Ground wire locking device state management system and method based on real-time monitoring

Abstract

The invention relates to the technical field of electric power safety real-time monitoring, and particularly discloses a grounding wire locking device state management system and method based on real-time monitoring. The system comprises an edge sensing and control terminal, a region convergence and decision node and a centralized management platform. According to the invention, a three-level decision-making architecture combining edge terminal local logic, regional node intelligent decision-making and centralized platform strategy management is constructed. The edge terminals have independent locking capability to guarantee basic safety, the regional nodes realize rapid regional cooperation and intelligent research and judgment, and the centralized platform is responsible for global optimization and learning evolution. The distribution and concentration combined architecture not only ensures the quick response and autonomous capability under the local fault condition, but also realizes the optimal configuration of global resources and the continuous accumulation of system intelligence, and enhances the overall robustness and intelligent level of the system.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power safety real-time monitoring, and particularly relates to a grounding wire locking device state management system and method based on real-time monitoring. BACKGROUND

[0002] In the field of power system safety protection, the grounding wire locking device is a key equipment for ensuring the safety of maintenance personnel and preventing misoperation, and its reliability and real-time performance is directly related to the safety of power grid operation and personal safety. With the deep integration of smart grid and Internet of Things technology, higher requirements are put forward for the remote monitoring and management of power safety equipment.

[0003] The grounding wire locking device state management system based on real-time monitoring aims to realize centralized control of device state and intelligent execution of locking logic through remote data acquisition and instruction issuance. The system needs to ensure that device state information can be transmitted to the monitoring center in a timely and accurate manner, and quickly respond to locking or unlocking instructions.

[0004] In the prior art, the transmission of state data relies on traditional distributed monitoring networks and general communication protocols. These protocols are usually designed with redundancy and complex data encapsulation levels, which introduces significant data processing and communication delays when transmitting real-time state signals of the grounding wire locking device.

[0005] The existing system cannot meet the millisecond-level response requirements of locking state judgment and instruction execution in power safety operations, and the delay problem is particularly prominent in high-concurrency or network fluctuation scenarios, which may directly lead to the lag of safety protection actions and pose a serious safety hazard. Therefore, there is an urgent need for a system and method that can realize low-delay and high-reliability state management. SUMMARY

[0006] The technical problem to be solved by the present application is to overcome the defect that the existing grounding wire locking device state management system has significant data processing and communication delays due to the use of traditional distributed monitoring networks and general communication protocols, which cannot meet the millisecond-level response requirements of power safety operations and thus poses a safety hazard.

[0007] The present application provides a grounding wire locking device state management system and method based on real-time monitoring, aiming to realize extremely low-delay, high-reliability transmission of device state information and fast execution of intelligent locking logic.

[0008] The present application provides a grounding wire locking device state management system based on real-time monitoring, comprising: an edge perception and control terminal deployed on the grounding wire locking device body, used to collect physical state signals of the grounding wire locking device and execute local locking logic; A regional aggregation and decision node is arranged in a regional substation or distribution center, which is used to collect the status data of multiple edge perception and control terminals in the jurisdiction, make preliminary data processing and intelligent decision, and generate lockout control instructions; A centralized management and display platform is arranged in a monitoring master station, which is used for global state monitoring, historical data storage and analysis, and issuing advanced management strategies to the regional aggregation and decision node; The edge perception and control terminal, the regional aggregation and decision node, and the centralized management and display platform use a lightweight real-time communication protocol stack designed for state monitoring and emergency control to transmit data.

[0009] Preferably, the edge perception and control terminal comprises: A high-precision state sensor is used to monitor the mechanical locking state of the grounding hook, the continuity of the conductive loop, and the action position of the electromagnetic lock mechanism in real time, and convert the analog signal into a digital signal; A microprocessor unit is built-in with a state signal preprocessing algorithm, which is used to filter and denoise the sensor digital signal and extract features to generate a standardized device state word; A local lockout logic execution unit stores a pre-configured lockout rule library, which is used to drive the electromagnetic lock mechanism to perform lockout or unlock operation according to the received state word and instructions from the regional aggregation and decision node; A protocol adaptation module is used to encapsulate the standardized device state word into a data frame format defined by the lightweight real-time communication protocol stack and send it to the regional aggregation and decision node, while parsing the control instruction data frame from the regional aggregation and decision node.

[0010] Preferably, the pre-configured lockout rule library of the local lockout logic execution unit contains multiple lockout condition judgment logics: The first judgment is based on the device's own state word, if the hook is not completely locked or the conductive loop resistance value exceeds the safety threshold, the local forced lockout is triggered immediately; The second judgment receives remote instructions issued by the regional aggregation and decision node, which has a higher priority than the local judgment; The third judgment involves a timeout mechanism, if no heartbeat signal or state confirmation instruction is received from the regional node within a preset time window, the preset safety strategy is executed automatically, and the lockout state is entered by default.

[0011] Preferably, the regional aggregation and decision node comprises: A multi-channel data acquisition interface is used to concurrently receive the status data frames uploaded from all edge perception and control terminals in the jurisdiction; An edge computing unit running a data fusion algorithm for time stamp alignment and spatial correlation verification of the received multiple terminal state data, to generate a regional comprehensive state evaluation result; A real-time decision engine loading an intelligent decision model issued by the centralized management and display platform, for researching and judging the locking strategy and generating a control instruction draft within 5 milliseconds based on the comprehensive state evaluation result, the power grid topology connection relationship and the preset safety rules; An instruction distribution module for encapsulating the control instruction draft into an instruction frame defined by a lightweight real-time communication protocol stack, and issuing it to the corresponding edge-aware and control terminal.

[0012] Preferably, the intelligent decision model loaded by the real-time decision engine is a hybrid model based on rules and real-time state reasoning: The hybrid model first matches a pre-defined safety rule base, which defines the safety constraint conditions for grounding line operation under various power grid operating modes; If the rule matching is successful, the corresponding control instruction is directly outputted; If a complex or conflicting working condition not covered by the rule base is encountered, the model starts a real-time state reasoning machine, which calculates the optimal locking strategy through a logical reasoning chain based on the current power grid topology, power flow distribution and adjacent device state.

[0013] Preferably, the lightweight real-time communication protocol stack adopts a layered structure design, including: A physical link layer selecting a low-delay wired Ethernet as the physical medium; A simplified transport layer adopting a fixed-length 64-byte data packet structure and a connectionless datagram transmission mode, each data packet containing a sequence number and a time stamp, and ensuring data integrity through forward error correction coding; An application semantic layer defining a set of message types dedicated to grounding line locking state management, including state reporting messages, emergency control messages, heartbeat maintenance messages and parameter configuration messages, each message type having a fixed data structure and parsing rules.

[0014] Preferably, the data fusion algorithm executed by the edge computing unit has the following execution process: Compensate the time stamps of all received state words based on the system clock of the regional node, so that they are unified to the same time coordinate system; Based on the stored power grid topology connection relationship graph, analyze whether the states of electrically connected devices under the same time slice are logically consistent; If a state contradiction is detected, generate a correlation verification alarm event and mark it in the comprehensive state evaluation result.

[0015] Preferably, the state signal preprocessing algorithm executed by the microprocessor unit comprises: Digital filtering is performed using a fourth-order Butterworth low-pass filter; Feature extraction is performed on the filtered signal, and the effective value, peak value and change trend slope of the signal in each sampling period are calculated; The feature values are combined into a 16-bit standardized device state word according to a predefined bit field mapping rule.

[0016] Preferably, the centralized management and display platform further comprises: A historical data analysis and self-learning module is configured to continuously record all state change events, control instruction execution results and system response times, and analyze historical operation data using a machine learning algorithm; The analysis results of the historical operation data include key operation characteristic indicators, potential risk pattern recognition results, system performance bottleneck prediction results and intelligent decision optimization correlation data; the analysis results of the historical operation data are used to periodically optimize the intelligent decision model parameters and rules of the regional node real-time decision engine, and adjust the configuration parameters of the lightweight real-time communication protocol stack.

[0017] The application also provides a grounding wire locking device state management method based on real-time monitoring, which uses the above-mentioned grounding wire locking device state management method based on real-time monitoring to manage the state of the grounding wire locking device. The method starts with the edge perception and control terminal periodically collecting device states and preprocessing to generate standardized state words. Then, the terminal uploads the state words to the regional aggregation and decision node through the lightweight real-time communication protocol stack.

[0018] After the regional node receives the data, data fusion and comprehensive state evaluation are performed, and a real-time decision engine is used to research and judge the locking strategy. The control instructions generated by the research and judgment are issued back to the corresponding edge terminal through the same protocol stack. The local locking logic execution unit of the edge terminal integrates local state judgment and remote instructions, and finally drives the execution mechanism to complete the locking or unlocking action. The total delay from state collection to instruction execution is strictly controlled within 20 milliseconds.

[0019] Compared with the prior art, the application has the following advantages: 1. The application reduces the encapsulation complexity and processing level of data transmission by designing and applying a lightweight real-time communication protocol stack optimized for state monitoring and emergency control, solves the problem of high delay caused by the introduction of existing general protocols, meets the stringent requirements of millisecond-level response for power safety operation, and greatly improves the real-time performance and reliability of the system.

[0020] 2. This invention constructs a three-tiered decision-making architecture that combines local logic on edge terminals, intelligent decision-making at regional nodes, and policy management on a centralized platform. Edge terminals possess independent locking capabilities to ensure basic security, regional nodes achieve rapid regional collaboration and intelligent analysis, and the centralized platform is responsible for global optimization and learning evolution.

[0021] 3. This architecture, which combines distributed and centralized approaches, ensures rapid response and autonomy in the event of local failures, while also achieving optimized allocation of global resources and continuous accumulation of system intelligence, thereby enhancing the overall robustness and intelligence of the system.

[0022] 4. The local interlocking logic execution unit of the present invention has a built-in multi-interlocking condition judgment mechanism and a timeout management strategy to ensure that the device can operate in accordance with preset safety principles under any network interruption, node failure or abnormal working conditions, providing multi-level safety redundancy.

[0023] 5. The real-time decision engine adopts a hybrid model of rules and reasoning, which balances decision-making efficiency with the ability to cope with complex working conditions, effectively preventing misoperations and omissions, and comprehensively improving the safety protection level of power maintenance operations. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall technical solution architecture of the grounding wire interlocking device status management system based on real-time monitoring proposed in this invention. Figure 2 This is a schematic diagram of the core principle framework of the lightweight real-time communication protocol stack in this invention; Figure 3 This is a logical flow diagram of the edge sensing and control terminal in this invention; Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow between regional aggregation and decision nodes in this invention; Figure 5 This is a schematic diagram comparing the interaction principle between the local locking logic execution unit and the real-time decision engine in this invention. Detailed Implementation

[0025] The real-time monitoring-based grounding wire interlocking device status management system aims to address the core requirement of millisecond-level response in power safety operations. Please refer to the attached document. Figure 1 To be continued Figure 5 The system's overall architecture consists of three core layers: an edge sensing and control terminal deployed on the grounding wire interlocking device itself; a regional aggregation and decision-making node set up in a regional substation or power distribution center; and a centralized management and display platform located at the monitoring master station.

[0026] These three layers interact with each other through a lightweight real-time communication protocol stack designed specifically for status monitoring and emergency control.

[0027] The edge sensing and control terminal acts as the system's nerve endings, directly interacting with the physical components of the grounding interlocking device. Please refer to the appendix. Figure 3 The terminal integrates four key sub-modules: a high-precision status sensor, a microprocessor unit, a local interlocking logic execution unit, and a protocol adaptation module.

[0028] The high-precision status sensor is responsible for monitoring the mechanical locking status of the grounding wire hook in real time. It uses a combination of strain gauges and Hall elements to detect micron-level displacement changes at the contact surface between the hook and the grounding stake, and converts this mechanical deformation into an analog voltage signal.

[0029] Simultaneously, the sensor also monitors the continuity of the conductive circuit. By injecting a microampere-level detection current and measuring the circuit resistance, a continuity anomaly is determined when the resistance value exceeds a preset safety threshold of 2 ohms. In addition, the sensor accurately senses the movement position of the electromagnetic lock mechanism, using a linear variable differential transformer to detect the extension and retraction state of the bolt, ensuring a position detection accuracy of 0.1 mm.

[0030] All analog signals are converted into digital signals by a 24-bit high-precision analog-to-digital converter, with a sampling frequency fixed at 1000 times per second to ensure the real-time performance and accuracy of the status data.

[0031] The microprocessor unit receives the raw digital signal from the high-precision status sensor and executes a status signal preprocessing algorithm. This preprocessing algorithm first performs digital filtering using a fourth-order Butterworth low-pass filter with a cutoff frequency set to 50 Hz to effectively suppress high-frequency noise introduced by electromagnetic interference in the field. The filtered signal then enters the feature extraction stage, where the microprocessor calculates the effective value, peak value, and slope of the signal for each sampling period.

[0032] For example, for a mechanical locking status signal, an effective value below 1 volt and an absolute slope of less than 0.5 seconds are considered incompletely locked. After feature extraction, the microprocessor combines these feature values ​​into a 16-bit standardized device status word according to predefined bit-field mapping rules.

[0033] Bits 0 to 3 of the status word represent the mechanical locking status code, bits 4 to 7 represent the resistance level of the conductive circuit, bits 8 to 11 represent the electromagnetic lock position code, and bits 12 to 15 are reserved for fault diagnosis. The generation cycle of the status word is strictly controlled within 1 millisecond to ensure the timeliness of the status information.

[0034] The local interlocking logic execution unit is the core decision-making component of the edge terminal, and it stores a pre-configured interlocking rule base. This rule base contains three layers of interlocking condition judgment logic, forming a multi-layered security protection system. The first layer of judgment is based entirely on the device's own status word.

[0035] The standardized device status word generated by the microprocessor unit is sent to the local locking logic execution unit for parsing in real time. The parsing logic first checks the mechanical locking status code of bits 0 to 3. If the code value is not equal to the predefined full locking code 15, the local forced locking is immediately triggered.

[0036] Simultaneously, the parsing logic checks the resistance level of the conductive loops in bits 4 to 7. If the actual resistance corresponding to the level value exceeds the safety threshold of 2 ohms, a local forced latch will be triggered immediately. This first-level judgment has an extremely short response time; the delay from status word update to latch logic output is no more than 1 millisecond, ensuring an instantaneous safe response when a local physical anomaly is detected.

[0037] The second layer of logic handles remote control commands from regional aggregation and decision-making nodes. Upon receiving a data frame from a regional node, the protocol adaptation module first performs frame check sequence verification to ensure data integrity. After successful verification, the protocol adaptation module parses the application semantic layer of the data frame and extracts the control command code. This command code contains a 2-bit priority field and a 6-bit opcode field.

[0038] When parsing remote commands, the local locking logic execution unit compares the command's priority field. If the priority field value is 3, it indicates the highest priority emergency control command. In this case, regardless of the result of the first local judgment, the local locking logic execution unit will unconditionally execute the remote command and immediately drive the electromagnetic lock mechanism to perform locking or unlocking operations.

[0039] This design enables regional nodes to transcend local logic when necessary, achieving collaborative security control across devices, such as forcibly blocking relevant grounding wires when a change in the regional power grid topology is detected.

[0040] The third layer of judgment logic involves the reliability guarantee mechanism of the communication link, namely the timeout management strategy. The local interlocking logic execution unit maintains a watchdog timer, which is synchronized with the heartbeat maintenance messages of the regional aggregation and decision nodes. Regional nodes periodically send heartbeat maintenance messages to all subordinate edge terminals through a lightweight real-time communication protocol stack, with a preset heartbeat period of 1000 milliseconds. The local interlocking logic execution unit resets its watchdog timer each time it receives a valid heartbeat message.

[0041] If the watchdog timer accumulates more than 1500 milliseconds without receiving a new heartbeat message, it is determined that the communication link with the regional node has been interrupted or the regional node itself has malfunctioned. At this time, the third judgment logic is activated, the local latching logic execution unit will automatically execute the preset security policy, forcibly enter the latching state by default, and illuminate the local fault indicator light.

[0042] This timeout mechanism ensures that the device can still adhere to the most conservative security principles in the event of network anomalies or failure of the upstream node, preventing the loss of security functions due to communication interruption.

[0043] The protocol adaptation module acts as a bridge between the edge terminal and external communication; its core function is to convert data formats and transmission protocols. Please refer to the appendix. Figure 2 This module strictly adheres to the specifications of a lightweight real-time communication protocol stack. In the data transmission direction, the protocol adaptation module receives standardized device status words from the microprocessor unit and encapsulates them into status reporting messages defined by the protocol stack.

[0044] The encapsulation process first constructs a message header at the application semantic layer. The message header contains a 2-byte message type code, and the type code for status reporting messages is fixed at 0x01.

[0045] Following the message header is a 16-bit device status word payload. Then, the data enters the simplified transport layer, where a sequence number and timestamp are added. The sequence number is a 16-bit circular counter that increments by 1 with each transmission, used by the receiver to detect packet loss. The timestamp is taken from the terminal's internal high-precision clock, with an accuracy of 1 microsecond, and is used for subsequent data fusion and time alignment.

[0046] Finally, the physical link layer adds forward error correction coding to the data packets, using Reed-Solomon coding, which can correct burst errors of up to 2 bytes. The entire encapsulation process is optimized to take less than 0.5 milliseconds.

[0047] In the data receiving direction, the protocol adaptation module performs the reverse process, parsing the control instruction data frames from the regional nodes, extracting the operation instructions from the application semantic layer, and passing them to the local locking logic execution unit.

[0048] Regional aggregation and decision-making nodes, as the intermediate layer of the system, undertake the key tasks of data aggregation and intelligent decision-making. Please refer to the appendix. Figure 4 The node comprises four core components: a multi-channel data acquisition interface, an edge computing unit, a real-time decision engine, and an instruction distribution module.

[0049] Multi-channel data acquisition interfaces typically have at least 32 independent communication channels, capable of concurrently receiving status data frames uploaded from all edge sensing and control terminals within the jurisdiction. Each communication channel is equipped with an independent protocol parsing hardware accelerator, enabling parallel processing of input data frames.

[0050] The processing includes forward error correction decoding at the physical link layer, sequence number verification and timestamp extraction at the simplified transport layer, and message type identification and payload extraction at the application semantic layer. Multiple data acquisition interfaces ensure that even under high-concurrency data inflow, the processing latency for each data frame does not exceed 1 millisecond, providing a data foundation for subsequent real-time decision-making.

[0051] The edge computing unit receives standardized device status words and their corresponding timestamps parsed from the multi-channel data acquisition interface. This unit runs a data fusion algorithm; its core task is to integrate and verify the status data of multiple terminals within its jurisdiction.

[0052] The data fusion algorithm first performs timestamp alignment. Because the built-in clocks of various edge terminals have microsecond-level deviations, the edge computing unit uses the system clock of the regional node as a reference to compensate for the timestamps of all received status words, unifying them to the same time coordinate system. The aligned status data then enters the spatial correlation verification stage.

[0053] The edge computing unit stores a diagram of the local power grid topology, which defines the electrical connections and logical relationships between various grounding interlocking devices. The verification logic analyzes whether the logical states of electrically connected devices are consistent within the same time slice.

[0054] For example, if the topology diagram shows two grounding devices connected in series at the same disconnect point, theoretically, the status words of these two devices should simultaneously indicate either locked or unlocked. If a status contradiction is detected, a correlation verification alarm event is generated and marked in the comprehensive status assessment result.

[0055] Finally, the edge computing unit outputs a region-level comprehensive status assessment result, which is a structure containing the calibrated status word of each device, the associated status flags between devices, and any alarm codes for detected anomalies. The execution time of the entire data fusion process is strictly controlled within 3 milliseconds.

[0056] The real-time decision engine is the intelligent core of the regional nodes, loaded with intelligent decision-making models distributed by the centralized management and display platform. Please refer to the appendix. Figure 5 This model is a hybrid model based on rules and real-time state reasoning, designed to balance decision-making speed with the ability to handle complex operating conditions.

[0057] The model first matches itself against a predefined security rule base. This rule base is a collection of rules maintained and periodically updated by a centralized management platform; rules are defined in the form of "if condition then action".

[0058] For example, a rule might be defined as sending a forced blocking command to all relevant grounding devices if the bus disconnector is detected to be in the open position and any grounding device on that bus is not locked. The rule engine uses the Rete algorithm for fast pattern matching, capable of scanning and matching all applicable rules within 1 millisecond. If a rule match is successful, the corresponding draft control command is directly output.

[0059] If the rule engine does not find a rule in the rule base that perfectly matches the current comprehensive state evaluation result, or if it matches multiple potentially conflicting rules, the model will activate the real-time state inference engine. The real-time state inference engine is a computational module based on logical reasoning chains.

[0060] It receives the current power grid topology, real-time power flow distribution data, and status information of adjacent substation equipment as inference context. The inference process is based on first-order logic predicates, and calculates the optimal blocking strategy through a series of logical derivation steps.

[0061] For example, when encountering a complex switching operation sequence not defined in the rule base, the inference engine may execute the following derivation chain: if line L1 is de-energized for maintenance and grounding wire E1 is connected, but the adjacent line L2 is still energized and the electrical distance between grounding wire E1 and L2 is less than the safe distance, then it is determined that there is a risk of induced electricity. Therefore, it is necessary to ensure that grounding wire E1 is reliably locked and it is recommended to install a temporary grounding wire on the L2 side.

[0062] The design goal of the real-time state inference engine is to complete such complex reasoning in a maximum of 3 milliseconds, ensuring that the total time from receiving the comprehensive state assessment results to generating the draft control instructions of the entire real-time decision engine does not exceed 5 milliseconds.

[0063] The command distribution module is responsible for converting the draft control commands generated by the real-time decision engine into command frames that can be transmitted over the network. This module first assigns a unique command identifier to each draft command and marks its priority. Then, the command distribution module constructs an emergency control message according to the specifications of a lightweight real-time communication protocol stack.

[0064] In the application semantic layer, the message type code is set to 0x02, indicating an emergency control message. The message payload contains the identifier of the target edge terminal, the opcode to be executed, and the expiration timestamp of the instruction. The opcode is a 6-bit field that defines in detail the specific action parameters for locking or unlocking, such as the magnitude of the driving current and the holding time of the electromagnetic lock. The constructed application layer message is then sent to the simplified transport layer, where a sequence number and timestamp are added.

[0065] Finally, after adding forward error correction coding at the physical link layer, the command frame is sent to the corresponding edge sensing and control terminal through the designated communication channel. The command distribution module ensures that the construction and sending latency of the command frame is less than 1 millisecond.

[0066] The centralized management and display platform is the highest level of the system, responsible for global monitoring, data persistence, and system optimization. This platform maintains communication with various regional aggregation and decision-making nodes through a lightweight real-time communication protocol stack, receiving comprehensive status assessment results, control command execution logs, and system performance metrics reported by each region.

[0067] One of the platform's core functional modules is the historical data analysis and self-learning module. This module continuously records all state change events, control command execution results, and system response times. The data is stored in a high-performance time-series database with a storage period of no less than 3 years. The self-learning module utilizes machine learning algorithms to analyze massive amounts of historical operational data.

[0068] The analysis process begins with feature engineering, extracting key features from historical data, such as average response latency under specific operation sequences, rule triggering frequency under specific operating conditions, and the spatiotemporal distribution pattern of communication error rates. Subsequently, clustering algorithms are used to identify potential risk patterns. For example, it was discovered that a certain switching operation sequence could cause multiple regional nodes to initiate high-priority blocking commands almost simultaneously, potentially triggering instantaneous network congestion.

[0069] In addition, regression analysis models can be used to predict system performance bottlenecks. For example, based on historical data, it can be predicted that after a 15% increase in grid load, existing communication bandwidth may become a latency bottleneck during peak hours.

[0070] The analysis results of historical operating data include: Key operational characteristic indicators cover the spatiotemporal distribution patterns of average system response latency, maximum response latency, control command execution success rate, rule base matching hit rate, communication error rate, and frequency of device status anomaly triggering under different operating scenarios. The results of potential risk pattern identification are used to mine working condition combinations, operation sequences and network fluctuation characteristics that are prone to safety hazards through clustering algorithms, such as the congestion risk pattern of multiple regional nodes issuing high-priority interlocking commands concurrently under a specific switching operation process. The system performance bottleneck prediction results, based on the regression analysis model, obtain the changes in communication bandwidth, computing resource load threshold and decision delay under different power grid load growth trends, and identify the key links that may restrict the real-time performance of the system in the future. Intelligent decision optimization is based on related data, including statistics on complex working conditions not covered by the existing rule base, evaluation results of the effectiveness of the reasoning path of the real-time state inference engine, and the weights of the impact of different decision parameters on the interlocking response speed and security.

[0071] These analytical results are transformed into optimization strategies, which regularly update the intelligent decision model parameters and rules issued to the real-time decision engine of regional nodes, and dynamically adjust the configuration parameters of the lightweight real-time communication protocol stack, such as the heartbeat cycle or the redundancy of forward error correction coding, thereby achieving continuous evolution and improvement of the overall system performance and security.

[0072] A lightweight real-time communication protocol stack is the cornerstone for achieving low-latency communication in a system. Please refer to the appendix. Figure 2 This protocol stack employs a strictly layered architecture, with its core consisting of the physical link layer, the simplified transport layer, and the application semantic layer. The physical link layer prioritizes low-latency communication media.

[0073] In fixed locations where cabling can be laid, use 100 Mbps or Gigabit industrial Ethernet with Category 5e or Category 6 shielded twisted-pair cable as the physical medium and RJ45 connectors for the interface to ensure link latency is less than 1 millisecond.

[0074] The physical link layer is also responsible for signal modulation and demodulation and physical encoding and decoding, ensuring the signal's anti-interference capability in harsh industrial electromagnetic environments.

[0075] The simplified transport layer is a key innovation in the protocol stack, as it eliminates the complex handshake and retransmission mechanisms of traditional transport control protocols. This layer uses a fixed-length data packet structure, with all data packets, whether uplink status data or downlink control commands, having a uniform length of 64 bytes.

[0076] This fixed-length design greatly simplifies buffer management and data parsing logic at the receiving end. Data transmission uses a connectionless datagram model, similar to the User Datagram Protocol (UDP), but without relying on the complex routing of the Internet protocol suite. Each data packet contains two key fields: a sequence number and a timestamp.

[0077] The sequence number is a 16-bit unsigned integer, maintained by the sender and incremented with each data packet. The receiver detects packet loss by checking the continuity of the sequence numbers. The timestamp field is 32 bits wide and has a precision of 1 microsecond.

[0078] To ensure data integrity without retransmission, the simplified transport layer introduces a forward error correction coding mechanism, which uses Reed-Solomon coding. This mechanism can automatically detect and correct errors of up to 2 bytes in a 64-byte data packet, thus ensuring correct data reception without retransmission under most channel conditions.

[0079] The application semantic layer directly addresses the business requirements of grounding wire interlocking status management and defines a proprietary set of message types. This layer primarily includes four message types: status reporting messages, emergency control messages, heartbeat maintenance messages, and parameter configuration messages. Status reporting messages are used by edge terminals to periodically report device status words to regional nodes; their message type code is 0x01, and the data payload is a 16-bit standardized device status word.

[0080] Emergency control messages are used by regional nodes to issue latching or unlocking commands to edge terminals. Their message type code is 0x02, and the payload includes the target terminal identifier, operation code, and priority. Heartbeat maintenance messages, with a type code of 0x03, are used for communication link keep-alive and network quality detection. Their payload is typically empty or contains very little diagnostic information. Parameter configuration messages, with a type code of 0x04, are used by the centralized management platform or regional nodes to issue configuration parameters to edge terminals or the regional nodes themselves, such as updating the latching rule base or adjusting sensor thresholds.

[0081] Each message type has a strict and fixed data structure and parsing rules. Application layer entities do not need to perform complex length calculations or type inferences when assembling and unpacking packets, which greatly reduces the complexity of data encapsulation and parsing time, and compresses the application layer processing latency to the sub-millisecond level.

[0082] The system's workflow begins with the edge sensing and control terminal periodically collecting the physical status of the grounding wire interlocking device. High-precision status sensors sample at a frequency of 1000 times per second to acquire the mechanical locking status, conductive circuit connectivity, and electromagnetic lock position signals.

[0083] The microprocessor unit performs preprocessing algorithms on the raw signal, including digital filtering and feature extraction, and generates a standardized 16-bit device status word within 1 millisecond. The local latching logic execution unit immediately performs the first latching judgment based on this status word, and if a local anomaly is detected, it instantly triggers forced latching.

[0084] Meanwhile, the protocol adaptation module encapsulates the status word into a status reporting message defined by a lightweight real-time communication protocol stack, a process that takes 0.5 milliseconds. The encapsulated data frame is then sent to the regional aggregation and decision-making node via the physical link.

[0085] The regional aggregation and decision-making node's multi-channel data acquisition interface concurrently receives data frames from multiple edge terminals within its jurisdiction. The hardware accelerator within the interface parses these frames in parallel, performing forward error correction decoding, extracting sequence numbers and timestamps, and identifying application layer message types. This process takes approximately 1 millisecond.

[0086] The parsed status word and its timestamp are sent to the edge computing unit. The edge computing unit runs a data fusion algorithm to perform timestamp alignment and spatial correlation verification, generating a region-level comprehensive status assessment result. This process takes 3 milliseconds. The comprehensive status assessment result is then received by the real-time decision engine.

[0087] The real-time decision engine first activates the rules engine to match the security rules base. If a match is found within 1 millisecond, a draft control instruction is generated. If no match is found or a conflict exists, the real-time state inference engine is activated, which calculates the optimal strategy through a logical inference chain within a maximum of 3 milliseconds.

[0088] The total decision-making time of the real-time decision engine does not exceed 5 milliseconds. The generated draft instructions are encapsulated into an emergency control message by the instruction distribution module within 1 millisecond and sent back to the corresponding edge terminal through the protocol stack. The downlink transmission latency of the instruction frame is similar to that of the uplink, less than 1 millisecond for wired connections and less than 5 milliseconds for wireless connections.

[0089] After receiving the instruction frame, the protocol adaptation module of the edge terminal parses it within 0.5 milliseconds, extracts the control instruction, and passes it to the local locking logic execution unit. The local locking logic execution unit integrates the remote instruction with the local second and third-level judgment logic, and finally drives the electromagnetic lock mechanism to perform locking or unlocking operations.

[0090] The action time of the electromagnetic lock mechanism is typically between 2 and 5 milliseconds. Throughout the entire process, from the start of edge terminal status acquisition to the final execution of the actuator, the total latency is strictly controlled to within 20 milliseconds under worst-case communication conditions, and is far lower under normal conditions, fully meeting the stringent requirements of millisecond-level response for power safety operations.

Claims

1. A ground wire lockout device status management system based on real-time monitoring, characterized by, The application relates to a ground wire locking device and a state monitoring and emergency control system thereof. The application comprises: An edge perception and control terminal arranged at the edge of the body of the ground wire locking device, which is used for collecting physical state signals of the ground wire locking device and executing local locking logic; A regional aggregation and decision node arranged at a regional substation or distribution center, which is used for collecting state data of a plurality of edge perception and control terminals in a jurisdiction, performing preliminary data processing and intelligent decision-making, and generating locking control instructions; A centralized management and display platform located at a monitoring master station, which is used for global state monitoring, historical data storage and analysis, and issuing advanced management strategies to the regional aggregation and decision node; 2. The real-time monitoring based ground lock status management system of claim 1, wherein, A lightweight real-time communication protocol stack designed for state monitoring and emergency control is used for data transmission between the edge perception and control terminal, the regional aggregation and decision node and the centralized management and display platform. The edge perception and control terminal comprises: High-precision state sensors used for monitoring the mechanical locking state of a ground wire hook, the continuity of a conductive loop and the action position of an electromagnetic lock mechanism in real time, and converting analog signals into digital signals; A microprocessor unit internally provided with a state signal preprocessing algorithm, which is used for filtering and denoising the sensor digital signals and extracting features to generate a standardized device state word; A local locking logic execution unit storing a preconfigured locking rule library, which is used for independently driving the electromagnetic lock mechanism to perform locking or unlocking operations according to the received state word and instructions from the regional aggregation and decision node; 3. The real-time monitoring based ground lock status management system of claim 2, wherein, A protocol adaptation module used for encapsulating the standardized device state word into a data frame format defined by the lightweight real-time communication protocol stack and sending the data frame to the regional aggregation and decision node, and simultaneously analyzing control instruction data frames from the regional aggregation and decision node. The preconfigured locking rule library of the local locking logic execution unit comprises multiple locking condition judgment logics: The first judgment is based on the state word of the device itself, if the hook is not completely locked or the resistance value of the conductive loop exceeds a preset safety threshold, local forced locking is triggered immediately; The second judgment receives remote instructions issued by the regional aggregation and decision node, and the priority of the remote instructions is higher than that of the local judgment; 4. The real-time monitoring based ground lock status management system of claim 1, wherein, The third judgment involves a timeout mechanism, if a heartbeat signal or a state confirmation instruction is not received from the regional node within a preset time window, a preset safety strategy is automatically executed, and the locking state is entered by default. The regional aggregation and decision node comprises: A multi-channel data acquisition interface used for concurrently receiving state data frames uploaded from all edge perception and control terminals in the jurisdiction; An edge computing unit running a data fusion algorithm, which is used for time stamp alignment and spatial correlation verification of the received state data of a plurality of terminals to generate a comprehensive state evaluation result at the regional level; A real-time decision engine loading an intelligent decision-making model issued by the centralized management and display platform, which is used for completing the research and judgment of the locking strategy based on the comprehensive state evaluation result, the connection relationship of the power grid topology and the preset safety rules within 5 milliseconds, and generating a control instruction draft; An instruction distribution module used for encapsulating the control instruction draft into an instruction frame defined by the lightweight real-time communication protocol stack and issuing the instruction frame to the corresponding edge perception and control terminal.

5. The real-time monitoring based ground lock status management system of claim 4, wherein, The intelligent decision model loaded by the real-time decision engine is a hybrid model based on rules and real-time state reasoning: The hybrid model first matches a pre-defined safety rule base, which defines safety constraints for grounding line operation under various power grid operating modes; If the rule matching is successful, the corresponding control instruction is directly output; If a complex or conflicting working condition not covered by the rule base is encountered, the model starts a real-time state reasoning machine, which calculates the optimal locking strategy through a logical reasoning chain based on the current power grid topology, power flow distribution, and adjacent device state.

6. The real-time monitoring based ground wire lockout device status management system of claim 1, wherein, The lightweight real-time communication protocol stack adopts a layered structure design, including: Physical link layer, which selects low-latency wired Ethernet as the physical medium; Simplified transport layer, which adopts a fixed-length 64-byte packet structure and a connectionless datagram transmission mode, each packet contains a sequence number and a timestamp, and the data integrity is guaranteed through forward error correction coding; Application semantic layer, which defines a set of message types dedicated to grounding line locking state management, including state reporting messages, emergency control messages, heartbeat maintenance messages, and parameter configuration messages, each message type has a fixed data structure and parsing rules.

7. The real-time monitoring based ground wire lockout device status management system of claim 4, wherein, The data fusion algorithm executed by the edge computing unit executes the following process: Compensate all received state word timestamps based on the system clock of the regional node to unify them to the same time coordinate system; Based on the stored power grid topology connection graph, analyze whether the states of the devices electrically connected at the same time slice are logically consistent; If a state contradiction is detected, generate a correlation check alarm event and mark it in the comprehensive state evaluation result.

8. The real-time monitoring based ground wire lockout device status management system of claim 2, wherein, The state signal preprocessing algorithm executed by the microprocessor unit includes: Digital filtering using a fourth-order Butterworth low-pass filter; Feature extraction on the filtered signal, calculating the effective value, peak value, and change trend slope of the signal in each sampling period; Map the feature values according to the pre-defined bit field mapping rules to combine them into a 16-bit standardized device state word.

9. The real-time monitoring based ground wire lockout device status management system of claim 1, wherein, The centralized management and display platform also includes: A historical data analysis and self-learning module for continuously recording all state change events, control instruction execution results, and system response times, and analyzing historical operation data using machine learning algorithms; The analysis results of historical operation data include: key operating feature indicators, potential risk pattern recognition results, system performance bottleneck prediction results, and intelligent decision optimization correlation data; the analysis results of historical operation data are used to periodically optimize the intelligent decision model parameters and rules of the regional node real-time decision engine, and adjust the configuration parameters of the lightweight real-time communication protocol stack.

10. A method for managing the state of a ground wire locking device based on real-time monitoring, characterized by, The grounding line locking device state management based on real-time monitoring according to any one of claims 1-9 is used to manage the state of the grounding line locking device.

Images