Discharge control system of high-power energy storage emergency power supply

Abstract

The invention relates to the technical field of energy storage power supply discharge control, and discloses a high-power energy storage emergency power supply discharge control system, which comprises a monitoring unit, a total current monitoring unit and a central discharge controller, and is characterized in that the central discharge controller is used for maintaining the connection of a battery cluster and instructing a load to reduce the power when the voltage of a battery cell reaches an early warning threshold value; the controller also monitors the total current to verify the load response, if the response fails, the switching strategy turns off the battery cluster, the conflict between the cut-off action of the battery management system and the stability of the system bus is solved, the failure mode is converted from instantaneous collapse to smooth dimensionality reduction, and the effectiveness when the control loop fails is ensured.

Description

Technical Field

[0001] This invention relates to a high-power energy storage emergency power supply discharge control system, belonging to the field of energy storage power supply discharge control technology. Background Technology

[0002] Currently, lithium-ion batteries are widely used in high-power energy storage systems. A battery management system (BMS) is standard equipment for these systems to ensure their operation. Its main function is to monitor the status of the cells within the system and initiate protective actions when a cell reaches a protection threshold. The inconsistency between individual lithium-ion cells (especially under high-current conditions) due to differences in internal resistance, capacity, and polarization characteristics is an objective physical characteristic. This inconsistency means that in a system composed of parallel battery clusters, the weakest cell will always reach the discharge cutoff voltage threshold preset by the BMS before the other cells. However, in high-power emergency discharge... Under a specific operating condition, the standard protection mechanism of the existing BMS can trigger a systemic failure propagation chain in its normal protection action: when the weakest cell in a parallel cluster reaches the cutoff threshold, the standard action of the BMS is to disconnect the cluster to protect the cell. This action instantly distributes the current carried by the cluster to the remaining N-1 parallel clusters in the system. For a system operating at high power, this instantaneous current surge will inevitably cause an instantaneous voltage drop on the DC bus. This voltage drop can cause sensitive loads to fail instantly. Due to the sudden increase in current, the weakest cells of the remaining N-1 clusters will also reach the cutoff threshold more quickly, causing the BMS to disconnect one after another, ultimately leading to a system-level cascading failure.

[0003] To address this issue, conventional improvement approaches, such as attempting to screen for cells with higher consistency or increasing system redundancy, are limited by electrochemical principles and cost, while the latter cannot avoid the physical process of bus voltage surges caused by the disconnection of the weakest cluster. This exposes a fundamental inherent contradiction: within the existing technological framework, the disconnection action performed to protect individual cells becomes the direct cause of system-level instability. This control logic, which sacrifices system stability to protect individual cells, is also reflected in existing emergency power supply patents. Its focus is on power switching and energy storage, rather than system robustness under high-power discharge conditions. For example, Chinese invention patent CN104269904B discloses a high-power mobile off-peak energy storage emergency power supply house. Although this solution provides a complete emergency power supply system, including vanadium battery stacks and charge / discharge control, its emergency logic is limited to detection and switching when the main system power suddenly stops supplying power.

[0004] Therefore, the technical problem to be solved by this invention is how to design a novel discharge control system that can solve the above-mentioned failure conduction chain problem without relying on absolute cell consistency or excessive redundancy, and can both utilize the cell's safety boundary information and avoid the instantaneous impact of the cut-off action on the system bus voltage. Summary of the Invention

[0005] This invention provides a high-power energy storage emergency power supply discharge control system, the main purpose of which is to solve the problem that there is a conflict between the cell protection action of the existing battery management system and the system-level power stability target, which easily leads to bus voltage drop and cascading failure.

[0006] To achieve the above objectives, this invention provides a high-power energy storage emergency power supply discharge control system, applicable to an emergency power supply comprising N parallel battery clusters and a controllable power load. The system includes a monitoring unit, a total current monitoring unit, and a central discharge controller. The monitoring unit is used to monitor the real-time voltage and real-time temperature of at least one cell in each of N parallel battery clusters. The total current monitoring unit is used to monitor the total output current of the emergency power supply system. The central discharge controller is connected to the monitoring unit, the total current monitoring unit, and the controllable power load via a communication interface; the central discharge controller is used for: Step a: In response to the emergency discharge, the controllable power load is instructed to operate at the initial power rating. Step b: Continuously receive real-time voltage and real-time temperature when the controllable power load is running; Step c: When at least one real-time voltage is detected to reach the preset warning reduction threshold for the first time, and the real-time temperature is lower than the preset temperature safety rejection threshold; Step d, proceed as follows: Step i, maintain the discharge connection of all N parallel battery clusters, and replace the disconnection of battery clusters that have reached the warning reduction threshold; Step ii, send a power reduction command to the controllable power load, so that the operating power of the controllable power load is reduced to a reduced power rating. Step e, after sending the power reduction command, is also used for: step e1, within the preset response confirmation window period, monitoring whether the total output current of the system has decreased as expected corresponding to the power reduction command; step e2, if the total output current of the system has not decreased as expected, switching the control strategy and instead instructing to disconnect the battery cluster that has reached the warning reduction threshold.

[0007] Preferably, after the controllable power load operates at the reduced power rating, the central discharge controller is also used to repeatedly send a power reduction command to the controllable power load if the real-time voltage is detected to reach the warning reduction threshold again, so that the operating power of the controllable power load is further reduced to a lower reduced power rating.

[0008] Preferably, before performing step d, the central discharge controller is also used to obtain the real-time temperature corresponding to the cell that has reached the warning reduction threshold, and to obtain the average temperature of N parallel battery clusters as the temperature reference; to determine whether the real-time temperature is lower than the reasonable temperature rise threshold determined by the sum of the temperature reference and the preset temperature rise margin; to perform step d only when the real-time temperature is determined to be not lower than the reasonable temperature rise threshold; and to suppress the performance of step d when the real-time temperature is determined to be lower than the reasonable temperature rise threshold.

[0009] Preferably, the central discharge controller is also used to: acquire the state of health (SOH) information of each of the N parallel battery clusters before emergency discharge is initiated; based on the SOH information, set different warning and reduction thresholds for the N parallel battery clusters with different SOH information; wherein step c specifically includes: when the real-time voltage of a certain battery cluster reaches its corresponding different warning and reduction threshold for the first time, and the real-time temperature is lower than the preset temperature safety rejection threshold, step d is triggered.

[0010] Preferably, the central discharge controller is used to, based on SOH information, via To define the relational expression for the first Different warning and reduction thresholds for each battery cluster ,in The preset baseline threshold, For the first State of Health (SOH) information for each battery cluster. This is the preset adjustment coefficient.

[0011] Preferably, the central discharge controller is also used to immediately disconnect the connection of the corresponding battery cluster if any real-time voltage is detected to be lower than a preset voltage safety rejection threshold at any time during emergency discharge; the central discharge controller is also used to immediately disconnect the connection of the corresponding battery cluster if any real-time temperature is detected to be higher than a preset temperature safety rejection threshold at any time during emergency discharge.

[0012] Preferably, the preset response confirmation window period is a fixed time window. The value of the time window is set to be greater than the communication delay between the central discharge controller and the controllable power load, and less than the time it takes for the voltage of the battery cluster that has reached the warning reduction threshold to drop to the preset safety rejection threshold at that power.

[0013] Preferably, the warning reduction threshold is set at the end of the voltage plateau region of the battery cell.

[0014] Preferably, the expected decrease includes: a decrease in the total output current of the system, and the amount of decrease reaches the current reference value, which is determined according to the reduced power rating in the power reduction instruction.

[0015] Preferably, the N parallel battery clusters are lithium-ion battery clusters.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. By adopting an active cross-domain control method, the inherent conflict between protection actions and system-level power stability in the field of battery management is resolved. The voltage of the weakest cell in the parallel battery cluster of the central discharge controller is used as the trigger source. Under the premise that the system bus voltage remains stable, the controller predictively sends power reduction commands to the controllable power load. The controller maintains the discharge connection of all battery clusters, replacing the conventional protection action of disconnecting weak clusters performed by the battery management system. The system failure mode is transformed from the instantaneous drop and cascading collapse of the bus voltage caused by the disconnection of weak clusters to a controllable and smooth step-by-step power reduction. Without impacting the load, the continuous discharge and maximum utilization of the energy storage capacity of all parallel clusters are achieved.

[0017] 2. Based on the control logic, a VT cross-validation mechanism based on the physical characteristics of the battery cell is implemented. This mechanism utilizes the existing temperature data from the monitoring unit. When the central discharge controller receives a signal that the battery cell voltage has reached the warning reduction threshold, it simultaneously verifies whether there is a reasonable temperature rise between the battery cell temperature and its reference temperature. When it is determined that the voltage signal and the temperature signal are physically inconsistent, the controller judges this as voltage sampling distortion or interference and actively suppresses the transmission of power reduction commands. This prevents the control logic of the main scheme from blindly trusting a single-dimensional voltage signal, and enables it to distinguish between true and false information sources. This avoids premature reduction of system functions due to false signals and ensures that the emergency power is maintained at the limit of the system's true physical limits.

[0018] 3. Before emergency discharge, the central discharge controller acquires the health status information of each parallel battery cluster and sets different warning and reduction thresholds for clusters with different health statuses, such as healthy clusters and aging clusters. For example, a lower voltage threshold is set for aging clusters. During emergency discharge, the main scheme's power reduction is triggered only when the cell voltage of a certain cluster touches its corresponding differentiated threshold. Attached Figure Description

[0019] Figure 1 This is a functional architecture and core data flow diagram of the control system of the present invention; Figure 2 This is a failure mode curve diagram of instantaneous bus voltage collapse in the prior art of this invention; Figure 3 This is a timing interaction diagram of the early warning and reduction closed-loop verification control logic of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0021] This invention provides a high-power energy storage emergency power supply discharge control system, applied to an emergency power supply comprising N parallel battery clusters and a controllable power load. The N parallel battery clusters are preferably lithium-ion battery clusters, and the controllable power load can be an inverter or a downstream DC-DC converter in the emergency power supply system. The system architecture mainly includes: a monitoring unit, a total current monitoring unit, and a central discharge controller. The monitoring unit is integrated with or connected to the N parallel battery clusters to acquire cell-level status data; the total current monitoring unit is deployed on the system's total output circuit to acquire system-level output data; and the central discharge controller serves as the system's... The control core establishes communication connections with the monitoring unit, the total current monitoring unit, and the controllable power load, respectively, to receive cell status data and system total current data, and sends control commands to the controllable power load according to the control logic of this invention. The monitoring unit is implemented by the battery management system (BMS) integrated in each parallel battery cluster. Each BMS has the ability to monitor all cells within the battery cluster it manages. The core function of the monitoring unit is to collect and report key status parameters within its cluster at high frequency (with a period of 10ms or faster), and the voltage of the weakest cell in each cluster (i.e., the real-time voltage, hereinafter referred to as the weakest cell). ) and the highest cell temperature (i.e., real-time temperature, hereinafter referred to as The central discharge controller, whose communication interface (such as CAN bus or daisy chain) is used to collect data from the BMS, monitors the voltage sensors in the monitoring unit. The voltage measurement accuracy should be sufficient to distinguish minute changes in cell voltage, not less than 5mV. Temperature sensors are located on the cell surface or near the tabs, with a measurement accuracy not less than 1... The total current monitoring unit can monitor the current installed on the emergency power supply DC bus ( A high-precision Hall current sensor or shunt at the total output, compared to existing BMS technologies that only focus on the battery cell ( This invention relates to the total output current of the system ( Monitoring the total current is the physical basis for achieving closed-loop verification of the subsequent control loop. The response speed of this total current monitoring unit needs to be at the millisecond level to capture instantaneous changes in current, and its range needs to cover the maximum output current of the emergency power supply, providing real-time... The numerical value (an analog voltage signal proportional to the current or a digitized CAN message) is sent to the central discharge controller; the central discharge controller (CDC) can be an industrial-grade microcontroller (MCU) or digital signal processor (DSP) with high-speed computing capabilities and multiple communication interfaces (such as multiple CAN, RS485 / Modbus, PMBus). This CDC is the carrier of the control logic of this invention, acting as the master station, actively polling or receiving data from N monitoring units (slave stations). and Data received from the total current monitoring unit (slave station) The data serves as the primary power source for sending power adjustment commands from the master station to the controllable power load (slave station).

[0022] In emergency discharge scenarios, the specific operating logic of the central discharge controller is as follows: Step a, emergency start: When the main power supply fails, the emergency power system (such as UPS) switching signal triggers the CDC to start emergency discharge. The CDC immediately sends a start command through its communication interface (such as PMBus or CANopen) with the controllable power load (such as inverter), and instructs the controllable power load to use its initial power rating at full load ( Step b, continuous monitoring: During emergency discharge, the CDC continuously receives and refreshes all data from N monitoring units at a high frequency (e.g., 10ms). and Data, synchronized (e.g., 1ms) monitoring The data is used to establish a real-time system status view; step c, early warning trigger determination: the core task of CDC is to prevent any weakest cell from reaching the hard cut-off threshold of BMS ( Systemic failures caused by factors such as 2.5V (e.g., 2.5V) are addressed by setting a warning threshold for voltage reduction in the CDC (CDC). );this Calibration is a crucial deterministic procedure aimed at finding a reliable inflection point in the voltage cliff of a lithium-ion cell during high-current discharge, where the voltage plateau ends and polarization increases sharply. A reproducible calibration procedure includes: selecting cells with different state of health (SOH, e.g., 100% and 80%), performing constant-current discharge at a preset maximum discharge rate (e.g., 1C), and recording the voltage-time curve at high frequency; then analyzing the second derivative of this curve (…). To determine the inflection point voltage where the curvature change is most drastic (e.g., measured at 2.7V), add a safety margin (e.g., 100mV) and the BMS sampling error to this inflection point voltage to finally set... (For example ),this (2.8V) is set higher than usual. (2.5V) and absolute safety veto threshold ( For example, 2.0V); at the same time, the CDC also sets a preset temperature safety veto threshold ( This value is based on the maximum safe operating temperature specified in the cell datasheet (e.g., 75°C). And deduct a safety margin to determine, for example, set to 70. During step b, the CDC continuously reports all data. and Comparison, will and In comparison, when the CDC detects at least one (e.g., cluster 1) First time reaching (2.8V) and the cluster reported (e.g. 45) (lower than) (70) When ), the triggering condition for step c is met.

[0023] Step d, core control action: Once step c is triggered, the CDC immediately performs the core control action of this invention: Step i (maintain grid connection), the CDC strictly prohibits the BMS of cluster 1 from disconnecting, and instead maintains the discharge connection of all N parallel battery clusters (including cluster 1). This fundamentally avoids the damage caused by disconnecting weak clusters. Instantaneous leap and Voltage drop.

[0024] Step ii (Active Power Degradation): The CDC immediately sends a power degradation command to the controllable power load via the communication interface. This command can be a standard PMBus or CANopen message, used to reduce the load's power rating from... (100kW) reduced to a reduced power rating ( The step size of the adjustment ( () can be a preset fixed value or percentage, for example, a 5% reduction. The load responds to this command within milliseconds, and the total system power demand ( The total current dropped to 95kW, causing the total current to decrease. The corresponding reduction is based on the physical characteristics of the battery cell. The current in the weakest cell in cluster 1 decreases, and its terminal voltage ( The voltage drops immediately due to weakened polarization, and may even rise slightly (e.g., stabilize at 2.81V), thus achieving a stable system. ) and protection cells ( The invention serves a dual purpose; the control logic can be further optimized to achieve progressive descent when the system is in... After operating stably at 95kW for a period of time, as power is consumed, the CDC may detect (possibly in cluster 2 or cluster 1 itself) Reached again (2.8V), the CDC will repeat step d(ii), sending a secondary power reduction command to further reduce the load power to 2.8V. (For example, 90kW), this process repeats itself, and the system's output power decreases smoothly in a step-like manner. ), until the battery is depleted or touched. Step e, closed-loop verification and strategy switching: The performance of step d depends on the normal operation of the communication link from the CDC to the load and the normal response of the load itself. In emergency situations, communication may be interrupted due to strong electromagnetic interference (EMI) or the load / inverter firmware may be stuck and unresponsive. This control loop failure will result in: the CDC sending a down-regulation command (step d(ii)), but the load still... The operation caused cluster 1 to... Accelerating downward Ultimately, a hard cut was made, reproducing the background technology. To address the power drop issue, a command-response closed-loop verification mechanism based on the total current monitoring unit is introduced. This mechanism is initiated immediately after the CDC performs step d(ii): Step e1 (verification in progress): While sending the power reduction command (e.g., from 100kW to 95kW), the CDC initiates a preset response confirmation window period ( );this The calibration (deterministic procedure) is crucial: the lower limit must be greater than the communication delay from CDC to the load plus the maximum load response time (e.g., measured at 20ms); the upper limit must be much smaller than that of the weak cell. At full power, from (2.8V) dropped to (2.0V) time (e.g., measured 500ms). It can be set to a fixed time window, such as 100ms.

[0025] During this 100ms window, CDC high-frequency monitoring is reported by the total current monitoring unit. The CDC's goal is to confirm Whether the expected decrease corresponding to the power reduction command has occurred; determination of expected decrease (deterministic procedure): the expected decrease can be quantified as, The decrease value reached based on A defined current reference value, such as the system bus voltage. If it stabilizes at 500V, then (100kW) corresponding Approximately 200A, (95kW) corresponding If the value is approximately 190A, then the expected decline can be determined as follows: The voltage drops by at least 8A (considering measurement error and dynamic response margin); Step e2 (strategy switching): At the end of the 100ms window, the CDC makes a judgment: Scenario A (loop normal): If detected The temperature has dropped to 191A (expected drop occurred). CDC determines: instruction successful, loop normal, system maintains the logic in step d, and continues... Operating at power; Scenario B (loop failure): If detected The status remains at 199A (no expected decrease). The CDC determines that the command failed (communication or load failure). The CDC immediately determines that the smooth dimensionality reduction path (step d) is no longer available and the control strategy must be switched. The CDC no longer attempts to control the load but instead sends a forced disconnect command to the BMS of cluster 1. This action sacrifices cluster 1 but protects the remaining clusters. Each cluster is protected from cascading failures due to insufficient load reduction; furthermore, to improve the reliability of the system of the present invention, the control logic of the present invention can also integrate the following two preferred cooperative mechanisms: one preferred cooperative mechanism is the VT cross-validation mechanism, used to address the problem of voltage signal distortion in the monitoring unit. In lithium-ion cells, during high-current discharge, the voltage (V) drop and internal resistance heat generation (T) are physically strongly coupled. For truly weak cells (high polarization, high internal resistance), their... touch At that time, its temperature rise ( It is inevitable, and conversely, if touch ,but If no reasonable temperature rise is observed (i.e., inconsistent VT), then it is highly likely that... The signal is distorted due to interference or sampling circuit failure.

[0026] Therefore, the central discharge controller can add a VT cross-validation step before step d: when step c ( touch When triggered, the CDC does not immediately perform step d (adjustment), but immediately obtains the real-time temperature of the cell. ), obtain temperature reference ( ), CDC judgment Is it below the reasonable temperature rise threshold? Calibration of reasonable temperature rise threshold (deterministic procedure): The calibration aims to determine the actual performance of low-voltage cells. The minimum temperature rise that should be present at that time Can be set to be based on and preset temperature rise margin ( The threshold of the sum of ) (e.g. 5) This can be calibrated through a thermal model of the battery cell or by actual measurement (see [reference]). Calibration procedure, synchronously record temperature rise); Logical judgment: IF ( ),For example , , ,but Determine: VT matches, signal is real, then proceed to step d (adjustment), IF ( ),For example If the VT (Virtual Transmission Time) is inconsistent and the signal is false, then step d is suppressed, the false signal is ignored, and a maintenance alarm is recorded. This mechanism avoids premature degradation of system functionality due to false signals. Another preferred collaborative mechanism is an asymmetric degradation strategy based on State of Health (SOH). Throughout the entire lifecycle, the SOH of parallel clusters will inevitably differ. If a unified... If the aging cluster with the lowest SOH is prematurely triggered, it will throttle the full power output capability of the healthy cluster.

[0027] To this end, the central discharge controller is also used for: Step i: Before emergency discharge is initiated (e.g., during system idle or charging), the CDC obtains the health status (SOH) information of each cluster from N monitoring units (BMS) (SOH is a standard estimation parameter of the BMS). Step j: Based on the SOH information, the CDC sets different warning and reduction thresholds for different clusters with different SOH information. The logic behind this differentiation is: the lower the SOH of an aging cluster, the deeper the discharge depth it can withstand. The lower the value, the better. This setting can be used for simple grouping (e.g., a healthy cluster with SOH > 80%). SOH 80% of the aging clusters, More preferably, this setting can employ a function based on SOH to achieve finer control, for example: ,in, It is the first The differential threshold for each cluster; It is their health status information. It is a preset baseline threshold (for example, the threshold when SOH=100%, set to 2.8V). This is a preset adjustment factor (this factor can be determined through calibration; for example, if the threshold is set to 2.6V when SOH=70%, it can be calculated backwards). Step c-1 (Asymmetric Trigger): During emergency discharge, the specific logic adjustment for step c is as follows: when a certain battery cluster (cluster) is detected... )of The first time it reaches its corresponding different warning reduction threshold ( Step d is triggered only when the SOH reaches 70% of its aging clusters (its...). This mechanism ensures that... The system will not trigger drooping when the voltage drops to 2.8V (at 2.6V), and will continue to operate at full power until the cluster voltage drops to 2.6V, at which point a first-level droop will be triggered. This greatly extends the system's full-power emergency time. Finally, to ensure the absolute safety of the system under any circumstances, the central discharge controller is also used for the highest priority safety veto logic: at any moment during emergency discharge, if any... Voltage below the preset safety threshold ( (e.g., 2.0V), immediately instruct the disconnection of the corresponding battery cluster; at any moment during emergency discharge, if any... Temperature exceeding the preset safety threshold ( For example, 70 The corresponding battery cluster is immediately disconnected. These two rejection logics have higher priority than all the above-mentioned downgrade logic, VT verification logic, and SOH strategy logic.

[0028] Example 1: In a specific application of a high-power energy storage emergency power supply discharge control system, the emergency power supply system consists of 20 parallel lithium-ion battery clusters. It supplies power to a data center through a controllable power load, i.e., an inverter. Its initial power rating is 1.2MW. The central discharge controller (CDC) in this system is set with an early warning and reduction threshold. Preset temperature safety veto threshold Preset temperature rise margin for VT cross-validation When the emergency discharge was initiated and the system was operating at full power (1.2MW) for 15 minutes, the base monitoring unit (BMS) of cluster 5 reported a distorted real-time voltage due to a transient strong electromagnetic interference (EMI) incident on its voltage sampling line. This value is lower than the set value. However, the physical voltage of the battery cell remained at 3.1V; the real-time temperature of the battery cell reported by the monitoring unit... At this time, the Central Discharge Controller (CDC) receives a signal from cluster 5. This value reached for the first time. Received its This value is lower than Before implementing the downsizing action, the CDC initiated the VT cross-validation mechanism to obtain the average temperature of 20 clusters as a temperature baseline. ,in accordance with Calculate the reasonable temperature rise threshold, obtain the VT inconsistency judgment for CDC, and report the data from cluster 5. That is, 30.1 , and the calculated That is, 35 The comparison was performed; the result was determined as follows: This result meets the criteria for determining VT inconsistency. Based on this determination, the CDC will [take further action]. touch The event was determined to be a false signal caused by voltage signal distortion. Accordingly, the CDC suppressed the originally planned step d, that is, it did not send a power reduction command to the controllable power load. The emergency power system continued to operate stably at the initial power rating of 1.2MW. The system bus voltage did not fluctuate. The CDC only recorded a maintenance alarm for abnormal voltage signal of cluster 5 monitoring unit to the background system. By using the existing voltage and temperature information in the monitoring unit for collaborative judgment, the problem between blindly trusting a single signal and misjudging system functions was solved, and unnecessary reduction of emergency power due to monitoring signal distortion was avoided.

[0029] Example 2: This example provides objective experimental data to compare the system bus stability of the discharge control system of this invention with that of the existing BMS standard protection mechanism when the weakest cell reaches the voltage threshold; Test platform construction: A high-power energy storage emergency power supply discharge test platform is built, which includes a battery system consisting of 10 parallel lithium-ion battery clusters of the same specifications, with the bus voltage calibrated at 500V; A controllable power load (high-frequency electronic load) is used to simulate constant power consumption under emergency conditions, with an initial power rating ( The power output is set to 250kW, and the platform is equipped with a high-frequency data acquisition system (sampling rate 1ms) to synchronously monitor the weakest cell voltage of each battery cluster. ) and system bus voltage ( To simulate the inconsistencies of lithium-ion cells, cluster 1 was replaced with a slightly aged cluster with a state of harmlessness (SOH) of 90%, making it the weakest battery cluster in the system. It will degrade faster than the other 9 healthy clusters under high current; Experimental group design: Control group (simulating existing technology): This group of experiments disables the active degradation function of the central discharge controller (CDC), and all 10 parallel clusters' monitoring units (BMS) adopt the standard protection logic of existing technology, that is, setting a rigid discharge cutoff voltage threshold. The voltage is 2.50V, when any cluster of When the voltage drops below 2.50V, the BMS of this battery cluster will immediately disconnect the battery cluster as a protection measure. In this invention's sample group (using the scheme of this invention): the central discharge controller (CDC) is activated in this test group. The CDC is calibrated according to the specific implementation procedure, and a warning and reduction threshold is set. Set a preset voltage safety veto threshold. (BMS) Also set to this value), the CDC communicates with the controllable power load, and the single power reduction quota of the power reduction command is set to 10% of the total power (i.e., 25kW); test process and data recording: both sets of tests start from a fully charged state, start emergency discharge, and the controllable power load is set to 250kW ( The initial power rating of cluster 1 is set, and the data acquisition system continuously monitors the data. and system bus voltage The timing data of key event points during the experiment were compared in Table 1 until the system failed.

[0030] Table 1: Comparison of system response between the control group and the present invention sample group when the weakest cell reaches the threshold.

[0031] In the control group, when cluster 1 It reached 2.50V at 600.15 seconds. At that time, its BMS performs a standard disconnect action; this action forces the remaining 9 clusters to bear the total load power of 250kW, and the current of each cluster jumps by 11.1% instantaneously. This leads to system bus voltage The voltage dropped instantaneously from 499.8V to 441.3V, a decrease of over 11%, causing the load to instantly shut down due to undervoltage, triggering a system-wide cascading failure; in the sample group of this invention, when cluster 1... The warning threshold of 2.80V was first reached at 540.12 seconds. At that time, the central discharge controller (CDC) did not disconnect any clusters, but maintained the discharge connection of all N battery clusters and sent a first-level power reduction command to the controllable power load; after the load responded, the total power smoothly decreased to 225.0kW; due to the total current Decrease, cluster 1 The voltage rose slightly to 2.811V due to reduced polarization; the system bus voltage... Only a transient disturbance of less than 0.2% occurred, and it quickly stabilized at 499.8V. At 680.14 seconds, the CDC repeated this warning reduction process, achieving a two-stage smooth power reduction. The test results show that the control system, through the coordinated action of warning reduction and grid connection maintenance, transforms the system failure mode from the instantaneous collapse of bus voltage caused by the disconnection of weak clusters to a controllable and smooth step-like power reduction, maintaining the continuous and stable operation of the emergency power supply system without impacting the load.

[0032] Example 3: This example combines Figures 1 to 3 A description of a high-power energy storage emergency power supply discharge control system, such as... Figure 1As shown, the system bus serves as an external physical data source, and its system physical current is monitored by the A2: total current monitoring unit processing process, which reports the total system output current to the A3: central discharge controller core processing process. N parallel battery clusters serve as external physical data sources / destination, and their cell physical status is monitored by the A1: monitoring unit processing process. A1 reports real-time voltage and temperature information to A3 and reports SOH information to the D1: health status SOH information module. D1 determines the differentiated warning reduction threshold based on the SOH information and inputs it into A3. The A3: central discharge controller core processing process, based on the received real-time voltage, real-time temperature, total system output current, and differentiated warning reduction threshold, sends a power reduction command to the controllable power load external physical data destination, and under specific conditions such as strategy switching, sends a disconnect command to the N parallel battery clusters.

[0033] like Figure 2 As shown in the figure, the graph uses time (s) as the horizontal axis, the system bus voltage (V) as the left vertical axis, and the weakest cell voltage (V) as the right vertical axis. The graph shows that the system bus voltage, represented by the solid line, remained stable at 500V before 540s, while the weakest cell voltage, represented by the dashed line, continuously decreased, reaching the 2.8V threshold at 540s. After 540s, the system bus voltage experienced a momentary collapse, plummeting from 500V to below 445V. The rate of decrease in the weakest cell voltage also changed and continued to decline. Figure 3 As shown, the process begins with emergency discharge startup, at which point the central discharge controller instructs the controllable power load to operate at the initial power rating. During the continuous monitoring phase, the monitoring unit reports the cell voltage and temperature to the central discharge controller, while the total current monitoring unit reports the total system output current. When a voltage reaches the warning value, the central discharge controller verifies whether the cell voltage has reached the warning reduction threshold for the first time and verifies that the temperature is below the safety rejection threshold. After confirmation, the central discharge controller sends a power reduction command to the controllable power load and simultaneously initiates response confirmation window monitoring to the total current monitoring unit. The total current monitoring unit confirms that the total current has decreased as expected and reports it to the central discharge controller. The central discharge controller verifies the change successfully and maintains the current strategy, allowing the system to continue operating at the reduced power.

[0034] Example 4: This example demonstrates the closed-loop verification and strategy switching mechanism of a control system when facing control loop failure; the test platform and parameter settings are the same as the sample group of the present invention in Example 2, with initial power rating. 250kW, preset response confirmation window period The timeframe is set to 100ms, and the expected current drop is set to a base value of 8A. At the 540.12th second of emergency discharge operation, cluster 1... First time reaching (2.80V), and its Below (70) The central discharge controller (CDC) determines that the triggering condition is met and immediately proceeds to step d, which involves maintaining the connection of all N clusters while simultaneously sending a first-level power reduction command (target power) to the controllable power load. At the moment the command was sent, the CAN communication bus between the CDC and the load experienced data frame corruption due to strong electromagnetic interference (EMI), and the load did not receive the power reduction command. The CDC then began monitoring the closed-loop verification logic in step e of the power reduction command initiation process. A preset response confirmation window timer of 100ms is started; since the load does not respond, the controllable power load continues to operate at [unspecified rate]. (250kW) Operation, total system output current It remained at approximately 200A, without the expected decrease (i.e., a decrease of at least 8A); within 100ms When the window period ends, the CDC determines If the expected decrease does not occur, this event is immediately classified as a loop failure. The CDC determines that the smooth dimensionality reduction path is no longer available and immediately switches the control strategy to strategy degradation. The CDC no longer attempts to send commands to the load but instead proceeds to step e2, actively sending a forced disconnect command to the cluster 1 monitoring unit (BMS). Cluster 1 is safely disconnected, and the system bus voltage... Therefore, the system experienced a momentary drop from 500.1V to 445.2V, but the remaining N-1 clusters withstood the drop and continued to operate. The system avoided further damage due to the continued voltage drop of cluster 1 cells. And this could ultimately lead to a cascading collapse, the worst-case scenario.

[0035] Example 5: In the control system of the present invention, the operation of the asymmetric sag strategy based on the state of health (SOH) and the VT cross-validation mechanism depends on the SOH information and the preset temperature rise margin. The pre-determination of these two key parameters ensures that a reproducible calibration procedure is performed before system deployment or during periodic maintenance to obtain the State of Health (SOH) information for each parallel battery cluster. Taking cluster 1 as an example, its initial nominal capacity is 200 Ah, and this cluster... 2 In a standard testing environment, the cells are fully charged to the cutoff condition, an example of which is constant current and constant voltage charging until the average cell voltage is 4.2V and the current is less than 0.05C, followed by resting for 1 hour; then, they are discharged at a constant current of 0.5C, i.e., 100A, until the voltage of the weakest cell in the cluster is reached. The discharge cutoff voltage of 2.5V was reached; during this process, the total charge released by the cluster was accurately recorded, with a measured value of 181.2 Ah; the SOH information of cluster 1 was determined to be... This procedure is repeated on all N clusters. The central discharge controller (CDC) stores a list containing SOH information for all clusters for use in asymmetric downregulation logic.

[0036] To determine the preset temperature rise margin required for the VT cross-validation mechanism High-rate discharge thermal characteristics were tested; a representative cluster with a state of harmlessness (SOH) of 95% was selected and placed at 25.0°C. In a constant temperature environment, wait for its internal temperature to equalize with the ambient temperature. All remained stable at 25.0 Then, a constant current discharge of 200A (1C rate) is applied; during this discharge process, the voltage of the weakest cell is monitored at high frequency. The real-time temperature corresponding to this battery cell ;when The voltage dropped for the first time and reached the warning threshold of 2.80V. Immediately lock the current time. The actual measurement was 30.7. ,Record It is 25.0 The actual temperature rise under this operating condition was calculated. Repeat this high-rate discharge test 5 times to obtain a set of actual temperature rise data, including 5.7. 5.5 5.9 5.6 5.8 We take the lower limit of the statistical value for this set of data, which is set here to be 5.7 points below the mean. conservative integer value 5 This value is used as the preset temperature rise margin. After the calibration procedure is completed, the Central Discharge Controller (CDC) internally stores the SOH information for each cluster, such as 90.6% for cluster 1, 98.2% for cluster 2, and the global SOH information. That is, 5 Based on this, the CDC performs asymmetric sag logic and VT cross-validation logic during emergency discharge.

[0037] Example 6: To determine the relationship in the asymmetric downsizing strategy based on the state of health (SOH) Required key parameters and Before system deployment, a multi-point calibration procedure should be performed; in this relationship, It is the first The warning reduction threshold for each cluster, It is a preset baseline threshold. It is the preset adjustment coefficient. It is the first The health status (SOH) information of each cluster is required; this procedure needs to obtain the SOH information of representative clusters, taking clusters A, B, and C as examples, through capacity testing (at 25°C). In a 2°C environment, the SOH was measured beforehand (discharged at a constant current of 0.5C to a cutoff of 2.5V), and the results were obtained. , and These three clusters were discharged at the same 1C high-rate, and their high-frequency monitoring was performed. To determine the inflection point voltage at the end of each voltage plateau region, the measured inflection point voltages were 2.80V for cluster A, 2.73V for cluster B, and 2.67V for cluster C. Based on this, the central discharge controller (CDC) calibrated its parameters and adjusted the voltage of cluster A (…). The inflection point voltage of 2.80V is set as the preset reference threshold. ,Right now Using data points from clusters B and C, a preset adjustment coefficient is calculated through linear regression fitting. ,get ,Should and The values ​​are stored in the CDC and used for real-time calculation of any SOH cluster during emergency discharge. .

[0038] Furthermore, to address the single preset temperature rise margin in the VT cross-validation mechanism To overcome the limitations under different operating conditions, a multi-dimensional thermal property calibration procedure can be developed. This procedure selects representative clusters of different SOH information, which are used here. and Taking two groups as examples, each cluster was subjected to discharge tests at multiple different constant discharge rates, including 0.5C, 1.0C, and 1.5C, while simultaneously testing under different ambient temperature conditions. Below, containing 15 With 25 Repeat the above tests; under each test combination, with 1.0C For example, all records are made when touch The actual temperature rise (determined by the aforementioned SOH function) All these measured temperature rise data are stored in a multidimensional lookup table of the CDC. During emergency discharge, the CDC uses the data from real-time monitoring... 1. Estimated current discharge rate (based on...) (Calculated from N) and ambient temperature The table can be dynamically queried to obtain a reasonable temperature rise threshold that matches the current operating conditions, replacing the fixed threshold. This allows it to adapt to changes in thermal properties under different operating conditions.

[0039] Example 7: After a high-power energy storage emergency power supply discharge control system is deployed at a specific project site, in order to match its control parameters with the physical characteristics of the battery clusters and controllable power loads at the site, a field calibration procedure can be performed. This procedure is used to determine the preset response confirmation window period. Its upper limit depends on the cell's safe drop time, and its lower limit depends on the system's communication and response latency. The first step in calibration is to measure the communication and response latency. This is achieved by sending 100 power reduction commands to the controllable power load through the central discharge controller (CDC), monitoring the actual load power response using a high-frequency acquisition device (10µs sampling rate), and recording the longest time from command transmission to response completion. The measured values ​​are then obtained. The second step in calibration is to determine the safe drop time, ensuring the system operates at full power. When the weakest cell is running, touch At (2.8V), suppress the droop command and monitor. from The voltage drops to the preset safety threshold. (Set here to 2.2V) The shortest time required, as measured, is... ,final Set to greater than And much smaller The value is set to 100ms here to provide a sufficient time window for control loop verification; this procedure is used to determine the single reduction quota of the power reduction command. The determination of this parameter aims to The magnitude of the rebound and Achieving a balance between voltage disturbances; at full power Next, when touch At (2.8V), different derating rates were tested respectively, from... The percentage increases from 2% to 15%; when the reduction is less than 4%, The recovery value is less than 8mV, making it susceptible to noise interference. When the reduction in the rating exceeds 10%, A sudden drop exceeding 1.2% could impact backend load, so an 8% reduction was ultimately selected. The rebound ranged from 15mV to 22mV. The dip can be controlled within 0.7%; this procedure verifies the preset voltage safety rejection threshold. and preset temperature safety veto threshold These two parameters are set according to the manufacturer's specifications for the lithium-ion battery cells used. The specifications state that the absolute minimum discharge voltage of the battery cell is 2.0V and the absolute maximum operating temperature is 80°C. To preserve a safety margin and account for the sampling delay of the BMS, Set to 2.2V (200mV higher than the lower limit). Set to 70 (Below the upper limit of 10) This calibration procedure ensures that all control actions of the system are performed within the safety boundaries of the cell manufacturer.

[0040] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A high-power energy storage emergency power supply discharge control system, applied in an emergency power supply comprising N parallel battery clusters and a controllable power load, characterized in that, The system includes a monitoring unit, a total current monitoring unit, and a central discharge controller. The monitoring unit is used to monitor the real-time voltage and real-time temperature of the cells in each of the N parallel battery clusters. The total current monitoring unit is used to monitor the total output current of the emergency power supply system. The central discharge controller is connected to the monitoring unit, the total current monitoring unit, and the controllable power load via a communication interface; the central discharge controller is used for: Step a: In response to the emergency discharge, the controllable power load is instructed to operate at the initial power rating. Step b: Continuously receive real-time voltage and real-time temperature when the controllable power load is running; Step c: When at least one real-time voltage is detected to reach the preset warning reduction threshold for the first time, and the real-time temperature is lower than the preset temperature safety rejection threshold; Step d, proceed as follows: Step i, maintain the discharge connection of all N parallel battery clusters, and replace the disconnection of battery clusters that have reached the warning reduction threshold; Step ii, send a power reduction command to the controllable power load, so that the operating power of the controllable power load is reduced to a reduced power rating. Step e, after sending the power reduction command, is also used for: step e1, within the preset response confirmation window period, monitoring whether the total output current of the system has decreased as expected corresponding to the power reduction command; step e2, if the total output current of the system has not decreased as expected, switching the control strategy and instead instructing to disconnect the battery cluster that has reached the warning reduction threshold.

2. The high-power energy storage emergency power supply discharge control system according to claim 1, characterized in that, After the controllable power load operates at the reduced power rating, the central discharge controller is also used to repeatedly send a power reduction command to the controllable power load if the real-time voltage is detected to reach the warning reduction threshold again, so that the operating power of the controllable power load is further reduced to a lower reduced power rating.

3. The high-power energy storage emergency power supply discharge control system according to claim 1, characterized in that, Before performing step d, the central discharge controller also acquires the real-time temperature of the cell that has reached the warning reduction threshold, and acquires the average temperature of N parallel battery clusters as the temperature reference; it determines whether the real-time temperature is lower than the reasonable temperature rise threshold determined by the sum of the temperature reference and the preset temperature rise margin; step d is performed only when the real-time temperature is determined to be not lower than the reasonable temperature rise threshold; when the real-time temperature is determined to be lower than the reasonable temperature rise threshold, step d is suppressed.

4. The high-power energy storage emergency power supply discharge control system according to claim 1, characterized in that, The central discharge controller is also used to: acquire the state of health (SOH) information of each of the N parallel battery clusters before emergency discharge is initiated; and set different warning and reduction thresholds for the N parallel battery clusters with different SOH information based on the SOH information. Step c specifically includes: when the real-time voltage of a certain battery cluster reaches its corresponding early warning reduction threshold for the first time, and the real-time temperature is lower than the preset temperature safety rejection threshold, step d is triggered.

5. The high-power energy storage emergency power supply discharge control system according to claim 4, characterized in that, The central discharge controller is used to, based on SOH information, via To define the relational expression for the first Different warning and reduction thresholds for each battery cluster ,in The preset baseline threshold, For the first State of Health (SOH) information for each battery cluster. This is the preset adjustment coefficient.

6. The high-power energy storage emergency power supply discharge control system according to claim 1, characterized in that, The central discharge controller is also used to immediately disconnect the connection of the corresponding battery cluster if any real-time voltage is detected to be lower than the preset voltage safety threshold at any time during emergency discharge; the central discharge controller is also used to immediately disconnect the connection of the corresponding battery cluster if any real-time temperature is detected to be higher than the preset temperature safety threshold at any time during emergency discharge.

7. The high-power energy storage emergency power supply discharge control system according to claim 1, characterized in that, The preset response confirmation window period is a fixed time window. The value of the time window is set to be greater than the communication delay between the central discharge controller and the controllable power load, and less than the time it takes for the voltage of the battery cluster that has reached the warning reduction threshold to drop to the preset safety rejection threshold at that power.

8. The high-power energy storage emergency power supply discharge control system according to claim 1, characterized in that, The warning threshold for voltage reduction is set at the end of the voltage plateau region of the battery cell.

9. A high-power energy storage emergency power supply discharge control system according to claim 1, characterized in that, The expected decrease includes: a decrease in the total output current of the system, and the amount of the decrease reaches the current reference value, which is determined according to the reduced power rating in the power reduction instruction.

10. A high-power energy storage emergency power supply discharge control system according to claim 1, characterized in that, N parallel battery clusters are lithium-ion battery clusters.

Images