Resilience improvement and autonomous fault recovery method and system for distributed optical storage and direct flexible system

Abstract

The application discloses a kind of distributed light storage direct flexible system's toughness promotion and autonomous fault recovery method and system.Method includes: in each node configuration edge fault detection unit, independent acquisition voltage, current, temperature and communication state signal, by pole mode transformation obtains line mode voltage and ground mode voltage and carries out distributed fault perception;Fault is divided into node internal fault, regional fault and system level fault three levels, and hierarchical autonomous isolation is executed;When three-level fault, switch to island operation mode, and execute black start recovery program;After external power grid recovers, executes grid connection smooth switching.System includes edge fault detection unit, isolation execution unit, regional coordination controller, system level mode switching unit and black start recovery unit.The application realizes distributed fault autonomous perception and hierarchical isolation, without central controller, fault isolation speed is fast, with complete black start recovery ability, can effectively improve the power supply toughness and reliability of light storage direct flexible system.

Description

Technical Field

[0001] This invention relates to the field of distributed energy system resilience enhancement and fault recovery technology, specifically to a method and system for enhancing the resilience and autonomous fault recovery of distributed photovoltaic-storage-DC-flexible systems. Background Technology

[0002] Distributed photovoltaic-storage-DC-flexible systems (PV-SHS-DC-flex systems) deeply integrate photovoltaic power generation, energy storage technology, DC power distribution, and flexible control technology, representing a crucial technological path for achieving efficient utilization of clean energy and flexible grid interaction. However, with the continuous expansion of the scale of distributed PV-SHS-DC-flex systems and the increasing complexity of application scenarios, their operational safety and power supply reliability face severe challenges.

[0003] First, the photovoltaic-storage-DC-flexible system uses DC power distribution. Because DC systems lack natural zero-crossing characteristics, short-circuit fault currents rise rapidly and reach high peak values, making it difficult for circuit breakers to extinguish arcs using natural zero-crossing as in AC systems. DC systems also have fast response times; voltage dips and current surges occur within milliseconds, requiring fault detection and protection actions to have sub-millisecond recognition capabilities. Traditional mechanical circuit breakers typically have response times ranging from several milliseconds to tens of milliseconds, which is insufficient to meet the rapid protection requirements of DC systems.

[0004] Secondly, existing photovoltaic-storage-direct-drive-flexible systems mostly employ a centralized architecture for fault handling, relying on a central controller for global fault diagnosis and decision-making. This architecture carries the risk of a single point of failure; if the central controller fails, the entire system's fault handling capability will be completely lost. Furthermore, the centralized architecture requires transmitting monitoring data from all nodes to the central controller, resulting in high communication latency and bandwidth consumption, making it difficult to meet real-time requirements in scenarios with a large number of distributed nodes.

[0005] Third, existing systems lack system-level black start and autonomous recovery capabilities. After a power outage caused by an external power grid failure, most systems rely on manual on-site operations for recovery, which often takes tens of minutes or even hours from the occurrence of the fault to the restoration of power. Although some systems have certain protection functions, they lack a tiered fault handling mechanism and a complete black start recovery process, making it impossible for them to autonomously resume operation and ensure priority power supply to critical loads after an external power grid outage.

[0006] Fourth, existing fault detection methods suffer from insufficient adaptability in photovoltaic-storage-DC-flexible systems. Fault characteristics of adjacent lines in a DC distribution system are often similar, making it difficult to distinguish between faults at the end of the protected line and faults outside the protected area. Traditional threshold methods suffer from low reliability due to the limited voltage drop amplitude and inconsistent slope of DC system voltage changes. While some existing technologies offer fault identification methods for photovoltaic-storage-DC-flexible systems, they primarily focus on fault detection and location, lacking a complete closed-loop processing capability encompassing fault perception, graded isolation, and black-start recovery.

[0007] To address the aforementioned issues, this invention proposes a method and system for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-flexible system. Summary of the Invention

[0008] The purpose of this invention is to provide a method and system for improving the resilience and autonomous fault recovery of a distributed optical-storage-direct-flexible system, so as to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a method for improving the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive flexible system, comprising the following steps: S1. Distributed Fault Detection: Edge fault detection units are configured at each generation node, energy storage node, and load node of the photovoltaic-storage-DC-flexible system. Each edge fault detection unit independently collects the DC bus voltage signal, current signal, temperature signal, and communication status signal of its own node; the collected voltage signals are subjected to pole mode transformation to obtain line mode voltage component and ground mode voltage component; when the line mode voltage component or the ground mode voltage component meets the preset protection start conditions, it is determined that a fault has occurred in the node, and a fault trigger signal is output. S2. Fault Type Classification and Identification: The edge fault detection unit of each node independently performs fault type classification and identification based on local detection information, dividing faults into three levels: Level 1 faults are internal node faults; Level 2 faults are regional faults; Level 3 faults are system-level faults. S3, Hierarchical Self-Isolation: For a Level 1 fault, the edge fault detection unit of this node directly sends an isolation command to the isolation execution unit of this node to disconnect the faulty node from the DC bus, while the remaining nodes continue to operate normally. For Level 2 faults, the edge fault detection units of each node in this region report fault information to the regional coordination controller through the regional communication bus. After the regional coordination controller confirms the fault range and severity by integrating the fault information of multiple nodes in this region, it issues isolation instructions to the isolation execution units of all nodes in the fault area to isolate the entire fault area. Non-faulty areas maintain operation through topology reconstruction. For Level 3 faults, each regional coordination controller reports system-level fault information to the system-level mode switching unit through the system communication bus. The system-level mode switching unit disconnects the grid-connected / islanded mode switching switch and switches the system to islanded operation mode. S4. Resilient Reconstruction and Black Boot Recovery Steps: After the system switches to isolated operation mode, execute the black boot recovery procedure: S41, Energy Storage Priority Access: The black start recovery unit prioritizes the closing of the isolation execution unit of the energy storage node, so that the energy storage unit is connected to the DC bus and the DC bus voltage is established through the DC / DC converter; S42. Gradual Photovoltaic Connection: After the DC bus voltage has stabilized, photovoltaic power generation nodes are connected sequentially. S43. Tiered load activation: Based on the power supply priority of each load node, loads at each level are activated sequentially in descending order of priority to gradually restore the system. S5. Smooth grid connection switching: The system-level mode switching unit continuously monitors the status of the external power grid. When the external power grid is detected to have returned to normal, the smooth grid connection switching procedure is executed to switch the system from islanded operation mode back to grid-connected mode.

[0010] Preferably, in step S1, the specific method for performing pole-mode conversion on the acquired voltage signal is as follows: ; in, This is the voltage between the positive busbar and ground. This is the voltage between the negative busbar and ground. Line-mode voltage, This is the ground mode voltage.

[0011] Preferably, in step S2, the fault type classification identification specifically includes: Within a preset time window T, the line-mode voltage and ground-mode voltage are integrated to obtain the integral value of the line-mode voltage. and ground mode voltage integral value ; Based on the integral value of the ground mode voltage With fault selection threshold The comparison results determine the fault type: when and An abnormal drop indicates a positive ground fault. when and An abnormal drop indicates a negative grounding fault. when and and When both are abnormal, it indicates a grounding fault in both the positive and negative terminals; Based on the integral value of the line mode voltage With region identification threshold The comparison results determine the faulty section: when When the fault is located within the current node area, it is determined that the fault is located within this node area. when At that time, the fault is determined to be located outside the local node area.

[0012] Preferably, in step S3, the specific method by which the regional coordination controller confirms the fault range by integrating fault information from multiple nodes within the region is as follows: The regional coordination controller receives fault trigger signals and fault type classification results reported by each node within its region, and constructs a regional fault information matrix. ,in This represents the total number of nodes in this region. For the first The fault status identifier of each node: when at least two adjacent nodes in the area simultaneously report a level 2 fault and the fault characteristics are related, the area coordination controller determines it as an area fault and issues an isolation command to the isolation execution unit of all nodes in the fault area.

[0013] Preferably, in step S41, the specific method for preferentially connecting energy storage to establish the DC bus voltage is as follows: The black start recovery unit first queries the state of charge (SOC) of each energy storage node through the system communication bus and selects the energy storage node with the highest SOC as the black start power source. The isolation actuator of the selected energy storage node is closed, and the energy storage node injects voltage into the DC bus in voltage control mode through the DC / DC converter. The voltage reference value is... Gradually increase according to the ramp function: Where: initial voltage Rated voltage time constant , For time; When the DC bus voltage reaches more than 95% of the rated voltage and remains stable for a certain period of time... After that, it is determined that the DC bus voltage has been established.

[0014] Preferably, the specific method for tiered load allocation in step S43 is as follows: The load nodes are pre-classified into three priority levels: Level 1 loads are critical loads such as fire-fighting equipment, communication equipment, and emergency lighting, with a time delay before activation. shortest; Secondary loads are general loads such as ordinary lighting and air conditioning, with a delayed start time. Next; Level 3 loads are interruptible loads such as ordinary sockets and non-critical equipment, with a delayed start-up time. longest.

[0015] Preferably, in step S5, the grid-connected smooth handover procedure specifically includes: The system-level mode switching unit detects the voltage amplitude, frequency, and phase of the external power grid in real time through the grid connection detection module; When the external power grid voltage amplitude, frequency, and phase are detected to have returned to the normal range and remained stable for a certain period of time. Then, pre-synchronization control is performed: the DC bus voltage of the photovoltaic-storage-DC-flexible system is adjusted to track the voltage amplitude, frequency and phase of the external power grid through the grid-connected inverter; When the voltage amplitude difference, frequency difference, and phase difference between the system side and the grid side are all less than the preset synchronization tolerance threshold, the grid-connected / islanded mode switching switch is closed to complete the grid-connected switching.

[0016] A system for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive-flexible system includes: Multiple edge fault detection units are respectively installed at each generation node, energy storage node, and load node of the photovoltaic-storage-DC-flexible system. Each edge fault detection unit includes a voltage sensor, a current sensor, a temperature sensor, a communication status monitoring module, and a local edge processor. The voltage sensor is connected in parallel between the positive and negative terminals of the node's DC bus to collect the node's DC bus voltage in real time; the current sensor is connected in series on the node's DC bus to collect the node's current in real time; the temperature sensor is installed on the surface of the node's power devices to monitor the equipment temperature; the communication status monitoring module is connected to the node's communication interface to monitor the communication status between the node and the control network. The outputs of the voltage sensor, current sensor, temperature sensor, and communication status monitoring module are all connected to the local edge processor. The local edge processor has a built-in pole mode conversion module, a fault type classification identification module, and a local isolation decision module, which are used to perform pole mode conversion in step S1, fault type classification identification in step S2, and local isolation decision for first-level faults in step S3, respectively.

[0017] Multiple isolation execution units are respectively located at each node and connected to the corresponding edge fault detection unit. Each isolation execution unit includes a solid-state DC circuit breaker connected in series on the DC bus of the node and a bypass switch connected in parallel with the solid-state DC circuit breaker. The solid-state DC circuit breaker is a silicon carbide MOSFET or IGCT-based solid-state DC circuit breaker with bidirectional current interruption capability and an interruption time of no more than 100 microseconds. The bypass switch is a mechanical bypass switch, which can be closed manually or remotely after the solid-state DC circuit breaker trips due to a fault, for bypass operation during node access restoration after fault clearance or equipment maintenance.

[0018] Multiple regional coordination controllers are located in each preset region. Each edge fault detection unit within the same region is connected to its corresponding regional coordination controller via a regional communication bus. The regional communication bus is redundant, employing either twisted-pair or fiber optic communication to ensure communication reliability.

[0019] The system-level mode switching unit connects to the regional coordination controllers via the system communication bus. This system communication bus is also redundant. The system-level mode switching unit includes a grid-connected / islanded mode switch, a DC bus voltage establishment module, and a grid-connected detection module. The grid-connected / islanded mode switch is located at the connection point between the photovoltaic-storage-DC-flexible system and the external power grid. The DC bus voltage establishment module includes a DC / DC converter and an energy storage battery interface connected to the DC / DC converter; the output of the DC / DC converter is connected to the DC bus of the photovoltaic-storage-DC-flexible system. The grid-connected detection module is connected to the external power grid to monitor its voltage, frequency, and phase in real time.

[0020] The black-start recovery unit is connected to each isolated execution unit and each area coordination controller. The black-start recovery unit includes an energy storage priority access control module and a tiered load input control module. The energy storage priority access control module is used to prioritize the connection of energy storage nodes to the DC bus to establish voltage after the system switches to islanded mode.

[0021] Preferably, the DC bus voltage establishment module is a DC / DC conversion unit inside the energy storage converter. Its input end is connected to the energy storage battery, and its output end is connected to the DC bus of the photovoltaic-energy storage DC-flexible system. It is used to convert the DC power from the energy storage battery into electrical energy that meets the DC bus voltage level requirements in islanded mode.

[0022] Compared with the prior art, the beneficial effects of the present invention are: This invention features independently configured edge fault detection units at each node, enabling all fault detection and preliminary judgment to be completed locally without relying on a central controller, fundamentally eliminating the single-point-of-failure risk of centralized architectures. It employs a pole-mode transformation method to analyze the DC bus voltage, using the integral values ​​of line-mode and ground-mode voltages for fault type identification and segment location, achieving fault detection and classification within milliseconds. Differentiated isolation strategies are adopted based on fault levels: Level 1 faults are isolated locally autonomously, Level 2 faults are isolated collaboratively in a region, and Level 3 faults involve system-level mode switching, achieving precise fault location and rapid isolation, effectively limiting the scope of fault propagation. A complete black-start recovery process is established, from priority voltage construction for energy storage and gradual photovoltaic integration to tiered load input, enabling the system to autonomously recover operation after external grid failures, significantly shortening outage time. Pre-synchronization control achieves a smooth switch from islanded mode to grid-connected mode, avoiding inrush current and voltage fluctuations during the switching process. Attached Figure Description Figure 1 This is an overall flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the overall structure of the system of the present invention. Figure 3 This is a block diagram of the edge fault detection unit; Figure 4 The circuit schematic for the isolated execution unit; Figure 5 Flowchart for fault type classification and identification; Figure 6 This is a sequence diagram for black boot recovery. Detailed Implementation

[0023] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] like Figures 1-6 As shown, this embodiment provides a resilience enhancement and autonomous fault recovery system for a distributed photovoltaic-storage-DC-flexible system, applicable to a photovoltaic-storage-DC-flexible system comprising M photovoltaic power generation nodes, N energy storage nodes, and L load nodes, where M, N, and L are all positive integers.

[0025] Each node is connected in parallel via a DC bus to form a common DC bus architecture. The DC bus voltage level is selected as 750V or 375V depending on the system size.

[0026] Each node is configured with an edge fault detection unit 100 and an isolation execution unit 200. Multiple nodes are divided into several regions according to geographical location, and each region is equipped with a region coordination controller 300.

[0027] The method for dividing the region is as follows: (a) Preliminary spatial grouping Based on the actual physical spatial distribution of each node in the photovoltaic-storage DC-flexible system, nodes with adjacent spatial locations are divided into the same candidate area, using the same building, the same factory area, or the same distribution substation as the basic unit. The scope of the candidate area is limited to a power supply radius of no more than 500m for the DC interconnection lines between nodes, to ensure that the voltage drop and line loss of the DC lines are within a reasonable range.

[0028] (II) Calculation of Electrical Characteristic Matching Degree Based on the initial spatial grouping, an electrical characteristic matching degree analysis was performed on the nodes within each candidate region. Historical operating data for each node was collected for 30 consecutive days, including photovoltaic power generation curves, load power consumption curves, and energy storage charging and discharging power curves. After normalizing the collected data, the power generation-power consumption complementarity index between any two nodes within the same candidate region was calculated. in, Let the complementarity between node i and node j be . Let be the photovoltaic power generation of node i at time t. Let be the load power of node j at time t. The closer the value is to 1, the higher the matching degree of power generation and power consumption time between the two nodes, and the better the complementarity.

[0029] when When node i and node j are identified as having complementary electrical characteristics, they are preferentially assigned to the same region. The number of nodes in each region shall not exceed 10, and the total installed capacity in each region shall not exceed 2MW, to ensure that the computing power and communication bandwidth of the regional coordination controller 300 meet the real-time control requirements. Meanwhile, the communication distance between nodes within the same region shall not exceed 200m, or the total communication distance after cascading via the regional communication bus shall not exceed 1000m, to ensure that the communication delay of the regional communication bus is no greater than 5ms.

[0030] (III) Dynamic Adjustment The regional division is not fixed. The system recalculates the complementarity index every quarter based on the actual operating data of each node. When it is found that the overall photovoltaic absorption rate of a certain region is lower than 70% for three consecutive months, the regional dynamic adjustment is triggered. The system-level mode switching unit 400 re-executes the above division process and sends the new regional division results to the regional coordination controllers 300.

[0031] The system is configured with a system-level mode switching unit 400 and a black boot recovery unit 500.

[0032] Each edge fault detection unit 100 is installed at each generation node, energy storage node, and load node of the photovoltaic-storage-DC-flexible system. Each edge fault detection unit 100 includes a voltage sensor 101, a current sensor 102, a temperature sensor 103, a communication status monitoring module 104, and a local edge processor 105. The voltage sensor 101 is connected in parallel between the positive and negative terminals of the node's DC bus to collect the node's DC bus voltage in real time; the current sensor 102 is connected in series on the node's DC bus to collect the node's current in real time; the temperature sensor 103 is installed on the surface of the node's power devices to monitor the equipment temperature; the communication status monitoring module 104 is connected to the node's communication interface to monitor the communication status between the node and the control network. The outputs of the voltage sensor 101, current sensor 102, temperature sensor 103, and communication status monitoring module 104 are all connected to the local edge processor 105.

[0033] The local edge processor 105 of the edge fault detection unit 100 adopts an embedded processor with an ARM Cortex-M7 or RISC-V architecture, with a main frequency of not less than 400MHz, and has a floating-point arithmetic unit and digital signal processing instruction set to meet the real-time requirements of polar mode transformation and integral calculation. The local edge processor 105 has a built-in polar mode transformation module 1051, a fault type classification identification module 1052, and a local isolation decision module 1053, which are used to perform polar mode transformation in step S1, fault type classification identification in step S2, and local isolation decision of first-level faults in step S3, respectively.

[0034] Multiple isolation execution units 200 are respectively installed at each node and connected to the corresponding edge fault detection unit 100. Each isolation execution unit 200 includes a solid-state DC circuit breaker 201 connected in series on the DC bus of the node and a bypass switch 202 connected in parallel with the solid-state DC circuit breaker 201. The solid-state DC circuit breaker 201 is a solid-state DC circuit breaker based on silicon carbide MOSFET or IGCT, with bidirectional current interruption capability and an interruption time of no more than 100 microseconds. The bypass switch 202 is a mechanical bypass switch, which can be closed manually or remotely after the solid-state DC circuit breaker trips due to a fault, for bypass operation during node access restoration after fault clearance or equipment maintenance.

[0035] The solid-state DC circuit breaker 201 of the isolation execution unit 200 uses silicon carbide MOSFETs as switching devices and has the following technical parameters: rated voltage 1200V, rated current configured according to node capacity, ranging from 100A to 1000A; turn-off time not greater than 100 microseconds; bidirectional current turn-off capability: supports the interruption of forward and reverse fault currents; the control signal adopts fiber optic isolation drive, and has electrical isolation and electromagnetic interference resistance capabilities.

[0036] Each edge fault detection unit 100 within the same area is connected to its corresponding regional coordination controller 300 via a regional communication bus. In this embodiment, the regional coordination controller 300 is an embedded industrial controller. Its core processor adopts the ARM Cortex-A72 architecture, with a main frequency of not less than 1.5GHz, and has at least two Ethernet communication interfaces and four RS-485 communication interfaces, with a built-in Linux real-time operating system. The regional coordination controller 300 can be a commercially available microgrid coordination controller product, such as the Acrel ACCU-100 or CET-7320-G microgrid controller. The specific model of the regional coordination controller 300 is not limited; any embedded industrial controller with data acquisition, logic judgment, and command issuance functions can implement the functions of the regional coordination controller 300 in this embodiment.

[0037] The system-level mode switching unit 400 is connected to the regional coordination controllers 300 via the system communication bus. The system-level mode switching unit 400 includes a grid-connected / islanded mode switching switch 402, a DC bus voltage establishment module 403, and a grid-connected detection module 404.

[0038] The system-level mode switching unit 400 is implemented collaboratively by the energy storage converter (PCS) and the energy management system (EMS). The grid-connected / islanded mode switching switch 402 is an AC-side solid-state switching switch within the energy storage converter, employing a thyristor anti-parallel configuration. It features bidirectional conduction and fast turn-off capabilities, with a switching time of no more than 10ms. One side of this switch is connected to the external grid AC bus, and the other side is connected to the AC input terminal of the energy storage converter.

[0039] The grid connection detection module 404 is a grid status monitoring module in the energy management system. It collects the voltage amplitude, frequency, and phase information of the external power grid in real time through voltage transformers and frequency detection circuits. Based on the detection results of the grid connection detection module 404, the energy management system sends a closing or opening control command to the grid-connected / islanded mode switching switch 402 via fiber optic communication. The DC bus voltage establishment module 403 is a DC / DC conversion unit inside the energy storage converter. Its input is connected to the energy storage battery, and its output is connected to the DC bus of the photovoltaic-storage-DC-flexible system. It is used to convert the DC power from the energy storage battery into electrical energy that meets the DC bus voltage level requirements in islanded mode. The energy storage converter can be a grid-type energy storage converter manufactured by Sungrow Power Supply Co., Ltd., and the energy management system can be the Sungrow EMS3000 energy management system.

[0040] The black-start recovery unit 500 is connected to each isolation execution unit 200 and each area coordination controller 300. The black-start recovery unit 500 includes an energy storage priority access control module 501 and a tiered load input control module 502. The black-start recovery unit 500 is a functional module built into the energy management system (EMS). This module interacts with the isolation execution units 200 of each energy storage node and the tiered load input control modules 502 of each load node via the IEC 61850 or Modbus communication protocol.

[0041] The implementation of the energy storage priority access control module 501 is as follows: After the system switches to islanded operation mode, the energy management system sends a SOC query command to each energy storage node through the system communication bus. After receiving the state of charge information returned by each energy storage node, the energy storage node with the highest SOC is selected as the black start power source through a sorting algorithm. Then, a closing command is sent to the isolation execution unit 200 of the energy storage node through the communication bus, and a voltage establishment command is sent to the DC / DC converter of the energy storage node at the same time.

[0042] The sorting algorithm can employ selection sort: iterate through the SOC values ​​of all energy storage nodes, recording the current maximum value and its corresponding node number. After the iteration is complete, the energy storage node with the highest SOC is obtained. When the number of energy storage nodes is small (usually no more than 10), the computational complexity of this method is sufficient to meet real-time requirements.

[0043] In this embodiment, the graded load input control module (502) includes multiple time-delay input switches, each of which is connected in series on load branches of different priority levels. Its specific structure is as follows: Each of the aforementioned time-delay switches consists of load switches (K1, K2, K3) and time-delay control units (TD1, TD2, TD3). The load switches (K1, K2, K3) are connected in series on the DC bus of the primary, secondary, and tertiary load branches, respectively, to connect or disconnect the electrical connection between the corresponding load branch and the DC bus. The input terminals of the time-delay control units (TD1, TD2, TD3) are connected to the control signal output terminals of the black-start recovery unit 500, and the output terminals of the time-delay control units (TD1, TD2, TD3) are connected to the control terminals of the corresponding load switches (K1, K2, K3).

[0044] The load switches (K1, K2, K3) are DC solid-state relays or DC contactors, and have the capability of rated voltage not lower than the rated voltage of the DC bus and rated current not lower than the maximum operating current of the corresponding load branch.

[0045] Each of the delay control units (TD1, TD2, TD3) has a built-in RC delay circuit. Each RC delay circuit consists of a resistor and a capacitor connected in series, and its time constant is determined by the product of the resistance and capacitance values. The time constants of the RC delay circuits of each delay control unit are different. Specifically, the delay control unit (TD1) corresponding to the primary load branch has the smallest time constant, followed by the delay control unit (TD2) corresponding to the secondary load branch, and the delay control unit (TD3) corresponding to the tertiary load branch has the largest time constant. As an alternative, the delay control unit can also use a digital timer chip (such as the NE555 timer IC) to implement the delay function, setting different delay times by selecting the values ​​of external resistors and capacitors.

[0046] The black-start recovery unit 500 simultaneously sends a unified activation command to the delay control units (TD1, TD2, TD3) of all delay activation switches via the system communication bus. Upon receiving the activation command, each delay control unit activates its respective RC delay circuit or digital timer for timing. Because the preset delay times of each delay control unit are different, the delay control unit (TD1) corresponding to the primary load branch first drives the load switch (K1) of the primary load branch to close after the shortest delay time (e.g., 1 second), activating the primary load. The delay control unit (TD2) corresponding to the secondary load branch drives the load switch (K2) of the secondary load branch to close after the second longest delay time (e.g., 5 seconds), activating the secondary load. The delay control unit (TD3) corresponding to the tertiary load branch drives the load switch (K3) of the tertiary load branch to close after the longest delay time (e.g., 15 seconds), activating the tertiary load.

[0047] With the aforementioned stepped delay closing structure, each delay switch does not need to rely on the communication network to send time-sharing control commands one by one. Instead, it can realize the graded delay connection of the load through its own independent delay control unit. This effectively avoids voltage drop and inrush current caused by all loads being connected to the DC bus at the same time, while improving the redundancy and reliability of the control system.

[0048] The tiered load activation control module 502 operates as follows: After the DC bus voltage is established, the energy management system reads the load tier configuration table pre-stored in the local database. This configuration table records the priority level (Level 1 / Level 2 / Level 3) and corresponding delay activation time for each load node. The energy management system sends a closing command to the corresponding load switch when the delay activation time for each load node arrives, according to the priority from highest to lowest, thus achieving tiered load activation. Specifically, the industrial server of the energy management system can be an Advantech IPC-610L industrial computer or an industrial control computer of equivalent performance.

[0049] The specific process of distributed fault perception and hierarchical identification is as follows: Each edge fault detection unit 100 uses a sampling frequency (Sampling period 100 microseconds) Continuously collect the DC bus positive terminal to ground voltage of this node. and negative terminal to ground voltage .

[0050] The pole-mode conversion module 1051 of the local edge processor 105 calculates the line-mode voltage in real time. and ground mode voltage : in: This is the voltage between the positive busbar and ground. Voltage of negative busbar to ground The preset action threshold is determined based on the voltage fluctuation range during normal system operation. For a 750V DC bus system, the line-mode voltage action threshold is set to... The ground mode voltage action threshold is set to When the line-mode voltage at three consecutive sampling points exceeds... Or the ground mode voltage exceeds This indicates that a fault has occurred at this node.

[0051] After detecting a fault trigger, the fault type classification and identification module 1052, within a time window Internally, the line mode voltages are respectively... and ground mode voltage Perform integration calculations: Among them, This is the time when the fault was triggered.

[0052] Fault Selectivity Threshold The settings are pre-defined based on the system grounding method and insulation level, with a typical value being... Region identification threshold Pre-set based on line parameters and the length of the protected section; typical value is... .

[0053] Fault type determination: when and An abnormal drop indicates a positive ground fault. when and An abnormal drop indicates a negative grounding fault. when and and When both are abnormal, it indicates a grounding fault in both the positive and negative terminals; when At that time, a fault occurred between the poles.

[0054] Fault section determination: when At that time, a fault occurred within the area; when At that time, an external fault occurred.

[0055] Fault severity classification: Level 1 fault: The fault type is an internal equipment fault (such as DC / DC converter fault, photovoltaic module fault, energy storage battery thermal runaway), and the fault section is determined to be within the zone; Level 2 fault: The fault type is a DC bus fault (such as a bus short circuit), and the fault section is determined to be within the zone, or the zone communication bus detects a communication interruption; Level 3 fault: Grid connection detection module 404 detected an abnormal external grid voltage that lasted for more than [time period missing]. Or the system communication bus detects a global communication interruption.

[0056] The process of tiered self-isolation: Level 1 Fault Isolation: When the local edge processor 105 determines a Level 1 fault, the local isolation decision module 1053 directly sends an isolation command to the isolation execution unit 200 of this node. After receiving the isolation command, the control terminal of the solid-state DC circuit breaker 201 turns off the silicon carbide MOSFET within 100 microseconds, disconnecting the faulty node from the DC bus. The entire time from detection and decision-making to isolation of the Level 1 fault is no more than 2ms.

[0057] Level 2 fault isolation: When the edge fault detection units 100 of at least two adjacent nodes within the region simultaneously report a Level 2 fault, each node reports the fault information to the regional coordination controller 300 via the regional communication bus. The regional coordination controller 300 receives the fault trigger signals and fault type classification results reported by each node within its region and constructs a regional fault information matrix. ,in This represents the total number of nodes in this region. For the first Fault status identifier for each node. When At least two adjacent elements in the fault status are identified as level two faults and their fault characteristics are related (e.g., the time difference between fault occurrences is less than 10 ... When the fault types are the same, the area coordination controller 300 determines it as an area fault and issues an isolation command to the isolation execution unit 200 of all nodes in the fault area.

[0058] Level 3 fault isolation: When the grid connection detection module 404 detects that the external grid voltage is lower than 80% of the rated value or the frequency deviation exceeds 0.5Hz continuously... When the fault is detected, it is determined to be a level three fault. Each regional coordination controller 300 reports the system-level fault information to the system-level mode switching unit 400 via the system communication bus. The grid-connected / islanded mode switching switch 402 of the system-level mode switching unit 400 is opened, switching the system to islanded operation mode.

[0059] The specific process of black boot recovery is as follows: Figure 6 As shown: Phase 1: Prioritize the access of energy storage (t=0 to t=50ms) The energy storage priority access control module 501 of the black start recovery unit 500 queries the state of charge (SOC) of each energy storage node through the system communication bus and selects the energy storage node with the highest SOC as the black start power source.

[0060] The isolation execution unit 200 of the selected energy storage node is closed. The energy storage node injects voltage into the DC bus in voltage control mode through the DC / DC converter, with a voltage reference value. Gradually increase according to the ramp function: Where: initial voltage Rated voltage time constant , For time.

[0061] initial voltage Rated voltage time constant ; When the DC bus voltage reaches The above, and continues to be stable. After that, it is determined that the DC bus voltage has been established.

[0062] Phase Two: Gradual Grid Connection of Photovoltaics (t=50ms to t=150ms) After the DC bus voltage is established, the black-start recovery unit 500 sequentially controls the isolation execution units 200 of each photovoltaic power generation node to close. After the photovoltaic nodes are connected, MPPT control mode is used, and the output power gradually increases in the following manner: in, The maximum output power of this photovoltaic node This refers to the connection time of the photovoltaic node. This represents the photovoltaic power rise time constant. The connection interval between each photovoltaic node is 20ms to avoid power surges caused by multiple photovoltaic nodes connecting simultaneously.

[0063] Phase 3: Tiered load deployment (t=150ms to t=500ms) The tiered load control module 502 sequentially assigns loads at different levels according to their priority: Level 1 load input ( ): Delayed input time , and put into critical loads such as fire-fighting equipment, communication equipment, and emergency lighting; Secondary load input ( ): Delayed input time (At 150ms intervals), general loads such as ordinary lighting and air conditioning are activated; Level 3 load input ( ): Delayed input time (At 200ms interval) Connect to interruptible loads such as ordinary sockets and non-critical equipment.

[0064] After each load level is put into operation, monitor the DC bus voltage fluctuation to ensure it does not exceed 5% of the rated value, and confirm that the system is operating stably.

[0065] At this point, the system has completed the entire process from the occurrence of the fault to full recovery, with a total recovery time of no more than 500ms.

[0066] Smooth Grid Connection Switching: The grid connection detection module 404 of the system-level mode switching unit 400 continuously monitors the external power grid status. When it detects that the external power grid voltage amplitude has recovered to 90%-110% of the rated value and the frequency has recovered to... Phase stabilized and remained stable Then, pre-synchronization control is executed.

[0067] The specific steps of pre-synchronization control are as follows: Adjust the output voltage amplitude of the grid-connected inverter of the photovoltaic-storage-DC-flexible system to ensure that the voltage amplitude difference with the grid voltage is less than 5%; Adjust the output frequency of the grid-connected inverter to make the frequency difference with the grid less than 0.1Hz; Adjust the output voltage phase of the grid-connected inverter and track the grid phase through a phase-locked loop to make the phase difference less than 5 degrees; When the above three conditions are met simultaneously, close the grid-connected / islanded mode switching switch 402 to complete the grid-connected switching. After the switch is completed, the system switches from voltage control mode to current control mode to achieve power exchange with the power grid.

[0068] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive flexible system, characterized in that, Includes the following steps: S1. Distributed Fault Detection: Edge fault detection units are configured at each generation node, energy storage node, and load node of the photovoltaic-storage-DC-flexible system. Each edge fault detection unit independently collects the DC bus voltage signal, current signal, temperature signal, and communication status signal of its own node; the collected voltage signals are subjected to pole mode transformation to obtain line mode voltage component and ground mode voltage component; when the line mode voltage component or the ground mode voltage component meets the preset protection start conditions, it is determined that a fault has occurred in the node, and a fault trigger signal is output. S2. Fault Type Classification and Identification Steps: The edge fault detection unit of each node independently performs fault type classification and identification based on local detection information, dividing the faults into three levels: Level 1 faults are internal node faults; Level 2 faults are regional faults; Level 3 faults are system-level faults. S3, Hierarchical Self-Isolation: For a Level 1 fault, the edge fault detection unit of this node directly sends an isolation command to the isolation execution unit of this node to disconnect the faulty node from the DC bus, while the remaining nodes continue to operate normally. For Level 2 faults, the edge fault detection units of each node in this region report fault information to the regional coordination controller through the regional communication bus. After the regional coordination controller confirms the fault range and severity by integrating the fault information of multiple nodes in this region, it issues isolation instructions to the isolation execution units of all nodes in the fault area to isolate the entire fault area. Non-faulty areas maintain operation through topology reconstruction. For Level 3 faults, each regional coordination controller reports system-level fault information to the system-level mode switching unit through the system communication bus. The system-level mode switching unit disconnects the grid-connected / islanded mode switching switch and switches the system to islanded operation mode. S4. Resilient Reconstruction and Black Boot Recovery Steps: After the system switches to isolated operation mode, execute the black boot recovery procedure: S41, Energy Storage Priority Access: The black start recovery unit prioritizes the closing of the isolation execution unit of the energy storage node, so that the energy storage unit is connected to the DC bus and the DC bus voltage is established through the DC / DC converter; S42. Gradual Photovoltaic Connection: After the DC bus voltage has stabilized, photovoltaic power generation nodes are connected sequentially. S43. Tiered load activation: Based on the power supply priority of each load node, loads at each level are activated sequentially in descending order of priority to gradually restore the system. S5. Smooth grid connection switching: The system-level mode switching unit continuously monitors the status of the external power grid. When the external power grid is detected to have returned to normal, the smooth grid connection switching procedure is executed to switch the system from islanded operation mode back to grid-connected mode.

2. The method for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive flexible system according to claim 1, characterized in that, In step S1, the specific method for performing pole-mode transformation on the acquired voltage signal is as follows: ； in, This is the voltage between the positive busbar and ground. This is the voltage between the negative busbar and ground. Line-mode voltage, This is the ground mode voltage.

3. The method for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive flexible system according to claim 2, characterized in that, In step S2, the fault type classification identification specifically includes: Within a preset time window T, the line-mode voltage and ground-mode voltage are integrated to obtain the integral value of the line-mode voltage. and ground mode voltage integral value ; Based on the integral value of the ground mode voltage With fault selection threshold The comparison results determine the fault type: when and An abnormal drop indicates a positive ground fault. when and An abnormal drop indicates a negative grounding fault. when and and When both are abnormal, it indicates a grounding fault in both the positive and negative terminals; Based on the integral value of the line mode voltage With region identification threshold The comparison results determine the faulty section: when When the fault is located within the current node area, it is determined that the fault is located within this node area. when At that time, the fault is determined to be located outside the local node area.

4. The method for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive flexible system according to claim 1, characterized in that, In step S3, the specific method by which the regional coordination controller confirms the fault range by integrating fault information from multiple nodes within the region is as follows: The regional coordination controller receives fault trigger signals and fault type classification results reported by each node within its region, and constructs a regional fault information matrix. ,in This represents the total number of nodes in this region. For the first The fault status identifier of each node: when at least two adjacent nodes in the area simultaneously report a level 2 fault and the fault characteristics are related, the area coordination controller determines it as an area fault and issues an isolation command to the isolation execution unit of all nodes in the fault area.

5. The method for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive flexible system according to claim 1, characterized in that, In step S41, the specific method for prioritizing the connection of energy storage to establish the DC bus voltage is as follows: The black start recovery unit first queries the state of charge (SOC) of each energy storage node through the system communication bus and selects the energy storage node with the highest SOC as the black start power source. The isolation actuator of the selected energy storage node is closed, and the energy storage node injects voltage into the DC bus in voltage control mode through the DC / DC converter. The voltage reference value is... Gradually increase according to the ramp function: Where: initial voltage Rated voltage time constant , For time; When the DC bus voltage reaches more than 95% of the rated voltage and remains stable for a certain period of time... After that, it is determined that the DC bus voltage has been established.

6. The method for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive flexible system according to claim 1, characterized in that, In step S43, the specific method for tiered load allocation is as follows: The load nodes are pre-classified into three priority levels: Level 1 loads are critical loads, with a delayed start-up time. shortest; Level 2 loads are general loads, with a delayed start time. Next; Level 3 loads are interruptible loads, with a delayed start-up time. longest.

7. The method for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive flexible system according to claim 1, characterized in that, In step S5, the grid-connected smooth handover procedure specifically includes: The system-level mode switching unit detects the voltage amplitude, frequency, and phase of the external power grid in real time through the grid connection detection module; When the external power grid voltage amplitude, frequency, and phase are detected to have returned to the normal range and remained stable for a certain period of time. Then, pre-synchronization control is performed: the DC bus voltage of the photovoltaic-storage-DC-flexible system is adjusted to track the voltage amplitude, frequency and phase of the external power grid through the grid-connected inverter; When the voltage amplitude difference, frequency difference, and phase difference between the system side and the grid side are all less than the preset synchronization tolerance threshold, the grid-connected / islanded mode switching switch is closed to complete the grid-connected switching.

8. A system for enhancing the resilience and autonomous fault recovery of a distributed optical-storage-direct-drive-flexible system, characterized in that, include: Multiple edge fault detection units (100) are respectively installed at each power generation node, energy storage node and load node of the photovoltaic-storage-direct-drive-flexible system. Each edge fault detection unit (100) includes a voltage sensor (101), a current sensor (102), a temperature sensor (103), a communication status monitoring module (104) and a local edge processor (105). The output terminals of the voltage sensor (101), current sensor (102), temperature sensor (103) and communication status monitoring module (104) are all connected to the local edge processor (105). The local edge processor (105) has a built-in polar mode conversion module (1051), a fault type classification identification module (1052) and a local isolation decision module (1053). Multiple isolation execution units (200) are respectively set at each node and connected to the corresponding edge fault detection unit (100). Each isolation execution unit (200) includes a solid-state DC circuit breaker (201) connected in series on the DC bus of the node and a bypass switch (202) connected in parallel with the solid-state DC circuit breaker (201). Multiple regional coordination controllers (300) are respectively set in each preset region, and each edge fault detection unit (100) in the same region is connected to the corresponding regional coordination controller (300) through a regional communication bus; The system-level mode switching unit (400) is connected to each of the regional coordination controllers (300) via a system communication bus. The system-level mode switching unit (400) includes a grid-connected / islanded mode switching switch (402), a DC bus voltage establishment module (403), and a grid-connected detection module (404). And a black start recovery unit (500), which is connected to each of the isolation execution units (200) and each of the regional coordination controllers (300), respectively. The black start recovery unit (500) includes an energy storage priority access control module (501) and a graded load input control module (502).

9. The resilience enhancement and autonomous fault recovery system for a distributed optical-storage-direct-drive flexible system according to claim 8, characterized in that, The solid-state DC circuit breaker (201) is a solid-state DC circuit breaker based on silicon carbide MOSFET or IGCT, which has bidirectional current interruption capability and an interruption time of no more than 100 microseconds.

10. The resilience enhancement and autonomous fault recovery system for a distributed optical-storage-direct-drive flexible system according to claim 8, characterized in that, The DC bus voltage establishment module (403) is a DC / DC conversion unit inside the energy storage converter. Its input end is connected to the energy storage battery, and its output end is connected to the DC bus of the photovoltaic-energy storage DC-flexible system. It is used to convert the DC power of the energy storage battery into electrical energy that meets the DC bus voltage level requirements in islanded mode.

Images