Event discovering and tracking method and system based on real-time signals of hydropower station

Abstract

The invention relates to the technical field of hydropower stations, and discloses an event discovery and tracking method and system based on real-time signals of a hydropower station, and the method comprises the steps: building a typical incidence relation event library of the hydropower station, obtaining a unique identification signal or a signal combination of an event, obtaining a mapping relation between the event and the signal, and carrying out the tracking of the event based on the relation between equipment and the event; establishing a device-event-signal relation graph; obtaining multi-source real-time data and carrying out data fusion to obtain a fused real-time data stream; based on an equipment-event-signal relation graph, performing event searching, event dynamic tracking and event overtime detection on the fused real-time data flow by adopting an event finding and tracking algorithm, judging whether event signals in the real-time data flow are found or not, if so, performing event shielding after grading according to importance grades of events, and otherwise, performing event shielding according to importance grades of the events. Judging whether the real-time data flow forms a new event or not; the operation efficiency, safety and reliability of the hydropower station are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hydropower stations, and in particular to an event discovery and tracking method and system based on real-time signals of a hydropower station. Background Art

[0002] Centralized and unified monitoring of the signals of the eight stations under the Dadu River faces challenges such as complex and diverse descriptions of event signals and extreme reliance on manual judgment and experience analysis. In addition, multiple events may overlap at the same time, resulting in a large influx of signals in a short period of time. However, there are still many problems in relying solely on experienced technicians to analyze signals and take corresponding measures. First, the tasks have a variety of equipment types, a large amount of signals, complex data processing, frequent alarms, difficult monitoring work, and high risks. Therefore, even experienced technicians need to spend a lot of time and human resources to analyze these signals. Secondly, operational safety faces challenges, and signal monitoring requires manual 24-hour uninterrupted monitoring. This high-intensity and high-pressure work requires technicians to always concentrate highly and keep a clear mind in order to detect abnormalities in time and take correct countermeasures. In addition, signal data is complex, and emergency and non-emergency events are intertwined. Considering the "zero fault tolerance" safety requirements of signal monitoring, even experienced technicians need to invest a lot of time to sort out and analyze these signals. Finally, the experience of technicians mainly comes from the accumulation of events that have been handled, so it usually takes a long time to train experienced technicians. Therefore, adopting the above method will lead to low operating efficiency of the hydropower station, and too high dependence on technical personnel will increase the workload of technical personnel. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides an event discovery and tracking method and system based on real-time signals of a hydropower station. Through the historical signal data and historical operation instruction data of the hydropower station, a relationship map of equipment-event-signal is constructed, and the system is used to analyze the collected data stream to realize automatic analysis and monitoring of the signal, so as to reduce the workload of technicians and improve the operating efficiency of the hydropower station.

[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0005] An event discovery and tracking method based on real-time signals of a hydropower station comprises the following steps:

[0006] S1. Based on the historical signal data and historical operation instruction data of the hydropower station, a typical association relationship event library is established to obtain the unique identification signal or signal combination of the event and obtain the mapping relationship between the event and the signal;

[0007] S2. According to the mapping relationship between the event and the signal obtained in step S1, and based on the relationship between the device and the event, a device-event-signal relationship graph is established;

[0008] S3, acquiring multi-source real-time data, and performing data fusion on the acquired multi-source real-time data to obtain a fused real-time data stream;

[0009] S4. According to the fused real-time data stream obtained in step S3, and based on the device-event-signal relationship map established in step S2, an event discovery and tracking algorithm is used to perform three processes of event search, event dynamic tracking, and event timeout detection on the fused real-time data stream to discover the event signal of the real-time data stream.

[0010] Furthermore, step S1 specifically includes:

[0011] S11. Obtain the events of various scenarios involved in the daily dispatching process of the historical hydropower station, analyze the correlation between the events and the characteristic signals, and based on the logical description of the correlation between the characteristic signals and the events, convert the characteristic signals into events, and establish a typical correlation event library;

[0012] S12, obtaining the operation instructions involved in the daily dispatching process of the historical hydropower station and the instruction events corresponding to the operation instructions, and based on the historical monitoring system signal data, using the frequent pattern mining method to obtain the signal data corresponding to the instruction events, and inputting the signal data into the typical association relationship event library established in step S11 to obtain the typical association relationship event library;

[0013] S13, collecting each event in the typical association relationship event library obtained in step S12, and obtaining a signal of each event;

[0014] S14, using a continuous digital coding method to encode the signal of each event obtained in step S13 to obtain an event-coded signal;

[0015] S15, according to the event coded signal obtained in step S14, perform full permutation and combination on the coded signal of each event to obtain various combinations of each event, and traverse them to obtain the number of combinations after the event traversal;

[0016] S16, determine whether the number of combinations after the event traversal obtained in step S15 is the same as the number of combinations of other events, if so, execute step S17, otherwise, stop looking for combinations exceeding the current length and obtain the signal combination length;

[0017] S17, continue to increase the number of event combinations until the number of event combinations is equal to the number of all signals included in the event, and obtain the signal combination length;

[0018] S18. Obtain a unique identification signal or signal combination of the event according to the signal combination length obtained in steps S16 and S17.

[0019] Furthermore, in step S12, the process of using the frequent pattern mining method to obtain the signal data corresponding to the instruction event and inputting the signal data into the typical association relationship event library established in step S11 is as follows:

[0020] The historical monitoring system signal data of the operation instructions in different time periods are obtained, and the historical monitoring system signal data are sorted in chronological order to obtain the signal data of the same operation instructions in different time periods. The frequent pattern mining method is used to find the signal data with higher frequency in different time periods, and these signal data are collected to form signal subsets, and it is determined whether these signal subsets constitute a complete event. If so, the signal data in the signal subsets are added to the typical association event library as a confirmed event.

[0021] Furthermore, the frequent pattern mining method used in step S12 is implemented based on the Apriori algorithm, and its algorithm steps are as follows:

[0022] First, the support of each item, i.e., a single signal, in the data set is calculated, and items with support higher than a preset threshold, i.e., the minimum support, are retained to generate candidate 1-item sets, and frequent 1-item sets are constructed.

[0023] Then, the constructed frequent 1-item sets are used to generate candidate 2-item sets, and the support of a single signal in the data set is used again to filter and obtain frequent 2-item sets;

[0024] Continue to generate larger candidate item sets and screen them until no new candidate item sets can be generated.

[0025] Furthermore, step S2 specifically includes:

[0026] S21, taking each device as a node, taking each event as a node, and using different identifiers to identify the nodes;

[0027] S22, using undirected edges to define the relationship between the main device nodes and the sub-device nodes, and using directed edges to define the relationship between the sub-device node and the main device node, and the relationship between the event node and the device node;

[0028] S23. According to the undirected edges and directed edges defined in step S22, all device nodes and event nodes are added to the graph, and directed edges between event nodes and device nodes, directed edges between sub-device nodes and main device nodes, undirected edges between main device nodes, and undirected edges between sub-device nodes are added to establish a device-event-signal relationship graph.

[0029] Furthermore, step S3 specifically includes:

[0030] S31, acquiring the signal data stream and operation instruction data generated during the daily dispatching process of the hydropower station to obtain multi-source real-time data;

[0031] S32, performing data fusion on the multi-source real-time data obtained in step S31, associating the signal data stream and the operation instruction data in the multi-source real-time data, and unifying the data formats of the signal data stream and the operation instruction data to obtain a fused real-time data stream.

[0032] Furthermore, the process of event search in step S4 is as follows:

[0033] Analyze the fused real-time data stream obtained in step S3, and if there is operation instruction data in the real-time data stream, identify the confirmed event corresponding to the real-time data stream through the operation instruction data;

[0034] If the real-time data stream is in a single form, the real-time data stream is matched with the events in the device-event-signal relationship map established in step S2, and a unique identification signal or signal combination of the matched confirmed event is determined;

[0035] If the real-time data stream corresponds to multiple events, storage space is enabled for each event in the cache, and the real-time data stream is backed up to each cache while continuing to accumulate the real-time data stream;

[0036] Determine whether a unique identification signal or signal combination exists in the currently accumulated real-time data stream according to the matched unique identification signal or signal combination of the confirmed event; if so, identify the confirmed event corresponding to the currently accumulated real-time data stream; if not, mark the currently accumulated real-time data stream as an event to be confirmed, and continue to accumulate the real-time data stream;

[0037] When the accumulated real-time data streams reach the set quantity threshold, the similarity between the event to be confirmed and the matched multiple confirmed events is calculated to obtain the event with the highest similarity that exceeds the threshold, and mark it as the event of the real-time data stream.

[0038] Furthermore, the process of event dynamic tracking in step S4 is as follows:

[0039] When the real-time data stream corresponds to multiple events, a cache of multiple events is created, and the cache of multiple events is monitored and maintained. The accumulation of the real-time data stream is backed up to the cache corresponding to the matched multiple events. If a unique event is identified, the cache of the identified event is retained, and the caches of other matched events are deleted.

[0040] Furthermore, the event timeout detection in step S4 includes fixed time limit timeout detection and dynamic time limit timeout detection, and the specific process is as follows:

[0041] Fixed time limit timeout detection sets a fixed time for the confirmed events according to the confirmed events identified in the real-time data stream. When the accumulated time of the real-time data stream exceeds the set fixed time, the current real-time data stream is submitted;

[0042] Dynamic time limit timeout detection determines the timeout period according to the actual situation of the real-time data stream. After the real-time data stream is identified as a confirmed event and the end signal of the event is identified, the current real-time data stream is no longer submitted according to a fixed time, but the accumulation of the real-time data stream is ended in advance.

[0043] Furthermore, in step S4, the three processes of event search, event dynamic tracking and event timeout detection are performed on the fused real-time data stream using the event discovery and tracking algorithm, and also include a process of determining whether an event signal in the real-time data stream is found, as follows:

[0044] If an event signal is found in the real-time data stream, an event shielding operation is added. Otherwise, the real-time data stream is transmitted to an external interface to determine whether the real-time data stream constitutes a new event. If so, it is defined as a new event and added to the typical association relationship event library. Otherwise, it is marked as a non-event signal.

[0045] Furthermore, the specific process of adding an event masking operation is as follows:

[0046] After the fused real-time data stream goes through the three processes of event search, event dynamic tracking, and event timeout detection, an event signal is found in the real-time data stream, that is, a confirmed event is found. Then, the event with the lowest importance level among the confirmed events and the number of occurrences exceeding the set threshold within the specified time range is shielded, and only one event signal information is retained within the specified time range to reduce the discovery and tracking of unimportant events.

[0047] Furthermore, the importance level is graded based on:

[0048] Based on the device-event-signal relationship graph, each device is sorted according to its importance, and the importance level of the events corresponding to the devices with high importance is set to the highest, and the importance level of the events corresponding to the devices with low importance is set to the lowest. The importance level of the events formed by the signals emitted by different devices is obtained, and the importance level information of the events is recorded in the typical association relationship event library.

[0049] Furthermore, the conditions for determining whether the real-time data stream constitutes a new event are:

[0050] Real-time data streams have event nature and are associated with devices;

[0051] The event nature refers to whether the real-time data stream has the cause signal, trigger signal, end signal, and key signal of the event.

[0052] An event discovery and tracking system based on real-time signals of a hydropower station, comprising:

[0053] The event library establishment module establishes a typical association relationship event library based on the historical signal data and historical operation instruction data of the hydropower station, and obtains the mapping relationship between events and signals by obtaining the unique identification signal or signal combination of the event;

[0054] The device event relationship map module establishes a device-event-signal relationship map based on the mapping relationship between events and signals in the typical association relationship event library established in the event library establishment module, and based on the relationship between devices and events;

[0055] The data real-time connection and preprocessing module obtains multi-source real-time data and performs fusion processing on the multi-source real-time data in time sequence to obtain the fused real-time data stream;

[0056] The event discovery and tracking module uses the event discovery and tracking algorithm to implement the three processes of event search, event dynamic tracking, and event timeout detection on the fused real-time data stream based on the fused real-time data stream obtained by the real-time data docking and preprocessing module, and the device-event-signal relationship map established based on the device event relationship map module. It discovers event signals in the real-time data stream, realizes event discovery, multi-cache construction and maintenance, and event tracking.

[0057] Furthermore, a new event adding module is added after the event discovery and tracking module, which is used to receive the real-time data stream transmitted by the event discovery and tracking module, and determine whether the real-time data stream constitutes a new event. If so, the new event is added to the typical association relationship event library of the event library establishment module; otherwise, it is marked as a non-event signal.

[0058] The present invention has the following beneficial effects:

[0059] The present invention proposes an event discovery and tracking method and system based on real-time signals of a hydropower station. By processing the real-time data stream of the hydropower station, it is possible to technically discover and track events, while reducing the continuous tracking of unimportant events, reducing the difficulty of monitoring events, and improving the operating efficiency, safety and reliability of the hydropower station. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic flow chart of an event discovery and tracking method based on real-time signals of a hydropower station proposed by the present invention;

[0061] Figure 2 This is a structural schematic diagram of an event discovery and tracking system based on real-time signals of a hydropower station proposed by the present invention;

[0062] Figure 3 Schematic diagram of the structure of the event discovery and tracking system for real-time signals of a hydropower station with modules added for adding new events. DETAILED DESCRIPTION

[0063] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0064] like Figure 1 As shown, an event discovery and tracking method based on real-time signals of a hydropower station includes the following steps S1-S4:

[0065] S1. Based on the historical signal data and historical operation instruction data of the hydropower station, a typical association event library is established to obtain the unique identification signal or signal combination of the event and obtain the mapping relationship between the event and the signal.

[0066] In this embodiment, firstly, various scenarios involved in the daily dispatching process of historical hydropower stations, such as start-up and shutdown, switching operations, accidents, etc., are studied, the events of each scenario are obtained, and the association relationship between events and characteristic signals is established, specifically, the mapping relationship between events and signals is established and the characteristic signals associated with each event are clarified, wherein the characteristic signals include the cause signal of the event, the trigger signal, the end signal, the key signal, etc. Secondly, the operation instructions involved in the daily dispatching process of historical hydropower stations and the events corresponding to the operation instructions are obtained. Specifically, the signal data corresponding to the operation instructions are obtained by using the signal data of the historical monitoring system and applying the frequent pattern mining technology, and these signal data are supplemented into the typical association relationship event library to further enrich the event information, so as to provide more accurate event discovery and tracking capabilities.

[0067] Specifically, step S1 specifically includes S11-S18:

[0068] S11. Obtain the events of various scenarios involved in the daily dispatching process of historical hydropower stations, analyze the correlation between events and characteristic signals, and based on the logical description of the correlation between characteristic signals and events, eventify the characteristic signals and establish a typical correlation event library.

[0069] S12. Obtain the operation instructions involved in the daily dispatching process of the historical hydropower station and the instruction events corresponding to the operation instructions, and based on the historical monitoring system signal data, use the frequent pattern mining method to obtain the signal data corresponding to the instruction events, and input the signal data into the typical association relationship event library established in step S11 to obtain the typical association relationship event library.

[0070] In this embodiment, the signal data obtained by the historical Kafka platform, i.e., the signal data of the historical monitoring system, is sorted to ensure that the signal data is arranged in chronological order, and according to the time information of the operation instruction, the time period in which the operation instruction is issued is determined, and the signal data of the same operation instruction in different time periods is obtained, and the signal data after sorting is analyzed by the frequent pattern mining method, and the signal data that frequently appears in different time periods is found, and these signal data are collected to form a signal subset, and these signal subsets are handed over to the technicians for verification and confirmation, and the technicians evaluate the relevance of the signal subset and the integrity of the event according to the knowledge in the field. And it is judged whether the signal subset constitutes a complete event. If so, the signal data in the signal subset is added to the typical association relationship event library and used as a confirmed event. In addition, the frequent pattern mining method used in this embodiment is implemented based on the Apriori algorithm, which can find the association relationship between data in a large-scale data set, thereby enhancing the accuracy of event matching.

[0071] Specifically, the process of using the frequent pattern mining method to obtain the signal data corresponding to the instruction event in step S12 and inputting the signal data into the typical association relationship event library established in step S11 is as follows:

[0072] The historical monitoring system signal data of the operation instructions in different time periods are obtained, and the historical monitoring system signal data are sorted in chronological order to obtain the signal data of the same operation instructions in different time periods. The frequent pattern mining method is used to find the signal data with higher frequency in different time periods, and these signal data are collected to form signal subsets, and it is determined whether these signal subsets constitute a complete event. If so, the signal data in the signal subsets are added to the typical association event library as a confirmed event.

[0073] Specifically, the frequent pattern mining method used in step S12 is implemented based on the Apriori algorithm, and its algorithm steps are as follows:

[0074] First, the support of each item, i.e., a single signal, in the data set is calculated, and items with support higher than a preset threshold, i.e., the minimum support, are retained to generate candidate 1-item sets and construct frequent 1-item sets.

[0075] Then, the constructed frequent 1-item sets are used to generate candidate 2-item sets, and the support of a single signal in the data set is used again to filter and obtain frequent 2-item sets.

[0076] Continue to generate larger candidate item sets and screen them until no new candidate item sets can be generated.

[0077] S13, collecting each event in the typical association relationship event library obtained in step S12, and obtaining the signal of each event.

[0078] S14. Encode the signal of each event acquired in step S13 using a continuous digital encoding method to obtain an event-coded signal.

[0079] S15. According to the event coded signal obtained in step S14, the coded signal of each event is fully permuted and combined to obtain various combinations of each event, and then traversed to obtain the number of combinations after the event traversal.

[0080] S16, determine whether the number of combinations after event traversal obtained in step S15 is the same as the number of combinations of other events, if so, execute step S17, otherwise, stop looking for combinations exceeding the current length and obtain the signal combination length.

[0081] S17. Continue to increase the number of event combinations until the number of event combinations is equal to the number of all signals included in the event, and obtain the signal combination length.

[0082] S18. Obtain a unique identification signal or signal combination of the event according to the signal combination length obtained in steps S16 and S17.

[0083] Steps S13-S18 are the process of acquiring a unique identification signal or signal combination for each event. The specific process is as follows: first, the signal of each event in the typical association relationship event library of this embodiment is collected, and a unique digital code is assigned to each signal to ensure that the signal has a unique identifier to avoid confusion and conflict; second, the collected signals are standardized and normalized, the signals are mapped to a unified form, and the signals are processed continuously so as to represent the signals as a compact continuous coding form; finally, for each event, its signals are fully arranged and combined, and various possible combinations are made for all signals of each event to explore the combination of different signals, and check whether there are other events with the same signal combination. If the current number of signal combinations can uniquely represent the event, stop looking for combinations that exceed the current signal combination length, and mark them as signal combinations belonging to the event. If the current number of signal combinations cannot uniquely represent the event, continue to increase the length of the signal combination number until the signal combination size is equal to the number of all signals contained in the event. If the signal combination length of the event is 1, then the signal combination is the unique identification signal of the current event, and there is no need to traverse all other combinations whose signal combination length exceeds 1 at this time; if the signal combination length of the event exceeds 1 and the combination does not exist in other events, then the signal combination represents the signal combination of the current event.

[0084] S2. According to the mapping relationship between events and signals obtained in step S1, and based on the relationship between devices and events, a device-event-signal relationship graph is established.

[0085] In this embodiment, the mapping relationship between events and signals has been obtained in the above steps, so as long as the association relationship between events and devices is clarified, the relationship diagram of devices, events and signals can be obtained. The specific process is: first, node definition, each device and event is regarded as a node, and is identified with different identifiers, such as: "Device1", "Device2", "Event1", "Event2". Secondly, the definition of edges, different relationships are represented by the definition of edges. Undirected edges point from one main device to another main device, or from one sub-device to another sub-device, which are used to represent the relationship between devices at the same level; directed edges point from sub-devices to main device nodes, which are used to represent the hierarchical relationship between main devices and sub-devices, and are used to indicate that sub-devices belong to main devices; in addition, there are directed edges from event nodes to device nodes, which are used to represent the association between events and devices, indicating that the device is associated with the event. Finally, the component diagram, that is, in the component diagram stage, all device and event nodes are added to the diagram, and directed edges between events and devices are added to represent the relationship between them. At the same time, directed edges between sub-devices and main devices are added to establish a hierarchical structure. In order to represent the devices of the same level, undirected edges between the main devices and undirected edges between the sub-devices are added. In addition, in the constructed device-event-signal relationship graph, the signal represents the attribute of the event node.

[0086] Specifically, step S2 specifically includes S21-S23:

[0087] S21. Take each device as a node, take each event as a node, and use different identifiers to identify the nodes.

[0088] S22. Use undirected edges to define the relationship between the main device nodes and the sub-device nodes, and use directed edges to define the relationship between the sub-device node and the main device node, and the relationship between the event node and the device node.

[0089] S23. According to the undirected edges and directed edges defined in step S22, all device nodes and event nodes are added to the graph, and directed edges between event nodes and device nodes, directed edges between sub-device nodes and main device nodes, undirected edges between main device nodes, and undirected edges between sub-device nodes are added to establish a device-event-signal relationship graph.

[0090] S3. Acquire multi-source real-time data, and perform data fusion on the acquired multi-source real-time data to obtain a fused real-time data stream.

[0091] In this embodiment, multi-source real-time data is obtained by acquiring the signal data stream and operation instruction data generated in the daily dispatching process of the hydropower station. Since the signal data flow generated in the daily dispatching process is large and involves multiple types of sensors and equipment, the obtained signal data stream includes signals generated by different devices of different systems. The "system" used in this embodiment is a way of saying different components, elements, parts, parts or assemblies at different levels. At the same time, the obtained multi-source real-time data is fused and processed, and the real-time data stream in a unified format is obtained by associating and unifying the format of the operation instruction data and the signal data stream in the multi-source real-time data. The specific process is to parse and clean the signal data stream from the Kafka platform, extract useful information, analyze the operation instructions to which the signal data stream belongs, obtain the time, object and instruction content of the operation instruction data, associate the signal data stream and the instruction data stream, and unify the format of the signal data stream and the operation instruction data of the Kafka platform, complete the unified formatting and association operation of multi-source real-time data of different sources and types, and obtain the fused real-time data stream.

[0092] Specifically, step S3 specifically includes S31-S32:

[0093] S31. Acquire the signal data stream and operation instruction data generated during the daily dispatching process of the hydropower station to obtain multi-source real-time data.

[0094] S32, performing data fusion on the multi-source real-time data obtained in step S31, associating the signal data stream and the operation instruction data in the multi-source real-time data, and unifying the data formats of the signal data stream and the operation instruction data to obtain a fused real-time data stream.

[0095] S4. According to the fused real-time data stream obtained in step S3, and based on the device-event-signal relationship map established in step S2, an event discovery and tracking algorithm is used to perform three processes of event search, event dynamic tracking, and event timeout detection on the fused real-time data stream to discover the event signal of the real-time data stream.

[0096] In this embodiment, the event discovery and tracking algorithm includes: real-time monitoring, collection and analysis of the acquired real-time data stream, and converting these real-time data streams into events, separating them in the form of events. In this process, the start, continuation and end of the tracking event are included. Therefore, the event discovery and tracking algorithm is used to convert the fused real-time data stream into events, including three processes: event search, event dynamic tracking and event timeout monitoring. In addition, in order to further improve the event discovery algorithm, the algorithm in this embodiment is calculated based on the similarity of the Jaccard similarity coefficient, which can accurately identify and classify the signal stream to determine the degree of conformity with the determined event. The Jaccard similarity coefficient is a calculation method for comparing the similarity between two sets, mainly used to measure the proportion of common elements of two sets in the total elements, so as to judge the similarity between the sets. The value range of the Jaccard similarity coefficient is between 0 and 1, where 0 means that the two sets have no common elements, that is, they are completely dissimilar; 1 means that the elements of the two sets are exactly the same, that is, they are completely similar. Therefore, in practical applications, the closer the similarity coefficient is to 1, the more similar the two sets are. Therefore, using this method, the event to which the current real-time data stream belongs can be determined. Most importantly, the results of the event discovery and tracking algorithm can also serve as preliminary data for scenarios such as fault diagnosis and anomaly detection in hydropower stations, and provide support for these scenarios.

[0097] The specific process of event search is as follows: when the operation instruction data is carried in the real-time data stream, the event corresponding to the current real-time data stream is directly identified through these operation instructions; when only a single real-time data stream arrives, the real-time data stream is matched with the event in the established device-event-signal relationship map, and the unique identification signal or combination signal of the matched determined event is determined. If the real-time data signal stream corresponds to multiple events, storage space for each event should be enabled in the cache, and it should be backed up in each cache, and the real-time data stream should continue to be accumulated. According to the unique identification signal or combination signal of the matched determined event, it is determined whether such an identification signal exists in the current real-time data stream. If it exists, the event corresponding to the current real-time data stream can be determined; if there is no unique identification signal or combination signal in the matched determined event, the current real-time data stream is marked as an event to be confirmed, and accumulation continues until the accumulated real-time data stream reaches the set threshold, then the event to be confirmed is calculated for similarity with the matched multiple known events to determine the event with the highest similarity and exceeding the threshold, and it is marked as the event to which the current real-time data stream belongs. In this embodiment, the event to be confirmed is calculated for similarity with multiple matched known events to determine the event with the highest similarity and exceeding the threshold. A similarity matching algorithm is adopted and the calculation process is as follows: extracting signals from the matched events as feature sets, and calculating the similarity between the signal set in the matched buffer and each feature set. Finally, the calculated similarity is compared with a preset threshold.

[0098] Specifically, the process of event search in step S4 is:

[0099] The fused real-time data stream obtained in step S3 is analyzed. If there is operation instruction data in the real-time data stream, the confirmed event corresponding to the real-time data stream is identified through the operation instruction data.

[0100] If the real-time data stream is in a single form, the real-time data stream is matched with the events in the device-event-signal relationship map established in step S2, and a unique identification signal or signal combination of the matched confirmed event is determined.

[0101] If the real-time data stream corresponds to multiple events, storage space is enabled in the cache for each event, and the real-time data stream is backed up in each cache while continuing to accumulate the real-time data stream.

[0102] Based on the unique identification signal or signal combination of the matched confirmed event, determine whether there is a unique identification signal or signal combination in the currently accumulated real-time data stream. If so, identify the confirmed event corresponding to the currently accumulated real-time data stream. If not, mark the currently accumulated real-time data stream as an event to be confirmed, and continue to accumulate real-time data streams.

[0103] When the accumulated real-time data streams reach the set quantity threshold, the similarity between the event to be confirmed and the matched multiple confirmed events is calculated to obtain the event with the highest similarity that exceeds the threshold, and mark it as the event of the real-time data stream.

[0104] In this embodiment, dynamic tracking monitors and maintains the cache of multiple events, and regularly monitors the cache status and updates the cache content to track the development of events while retaining signal data related to the events. Through unique event confirmation and cache management strategies, only necessary information is retained and cache usage is optimized. The entire dynamic tracking process covers the entire life cycle of signal events from identification, gradual accumulation to final end. Its purpose is to monitor signal flow.

[0105] Specifically, the process of event dynamic tracking in step S4 is:

[0106] When the real-time data stream corresponds to multiple events, a cache of multiple events is created, and the cache of multiple events is monitored and maintained. The accumulation of the real-time data stream is backed up to the cache corresponding to the matched multiple events. If a unique event is identified, the cache of the identified event is retained, and the caches of other matched events are deleted.

[0107] In this embodiment, during the event search process, when the real-time data stream is identified as a confirmed event, but the real-time data stream has not received an end signal, due to real-time constraints, the signal cannot be accumulated indefinitely, so timeout detection is introduced. Once the confirmed event is confirmed, the time limit information preset in the event itself is used to determine when the event will end, and the event signal is submitted. Timeout detection includes two modes, namely fixed time limit and dynamic time limit.

[0108] Specifically, the event timeout detection in step S4 includes fixed time limit timeout detection and dynamic time limit timeout detection, and the specific process is as follows:

[0109] Fixed time limit timeout detection sets a fixed time for the confirmed events according to the confirmed events identified in the real-time data stream. When the accumulated time of the real-time data stream exceeds the set fixed time, the current real-time data stream is submitted;

[0110] Dynamic time limit timeout detection determines the timeout period according to the actual situation of the real-time data stream. After the real-time data stream is identified as a confirmed event and the end signal of the event is identified, the current real-time data stream is no longer submitted according to a fixed time, but the accumulation of the real-time data stream is ended in advance.

[0111] Specifically, in step S4, the event discovery and tracking algorithm is used to perform event search, event dynamic tracking, and event timeout detection on the fused real-time data stream, and the process of determining whether an event signal in the real-time data stream is found is also included, as follows:

[0112] If an event signal is found in the real-time data stream, an event shielding operation is added. Otherwise, the real-time data stream is transmitted to an external interface to determine whether the real-time data stream constitutes a new event. If so, it is defined as a new event and added to the typical association relationship event library. Otherwise, it is marked as a non-event signal.

[0113] In this embodiment, when the event discovery and tracking algorithm is used to perform the three processes of event search, event dynamic tracking and event timeout detection on the fused real-time data stream, a judgment process for discovering event signals in the real-time data stream is added, that is, it is judged whether there is an event signal in the real-time data stream. If it exists, an event shielding operation is added to reduce the continuous tracking of unimportant events. If it does not exist, it means that the real-time data stream does not match any event in the typical association relationship event library, that is, it is not identified as any event. Since the reason for not being identified as any event may be the incompleteness of the typical association relationship event library, the real-time data stream can be transmitted to the computer through an external interface, that is, it is manually judged whether it can constitute a new event. If so, it can be defined as an event and added to the typical association relationship event library, thereby realizing the expansion of the typical association relationship event library. Otherwise, it is marked as a non-event signal, and after being marked as a non-event signal, no judgment is required when it is encountered again.

[0114] Specifically, the specific process of adding an event shielding operation is as follows:

[0115] After the fused real-time data stream goes through the three processes of event search, event dynamic tracking, and event timeout detection, an event signal is found in the real-time data stream, that is, a confirmed event is found. Then, the event with the lowest importance level among the confirmed events and the number of occurrences exceeding the set threshold within the specified time range is shielded, and only one event signal information is retained within the specified time range to reduce the discovery and tracking of unimportant events.

[0116] Specifically, the importance level is based on:

[0117] Based on the device-event-signal relationship graph, each device is sorted according to its importance, and the importance level of the events corresponding to the devices with high importance is set to the highest, and the importance level of the events corresponding to the devices with low importance is set to the lowest. The importance level of the events formed by the signals emitted by different devices is obtained, and the importance level information of the events is recorded in the typical association relationship event library.

[0118] In this embodiment, the level of events generated by signals sent by different devices is determined by judging the correlation between events, devices and signals. The level of events generated by important devices is higher than that of events generated by unimportant devices.

[0119] Specifically, the conditions for determining whether a real-time data stream constitutes a new event are:

[0120] The real-time data stream has event properties and is associated with the device; the event properties refer to the real-time data stream having the cause signal, trigger signal, end signal, and key signal of the event.

[0121] like Figure 2 As shown, an event discovery and tracking system based on real-time signals of a hydropower station includes:

[0122] The event library establishment module establishes a typical association relationship event library based on the historical signal data and historical operation instruction data of the hydropower station, and obtains the mapping relationship between events and signals by obtaining the unique identification signal or signal combination of the event.

[0123] In this embodiment, the event library building module is responsible for collecting and storing the event signal data of the historical hydropower station, and is used to construct the event signal of the hydropower station. The module includes an event library probability table, which covers all events and corresponding time limits, etc. At the same time, the module also includes an event signal list, which records the key signal, cause signal, end signal, trigger signal, and signal timing and other related relationships and attribute information of the event.

[0124] The device event relationship map module establishes the mapping relationship between events and signals in the typical association relationship event library established in the event library establishment module, and establishes a device-event-signal relationship map based on the relationship between devices and events.

[0125] In this embodiment, the device event relationship map module creates a device event relationship map of the hydropower station based on the correspondence between events and devices, and determines the events associated with each device and the signals associated with the events by analyzing the map.

[0126] The data real-time connection and preprocessing module obtains multi-source real-time data, and performs fusion processing on the multi-source real-time data in time series to obtain the fused real-time data stream.

[0127] In this embodiment, the data real-time connection and preprocessing module is responsible for acquiring the signal data stream and operation instruction data generated by relevant equipment from the real-time monitoring system of the hydropower station, and performing fusion processing of multi-source signal streams in time sequence.

[0128] The event discovery and tracking module uses the event discovery and tracking algorithm to implement the three processes of event search, event dynamic tracking, and event timeout detection on the fused real-time data stream based on the fused real-time data stream obtained by the real-time data docking and preprocessing module, and the device-event-signal relationship map established based on the device event relationship map module. It discovers event signals in the real-time data stream, realizes event discovery, multi-cache construction and maintenance, and event tracking.

[0129] In this embodiment, the event discovery and tracking module realizes signal event discovery, multi-cache construction and maintenance, and event tracking based on the device-event-signal relationship map established by the device event relationship map module based on the fused real-time data stream, thereby realizing intelligent monitoring of a large number of real-time data streams, quickly discovering and tracking event status, and assisting the technicians of the hydropower station to operate the working status of the hydropower station to improve work efficiency, prevent accidents and make emergency responses in a timely manner. In addition, the module also has the function of judging whether the event signal in the real-time data stream is found, that is, after going through the three processes of event search, event dynamic tracking, and event timeout detection, the event signal in the real-time data stream is found, and an event shielding process can also be added to reduce the continuous tracking of unimportant events; if the event signal in the real-time data stream is not found, the real-time data stream can be transmitted to the computer through an external interface, and manual review and feedback are used.

[0130] Specifically, a new event adding module is added after the event discovery and tracking module, which is used to receive the real-time data stream transmitted by the event discovery and tracking module, and determine whether the real-time data stream constitutes a new event. If so, the new event is added to the typical association relationship event library of the event library establishment module; otherwise, it is marked as a non-event signal.

[0131] like Figure 3 As shown, Figure 3 A new event adding module is included. If the event discovery and tracking module does not find the event signal in the real-time data stream after performing event search, event dynamic tracking, and event timeout detection, it is necessary to add a new event adding module after the event discovery and tracking module, that is, the new event adding module is responsible for extracting the signal set that is not identified as an event in the real-time data stream, transmitting it to the computer through an external interface, manually reviewing and providing feedback, recording the review results, interacting with the relevant data tables of the event library, and obtaining the unique identification signal or signal combination of the new event.

[0132] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0133] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. An event discovery and tracking method based on real-time signals of hydropower stations, It is characterized in that The following steps are involved: S1. Based on the historical signal data and historical operation instruction data of the hydropower station, a typical association relationship event library is established to obtain the unique identification signal or signal combination of the event and obtain the mapping relationship between the event and the signal; S2. According to the mapping relationship between the event and the signal obtained in step S1, and based on the relationship between the device and the event, a device-event-signal relationship map is established; S3, acquiring multi-source real-time data, and performing data fusion on the acquired multi-source real-time data to obtain a fused real-time data stream; S4. According to the fused real-time data stream obtained in step S3, and based on the device-event-signal relationship map established in step S2, an event discovery and tracking algorithm is used to perform three processes of event search, event dynamic tracking, and event timeout detection on the fused real-time data stream to discover the event signal of the real-time data stream.

2. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 1, It is characterized in that Step S1 specifically includes: S11. Obtain the events of various scenarios involved in the daily dispatching process of the historical hydropower station, analyze the correlation between the events and the characteristic signals, and based on the logical description of the correlation between the characteristic signals and the events, convert the characteristic signals into events, and establish a typical correlation event library; S12, obtaining the operation instructions involved in the daily dispatching process of the historical hydropower station and the instruction events corresponding to the operation instructions, and based on the historical monitoring system signal data, using the frequent pattern mining method to obtain the signal data corresponding to the instruction events, and inputting the signal data into the typical association relationship event library established in step S11 to obtain the typical association relationship event library; S13, collecting each event in the typical association relationship event library obtained in step S12, and obtaining a signal of each event; S14, using a continuous digital coding method to encode the signal of each event obtained in step S13 to obtain an event-coded signal; S15, according to the event coded signal obtained in step S14, perform full permutation and combination on the coded signal of each event to obtain various combinations of each event, and traverse them to obtain the number of combinations after the event traversal; S16, determine whether the number of combinations after the event traversal obtained in step S15 is the same as the number of combinations of other events, if so, execute step S17, otherwise, stop looking for combinations exceeding the current length and obtain the signal combination length; S17, continue to increase the number of event combinations until the number of event combinations is equal to the number of all signals included in the event, and obtain the signal combination length; S18. Obtain a unique identification signal or signal combination of the event according to the signal combination length obtained in steps S16 and S17.

3. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 2, It is characterized in that The process of using the frequent pattern mining method to obtain the signal data corresponding to the instruction event in step S12 and inputting the signal data into the typical association relationship event library established in step S11 is as follows: The historical monitoring system signal data of the operation instructions in different time periods are obtained, and the historical monitoring system signal data are sorted in chronological order to obtain the signal data of the same operation instructions in different time periods. The frequent pattern mining method is used to find the signal data with higher frequency in different time periods, and these signal data are collected to form signal subsets, and it is determined whether these signal subsets constitute a complete event. If so, the signal data in the signal subsets are added to the typical association event library as a confirmed event.

4. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 2, It is characterized in that The frequent pattern mining method used in step S12 is implemented based on the Apriori algorithm, and its algorithm steps are as follows: First, the support of each item, i.e., a single signal, in the data set is calculated, and items with support higher than a preset threshold, i.e., the minimum support, are retained to generate candidate 1-item sets, and frequent 1-item sets are constructed. Then, the constructed frequent 1-item sets are used to generate candidate 2-item sets, and the support of a single signal in the data set is used again to filter and obtain frequent 2-item sets; Continue to generate larger candidate item sets and screen them until no new candidate item sets can be generated.

5. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 1, It is characterized in that Step S2 specifically includes: S21, taking each device as a node, taking each event as a node, and using different identifiers to identify the nodes; S22, using undirected edges to define the relationship between the main device nodes and the sub-device nodes, and using directed edges to define the relationship between the sub-device node and the main device node, and the relationship between the event node and the device node; S23. According to the undirected edges and directed edges defined in step S22, all device nodes and event nodes are added to the graph, and directed edges between event nodes and device nodes, directed edges between sub-device nodes and main device nodes, undirected edges between main device nodes, and undirected edges between sub-device nodes are added to establish a device-event-signal relationship graph.

6. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 1, It is characterized in that Step S3 specifically includes: S31, acquiring the signal data stream and operation instruction data generated during the daily dispatching process of the hydropower station to obtain multi-source real-time data; S32, performing data fusion on the multi-source real-time data obtained in step S31, associating the signal data stream and the operation instruction data in the multi-source real-time data, and unifying the data formats of the signal data stream and the operation instruction data to obtain a fused real-time data stream.

7. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 1, It is characterized in that The process of event search in step S4 is: Analyze the fused real-time data stream obtained in step S3, and if there is operation instruction data in the real-time data stream, identify the confirmed event corresponding to the real-time data stream through the operation instruction data; If the real-time data stream is in a single form, the real-time data stream is matched with the events in the device-event-signal relationship map established in step S2, and a unique identification signal or signal combination of the matched confirmed event is determined; If the real-time data stream corresponds to multiple events, storage space is enabled for each event in the cache, and the real-time data stream is backed up to each cache while continuing to accumulate the real-time data stream; Determine whether a unique identification signal or signal combination exists in the currently accumulated real-time data stream according to the matched unique identification signal or signal combination of the confirmed event; if so, identify the confirmed event corresponding to the currently accumulated real-time data stream; if not, mark the currently accumulated real-time data stream as an event to be confirmed, and continue to accumulate the real-time data stream; When the accumulated real-time data streams reach the set quantity threshold, the similarity between the event to be confirmed and the matched multiple confirmed events is calculated to obtain the event with the highest similarity that exceeds the threshold, and mark it as the event of the real-time data stream.

8. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 1, It is characterized in that The process of event dynamic tracking in step S4 is as follows: When the real-time data stream corresponds to multiple events, a cache of multiple events is created, and the cache of multiple events is monitored and maintained. The accumulation of the real-time data stream is backed up to the cache corresponding to the matched multiple events. If a unique event is identified, the cache of the identified event is retained, and the caches of other matched events are deleted.

9. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 1, It is characterized in that The event timeout detection in step S4 includes fixed time limit timeout detection and dynamic time limit timeout detection, and the specific process is as follows: Fixed time limit timeout detection sets a fixed time for the confirmed events according to the confirmed events identified in the real-time data stream. When the accumulated time of the real-time data stream exceeds the set fixed time, the current real-time data stream is submitted; Dynamic time limit timeout detection determines the timeout period according to the actual situation of the real-time data stream. After the real-time data stream is identified as a confirmed event and the end signal of the event is identified, the current real-time data stream is no longer submitted according to a fixed time, but the accumulation of the real-time data stream is ended in advance.

10. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 1, It is characterized in that In step S4, the event discovery and tracking algorithm is used to perform event search, event dynamic tracking, and event timeout detection on the fused real-time data stream, and the process of determining whether an event signal in the real-time data stream is found is also included, as follows: If an event signal is found in the real-time data stream, an event shielding operation is added. Otherwise, the real-time data stream is transmitted to an external interface to determine whether the real-time data stream constitutes a new event. If so, it is defined as a new event and added to the typical association relationship event library. Otherwise, it is marked as a non-event signal.

11. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 10, It is characterized in that The specific process of adding event masking operation is as follows: After the fused real-time data stream goes through the three processes of event search, event dynamic tracking, and event timeout detection, an event signal is found in the real-time data stream, that is, a confirmed event is found. Then, the event with the lowest importance level among the confirmed events and the number of occurrences exceeding the set threshold within the specified time range is shielded, and only one event signal information is retained within the specified time range to reduce the discovery and tracking of unimportant events.

12. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 11, It is characterized in that The importance rating is based on: Based on the device-event-signal relationship graph, each device is sorted according to its importance, and the importance level of the events corresponding to the devices with high importance is set to the highest, and the importance level of the events corresponding to the devices with low importance is set to the lowest. The importance level of the events formed by the signals emitted by different devices is obtained, and the importance level information of the events is recorded in the typical association relationship event library.

13. The event discovery and tracking method based on real-time signals of a hydropower station according to claim 12, It is characterized in that The conditions for determining whether a real-time data stream constitutes a new event are: Real-time data streams have event nature and are associated with devices; The event nature refers to whether the real-time data stream has the cause signal, trigger signal, end signal, and key signal of the event.

14. An event discovery and tracking system based on real-time signals from hydropower stations, It is characterized in that include: The event library establishment module establishes a typical association relationship event library based on the historical signal data and historical operation instruction data of the hydropower station, and obtains the mapping relationship between events and signals by obtaining the unique identification signal or signal combination of the event; The device event relationship map module establishes a device-event-signal relationship map based on the mapping relationship between events and signals in the typical association relationship event library established in the event library establishment module, and based on the relationship between devices and events; The data real-time connection and preprocessing module obtains multi-source real-time data and performs fusion processing on the multi-source real-time data in time sequence to obtain the fused real-time data stream; The event discovery and tracking module uses the event discovery and tracking algorithm to implement the three processes of event search, event dynamic tracking, and event timeout detection on the fused real-time data stream based on the fused real-time data stream obtained by the real-time data docking and preprocessing module, and the device-event-signal relationship map established based on the device event relationship map module. It discovers event signals in the real-time data stream, realizes event discovery, multi-cache construction and maintenance, and event tracking.

15. The event discovery and tracking system based on real-time signals of a hydropower station according to claim 14, It is characterized in that A new event adding module is added after the event discovery and tracking module, which is used to receive the real-time data stream transmitted by the event discovery and tracking module, and determine whether the real-time data stream constitutes a new event. If so, the new event is added to the typical association relationship event library of the event library establishment module; otherwise, it is marked as a non-event signal.

Images