Fault automatic tracing method, system, device and medium based on topology coding

Abstract

The application relates to the technical field of data processing, and specifically provides a fault automatic tracing method, system, device and medium based on topology coding, which comprises the following steps: allocating a unique topology code reflecting the hierarchical position of a physical device in a power distribution system of a factory; collecting real-time operation data through an edge collection terminal and binding the topology code to form a real-time data stream with a label; constructing a topology dependency model, defining the inherent failure rate of each node and the dependency on the upper node; when a target device failure is detected, the upstream nodes are traversed in reverse along the topology path based on the model, and the probability propagation algorithm is used to calculate the fault suspicion probability of each upstream node; the probability is normalized and fused with real-time state data for weighted correction, and finally a fault diagnosis report sorted by probability is generated. Through the introduction of a multilevel topology coding system and a probability propagation algorithm, the application realizes rapid and accurate positioning and intelligent diagnosis of faults.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, specifically relating to an automatic fault tracing method, system, device, and medium based on topology coding. Background Technology

[0002] In the field of factory power distribution operation and maintenance, existing technologies mainly fall into two modes: one relies on manual inspection, which is inefficient, costly, and lacks real-time response; the other uses industrial buses for local monitoring, but these systems are often isolated and can only provide simple threshold alarms. The core flaw of these methods lies in the lack of a unified digital mapping and binding of the multi-level topology of "distribution room-distribution box-equipment-meters". When a meter fails, the system cannot automatically and quickly trace its upstream root cause (such as distribution box tripping or communication interruption), causing fault location to rely entirely on the experience of maintenance personnel, resulting in lengthy troubleshooting and severely impacting production continuity. Furthermore, existing solutions have simplistic diagnostic logic, failing to integrate real-time data with historical equipment failure probabilities for comprehensive intelligent analysis, leading to a high false alarm rate and an inability to provide multi-possibility ranking diagnostic results. The accuracy and efficiency of operation and maintenance support urgently need improvement. Summary of the Invention

[0003] In view of the above-mentioned shortcomings of the prior art, the present invention provides a method, system, device and medium for automatic fault tracing based on topology coding to solve the above-mentioned technical problems.

[0004] In a first aspect, the present invention provides an automatic fault tracing method based on topological coding, comprising: Assign unique topology codes to multiple physical devices in the factory power distribution system. The topology codes are based on a multi-level device structure, and each topology code uniquely identifies a physical device and reflects its hierarchical position. The real-time operating data of the physical device is collected by the edge acquisition terminal, and corresponding topology codes are attached to the collected data to form a real-time data stream with topology tags. Construct a topology dependency model, defining the inherent failure rate and dependency on the parent node for each node; When a fault is detected in the target device based on the real-time data stream, the upstream nodes are traversed backward along the topology path starting from the target device based on the topology dependency model, and the probability of fault suspicion of the traversed upstream nodes is calculated based on the probability propagation algorithm. The probability of failure for all upstream nodes is normalized and weighted by incorporating real-time status data to generate a fault diagnosis report sorted by probability.

[0005] In one optional implementation, a unique topology code is assigned to multiple physical devices in the factory power distribution system, including: A segmented coding structure is adopted, which includes a power distribution room code segment, a floor or area code segment, a workshop equipment code segment and a branch meter code segment connected in sequence, with each code segment connected by a preset separator; The power distribution room code segment is used to uniquely identify a power distribution room, the floor or area code segment is used to identify the power supply area under the power distribution room, the workshop equipment code segment is used to identify the terminal electrical equipment under the power supply area, and the branch meter code segment is used to identify the energy meter or sensor installed on the power supply circuit of the electrical equipment.

[0006] In an optional implementation, real-time operational data of the physical device is collected via an edge acquisition terminal, and corresponding topology codes are appended to the collected data to form a real-time data stream with topology tags, including: Edge acquisition terminals are deployed at each level of the factory's power distribution room, and these terminals are connected to smart meters within their jurisdiction via an industrial bus network. Configure the edge acquisition terminal so that when it reads real-time operating data from the smart meter via Modbus-RTU, Modbus-TCP or DL / T645 industrial protocol, it actively encapsulates the unique topology code corresponding to the smart meter into the same data packet; The edge acquisition terminal transmits the encapsulated real-time data stream with topology tags to the central operation and maintenance support system via the factory intranet or private network. The generation method of real-time data stream with topology tags includes: the edge acquisition terminal is configured to actively bind the unique topology code of any physical device when it acquires real-time running data items, and add a timestamp to form a structured data record; each data record contains at least three core fields: device topology code, real-time running data value, and data acquisition timestamp.

[0007] In an optional implementation, a topology dependency model is constructed, defining an inherent failure rate and dependency on parent nodes for each node, including: Based on the hierarchical connection relationship defined by the unique topology code, a node network topology graph is generated; An inherent failure rate is preset for each node in the topology diagram, and its value is determined based on the device type and historical operation and maintenance data of the physical device corresponding to the node; For each pair of parent and child nodes with a direct connection in the topology graph, define and assign a dependency degree. The dependency degree is used to quantify the degree to which the normal operation of the child node depends on the direct parent node.

[0008] In one optional implementation, the inherent failure rate of the node is dynamically optimized and updated using a machine learning algorithm, specifically including the following steps: Record historical fault events, and the dataset of historical fault events shall include at least: fault phenomena, initial root cause diagnosis reports generated by the system, and the actual root causes confirmed by on-site operation and maintenance. The initial fault suspicion probability of each candidate node in the initial root cause diagnosis report is used as a feature, and the real root cause is used as a label to form a supervised learning training sample. Using the training samples, an optimization model is trained through a supervised learning algorithm. This optimization model learns the complex relationship between the initial diagnostic probability and the actual cause of the failure, and feeds this knowledge back to adjust the inherent failure rate benchmark for the corresponding equipment type.

[0009] In an optional implementation, when a fault is detected in the target device based on the real-time data stream, the upstream nodes are traversed backward along the topology path starting from the target device, based on the topology dependency model, and the probability of fault suspicion of the traversed upstream nodes is calculated based on the probability propagation algorithm, including: When the real-time data stream indicates a communication interruption or abnormal electrical parameters of a target device, it is determined that the target device has malfunctioned, and its initial fault suspicion probability is set to 1.0. Starting from the faulty target device, according to the connection relationship defined in the topology dependency model, all its direct and indirect upstream nodes are traversed in reverse order along the power supply or communication path. During the traversal, for each upstream node reached, its probability of being suspected of failure is calculated according to the probability propagation formula. The calculation formula for the probability propagation algorithm is as follows:

[0010] in, This indicates the probability of a fault in the upstream node. This indicates the probability of a downstream node being suspected of failure. This indicates the degree of dependence of downstream nodes on upstream nodes. This represents the inherent failure rate of the upstream node. This represents the inherent failure rate of the downstream node.

[0011] In one optional implementation, the probability of failure for all upstream nodes is normalized and weighted by incorporating real-time status data to generate a probability-ranked fault diagnosis report, including: The initial fault probability of all upstream nodes obtained by traversing and calculating is summed, and the fault probability of each node is divided by this sum to convert it into a normalized probability with a sum of 1. The real-time status data of each suspected node is queried, and the normalized probability of the node is dynamically weighted and adjusted according to preset rules. The real-time status data includes at least: the communication status of the node itself; the key electrical parameters of the node; and the operating status of the downstream associated equipment group of the node. Based on the weighted and corrected final fault probability, all suspected nodes are sorted in descending order, and a diagnostic report containing the most likely root cause and its corresponding probability is generated and output.

[0012] Secondly, the present invention provides an automatic fault tracing system based on topology coding, comprising: The topology coding module is used to assign unique topology codes to multiple physical devices in the factory power distribution system. The topology codes are based on a multi-level device structure, and each topology code uniquely identifies a physical device and reflects its hierarchical position. The data acquisition module is used to collect real-time operating data of the physical device through the edge acquisition terminal, and to attach corresponding topology codes when collecting data to form a real-time data stream with topology tags; The model building module is used to build a topology dependency model, defining the inherent failure rate and dependency on the parent node for each node; The probability calculation module is used to, when a fault is detected in the target device based on the real-time data stream, traverse upstream nodes in reverse along the topological path starting from the target device based on the topological dependency model, and calculate the fault suspicion probability of the traversed upstream nodes based on the probability propagation algorithm. The report generation module is used to normalize the probability of failure for all upstream nodes and incorporate real-time status data for weighted correction, generating a fault diagnosis report sorted by probability.

[0013] Thirdly, a device is provided, comprising: Memory for storing automatic fault tracing programs based on topology coding; A processor is configured to implement the steps of the topology-based automatic fault tracing method provided in the first aspect when executing the topology-based automatic fault tracing program.

[0014] Fourthly, a computer-readable medium is provided, on which a topology-based automatic fault tracing program is stored, wherein when executed by a processor, the topology-based automatic fault tracing program implements the steps of the topology-based automatic fault tracing method provided in the first aspect.

[0015] The beneficial effects of this invention are as follows: The automatic fault tracing method, system, equipment, and medium based on topology coding provided by this invention achieve rapid and accurate fault location and intelligent diagnosis by introducing a multi-level topology coding system and a probability propagation algorithm. Its beneficial effects are: 1. Improved operation and maintenance efficiency: Through automatic tracing of topological relationships, fault location time is shortened from hours to minutes, significantly reducing downtime losses; 2. Enhanced system intelligence: By using a probabilistic model to fuse real-time data, diagnostic results are output in order of probability, guiding operation and maintenance personnel to troubleshoot according to priority, significantly reducing false alarm and missed alarm rates; 3. Optimized resource allocation: Accurate root cause judgment reduces unnecessary on-site inspection frequency, effectively reducing manpower and time costs. This system transforms traditional passive operation and maintenance into a high-availability mode combining proactive early warning and accurate diagnosis, providing strong intelligent operation and maintenance support for factory power distribution systems. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart of a method according to an embodiment of the present invention.

[0018] Figure 2 This is a schematic block diagram of a system according to an embodiment of the present invention.

[0019] Figure 3 This is a schematic diagram of the structure of a device provided in an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0022] The automatic fault tracing method based on topology coding provided in this embodiment of the invention is executed by a computer device, and correspondingly, the automatic fault tracing system based on topology coding runs in the computer device.

[0023] Figure 1 This is a schematic flowchart illustrating a method according to an embodiment of the present invention. Wherein, Figure 1 The executing entity can be an automated fault tracing system based on topology coding. Depending on different requirements, the order of the steps in this flowchart can be changed, and some can be omitted.

[0024] like Figure 1 As shown, the method includes: S1. Assign a unique topology code to multiple physical devices in the factory power distribution system. The topology code is based on a multi-level device structure, and each topology code uniquely identifies a physical device and reflects its hierarchical position. S2. Collect real-time operating data of the physical device through the edge acquisition terminal, and attach corresponding topology codes when collecting data to form a real-time data stream with topology tags; S3. Construct a topology dependency model, defining the inherent failure rate and dependency on the parent node for each node; S4. When a fault is detected in the target device based on the real-time data stream, based on the topology dependency model, the upstream nodes are traversed backward along the topology path starting from the target device, and the fault suspicion probability of the traversed upstream nodes is calculated based on the probability propagation algorithm. S5. Normalize the probability of failure for all upstream nodes and incorporate real-time status data for weighted correction to generate a fault diagnosis report sorted by probability.

[0025] In one embodiment of the present invention, based on step S1, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0026] A segmented coding structure is adopted, consisting of four code segments connected sequentially: power distribution room code segment, floor or area code segment, workshop equipment code segment, and branch meter code segment. Each code segment is connected by a hyphen "-" as a preset separator to ensure overall readability and system parsability.

[0027] Each code segment serves a specific identification function: The power distribution room code segment is used to uniquely identify a power distribution room or power distribution center within the factory. For example, it can be in the format of "2 letters + 2 numbers", where the letters represent functional areas (e.g., DF represents the main power distribution room, MF represents the manufacturing workshop power distribution room), and the numbers represent serial numbers (e.g., 01, 02).

[0028] Floor or area code segments are used to identify the specific power supply area under the power distribution room, which can be a physical floor or a logical functional area. For example, the format is "1 letter L + 2 numbers", such as L01 representing the first floor and L02 representing the second floor.

[0029] The workshop equipment code segment is used to uniquely identify a terminal electrical device within the power supply area. For example, the format is "1 letter E + 3 numbers", such as E101 representing equipment number 1 on production line A (such as an injection molding machine).

[0030] The branch meter code segment is used to uniquely identify a specific energy meter or sensor installed on the power supply circuit of the electrical equipment. For example, the format is "1 letter M + 3 numbers", such as E101-M001 representing the total energy consumption meter of the injection molding machine, and E101-M002 representing a dedicated branch meter for its heating module.

[0031] Recommended implementation process: First, inventory all power distribution assets in the plant, clarifying their physical locations and electrical connections. Then, generate a unique code for each device according to the aforementioned rules, and manage them uniformly using a database. Finally, convert the codes into QR codes or RFID tags, affix them to the corresponding devices, and import them into this operation and maintenance support system to complete the initial configuration of the topology model.

[0032] In one embodiment of the present invention, based on step S2, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0033] First, hardware deployment and networking are carried out. Edge acquisition terminals are deployed at key nodes such as factory power distribution rooms and floor distribution boxes. These edge acquisition terminals, such as common TTUs, are physically connected to multiple smart meters, sensors, and other physical devices within their jurisdiction via industrial bus networks (such as RS-485 buses) to establish reliable data acquisition channels.

[0034] Secondly, terminal software configuration and data encapsulation are performed. The edge acquisition terminal is configured to perform a core data binding operation when it cyclically reads real-time operating data such as voltage, current, and energy consumption from smart meters via industrial protocols such as Modbus-RTU, Modbus-TCP, or DL / T645. Specifically, after issuing a data read request or receiving a response, the terminal actively queries its local configuration for the unique topology code corresponding to the smart meter and encapsulates this code along with the read real-time operating data value in the same uplink data packet. This process completes the binding of data and identity at the source of data generation.

[0035] Subsequently, data stream transmission takes place. The edge acquisition terminal continuously transmits the encapsulated, topology-tagged real-time data stream to the backend central operation and maintenance support system via the factory's wired local area network or industrial wireless private network for subsequent storage, analysis, and diagnosis.

[0036] The core technology for generating the real-time data stream with topology tags lies in constructing structured data records. The edge acquisition terminal is programmed to immediately create a structured data record whenever a data item (such as phase A current) from any physical device is successfully acquired. This record contains at least the following three core fields: Device topology code: Uniquely identifies the device from which the data originates, such as "MF01-L02-E101-M001".

[0037] Real-time running data value: specific measurement value, such as "225.6" (unit: A).

[0038] Data collection timestamp: Precisely records the time of data collection, such as "2023-10-27 14:05:30.550".

[0039] Through the above implementation methods, it is ensured from the source of data generation that each piece of operational data accurately carries its "location" and "time" information in the factory's power distribution network, providing a high-quality and highly reliable data foundation for subsequent automatic fault tracing and intelligent analysis based on precise topology.

[0040] In one embodiment of the present invention, based on step S3, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0041] S301, Construction of Topological Dependency Model First, based on the unique topology code and its hierarchical relationship defined in the early stage, the system constructs a node network topology graph in memory. This graph is stored in the form of graph structure data, where nodes represent various physical devices (such as power distribution rooms, distribution boxes, equipment, meters), and edges represent the electrical connections and communication relationships between them.

[0042] Subsequently, an inherent failure rate (λ) is preset for each node in the topology graph. This value is determined based on two types of information: Device Type: Set a base failure rate for different types of devices. For example, the inherent failure rate of a communication module (e.g., 0.05) is higher than that of a smart meter (e.g., 0.02), and the failure rate of a meter is higher than that of an air switch (e.g., 0.01).

[0043] Historical maintenance data: By referencing historical records such as the number of equipment failures and mean time between failures (MTBF), the baseline failure rate is fine-tuned. This process can be achieved through pre-configured data tables.

[0044] Simultaneously, a dependency (D) is defined for each pair of directly connected parent and child nodes in the diagram. This value, between 0 and 1, quantifies the degree to which the normal operation of a lower-level node depends on its direct parent node. In typical power supply relationships, the dependency is usually preset to 1.0. For example, if meter M001 is completely dependent on the power supply of its direct device E101, then D(M001 -> E101) = 1.0.

[0045] S302, Dynamic Optimization of Model Parameters To enable the model to have self-learning and evolutionary capabilities, the system dynamically optimizes the inherent failure rate of nodes through the following steps: Historical data collection: The system continuously runs and records all historical failure events, forming a structured dataset. Each record contains three key pieces of information: Fault symptoms: such as "Communication interruption of meter DF01-L02-E101-M001".

[0046] Initial diagnostic report: A list generated by the algorithm at that time, containing each candidate node and its initial probability of being suspected of failure.

[0047] Confirmed root cause: The cause of the fault is finally verified by the on-site maintenance personnel and entered into the system, such as "fault of communication module L02 in the floor distribution box".

[0048] Training sample construction: The data processing module extracts information from the historical dataset, takes the initial diagnosis report of each fault as the input feature (i.e., the suspicion probability of each candidate node at that time), and takes the real root cause node corresponding to the fault as the target label to form a training sample for supervised learning.

[0049] Model Training and Parameter Updates: The system periodically (e.g., monthly) uses accumulated training samples to train an optimized model through a supervised learning algorithm such as logistic regression or gradient boosting decision trees. The learning objective of this optimized model is to find the deep correlation between the initial diagnostic probability and the true cause of the failure. After training, the knowledge output by the model (such as feature importance) is converted into adjustments to the baseline values ​​of the inherent failure rates of various types of equipment. For example, if the model finds that the probability of suspicion for the "communication module" is often underestimated in historical diagnoses, the system will automatically increase the preset value of the inherent failure rate for all similar communication modules.

[0050] Through the above implementation methods, the topology dependency model evolves from a static pre-configured model into a dynamic intelligent system that can continuously learn from actual operation and maintenance experience and continuously improve diagnostic accuracy, providing core reasoning capabilities for achieving accurate automatic fault tracing.

[0051] In one embodiment of the present invention, based on step S4, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0052] S401. Fault Detection and Initialization The system continuously monitors real-time data streams from edge acquisition terminals. When the system detects that the communication heartbeat signal loss of a target device (such as meter DF01-L02-E101-M001) exceeds a preset threshold (e.g., 3 acquisition cycles) for an extended period, or that its electrical parameters (e.g., voltage value of 0) remain abnormal, it determines that the target device has malfunctioned. The system then creates a fault diagnosis task and sets the initial fault suspicion probability of the target device to P(target device) = 1.0.

[0053] S402. Reverse Topological Traversal The system starts with the faulty device as the initial node and, based on the hierarchical connections defined in the topology dependency model, traverses all its upstream nodes in reverse order along the power supply or communication path. For example, starting with the faulty meter M001, it first traverses to its direct superior device E101, then continues to the floor distribution box L02 where that device is located, and finally reaches the top-level distribution room DF01, forming a complete reverse diagnostic path. The algorithm traverses all possible paths to ensure that no potential fault source is missed.

[0054] S403. Probability Propagation Calculation During the traversal, for each upstream node reached, the system invokes the probability propagation algorithm to calculate its probability of being suspected of failure. Specifically, the following calculation formula is used:

[0055] in, This indicates the probability of a fault in the upstream node. This indicates the probability of a downstream node being suspected of failure. This indicates the degree of dependence of downstream nodes on upstream nodes. This represents the inherent failure rate of the upstream node. This represents the inherent failure rate of the downstream node.

[0056] Calculation example: Assumptions: The initial probability of faulty meter M001 is P(M001) = 1.0, the dependence of M001 on its upstream device E101 is D(M001→E101) = 1.0, the inherent failure rate of device E101 is λ_E101 = 0.03, and the inherent failure rate of meter M001 is λ_M001 = 0.02.

[0057] The probability of suspicion for device E101 is calculated as follows: P(E101)=1.0×1.0×(0.03 / 0.02)=1.5 The system calculates upstream according to this formula, recording the suspicion probability of each upstream node. All calculations are performed during graph traversal in memory, ensuring efficient real-time diagnostic performance.

[0058] In one embodiment of the present invention, based on step S5, a possible embodiment will be given below, and its specific implementation will be described in a non-limiting manner.

[0059] S501. Probability Normalization Process After calculating the initial suspected fault probabilities for all upstream nodes, the system performs a normalization process. Specifically, the initial suspected fault probabilities of all upstream nodes are summed to obtain a total probability value S = ΣP_i, where P_i is the initial suspected fault probability of the i-th node. Then, the suspected fault probability of each node is divided by this sum, i.e., P_i' = P_i / S, thus converting the probabilities of all nodes into normalized probabilities with a sum of 1. This mathematical processing makes the suspected fault levels of each node comparable, laying the foundation for subsequent priority ranking.

[0060] S502. Real-time data weighting correction The system then queries the real-time status data of each suspected node and dynamically adjusts the normalized probability according to preset weighting rules: Communication status check: If the suspect node's own communication module is online and its heartbeat is normal, its probability is multiplied by a weight of 0.1; if communication is interrupted, the probability value remains unchanged or is appropriately increased.

[0061] Electrical parameter verification: Check the key electrical parameters of the suspected node, such as the output current of the main switch. If the parameter is normal (e.g., there is current output), multiply the probability by 0.3; if there is no output current, multiply the probability value.

[0062] Downstream correlation analysis: Check the operating status of downstream related device groups of the suspected node. If a "one-down, all-down" phenomenon occurs (i.e., all downstream devices of the node go offline at the same time), the probability value of that node is significantly increased (e.g., multiplied by 5 times the weight); if only a few devices go offline, the probability value remains unchanged.

[0063] S503. Diagnostic Report Generation and Output The system sorts all suspected nodes in descending order based on the weighted and corrected final fault probability, generating a structured intelligent diagnostic report. The core content of the report includes: a list of the most likely root cause nodes sorted from highest to lowest probability; the final fault probability (displayed as a percentage) for each candidate root cause; and key supporting evidence for each conclusion, such as quantitative descriptions like "all devices under this distribution box are offline simultaneously" and "the node itself has communication interruption".

[0064] In some embodiments, the topology-based automatic fault tracing system may include multiple functional modules composed of computer program segments. The computer programs for each program segment in the topology-based automatic fault tracing system may be stored in the memory of a computer device and executed by at least one processor to perform (see details). Figure 1 (Description) The function of automatic fault tracing based on topology coding.

[0065] In this embodiment, the topology-based automatic fault tracing system can be divided into multiple functional modules according to its functions, such as... Figure 2 As shown. The module referred to in this invention is a series of computer program segments that can be executed by at least one processor and perform a fixed function, and is stored in memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0066] The topology coding module is used to assign unique topology codes to multiple physical devices in the factory power distribution system. The topology codes are based on a multi-level device structure, and each topology code uniquely identifies a physical device and reflects its hierarchical position. The data acquisition module is used to collect real-time operating data of the physical device through the edge acquisition terminal, and to attach corresponding topology codes when collecting data to form a real-time data stream with topology tags; The model building module is used to build a topology dependency model, defining the inherent failure rate and dependency on parent nodes for each node; The probability calculation module is used to, when a fault is detected in the target device based on the real-time data stream, traverse upstream nodes in reverse along the topological path starting from the target device based on the topological dependency model, and calculate the fault suspicion probability of the traversed upstream nodes based on the probability propagation algorithm. The report generation module is used to normalize the probability of failure for all upstream nodes and incorporate real-time status data for weighted correction, generating a fault diagnosis report sorted by probability.

[0067] Figure 3The topology-based automatic fault tracing method provided in this application embodiment can be applied to devices. Those skilled in the art will understand that the device structure involved in the embodiments of this invention does not constitute a limitation on the device. A device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of this invention, the device includes, but is not limited to, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of this application described and / or claimed herein.

[0068] The device 300 may include a processor 310, a memory 320, and a communication unit 330. These components communicate via one or more buses. Those skilled in the art will understand that the server structure shown in the figure does not constitute a limitation of the present invention. It may be a bus topology or a star topology, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0069] The memory 320 can be used to store execution instructions of the processor 310. The memory 320 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. When the execution instructions in the memory 320 are executed by the processor 310, the device 300 is able to perform some or all of the steps in the above method embodiments.

[0070] The processor 310 serves as the control center of the storage device, connecting various parts of the electronic device via various interfaces and lines. It executes software programs and / or modules stored in the memory 320, and calls data stored in the memory to perform various functions of the electronic device and / or process data. The processor can be composed of integrated circuits (ICs), such as a single packaged IC or multiple packaged ICs with the same or different functions connected together. For example, the processor 310 may consist only of a central processing unit (CPU). In this embodiment of the invention, the CPU may have a single processing core or include multiple processing cores.

[0071] The communication unit 330 is used to establish a communication channel, enabling the storage device to communicate with other devices. It can receive user data sent by other devices or send user data to other devices.

[0072] The present invention also provides a computer medium, wherein the computer medium may store a program, which, when executed, may include some or all of the steps provided in the embodiments of the present invention. The medium may be a magnetic disk, an optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0073] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a medium such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other medium capable of storing program code. It includes several instructions to cause a computer device (which may be a personal computer, a server, or a second device, network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0074] The same or similar parts between the various embodiments in this specification can be referred to mutually. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description in the method embodiments.

[0075] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or modules may be electrical, mechanical, or other forms.

[0076] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0077] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0078] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the present invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should also be covered within the protection scope of the present invention.

Claims

1. An automatic fault tracing method based on topological coding, characterized in that, include: Assign unique topology codes to multiple physical devices in the factory power distribution system. The topology codes are based on a multi-level device structure, and each topology code uniquely identifies a physical device and reflects its hierarchical position. The real-time operating data of the physical device is collected by the edge acquisition terminal, and corresponding topology codes are attached to the collected data to form a real-time data stream with topology tags. Construct a topology dependency model, defining the inherent failure rate and dependency on the parent node for each node; When a fault is detected in the target device based on the real-time data stream, the upstream nodes are traversed backward along the topology path starting from the target device based on the topology dependency model, and the probability of fault suspicion of the traversed upstream nodes is calculated based on the probability propagation algorithm. The probability of failure for all upstream nodes is normalized and weighted by incorporating real-time status data to generate a fault diagnosis report sorted by probability. When a fault is detected in the target device based on the real-time data stream, the upstream nodes are traversed backward along the topology path starting from the target device, based on the topology dependency model. The probability of fault suspicion for each traversed upstream node is calculated using a probability propagation algorithm, including: When the real-time data stream indicates a communication interruption or abnormal electrical parameters of a target device, it is determined that the target device has malfunctioned, and its initial fault suspicion probability is set to 1.

0. Starting from the faulty target device, according to the connection relationship defined in the topology dependency model, all its direct and indirect upstream nodes are traversed in reverse order along the power supply or communication path. During the traversal, for each upstream node reached, its probability of being suspected of failure is calculated according to the probability propagation formula. The calculation formula for the probability propagation algorithm is as follows: in, This indicates the probability of a fault in the upstream node. This indicates the probability of a downstream node being suspected of failure. This indicates the degree of dependence of downstream nodes on upstream nodes. This represents the inherent failure rate of the upstream node. This represents the inherent failure rate of the downstream node.

2. The method according to claim 1, characterized in that, Assign unique topology codes to multiple physical devices in the factory's power distribution system, including: A segmented coding structure is adopted, which includes a power distribution room code segment, a floor or area code segment, a workshop equipment code segment and a branch meter code segment connected in sequence, with each code segment connected by a preset separator; The power distribution room code segment is used to uniquely identify a power distribution room, the floor or area code segment is used to identify the power supply area under the power distribution room, the workshop equipment code segment is used to identify the terminal electrical equipment under the power supply area, and the branch meter code segment is used to identify the energy meter or sensor installed on the power supply circuit of the electrical equipment.

3. The method according to claim 1, characterized in that, The real-time operating data of the physical device is collected by an edge acquisition terminal, and corresponding topology codes are appended to the collected data to form a real-time data stream with topology tags, including: Edge acquisition terminals are deployed at each level of the factory's power distribution room, and these terminals are connected to smart meters within their jurisdiction via an industrial bus network. Configure the edge acquisition terminal so that when it reads real-time operating data from the smart meter via Modbus-RTU, Modbus-TCP or DL / T645 industrial protocol, it actively encapsulates the unique topology code corresponding to the smart meter into the same data packet; The edge acquisition terminal transmits the encapsulated real-time data stream with topology tags to the central operation and maintenance support system via the factory intranet or private network. The methods for generating real-time data streams with topology labels include: The edge acquisition terminal is configured to actively bind the unique topology code of any physical device when it acquires real-time operating data items, and add a timestamp to form a structured data record; each data record contains at least three core fields: device topology code, real-time operating data value, and data acquisition timestamp.

4. The method according to claim 1, characterized in that, Construct a topology dependency model, defining the inherent failure rate and dependency on parent nodes for each node, including: Based on the hierarchical connection relationship defined by the unique topology code, a node network topology graph is generated; An inherent failure rate is preset for each node in the topology diagram, and its value is determined based on the device type and historical operation and maintenance data of the physical device corresponding to the node; For each pair of parent and child nodes with a direct connection in the topology graph, define and assign a dependency degree. The dependency degree is used to quantify the degree to which the normal operation of the child node depends on the direct parent node.

5. The method according to claim 4, characterized in that, The inherent failure rate of nodes is dynamically optimized and updated using machine learning algorithms, specifically including the following steps: Record historical fault events, and the dataset of historical fault events shall include at least: fault phenomena, initial root cause diagnosis reports generated by the system, and the actual root causes confirmed by on-site operation and maintenance. The initial fault suspicion probability of each candidate node in the initial root cause diagnosis report is used as a feature, and the real root cause is used as a label to form a supervised learning training sample. Using the training samples, an optimization model is trained through a supervised learning algorithm. This optimization model learns the complex relationship between the initial diagnostic probability and the actual cause of the failure, and feeds this knowledge back to adjust the inherent failure rate benchmark for the corresponding equipment type.

6. The method according to claim 1, characterized in that, The probability of failure for all upstream nodes is normalized and weighted by incorporating real-time status data to generate a probability-ranked fault diagnosis report, including: The initial fault probability of all upstream nodes obtained by traversing and calculating is summed, and the fault probability of each node is divided by this sum to convert it into a normalized probability with a sum of 1. The real-time status data of each suspected node is queried, and the normalized probability of the node is dynamically weighted and adjusted according to preset rules. The real-time status data includes at least: the communication status of the node itself; the key electrical parameters of the node; and the operating status of the downstream associated equipment group of the node. Based on the weighted and corrected final fault probability, all suspected nodes are sorted in descending order, and a diagnostic report containing the most likely root cause and its corresponding probability is generated and output.

7. An automatic fault tracing system based on topology coding, characterized in that, include: The topology coding module is used to assign unique topology codes to multiple physical devices in the factory power distribution system. The topology codes are based on a multi-level device structure, and each topology code uniquely identifies a physical device and reflects its hierarchical position. The data acquisition module is used to collect real-time operating data of the physical device through the edge acquisition terminal, and to attach corresponding topology codes when collecting data to form a real-time data stream with topology tags; The model building module is used to build a topology dependency model, defining the inherent failure rate and dependency on the parent node for each node; The probability calculation module is used to, when a fault is detected in the target device based on the real-time data stream, traverse upstream nodes in reverse along the topological path starting from the target device based on the topological dependency model, and calculate the fault suspicion probability of the traversed upstream nodes based on the probability propagation algorithm. The report generation module is used to normalize the probability of failure for all upstream nodes and incorporate real-time status data for weighted correction, generating a fault diagnosis report sorted by probability. When a fault is detected in the target device based on the real-time data stream, the upstream nodes are traversed backward along the topology path starting from the target device, based on the topology dependency model. The probability of fault suspicion for each traversed upstream node is calculated using a probability propagation algorithm, including: When the real-time data stream indicates a communication interruption or abnormal electrical parameters of a target device, it is determined that the target device has malfunctioned, and its initial fault suspicion probability is set to 1.

0. Starting from the faulty target device, according to the connection relationship defined in the topology dependency model, all its direct and indirect upstream nodes are traversed in reverse order along the power supply or communication path. During the traversal, for each upstream node reached, its probability of being suspected of failure is calculated according to the probability propagation formula. The calculation formula for the probability propagation algorithm is as follows: in, This indicates the probability of a fault in the upstream node. This indicates the probability of a downstream node being suspected of failure. This indicates the degree of dependence of downstream nodes on upstream nodes. This represents the inherent failure rate of the upstream node. This represents the inherent failure rate of the downstream node.

8. An automatic fault tracing device based on topology coding, characterized in that, include: Memory for storing automatic fault tracing programs based on topology coding; A processor, configured to implement the steps of the topology-based automatic fault tracing method as described in any one of claims 1-6 when executing the topology-based automatic fault tracing program.

9. A computer-readable medium storing a computer program, characterized in that, The readable medium stores an automatic fault tracing program based on topology coding, which, when executed by a processor, implements the steps of the automatic fault tracing method based on topology coding as described in any one of claims 1-6.

Images