Distributed cooperative protection system based on space-time synchronization and protection method thereof

Abstract

This application discloses a distributed collaborative protection system and method based on spatiotemporal synchronization. It includes multiple protection nodes connected in series along the power supply direction of the power distribution system in each level of circuit. Nodes located upstream in the power supply direction are the superior nodes of downstream nodes, and nodes located downstream in the power supply direction are the inferior nodes of upstream nodes. According to the embodiments of this application, the distributed collaborative protection system based on spatiotemporal synchronization establishes microsecond-level time synchronization by connecting multiple protection nodes in series along the power supply direction. This allows the edge controller to accurately determine the node closest to the fault point based on the timestamp and amplitude of current mutations reported by each node, and instruct it to prioritize cutting off the circuit. Its superior node only waits in backup mode, thus completely solving the problem of cascading tripping caused by lack of coordination in traditional power distribution protection, and reducing the cascading tripping failure rate.

Description

Technical Field

[0001] This application relates to the technical field of circuit protection, and in particular to a distributed collaborative protection system and protection method based on spatiotemporal synchronization. Background Technology

[0002] Prior art document 202510136556.1 discloses a self-resetting sensitive protector for current overload and overtemperature. It employs a mechanical structure including a ceramic tube, a cross-shaped mandrel, a memory spring, a steel wire spring, and a flared sleeve to achieve single-circuit overcurrent and overheat self-resetting protection, solving the problems of short lifespan and large size associated with traditional one-time fuses. However, this protector has the following shortcomings in use: when multiple protectors are used in series, they operate independently and cannot achieve coordinated operation between higher and lower levels. When a fault occurs in one level of circuit, the higher-level protectors often trip simultaneously, leading to cascading tripping and power outages in non-faulty circuits. Summary of the Invention

[0003] This application aims to at least partially address one of the technical problems in the related art.

[0004] Therefore, one objective of this application is to provide a distributed collaborative protection system based on spatiotemporal synchronization. By connecting multiple protection nodes in series along the power supply direction and establishing microsecond-level time synchronization, the edge controller can accurately determine the node closest to the fault point based on the timestamp and amplitude of current mutation reported by each node, and instruct it to prioritize cutting off the circuit. Its superior node only waits in backup, thereby completely solving the problem of cascading tripping caused by lack of coordination in traditional power distribution protection and reducing the cascading tripping failure rate.

[0005] To achieve the above objectives, a first aspect of this application proposes a distributed cooperative protection system based on spatiotemporal synchronization, comprising: Multiple protection nodes are connected in series in each level of the power distribution system along the power supply direction. The node located upstream in the power supply direction is the upper-level node of the downstream node, and the node located downstream in the power supply direction is the lower-level node of the upstream node. Each of the protection nodes includes a loop switching actuator, a current detection unit, and a communication unit; An edge controller is communicatively connected to each of the protection nodes; The edge controller is configured to: The system establishes time synchronization with each of the aforementioned protection nodes, with a synchronization accuracy better than 100 microseconds; Receive current data with timestamps reported in real time by each of the protection nodes; When a fault current is detected, the fault location is automatically determined based on the timestamp and amplitude of the current change reported by each node. The node closest to the fault point is the node with the earliest current change time and the largest change amplitude. The instruction states that the node closest to the fault point should take priority in performing protection actions and disconnecting the faulty circuit; The instruction commands the parent node of the node closest to the fault point to enter a backup waiting state, and the parent node performs backup protection actions when the node closest to the fault point fails to operate.

[0006] According to the distributed collaborative protection system based on spatiotemporal synchronization in this application embodiment, multiple protection nodes are connected in series along the power supply direction and a microsecond-level time synchronization is established. This enables the edge controller to accurately determine the node closest to the fault point based on the timestamp and amplitude of the current change reported by each node, and instruct it to prioritize cutting off the circuit. Its superior node only waits in backup, thereby completely solving the problem of over-level tripping caused by lack of coordination in traditional power distribution protection and reducing the over-level tripping failure rate.

[0007] In addition, the distributed cooperative protection system based on spatiotemporal synchronization proposed in this application may also have the following additional technical features: In one embodiment of this application, the time synchronization is achieved in the following manner: The edge controller broadcasts a synchronization message to all protection nodes via the communication bus at predetermined intervals. The synchronization message contains the current timestamp of the edge controller. After receiving the synchronization message, each protection node calibrates its local clock according to the timestamp in the synchronization message; The local clock is implemented by the internal timer of the MCU of the protection node, and the clock source of the internal timer of the MCU is a crystal oscillator.

[0008] In one embodiment of this application, the communication unit is a single-bus communication chip, and the single-bus communication chip is connected to the general-purpose input / output pins of the microcontroller unit of the protection node; The data pins and power supply voltage pins of the single-bus communication chip are connected in parallel to the same physical bus. The protection node obtains operating power from the physical bus through the parasitic power supply method of the single-bus communication chip. The parasitic power supply method is as follows: the high-level signal on the physical bus charges the energy storage capacitor inside the protection node through the power supply voltage pin of the single-bus communication chip.

[0009] In one embodiment of this application, the current detection unit includes a sampling resistor connected in series in the loop of the protection node, and the sampling resistor is connected to the ADC pin of the MCU of the protection node.

[0010] In one embodiment of this application, the current detection unit includes a current transformer, the primary coil of which is connected in series in the circuit of the protection node, and the secondary coil of which is connected to the ADC pin of the MCU of the protection node.

[0011] In one embodiment of this application, the protection node further includes a non-volatile memory chip, which is connected to the MCU of the protection node via a bus; The MCU is configured to cache, in the form of a circular buffer, full current, voltage, temperature, and action timing data in the non-volatile memory chip, including the time before the fault, the fault process, and the full current, voltage, temperature, and action timing data after the fault recovery.

[0012] In one embodiment of this application, the protection node further includes a non-volatile memory chip, which is connected to the MCU of the protection node via a bus; The MCU is configured to cache, in the form of a circular buffer, full current, voltage, temperature, and action timing data in the non-volatile memory chip, including the time before the fault, the fault process, and the full current, voltage, temperature, and action timing data after the fault recovery.

[0013] In one embodiment of this application, the automatic determination of the fault location includes: Compare the timestamps of current surges reported by each node, and determine the node with the earliest current surge time as the node closest to the fault point; When the difference in timestamps of current surges among multiple nodes is less than a preset threshold, the magnitudes of current surges are further compared, and the node with the largest surge magnitude is determined to be the node closest to the fault point.

[0014] In one embodiment of this application, the preset delay of the backup protection action is dynamically configured according to the topology hierarchy of the power distribution system. The delay time of the upper-level node is greater than the delay time of the lower-level node, and the delay time difference between adjacent level nodes is not less than the action time margin of the protection node.

[0015] The second aspect of this application proposes a distributed collaborative protection method based on spatiotemporal synchronization, applied to the distributed collaborative protection system based on spatiotemporal synchronization proposed in the first aspect, comprising the following steps: S1. The edge controller establishes communication connections with multiple protection nodes and establishes time synchronization with each protection node, with a synchronization accuracy better than 100 microseconds; S2. Each protection node collects the current waveform data of its circuit in real time and reports the current data with timestamps to the edge controller. S3. The edge controller receives data reported by each node. When a fault current is detected, it automatically determines the fault location based on the timestamp and amplitude of the current change reported by each node. S4. The edge controller instructs the node closest to the fault point to perform protection actions first, and at the same time instructs the upstream node of that node to enter a backup waiting state. S5. If the node closest to the fault point successfully operates, the collaborative protection process ends; if the node closest to the fault point fails to operate within a preset delay, the superior node performs the backup protection action.

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

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the structure of a distributed collaborative protection system based on spatiotemporal synchronization and the protection system according to an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a distributed collaborative protection system based on spatiotemporal synchronization and the protection system according to another embodiment of this application. Detailed Implementation

[0018] Embodiments of this application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. Rather, embodiments of this application include all variations, modifications, and equivalents falling within the spirit and scope of the appended claims.

[0019] The following is in conjunction with the appendix Figures 1-2 This application describes a distributed collaborative protection system based on spatiotemporal synchronization, according to an embodiment of the present application.

[0020] The distributed collaborative protection system based on spatiotemporal synchronization provided in this application includes: Multiple protection nodes are connected in series in each level of the power distribution system along the power supply direction. The node located upstream in the power supply direction is the upper-level node of the downstream node, and the node located downstream in the power supply direction is the lower-level node of the upstream node. Each protection node includes a loop switching actuator, a current detection unit, and a communication unit; The edge controller communicates with each protection node. The edge controller is configured as follows: Establish time synchronization with each protection node, with a synchronization accuracy better than 100 microseconds. The synchronization accuracy is configured to be less than an order of magnitude of the time difference required for the fault current to propagate along the conductor between adjacent protection nodes. Receive current data with timestamps reported in real time by each protection node; When a fault current is detected, the fault location is automatically determined based on the timestamp and amplitude of the current change reported by each node. The node closest to the fault point is the node with the earliest current change time and the largest change amplitude. The instruction prioritizes the node closest to the fault point to perform protection actions and disconnect the faulty circuit. The instruction commands the parent node of the node closest to the fault point to enter a backup waiting state, and if the node closest to the fault point fails to operate, the parent node will perform the backup protection action.

[0021] Specifically, in this system, multiple such protectors are connected in series at different locations along the power supply direction. For example, in a DC power distribution trunk line, the protector closer to the power source acts as the upstream node, and the protector closer to the load acts as the downstream node, forming a clear upstream-downstream relationship. Each protector, in addition to its original structure, includes a microcontroller (MCU), a current sampling circuit (using a three-jaw leaf spring to obtain the loop current by detecting the voltage across its terminals), a 1-Wire single-bus communication chip, and an energy storage capacitor. All protectors' communication chips are connected to an edge controller (such as a microcomputer installed in a distribution box) via the same twisted-pair bus.

[0022] After the system powers on, the edge controller broadcasts a synchronization message via the bus at regular intervals (e.g., every second). Upon receiving the message, each protector's MCU adjusts its local clock, ultimately ensuring that the clock deviation of all protectors is better than 100 microseconds. This accuracy is set to be an order of magnitude less than the time difference required for fault current to propagate between adjacent protectors. For example, the wire length between two adjacent protectors is typically tens of meters, and the fault current propagation time difference is approximately 0.5 microseconds. A synchronization accuracy of 100 microseconds is far less than one-tenth of this time difference, thus sufficient to accurately distinguish which protector detects the current surge first.

[0023] During routine monitoring, each protector's MCU continuously collects the voltage across the three-jaw leaf spring, converts it into a current value, timestamps it locally, and then sends it to the edge controller via the bus. When a short-circuit fault occurs in a certain stage of the circuit, the fault current appears instantaneously. The protector closest to the fault point (e.g., the downstream node) detects the current surge first, and its MCU records the surge time T1 and surge magnitude I1. Subsequently, the fault current propagates to the upstream node, and the upstream node's MCU records a slightly later time T2 and magnitude I2 (I2 is usually less than I1). Both nodes immediately report the timestamped fault data to the edge controller.

[0024] Upon receiving the data, the edge controller automatically compares it: since T1 is earlier than T2 and I1 is greater than I2, it determines that the lower-level node is the closest to the fault point. The edge controller then instructs the lower-level node to prioritize the protection action via the bus. The node's MCU drives a MOSFET connected in series in the loop (or directly utilizes the tripping mechanism of the original protector) to quickly disconnect the loop, isolating the fault. Simultaneously, the edge controller instructs the upper-level node of this lower-level node to enter a "backup standby" state, meaning the upper-level node does not act immediately but continues monitoring.

[0025] If the downstream node successfully disconnects the circuit, the entire protection process ends, the upstream node remains closed, and the power supply to other non-faulty circuits is unaffected. If the downstream node fails to operate due to mechanical jamming, electronic circuit damage, or other reasons and fails to disconnect within the preset short delay, the upstream node will automatically execute backup protection actions, disconnecting its upstream circuit to ensure that the fault does not continue to escalate.

[0026] Through the above process, this system achieves precise coordination of "prioritizing lower-level disconnection for lower-level faults, with upper-level protection only providing backup," avoiding the cascading tripping phenomenon common in traditional power distribution protection. The entire judgment and execution are completed in milliseconds, without requiring manual setting of settings or reliance on cloud commands, thus improving the power supply reliability of the power distribution system.

[0027] It should be noted that the specific installation locations of the microcontroller and sampling circuit in this application on the "current overload and over-temperature self-resetting sensitive protector" are not limited to the "current overload and over-temperature self-resetting sensitive protector".

[0028] I. Microcontroller settings.

[0029] The microcontroller (MCU) is mounted on a small PCB board, which is fixed to the middle of the outer wall of the ceramic tube by epoxy resin adhesive or clips, located in the blank area between the upper and lower conductive caps. This location was chosen because there are no other components on the outer wall of the ceramic tube at this point, providing sufficient space, and it is far from the pins at both ends and the conductive caps, preventing high-voltage discharge from interfering with the MCU. The MCU's pins are connected to the sampling unit, communication chip, memory chip, and MOSFET control terminals via copper traces on the PCB board.

[0030] II. Location of the current sampling circuit.

[0031] The core of the current sampling circuit is the three-jaw leaf spring from the original patent. This spring is installed inside the ceramic tube, specifically between the shoulder of the cross-shaped mandrel and the flared sleeve, with the three jaws of the spring in contact with the outer wall of the flared sleeve. When current flows through the protector, a voltage difference proportional to the current is generated across the three-jaw leaf spring.

[0032] To obtain this voltage signal, two thin wires are soldered to each end of the three-jaw leaf spring. These two wires run along the inner wall of the ceramic tube, exiting through a pre-drilled hole at the bottom of the ceramic tube (near the flared sleeve) to the outside of the tube. They are then soldered to two pads on the PCB board and connected to the two ADC input pins of the MCU via PCB traces. By detecting the voltage difference between these two pins and combining this with the known resistance value of the three-jaw leaf spring, the MCU can calculate the current flowing through the protector's loop.

[0033] To ensure signal stability, the two leads are fitted with high-temperature resistant insulating sleeves inside the ceramic tube to prevent contact or short circuits with internal moving parts such as the cross-shaped spindle and memory spring.

[0034] III. Location of the overall layout.

[0035] Three-jaw leaf spring (sampling element): located inside the ceramic tube, between the shoulder and the flared sleeve.

[0036] Lead wires: Lead out from the small hole at the bottom of the ceramic tube and connect to the external PCB board.

[0037] PCB board: fixed to the middle of the outer wall of the ceramic tube.

[0038] MCU (Microcontroller): Soldered on the PCB board, it receives voltage signals from a three-jaw leaf spring through the ADC pin.

[0039] In one embodiment of the application, time synchronization is achieved in the following way: The edge controller broadcasts a synchronization message to all protection nodes via the communication bus at predetermined intervals. The synchronization message contains the current timestamp of the edge controller. After receiving the synchronization message, each protection node calibrates its local clock according to the timestamp in the synchronization message; The local clock is implemented by the internal timer of the MCU in the protection node, and the clock source of the internal timer of the MCU is a crystal oscillator.

[0040] Specifically, in actual use, time synchronization is achieved as follows: The edge controller is connected to all protection nodes via the same communication bus. After the system is powered on, the edge controller has a high-precision clock source (such as a real-time clock with a temperature-compensated crystal oscillator) inside, which serves as the time reference for the entire system.

[0041] Every fixed interval (e.g., 1 second), the edge controller will proactively broadcast a special synchronization message onto the bus. This message contains a crucial piece of information: the edge controller's current timestamp, accurate to the microsecond level.

[0042] Every protection node on the bus receives this broadcast message. Each protection node has an internal MCU, and the MCU has a timer whose clock source is a crystal oscillator soldered onto the PCB board.

[0043] Upon receiving the synchronization message, the MCU of the protection node immediately records the current value of its local timer and then parses the timestamp of the edge controller from the message.

[0044] Next, the MCU calculates the difference between the local time and the edge controller time. For example, if the timestamp in the message is 1000.000 seconds, while the local timer displays 999.999 seconds, then the deviation is -0.001 seconds. The MCU saves this deviation value as compensation when timestamping all subsequent sampled data.

[0045] In this way, the local clock of each protection node is synchronized with the clock of the edge controller. Due to the high short-term stability of the crystal oscillator, the clock drift of each node is minimal within the 1-second interval between two synchronization messages. Through this periodic broadcast calibration method, the time deviation between all protection nodes is controlled within a range better than 100 microseconds, which is sufficient to accurately distinguish the order in which fault currents arrive at different nodes. The entire synchronization process is automatic and periodic, requiring no manual intervention.

[0046] In one embodiment of the application, the communication unit is a 1-Wire single-bus communication chip, which is connected to the GPIO pin of the MCU of the protection node; The data pin and VDD pin of the 1-Wire single-bus communication chip are connected in parallel to the same physical bus; The protection node obtains its operating power from the physical bus through the parasitic power supply method of the 1-Wire single-bus communication chip. The parasitic power supply method is as follows: the high-level signal on the physical bus charges the energy storage capacitor inside the protection node through the VDD pin of the 1-Wire single-bus communication chip.

[0047] In practical use, the protection node contains a 1-Wire single-bus communication chip. This chip is responsible for connecting the protection node to the bus and exchanging data with the edge controller. The entire communication bus has only two wires: one is a signal line shared by data and power, and the other is a ground wire. The protection node does not need a separate power supply line; its operating power is directly obtained from the bus signal line.

[0048] The specific working process is as follows: When the bus is idle, the edge controller maintains the bus level at a high 5V. This high-level signal continuously charges a small energy storage capacitor on the protection node circuit board through the internal circuitry of the communication chip. The energy storage capacitor stores the electrical energy, which is then regulated and supplied to the MCU and other circuits on the protection node.

[0049] When the edge controller needs to send data, it pulls the bus level low or releases the high level according to the 1-Wire protocol, forming a specific pulse sequence. The communication chip detects these level changes, converts them into digital signals that the MCU can recognize, and transmits them to the MCU through its pins. Conversely, when the protection node needs to report data, the MCU sends the data to the communication chip, which then transmits the data back to the edge controller by controlling the bus level changes.

[0050] During data transmission, the bus voltage level will briefly drop. At this time, the energy stored in the energy storage capacitor will be automatically released to maintain power supply to the protection node during the low-level period, preventing the node from losing power due to the bus voltage drop. After data transmission ends, the bus returns to a high level, and the energy storage capacitor begins to recharge, preparing for the next communication.

[0051] In this way, all protection nodes are connected in parallel to the edge controller through the same twisted pair cable, which not only realizes data communication but also solves the power supply problem. Only this pair of cables needs to be laid during on-site installation.

[0052] It should be noted that the edge controller acts as the master clock, periodically (e.g., every 1 second) broadcasting synchronization messages to the bus, with the messages carrying the current time of the master clock. .

[0053] node Record the local clock when a message is received. Calculate the deviation .

[0054] node Deviation Storage, and the timestamps of all subsequent locally sampled data will be compensated to .

[0055] Considering crystal oscillator drift, clock deviations between two adjacent synchronization intervals are compensated using linear interpolation.

[0056] The final output is data reported by all nodes with a unified timestamp traceable to the same baseline and with an accuracy better than 100 microseconds.

[0057] In one embodiment of the application, the current detection unit includes a sampling resistor connected in series in the loop of the protection node and connected to the ADC pin of the MCU of the protection node.

[0058] In practical applications, the current detection unit of the protection node uses a low-resistance sampling resistor. This resistor is directly connected in series in the main current circuit controlled by the protection node. That is, the current flows in from one end of the circuit, passes through the sampling resistor, and then flows out from the other end. The sampling resistor bears the entire circuit current.

[0059] The two ends of the sampling resistor are connected to the two ADC (analog-to-digital converter) input pins of the MCU inside the protection node via two wires. When current flows through the circuit, a small voltage difference is generated across the sampling resistor, the magnitude of which is proportional to the current.

[0060] The MCU measures this voltage difference in real time through the ADC pin, and then calculates the actual current value in the current circuit based on the known resistance value of the sampling resistor.

[0061] Because the resistance of the sampling resistor is very small (usually in the milliohm range), its influence on the loop current is negligible, and its power consumption is very low, so it does not generate significant heat.

[0062] The MCU reads the ADC value at a fixed sampling frequency (e.g., 1000 times per second) to obtain continuous, real-time current waveform data, which is used to determine whether an overload or short-circuit fault has occurred.

[0063] In one embodiment of the application, the current detection unit includes a current transformer, the primary coil of which is connected in series in the circuit of the protection node, and the secondary coil of which is connected to the ADC pin of the MCU of the protection node.

[0064] In practical applications, the current detection of the protection node uses a current transformer. The primary coil of the current transformer has very few turns (usually only one or a few turns) and is directly connected in series in the main current circuit controlled by the protection node. That is to say, the circuit conductor passes through or is wound around the center of the transformer once, and all circuit current flows through the primary coil.

[0065] The secondary coil of the current transformer has a large number of turns, and its two ends are connected to the two ADC input pins of the MCU inside the protection node via wires. When an alternating current flows through the main circuit, the primary coil generates an alternating magnetic field, and the secondary coil induces an alternating voltage signal proportional to the primary current.

[0066] The MCU acquires this voltage signal through the ADC pin, and after rectification, filtering and conversion, obtains the actual current value of the main circuit.

[0067] For DC circuits, Hall effect current transformers (or Hall current sensors) are used, with a similar working principle: the magnetic field generated by the primary current is detected by the Hall element, outputting a voltage signal proportional to the current, which is also connected to the MCU's ADC pin for measurement. The MCU reads the ADC value at a fixed sampling frequency (e.g., 1000 times per second) to obtain continuous, real-time current waveform data, used to determine whether an overload or short-circuit fault has occurred. The current transformer itself is not directly electrically connected to the main circuit, thus having a natural isolation advantage, making it suitable for high-voltage or high-current applications.

[0068] In one embodiment of the application, the protection node further includes a non-volatile memory chip, which is connected to the MCU of the protection node via an SPI bus; The MCU is configured to cache, in the form of a circular buffer, full current, voltage, temperature, and timing data in a non-volatile memory chip, including the time before the fault, the fault process, and the full current, voltage, temperature, and operation timing data after the fault recovery.

[0069] In practical use, each protection node's circuit board is also equipped with a non-volatile memory chip (such as Flash memory). This chip is connected to the MCU via the SPI bus. Specifically, the MCU's four pins, SCK, MISO, MOSI, and CS, are connected to the corresponding pins of the memory chip. The MCU writes or reads data from the memory chip through these four lines.

[0070] The program running inside the MCU manages the memory chip in the following way: it allocates a fixed-size memory area as a circular buffer. That is, the MCU continuously writes the latest data into this buffer, and when the entire area is full, it automatically goes back to the beginning to overwrite the oldest data, always keeping the buffer containing data from the most recent period.

[0071] What specific information is cached? The MCU continuously records three aspects of information: first, the electrical parameters of the circuit, including current, voltage, and temperature; second, the operational status of the protection nodes, such as whether a protection action has occurred and the time of the action; and third, the detailed timing of each action, such as the time elapsed from detecting an anomaly to executing the cutoff. All this data is stored in time-series format, with each data entry carrying a precise timestamp.

[0072] The buffer size is set to store a predetermined time before the failure (e.g., 10 seconds before the failure), the duration of the failure, and all data after the failure is resolved. In other words, when a failure event occurs, the memory chip retains not only the data at the moment of the failure, but also the normal data for a period of time before the failure, as well as the data during the system recovery process after the failure is resolved.

[0073] When a fault occurs, the MCU reports the cached data to the edge controller via the communication bus. Because it is non-volatile memory, the data in the storage chip is not lost even if the protection node itself loses power; it can still be read after power is restored for post-fault analysis of the cause of the fault and assessment of the protection node's health. This circular buffer design ensures that the storage chip capacity does not need to be large (typically a few megabytes are sufficient) while continuously recording the most valuable pre- and post-fault data.

[0074] In one embodiment of the application, the protection node further includes a non-volatile memory chip, which is connected to the MCU of the protection node via an I²C bus; The MCU is configured to cache, in the form of a circular buffer, full current, voltage, temperature, and timing data in a non-volatile memory chip, including the time before the fault, the fault process, and the full current, voltage, temperature, and operation timing data after the fault recovery.

[0075] In practical use, each protection node's circuit board is also equipped with a non-volatile memory chip (such as EEPROM or ferroelectric memory).

[0076] This chip connects to the MCU via the I²C bus. Specifically, the MCU's clock line SCL and data line SDA are connected to the corresponding pins of the memory chip. The MCU writes or reads data from the memory chip through these two lines. Compared to the SPI bus, the I²C bus requires only two signal lines, occupying fewer MCU pins, making it suitable for protection nodes with limited pin resources.

[0077] The program running inside the MCU manages the memory chip in the following way: it allocates a fixed-size memory area as a circular buffer. That is, the MCU continuously writes the latest data into the buffer, and when the entire area is full, it automatically returns to the beginning to overwrite the oldest data, always ensuring that the buffer contains data from the most recent period.

[0078] The specific cached content includes: circuit current, voltage, temperature, and the operating status of protection nodes and the detailed timing of each action. All data is accurately timestamped. The buffer capacity is set to store a predetermined time before the fault, the duration of the fault, and all data after the fault recovery. Thus, when a fault occurs, the memory chip retains not only the data at the moment of the fault, but also the normal data for a period before the fault and the recovery process data after the fault.

[0079] After a fault occurs, the MCU reports the cached data to the edge controller via the communication bus. Because it uses non-volatile storage, the data is not lost even if the protection node loses power and can be read again after power is restored for post-fault analysis and node health assessment. The circular buffer design allows a small-capacity storage chip to continuously record the most valuable pre- and post-fault data.

[0080] In one embodiment of the application, automatically determining the fault location includes: Compare the timestamps of current surges reported by each node, and determine the node with the earliest current surge time as the node closest to the fault point; When the difference in timestamps of current surges among multiple nodes is less than a preset threshold, the magnitudes of current surges are further compared, and the node with the largest surge magnitude is determined to be the node closest to the fault point.

[0081] In practical use, after receiving fault data reported by multiple protection nodes, the edge controller will automatically determine the fault location according to the following steps: The first step is to compare the timestamps of current surges in all reported data. The edge controller identifies the node with the earliest timestamp; this node was the first to detect the current surge and is therefore determined to be the node closest to the fault point.

[0082] The second step involves comparing the current surge timestamps of two or more nodes that are very close, with a difference less than a pre-set threshold, where the timestamps alone cannot clearly distinguish the order of events. In this case, the edge controller further compares the current surge amplitudes of these nodes. During the propagation of the fault current, nodes closer to the fault point measure larger current amplitudes. Therefore, the edge controller selects the node with the largest surge amplitude and determines it as the node closest to the fault point.

[0083] It should be noted that the fault current detection function quickly and accurately identifies the start time and sudden amplitude of the fault current from continuous current sampling values. Specifically: The node collects current values ​​at a fixed frequency (e.g., 1 kHz). ,in Indicates the first One sampling point.

[0084] Calculate the difference between the current sampling point and the previous period (e.g., 20ms ago). , The number of periodic points.

[0085] like If the current exceeds a preset threshold (such as 3 times the rated current), it is determined to be a sudden current change.

[0086] Mutation time The timestamp corresponding to the sampling point that first exceeds the threshold.

[0087] The amplitude of the sudden change is taken as the effective value or peak value of the current in the first steady-state cycle after the sudden change.

[0088] Sampling frequency: The rate at which the MCU collects current, for example, 1kHz means 1000 samples per second.

[0089] Cycle length: Take the time of one power frequency cycle, for example, 20 milliseconds (corresponding to a complete cycle of 50Hz AC).

[0090] in , Indicates the current sampling point The current value at the sampling point one power frequency cycle (20 milliseconds) prior. By comparing the current with the current one cycle ago, it is possible to determine whether a sudden change has occurred, and at the same time eliminate the influence of normal power frequency fluctuations.

[0091] Then, the fault location is determined: based on fault data reported by multiple nodes, the node closest to the fault point is accurately identified. Specifically: Time-priority determination: Collect data from all nodes that report faults and extract the timestamps of their current surges.

[0092] Select the value with the smallest current change timestamp among all reporting nodes, and denote it as . .

[0093] Select Corresponding node set ,in, This indicates the point in time when the current surge is first detected. Since the fault current propagates from the fault point to both sides, the node closest to the fault point will detect the current surge first; therefore, this node's... Minimum.

[0094] If set S1 contains only one node, that is This node is the closest node to the fault point, and the algorithm ends.

[0095] If set S1 has multiple nodes, that is (i.e., the difference between the timestamps of multiple nodes is less than a preset threshold) For example, 1 microsecond This is a preset time threshold. When the timestamp difference between multiple nodes is less than this threshold, they are considered to have detected the fault current almost simultaneously. This situation may occur when multiple nodes are very close to the fault point, or when the sampling accuracy is limited: proceed to amplitude-assisted determination.

[0096] Amplitude auxiliary determination: in the set In the middle, compare the amplitude of current surges at each node. Select The corresponding node is the node closest to the fault point, where, It is the maximum value among all current surge amplitudes.

[0097] Output: Fault location determination result (node ​​number).

[0098] Through this dual judgment method of "time priority and amplitude assistance," the edge controller can accurately determine the fault location, providing a basis for issuing subsequent protection commands. The entire judgment process is completed automatically by software without manual intervention.

[0099] In one embodiment of the application, the preset delay of the backup protection action is dynamically configured according to the topology hierarchy of the power distribution system. The delay time of the upper-level node is greater than the delay time of the lower-level node, and the delay time difference between adjacent level nodes is not less than the action time margin of the protection node.

[0100] In practical use, the preset delay for backup protection actions is not a fixed value, but is automatically calculated and configured by the edge controller according to the topology of the power distribution system for each protection node.

[0101] The specific rule is: the latency of the parent node is greater than the latency of the child node. For example, the latency of the lowest child node (closest to the load) is set to the shortest (e.g., 2 milliseconds), its direct parent node's latency is set to a longer one (e.g., 5 milliseconds), the next higher-level node's latency is set to 8 milliseconds, and so on. This way, when a fault occurs, the node closest to the fault point will act first; if it refuses to act, its parent node will act after a slightly longer delay, but will not wait for the next higher-level node to act first.

[0102] Meanwhile, the time difference between adjacent level nodes cannot be less than the "action time margin" of the protection node itself. The action time margin refers to the maximum time required for a protection node to complete mechanical disconnection from receiving the command (e.g., 1 millisecond, including MCU processing time, MOSFET or tripping mechanism action time). This constraint is set to ensure that when a lower-level node fails to operate, the upper-level node has sufficient time margin to recognize that the lower-level node has not successfully disconnected before initiating its own backup action, and avoids a situation where the upper-level node operates simultaneously with the lower-level node due to too small a time difference.

[0103] When the system powers on, the edge controller automatically identifies the topology of the entire power distribution network (how many levels there are, and which nodes are in each level). Then, it calculates the appropriate backup delay time for each node according to the aforementioned rules and sends it to each node for storage via the communication bus. The entire process requires no manual configuration, and when the topology changes (such as adding or removing nodes), the edge controller recalculates and updates the configuration.

[0104] It should be noted that after the system powers on, the edge controller first acquires the topology information of the entire power distribution network. There are two methods for acquisition: one is that each protection node actively reports its installation location information via the communication bus (e.g., "I am connected downstream of node A"); the other is through a configuration file pre-stored in the edge controller. Based on this information, the edge controller constructs a directed graph, clarifying the positional relationship of each node in the power supply direction, and assigns a hierarchical number to each node. The rule is: the level of the end node closest to the load side is... For each level above the power supply side, the level increments by 1, meaning it is the node directly above the power supply. one more level And so on.

[0105] Next, the edge controller sets a basic delay unit. This value needs to be greater than the protection node's own action time margin. The action time margin refers to the maximum time (e.g., 0.5 milliseconds) required for a node to actually complete the mechanical disconnection after receiving the disconnection command. Typically, 1 millisecond is used to ensure sufficient margin.

[0106] Then, the edge controller calculates the backup protection delay time for each node. The calculation formula is: In other words, the end node ( The backup latency is 0 milliseconds, and its direct parent node ( The delay is 1 millisecond, and the next level up ( The delay is 2 milliseconds, and so on. Delay time for backup protection.

[0107] With this configuration, the latency difference between adjacent level nodes is exactly equal to For example, the latency of a lower-level node is 0 milliseconds, while the latency of a higher-level node is 1 millisecond, a difference of 1 millisecond. This time difference ensures that the lower-level node has enough time to complete its action (action time margin of 0.5 milliseconds). If the lower-level node refuses to act, the higher-level node will initiate a backup action after 1 millisecond, preventing a situation where both levels act simultaneously.

[0108] After completing the calculations, the edge controller distributes the backup delay time for each node via the communication bus and stores it within the node. When a fault occurs, the node closest to the fault (the lower-level node) immediately performs a disconnection with a 0-millisecond delay; if it refuses to operate, its upper-level node automatically operates after 1 millisecond; if the upper-level node also refuses to operate, the next higher-level node operates after 2 milliseconds. This hierarchical approach ensures both rapid fault isolation and a complete backup protection chain.

[0109] If the topology of the power distribution system changes (e.g., nodes are added or removed), the edge controller will re-identify the topology upon the next power-on and recalculate and distribute the backup delay time for all nodes, always keeping the configuration matched to the current topology.

[0110] This application provides a distributed collaborative protection method based on spatiotemporal synchronization, applied to a distributed collaborative protection system based on spatiotemporal synchronization, comprising the following steps: S1. The edge controller establishes communication connections with multiple protection nodes and establishes time synchronization with each protection node, with a synchronization accuracy better than 100 microseconds; S2. Each protection node collects the current waveform data of its circuit in real time and reports the current data with timestamps to the edge controller. S3. The edge controller receives data reported by each node. When a fault current is detected, it automatically determines the fault location based on the timestamp and amplitude of the current change reported by each node. S4. The edge controller instructs the node closest to the fault point to perform protection actions first, and at the same time instructs the upstream node of that node to enter a backup waiting state. S5. If the node closest to the fault point successfully operates, the collaborative protection process ends; if the node closest to the fault point fails to operate within the preset delay, the superior node will perform the backup protection action.

[0111] This application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements a distributed collaborative protection method based on spatiotemporal synchronization.

[0112] In summary, the distributed collaborative protection system based on spatiotemporal synchronization in this application establishes microsecond-level time synchronization by connecting multiple protection nodes in series along the power supply direction. This enables the edge controller to accurately determine the node closest to the fault point based on the timestamp and amplitude of current mutation reported by each node, and instruct it to prioritize cutting off the circuit. Its superior node only waits in backup, thereby completely solving the problem of cascading tripping caused by lack of coordination in traditional power distribution protection and reducing the cascading tripping failure rate.

[0113] In the description of this specification, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0114] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0115] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A distributed collaborative protection system based on spatiotemporal synchronization, characterized in that, include: Multiple protection nodes are connected in series in each level of the power distribution system along the power supply direction. The node located upstream in the power supply direction is the upper-level node of the downstream node, and the node located downstream in the power supply direction is the lower-level node of the upstream node. Each of the protection nodes includes a loop switching actuator, a current detection unit, and a communication unit; An edge controller is communicatively connected to each of the protection nodes; The edge controller is configured to: The system establishes time synchronization with each of the aforementioned protection nodes, with a synchronization accuracy better than 100 microseconds; Receive current data with timestamps reported in real time by each of the protection nodes; When a fault current is detected, the fault location is automatically determined based on the timestamp and amplitude of the current change reported by each node. The node closest to the fault point is the node with the earliest current change time and the largest change amplitude. The instruction states that the node closest to the fault point should take priority in performing protection actions and disconnecting the faulty circuit; The instruction commands the parent node of the node closest to the fault point to enter a backup waiting state, and the parent node performs backup protection actions when the node closest to the fault point fails to operate.

2. The distributed collaborative protection system based on spatiotemporal synchronization according to claim 1, characterized in that, The time synchronization is achieved in the following ways: The edge controller broadcasts a synchronization message to all protection nodes via the communication bus at predetermined intervals. The synchronization message contains the current timestamp of the edge controller. After receiving the synchronization message, each protection node calibrates its local clock according to the timestamp in the synchronization message; The local clock is implemented by the internal timer of the MCU of the protection node, and the clock source of the internal timer of the MCU is a crystal oscillator.

3. The distributed collaborative protection system based on spatiotemporal synchronization according to claim 1, characterized in that, The communication unit is a single-bus communication chip, which is connected to the general-purpose input / output pins of the microcontroller unit of the protection node. The data pins and power supply voltage pins of the single-bus communication chip are connected in parallel to the same physical bus. The protection node obtains operating power from the physical bus through the parasitic power supply method of the single-bus communication chip. The parasitic power supply method is as follows: the high-level signal on the physical bus charges the energy storage capacitor inside the protection node through the power supply voltage pin of the single-bus communication chip.

4. The distributed collaborative protection system based on spatiotemporal synchronization according to claim 1, characterized in that, The current detection unit includes a sampling resistor, which is connected in series in the loop of the protection node and is connected to the ADC pin of the MCU of the protection node.

5. The distributed collaborative protection system based on spatiotemporal synchronization according to claim 1, characterized in that, The current detection unit includes a current transformer, the primary coil of which is connected in series in the circuit of the protection node, and the secondary coil of which is connected to the ADC pin of the MCU of the protection node.

6. The distributed cooperative protection system based on spatiotemporal synchronization according to claim 1, characterized in that, The protection node also includes a non-volatile memory chip, which is connected to the MCU of the protection node via a bus; The MCU is configured to cache, in the form of a circular buffer, full current, voltage, temperature, and action timing data in the non-volatile memory chip, including the time before the fault, the fault process, and the full current, voltage, temperature, and action timing data after the fault recovery.

7. The distributed cooperative protection system based on spatiotemporal synchronization according to claim 1, characterized in that, The protection node also includes a non-volatile memory chip, which is connected to the MCU of the protection node via a bus; The MCU is configured to cache, in the form of a circular buffer, full current, voltage, temperature, and action timing data in the non-volatile memory chip, including the time before the fault, the fault process, and the full current, voltage, temperature, and action timing data after the fault recovery.

8. The distributed cooperative protection system based on spatiotemporal synchronization according to claim 1, characterized in that, The automatic fault location determination includes: Compare the timestamps of current surges reported by each node, and determine the node with the earliest current surge time as the node closest to the fault point; When the difference in timestamps of current surges among multiple nodes is less than a preset threshold, the magnitudes of current surges are further compared, and the node with the largest surge magnitude is determined to be the node closest to the fault point.

9. The distributed collaborative protection system based on spatiotemporal synchronization according to claim 1, characterized in that, The preset delay of the backup protection action is dynamically configured according to the topology hierarchy of the power distribution system. The delay time of the upper-level node is greater than the delay time of the lower-level node, and the delay time difference between adjacent level nodes is not less than the action time margin of the protection node.

10. A distributed collaborative protection method based on spatiotemporal synchronization, applied to the system described in any one of claims 1 to 10, characterized in that, Includes the following steps: S1. The edge controller establishes communication connections with multiple protection nodes and establishes time synchronization with each protection node, with a synchronization accuracy better than 100 microseconds; S2. Each protection node collects the current waveform data of its circuit in real time and reports the current data with timestamps to the edge controller. S3. The edge controller receives data reported by each node. When a fault current is detected, it automatically determines the fault location based on the timestamp and amplitude of the current change reported by each node. S4. The edge controller instructs the node closest to the fault point to perform protection actions first, and at the same time instructs the upstream node of that node to enter a backup waiting state. S5. If the node closest to the fault point successfully operates, the collaborative protection process ends; if the node closest to the fault point fails to operate within a preset delay, the superior node performs the backup protection action.

Images