On-line detection system and method for current collection circuit

Abstract

The invention provides a current collection line on-line detection system and method, and belongs to the technical field of power detection, and the system comprises an acquisition module which obtains the historical operation data of a current collection line, and collects the real-time operation data of the current collection line in real time; the construction module is used for determining the normal operation range of the current collection line based on the historical operation data and constructing a fault detection model; the determination module is used for determining real-time fault data of the current collection line based on the real-time operation data and the normal operation range; and the fault module is used for performing fault detection on the real-time fault data of the current collection line based on the fault detection model and generating a fault report. Real-time accurate operation data acquisition and processing can be realized, fault key nodes can be accurately identified and positioned, high efficiency and accuracy of intelligent fault detection are realized, safety and stability of a current collection line are improved, shutdown risks are reduced, and configuration and maintenance strategies of maintenance resources are optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of power detection, and in particular to an online detection system and method for a collector line. Background Art

[0002] Online monitoring of collector lines is a critical technology used in the power industry to ensure stable operation. With increasing power loads and aging equipment, the risk of failure in collector lines is increasing. Traditional manual inspections are not only inefficient but also difficult to detect potential faults in real time. To address this issue, as early as the late 20th century, the power industry began introducing online monitoring technology to monitor the operating status of collector lines in real time. By combining sensors with data acquisition equipment, real-time data on line electrical parameters can be collected. With advances in technologies such as big data analysis and artificial intelligence, fault prediction and intelligent diagnosis have become core functions of online monitoring systems. While existing monitoring systems can provide basic monitoring functions, they still lack intelligent fault diagnosis and preventive maintenance.

[0003] Therefore, the present invention provides a system and method for online detection of a collector line. Summary of the Invention

[0004] The present invention provides an online detection system and method for a collector line. By acquiring historical operating data of the collector line, collecting real-time operating data of the collector line, determining the normal operating range, and constructing a fault detection model, the real-time fault data of the collector line is determined according to the real-time operating data and the normal operating range. Fault detection is performed in combination with the fault detection model and a fault report is generated. This system can realize real-time and accurate operation data collection and processing, accurately identify and locate key fault nodes, realize the efficiency and accuracy of intelligent fault detection, improve the safety and stability of the collector line, reduce the risk of outage, and optimize the configuration of maintenance resources and maintenance strategies.

[0005] In one aspect, the present invention provides an online detection system for a collector line, comprising: Collection module: obtains historical operation data of the collector line and collects real-time operation data of the collector line; Construction module: Determine the normal operating range of the collector line based on historical operating data and build a fault detection model; Determination module: determines the real-time fault data of the collector line based on the real-time operation data and the normal operation range; Fault module: Based on the fault detection model, it performs fault detection on the real-time fault data of the collector line and generates a fault report.

[0006] According to the present invention, a current collection line online detection system, a collection module, includes: Historical sub-operation data unit: obtains historical sub-operation data of the collector line in multiple historical specified time periods, wherein the historical sub-operation data includes historical normal sub-operation data and historical fault sub-operation data. The historical normal sub-operation data includes a historical normal operation matrix, and the historical fault sub-operation data includes a historical fault operation vector and a historical fault type. Historical normal operation data and historical fault operation data unit: determines historical normal operation data based on historical normal sub-operation data of the collector line in all historical sub-operation data of the historical specified time period, and at the same time, determines historical fault operation data based on historical fault sub-operation data of the collector line in all historical sub-operation data of the historical specified time period; Historical operation data unit: determines historical normal operation data and historical fault operation data.

[0007] According to the present invention, a current collection line online detection system, a collection module, includes: Operation area unit: determining a plurality of operation areas based on historical fault operation data in the historical operation data; Key node data unit: Based on the topological structure of the collector line and all operating areas, key node data is determined. The key node data includes multiple key nodes, the data collection requirements of each key node, and the location of the key nodes. Installation unit: Install sensor groups at corresponding key node locations based on the data collection requirements of each key node; Real-time operation data unit: collects real-time operation sub-data of each key node of the collector line based on the sensor group, pre-processes the real-time operation sub-data of each key node, and determines the real-time operation data based on the pre-processed real-time operation sub-data of all key nodes.

[0008] According to the present invention, a collector line online detection system is provided, and a building module is provided, including: Key node category unit: performing cluster analysis based on historical normal operation vectors in all historical normal sub-operation data in the historical normal operation data to determine multiple key node categories, wherein each key node category includes multiple key nodes; Node category data unit: extracts historical normal operation vectors of all key nodes in each key node category from the historical normal operation matrix of each historical normal sub-operation data, and determines node category data of each key node category based on all historical normal operation vectors of each key node category extracted from the historical normal operation matrix of all historical normal sub-operation data; Normal operation matrix unit: based on the node category data of each key node category, determining the normal operation matrix of each node category data; Normal operation range unit: determines the normal operation range of the collector line based on the normal operation matrix of the node category data of all key node categories; Construction unit: taking the historical fault operation vectors in all the historical fault sub-operation data in the historical fault operation data as the input of the fault detection model, and taking the historical fault types in all the historical fault sub-operation data in the historical fault operation data as the output of the fault detection model; Training unit: training a fault detection model based on historical fault operation data in the historical operation data.

[0009] According to the present invention, a collector line online detection system is provided, wherein the normal operation matrix unit comprises: ; ; ; ; ; ; in, represents the normal operation matrix of the k-th node category data, Respectively represent the lower limit and upper limit of the first feature of the k-th node category data, They represent the lower limit and upper limit of the jth feature of the kth node category data, respectively. They represent the lower limit and upper limit of the N1th feature of the kth node category data, respectively. N1 represents the number of features of the historical normal operation vector. represents the average value of the jth feature of the kth node category data, represents the standard deviation of the jth feature of the kth node category data, The eigenvalue of the jth feature of the centroid running vector of the kth node category data, The characteristic value of the jth characteristic of the historical normal operation vector of the i-th key node in the key node category of the k-th node category data in the t-th historical specified time period, represents the number of historical normal operation vectors in the k-th node category data, The eigenvalue of the jth feature of the historical normal operation vector of the i-th key node in the key node category of the k-th node category data in all historical specified time periods and the eigenvalue of the jth feature of the centroid running vector of the kth node category data The similarity value of represents the time smoothing factor, represents the first adjustment parameter, represents the second adjustment parameter, represents the third adjustment parameter, Represents the median of all j-th features of the k-th node category data after sorting, represents the 5% quantile after sorting all j-th features of the k-th node category data, Represents the 95th percentile after sorting all j-th features of the k-th node category data.

[0010] According to the present invention, a current collector line online detection system is provided, wherein the determination module includes: Real-time operation vector unit: extracts features from the pre-processed real-time operation sub-data of each key node in the real-time operation data, and determines the real-time operation vector of each key node; Faulty key node unit: judge the real-time operation vector of each key node against the normal operation matrix of the node category data of the key node category corresponding to each key node in the normal operation range. If the eigenvalue of any feature of the real-time operation vector of the key node is not between the lower limit and upper limit of the corresponding normal operation matrix, it is determined to be a faulty key node; otherwise, it is determined to be a normal key node; Real-time fault data unit: determines real-time fault data based on all fault-critical nodes and the real-time operation vector of each fault-critical node.

[0011] According to the present invention, a collector line online detection system and a fault module are provided, comprising: Fault prediction unit: Inputs all real-time operation vectors in real-time fault data into the fault detection model, and determines the predicted fault type and fault maintenance optimization suggestions for each key fault node based on the output structure of the fault detection model; Maintenance optimization unit: performs maintenance optimization for each critical fault node based on the predicted fault type and fault maintenance optimization suggestions; Generation unit: Generates fault reports based on real-time fault data, predicted fault types of all key fault nodes, fault maintenance optimization suggestions, and maintenance optimization results.

[0012] On the other hand, the present invention also provides a method for online detection of a collector line, comprising: Step 1: Obtain historical operating data of the collector line and collect real-time operating data of the collector line; Step 2: Determine the normal operating range of the collector line based on historical operating data and build a fault detection model; Step 3: Determine the real-time fault data of the collector line based on the real-time operation data and the normal operation range; Step 4: Based on the fault detection model, perform fault detection on the real-time fault data of the collector line and generate a fault report.

[0013] Compared with the prior art, the present invention has the following advantages: By acquiring the historical operating data of the collector line, collecting the real-time operating data of the collector line, determining the normal operating range, building a fault detection model, determining the real-time fault data of the collector line based on the real-time operating data and the normal operating range, and combining the fault detection model to perform fault detection and generate a fault report, it is possible to achieve real-time and accurate operation data collection and processing, accurately identify and locate key fault nodes, achieve the efficiency and accuracy of intelligent fault detection, improve the safety and stability of the collector line, reduce the risk of outages, and optimize the allocation of maintenance resources and maintenance strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0015] Figure 1 It is a structural schematic diagram of an online detection system for a collector line provided by an embodiment of the present invention.

[0016] Figure 2 It is a flow chart of an online detection method for a collector line provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0018] Example 1: The embodiment of the present invention provides a collector line online detection system, such as Figure 1 Shown, including: Collection module: obtains historical operation data of the collector line and collects real-time operation data of the collector line; Construction module: Determine the normal operating range of the collector line based on historical operating data and build a fault detection model; Determination module: determines the real-time fault data of the collector line based on the real-time operation data and the normal operation range; Fault module: Based on the fault detection model, it performs fault detection on the real-time fault data of the collector line and generates a fault report.

[0019] In this embodiment, historical and real-time operating data of the collector circuit are acquired. The historical operating data includes information such as the equipment's past operating status and fault conditions, while the real-time operating data is real-time parameters (such as voltage, current, and temperature) dynamically collected during actual operation.

[0020] In this embodiment, the normal operating status of the collector circuit is analyzed based on collected historical operating data to determine its normal operating range. This range includes the upper and lower limits of all electrical parameters and represents the range of parameter fluctuations during normal operation. A fault detection model is also required to identify potential faults based on real-time fault data.

[0021] In this embodiment, the operating status of the collector circuit is monitored in real time by combining real-time operating data with a determined normal operating range to identify and determine whether a fault exists. By comparing the real-time data with the normal operating range, if the data exceeds the normal range, real-time fault data of the collector circuit is determined.

[0022] In this embodiment, the constructed fault detection model is used to analyze real-time fault data, and a fault report is generated based on the model results.

[0023] The beneficial effects of the above technical solution are: by obtaining the historical operation data of the collector line, collecting the real-time operation data of the collector line, determining the normal operation range, building a fault detection model, determining the real-time fault data of the collector line based on the real-time operation data and the normal operation range, and combining the fault detection model to perform fault detection and generate a fault report, real-time and accurate operation data collection and processing can be achieved, key fault nodes can be accurately identified and located, the efficiency and accuracy of intelligent fault detection can be achieved, the safety and stability of the collector line can be improved, the risk of downtime can be reduced, and the allocation of maintenance resources and maintenance strategies can be optimized.

[0024] Example 2: The embodiment of the present invention provides a collection line online detection system, a collection module, including: Historical sub-operation data unit: obtains historical sub-operation data of the collector line in multiple historical specified time periods, wherein the historical sub-operation data includes historical normal sub-operation data and historical fault sub-operation data. The historical normal sub-operation data includes a historical normal operation matrix, and the historical fault sub-operation data includes a historical fault operation vector and a historical fault type. Historical normal operation data and historical fault operation data unit: determines historical normal operation data based on historical normal sub-operation data of the collector line in all historical sub-operation data of the historical specified time period, and at the same time, determines historical fault operation data based on historical fault sub-operation data of the collector line in all historical sub-operation data of the historical specified time period; Historical operation data unit: determines historical normal operation data and historical fault operation data.

[0025] In this embodiment, historical sub-operation data of the collector line within multiple historical specified time periods are obtained, and these data are divided into two categories: historical normal sub-operation data: including detailed data of the collector line during normal operation, specifically expressed as a historical normal operation matrix, which is a two-dimensional data structure that records the normal values of various indicators (such as current, temperature, etc.) at different time points; historical fault sub-operation data: including detailed data of the collector line when a fault occurs, specifically expressed as a historical fault operation vector, which is a one-dimensional data structure that describes the key indicators when the fault occurs; at the same time, it also includes historical fault types, that is, the type of fault (such as short circuit, overload, etc.).

[0026] In this embodiment, based on the historical normal sub-operation data in all historical specified time periods, complete historical normal operation data is determined to provide benchmark data for subsequent analysis. At the same time, based on the historical fault sub-operation data in all historical specified time periods, complete historical fault operation data is determined to identify and warn of potential faults in the future.

[0027] In this embodiment, the historical normal operation data and the historical fault operation data are integrated to form the final historical operation data.

[0028] The beneficial effects of the above technical solution are: obtaining historical operating data of the collector line can provide an accurate data basis for determining the normal operating range and building a fault detection model, thereby improving the accuracy of fault detection.

[0029] Example 3: The embodiment of the present invention provides a collection line online detection system, a collection module, including: Operation area unit: determining a plurality of operation areas based on historical fault operation data in the historical operation data; Key node data unit: Based on the topological structure of the collector line and all operating areas, key node data is determined. The key node data includes multiple key nodes, the data collection requirements of each key node, and the location of the key nodes. Installation unit: Install sensor groups at corresponding key node locations based on the data collection requirements of each key node; Real-time operation data unit: collects real-time operation sub-data of each key node of the collector line based on the sensor group, pre-processes the real-time operation sub-data of each key node, and determines the real-time operation data based on the pre-processed real-time operation sub-data of all key nodes.

[0030] In this embodiment, multiple operating areas are determined by analyzing the fault data of the collector line during historical operation. The operating area can be a place where the load changes greatly: for some areas with large load fluctuations, such as large shopping malls, factory production workshops, etc., the cables need to be monitored at the nodes in these areas. Sudden changes in load may cause large fluctuations in the current and voltage of the cable, affecting the insulation of the cable and the operation of the equipment. By installing sensors at these nodes, the load changes can be monitored in real time, and the power supply strategy can be adjusted in time to avoid overloaded operation of the cable; the middle position of long-distance cables: long-distance cables will have problems such as voltage drop and heat generation during transmission. The operating conditions in the middle position have an important impact on the performance of the entire cable line. By setting key nodes in the middle position of long-distance cables and installing sensors to monitor the temperature, voltage and other parameters of the cable, abnormal conditions in the middle part of the cable can be discovered in time, and corresponding measures can be taken to deal with them to ensure stable operation of the cable.

[0031] In this embodiment, key node data is determined based on the topology of the collector line. Key nodes are important points in the collector line that affect system operation, such as substations and connection points. The data collection requirements for each key node include the type and number of sensors, as well as the parameters they need to collect (such as current, voltage, and temperature). The locations of key nodes are also determined in this step to ensure precise installation.

[0032] In this embodiment, based on the data collection requirements of each key node, the system will install appropriate sensor groups at the corresponding key node locations. These sensors are used to collect various operating data of the collector line in real time.

[0033] In this embodiment, after the sensor group is installed, the sensor begins to collect real-time operating sub-data for each key node. This data undergoes pre-processing (such as denoising and normalization) and is combined with the data from all key nodes to form the final real-time operating data.

[0034] The beneficial effects of the above technical solution are: real-time collection of real-time operating data of the collector line can ensure real-time monitoring of key points of the collector line, realize real-time and accurate operation data collection and processing, and provide data basis for determining real-time fault data.

[0035] Example 4: The embodiment of the present invention provides a collector line online detection system, comprising a building block, including: Key node category unit: performing cluster analysis based on historical normal operation vectors in all historical normal sub-operation data in the historical normal operation data to determine multiple key node categories, wherein each key node category includes multiple key nodes; Node category data unit: extracts historical normal operation vectors of all key nodes in each key node category from the historical normal operation matrix of each historical normal sub-operation data, and determines node category data of each key node category based on all historical normal operation vectors of each key node category extracted from the historical normal operation matrix of all historical normal sub-operation data; Normal operation matrix unit: based on the node category data of each key node category, determining the normal operation matrix of each node category data; Normal operation range unit: determines the normal operation range of the collector line based on the normal operation matrix of the node category data of all key node categories; Construction unit: taking the historical fault operation vectors in all the historical fault sub-operation data in the historical fault operation data as the input of the fault detection model, and taking the historical fault types in all the historical fault sub-operation data in the historical fault operation data as the output of the fault detection model; Training unit: training a fault detection model based on historical fault operation data in the historical operation data.

[0036] In this embodiment, cluster analysis is used to identify and determine multiple key node categories based on the historical normal operation sub-data within the historical normal operation data, particularly the historical normal operation vectors. Cluster analysis groups nodes with similar operating characteristics to create different key node categories. Each category includes multiple key nodes with similar characteristics.

[0037] In this embodiment, all historical normal operation vectors under each key node category are extracted from the historical normal operation matrix, and node category data of each key node category is determined based on these data.

[0038] In this embodiment, the node category data of each key node category is used to construct a normal operation matrix for each node category. The normal operation matrix records the operating status data of each node under normal conditions, including upper and lower limits of parameters such as current, temperature, and pressure.

[0039] In this embodiment, the overall normal operating range of the collector line is determined based on the normal operating matrices of all key node categories. This range is derived from the matrix data of each node category and represents the normal operating conditions of the collector line.

[0040] In this embodiment, the historical fault operation vectors in the historical fault operation data are used as input and the historical fault types are used as output to construct a fault detection model. The goal of this model is to predict possible future faults by learning the characteristics of the fault operation data.

[0041] In this embodiment, a machine learning algorithm is used to train a fault detection model based on historical fault operation data. The training process involves adjusting model parameters so that it can accurately identify historical fault types and provide effective fault warnings in actual applications.

[0042] The beneficial effects of the above technical solution are: determining the normal operating range of the collector line based on historical operating data and building a fault detection model, which can accurately identify the normal operating matrix of different nodes, realize intelligent and efficient fault detection, improve the safety and stability of the collector line, reduce the risk of outage and optimize maintenance strategies.

[0043] Example 5: An embodiment of the present invention provides a collector line online detection system, a normal operation matrix unit, comprising: ; ; ; ; ; ; in, represents the normal operation matrix of the k-th node category data, Respectively represent the lower limit and upper limit of the first feature of the k-th node category data, They represent the lower limit and upper limit of the jth feature of the kth node category data, respectively. They represent the lower limit and upper limit of the N1th feature of the kth node category data, respectively. N1 represents the number of features of the historical normal operation vector. represents the average value of the jth feature of the kth node category data, represents the standard deviation of the jth feature of the kth node category data, The eigenvalue of the jth feature of the centroid running vector of the kth node category data, The characteristic value of the jth characteristic of the historical normal operation vector of the i-th key node in the key node category of the k-th node category data in the t-th historical specified time period, represents the number of historical normal operation vectors in the k-th node category data, The eigenvalue of the jth feature of the historical normal operation vector of the i-th key node in the key node category of the k-th node category data in all historical specified time periods and the eigenvalue of the jth feature of the centroid running vector of the kth node category data The similarity value of represents the time smoothing factor, represents the first adjustment parameter, represents the second adjustment parameter, represents the third adjustment parameter, Represents the median of all j-th features of the k-th node category data after sorting, represents the 5% quantile after sorting all j-th features of the k-th node category data, Represents the 95th percentile after sorting all j-th features of the k-th node category data.

[0044] In this embodiment, the first adjustment parameter is the standard deviation of the jth feature based on the kth node category data The adjustment parameters.

[0045] In this embodiment, the second adjustment parameter The number is the average value of the jth feature based on the kth node category data and the eigenvalue of the jth feature of the centroid running vector of the kth node category data The absolute value of the difference is the adjustment parameter.

[0046] In this embodiment, the third adjustment parameter The adjustment parameter is the difference between the median and the 5% quantile after sorting all j-th features of the k-th node category data, and the 95% quantile and the median after sorting all j-th features of the k-th node category data.

[0047] The beneficial effects of the above technical solution are: based on the node category data of each key node category, the normal operation matrix of each node category data is determined, which can provide an accurate data basis for determining the normal operation range and realize intelligent and efficient fault detection.

[0048] Example 6: An embodiment of the present invention provides a system for online detection of a collector line, including a determination module, comprising: Real-time operation vector unit: extracts features from the pre-processed real-time operation sub-data of each key node in the real-time operation data, and determines the real-time operation vector of each key node; Faulty key node unit: judge the real-time operation vector of each key node against the normal operation matrix of the node category data of the key node category corresponding to each key node in the normal operation range. If the eigenvalue of any feature of the real-time operation vector of the key node is not between the lower limit and upper limit of the corresponding normal operation matrix, it is determined to be a faulty key node; otherwise, it is determined to be a normal key node; Real-time fault data unit: determines real-time fault data based on all fault-critical nodes and the real-time operation vector of each fault-critical node.

[0049] In this embodiment, after preprocessing the real-time operating sub-data of each key node, key features are extracted to form a real-time operating vector. The real-time operating vector includes the characteristic values of all relevant parameters of each key node during real-time monitoring, such as current, voltage, and temperature, to reflect the health status of the node.

[0050] In this embodiment, the real-time operation vector of each key node is compared with its corresponding normal operation matrix. If a certain eigenvalue in the real-time operation vector exceeds the range formed by the upper and lower limits of the normal operation matrix for that node category, the node is determined to be a faulty key node. If all eigenvalues are within the normal range, the node is considered a normal key node.

[0051] In this embodiment, real-time fault data is generated by using all key fault nodes and their real-time operation vectors, which include characteristic information of all key fault nodes.

[0052] The beneficial effects of the above technical solution are: based on real-time operating data and normal operating range to determine the real-time fault data of the collector line, it can accurately identify and locate the key fault nodes, improve the operating efficiency and fault response speed of the collector line, optimize the maintenance strategy, reduce system downtime and maintenance costs, and enhance the stability and safety of the line.

[0053] Example 7: An embodiment of the present invention provides a collector line online detection system, a fault module, including: Fault prediction unit: Inputs all real-time operation vectors in real-time fault data into the fault detection model, and determines the predicted fault type and fault maintenance optimization suggestions for each key fault node based on the output structure of the fault detection model; Maintenance optimization unit: performs maintenance optimization for each critical fault node based on the predicted fault type and fault maintenance optimization suggestions; Generation unit: Generates fault reports based on real-time fault data, predicted fault types of all key fault nodes, fault maintenance optimization suggestions, and maintenance optimization results.

[0054] In this embodiment, all real-time operating vectors from real-time fault data are input into a trained fault detection model. Using the model's output, the system predicts the likely fault type (e.g., overload, short circuit, etc.) at each critical fault node and generates optimized maintenance recommendations based on the predictions. These recommendations may include repairs, component replacements, increased monitoring, and other actions, aiming to optimize maintenance processes and reduce downtime.

[0055] In this embodiment, specific maintenance optimization measures are formulated based on the predicted fault type and fault maintenance optimization suggestions for each critical fault node. These measures help to effectively maintain the critical fault node, ensure that the system returns to normal operation, and minimize the impact of the fault on the system.

[0056] In this embodiment, a fault report is generated based on real-time fault data, predicted fault types for all critical fault nodes, fault maintenance optimization suggestions, and maintenance optimization results. The report details the fault diagnosis results, predicted fault types, maintenance optimization suggestions, and implemented maintenance measures.

[0057] The beneficial effects of the above technical solution are: fault detection and fault report generation based on the real-time fault data of the collector line based on the fault detection model can improve the accuracy of fault detection, and the proactive maintenance strategy can effectively reduce system downtime, optimize the allocation of maintenance resources, reduce maintenance costs, and improve the operational safety and stability of the collector line.

[0058] Example 8: The embodiment of the present invention provides a method for online detection of a collector circuit, such as Figure 2 As shown, including: Step 1: Obtain historical operating data of the collector line and collect real-time operating data of the collector line; Step 2: Determine the normal operating range of the collector line based on historical operating data and build a fault detection model; Step 3: Determine the real-time fault data of the collector line based on the real-time operation data and the normal operation range; Step 4: Based on the fault detection model, perform fault detection on the real-time fault data of the collector line and generate a fault report.

[0059] The beneficial effects of the above technical solution are: by obtaining the historical operation data of the collector line, collecting the real-time operation data of the collector line, determining the normal operation range, building a fault detection model, determining the real-time fault data of the collector line based on the real-time operation data and the normal operation range, and combining the fault detection model to perform fault detection and generate a fault report, real-time and accurate operation data collection and processing can be achieved, key fault nodes can be accurately identified and located, the efficiency and accuracy of intelligent fault detection can be achieved, the safety and stability of the collector line can be improved, the risk of downtime can be reduced, and the allocation of maintenance resources and maintenance strategies can be optimized.

[0060] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0061] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a ROM / RkM, a magnetic disk, or an optical disk, and includes a number of instructions for enabling a computer device (such as a personal computer, a server, or a network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A collector line online detection system, characterized in that: include: Collection module: obtains historical operation data of the collector line and collects real-time operation data of the collector line; Construction module: Determine the normal operating range of the collector line based on historical operating data and build a fault detection model; Determination module: determines the real-time fault data of the collector line based on the real-time operation data and the normal operation range; Fault module: Based on the fault detection model, it performs fault detection on the real-time fault data of the collector line and generates a fault report.

2. The online detection system for collector lines according to claim 1, characterized in that: Acquisition module, including: Historical sub-operation data unit: obtains historical sub-operation data of the collector line in multiple historical specified time periods, wherein the historical sub-operation data includes historical normal sub-operation data and historical fault sub-operation data. The historical normal sub-operation data includes a historical normal operation matrix, and the historical fault sub-operation data includes a historical fault operation vector and a historical fault type. Historical normal operation data and historical fault operation data unit: determines historical normal operation data based on historical normal sub-operation data of the collector line in all historical sub-operation data of the historical specified time period, and at the same time, determines historical fault operation data based on historical fault sub-operation data of the collector line in all historical sub-operation data of the historical specified time period; Historical operation data unit: determines historical normal operation data and historical fault operation data.

3. The online detection system for collector lines according to claim 2, characterized in that: Acquisition module, including: Operation area unit: determining a plurality of operation areas based on historical fault operation data in the historical operation data; Key node data unit: Based on the topological structure of the collector line and all operating areas, key node data is determined. The key node data includes multiple key nodes, the data collection requirements of each key node, and the location of the key nodes. Installation unit: Install sensor groups at corresponding key node locations based on the data collection requirements of each key node; Real-time operation data unit: collects real-time operation sub-data of each key node of the collector line based on the sensor group, pre-processes the real-time operation sub-data of each key node, and determines the real-time operation data based on the pre-processed real-time operation sub-data of all key nodes.

4. The online detection system for collector lines according to claim 2, characterized in that: Building blocks, including: Key node category unit: performing cluster analysis based on historical normal operation vectors in all historical normal sub-operation data in the historical normal operation data to determine multiple key node categories, wherein each key node category includes multiple key nodes; Node category data unit: extracts historical normal operation vectors of all key nodes in each key node category from the historical normal operation matrix of each historical normal sub-operation data, and determines node category data of each key node category based on all historical normal operation vectors of each key node category extracted from the historical normal operation matrix of all historical normal sub-operation data; Normal operation matrix unit: based on the node category data of each key node category, determining the normal operation matrix of each node category data; Normal operation range unit: determines the normal operation range of the collector line based on the normal operation matrix of the node category data of all key node categories; Construction unit: taking the historical fault operation vectors in all the historical fault sub-operation data in the historical fault operation data as the input of the fault detection model, and taking the historical fault types in all the historical fault sub-operation data in the historical fault operation data as the output of the fault detection model; Training unit: training a fault detection model based on historical fault operation data in the historical operation data.

5. The online detection system for collector lines according to claim 4, characterized in that: Normal operation of the matrix unit includes: ； ； ； ； ； ； in, represents the normal operation matrix of the k-th node category data, Respectively represent the lower limit and upper limit of the first feature of the k-th node category data, They represent the lower limit and upper limit of the jth feature of the kth node category data, respectively. They represent the lower limit and upper limit of the N1th feature of the kth node category data, respectively. N1 represents the number of features of the historical normal operation vector. represents the average value of the jth feature of the kth node category data, represents the standard deviation of the jth feature of the kth node category data, The eigenvalue of the jth feature of the centroid running vector of the kth node category data, The characteristic value of the jth characteristic of the historical normal operation vector of the i-th key node in the key node category of the k-th node category data in the t-th historical specified time period, represents the number of historical normal operation vectors in the k-th node category data, The eigenvalue of the jth feature of the historical normal operation vector of the i-th key node in the key node category of the k-th node category data in all historical specified time periods and the eigenvalue of the jth feature of the centroid running vector of the kth node category data The similarity value of represents the time smoothing factor, represents the first adjustment parameter, represents the second adjustment parameter, represents the third adjustment parameter, Represents the median of all j-th features of the k-th node category data after sorting, represents the 5% quantile after sorting all j-th features of the k-th node category data, Represents the 95th percentile after sorting all j-th features of the k-th node category data.

6. The online detection system for collector lines according to claim 4, characterized in that: Identify modules, including: Real-time operation vector unit: extracts features from the pre-processed real-time operation sub-data of each key node in the real-time operation data, and determines the real-time operation vector of each key node; Faulty key node unit: judge the real-time operation vector of each key node against the normal operation matrix of the node category data of the key node category corresponding to each key node in the normal operation range. If the eigenvalue of any feature of the real-time operation vector of the key node is not between the lower limit and upper limit of the corresponding normal operation matrix, it is determined to be a faulty key node; otherwise, it is determined to be a normal key node; Real-time fault data unit: determines real-time fault data based on all fault-critical nodes and the real-time operation vector of each fault-critical node.

7. The online detection system for collector lines according to claim 1, characterized in that: Fault modules, including: Fault prediction unit: Inputs all real-time operation vectors in real-time fault data into the fault detection model, and determines the predicted fault type and fault maintenance optimization suggestions for each key fault node based on the output structure of the fault detection model; Maintenance optimization unit: performs maintenance optimization for each critical fault node based on the predicted fault type and fault maintenance optimization suggestions; Generation unit: Generates fault reports based on real-time fault data, predicted fault types of all key fault nodes, fault maintenance optimization suggestions, and maintenance optimization results.

8. A method for online detection of a collector line, characterized in that: include: Step 1: Obtain historical operating data of the collector line and collect real-time operating data of the collector line; Step 2: Determine the normal operating range of the collector line based on historical operating data and build a fault detection model; Step 3: Determine the real-time fault data of the collector line based on the real-time operation data and the normal operation range; Step 4: Based on the fault detection model, perform fault detection on the real-time fault data of the collector line and generate a fault report.

Images