Super capacitor energy storage system and hierarchical management system thereof

Abstract

The invention relates to a super-capacitor energy storage system and a hierarchical management system thereof, and belongs to the technical field of energy storage system management. The super-capacitor energy storage system comprises N + X independent super-capacitor clusters which are cascaded through a bus, N is the number of working clusters, X is the number of standby clusters, and a hierarchical management system of a three-level architecture comprising a stack management unit, a cluster calculation unit and a super-capacitor acquisition unit is adopted. And the stack management unit is used for judging whether the super capacitor cluster fails or not according to the super capacitor cluster maximum and minimum value sent by each cluster calculation unit, and controlling fault cluster isolation and standby cluster input according to the state and position of the standby cluster and the position of the fault cluster. According to the invention, a cluster-level redundancy design is adopted, switching control is carried out when a working cluster breaks down, isolation of a fault cluster and input of a standby cluster are realized, the influence of a single-cluster fault is reduced to the minimum, and continuous power supply of the super-capacitor energy storage system is ensured.

Description

Technical Field

[0001] This invention relates to a supercapacitor energy storage system and its hierarchical management system, belonging to the field of energy storage system management technology. Background Technology

[0002] With the increasing proportion of new energy power generation and the surge in demand for flexible grid regulation capabilities, large-capacity supercapacitor energy storage systems are playing an increasingly important role in power systems due to their high power density and rapid charge / discharge characteristics. Currently, most mainstream supercapacitor energy storage systems adopt a hierarchical management architecture. For example, Chinese patent application CN120749856A discloses a supercapacitor energy storage system, its control equipment, and its energy control method. The supercapacitor management system unit has three levels: the first-level supercapacitor management unit is connected to both the high-voltage box unit and the supercapacitor unit; the second-level supercapacitor management unit is located inside the high-voltage box unit; and the third-level supercapacitor management unit is connected to the energy management system unit, the high-voltage box unit, and the current conversion unit. This solution can achieve safe and efficient management of supercapacitor energy, but existing technologies still have the following drawbacks: 1. Insufficient reliability: Traditional energy storage systems mostly adopt a single cluster parallel or simple series structure. When a cluster fails due to a fault, the entire energy storage system may be taken out of operation, affecting the continuity of power supply. 2. Low thermal management efficiency: Supercapacitors are prone to heat generation during operation. Traditional thermal management (such as air cooling or fixed power liquid cooling) relies on manually set parameters and cannot be dynamically adjusted according to real-time load and temperature, which can easily lead to energy waste or local overheating. 3. Poor communication and addressing flexibility: In the existing system, the station address of the cluster needs to be manually configured on-site, which is inefficient; 4. Complex upgrade and maintenance: Upgrading the firmware across the entire site requires cluster-by-cluster operation, which is time-consuming and labor-intensive. Summary of the Invention

[0003] The purpose of this invention is to provide a supercapacitor energy storage system and its hierarchical management system to solve the problem of low power supply reliability caused by the entire energy storage system shutting down when a cluster failure occurs in current supercapacitor energy storage systems.

[0004] To address the aforementioned technical problems, this invention provides a hierarchical management system for a supercapacitor energy storage system. This system employs a three-tiered architecture comprising a stack management unit, a cluster computing unit, and a supercapacitor acquisition unit. Each supercapacitor is equipped with a corresponding supercapacitor acquisition unit to collect its status data. The cluster computing unit performs maximum / minimum value calculations based on the status data collected by the acquisition units within its cluster. The stack management unit determines whether a supercapacitor cluster is faulty based on the maximum / minimum values ​​submitted by each cluster computing unit, and controls the isolation of faulty clusters and the activation of backup clusters based on the status and location of backup clusters and the location of faulty clusters. The supercapacitor energy storage system managed by this hierarchical management system comprises N+X independent supercapacitor clusters cascaded via a bus, where N is the number of working clusters and X is the number of backup clusters.

[0005] Furthermore, the heap management unit uses a dynamic allocation mechanism based on power-on sequence and protocol handshake to allocate addresses to each cluster computing unit. The allocation process is as follows: After the heap management unit starts up, it broadcasts an address and IP call frame to the cluster computing unit. The cluster computing unit sends its own IP and address according to the received IP call frame. The heap management unit determines whether the address and IP of the cluster computing unit conform to the plan. If they do not conform, it powers down all cluster computing units and powers up the cluster computing unit. The heap management unit sends an address precoding frame to the cluster computing unit. After the cluster computing unit confirms, the heap management unit sends an address change frame. The cluster computing unit changes its IP and address according to the address change frame.

[0006] Furthermore, the stack management unit is also used to remotely upgrade the firmware of all cluster computing units and supercapacitor acquisition units, and the upgrade adopts a differential upgrade+ and breakpoint resume upgrade mechanism.

[0007] Furthermore, the stack management unit is also used to communicate with the fire protection system, environmental monitoring system and thermal management system to realize the linkage control of energy storage-fire protection-thermal management.

[0008] Furthermore, the stack management unit integrates a thermal management module. The supercapacitor acquisition unit is used to upload the acquired supercapacitor surface temperature, liquid cooling pipeline inlet / outlet temperature, and ambient temperature to the cluster computing unit. The cluster computing unit uploads the above information to the stack management unit. The stack management unit is used to determine the cooling strategy according to the grid dispatch instructions and / or according to the information uploaded by the cluster computing unit.

[0009] Furthermore, the cooling strategy formulated by the stack management unit includes: When the surface temperature of the supercapacitor is less than the first set threshold and the temperature difference between the inlet and outlet of the liquid cooling pipeline is less than the second set threshold, the speed of the liquid cooling pump is reduced to put it into energy-saving mode. When the surface temperature of the supercapacitor is not less than the first set threshold or the temperature difference between the inlet and outlet of the liquid cooling pipeline is greater than the third set threshold, the speed of the liquid cooling pump is controlled to increase so that it is in standard mode. When the surface temperature of the supercapacitor exceeds the fourth set threshold or the temperature difference of a local module exceeds the fifth set threshold, the liquid cooling pump speed is controlled to reach the maximum, and the air conditioner is started to cool. The first set threshold is less than the fourth set threshold, the second set threshold is less than the third set threshold, the third set threshold is less than the fifth set threshold, and the fifth set threshold is less than the first set threshold.

[0010] Furthermore, when controlling the liquid cooling pump and / or air conditioning, the stack management unit updates the strategy at set intervals and judges the cooling effect based on the surface temperature of the supercapacitor. If the requirement is not met within a certain period of time, an alarm is triggered.

[0011] Furthermore, the stack management unit controls the isolation of faulty clusters and the activation of backup clusters by sending corresponding action commands to the input-output expansion device, which is used to control the connection and disconnection between each supercapacitor cluster and the PCS.

[0012] Furthermore, the stack management unit communicates with the cluster computing unit via the UDP protocol, and the cluster computing unit communicates with the supercapacitor acquisition unit via the CAN protocol.

[0013] This invention also provides a supercapacitor energy storage system, which includes N+X independent supercapacitor clusters cascaded via a bus and a hierarchical management system. The hierarchical management system adopts a three-tier architecture comprising a stack management unit, a cluster computing unit, and a supercapacitor acquisition unit. Each supercapacitor is equipped with a corresponding supercapacitor acquisition unit for collecting its status data. The cluster computing unit performs maximum / minimum value calculations based on the status data collected by the acquisition units of each supercapacitor cluster within its cluster. The stack management unit determines whether a supercapacitor cluster is faulty based on the maximum / minimum values ​​of the supercapacitor clusters sent by each cluster computing unit, and controls the isolation of faulty clusters and the activation of backup clusters based on the status and location of backup clusters and the location of faulty clusters. The supercapacitor energy storage system managed by this hierarchical management system comprises N+X independent supercapacitor clusters cascaded via a bus, where N is the number of working clusters and X is the number of backup clusters.

[0014] Furthermore, the heap management unit uses a dynamic allocation mechanism based on power-on sequence and protocol handshake to allocate addresses to each cluster computing unit. The allocation process is as follows: After the heap management unit starts up, it broadcasts an address and IP call frame to the cluster computing unit. The cluster computing unit sends its own IP and address according to the received IP call frame. The heap management unit determines whether the address and IP of the cluster computing unit conform to the plan. If they do not conform, it powers down all cluster computing units and powers up the cluster computing unit. The heap management unit sends an address precoding frame to the cluster computing unit. After the cluster computing unit confirms, the heap management unit sends an address change frame. The cluster computing unit changes its IP and address according to the address change frame.

[0015] Furthermore, the stack management unit is also used to remotely upgrade the firmware of all cluster computing units and supercapacitor acquisition units, and the upgrade adopts a differential upgrade+ and breakpoint resume upgrade mechanism.

[0016] Furthermore, the stack management unit is also used to communicate with the fire protection system, environmental monitoring system and thermal management system to realize the linkage control of energy storage-fire protection-thermal management.

[0017] Furthermore, the stack management unit integrates a thermal management module. The supercapacitor acquisition unit is used to upload the acquired supercapacitor surface temperature, liquid cooling pipeline inlet / outlet temperature, and ambient temperature to the cluster computing unit. The cluster computing unit uploads the above information to the stack management unit. The stack management unit is used to determine the cooling strategy according to the grid dispatch instructions and / or according to the information uploaded by the cluster computing unit.

[0018] Furthermore, the cooling strategy formulated by the stack management unit includes: When the surface temperature of the supercapacitor is less than the first set threshold and the temperature difference between the inlet and outlet of the liquid cooling pipeline is less than the second set threshold, the speed of the liquid cooling pump is reduced to put it into energy-saving mode. When the surface temperature of the supercapacitor is not less than the first set threshold or the temperature difference between the inlet and outlet of the liquid cooling pipeline is greater than the third set threshold, the speed of the liquid cooling pump is controlled to increase so that it is in standard mode. When the surface temperature of the supercapacitor exceeds the fourth set threshold or the temperature difference of a local module exceeds the fifth set threshold, the liquid cooling pump speed is controlled to reach the maximum, and the air conditioner is started to cool. The first set threshold is less than the fourth set threshold, the second set threshold is less than the third set threshold, the third set threshold is less than the fifth set threshold, and the fifth set threshold is less than the first set threshold.

[0019] Furthermore, when controlling the liquid cooling pump and / or air conditioning, the stack management unit updates the strategy at set intervals and judges the cooling effect based on the surface temperature of the supercapacitor. If the requirement is not met within a certain period of time, an alarm is triggered.

[0020] Furthermore, the stack management unit controls the isolation of faulty clusters and the activation of backup clusters by sending corresponding action commands to the input-output expansion device, which is used to control the connection and disconnection between each supercapacitor cluster and the PCS.

[0021] Furthermore, the stack management unit communicates with the cluster computing unit via the UDP protocol, and the cluster computing unit communicates with the supercapacitor acquisition unit via the CAN protocol.

[0022] The beneficial effects of this invention are as follows: The supercapacitor energy storage system of this invention includes N+X independent supercapacitor clusters cascaded via a bus, where N is the number of working clusters and X is the number of standby clusters. It employs a hierarchical management system with a three-tiered architecture, comprising a stack management unit, a cluster calculation unit, and a supercapacitor acquisition unit. The supercapacitor acquisition unit collects the status data of each supercapacitor, the cluster calculation unit calculates the maximum and minimum values ​​of the supercapacitor cluster based on the status data of each supercapacitor, and the stack management unit determines whether a supercapacitor cluster is faulty based on the maximum and minimum values ​​of the supercapacitor clusters sent by each cluster calculation unit. It also controls the isolation of faulty clusters and the activation of standby clusters based on the status and location of standby clusters and the location of faulty clusters. This invention adopts a cluster-level redundancy design, performing switching control when a working cluster fails to isolate the faulty cluster and activate the standby cluster, minimizing the impact of a single cluster failure and ensuring continuous power supply to the supercapacitor energy storage system. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the hierarchical management system architecture of the supercapacitor energy storage system of the present invention; Figure 2 This is a flowchart of the dynamic address allocation process used in the hierarchical management system of the supercapacitor energy storage system of this invention. Figure 3 This is a schematic diagram of the cluster redundancy switching logic used in the hierarchical management system of the supercapacitor energy storage system of this invention. Detailed Implementation

[0024] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0025] This invention employs a cluster-level redundancy design, which performs switching control when a working cluster fails, thereby isolating the faulty cluster and putting the backup cluster into operation. This minimizes the impact of a single cluster failure and ensures continuous power supply to the supercapacitor energy storage system.

[0026] Supercapacitor Energy Storage System Implementation The supercapacitor energy storage system comprises N+X independent supercapacitor clusters cascaded via a bus. Each supercapacitor cluster is equipped with a corresponding cluster computing unit. Each supercapacitor cluster includes multiple supercapacitors, and each supercapacitor is equipped with a corresponding supercapacitor acquisition unit for collecting status data. The cluster computing unit performs maximum / minimum value calculations based on the status data collected by the acquisition units of each supercapacitor cluster within its cluster. The stack management unit determines whether a supercapacitor cluster is faulty based on the maximum / minimum values ​​submitted by each cluster computing unit, and isolates the faulty cluster and activates the backup clusters based on their status and location. Here, N represents the number of working clusters, and X represents the number of backup clusters. The stack management unit, cluster computing unit, and supercapacitor acquisition unit constitute the hierarchical management system of this supercapacitor energy storage system.

[0027] This hierarchical management system adopts a three-tiered architecture of "management-computation-data acquisition," consisting of a management layer, a computation layer, and a data acquisition layer. Figure 1 As shown. Management Layer: The stack management unit is the core management unit of the hierarchical management system, responsible for global coordination, communication management, fault diagnosis, thermal management strategy formulation, and auxiliary control device integration. It communicates with the cluster computing unit layer via UDP protocol, supporting broadcast or multicast modes; it has a built-in web server and OTA upgrade module, supporting remote operation and maintenance. Computing Layer: The cluster computing unit manages all supercapacitor acquisition units within the cluster, responsible for calculating the maximum and minimum values ​​of the supercapacitor modules within the cluster (individual cell voltage, individual cell temperature, individual cell temperature rise), and performing cluster balancing control functions; it communicates with the stack management unit via UDP protocol (heartbeat packets + status reporting) and with the acquisition layer via CAN 2.0B protocol (periodic data synchronization). Acquisition Layer: The supercapacitor acquisition unit, also called the PACK acquisition unit (also... Figure 1 The supercapacitor acquisition unit (in the PACK) is directly connected to the supercapacitor PACK to collect data such as temperature and voltage of individual supercapacitor cells in real time, and sends the data to the cluster computing unit via the CAN bus.

[0028] The stack management unit (CMS) is also used to communicate with fire protection systems, environmental monitoring systems, and thermal management systems to achieve coordinated control of energy storage, fire protection, and thermal management. The CMS continuously communicates with these auxiliary control systems (typically using Modbus RTU communication in the field) to acquire environmental information in real time. Then, based on the collected temperature information of the supercapacitor, the CMS controls the thermal management (air conditioning) module to activate heating or cooling. For example, when the CMS detects that the supercapacitor temperature exceeds 90°C and the temperature rise is abnormal (out of control), it controls the fire protection control panel to activate alarms and sprinklers.

[0029] The supercapacitor energy storage system adopts an "N+X redundant cluster" configuration (N is the number of working clusters, and X is the number of standby clusters). An input-output expansion device is used to control the connection and disconnection between each supercapacitor cluster and the PCS. The stack management unit controls the input-output expansion device to achieve rapid isolation of faulty clusters and seamless switching to standby clusters. Its control process is as follows: Figure 3 As shown, the specific logic is as follows: Fault Detection: The stack management unit determines whether a cluster is prohibited from charging or discharging based on the maximum and minimum values ​​sent by the cluster calculation unit (referring to the calculation unit of the working cluster). These maximum and minimum values ​​include the maximum voltage (charging or discharging), the maximum temperature, and the maximum temperature rise. The maximum and minimum values ​​are compared with setpoints. If the voltage or temperature exceeds the setpoint range, charging and discharging are prohibited. If both charging and discharging are prohibited simultaneously, it is definitely a fault. Prohibition of discharging during the discharging phase is a fault, and prohibition of charging during the charging phase is a fault.

[0030] Switching command calculation: The stack management unit sends corresponding action commands to the input / output expansion device based on the status and location of the standby cluster and the location of the faulty cluster. Communication between the standby cluster and the stack management unit is continuous, only the power line is disconnected. The status of the standby cluster is obtained through communication (RJ45 UDP or TCP).

[0031] Fault cluster isolation and backup cluster activation: The input / output extension device executes commands to isolate the fault cluster and activate the backup cluster, thereby ensuring the stability of the high-voltage bus voltage.

[0032] Traditional energy storage systems lack cluster-level redundancy. This invention minimizes the impact of single-cluster failures and ensures continuous power supply through N+X redundancy and automatic switching.

[0033] The heap management unit addresses cluster computing units based on a dynamic allocation mechanism of "power-on sequence + protocol handshake," such as... Figure 2 As shown, after the heap management unit starts up, it broadcasts an address and IP call frame to the cluster computing unit. The cluster computing unit then sends its own IP address and address based on the received IP call frame. The heap management unit determines whether the address and IP of the cluster computing unit conform to the plan. If not, it powers down all cluster computing units and powers on the current cluster computing unit. The heap management unit sends an address precoding frame to the current cluster computing unit. After confirmation, the cluster computing unit sends an address change frame, and the cluster computing unit changes its IP address and address based on the address change frame. The specific process of the heap management unit allocating addresses to each cluster computing unit using a dynamic allocation mechanism based on power-on sequence and protocol handshake is as follows: The normal communication IPs designed are as follows: the heap management unit IP is 192.168.10.222, and the cluster computing unit IPs are 192.168.10.1 to 192.168.10.30 (30 clusters in total). The cluster address is the low byte of the IP.

[0034] Step 1: After the heap management unit starts up, it broadcasts an address and IP call frame to the cluster management unit; Step 2: After receiving the request, the cluster computing unit sends its own IP address, where the address refers to the low byte of the IP address; Step 3: Check if the address and IP of the heap management unit's computing cluster computing unit conform to the plan. If the address or IP does not conform to the plan, the heap management unit starts the encoding program (subsequent steps); otherwise, it enters the running program. There are two situations where the address does not conform to the plan: 1) The low byte of the IP address is not equal to the address; 2. The heap management unit receives a reply packet with a duplicate address / IP. Step 4: Power off all cluster computing units; Step 5: Power on the nth cluster computing unit (i.e., the cluster computing unit that does not conform to the plan); Step 6: The heap management unit sends a precoded frame with address x, and the nth cluster computing unit replies with the same frame; Step 7: The heap management unit sends a change frame for address x, the nth cluster calculation unit changes the IP and address, and replies that the encoding is complete.

[0035] Repeat steps 5, 6, and 7 until all cluster computation units are encoded (X from 1 to 30).

[0036] The power-on encoding process for clusters 1 to 30 is as follows: 1) The heap management unit broadcasts to restore all clusters to their initial IP, which is 235 (much larger than the actual number of clusters). 2) Power off all clusters and power on the first cluster (ensuring that only one cluster with the same IP address is online); 3) The first cluster is coded as 1; 4) Power on the second cluster, encode the second cluster as 2, and power on the subsequent clusters in sequence; 5) Complete all coding.

[0037] Meanwhile, to achieve intelligent thermal management, the stack management unit of this invention integrates a thermal management module. Through a multi-objective optimization algorithm based on "temperature-load-energy consumption," it dynamically adjusts the liquid cooling pump speed and air conditioning power to maximize energy efficiency. The supercapacitor acquisition unit uploads the collected supercapacitor surface temperature, liquid cooling pipe inlet / outlet temperature, and ambient temperature to the cluster computing unit. The cluster computing unit then uploads this information to the stack management unit. The stack management unit determines the cooling strategy based on grid dispatch instructions and / or the information uploaded by the cluster computing unit. The cooling strategy formulated by the stack management unit based on the information uploaded by the cluster computing unit includes: When the supercapacitor surface temperature is below a first set threshold and the temperature difference between the inlet and outlet of the liquid cooling pipeline is less than a second set threshold, the liquid cooling pump speed is reduced to enter energy-saving mode. When the supercapacitor surface temperature is not less than the first set threshold or the temperature difference between the inlet and outlet of the liquid cooling pipeline is greater than a third set threshold, the liquid cooling pump speed is increased to enter standard mode. When the supercapacitor surface temperature is greater than a fourth set threshold or the temperature difference between local modules is greater than a fifth set threshold, the liquid cooling pump speed is increased to its maximum, and the air conditioning cooling system is activated. The first set threshold is less than the fourth set threshold, the second set threshold is less than the third set threshold, the third set threshold is less than the fifth set threshold, and the fifth set threshold is less than the first set threshold. These set thresholds can be set in the stack management unit, and they are set based on the characteristics of the supercapacitor.

[0038] Data acquisition: The PACK acquisition unit uploads the surface temperature of the supercapacitor module, the inlet / outlet temperature of the liquid cooling pipeline, and the ambient temperature in real time; the stack management unit collects environmental parameters of the battery compartment through temperature and humidity sensors.

[0039] Load forecasting: The stack management unit combines grid dispatch instructions (such as the charging and discharging power plan for the next hour) and historical operating data (such as daily peak and valley periods) to predict the system load for the next 30 minutes.

[0040] In practical applications, a cooling strategy can be formulated based on the temperature of the supercapacitor cells and the temperature difference between the liquid-cooled inlet and outlet. Alternatively, the temperature of the supercapacitor cluster can be pre-adjusted to a suitable temperature based on the pre-charge / discharge command. The following explanation uses the formulation of a cooling strategy based on the temperature of the supercapacitor cells and the temperature difference between the liquid-cooled inlet and outlet as an example.

[0041] Cooling strategy calculation: When T_cell < 40℃ and ΔT < 5℃, reduce the liquid cooling pump speed to 30% (energy saving mode). When 40℃≤T_cell<50℃ or ΔT≥8℃, increase the pump speed to 70% (standard mode). When T_cell ≥ 50℃ or the temperature difference of a local module > 10℃, the pump speed increases to 100%, the air conditioner starts cooling (set temperature 25℃), and an alarm is triggered. When the ambient humidity is greater than 70%, the air conditioner will simultaneously turn on the dehumidification mode (power is reduced by 20% to avoid over-cooling).

[0042] Execution and Feedback: The stack management unit sends control commands to the liquid cooling pump controller and air conditioning PLC via the Modbus protocol, and updates the strategy every 5 minutes; at the same time, it monitors the cooling effect (such as whether the T_cell temperature drops), and if there is no improvement within 3 minutes, it triggers a level 2 alarm (notifying maintenance personnel).

[0043] Traditional thermal management relies on fixed threshold control and is not linked with energy storage systems. In complex environments, it can even lead to conflicting adjustment modes between air conditioning and liquid cooling units. This method avoids "overcooling" or "overheating" through load prediction and multi-parameter collaborative optimization, and is expected to reduce thermal management energy consumption by 15%-20%.

[0044] Where T_cell is the temperature of a single supercapacitor cell, and ΔT is the temperature difference between the liquid inlet and outlet of the liquid coolant.

[0045] In addition, the heap management unit supports remote firmware upgrades for the entire cluster computing unit and PACK acquisition unit, using a "differential upgrade + breakpoint resume" mechanism. The specific process is as follows: Step 1: The heap management unit receives the upgrade package pushed from the cloud (including differential files of the cluster computing unit firmware and the PACK acquisition unit firmware), verifies the hash value, and stores it in the local Flash. Step 2: The heap management unit sends an "upgrade notification frame" (containing upgrade version, time window, and file size) via UDP broadcast; the cluster computing unit receives the frame and replies with a "ready frame". Step 3: The heap management unit sends the cluster computing unit firmware differential file to the cluster computing unit via unicast according to the cluster computing unit address order (block transmission, 1KB per block, CRC check); after receiving the file, the cluster computing unit writes it to the backup partition and replies with a "cluster computing unit upgrade complete frame"; Step 4: The cluster computing unit broadcasts the PACK acquisition unit firmware differential file to the PACK acquisition unit via the CAN protocol. The PACK acquisition unit receives, verifies, and writes the file in blocks in the same manner. Step 5: After all devices have been upgraded, the stack management unit sends a "reboot command", and the cluster computing unit / PACK acquisition unit switches to the new firmware partition to run; if the upgrade fails (e.g., verification error), it automatically rolls back to the original partition and reports to the stack management unit.

[0046] Therefore, the supercapacitor energy storage system of the present invention has the following effects: High reliability: Cluster-level N+1 redundancy design. When the number of faulty clusters is less than the number of redundant clusters, the faulty clusters can be automatically isolated and the backup clusters can be put into operation to avoid the entire station downtime.

[0047] High efficiency: The intelligent thermal management strategy dynamically adjusts the cooling power, which is expected to reduce thermal management energy consumption by 15%-20% and extend the life of the supercapacitor.

[0048] Easy to maintain: Dynamic addressing supports plug and play, and OTA differential upgrades enable batch remote maintenance, reducing on-site operation time; Flexible expansion: UDP communication supports broadcast / multicast, and cluster computing unit addresses are dynamically allocated, facilitating system expansion.

[0049] Implementation of a hierarchical management system for supercapacitor energy storage systems The hierarchical management system of this invention adopts a three-level architecture including a stack management unit, a cluster computing unit, and a supercapacitor acquisition unit. Each supercapacitor is equipped with a corresponding supercapacitor acquisition unit for collecting the status data of the corresponding supercapacitor. The cluster computing unit is used to calculate the maximum and minimum values ​​based on the status data collected by the supercapacitor acquisition units in its supercapacitor cluster. The stack management unit is used to determine whether the supercapacitor cluster is faulty based on the supercapacitor cluster maximum and minimum values ​​sent by each cluster computing unit, and to control the isolation of the faulty cluster and the activation of the backup cluster based on the status and location of the backup cluster and the location of the faulty cluster. The supercapacitor energy storage system managed by this hierarchical management system includes N+X independent supercapacitor clusters cascaded through a bus, where N is the number of working clusters and X is the number of backup clusters.

[0050] Specifically, such as Figure 1 As shown, the three-tiered architecture consists of a management layer, a computing layer, and a data acquisition layer. The management layer comprises the stack management unit, the core management unit of the hierarchical management system, responsible for global coordination, communication management, fault diagnosis, thermal management strategy formulation, and auxiliary control device integration. It communicates with the cluster computing unit layer via UDP protocol, supporting broadcast or multicast modes; it has a built-in web server and OTA upgrade module, supporting remote operation and maintenance. The computing layer comprises the cluster computing units. Each cluster computing unit manages all supercapacitor acquisition units within the cluster, responsible for calculating the maximum and minimum values ​​of the supercapacitor modules within the cluster (individual cell voltage, individual cell temperature, individual cell temperature rise), and performing cluster balancing control functions; it communicates with the stack management unit via UDP protocol (heartbeat packets + status reporting) and with the data acquisition layer via CAN2.0B protocol (periodic data synchronization). The data acquisition layer comprises the supercapacitor acquisition units, also called PACK acquisition units (also... Figure 1 The supercapacitor acquisition unit (in the PACK) is directly connected to the supercapacitor PACK to collect data such as temperature and voltage of individual supercapacitor cells in real time, and sends the data to the cluster computing unit via the CAN bus.

[0051] The stack management unit (SMU) achieves rapid isolation of faulty clusters and seamless switching to standby clusters by controlling the input / output expansion devices. The SMU determines whether a supercapacitor cluster is faulty based on the maximum / minimum values ​​of the supercapacitor clusters sent by each cluster computing unit, and controls the isolation of faulty clusters and the activation of standby clusters based on the status and location of the standby clusters and the location of the faulty clusters. Its control flow is as follows: Figure 3 As shown, the specific logic is as follows: Fault detection: The stack management unit determines whether a cluster is prohibited from charging or discharging based on the maximum and minimum values ​​sent by the cluster calculation unit (referring to the calculation unit of the working cluster). The maximum and minimum values ​​include the maximum voltage (charging or discharging), the maximum temperature, and the maximum temperature rise.

[0052] Switching command calculation: The stack management unit sends corresponding action commands to the input / output expansion device based on the status and location of the standby cluster and the location of the faulty cluster.

[0053] Fault cluster isolation and backup cluster activation: The input / output extension device executes commands to isolate the fault cluster and activate the backup cluster, thereby ensuring the stability of the high-voltage bus voltage.

[0054] Traditional energy storage systems lack cluster-level redundancy. This invention minimizes the impact of single-cluster failures and ensures continuous power supply through N+X redundancy and automatic switching.

[0055] The heap management unit addresses cluster computing units based on a dynamic allocation mechanism of "power-on sequence + protocol handshake," such as... Figure 2 As shown, after the heap management unit starts up, it broadcasts an address and IP call frame to the cluster computing unit. The cluster computing unit sends its own IP and address according to the received IP call frame. The heap management unit determines whether the address and IP of the cluster computing unit conform to the plan. If they do not conform, it shuts down all cluster computing units and powers on the cluster computing unit. The heap management unit sends an address precoding frame to the cluster computing unit. After the cluster computing unit confirms, the heap management unit sends an address change frame. The cluster computing unit changes its IP and address according to the address change frame.

[0056] Meanwhile, to achieve intelligent thermal management, the stack management unit of this invention integrates a thermal management module. Through a multi-objective optimization algorithm based on "temperature-load-energy consumption," it dynamically adjusts the liquid cooling pump speed and air conditioning power to maximize energy efficiency. The supercapacitor acquisition unit uploads the collected supercapacitor surface temperature, liquid cooling pipe inlet / outlet temperature, and ambient temperature to the cluster computing unit. The cluster computing unit then uploads this information to the stack management unit. The stack management unit predicts the load for a future period based on grid dispatch instructions and historical operating data, and determines the cooling strategy based on the information uploaded by the cluster computing unit. The cooling strategy formulated by the stack management unit includes: When the surface temperature of the supercapacitor is less than the first set threshold and the temperature difference between the inlet and outlet of the liquid cooling pipe is less than the second set threshold, the speed of the liquid cooling pump is reduced to put it into energy-saving mode; when the surface temperature of the supercapacitor is not less than the first set threshold or the temperature difference between the inlet and outlet of the liquid cooling pipe is greater than the third set threshold, the speed of the liquid cooling pump is increased to put it into standard mode; when the surface temperature of the supercapacitor is greater than the fourth set threshold or the temperature difference of the local modules is greater than the fifth set threshold, the speed of the liquid cooling pump is increased to the maximum, and the air conditioner is started to cool; wherein the first set threshold is less than the fourth set threshold, the second set threshold is less than the third set threshold, the third set threshold is less than the fifth set threshold, and the fifth set threshold is less than the first set threshold.

[0057] In addition, the heap management unit supports remote firmware upgrades for the entire cluster computing unit and PACK acquisition unit, using a "differential upgrade + breakpoint resume" mechanism. The specific process is as follows: Step 1: The heap management unit receives the upgrade package pushed from the cloud (including differential files of the cluster computing unit firmware and the PACK acquisition unit firmware), verifies the hash value, and stores it in the local Flash. Step 2: The heap management unit sends an "upgrade notification frame" (containing upgrade version, time window, and file size) via UDP broadcast; the cluster computing unit receives the frame and replies with a "ready frame". Step 3: The heap management unit sends the cluster computing unit firmware differential file to the cluster computing unit via unicast according to the cluster computing unit address order (block transmission, 1KB per block, CRC check); after receiving the file, the cluster computing unit writes it to the backup partition and replies with a "cluster computing unit upgrade complete frame"; Step 4: The cluster computing unit broadcasts the PACK acquisition unit firmware differential file to the PACK acquisition unit via the CAN protocol. The PACK acquisition unit receives, verifies, and writes the file in blocks in the same manner. Step 5: After all devices have been upgraded, the stack management unit sends a "reboot command", and the cluster computing unit / PACK acquisition unit switches to the new firmware partition to run; if the upgrade fails (e.g., verification error), it automatically rolls back to the original partition and reports to the stack management unit.

Claims

1. A hierarchical management system for a supercapacitor energy storage system, characterized in that, The hierarchical management system adopts a three-tier architecture including a stack management unit, a cluster computing unit, and a supercapacitor acquisition unit. Each supercapacitor is equipped with a corresponding supercapacitor acquisition unit to collect the status data of the corresponding supercapacitor. The cluster computing unit is used to calculate the maximum and minimum values ​​based on the status data collected by the supercapacitor acquisition units in its supercapacitor cluster. The stack management unit is used to determine whether the supercapacitor cluster is faulty based on the supercapacitor cluster maximum and minimum values ​​sent by each cluster computing unit, and to control the isolation of the faulty cluster and the activation of the backup cluster based on the status and location of the backup cluster and the location of the faulty cluster. The supercapacitor energy storage system managed by the hierarchical management system includes N+X independent supercapacitor clusters cascaded through a bus, where N is the number of working clusters and X is the number of backup clusters.

2. The hierarchical management system for the supercapacitor energy storage system according to claim 1, characterized in that, The heap management unit uses a dynamic allocation mechanism based on power-on sequence and protocol handshake to allocate addresses to each cluster computing unit. The allocation process is as follows: After the heap management unit starts up, it broadcasts an address and IP call frame to the cluster computing unit. The cluster computing unit sends its own IP and address according to the received IP call frame. The heap management unit determines whether the address and IP of the cluster computing unit conform to the plan. If they do not conform, it powers down all cluster computing units and powers up the cluster computing unit. The heap management unit sends an address precoding frame to the cluster computing unit. After the cluster computing unit confirms, the heap management unit sends an address change frame. The cluster computing unit changes its IP and address according to the address change frame.

3. The hierarchical management system for the supercapacitor energy storage system according to claim 1, characterized in that, The stack management unit is also used to remotely upgrade the firmware of all cluster computing units and supercapacitor acquisition units. The upgrade adopts a differential upgrade+ and breakpoint resume upgrade mechanism.

4. The hierarchical management system for the supercapacitor energy storage system according to claim 1, characterized in that, The stack management unit is also used to communicate with the fire protection system, environmental monitoring system and thermal management system to realize the linkage control of energy storage-fire protection-thermal management.

5. The hierarchical management system for the supercapacitor energy storage system according to claim 1, characterized in that, The stack management unit integrates a thermal management module. The supercapacitor acquisition unit is used to upload the acquired supercapacitor surface temperature, liquid cooling pipeline inlet / outlet temperature, and ambient temperature to the cluster computing unit. The cluster computing unit uploads the above information to the stack management unit. The stack management unit is used to determine the cooling strategy according to the grid dispatch instructions and / or the information uploaded by the cluster computing unit.

6. The hierarchical management system for the supercapacitor energy storage system according to claim 5, characterized in that, The cooling strategy formulated by the stack management unit includes: When the surface temperature of the supercapacitor is less than the first set threshold and the temperature difference between the inlet and outlet of the liquid cooling pipeline is less than the second set threshold, the speed of the liquid cooling pump is reduced to put it into energy-saving mode. When the surface temperature of the supercapacitor is not less than the first set threshold or the temperature difference between the inlet and outlet of the liquid cooling pipeline is greater than the third set threshold, the speed of the liquid cooling pump is controlled to increase so that it is in standard mode. When the surface temperature of the supercapacitor exceeds the fourth set threshold or the temperature difference of a local module exceeds the fifth set threshold, the liquid cooling pump speed is controlled to reach the maximum, and the air conditioner is started to cool. The first set threshold is less than the fourth set threshold, the second set threshold is less than the third set threshold, the third set threshold is less than the fifth set threshold, and the fifth set threshold is less than the first set threshold.

7. The hierarchical management system for the supercapacitor energy storage system according to claim 6, characterized in that, When controlling the liquid cooling pump and / or air conditioning, the stack management unit updates the strategy at set intervals and judges the cooling effect based on the surface temperature of the supercapacitor. If the requirement is not met within a certain period of time, an alarm is triggered.

8. The hierarchical management system for the supercapacitor energy storage system according to claim 1, characterized in that, The stack management unit controls the isolation of faulty clusters and the activation of backup clusters by sending corresponding action commands to the input-output expansion device, which is used to control the connection and disconnection between each supercapacitor cluster and the PCS.

9. The hierarchical management system for the supercapacitor energy storage system according to claim 1, characterized in that, The stack management unit communicates with the cluster computing unit via the UDP protocol, and the cluster computing unit communicates with the supercapacitor acquisition unit via the CAN protocol.

10. A supercapacitor energy storage system, characterized in that, The system includes N+X independent supercapacitor clusters cascaded via a bus and a hierarchical management system, wherein the hierarchical management system adopts the hierarchical management system of the supercapacitor energy storage system as described in any one of claims 1-9.

Images