A method, system, device, and medium for data link fault analysis

Abstract

The application discloses a kind of data link fault analysis method, system, equipment and medium, specifically related to fault diagnosis field, its technical points are in at:call data accuracy diagnostic algorithm is diagnosed to the accuracy of the link transmission data of abnormal data link, if accurate data is determined that the link transmission data is normal data, if inaccurate data is determined that the link transmission data is abnormal data, and corresponding abnormal data quantity is superimposed, when abnormal data quantity exceeds abnormal data quantity threshold, data accuracy diagnostic result is obtained, wherein, abnormal data quantity threshold is dynamically adjusted by pre-constructed threshold update model;After obtaining the data accuracy diagnostic result of data link, call link information consistency diagnostic algorithm is diagnosed to the link node configuration information consistency of the data abnormal data link, link node configuration diagnostic result is obtained, based on link node configuration diagnostic result, optimize data link node configuration.

Description

Technical Field

[0001] This invention relates to the field of fault diagnosis, and specifically to a data link fault analysis method, system, device, and medium. Background Technology

[0002] Real-time data from oil and gas production plays a crucial role. First, this data is fundamental to ensuring safe production in oil and gas fields. It provides critical real-time information to help monitor and control each stage of production, promptly identify and address potential safety risks and production anomalies, and guarantee the safe and stable operation of the production process. Second, this data is also an important resource for driving the digital transformation of oil and gas fields, providing a solid foundation for data-driven decision-making, intelligent management, and efficiency improvements.

[0003] The real-time production data collected by various sensors deployed at the well site needs to pass through two network systems, the production network and the office network, to reach the user application. The entire data transmission link involves various professional equipment or systems such as PLCs, switches, firewalls, SCADA host computers, OPC servers, and real-time databases, involving more than 10 equipment nodes. The diversity and dispersion of the equipment increases maintenance costs. Failure of any node will affect the accuracy and continuity of production data. Troubleshooting is difficult and requires a lot of time to troubleshoot step by step.

[0004] Therefore, the present invention aims to provide a data link fault analysis method, system, device and medium to solve the aforementioned related problems. Summary of the Invention

[0005] The technical problem this invention aims to solve is the difficulty in troubleshooting and quickly diagnosing faults when device nodes in a data link fail. The goal is to provide a data link fault analysis method, system, device, and medium. The data link fault analysis method provided by this invention enables real-time diagnosis of the operational status of various devices and systems in a data link. Through sub-algorithms such as idle link diagnosis, failed link diagnosis, device fault diagnosis, link information consistency diagnosis, and data accuracy diagnosis, it completes the analysis of the real-time status, data transmission rate, and fault conditions of various devices, achieving full-coverage diagnosis of the data link, quickly locating faults, significantly reducing maintenance costs, and improving system stability and data transmission reliability. Furthermore, this invention innovatively integrates the various sub-algorithms, realizing the intelligence, collaboration, and efficiency of the full-link fault diagnosis system, significantly improving the system's diagnostic capabilities, management efficiency, and data quality, and solving the problems of difficulty in troubleshooting and quickly diagnosing faults when device nodes in a data link fail in existing technologies.

[0006] This invention is achieved through the following technical solution:

[0007] A data link fault analysis method, comprising:

[0008] The idle link diagnosis algorithm is called to diagnose the idle state of the data link and obtain the idle state diagnosis result. After obtaining the link idle state diagnosis result, the failure link diagnosis algorithm is called to diagnose the failure state of the idle data link and obtain the failure state diagnosis result. After obtaining the failure state diagnosis result, the node fault diagnosis algorithm is called to diagnose the link node fault of the failure data link and obtain the node anomaly diagnosis result.

[0009] After obtaining the node anomaly diagnosis result, the data accuracy diagnosis algorithm is called to diagnose the accuracy of the data transmitted by the abnormal data link. If the data is accurate, the data transmitted by the link is determined to be normal data. If the data is not accurate, the data transmitted by the link is determined to be abnormal data, and the number of abnormal data is added accordingly. When the number of abnormal data exceeds the abnormal data number threshold, the data accuracy diagnosis result is obtained. The abnormal data number threshold is dynamically adjusted through a pre-built threshold update model.

[0010] After obtaining the data accuracy diagnosis results of the data link, the link information consistency diagnosis algorithm is called to diagnose the consistency of the link node configuration information of the data link with abnormal data, and the link node configuration diagnosis results are obtained. Based on the link node configuration diagnosis results, the data link node configuration is optimized.

[0011] Furthermore, the idle link diagnostic algorithm is invoked to diagnose the idle state of the data link, and the idle state diagnostic results are obtained, specifically:

[0012] Obtain the data transmission timestamp information for each data link, calculate the time interval between the data transmission timestamp information and the current time, and obtain the idle time data;

[0013] The idle time data is compared with a preset first time threshold. If the idle time data exceeds the first time threshold, the corresponding data link is determined to be an idle data link; otherwise, it is a non-idle data link.

[0014] Furthermore, the failure link diagnosis algorithm is invoked to diagnose the failure status of the idle data link, and the failure status diagnosis result is obtained, specifically:

[0015] By treating each system device with idle data links as a node and the data link relationships between each system device as edges, a data link topology is constructed.

[0016] The probe packets are continuously sent to each node in the data link topology. If a node reply signal is received within a preset second time threshold, the corresponding node is determined to be a normal node; otherwise, it is an abnormal node.

[0017] When there is only one data link between two nodes, if there is at least one abnormal node among the two nodes, the data link is determined to be in a failed state; otherwise, it is in a normal state.

[0018] When there are multiple data links between two nodes, if there is at least one abnormal node between the two nodes and at least one other data link between the two nodes is in a normal state, then the data link is determined to be in a failed state; otherwise, it is in a normal state.

[0019] Furthermore, probe packets are continuously sent to each node in the data link topology, with the probe packets being sent to each node using the ICMP protocol.

[0020] Furthermore, the node fault diagnosis algorithm is invoked to diagnose the faults of the link nodes in the failed data link, and the node anomaly diagnosis results are obtained, specifically:

[0021] Real-time acquisition of operational parameter data from each node on the failed data link;

[0022] Determine whether the operating parameter data of each node is within the preset parameter threshold range. If the operating parameter data is within the parameter threshold range, the corresponding node is determined to be a normal node; otherwise, it is a faulty node.

[0023] Furthermore, the system collects operational parameter data of each node on the failed data link in real time, including CPU utilization, memory utilization, and device temperature.

[0024] The preset parameter threshold ranges include CPU utilization threshold range, memory utilization threshold range, and device temperature threshold range;

[0025] Each node's CPU utilization, memory utilization, and device temperature are checked to see if they are within the corresponding threshold ranges. If all three are within their respective threshold ranges, the node is considered a normal node; otherwise, it is considered a faulty node.

[0026] Furthermore, a data accuracy diagnostic algorithm is invoked to diagnose the accuracy of data transmitted through abnormal data links. If the data is accurate, the data transmitted through that link is determined to be normal; if the data is inaccurate, it is determined to be abnormal, and the number of abnormal data is incremented accordingly. When the number of abnormal data exceeds an abnormal data threshold, a data accuracy diagnostic result is obtained. Specifically, the abnormal data threshold is dynamically adjusted using a pre-built threshold update model.

[0027] Acquire the link transmission data of each abnormal data link, and determine whether the link transmission data of each abnormal data link remains within the preset data accuracy threshold range and maintains the preset third time threshold. If it remains within the preset data accuracy threshold range and maintains the third time threshold, the corresponding link transmission data is determined to be normal data; otherwise, it is abnormal data.

[0028] The number of abnormal data in each abnormal data link is accumulated and superimposed. When the number of abnormal data exceeds the threshold, the corresponding abnormal data link is determined to be an inaccurate data link. The pre-built threshold update model is obtained by training an SVM model.

[0029] Furthermore, the pre-built threshold update model is obtained by training an SVM model, specifically as follows:

[0030] Obtain historical link data and accuracy results after data annotation, and construct model training set and model validation set;

[0031] The model training set is input into the SVM model to train and obtain the initial threshold to update the model;

[0032] By using a model validation set combined with cross-validation, the model parameters of the initial threshold update model are iteratively optimized to obtain the optimized threshold update model.

[0033] Furthermore, the threshold for the number of outlier data is dynamically adjusted using a pre-built threshold update model, specifically as follows:

[0034] Obtain the link transmission data for each abnormal data link, input the link transmission data into the threshold update model, and process it to obtain the model output result;

[0035] The model output is thresholded based on a pre-built linear mapping function to obtain the threshold for the number of outliers.

[0036] Calculate the threshold difference between the converted abnormal data quantity threshold and the current abnormal data quantity threshold. If the threshold difference exceeds the preset difference threshold, replace the current abnormal data quantity threshold with the converted abnormal data quantity threshold to update the current abnormal data quantity threshold.

[0037] Furthermore, the link information consistency diagnostic algorithm is invoked to diagnose the consistency of link node configuration information in data links with abnormal data, and the link node configuration diagnostic results are obtained, specifically:

[0038] Obtain the link node configuration information of each node on each data link with data anomalies. The link node configuration information includes point description information and unit information.

[0039] The string matching algorithm is used to determine whether the location description information of each node is consistent. If the location description information of each node is consistent, the location description information of the corresponding data link is determined to be consistent; otherwise, the location description information of the corresponding data link is determined to be inconsistent.

[0040] The system uses a predefined list of unit information to determine whether the unit information of each node is consistent. If the unit information of each node is consistent, the system determines that the unit information of the corresponding data link is consistent; otherwise, the system determines that the unit information of the corresponding data link is inconsistent.

[0041] The number of inconsistencies in each data link is accumulated. When the number of inconsistencies exceeds the preset information consistency threshold, the corresponding data link is determined to be an information configuration abnormal link; otherwise, it is an information configuration normal link.

[0042] The present invention also provides a data link collaborative fault diagnosis system, which is used in the data link fault analysis method described in any one of the above claims, the system comprising:

[0043] The first diagnostic analysis module is used to call the idle link diagnostic algorithm to diagnose the idle state of the data link and obtain the idle state diagnostic result. After obtaining the link idle state diagnostic result, the module calls the failed link diagnostic algorithm to diagnose the failed state of the idle data link and obtain the failed state diagnostic result. After obtaining the failed state diagnostic result, the module calls the node fault diagnostic algorithm to diagnose the link node fault of the failed data link and obtain the node anomaly diagnostic result.

[0044] The second diagnostic analysis module is used to call the data accuracy diagnostic algorithm to diagnose the accuracy of the data transmitted by the abnormal data link after obtaining the node anomaly diagnostic result. If the data is accurate, the data transmitted by the link is determined to be normal data; if the data is not accurate, the data transmitted by the link is determined to be abnormal data, and the number of abnormal data is added accordingly. When the number of abnormal data exceeds the abnormal data number threshold, the data accuracy diagnostic result is obtained. The abnormal data number threshold is dynamically adjusted through a pre-built threshold update model.

[0045] The third diagnostic analysis module is used to call the link information consistency diagnostic algorithm to diagnose the consistency of the link node configuration information of the data link with abnormal data after obtaining the data accuracy diagnostic results of the data link, obtain the link node configuration diagnostic results, and optimize the data link node configuration based on the link node configuration diagnostic results.

[0046] The present invention also provides a computer device, including a system memory and a processor, wherein the system memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods described above.

[0047] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods described above.

[0048] The present invention also provides a computer program product containing instructions that, when executed by a cluster of computer devices, cause the cluster of computer devices to perform the method described in any of the preceding claims.

[0049] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0050] 1. In this invention, the data link fault analysis method provided by this invention performs real-time diagnosis of the operating status of various devices and systems in the data link. Through sub-algorithms such as idle link diagnosis, failed link diagnosis, device fault diagnosis, link information consistency diagnosis, and data accuracy diagnosis, the real-time status, data transmission rate, and fault conditions of various devices are analyzed, achieving full-coverage diagnosis of the data link, rapid fault location, significantly reducing maintenance costs, and improving system stability and data transmission reliability. Simultaneously, this invention innovatively organically links the various sub-algorithms, realizing the intelligence, collaboration, and efficiency of the full-link fault diagnosis system, significantly improving the system's diagnostic capabilities, management efficiency, and data quality, and solving the problems in existing technologies where troubleshooting is difficult when device nodes in the data link fail, and where rapid fault diagnosis is challenging.

[0051] 2. In the data accuracy diagnosis algorithm of the present invention, the threshold update model obtained by training the SVM model achieves the technical effect of dynamic threshold update. This algorithm can not only adapt to changes in real-time data and autonomously adjust the range of data accuracy, but also effectively avoid false alarms and false negatives caused by fixed thresholds, ensuring the sensitivity and accuracy of diagnosis, improving the system's adaptability in complex data environments, making production management more intelligent, improving management efficiency and the scientific nature of decision-making, and further promoting its digital transformation. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0053] Figure 1 This is a schematic diagram of the entire data transmission link provided in this embodiment;

[0054] Figure 2This is a flowchart of a data link fault analysis method in this embodiment;

[0055] Figure 3 This is a schematic diagram of the system modules of a data link fault analysis system in this embodiment;

[0056] Figure 4 This is a schematic diagram of the structure of a computer device in this embodiment. Detailed Implementation

[0057] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0058] In this disclosure, unless otherwise stated, the use of terms such as "first," "second," etc., to describe various elements is not intended to limit the positional, temporal, or importance relationships of these elements; such terms are merely used to distinguish one element from another. In some examples, the first element and the second element may refer to the same instance of that element, while in other cases, based on the context, they may refer to different instances.

[0059] The terminology used in the description of the various examples in this disclosure is for the purpose of describing particular examples only and is not intended to be limiting. Unless the context explicitly indicates otherwise, an element may be one or more unless the number of elements is specifically limited. Furthermore, the term "and / or" as used in this disclosure covers any one of the listed items and all possible combinations thereof.

[0060] Example 1

[0061] See Figure 1The diagram shows the entire data transmission link, which includes multiple data transmission devices, specifically sensors, PLC devices, RTU devices, switches, firewalls, routers, and a central database, as well as the data links of each transmission device bracket; the sensors are connected to the corresponding devices through different interfaces (such as RS485, RS232), including PLCs (Programmable Logic Controllers) and RTUs (Remote Terminal Units). The data in the well station needs to pass through the central station and the work area level, and finally the data is collected into the gas field for centralized storage. The specific path is as follows: (1) Sensor data is transmitted to the central station through the PLC and switch of the well station; (2) it is transmitted to the work area through the switch and firewall of the central station; (3) it is transmitted to the gas field through the switch and firewall of the work area; (4) the data is finally collected into the central database of the gas field for storage. The entire link involves many nodes and is complex.

[0062] In this embodiment, the data acquisition program sequentially collects data from the PLC, switch, firewall, and central database along the data link according to preset acquisition rules and interface protocols. The PLC uses the Modbus TCP protocol for data acquisition. The switch, firewall, and router acquire data via the SNMP protocol, collecting data items including device CPU, memory, and current port traffic. The central database acquires data via the manufacturer-provided HTTP API interface, collecting data items including real-time data points, descriptions, and current values. The acquisition program actively sends acquisition commands to the devices, with an acquisition frequency set to once every 5 minutes. After receiving the data, the acquisition server further preprocesses it, such as converting the data format to unify the format of data from different devices for subsequent analysis.

[0063] See Figure 2 A flowchart of a data link fault analysis method is shown, which includes:

[0064] S1: Call the idle link diagnosis algorithm to diagnose the idle state of the data link and obtain the idle state diagnosis result. After obtaining the link idle state diagnosis result, call the failed link diagnosis algorithm to diagnose the failed state of the idle data link and obtain the failed state diagnosis result. After obtaining the failed state diagnosis result, call the node fault diagnosis algorithm to diagnose the link node fault of the failed data link and obtain the node anomaly diagnosis result.

[0065] It should be noted that in this embodiment, the idle link diagnosis algorithm can promptly detect idle links that have not transmitted data for a long time by monitoring the real-time data transmission rate of the data link. Once an idle link is detected, the system will immediately trigger the subsequent detection process, which will link the failed link diagnosis algorithm and the equipment fault diagnosis algorithm to check the operating status of the nodes associated with the idle link.

[0066] Meanwhile, in this embodiment, the failed link diagnosis algorithm can monitor the status of each node in the link and the link's connectivity in real time. When a link failure is detected, the status of each node at the end of the link is further checked to determine the cause of the link failure. If the node status is abnormal, the device fault diagnosis algorithm will be triggered to perform a deep inspection of the relevant devices. It should be noted that, in addition to performing failure status diagnosis only on data links in the idle state after obtaining the idle state diagnosis result, the failed link diagnosis algorithm of this technical solution can also directly perform failure status diagnosis on all data link states individually. The diagnosis process is the same and will not be described in detail here.

[0067] Meanwhile, in this embodiment, the equipment fault diagnosis algorithm can accurately determine whether the equipment has a fault by real-time monitoring of the equipment's key operating parameters. Once a fault is detected, it will be linked with the data accuracy diagnosis algorithm to verify the accuracy of the data transmitted by the equipment, ensuring the reliability and integrity of the data.

[0068] S2: After obtaining the node anomaly diagnosis result, the data accuracy diagnosis algorithm is called to diagnose the accuracy of the data transmitted by the abnormal data link. If the data is accurate, the data transmitted by the link is determined to be normal data. If the data is not accurate, the data transmitted by the link is determined to be abnormal data, and the number of abnormal data is added accordingly. When the number of abnormal data exceeds the abnormal data number threshold, the data accuracy diagnosis result is obtained. The abnormal data number threshold is dynamically adjusted through a pre-built threshold update model.

[0069] It should be noted that in this embodiment, the data accuracy diagnostic algorithm monitors and analyzes the transmitted data in real time to identify abnormal data. If data anomalies are detected, the link information configuration will be checked in conjunction with the link information consistency diagnostic algorithm to determine whether the data anomalies are caused by link configuration errors.

[0070] S3: After obtaining the data accuracy diagnosis results of the data link, call the link information consistency diagnosis algorithm to diagnose the consistency of the link node configuration information of the data link with abnormal data, obtain the link node configuration diagnosis results, and optimize the data link node configuration based on the link node configuration diagnosis results.

[0071] It should be noted that in this embodiment, the link information consistency diagnosis algorithm compares the configuration information of different nodes and finds inconsistencies. If the link information is inconsistent, it will affect the accuracy of the data and the normal operation of the device. The system will issue an alarm in time, and the operation and maintenance personnel will make adjustments.

[0072] Specifically, in this embodiment, through the coordinated operation of various algorithms, the system can diagnose the entire link from multiple dimensions. The idle link diagnosis algorithm can detect potential problems in advance, the failed link diagnosis algorithm can quickly locate the faulty link, and the equipment fault diagnosis algorithm can accurately identify equipment faults. The comprehensive analysis of this information greatly improves the efficiency and accuracy of fault diagnosis analysis.

[0073] Meanwhile, the collaborative diagnostic mode simplifies the fault response and handling process. When the system detects a fault, it automatically generates a detailed work order, including information on the faulty device, faulty link details, and fault type, and promptly pushes it to relevant maintenance personnel. Maintenance personnel can view real-time monitoring data, historical data, and relevant fault solutions from the knowledge base through a unified fault response interface, improving the efficiency and quality of fault handling. For example, in handling a link failure fault, the network operations team and the equipment maintenance team quickly collaborated through the platform, utilizing the detailed information provided by the correlation diagnostic algorithm to rapidly locate the fault point and promptly repair the fault, significantly shortening the fault handling time.

[0074] As one possible implementation method,

[0075] The idle link diagnostic algorithm is invoked to diagnose the idle status of the data link, and the idle status diagnostic results are obtained as follows:

[0076] Obtain the data transmission timestamp information for each data link, calculate the time interval between the data transmission timestamp information and the current time, and obtain the idle time data;

[0077] It should be noted that, in this embodiment, an ArrayList data structure is maintained for each data link to record real-time data timestamp information. When data is collected from the data link, the data timestamp information is also obtained and added to the end of the corresponding list. At the same time, for each data link, the idle time data is calculated as the time interval from the last real-time data timestamp with data transmission to the current time.

[0078] The idle time data is compared with a preset first time threshold. If the idle time data exceeds the first time threshold, the corresponding data link is determined to be an idle data link; otherwise, it is a non-idle data link.

[0079] It should be noted that, in this embodiment, the preset first time threshold is an empirical value set according to the production environment and business needs, used to determine whether the link is in an idle state. It can be set to 2400s, 3000s, 3600s or other times. In this embodiment, 3600 seconds is used as the first time threshold.

[0080] For example, for a certain link, 10 real-time data timestamps have been collected. When the interval between the last time data was transmitted to the current time is calculated to be 4500 seconds (greater than the first time threshold), the data link is determined to be an idle data link.

[0081] Specifically, in this embodiment, by detecting the data transmission rate in the data link, data links that have not transmitted data for a long time are identified, thus avoiding resource waste. Automated diagnosis reduces the workload of manual operation and maintenance and improves the overall production management level.

[0082] As one possible implementation method,

[0083] The failure link diagnosis algorithm is invoked to diagnose the failure status of idle data links, and the failure status diagnosis results are obtained as follows:

[0084] By treating each system device with idle data links as a node and the data link relationships between each system device as edges, a data link topology is constructed.

[0085] It should be noted that, in this embodiment, the system devices include PLCs, switches, firewalls, host computers, communication devices, and a central database, etc. These devices are treated as nodes, and the data link relationships between these devices are treated as edges to construct a data link topology. The constructed data link topology is used to store relevant information about nodes and links.

[0086] The probe packets are continuously sent to each node in the data link topology. If a node reply signal is received within a preset second time threshold, the corresponding node is determined to be a normal node; otherwise, it is an abnormal node.

[0087] It should be noted that in this embodiment, the probe packet is sent to each node using the ICMP protocol. When the ICMP message is sent, it carries a "type" field and a "code" field to indicate the specific message type, and the return time is used as the basis for diagnosis.

[0088] Meanwhile, in this embodiment, the second time threshold can be set to 3s, 5s, 10s, or determined according to the actual situation. In this embodiment, the second time threshold is set to 5s.

[0089] When there is only one data link between two nodes, if there is at least one abnormal node among the two nodes, the data link is determined to be in a failed state; otherwise, it is in a normal state.

[0090] When there are multiple data links between two nodes, if there is at least one abnormal node between the two nodes and at least one other data link between the two nodes is in a normal state, then the data link is determined to be in a failed state; otherwise, it is in a normal state.

[0091] It should be noted that in this embodiment, when there is only one data link between two nodes, the data link is in a failed state if either of the two nodes is abnormal. In other words, the data link is in a normal state only if both nodes are normal. When there are multiple data links between two nodes, the remaining data links are in a failed state if either of the two nodes is abnormal and there are normal data links between the two nodes. This is because if one node is abnormal and the link is not working, but there is an alternative path and the alternative path is normal, then the original link can be determined to be failed.

[0092] For example, if no response is received within 5 seconds after sending a probe packet to a switch, the node status of the switch is marked as abnormal. The data link connecting two nodes (such as a well station PLC and an adjacent switch) is judged. If at least one of the nodes at both ends of the link is abnormal and the link itself is not connected, or if one end of the node is abnormal and the link is not connected but there is an alternative path and the alternative path is normal, the link is determined to be a failed link.

[0093] As one possible implementation method,

[0094] The node fault diagnosis algorithm is invoked to diagnose the faults of the failed data link nodes, and the node anomaly diagnosis results are obtained, specifically:

[0095] Real-time acquisition of operational parameter data from each node on the failed data link;

[0096] It should be noted that the operating parameter data includes CPU utilization, memory utilization, and device temperature, etc.

[0097] Determine whether the operating parameter data of each node is within the preset parameter threshold range. If the operating parameter data is within the parameter threshold range, the corresponding node is determined to be a normal node; otherwise, it is a faulty node.

[0098] It should be noted that, in this embodiment, the preset parameter threshold ranges include CPU utilization threshold ranges, memory utilization threshold ranges, and device temperature threshold ranges. The specific value ranges are determined according to the actual situation and are not subject to much restriction here. To determine whether the running parameter data of each node is within the preset parameter threshold range, the CPU utilization, memory utilization, and device temperature of each node are determined separately to see if they are within the corresponding CPU utilization threshold range, memory utilization threshold range, and device temperature threshold range. If the CPU utilization, memory utilization, and device temperature are all within the corresponding threshold range, the corresponding node is determined to be a normal node; otherwise, it is a faulty node.

[0099] Specifically, in this embodiment, a node fault diagnosis algorithm is used to identify abnormal operating conditions of network devices, communication devices, host computers, and slave computers in the production chain. This algorithm monitors key operating parameters of the devices (such as CPU load, memory usage, and network status) to determine if a device has malfunctioned and provides real-time alarms and fault location functions.

[0100] For example, if a host computer's CPU utilization is 95%, while its normal operating parameter range is 20%-80%, then the host computer is determined to be faulty. For complex devices, such as SCADA host computers, multiple operating parameters (such as CPU utilization, memory utilization, network connection status, etc.) are considered together for judgment. When any of these parameters exceeds the normal range, the device is determined to be faulty, a real-time alarm is issued, and the faulty device is located on the fault diagnosis interface.

[0101] As one possible implementation method,

[0102] A data accuracy diagnostic algorithm is invoked to diagnose the accuracy of data transmitted through abnormal data links. If the data is accurate, the data transmitted through that link is determined to be normal; otherwise, it is determined to be abnormal, and the number of abnormal data is incremented accordingly. When the number of abnormal data exceeds an abnormal data threshold, a data accuracy diagnostic result is obtained. Specifically, the abnormal data threshold is dynamically adjusted using a pre-built threshold update model.

[0103] Acquire the link transmission data of each abnormal data link, and determine whether the link transmission data of each abnormal data link remains within the preset data accuracy threshold range and maintains the preset third time threshold. If it remains within the preset data accuracy threshold range and maintains the third time threshold, the corresponding link transmission data is determined to be normal data; otherwise, it is abnormal data.

[0104] It should be noted that in this embodiment, the data accuracy threshold range is determined based on historical data, equipment performance indicators, and production process requirements, while the third time threshold is determined according to the actual situation and is not subject to excessive restrictions here. For a data link, the data transmitted by the link is compared with the preset data accuracy threshold range one by one. If the data transmitted by the link is within the data accuracy threshold range and the data is consistently accurate to the third time threshold, then the data is considered normal data; otherwise, it is considered abnormal data.

[0105] The number of abnormal data in each abnormal data link is accumulated and superimposed. When the number of abnormal data exceeds the threshold, the corresponding abnormal data link is determined to be an inaccurate data link. The pre-built threshold update model is obtained by training an SVM model.

[0106] Specifically, in this embodiment, the pre-built threshold update model is obtained by training an SVM model. Specifically, data preprocessing is performed first to obtain historical link data and the accuracy results after data annotation. A model training set and a model validation set are then constructed. These historical link data cover different production scenarios, equipment states, and time periods to ensure the diversity and representativeness of the training set.

[0107] A comprehensive data cleaning process was performed on historical data to remove obviously erroneous data, such as data values ​​outside the reasonable range (e.g., negative temperature data that does not conform to actual physical conditions), data with incorrect formats (e.g., string data appearing in numeric fields), and data records with an excessively high proportion of missing values. For missing values, mean imputation, median imputation, or model-based imputation methods (e.g., using the K-nearest neighbor algorithm to imput missing values ​​based on similar data points) were used to ensure data integrity. The Z-score standardization method was employed, with the formula: Where x represents the original data, μ is the mean of the data, and σ is the standard deviation of the data. new The data is standardized. Data features are transformed to the same scale range for better model training. Then, the preprocessed historical data is divided into training and validation sets in a 7:3 ratio.

[0108] The initial threshold update model is obtained by inputting the training set into an SVM model. Then, using a validation set and cross-validation, the model parameters are iteratively optimized to obtain the optimized threshold update model. Specifically, cross-validation is used: the training set is divided into K subsets. Each time, K-1 subsets are used as training data, and the remaining subset is used as validation data. This process is repeated K times, and the accuracy of each validation is calculated. The average accuracy is used as the model's evaluation metric. The model parameters are adjusted based on the evaluation metric until the model performance reaches its optimal level.

[0109] It should be noted that, in this embodiment, the initial value of the abnormal data quantity threshold is determined comprehensively based on factors such as the importance of the link, the characteristics of the data, and the requirements for accuracy. No restrictions are imposed here. This threshold can be dynamically adjusted and updated based on a pre-built threshold update model.

[0110] In this embodiment, the threshold update model uses Support Vector Machine (SVM) as the base model for threshold prediction, primarily due to its advantages in handling small sample sizes and nonlinear data. SVM can effectively learn the distribution patterns of historical data to predict thresholds. Furthermore, considering the complexity and dynamism of the data, a neural network is used as an auxiliary model for validation and supplementation to improve the accuracy and reliability of the prediction.

[0111] Specifically, in this embodiment, a data accuracy diagnostic algorithm is used to ensure that the data transmitted in the production chain remains accurate and effective throughout the entire process. Through real-time monitoring and analysis of the transmitted data, the algorithm identifies abnormal data in the chain and processes it quickly, ensuring the reliability and integrity of the production data. It can also quickly identify and classify fault types, providing detailed fault reports to help maintenance personnel respond to problems rapidly.

[0112] For example, the initial value for the threshold of abnormal data quantity is set to 10, and the third time threshold is set to 60 seconds. For the wellhead oil pressure data of a certain link, the collected data is compared with the data accuracy threshold range. If 15 data points are found to be outside the data accuracy threshold range (abnormal data) within 60 seconds, the data link is determined to be an inaccurate link.

[0113] As one possible implementation method,

[0114] The threshold for the number of outliers is dynamically adjusted using a pre-built threshold update model, specifically as follows:

[0115] Obtain the link transmission data for each abnormal data link, input the link transmission data into the threshold update model, and process it to obtain the model output result;

[0116] It should be noted that, in this embodiment, after obtaining the link transmission data of the data link, the link transmission data is preprocessed, and the processed link transmission data is input into the threshold update model to obtain the model output result.

[0117] The model output is thresholded based on a pre-built linear mapping function to obtain the threshold for the number of outliers.

[0118] It should be noted that in this embodiment, if the range of the model output is [0,1], and it is desired that the threshold for the number of outliers varies within a certain range (e.g., [10,50]), then the linear mapping function is set to f(y). sum ) = 10 + 40y sum , where y sum f(y) represents the model output. sum The number of abnormal data after mapping is denoted as ). The linear mapping function can also be set according to the actual situation, which will not be elaborated on here.

[0119] Calculate the threshold difference between the converted abnormal data quantity threshold and the current abnormal data quantity threshold. If the threshold difference exceeds the preset difference threshold, replace the current abnormal data quantity threshold with the converted abnormal data quantity threshold to update the current abnormal data quantity threshold.

[0120] It should be noted that in this embodiment, the difference threshold is set to 5. In other embodiments, it can be determined according to the actual situation, and no restrictions are imposed here. The calculated and converted abnormal data quantity threshold is compared with the current abnormal data quantity threshold. If the difference between the calculated and converted abnormal data quantity threshold and the current abnormal data quantity threshold exceeds the difference threshold, the calculated and converted abnormal data quantity threshold is updated to the current abnormal data quantity threshold. The updated threshold will be used for subsequent data accuracy judgment.

[0121] As one possible implementation method,

[0122] The link information consistency diagnostic algorithm is invoked to diagnose the consistency of link node configuration information in data links with abnormal data, and the link node configuration diagnostic results are obtained as follows:

[0123] Obtain the link node configuration information of each node on each data link with data anomalies. The link node configuration information includes point description information and unit information.

[0124] The string matching algorithm is used to determine whether the location description information of each node is consistent. If the location description information of each node is consistent, the location description information of the corresponding data link is determined to be consistent; otherwise, the location description information of the corresponding data link is determined to be inconsistent.

[0125] The system uses a predefined list of unit information to determine whether the unit information of each node is consistent. If the unit information of each node is consistent, the system determines that the unit information of the corresponding data link is consistent; otherwise, the system determines that the unit information of the corresponding data link is inconsistent.

[0126] The number of inconsistencies in each data link is accumulated. When the number of inconsistencies exceeds the preset information consistency threshold, the corresponding data link is determined to be an information configuration abnormal link; otherwise, it is an information configuration normal link.

[0127] It should be noted that, in this embodiment, the value of the information consistency threshold is determined according to the complexity of the link and the requirements for information consistency. It can be set to 1, 3 or other numbers. No restrictions are imposed here. The information consistency threshold represents the upper limit of the number of inconsistent information allowed.

[0128] Specifically, in this embodiment, a link information consistency diagnostic algorithm is used to identify the consistency of configuration information (such as location descriptions, units, etc.) of each node in the data link, ensuring that the information of all nodes in the link remains consistent. This algorithm quickly detects and reports inconsistencies by comparing the configuration information of different nodes.

[0129] Example 2

[0130] The present invention also provides a data link collaborative fault diagnosis system, which is used in the data link fault analysis method described in any one of the above claims, the system comprising:

[0131] The first diagnostic analysis module 100 is used to call the idle link diagnostic algorithm to diagnose the idle state of the data link and obtain the idle state diagnostic result. After obtaining the link idle state diagnostic result, it calls the failed link diagnostic algorithm to diagnose the failed state of the idle data link and obtain the failed state diagnostic result. After obtaining the failed state diagnostic result, it calls the node fault diagnostic algorithm to diagnose the link node fault of the failed data link and obtain the node anomaly diagnostic result.

[0132] The second diagnostic analysis module 200 is used to call the data accuracy diagnostic algorithm to diagnose the accuracy of the data transmitted by the abnormal data link after obtaining the node anomaly diagnostic result. If the data is accurate, the data transmitted by the link is determined to be normal data. If the data is not accurate, the data transmitted by the link is determined to be abnormal data, and the number of abnormal data is added accordingly. When the number of abnormal data exceeds the abnormal data number threshold, the data accuracy diagnostic result is obtained. The abnormal data number threshold is dynamically adjusted by a pre-built threshold update model.

[0133] The third diagnostic analysis module 300 is used to call the link information consistency diagnostic algorithm to diagnose the consistency of the link node configuration information of the data link with abnormal data after obtaining the data accuracy diagnostic result of the data link, obtain the link node configuration diagnostic result, and optimize the data link node configuration based on the link node configuration diagnostic result.

[0134] It should be noted that the modules in the system of Embodiment 2 correspond to the steps in the method of Embodiment 1. The steps in the method of Embodiment 1 have been described in detail in Embodiment 1, and the module content in the system will not be described in detail in this Embodiment 2.

[0135] Example 3

[0136] This embodiment also provides a computer device, including a system memory 1005 and a processor 1001. The system memory 1005 stores a computer program, and the processor 1001 executes the computer program to implement the steps of any of the methods described above.

[0137] It should be noted that the processor 1001 is used to execute the steps in the above method embodiments according to the instructions in the program code. Alternatively, when the processor 1001 executes the computer program, it implements the functions of each module / unit in the above system / device embodiments.

[0138] Specifically, in this embodiment, the computer program can be divided into one or more modules / units. One or more modules / units are stored in the system memory 1005 and executed by the processor 1001 to complete this application. One or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device.

[0139] The terminal device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor 1001 and a system memory 1005. Those skilled in the art will understand that this does not constitute a limitation on the terminal device, which may include more or fewer components than shown, or a combination of certain components, or different components. For example, the terminal device may also include an input / output device 1003, a network access device 1002, a bus 1006, etc.

[0140] The processor 1001 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0141] System memory 1005 can be an internal storage unit of the terminal device, such as a hard drive or RAM. System memory 1005 can also be a storage device 1004 of the terminal device, such as an external hard drive, SmartMedia Card (SMC), Secure Digital (SD) card, or FlashCard. Furthermore, system memory 1005 can include both internal storage units and storage device 1004. System memory 1005 is used to store computer programs and other programs and data required by the terminal device. System memory 1005 can also be used to temporarily store data that has been output or will be output.

[0142] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0143] Example 4

[0144] This embodiment provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0145] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), registers, hard disks, optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof, or any other form of computer-readable storage medium in the art.

[0146] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside within an application-specific integrated circuit (ASIC). In embodiments of the invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device.

[0147] Example 5

[0148] This embodiment also provides a computer program product containing instructions that, when executed by a cluster of computer devices, cause the cluster of computer devices to perform the method described in Embodiment 1.

[0149] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A data link fault analysis method, characterized in that, include: The idle link diagnosis algorithm is called to diagnose the idle state of the data link and obtain the idle state diagnosis result. After obtaining the link idle state diagnosis result, the failure link diagnosis algorithm is called to diagnose the failure state of the idle data link and obtain the failure state diagnosis result. After obtaining the failure state diagnosis result, the node fault diagnosis algorithm is called to diagnose the link node fault of the failure data link and obtain the node anomaly diagnosis result. After obtaining the node anomaly diagnosis result, the data accuracy diagnosis algorithm is called to diagnose the accuracy of the data transmitted by the abnormal data link. If the data is accurate, the data transmitted by the link is determined to be normal data. If the data is not accurate, the data transmitted by the link is determined to be abnormal data, and the number of abnormal data is added accordingly. When the number of abnormal data exceeds the abnormal data number threshold, the data accuracy diagnosis result is obtained. The abnormal data number threshold is dynamically adjusted through a pre-built threshold update model. After obtaining the data accuracy diagnosis results of the data link, the link information consistency diagnosis algorithm is called to diagnose the consistency of the link node configuration information of the data link with abnormal data, and the link node configuration diagnosis results are obtained. Based on the link node configuration diagnosis results, the data link node configuration is optimized.

2. The data link fault analysis method according to claim 1, characterized in that, The idle link diagnostic algorithm is invoked to diagnose the idle status of the data link, and the idle status diagnostic results are obtained as follows: Obtain the data transmission timestamp information for each data link, calculate the time interval between the data transmission timestamp information and the current time, and obtain the idle time data; The idle time data is compared with a preset first time threshold. If the idle time data exceeds the first time threshold, the corresponding data link is determined to be an idle data link; otherwise, it is a non-idle data link.

3. The data link fault analysis method according to claim 1, characterized in that, The failure link diagnosis algorithm is invoked to diagnose the failure status of idle data links, and the failure status diagnosis results are obtained as follows: By treating each system device with idle data links as a node and the data link relationships between each system device as edges, a data link topology is constructed. The probe packets are continuously sent to each node in the data link topology. If a node reply signal is received within a preset second time threshold, the corresponding node is determined to be a normal node; otherwise, it is an abnormal node. When there is only one data link between two nodes, if there is at least one abnormal node among the two nodes, the data link is determined to be in a failed state; otherwise, it is in a normal state. When there are multiple data links between two nodes, if there is at least one abnormal node between the two nodes and at least one other data link between the two nodes is in a normal state, then the data link is determined to be in a failed state; otherwise, it is in a normal state.

4. The data link fault analysis method according to claim 3, characterized in that, Probe packets are continuously sent to each node in the data link topology, with the probe packets being sent to each node using the ICMP protocol.

5. The data link fault analysis method according to claim 1, characterized in that, The node fault diagnosis algorithm is invoked to diagnose the faults of the failed data link nodes, and the node anomaly diagnosis results are obtained, specifically: Real-time acquisition of operational parameter data from each node on the failed data link; Determine whether the operating parameter data of each node is within the preset parameter threshold range. If the operating parameter data is within the parameter threshold range, the corresponding node is determined to be a normal node; otherwise, it is a faulty node.

6. The data link fault analysis method according to claim 5, characterized in that, Real-time acquisition of operational parameter data for each node on the failed data link, including CPU utilization, memory utilization, and device temperature; The preset parameter threshold ranges include CPU utilization threshold range, memory utilization threshold range, and device temperature threshold range; Each node's CPU utilization, memory utilization, and device temperature are checked to see if they are within the corresponding threshold ranges. If all three are within their respective threshold ranges, the node is considered a normal node; otherwise, it is considered a faulty node.

7. The data link fault analysis method according to claim 1, characterized in that, A data accuracy diagnostic algorithm is invoked to diagnose the accuracy of data transmitted through abnormal data links. If the data is accurate, the data transmitted through that link is determined to be normal; otherwise, it is determined to be abnormal, and the number of abnormal data is incremented accordingly. When the number of abnormal data exceeds an abnormal data threshold, a data accuracy diagnostic result is obtained. Specifically, the abnormal data threshold is dynamically adjusted using a pre-built threshold update model. Acquire the link transmission data of each abnormal data link, and determine whether the link transmission data of each abnormal data link remains within the preset data accuracy threshold range and maintains the preset third time threshold. If it remains within the preset data accuracy threshold range and maintains the third time threshold, the corresponding link transmission data is determined to be normal data; otherwise, it is abnormal data. The number of abnormal data in each abnormal data link is accumulated and superimposed. When the number of abnormal data exceeds the threshold, the corresponding abnormal data link is determined to be an inaccurate data link. The pre-built threshold update model is obtained by training an SVM model.

8. The data link fault analysis method according to claim 7, characterized in that, The pre-built threshold update model is obtained by training an SVM model, specifically as follows: Obtain historical link data and accuracy results after data annotation, and construct model training set and model validation set; The model training set is input into the SVM model to train and obtain the initial threshold to update the model; By using a model validation set combined with cross-validation, the model parameters of the initial threshold update model are iteratively optimized to obtain the optimized threshold update model.

9. A data link fault analysis method according to claim 1 or 7, characterized in that, The threshold for the number of outlier data is dynamically adjusted using a pre-built threshold update model, specifically as follows: Obtain the link transmission data for each abnormal data link, input the link transmission data into the threshold update model, and process it to obtain the model output result; The model output is thresholded based on a pre-built linear mapping function to obtain the threshold for the number of outliers. Calculate the threshold difference between the converted abnormal data quantity threshold and the current abnormal data quantity threshold. If the threshold difference exceeds the preset difference threshold, replace the current abnormal data quantity threshold with the converted abnormal data quantity threshold to update the current abnormal data quantity threshold.

10. A data link fault analysis method according to claim 1, characterized in that, The link information consistency diagnostic algorithm is invoked to diagnose the consistency of link node configuration information in data links with abnormal data, and the link node configuration diagnostic results are obtained as follows: Obtain the link node configuration information of each node on the data link for each data anomaly. The link node configuration information includes point description information and unit information. The string matching algorithm is used to determine whether the location description information of each node is consistent. If the location description information of each node is consistent, the location description information of the corresponding data link is determined to be consistent; otherwise, the location description information of the corresponding data link is determined to be inconsistent. The system uses a predefined list of unit information to determine whether the unit information of each node is consistent. If the unit information of each node is consistent, the system determines that the unit information of the corresponding data link is consistent; otherwise, the system determines that the unit information of the corresponding data link is inconsistent. The number of inconsistencies in each data link is accumulated. When the number of inconsistencies exceeds the preset information consistency threshold, the corresponding data link is determined to be an information configuration abnormal link; otherwise, it is an information configuration normal link.

11. A data link collaborative fault diagnosis system, characterized in that, This system is used in a data link fault analysis method according to any one of claims 1-10, the system comprising: The first diagnostic analysis module is used to call the idle link diagnostic algorithm to diagnose the idle state of the data link and obtain the idle state diagnostic result. After obtaining the link idle state diagnostic result, the module calls the failed link diagnostic algorithm to diagnose the failed state of the idle data link and obtain the failed state diagnostic result. After obtaining the failed state diagnostic result, the module calls the node fault diagnostic algorithm to diagnose the link node fault of the failed data link and obtain the node anomaly diagnostic result. The second diagnostic analysis module is used to call the data accuracy diagnostic algorithm to diagnose the accuracy of the data transmitted by the abnormal data link after obtaining the node anomaly diagnostic result. If the data is accurate, the data transmitted by the link is determined to be normal data; if the data is not accurate, the data transmitted by the link is determined to be abnormal data, and the number of abnormal data is added accordingly. When the number of abnormal data exceeds the abnormal data number threshold, the data accuracy diagnostic result is obtained. The abnormal data number threshold is dynamically adjusted through a pre-built threshold update model. The third diagnostic analysis module is used to call the link information consistency diagnostic algorithm to diagnose the consistency of the link node configuration information of the data link with abnormal data after obtaining the data accuracy diagnostic results of the data link, obtain the link node configuration diagnostic results, and optimize the data link node configuration based on the link node configuration diagnostic results.

12. A computer device comprising a system memory and a processor, wherein the system memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1 to 10.

14. A computer program product containing instructions, characterized in that, When the instruction is executed by a cluster of computer devices, the cluster of computer devices causes the cluster of computer devices to perform the method as described in any one of claims 1 to 10.

Images