Concentrator operation management method and system based on data processing

Abstract

The application discloses a concentrator operation management method and system based on data processing and belongs to the technical field of data processing. The system comprises a data sensing module, an event analysis module, an operation management module and a reporting transmission module. The data sensing module collects historical records, operation parameters and triggered events of electric energy meters through the concentrator; the event analysis module establishes a causal relationship chain according to the historical records, marks the triggered events and calculates a weight index in combination with the causal relationship chain; the operation management module is used for evaluating event states and performing screening, calculating priority indexes of the electric energy meters and generating an event record table according to the weight indexes and the priority indexes; and the reporting transmission module reports the marked events in the event record table one by one, and if the reporting fails, calculates a transmission kinetic index of the corresponding marked event, so as to determine whether to continue reporting. The application optimizes the event reporting process by the dynamic priority management and the adaptive reporting mechanism based on the event relationship chain, and improves the event management efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a concentrator operation management method and system based on data processing. BACKGROUND

[0002] In modern power systems, concentrators, as key data acquisition and transmission hubs, are indispensable to the realization of intelligent and automated operation of power grids. They efficiently collect massive amounts of electric energy meter data and accurately upload them to communication master stations, providing a solid data foundation for power companies to conduct electricity analysis, load forecasting, line loss management, and fault diagnosis. The stable operation of concentrators has a profound impact on the reliability, economy, and safety of power systems.

[0003] At present, the operation management of concentrators usually relies on manual experience and static rules, and event management and reporting are achieved through manual screening or preset strategies. This method has obvious drawbacks. Specifically: 1. Lack of event causal relationship: traditional solutions only analyze isolated events based on preset thresholds or manual rules, and cannot automatically build temporal logical chains between events, which may mislabel transient fluctuations caused by environmental disturbances as critical failures, or miss high-risk root cause events due to neglecting event propagation paths. 2. Static priority allocation and timeliness rigidity: event sorting relies on fixed weights without considering real-time state and time decay effects, leading to resource mismatch. Old and low-risk events are long-term dominant, while sudden high-risk events are delayed in response. 3. Inefficient handling mechanism for reporting failures: fixed number or interval retransmission strategies do not dynamically adjust according to event value, causing network congestion and loss of critical events, as well as repeated retries of low-weight events occupying bandwidth, and high-value events being discarded due to communication failure. The existing method cannot quantify event causality, ignores time decay, and lacks a value-aware retransmission mechanism, resulting in low efficiency of large-scale electric energy meter cluster operation and maintenance and insufficient accuracy of fault location. SUMMARY

[0004] The present application aims to provide a concentrator operation management method and system based on data processing to solve the problems raised in the background.

[0005] To solve the above technical problems, the present application provides a concentrator operation management method based on data processing, which includes dynamic priority management and adaptive reporting mechanism based on event relationship chain.

[0006] Through dynamic modeling of event correlation and adaptive allocation of resources, intelligent optimization of event reporting in power Internet of Things is achieved. The causal chain in the physical world is converted into a computable model in the data world, and communication resources are dynamically regulated accordingly. Specifically, it includes:

[0007] S100, collecting the historical records of electric energy meters, as well as real-time operation parameters and triggered events through the concentrator.

[0008] The concentrator refers to a device for centralized processing, storage, management and bidirectional transmission of meter reading data of the electric energy meter.

[0009] The historical record includes historical trigger events. The running parameter includes various quantitative indicators for describing the performance and state of the electric energy meter during operation.

[0010] The running condition of the electric energy meter is monitored through analysis and judgment of the running parameter. The running condition includes: electric energy meter parameter change, electric energy meter time difference, electric meter fault information, electric energy meter scale drop, electric energy difference, electric energy meter flyaway, electric energy meter stop, phase sequence anomaly, electric energy meter cover record, electric energy meter running state word shift, etc.

[0011] The event refers to the state change record triggered during the operation of the electric energy meter, which belongs to the automatically generated log.

[0012] The concentrator collects the historical trigger events, running parameters and automatically generated event logs of the electric energy meter in real time, and builds a complete data foundation.

[0013] As the core device, the concentrator uniformly processes, stores and transmits data, ensures the real-time and data integrity of subsequent analysis, and provides reliable input for event analysis.

[0014] S200, according to the historical record, the logical relationship between events is analyzed, and the causal relationship chain is established. The current triggered event is marked, the triggering probability of the event is analyzed combined with the causal relationship chain, and the weight index of each marked event is calculated. Specifically, it includes:

[0015] S201, the historical record of all electric energy meters is obtained, and the historical trigger events contained in the historical record are analyzed. The triggering time corresponding to each event is analyzed, and all events in each historical record are sorted and an event chain is established according to the triggering time in chronological order.

[0016] Each electric energy meter corresponds to a historical record, and each historical record corresponds to an event chain.

[0017] S202, the triggering time interval of the adjacent two events in each event chain is calculated, the adjacent events in each event chain with a triggering time interval greater than a threshold value are segmented, and the triggering time interval of the adjacent events in the segmented event chain is not greater than the threshold value.

[0018] Based on the time interval threshold, the event chain is segmented, irrelevant events are removed, the events in the chain have strong time sequence correlation, and noise interference is reduced.

[0019] S203, a relationship database is established, and event chains with more than one event are selected and put into the relationship database. The adjacent two events in the relationship database are collected as a cause-effect pair, and the events between different cause-effect pairs are not completely the same or the time sequence is different.

[0020] Each causal pair contains two adjacent events with a time sequence. The events in different causal pairs are not exactly the same, or the time sequence is different.

[0021] S204: Analyze the frequency and interval length of each causal pair in the relationship database and calculate the synchronization index. Filter out causal pairs with synchronization indexes greater than a threshold, thereby establishing a causal relationship chain. This includes:

[0022] S2041: Count the number of event chains belonging to each causal pair in the relationship database as the number of occurrences of the corresponding causal pair. Divide the number of occurrences of each causal pair by the total number of occurrences of all causal pairs to obtain the frequency of occurrence of the corresponding causal pair.

[0023] S2042. Obtaining causal pairs Number of occurrences and frequency of occurrence , and contains events and Causal pairs with a prior order Number of occurrences , substitute into the formula to calculate the causal pair The coincidence index :

[0024] ;

[0025] Where, is a constant greater than 1, is the average frequency of all causal pairs, and is a constant, For the causal pair in the relation database All events in the event chain and The standard deviation of the trigger interval.

[0026] Used to penalize causal pairs with large time interval fluctuations, The value determines the weight of the time interval standard deviation in the synchronization index.

[0027] Used to amplify the value of low-frequency causal pairs.

[0028] The value is greater than 0 to prevent division by zero errors caused by a standard deviation of zero in theoretical calculations and to provide dimensional consistency.

[0029] The synchronization index is used to quantify the strength of the association between causal pairs, screen statistically significant and temporally stable causal pairs, and exclude accidental associations and fluctuation noise.

[0030] S2043: Similarly, calculate the synchronization index of each causal pair, select the causal pairs whose synchronization index is greater than the threshold, and analyze whether these causal pairs have the same event.

[0031] S2044. Connect two causal pairs that have the same event but are located in different positions end to end to form a causal chain.

[0032] S2045. Continue to merge causal chains with the same events until there are no identical events between the causal chains.

[0033] Highly correlated causal pairs are screened through the synchronization index formula to ensure the statistical significance of the causal chain.

[0034] S205: Mark the events triggered by each electric energy meter within the last time period k, analyze the triggering probability of each event, and calculate the weight index of each marked event based on the synchronization index in the causal relationship chain. Specifically, it includes:

[0035] S2051, analyze the causal relationship chain to which each marked event belongs, and mark the event in the causal relationship chain All subsequent events as The affected objects.

[0036] S2052. Extract the influencing objects in the causal relationship chain With Marker Events For all causal pairs between The trigger probability The triggering probabilities of all causal pairs are multiplied to obtain the corresponding affected objects. The trigger probability.

[0037] When there is no event between the influencing object and the marked event, the trigger probability is the trigger probability of the corresponding causal pair of the influencing object and the marked event.

[0038] When there are multiple events and multiple paths between the influencing object and the marked event, the triggering probabilities of all causal pairs in each path are multiplied, and the triggering probability of the path with the largest result is selected as the triggering probability of the influencing object.

[0039] S2053. Filter out all included events in the causal chain and Causal pair with later order , calculate the average value of the synchronization index of these causal pairs as the influencing object Impact Index .

[0040] S2054, respectively calculate the marking events The trigger probability and influence index of each influence object of the marked event, and the occurrence number of each event in all historical records, taking the average value of the occurrence number of all events as the reference number .

[0041] S2055, obtaining the marked event occurrence number , substituting the formula to calculate the weight index of the marked event :

[0042] ;

[0043] In the formula, and are constants, is the number of influence objects of the marked event , and and are the trigger probability and influence index of the first influence object, respectively.

[0044] The trigger probability and influence index of the marked event , are used to adjust the output range.

[0045] The control event frequency affects the weight. The occurrence frequency of the marked event , the lower the frequency, the more rare, the higher the weight.

[0046] Through the calculation of the weight index, it is ensured that the high-influence and low-frequency event obtains higher weight, and the priority judgment accuracy is improved.

[0047] Based on the causal relationship chain, the event trigger probability and influence index are calculated, and finally the event importance is quantified through the weight index formula.

[0048] S300, analyze the running parameters to evaluate the event state, and screen the marked events through the state. Calculate the priority index of each electric energy meter, and generate an event record table combining the weight index of the marked event and the priority index of the electric energy meter. Specifically, it includes:

[0049] S301, obtain the current running parameters of each marked event, set the preset value and compare to judge, and then make event state judgment one by one, including event occurrence and event recovery.

[0050] ​Event occurrence refers to determining the abnormal state of the marked event record as true after comparing and analyzing the preset value and the real-time running parameter.

[0051] Event recovery refers to determining the normal state of the marked event record as not true after comparing and analyzing the preset value and the real-time running parameter.

[0052] Compare the running parameter with the preset value to distinguish event occurrence and event recovery, and ensure that only effective abnormal events are processed.

[0053] S302, screen out the marked events in the event occurrence state, classify these marked events according to the corresponding electric energy meter, and calculate the priority index according to the trigger time and weight index of the marked events under each electric energy meter .

[0054] Set different reference time lengths and time decay coefficients for each marked event, respectively calculate the time length of the trigger time of each marked event from the current time, and substitute it into the formula to calculate the priority index of the corresponding electric energy meter:

[0055] ;

[0056] In the formula, is the number of unreported marked events corresponding to the electric energy meter, is the weight index of the i-th marked event, and are the time decay coefficient and reference time length of the i-th marked event, respectively. is a constant,

[0057] is the time length of the trigger time of the i-th marked event from the current time. Through the priority index, the comprehensive priority of all unprocessed events of a single electric energy meter is dynamically calculated. Superimpose high-weight events and dynamically decay according to fault types to ensure that old faults are down-weighted and new urgent faults are prioritized.

[0058]

[0059] S303, respectively for each electric energy meter to establish a group, according to the order from large to small of the weight index, sequentially put the marked events into the corresponding group.

[0060] S304, all groups are sorted according to the priority index of the corresponding electric energy meter from large to small, and all marked events in the first v groups are selected to generate an event record table.

[0061] S304, all groups are sorted according to the priority index of the corresponding electric energy meter from large to small, and all marked events in the first v groups are selected to generate an event record table.

[0062] ​​S400, save the event record table in the data center of the concentrator, and report one by one according to the arrangement order in the event record table. Calculate the transmission kinetic energy index of the corresponding marked event each time the reporting fails, so as to judge whether to continue reporting. Specifically, it includes:

[0063] S401, save the event record table in the data center of the concentrator, and generate reporting tasks one by one according to the arrangement order of each marked event in the event record table.

[0064] S402, analyze the reporting task, and combine and report the content to be reported according to the reporting format specified in the 698 protocol.

[0065] The reported data can be transmitted in compressed and uncompressed modes. For data transmitted in compressed mode, lossless compression algorithm must be used to ensure that the decompressed data is exactly the same as the original data.

[0066] S403, after the communication master station receives the reported content and returns the reporting confirmation frame, the reporting task and the corresponding marked event in the event record table are automatically deleted.

[0067] S404, if the concentrator does not receive the reporting confirmation frame for more than a preset time, calculate the transmission kinetic energy index of the corresponding marked event of the reporting task, and automatically switch to the next reporting task for combination and reporting of the reporting frame.

[0068] Set the standard time Transmission kinetic energy index The calculation formula is as follows:

[0069] ;

[0070] In the formula, is a constant greater than 1, is the time length from the last reporting failure time of the marked event corresponding to the reporting task to the current time.

[0071] is the priority index of the marked event corresponding to the electric energy meter, is the average value of the priority indexes of all electric energy meters in the event record table.

[0072] is the weight index of the marked event, is the sum of the weight indexes of all marked events remaining to be reported in the event record table corresponding to the electric energy meter of the marked event.

[0073] Used to measure the priority of the device. Used to measure the weight proportion of the event.

[0074] Through And Add and subtract weight to adjust the importance of marked events, dynamically abandon low-value tasks, and reduce network congestion.

[0075] S405, calculate the transmission energy index of the marked event corresponding to the reporting task once for each reporting failure, and automatically delete the reporting task and the corresponding marked event in the event record table when the transmission energy index is less than a threshold.

[0076] The application also provides a concentrator operation management system based on data processing, including a data sensing module, an event analysis module, an operation management module and a reporting transmission module.

[0077] The data sensing module collects the historical records, operating parameters and triggered events of the electric energy meter through the concentrator.

[0078] The historical records, operating parameters and triggered events of the electric energy meter are collected in real time through the concentrator, wherein the concentrator is responsible for centralized processing, storage and bidirectional transmission of data.

[0079] A comprehensive and real-time data foundation is provided to ensure that the system can accurately capture the operating state and event information of the electric energy meter, lay a reliable data support for subsequent analysis, and improve the real-time and completeness of monitoring.

[0080] The event analysis module establishes a causal relationship chain according to the historical records, marks the triggered events and calculates the weight index in combination with the causal relationship chain.

[0081] Based on the historical records, the logical relationship between events is analyzed, including splitting the event chain, establishing a causal relationship chain, marking the current event, analyzing the triggering probability and impact index in combination with the causal relationship chain, and finally calculating the weight index.

[0082] By quantifying the causal relationship and probability between events, high-impact events are identified, the priority evaluation of events is optimized, the accuracy and efficiency of event analysis are improved, and the risk of misjudgment is reduced.

[0083] The operation management module is used to evaluate the event state and perform screening, calculate the priority index of each electric energy meter, and generate an event record table according to the weight index and priority index.

[0084] The operating parameters are analyzed to evaluate the event state, and the marked events of the event occurrence state are screened out; the priority index of the electric energy meter is calculated, and the events are sorted according to the weight index to generate an event record table.

[0085] Intelligent screening and sorting of events ensure that high-weight and high-priority events are processed first, optimize resource allocation and event response, and improve the processing efficiency of the system and the orderliness of event management.

[0086] The reporting transmission module reports the marked events in the event record table one by one, and if the reporting fails, the transmission kinetic energy index of the corresponding marked event is calculated, so as to determine whether to continue reporting.

[0087] The event record table is stored in a data center and is reported one by one in sequence; when the reporting fails, the transmission kinetic energy index is calculated to determine whether to continue reporting or delete the task.

[0088] The reporting failure is dynamically processed, invalid retries are avoided, network congestion is reduced, reliable transmission of key events is ensured, and the communication efficiency and overall reliability of the system are improved.

[0089] Compared with the prior art, the beneficial effects achieved by the present application are:

[0090] Causal correlation quantitative advantage: the traditional scheme relies on manual rule analysis of isolated events, and the present technology constructs a causal relationship chain through event chain segmentation (based on a time interval threshold) and causal pair screening (calculating a synchronization index), automatically identifies the timing logic and propagation path between events, significantly improves the accuracy of root cause positioning, and avoids misjudgment of non-key events such as environmental interference.

[0091] Dynamic priority decision advantage: the existing method uses fixed event weights, and the present technology combines event state judgment (occurrence / restoration) and time decay mechanism (dynamically calculating priority index based on trigger duration and reference duration), realizes resource tilt to high-weight and high-urgency events, solves the problem of resource occupation by old and low-risk events, and improves the response efficiency of sudden faults.

[0092] Adaptive reporting mechanism advantage: the traditional reporting failure adopts a fixed retry strategy, and the present technology introduces a transmission kinetic energy index (comprehensive event value, waiting time and network load), dynamically gives up low-value retry tasks, avoids network congestion, ensures that high-priority events are reached, and optimizes the utilization rate of communication resources.

[0093] System coordination closed loop advantage: breaking through the limitation of data fragmentation, through four-module linkage (data perception→event analysis→operation management→reporting transmission), an intelligent decision-making closed loop is formed: historical causal chain drives current event analysis, dynamic priority guides reporting sequence, and efficient operation and maintenance of large-scale power meter clusters is realized. BRIEF DESCRIPTION OF DRAWINGS

[0094] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0095] Figure 1 is a flow diagram of the concentrator operation management method based on data processing of the present application;

[0096] Figure 2is the event processing flowchart of the concentrator operation management method based on data processing of the application;

[0097] Figure 3 is the message processing flowchart of the concentrator operation management method based on data processing of the application;

[0098] Figure 4 is the structure schematic diagram of the concentrator operation management system based on data processing of the application. DETAILED DESCRIPTION

[0099] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0100] Please refer to Figure 1 The application provides a concentrator operation management method based on data processing, dynamic priority management and adaptive reporting mechanism based on event relationship chain.

[0101] Through dynamic modeling of event correlation and resource adaptive allocation, intelligent optimization of event reporting in the power internet of things is realized. The causal chain of the physical world is converted into a computable model in the data world, and the communication resources are dynamically regulated and controlled accordingly. Specifically, it includes:

[0102] S100, collecting the historical records of the electric energy meter, and real-time operation parameters and triggered events through the concentrator.

[0103] The concentrator refers to a device for centralized processing, storage, management and bidirectional transmission of meter reading data of the electric energy meter.

[0104] The historical records include historical triggered events. The operation parameters include various quantitative indicators for describing the performance and state of the electric energy meter during the operation process.

[0105] The specific content of the operation parameters is shown in Table 1:

[0106] Table 1

[0107] Serial number Data item 1 Voltage 2 Current 3 Zero sequence current 4 Total and phase active power 5 Total and phase reactive power 6 Power factor 7 Active maximum demand and time of occurrence 8 Voltage, current phase angle 9 Forward active energy indication 10 Forward reactive energy indication 11 Reverse active energy indication 12 Reverse reactive energy indication 13 One quadrant reactive energy indication 14 Two quadrant reactive energy indication 15 Three quadrant reactive energy indication 16 Four quadrant reactive energy indication 17 Combined active energy indication 18 Calendar clock

[0108] The running status of the electric energy meter is monitored through the analysis and judgment of the operation parameters. The running status includes: electric energy meter parameter change, electric energy meter time difference, electric meter fault information, electric energy meter scale drop, electric energy difference, electric energy meter flying, electric energy meter stop, phase sequence abnormality, electric energy meter cover record, electric energy meter running state word displacement, etc.

[0109] An event is a record of a state change triggered during the operation of an energy meter and is a passively generated log. The specific contents are shown in Table 2:

[0110] Table 2

[0111] Serial number Event name 1 Energy meter loss of voltage 2 Energy meter undervoltage 3 Energy meter overvoltage 4 Energy meter open phase 5 Energy meter loss of current 6 Energy meter overcurrent 7 Energy meter reverse power 8 Energy meter total loss of voltage 9 Energy meter reverse phase sequence of voltage 10 Energy meter loss of power 11 Energy meter programming 12 Energy meter zero reset 13 Energy meter demand zero reset 14 Energy meter event zero reset 15 Energy meter time setting 16 Energy meter cover open 17 Energy meter end button box open 18 Energy meter voltage unbalance 19 Energy meter tripping 20 Energy meter closing 21 Energy meter constant magnetic field interference 22 Energy meter load switch malfunction 23 Energy meter power supply abnormality

[0112] The concentrator collects the historical trigger events, operating parameters (voltage, current and other quantitative indicators) and passively generated event logs of the electricity meter in real time to build a complete data foundation.

[0113] As the core device, the concentrator uniformly processes, stores and transmits data, ensuring the real-time and data integrity of subsequent analysis and providing reliable input for event analysis.

[0114] S200: Analyze the logical relationship between events based on historical records to establish a causal relationship chain. Mark the currently triggered event, analyze the triggering probability of the event based on the causal relationship chain, and calculate the weight index of each marked event. Specifically including:

[0115] S201: Obtain historical records of all electric energy meters, analyze historical trigger events contained in the historical records, analyze the trigger time corresponding to each event, sort all events in each historical record in chronological order of trigger time, and establish an event chain.

[0116] Each electricity meter corresponds to a historical record, and each historical record corresponds to an event chain.

[0117] S202: Calculate the triggering time interval between two adjacent events in each event chain, split the adjacent events in each event chain whose triggering time interval is greater than a threshold, and ensure that the triggering time interval of adjacent events in the split event chain is not greater than the threshold.

[0118] The event chain is segmented based on the time interval threshold, and irrelevant events are eliminated to ensure that the events in the chain have strong temporal correlation and reduce noise interference.

[0119] S203: Establish a relationship database, select event chains with more than 1 event and put them into the relationship database. Collect two adjacent events in the relationship database as causal pairs. The events in different causal pairs are not exactly the same or have different time sequences.

[0120] Each causal pair contains two adjacent events with a time sequence. The events in different causal pairs are not exactly the same, or the time sequence is different.

[0121] S204: Analyze the frequency and interval length of each causal pair in the relationship database and calculate the synchronization index. Filter out causal pairs with synchronization indexes greater than a threshold, thereby establishing a causal relationship chain. This includes:

[0122] S2041. Count the number of event chains belonging to each causal pair in the relationship database as the number of occurrences of the corresponding causal pair. Divide the number of occurrences of each causal pair by the total number of occurrences of all causal pairs to obtain the frequency of occurrence of the corresponding causal pair.

[0123] S2042. Obtaining causal pairs Number of occurrences and frequency of occurrence , and contains events and Causal pairs with a prior order Number of occurrences , substitute into the formula to calculate the causal pair The coincidence index :

[0124] ;

[0125] Where, is a constant greater than 1, is the average frequency of all causal pairs, and is a constant, For the causal pair in the relation database All events in the event chain and The standard deviation of the trigger interval.

[0126] Used to penalize causal pairs with large time interval fluctuations ( Adjusting fluctuation sensitivity), The value determines the weight of the time interval standard deviation in the synchronization index.

[0127] Used to amplify the value of low-frequency causal pairs ( Suppressing the advantage of high-frequency events).

[0128] The value is greater than 0 to prevent division by zero errors caused by a standard deviation of zero in theoretical calculations and to provide dimensional consistency.

[0129] The synchronization index is used to quantify the strength of the association between causal pairs, screen statistically significant and temporally stable causal pairs, and exclude accidental associations and fluctuation noise.

[0130] S2043: Similarly, calculate the synchronization index of each causal pair, select the causal pairs whose synchronization index is greater than the threshold, and analyze whether these causal pairs have the same event.

[0131] S2044. Connect two causal pairs that have the same event but are located in different positions end to end to form a causal chain.

[0132] S2045. Continue to merge causal chains with the same events until there are no identical events between the causal chains.

[0133] Highly correlated causal pairs are screened using the synchronization index formula (the formula includes frequency and standard deviation weights) to ensure the statistical significance of the causal chain.

[0134] S205: Mark the events triggered by each electric energy meter within the last time period k, analyze the triggering probability of each event, and calculate the weight index of each marked event based on the synchronization index in the causal relationship chain. Specifically, it includes:

[0135] S2051, analyze the causal relationship chain to which each marked event belongs, and mark the event in the causal relationship chain All subsequent events as The affected objects.

[0136] S2052. Extract the influencing objects in the causal relationship chain With Marker Events All causal pairs between, calculate each causal pair The trigger probability The triggering probabilities of all causal pairs are multiplied to obtain the corresponding affected objects. The trigger probability.

[0137] When there is no event between the influencing object and the marked event, the trigger probability is the trigger probability of the corresponding causal pair of the influencing object and the marked event.

[0138] When there are multiple events and multiple paths between the influencing object and the marked event, the triggering probabilities of all causal pairs in each path are multiplied, and the triggering probability of the path with the largest result is selected as the triggering probability of the influencing object.

[0139] S2053. Filter out all included events in the causal chain and Causal pair with later order , calculate the average value of the synchronization index of these causal pairs as the influencing object Impact Index .

[0140] S2054, respectively calculate the marking events The trigger probability and impact index of each affected object, as well as the number of occurrences of each event in all historical records, and the average number of occurrences of all events is used as the reference number .

[0141] S2055. Get marker event Number of occurrences , substitute the formula to calculate the weight index of the marked event :

[0142] ;

[0143] In the formula, and are constants, is the number of affected objects of the marked event , and and are the triggering probability and impact index of the nth affected object, respectively.

[0144] The triggering probability and the impact index of the marked event on all affected objects are added up to reflect the event's diffusion ability.

[0145] The control event frequency affects the strength of the weight The smaller the frequency weight is, the higher the frequency weight is. The frequency of the reaction marked event is lower, the rarer it is, and the higher the weight is.

[0146] By calculating the weight index, high-impact, low-frequency events are given higher weights to improve the accuracy of priority judgment.

[0147] Based on the causal relationship chain, the event triggering probability (maximum probability of multiple paths) and the impact index (average of synchronous index of cause and effect) are calculated, and finally the importance of the event is quantified by the weight index formula.

[0148] S300, analyze the running parameters to evaluate the event state, and filter the marked events by state. Calculate the priority index of each electric energy meter, and generate an event record table by combining the weight index of the marked event and the priority index of the electric energy meter.

[0149] Please refer to Figure 2 , which specifically includes:

[0150] S301, obtain the current running parameters of each marked event, set the preset value and compare to judge the event state one by one, including event occurrence and event recovery.

[0151] Event occurrence refers to determining the abnormal state of the marked event record under the current circumstances after comparing and analyzing the preset value with the real-time running parameters.

[0152] ​​​Event recovery refers to determining that the marked event record is not true in the normal state according to the preset value and the real-time operation parameter after rule comparison and analysis.

[0153] Compare the operation parameter with the preset value, distinguish between event occurrence (abnormal) and event recovery (normal), and ensure that only effective abnormal events are processed.

[0154] S302, screen out the marked events in the event occurrence state, classify these marked events according to the electric energy meter they belong to, and calculate the priority index according to the trigger time and weight index of the marked events under each electric energy meter .

[0155] Set different reference time lengths and time decay coefficients for each marked event, respectively calculate the time length of the trigger time of each marked event from the current time, and substitute it into the formula to calculate the priority index of the corresponding electric energy meter:

[0156] ;

[0157] In the formula, is the number of unreported marked events corresponding to the electric energy meter, is the weight index of the i-th marked event, and are the time decay coefficient and reference time length of the i-th marked event, respectively. is a constant,

[0158] is the time length of the trigger time of the i-th marked event from the current time. Through the priority index, the comprehensive priority of all unprocessed events of a single electric energy meter is dynamically calculated. Superimpose high-weight events (

[0159] ) and dynamically decay ( and

[0160] ) according to the fault type, to ensure that old faults are down-weighted and new urgent faults are prioritized. S303, respectively for each electric energy meter, establish a group, and sequentially put the marked events into the corresponding group according to the order of the weight index from large to small.

[0161] S304, all groups are sorted according to the priority index of the corresponding electric energy meter in the order from large to small, and all marked events in the first v groups are collectively generated into an event record table.

[0162]

[0163] ​​​S400, save the event record table in the data center of the concentrator, and report one by one according to the arrangement order in the event record table. Calculate the transmission kinetic energy index of the corresponding marked event each time the reporting fails, so as to judge whether to continue reporting. Specifically, it includes:

[0164] S401, save the event record table in the data center of the concentrator, and generate reporting tasks one by one according to the arrangement order of each marked event in the event record table.

[0165] S402, analyze the reporting task, and combine and report the content to be reported according to the reporting format specified in the 698 protocol.

[0166] The reported data can be transmitted in compressed and uncompressed modes. For data transmitted in compressed mode, lossless compression algorithm must be used to ensure that the decompressed data is exactly the same as the original data.

[0167] S403, after the communication master station receives the reported content and returns the reporting confirmation frame, automatically delete the reporting task and the corresponding marked event in the event record table.

[0168] Please refer to Figure 3 , the message issued by the communication master station is parsed according to the content of the 698 protocol, and the corresponding work is performed according to the specific content parsed.

[0169] After completing the corresponding work, the work result content corresponding to the issued message needs to be framed according to the content of the 698 protocol, and the framed data frame is sent to the communication master station.

[0170] S404, if the concentrator does not receive the reporting confirmation frame for more than a preset time, calculate the transmission kinetic energy index of the corresponding marked event of the reporting task, and automatically switch to the next reporting task for frame combination and reporting.

[0171] Set the standard time , the transmission kinetic energy index The calculation formula is as follows:

[0172] ;

[0173] In the formula, is a constant greater than 1, is the time length from the last reporting failure time of the marked event corresponding to the reporting task to the current time.

[0174] is the priority index of the marked event corresponding to the electric energy meter, is the average value of the priority indexes of all electric energy meters in the event record table.

[0175] is the weight index of the marking event, It is the sum of the weight indices of all remaining unreported marked events in the event record table of the electric energy meter corresponding to the marked event.

[0176] Used to measure device priority compared to the event log average. Used to measure the event weight ratio (the sum of the weights of events not reported by the current device).

[0177] pass and Adding and subtracting weights adjusts the importance of marked events, dynamically abandons low-value tasks, and reduces network congestion.

[0178] S405 , each time a report fails, a transmission kinetic energy index of a marking event corresponding to the reporting task is calculated. When the transmission kinetic energy index is less than a threshold, the reporting task and the corresponding marking event in the event record table are automatically deleted.

[0179] See also Figure 4 The present invention also provides a concentrator operation and management system based on data processing, including a data perception module, an event analysis module, an operation management module and a reporting and transmission module.

[0180] The data perception module collects the historical records, operating parameters and triggered events of the electricity meter through the concentrator.

[0181] The concentrator collects the meter's historical records (including historical trigger events), operating parameters (such as voltage, current and other quantitative indicators), and triggered events (passively generated state change logs) in real time. The concentrator is responsible for centralized data processing, storage, and bidirectional transmission.

[0182] Providing a comprehensive and real-time data foundation ensures that the system can accurately capture the operating status and event information of the electricity meter, laying a reliable data support for subsequent analysis, and improving the real-time and integrity of monitoring.

[0183] The event analysis module establishes a causal chain based on historical records, marks the triggered events and calculates the weight index based on the causal chain.

[0184] Analyze the logical relationship between events based on historical records, including segmenting event chains (according to the trigger time interval threshold), establishing causal chains (by calculating the synchronization index of causal pairs and screening highly correlated event chains), marking the current event, analyzing the trigger probability and impact index in combination with the causal chain, and finally calculating the weight index (such as the formula for calculating BAW).

[0185] By quantifying the causal relationships and probabilities between events, high-impact events can be identified, event priority assessment can be optimized, the accuracy and efficiency of event analysis can be improved, and the risk of misjudgment can be reduced.

[0186] The operation management module is used for evaluating event status and screening, calculating the priority index of each electric energy meter, and generating an event record table according to the weight index and the priority index size.

[0187] The analysis operation parameter evaluates the event status (such as event occurrence or recovery), screens out the marker event in the event occurrence state, calculates the priority index of the electric energy meter (based on the weight index, trigger time decay formula, etc.), and sorts the events according to the weight index to generate an event record table (selecting a high priority group).

[0188] Intelligent screening and sorting of events ensure that high-weight and high-priority events are processed first, optimize resource allocation and event response, and improve system processing efficiency and the orderliness of event management.

[0189] The reporting transmission module reports each marker event in the event record table one by one, and calculates the transmission kinetic index of the corresponding marker event if the reporting fails, so as to determine whether to continue reporting.

[0190] The event record table is stored in the data center and is reported one by one in order (reported in combination with the 698 protocol frame); when the reporting fails, the transmission kinetic index is calculated (based on the failure time, priority index, and weight index formula), and it is determined whether to continue reporting or delete the task.

[0191] Dynamic processing of reporting failure avoids invalid retries, reduces network congestion, ensures reliable transmission of critical events, and improves system communication efficiency and overall reliability.

[0192] Example 1: Assuming that the time length of the last reporting failure of marker event A1 from the current time is 2 min, the standard time length is 5 min, the priority index of the corresponding electric energy meter is 2.5, and the average value of the priority index of all electric energy meters in the event record table is 3.

[0193] The weight index of marker event A1 is 2, the sum of the weight indexes of all marker events remaining unreported in the event record table corresponding to the electric energy meter is 8, and the constant is 1.5, the transmission kinetic index is calculated by substituting the formula:

[0194] ;

[0195] The transmission kinetic index of marker event A1 is 0.70.

[0196] It is to be noted that, in the present text, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0197] Finally, it should be noted that the above-mentioned only constitutes the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art will appreciate that modifications can be made to the technical solutions described in the foregoing embodiments, or equivalent replacements can be made to some of the technical features. Any modifications, equivalent replacements, improvements, and the like made within the spirit and principle of the present application shall fall within the scope of the protection of the present application.

Claims

1. A concentrator operation management method based on data processing, characterized in that: The method includes: S100, collecting the historical records of the electric energy meter, as well as the real-time operating parameters and triggered events through the concentrator; S200: Analyze the logical relationship between events based on historical records to establish a causal relationship chain; mark the currently triggered event, analyze the triggering probability of the event based on the causal relationship chain, and calculate the weight index of each marked event; S300: Analyze the operating parameters and evaluate the event status, and filter the marked events according to the status; calculate the priority index of each electric energy meter, and generate an event record table by combining the weight index of the marked event and the priority index of the electric energy meter; specifically, including: S301, obtain the current operating parameters of each marked event, set the preset values ​​and make event status judgments one by one after comparison and judgment, the status includes event occurrence and event recovery; S302: Filter out the marked events of the event occurrence status, classify these marked events according to the electric energy meters they belong to, and calculate the priority index according to the triggering time and weight index of the marked event under each electric energy meter. ; Set different reference durations and time decay coefficients for each marker event, calculate the time between the trigger time of each marker event and the current time, and substitute them into the formula to calculate the priority index of the corresponding energy meter: ； Where, is the number of unreported marking events corresponding to the electric energy meter, For the The weight index of the marked event, and Respectively The time decay coefficient and reference duration of each marker event; is a constant, For the The length of time between the trigger time of a marked event and the current time; S303: Create a group for each electric energy meter and place the marked events into the corresponding group in descending order of weight index; S304: sort all groups in descending order according to the priority index of the corresponding electric energy meter, and select all marked events in the first v groups to jointly generate an event record table; S400: Save the event record table in the data center of the concentrator, and report one by one according to the arrangement order in the event record table; calculate the transmission kinetic energy index of the corresponding marked event each time a report fails, so as to determine whether to continue reporting.

2. The concentrator operation management method based on data processing according to claim 1, characterized in that: In S100, the concentrator refers to a device that centrally processes, stores, manages and bidirectionally transmits meter reading data of electric energy meters; Historical records include historical trigger events; operating parameters include various quantitative indicators used to describe the performance and status of the electricity meter during operation; An event is a record of state changes triggered during the operation of an electricity meter and is a passively generated log.

3. The concentrator operation management method based on data processing according to claim 2, characterized in that: S200 includes: S201. Obtain historical records of all electric energy meters and parse the historical trigger events contained in the historical records; analyze the trigger time corresponding to each event, sort all events in each historical record in chronological order of trigger time, and establish an event chain; S202: Calculate the triggering time interval between two adjacent events in each event chain, split the adjacent events in each event chain whose triggering time interval is greater than a threshold, and ensure that the triggering time interval of adjacent events in the split event chain is not greater than the threshold; S203: Establish a relationship database, select event chains with more than 1 event and put them into the relationship database; collect two adjacent events in the relationship database as causal pairs, and the events in different causal pairs are not completely the same or have different time sequences; S204: Analyze the frequency of occurrence and interval duration of each causal pair in the relationship database and calculate the synchronization index; select causal pairs with synchronization indexes greater than a threshold, thereby establishing a causal relationship chain; S205 , marking the events triggered by each electric energy meter within the most recent time period k, analyzing the triggering probability of each event, and calculating the weight index of each marked event according to the synchronization index in the causal relationship chain.

4. The concentrator operation management method based on data processing according to claim 3, characterized in that: 204 includes: S2041. Count the number of event chains to which each causal pair belongs in the relationship database as the number of occurrences of the corresponding causal pair; divide the number of occurrences of each causal pair by the total number of occurrences of all causal pairs to obtain the frequency of occurrence of the corresponding causal pair; S2042. Obtaining causal pairs Number of occurrences and frequency of occurrence , and contains events and Causal pairs with a prior order Number of occurrences , substitute into the formula to calculate the causal pair The coincidence index : ； Where, is a constant greater than 1, is the average frequency of all causal pairs, and is a constant, For the causal pair in the relation database All events in the event chain and Standard deviation of trigger intervals; S2043, and so on, respectively calculating the synchronization index of each causal pair; screening out causal pairs with synchronization indexes greater than a threshold, and analyzing whether these causal pairs have the same event; S2044. Connect two causal pairs that have the same event but are located in different positions end to end to form a causal relationship chain; S2045. Continue to merge causal chains with the same events until there are no identical events between the causal chains.

5. The concentrator operation management method based on data processing according to claim 4 is characterized in that: 205 includes: S2051, analyze the causal relationship chain to which each marked event belongs, and mark the event in the causal relationship chain All subsequent events as the objects of influence; S2052. Extract the influencing objects in the causal relationship chain With Marker Events All causal pairs between, calculate each causal pair The trigger probability ; Multiply the trigger probabilities of all causal pairs to get the corresponding affected objects The trigger probability of S2053. Filter out all included events in the causal chain and Causal pair with later order , calculate the average value of the synchronization index of these causal pairs as the influencing object Impact Index ; S2054, respectively calculate the marking events The trigger probability and impact index of each affected object, as well as the number of occurrences of each event in all historical records, and the average number of occurrences of all events is used as the reference number ; S2055. Get marker event Number of occurrences , substitute the formula to calculate the marking event Weight index : ； Where, and is a constant, To mark an event The number of affected objects, and Respectively The trigger probability and impact index of each affected object.

6. The concentrator operation management method based on data processing according to claim 1, characterized in that: S400 includes: S401, save the event record table in the data center of the concentrator, and generate reporting tasks one by one according to the order of each marked event in the event record table; S402, analyzing the reporting task, combining and reporting the content to be reported in the reporting format specified in the 698 protocol; S403, when the communication master station receives the report content and returns the report confirmation frame, it automatically deletes the report task and the corresponding marked event in the event record table; S404: If the concentrator does not receive the report confirmation frame within a preset time period, it calculates the transmission kinetic energy index of the marking event corresponding to the reporting task and automatically switches to the next reporting task to combine and report the reporting frame; S405 , each time a report fails, a transmission kinetic energy index of a marking event corresponding to the reporting task is calculated. When the transmission kinetic energy index is less than a threshold, the reporting task and the corresponding marking event in the event record table are automatically deleted.

7. The concentrator operation management method based on data processing according to claim 6, characterized in that: In S404, set the standard duration , transmission kinetic energy index The calculation formula is as follows: ； Where, is a constant greater than 1, The time interval between the last failed report time of the reporting task corresponding to the marked event and the current time; is the priority index of the energy meter corresponding to the marking event, The average value of the priority index of all electric energy meters in the event record table; is the weight index of the marking event, It is the sum of the weight indices of all remaining unreported marked events in the event record table of the electric energy meter corresponding to the marked event.

8. A concentrator operation management system based on data processing, applied to the concentrator operation management method based on data processing as claimed in claim 1, characterized in that: The system includes a data perception module, an event analysis module, an operation management module and a reporting and transmission module; The data perception module collects the historical records, operating parameters and triggered events of the electricity meter through the concentrator; The event analysis module establishes a causal chain based on historical records, marks the triggered events, and calculates the weight index based on the causal chain; The operation management module is used to evaluate the event status and screen it, calculate the priority index of each electricity meter, and generate an event record table according to the weight index and priority index size; The reporting and transmission module reports the marked events in the event record table one by one. If the reporting fails, the transmission kinetic energy index of the corresponding marked event is calculated to determine whether to continue reporting.

Images