

[0226] In this embodiment, an adaptive threshold adjustment mechanism is formed through a multi-dimensional operating condition evaluation model to dynamically adapt to complex operating conditions such as low temperature, high temperature, overload, and power grid fluctuations, thereby reducing the false triggering rate and missed triggering rate of protection measures at all levels and solving the problem of poor adaptability of traditional static thresholds.

[0227] In some embodiments, the method provided in the application further includes: if the triggering condition of primary protection or secondary protection is met according to the judgment result, then the connection between the faulty battery cluster and the power grid is cut off through the PCS, and the charging and discharging permissions of the faulty battery cluster and its adjacent battery clusters are locked.

[0228] According to GB / T 36572-2018 "Technical Specification for Energy Storage Converters in Electrochemical Energy Storage Systems", the power conversion system (PCS) is a key device used to realize AC / DC energy conversion and power regulation between the energy storage system and the power grid. For example, the high-voltage energy storage system sends coordinated control commands to the PCS via the CAN bus and controls the PCS to perform disconnection actions; when the battery cluster triggers the first-level protection (personal safety level), the charging and discharging permissions are locked for 10 minutes; when the battery cluster triggers the second-level protection (equipment protection level), the charging and discharging permissions are locked for 5 minutes.

[0229] In this embodiment, fault cascading suppression measures are established for primary and secondary protection, so that when a single battery cluster triggers primary or secondary protection, it not only cuts off its own circuit, but also locks the charging and discharging permissions of its adjacent battery clusters, thereby suppressing fault escalation and reducing the risk of the entire station shutting down due to fault propagation.

[0230] Furthermore, since Level 3 protection (performance warning level) has a low priority and relatively lenient triggering conditions, when only Level 3 protection is triggered, the corresponding fault is unlikely to escalate to the fault conditions required to trigger Level 1 or Level 2 protection. The above embodiments do not perform fault cascading suppression for Level 3 protection, avoiding unnecessary disconnection actions and contributing to maintaining the operational continuity of the battery cluster.

[0231] In some embodiments, S400 specifically includes: repairing the fault of the faulty battery cluster based on the fault data of the faulty battery cluster, and restarting the faulty battery cluster in three stages: pre-charge stage, low-power discharge stage, and rated power recovery stage.

[0232] The pre-charge phase specifically includes: closing the pre-charge circuit of the faulty battery cluster within 5 seconds, limiting the pre-charge current of the faulty battery cluster to less than or equal to the first restart threshold, until the total voltage change rate of the faulty battery cluster is less than or equal to the second restart threshold.

[0233] The low-power discharge phase specifically includes: limiting the discharge current of the faulty battery cluster to the third restart threshold within 10 seconds, until the difference between the maximum and minimum voltages of a single string of cells in the faulty battery cluster is ≤ the fourth restart threshold and the temperature change rate of a single string of cells in the faulty battery cluster is ≤ the fifth restart threshold.

[0234] The rated power recovery phase specifically includes: gradually increasing the charging and discharging power of the faulty battery cluster within 30 seconds; if the faulty battery cluster no longer malfunctions, then the normal operation of the faulty battery cluster is restored.

[0235] For a 1500V battery cluster consisting of 416 cells connected in series, the first restart threshold is preferably 5A, the second restart threshold is preferably 5V / s, the third restart threshold is preferably 0.2C, the fourth restart threshold is preferably 0.3V, and the fifth restart threshold is preferably 0.11℃ / 10s; during the rated power recovery phase, the rated power is increased by 20% every 5s within 30s.

[0236] In practical applications, maintenance personnel read the fault data of the faulty battery cluster through the host computer, troubleshoot and repair the fault; if the battery cluster is subject to fault cascading suppression, which locks the faulty battery cluster and its adjacent battery clusters, the lock is first released by the reset button on the host computer before the fault is repaired.

[0237] In this embodiment, a tiered restart process, from pre-charging to low-power discharging and then to rated power recovery, is used to verify the fault repair results of the battery cluster in a stepwise manner. This avoids secondary faults caused by the faulty battery cluster running at full load directly after repair, thereby improving the reliability of fault repair and restart.

[0238] In some embodiments, S100 specifically includes: synchronously acquiring voltage data of each string of cells in the battery cluster through multiple first main channels and multiple first backup channels; synchronously acquiring charging and discharging current data of the battery cluster through one second main channel and one second backup channel; synchronously acquiring temperature data of each string of cells in the battery cluster through multiple third main channels and multiple third backup channels; and synchronously acquiring insulation resistance data of the positive and negative electrodes of the battery cluster to ground through two fourth channels.

[0239] Among them, the first main channel and the first backup channel of the Nth channel synchronously collect the voltage data of the Nth string of cells in the battery cluster; the second main channel and the second backup channel are connected in parallel to the charging and discharging current loop of the battery cluster; the third main channel and the third backup channel of the Nth channel synchronously collect the temperature data of the Nth string of cells in the battery cluster; one fourth channel constructs the positive electrode to ground balance bridge circuit of the battery cluster, and another fourth channel constructs the negative electrode to ground balance bridge circuit of the battery cluster.

[0240] For a 1500V battery cluster consisting of 416 series cells, the preferred first main channel and the preferred first backup channel are 26 ADS1256 chips, with each ADS1256 chip corresponding to 16 series cells, and the 26 ADS1256 chips covering 416 series cells, with a sampling rate of 1kHz; the preferred second main channel is ACS758, and the preferred second backup channel is LA55-P, with ACS758 and LA55-P connected in parallel to the DC bus of the battery cluster, with a range of ±1000A; the preferred third main channel is DS18B20, and the preferred third backup channel is PT100, with one DS18B20 and one PT100 installed for each series of cells; the preferred fourth channel is ADUM1400, with two ADUM1400 chips respectively constructing a positive-to-ground and negative-to-ground balanced bridge circuit for the battery cluster.

[0241] In practical applications, the CUDA parallel computing architecture can be used to verify the data acquisition deviation between the first main channel and the first backup channel, the second main channel and the second backup channel, the third main channel and the third backup channel, or the two fourth channels in real time, and to determine the channel failure status; when one channel fails, the other channel can still continue to perform the data acquisition task.

[0242] It should be understood that the aforementioned ADS1256, ACS758, LA55-P, DS18B20, PT100 and ADUM1400 are all device or chip models, and their structures and circuit connections are all known technologies.

[0243] In this embodiment, battery cluster data is synchronously collected and redundantly backed up using a dual-channel approach, thereby improving the efficiency and reliability of battery cluster data collection.

[0244] In some embodiments, S100 further includes: preprocessing the battery cluster data using a combination algorithm of sliding window filtering and Kalman filtering.

[0245] The sliding window filter preferably has a window length of 3-5, and the Kalman filter preferably uses the 3σ rule to remove outliers from the battery cluster data, thereby eliminating pulse interference and random noise in the battery cluster data acquired through the dual-channel acquisition method.

[0246] See Figure 5 This application provides a high-voltage energy storage system protection and control device, including a data acquisition unit 501, a data processing unit 502, a fault diagnosis unit 503, and a fault repair unit 504.

[0247] The data acquisition unit 501 is used to acquire battery cluster data of the high-voltage energy storage system; the data processing unit 502 is used to process the battery cluster data based on the CUDA parallel computing architecture to obtain battery cluster parameters; the fault diagnosis unit 503 is used to diagnose the faults of the battery clusters based on the battery cluster parameters and the DS evidence theory, and record the fault data of the faulty battery clusters; the fault repair unit 504 is used to repair the faults of the faulty battery clusters based on the fault data of the faulty battery clusters, and restart the faulty battery clusters in multiple stages.

[0248] It should be noted that the device provided in the above embodiments is only illustrated by the division of the above functional modules when executing the high-voltage energy storage system protection and control method. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the equipment can be divided into different functional modules to complete all or part of the functions described above. In addition, the device provided in the above embodiments and the high-voltage energy storage system protection and control method embodiments belong to the same concept, and its implementation process can be found in the method embodiments, which will not be repeated here.

[0249] See Figure 6 This application provides a hardware architecture for a high-voltage energy storage system, including a display layer, a core layer, a redundant acquisition layer, a control layer, a power supply layer, and a support service layer.

[0250] In the core layer, Jetson Nano serves as the main control unit, used to acquire battery cluster data from the high-voltage energy storage system; it processes the battery cluster data based on the CUDA parallel computing architecture to obtain battery cluster parameters; based on the battery cluster parameters, it diagnoses battery cluster faults based on DS evidence theory and records fault data of faulty battery clusters; the Jetson Nano's interface can be implemented based on SPI, CAN, ADC, UART, Ethernet, etc.

[0251] The redundant acquisition layer consists of a voltage acquisition module, a current acquisition module, a temperature acquisition module, and an insulation monitoring module. Specifically, the voltage acquisition module connects to the Jetson Nano's SPI0 / SPI1 port via an SPI interface to acquire the battery cluster's voltage data; the current acquisition module connects to the Jetson Nano's ADC0 / ADC1 port via an ADC interface to acquire the battery cluster's current data; the temperature acquisition module connects to the Jetson Nano's GPIO2 / GPIO3 port via a single-bus protocol to acquire the battery cluster's temperature data; and the insulation monitoring module connects to the Jetson Nano's SPI2 / SPI3 port via an SPI interface to acquire the battery cluster's insulation data.

[0252] In the control layer, the input of the optocoupler isolation relay driver board is connected to the GPIO4-GPIO7 ports of the Jeston Nano, and the output is connected to the main contactor of the battery cluster, the precharge contactor, the fire-fighting linkage device of the high-voltage energy storage system, and the audible and visual alarm device of the high-voltage energy storage system.

[0253] The support service layer, acting as a communication module, establishes communication connections between Jetson Nano and the display and control layers via CAN bus, RS485 bus, or RJ45 Ethernet, and enables bidirectional communication between modules through shielded cables.

[0254] The power supply layer, as a power supply module, is preferably a DC-DC wide-voltage DC power supply. Its input end is connected to the auxiliary power supply of the energy storage system, and its output end is connected to the power interface of each of the above layers, so as to provide stable power supply to each of the above layers and suppress electromagnetic interference.

[0255] To optimize the reliability of the methods provided in the foregoing embodiments, this application also provides an optimization method, including: configuring a software architecture; the software architecture is based on the Ubuntu 20.04 LTS operating system and loads the PREEMPT_RT real-time kernel (preferably version 5.4.191-rt100), including thread management, memory optimization, process protection, and fault logging.

[0256] In thread management, threads under Level 1 protection are bound to CPU core 0 (exclusive core), threads under Level 2 protection are bound to CPU core 1, threads under Level 3 protection are bound to CPU core 2, and computational tasks of the CUDA parallel computing architecture are bound to CPU core 3 and GPU (128-core Maxwell architecture).

[0257] In memory optimization, the mlockall(MCL_CURRENT\MCL_FUTURE) function is used to lock all code and data pages, disabling Linux memory swapping and avoiding response delays caused by page faults.

[0258] In process protection, a protection unit for the logical process is created through systemd, and the process is configured to automatically restart after a crash. After restarting, the most recently saved battery cluster parameters and operating condition parameters are automatically loaded.

[0259] The fault log uses a storage method that combines local SD card storage and cloud backup to record battery cluster data, battery cluster parameters, the status of the first-level protection thread, the status of the second-level protection thread, the status of the third-level protection thread, and operating parameters for 10 seconds before and after the fault. The log retains the most recent 100 entries and supports cyclic overwriting.

[0260] See Figure 7 This application provides an electronic device 700, including a processor 701 and a memory 702.

[0261] In this embodiment, the processor 701 is the control center of the computer system, and can be a processor of a physical machine or a processor of a virtual machine. The processor 701 may include one or more processing cores, such as a 4-core processor or an 8-core processor. The processor 701 can be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array).

[0262] Processor 701 may also include a main processor and a coprocessor. The main processor is a processor used to process data in the wake-up state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor used to process data in the standby state.

[0263] Memory 702 may include one or more computer-readable storage media. Memory 702 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments of this application, the computer-readable storage media in memory 702 are used to store at least one instruction, which is executed by processor 701 to implement the methods in the embodiments of this application.

[0264] In some embodiments, the electronic device 700 further includes a peripheral device interface 703 and at least one peripheral device 704. The processor 701, memory 702, and peripheral device interface 703 can be connected via a bus or signal line. Each peripheral device 704 can be connected to the peripheral device interface 703 via a bus, signal line, or circuit board. Specifically, the peripheral device 704 includes: a display screen, a camera, and audio circuitry. The peripheral device interface 703 can be used to connect at least one I / O (Input / Output) related peripheral device to the processor 701 and memory 702.

[0265] In some embodiments of this application, the processor 701, memory 702, and peripheral device interface 703 are integrated on the same chip or circuit board; in other embodiments of this application, any one or two of the processor 701, memory 702, and peripheral device interface 703 can be implemented on separate chips or circuit boards. This application does not specifically limit the implementation in this regard.

[0266] The schematic diagram of the electronic device shown in the embodiments of this application does not constitute a limitation on the electronic device 700. The electronic device 700 may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0267] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the methods in any of the foregoing embodiments. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0268] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A protection and control method for a high-voltage energy storage system, characterized in that, The method includes: Acquire battery cluster data from a high-voltage energy storage system; The battery cluster data is processed using the CUDA parallel computing architecture to obtain battery cluster parameters. Based on the battery cluster parameters, the faults of the battery clusters are diagnosed using DS evidence theory, and the fault data of the faulty battery clusters are recorded. Based on the fault data of the faulty battery cluster, the faulty battery cluster is repaired and restarted in multiple stages.

2. The high-voltage energy storage system protection and control method according to claim 1, characterized in that, The battery cluster parameters include at least one of the following: voltage parameters, current parameters, temperature parameters, and insulation parameters.

3. The high-voltage energy storage system protection and control method according to claim 2, characterized in that, The step of diagnosing battery cluster faults based on the battery cluster parameters and using DS evidence theory, and recording fault data of the faulty battery clusters, includes: Based on the battery cluster parameters, establish multiple corresponding battery cluster sub-criteria; Based on the DS evidence theory, a basic probability allocation is set for each of the corresponding battery cluster sub-criteria. By fusing the basic probability assignments of each of the battery cluster sub-criteria, the fusion confidence of the corresponding multiple battery cluster criteria is obtained; Based on the fusion confidence of the multiple battery cluster sub-criteria, a corresponding multi-parameter fusion criterion is established; The faults of battery clusters are diagnosed and the fault data of faulty battery clusters are recorded by using the multi-parameter fusion criteria.

4. The high-voltage energy storage system protection and control method according to claim 3, characterized in that, The plurality of battery cluster sub-criteria includes at least one of a plurality of voltage sub-criteria, a plurality of current sub-criteria, and a plurality of temperature sub-criteria; The plurality of voltage sub-criteria are established based on the voltage parameter, the plurality of current sub-criteria are established based on the current parameter, and the plurality of temperature sub-criteria are established based on the temperature parameter; The multi-parameter fusion criteria include at least one of the following: overvoltage protection criteria, overcurrent protection criteria, and temperature rise protection criteria; The overvoltage protection criterion is established based on the fusion confidence level of the multiple voltage sub-criterions, the overcurrent protection criterion is established based on the fusion confidence level of the current sub-criterion, and the temperature rise protection criterion is established based on the fusion confidence level of the multiple temperature sub-criterions.

5. The high-voltage energy storage system protection and control method according to claim 4, characterized in that, The plurality of voltage sub-criteria include: First overvoltage criterion: The maximum voltage of a single cell in a battery cluster is greater than or equal to the first voltage threshold. Second overvoltage criterion: Total voltage of battery cluster ≥ second voltage threshold; Third overvoltage criterion: Overvoltage duration of battery cluster ≥ third voltage threshold; Wherein, the maximum voltage of a single cell string in the battery cluster, the total voltage of the battery cluster, and the overvoltage duration of the battery cluster are all voltage parameters; The formula for calculating the fusion confidence of the multiple voltage sub-criteria is as follows: ； In the formula: This represents the fusion confidence level of the multiple voltage sub-criteria; This represents the fusion value of the basic probability allocation of the first overvoltage subcriterion and the second overvoltage subcriterion; This represents the basic probability allocation of the third overpressure criterion; This represents the conflict coefficient between the first and second overvoltage sub-criteria after their basic probability allocation fusion and the third overvoltage sub-criteria; " indicates the XOR operation; The overvoltage protection criterion is established as follows: the fusion confidence level of the multiple voltage sub-criterions is greater than or equal to the first confidence threshold; if the overvoltage protection criterion is valid, then the fault of the battery cluster is an overvoltage fault.

6. The high-voltage energy storage system protection and control method according to claim 4, characterized in that, The plurality of current-electron criteria include: First overcurrent criterion: The charging and discharging current of the battery cluster is greater than or equal to the first current threshold. Second current criterion: The rate of increase of the charging and discharging current of the battery cluster is greater than or equal to the second current threshold. Third overcurrent criterion: Overcurrent duration of battery cluster ≥ third current threshold; Wherein, the charging and discharging current of the battery cluster, the rate of increase of the charging and discharging current of the battery cluster, and the overcurrent duration of the battery cluster are all current parameters. The formula for calculating the fusion confidence of the multiple current sub-criteria is as follows: ； In the formula: This represents the fusion confidence level of the multiple current sub-criteria; This represents the fusion value of the basic probability allocation of the first overcurrent sub-criteria and the second overcurrent sub-criteria; This represents the basic probability allocation of the third overcurrent sub-criterion; This represents the conflict coefficient between the first and second overcurrent sub-criteria after their basic probability allocation fusion and the third overcurrent sub-criteria; " indicates the XOR operation; The overcurrent protection criterion is established as follows: the fusion confidence level of the multiple current sub-criterions is greater than or equal to the second confidence threshold; if the overcurrent protection criterion is valid, then the fault of the battery cluster is an overcurrent fault.

7. The high-voltage energy storage system protection and control method according to claim 4, characterized in that, The plurality of temperature sub-criteria include: First temperature rise criterion: The maximum temperature of a single cell string in the battery cluster is ≥ the first temperature threshold. Second temperature rise criterion: The rate of temperature rise of a single cell string within a battery cluster is greater than or equal to the second temperature threshold. The third temperature rise criterion is: the temperature difference between three adjacent strings of cells within a battery cluster is greater than or equal to the third temperature threshold. The temperature parameters are: the maximum temperature of a single cell string within the battery cluster, the temperature rise rate of a single cell string within the battery cluster, and the temperature difference between three adjacent cells strings within the battery cluster. The formula for calculating the fusion confidence of the multiple temperature sub-criterions is as follows: ； In the formula: This represents the fusion confidence level of the multiple temperature sub-criterions; This represents the fusion value of the basic probability allocation of the first temperature riser criterion and the second temperature riser criterion; This represents the basic probability allocation of the third temperature riser criterion; This represents the conflict coefficient between the first and second temperature riser criteria and the third temperature riser criterion after their basic probability allocation is fused; " indicates the XOR operation; The temperature rise protection criterion is established as follows: the fusion confidence level of the multiple temperature sub-criterions is greater than or equal to the third confidence threshold; if the temperature rise protection criterion is valid, then the fault of the battery cluster is a temperature rise fault.

8. The high-voltage energy storage system protection and control method according to claim 2, characterized in that, The method further includes: It provides Level 1 protection, Level 2 protection, and Level 3 protection, along with their respective triggering conditions and response time limits; the Level 1 protection has a higher priority than the Level 2 protection, and the Level 2 protection has a higher priority than the Level 3 protection. Based on the fault data of the battery cluster, the triggering conditions of the first-level protection, second-level protection and third-level protection are determined; Based on the judgment result, the first-level protection, the second-level protection, or the third-level protection are executed in descending order of priority.

9. The high-voltage energy storage system protection and control method according to claim 8, characterized in that, The method further includes: Acquire operational data from the high-voltage energy storage system; Based on the operating data of the high-voltage energy storage system and the parameters of the battery cluster, the operating parameters are obtained by analysis using the CUDA parallel computing architecture. Based on the battery cluster parameters and the operating condition parameters, a multi-dimensional operating condition evaluation model is established. Based on the multi-dimensional operating condition evaluation model, the triggering conditions or response time limits of the first-level protection, the second-level protection, or the third-level protection are adjusted.

10. The high-voltage energy storage system protection and control method according to claim 9, characterized in that, The first-level protection includes: shutting off the main contactor and pre-charge contactor of the battery cluster and triggering the fire-fighting linkage device of the high-voltage energy storage system; The triggering conditions for the first-level protection include at least one of the following: the maximum voltage of a single cell string in the battery cluster is ≥ the first protection threshold, the minimum voltage of a single cell string in the battery cluster is ≤ the second protection threshold, the charging and discharging current of the battery cluster is ≥ the third protection threshold, the temperature rise rate of a single cell string in the battery cluster is ≥ the fourth protection threshold, and the rate of change of the insulation resistance of the positive and negative electrodes of the battery cluster to ground is ≥ the fifth protection threshold. The secondary protection includes: cutting off the charging circuit of the battery cluster, limiting the discharge power of the battery cluster to ≤20% of the rated power of the battery cluster, and sending a fault alarm signal to the BMS of the high-voltage energy storage system. The triggering conditions for the secondary protection include at least one of the following: the total voltage of the battery cluster ≥ the sixth protection threshold, the total voltage of the battery cluster ≤ the seventh protection threshold, the charge / discharge current rate of the battery cluster ≥ the eighth protection threshold, and the overcurrent duration of the battery cluster ≥ the ninth protection threshold. The three-level protection includes: triggering the audible and visual alarm device of the high-voltage energy storage system and limiting the discharge power of the battery cluster to ≤50% of the rated power of the battery cluster; The triggering conditions for the three-level protection include at least one of the following: the deviation of the equalization current of the battery cluster is greater than or equal to the tenth protection threshold, the communication interruption time of the high-voltage energy storage system is greater than or equal to the eleventh protection threshold, and the cooling system of the high-voltage energy storage system fails. The response time limit of the first-level protection is shorter than that of the second-level protection, and the response time limit of the second-level protection is shorter than that of the third-level protection; in: The maximum voltage of a single cell string within the battery cluster, the minimum voltage of a single cell string within the battery cluster, and the total voltage of the battery cluster are all the voltage parameters. The charging and discharging current of the battery cluster, the charging and discharging current ratio of the battery cluster, the overcurrent duration of the battery cluster, and the equilibrium current deviation of the battery cluster are all current parameters. The temperature rise rate of a single string of cells within the battery cluster is the temperature parameter. The rate of change of the insulation resistance of the positive and negative electrodes of the battery cluster to ground is the insulation parameter; The data on the communication interruption time of the high-voltage energy storage system and the data on the cooling system of the high-voltage energy storage system are both operational data of the high-voltage energy storage system.

11. The high-voltage energy storage system protection and control method according to claim 10, characterized in that, The multi-dimensional working condition assessment model includes: Low temperature condition: The minimum temperature of a single cell string within the battery cluster is ≤ the threshold of the first operating condition; High-temperature operating conditions: The maximum temperature of a single cell string within the battery cluster is ≥ the threshold of the second operating condition; Overload condition: The charge / discharge current rate of the battery cluster is ≥ the threshold of the third condition; Low SOC condition: SOC of battery cluster ≤ fourth condition threshold; High SOC condition: SOC of the battery cluster ≥ threshold of condition 5; Power grid fluctuation condition: Power grid voltage fluctuation amplitude ≥ threshold of condition 6; in: The minimum temperature of a single cell string within the battery cluster and the maximum temperature of a single cell string within the battery cluster are the temperature parameters. The SOC of the battery cluster and the voltage fluctuation of the grid are both operating parameters.

12. The high-voltage energy storage system protection and control method according to claim 11, characterized in that, The adjustment of the triggering conditions or response time limits of the primary protection, the secondary protection, or the tertiary protection based on the multi-dimensional operating condition evaluation model includes: Based on the aforementioned low-temperature operating condition, when the low-temperature operating condition is established, the first protection threshold and the eighth protection threshold are lowered. Based on the aforementioned high-temperature operating condition, when the high-temperature operating condition is established, the fourth protection threshold is lowered. Based on the overload condition, when the overload condition is established, the response time limit of the sixth protection threshold, the ninth protection threshold, and the secondary protection is reduced. Based on the low SOC condition, when the low SOC condition is established, the second protection threshold is increased. Based on the high SOC condition, when the high SOC condition is established, the first protection threshold is lowered. Based on the power grid fluctuation condition, when the power grid fluctuation condition is established, the anti-jitter window of the fifth protection threshold is extended.

13. The high-voltage energy storage system protection and control method according to claim 10, characterized in that, The method further includes: Based on the judgment result, if the triggering conditions of the first-level protection or the second-level protection are met, the connection between the faulty battery cluster and the power grid will be cut off by the PCS, and the charging and discharging permissions of the faulty battery cluster and its adjacent battery clusters will be locked.

14. The high-voltage energy storage system protection and control method according to claim 1, characterized in that, The process of repairing the faulty battery cluster based on its fault data, and restarting the faulty battery cluster in multiple stages, includes: Based on the fault data of the faulty battery cluster, the faulty battery cluster is repaired by restarting the faulty battery cluster in three stages: pre-charge stage, low-power discharge stage, and rated power recovery stage. in: The pre-charge phase includes: closing the pre-charge circuit of the faulty battery cluster within 5 seconds, limiting the pre-charge current of the faulty battery cluster to ≤ the first restart threshold, until the total voltage change rate of the faulty battery cluster is ≤ the second restart threshold. The low-power discharge stage includes: limiting the discharge current of the faulty battery cluster to the third restart threshold within 10 seconds, until the difference between the maximum and minimum voltages of the single-string cells in the faulty battery cluster is ≤ the fourth restart threshold and the temperature change rate of the single-string cells in the faulty battery cluster is ≤ the fifth restart threshold. The rated power recovery phase includes: gradually increasing the charging and discharging power of the faulty battery cluster within 30 seconds; if the faulty battery cluster no longer malfunctions, then the normal operation of the faulty battery cluster is restored.

15. The protection and control method for a high-voltage energy storage system according to claim 1, characterized in that, The acquisition of battery cluster data of the high-voltage energy storage system includes: Voltage data of each cell in the battery cluster is collected synchronously through multiple primary channels and multiple backup channels. The charging and discharging current data of the battery cluster are collected synchronously through one second main channel and one second backup channel; Temperature data of each cell in the battery cluster is collected synchronously through multiple third main channels and multiple third backup channels; The insulation resistance data of the positive and negative terminals of the battery cluster to ground are collected simultaneously through two fourth channels. in: The first main channel of the Nth channel and the first backup channel of the Nth channel simultaneously collect the voltage data of the Nth string of cells in the battery cluster; The second main channel and the second backup channel are connected in parallel to the charging and discharging current circuit of the battery cluster; The third main channel and the third backup channel of the Nth path synchronously collect temperature data of the Nth string of cells in the battery cluster; One of the fourth channels constructs a positive-to-ground balance bridge circuit for the battery cluster, and the other of the fourth channels constructs a negative-to-ground balance bridge circuit for the battery cluster.

16. The protection and control method for a high-voltage energy storage system according to claim 1, characterized in that, The acquisition of battery cluster data of the high-voltage energy storage system also includes: The battery cluster data is preprocessed using a combination algorithm of sliding window filtering and Kalman filtering.

17. A protection and control device for a high-voltage energy storage system, performing the method as described in any one of claims 1-16, characterized in that, The device includes: The data acquisition unit is used to acquire battery cluster data of the high-voltage energy storage system; The data processing unit is used to process the battery cluster data based on the CUDA parallel computing architecture to obtain battery cluster parameters; The fault diagnosis unit is used to diagnose the faults of the battery clusters based on the battery cluster parameters and the DS evidence theory, and to record the fault data of the faulty battery clusters. The fault repair unit is used to repair the fault of the faulty battery cluster based on the fault data of the faulty battery cluster, and restart the faulty battery cluster in multiple stages.

18. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-16.

19. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-16.

20. An optimization method, based on the method as described in any one of claims 9-13, characterized in that, include: Configure software architecture; The software architecture is based on the Ubuntu 20.04 LTS operating system and loads the PREEMPT_RT real-time kernel, including: Thread management: The first-level protected thread is bound to CPU core 0, the second-level protected thread is bound to CPU core 1, the third-level protected thread is bound to CPU core 2, and the computing tasks of the CUDA parallel computing architecture are bound to CPU core 3 and GPU. Memory optimization: Lock all code and data pages using the mlockall(MCL_CURRENT\MCL_FUTURE) function to prevent Linux memory paging; Process Guardian: Create a guardian unit to protect the logical process through systemd, configure automatic restart when the process crashes, and automatically load the most recently saved battery cluster parameters and operating condition parameters after restarting; Fault Log: The storage method combines local SD card storage and cloud backup to record the battery cluster data, battery cluster parameters, status of the first-level protection thread, status of the second-level protection thread, status of the third-level protection thread, and operating parameters within 10 seconds before and after the fault. The log retains the most recent 100 entries and supports cyclic overwriting.

Images