Hydroelectric power generation equipment reliability event identification method based on production monitoring data

Abstract

The invention discloses a hydroelectric power generation equipment reliability event identification method based on production monitoring data, and the method comprises the steps: firstly constructing a data set for reliability event identification, then carrying out the data preprocessing and data marking of the data in the data set, and obtaining the marked data; generating a start-stop record of the water-turbine generator set according to the marked data, and judging the shutdown type based on the start-stop record of the water-turbine generator set; and finally, generating a reliability event list of the water-turbine generator set based on the event switching mode according to the start-stop record and the maintenance record of the water-turbine generator set. According to the method, electric power system operation data and electric power equipment monitoring data are applied to carry out analysis and evaluation, on the premise that information safety is guaranteed, automatic collection, intelligent analysis, traceable management and transparent supervision of reliability data are achieved, real-time monitoring of the reliability of power generation equipment is achieved, and the accuracy, timeliness and integrity of the reliability data are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of power generation system reliability, and in particular to a method for identifying reliability events of hydropower generation equipment based on production monitoring data. Background Art

[0002] The reliability of power generation equipment refers to the ability of the equipment to complete the specified functions under specified conditions and within a specified time period. The power generation reliability management information system of the existing technology requires each station to enter the reliability data of the previous month into the system at the beginning of each month. When entering, it mainly refers to the relevant data in the start-up and shutdown record ledger and maintenance application form manually recorded by the power production management system. Although the reliability management system based on manual reporting can effectively avoid the misjudgment of key events, it has the problem of low reporting timeliness, the data cannot be collected automatically, and there is a lag in timeliness. The judgment of special events mainly depends on the management experience of on-site personnel. The accuracy of the judgment of previous events is highly related to the experience of personnel, and the judgment basis comes from manual interpretation. There are no established reference rules and monitoring alarms, and transparent supervision and control cannot be achieved. The logic of manual judgment varies from person to person, and has not been systematically standardized. The results cannot be traced, and transparent supervision cannot be achieved. The operation of hydropower generator sets with abnormal characteristics is easily missed. Therefore, it is necessary to design a reliability event identification method for hydropower equipment based on production monitoring data to solve the above problems. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method for identifying reliability events of hydropower equipment based on production monitoring data, so as to solve the problem that the reliability management of hydropower equipment in the prior art is highly dependent on management information systems and manual experience, resulting in low timeliness and opaque supervision. The method has the characteristic of enabling power generation enterprises to fully and timely grasp the reliability status of hydropower equipment through digital monitoring means.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is: A method for identifying reliability events of hydropower equipment based on production monitoring data, comprising: S1, construct a data set for reliability event identification, which includes equipment monitoring data, maintenance records of hydro-generator units, and reliability events reported in the reliability system; S2, performing data preprocessing and data labeling on the data in the data set to obtain labeled data; S3, generating a start and stop record of a hydro-generator set according to the marked data; S4, determining the shutdown category based on the start and stop records of the hydro-generator set; S5, generating a reliability event list of the hydro-generator set based on the event switching mode according to the start-stop record of the hydro-generator set and the maintenance record of the hydro-generator set in S1-S4; S6, compare the generated list of reliability events of the hydro-turbine generator set with the reported records in the reliability system for supplementary verification.

[0005] Preferably, in step S1, the equipment monitoring data includes power data of the hydro-generator set and equipment status data; Power data includes active power and reactive power. A power measurement point is set for each hydro-turbine generator set. The power data has a frequency of seconds during the operation of the hydro-turbine generator set. During shutdown, each data is recorded every 20 minutes.

[0006] The equipment status refers to the status of the hydro-turbine generator set, including eight equipment statuses: shutdown status, idling status, no-load status, grid-connected status, transition status, water system operation status, maintenance status, and unknown status. One equipment status measurement point is set for each status, and eight equipment status measurement points are set for each hydro-turbine generator set. When the unit is in this state, the corresponding measurement point value is 1, otherwise it is 0. The data frequency is not synchronized with the power measurement points, and is about 10-20 minutes per item.

[0007] Preferably, in step S1, the maintenance record of the hydro-turbine generator set includes a maintenance application form, which includes the equipment name, maintenance level, work content, adjustment order time and completion report time; the maintenance application form belongs to power production management data, and the declared maintenance plan takes effect after approval by the dispatcher.

[0008] Preferably, the reliability events reported in the reliability system include: Reliability event types include operation events and shutdown events; Operation events consist of full-output operation and reduced-output operation; shutdown events consist of standby, planned shutdown, unplanned shutdown and deactivation events; the basis for reporting reliability events is the start-up and shutdown records and maintenance application forms in the power production management system.

[0009] Furthermore, reliability events are reported regularly every month by a dedicated person; operation events consist of full-output operation and reduced-output operation. Full-output operation events are generally not reported, and are only reported manually when no other events occur in the unit that month. The required information for event reporting includes the unit number, event start and end time, and event status; reduced-output operation events cannot be technically determined at present, and this status event has not yet been reported in the reliability system.

[0010] Furthermore, shutdown events are composed of standby, planned shutdown, unplanned shutdown and deactivation events. The reference basis is the start and shutdown records and maintenance application forms in the power production management system. Unplanned shutdown events have no system data support and need to be reported in the reliability system after manual judgment.

[0011] Preferably, data preprocessing and data labeling are performed on the data in the data set, and the labeled data obtained includes: The second-level data of active power in the data set is processed into minute-level data, and the minute node of the startup or shutdown operation is confirmed by marking; Set the minimum threshold for the operation of the hydro-generator set Based on minute-level data, the active power of the hydro-generator unit in operation is greater than the threshold. The data is recorded as the operating state and marked as S. The active power of the hydro-generator set in the operating state is less than the threshold The data record is in outage state, marked as R.

[0012] Furthermore, the selection of the minimum operating threshold can be set according to the unit capacity and experience, or by using a grid search to compare the output result corresponding to the threshold with the reported result.

[0013] Preferably, in step S3, generating a start and stop record of a hydro-generator set according to the marked data includes: According to the data mark in step S2, the start time and the end time of shutdown within the generation cycle are generated, and the generation rule is: The first recorded time node of N consecutive data marked as R is the shutdown start time of the hydro-generator set; The first recorded time node of N consecutive data marked as S is the start time of the hydro-generator set; Perform data correction on the marked data.

[0014] Preferably, in step S4, determining the shutdown category based on the start and stop records of the hydro-generator set includes: On the basis of the shutdown termination time identified in step S3, it is determined whether the shutdown process is a normal shutdown or an abnormal shutdown based on the characteristics of the original data of the active power of the hydro-generator set, wherein the original data is the data collected in step S1 with a frequency of seconds; the specific determination method is: S301, setting the benchmark change range and interval duration threshold ; S302, taking the shutdown end time as the reference time point, t minutes ahead of the reference time point to form a time period T, and cyclically calculating the time nodes within the time period T Active power data and the previous time node Active power data The range of change And the interval duration , for the change range and interval duration Execution judgment: If the change Exceeding the baseline change , and the interval time Not less than the interval duration threshold , then the shutdown is judged to be an abnormal shutdown due to data loss. The calculation formula is as follows: ; If the change Exceeding the baseline change , and the interval time Less than the interval duration threshold , then the shutdown is judged as a general abnormal shutdown, and the calculation formula is as follows; ; If the change No more than the benchmark change , then the shutdown is judged to be a normal shutdown, and the calculation formula is as follows: ; S303, set the cycle stop condition: When the following conditions are met for the first time, the loop is stopped and the time node of the data is reassigned to the shutdown start time of the hydro-generator set; ; in The current time node The active power, The minimum operating threshold of the hydro-generator set set in step S2; is the range of change of the set benchmark; The loop stops after all the data in period T has been circulated; If no data is retrieved within the T period, it will be directly judged as an abnormal shutdown due to data loss.

[0015] Furthermore, when marking data and determining the downtime category, the data used is the original data at the second level, but not all the second-level data in the T period is used. Instead, part of the second-level data in a fixed interval before and after each data node of the minute-level data is used. This ensures the use of second-level data to achieve higher accuracy and avoids the problem of slow processing speed caused by processing a large amount of second-level data.

[0016] Preferably, the reference variation range is the difference between the active power data of two adjacent time nodes during a normal shutdown process; the data of the normal shutdown process is obtained through the planned shutdown history data in the shutdown events in the data set.

[0017] Preferably, in step S5, according to the start-stop record of the hydro-generator set and the maintenance record of the hydro-generator set in S1 to S4, the method for generating a reliability event list of the hydro-generator set based on the event switching mode includes: S501, based on historical data and unit operating conditions, summarizes three common event switching modes, including: First mode: operation, standby, operation mode; Second mode: operation, standby, planned outage, standby, operation mode; The third mode: operation, unplanned outage, planned outage, standby, and operation mode; S502, generating a reliability event list of the hydro-generator set according to the start-up and shutdown records of the hydro-generator set and the maintenance records of the hydro-generator set in S1-S4 by following the steps below: a. All time periods from “start-up to shutdown” are considered as operation events; b. For the "shutdown to restart" time period and the shutdown type is normal shutdown, if there is no maintenance record in this time period, it is judged as a backup event; if there is a maintenance record in this time period, the reliability events in this time period are backup, planned shutdown and backup in order; c. For the "shutdown to restart" time period and the shutdown type is a general abnormal shutdown, search for maintenance records within this time period. The reliability events within this time period are unplanned shutdown, planned shutdown, and standby in order; if no maintenance records are found, it means that the maintenance records are missing, and the maintenance records are added to the data set; d. For the "shutdown to restart" time period and the shutdown type is abnormal shutdown due to data loss, if there is a maintenance record in this time period, the reliability events in this time period are unplanned shutdown due to lack of data, planned shutdown and standby in order; if there is no maintenance record in this time period, the reliability event in this time period is unplanned shutdown due to lack of data; Among them, the "start-up to shutdown" time period refers to the time period between the start time of startup and the start time of shutdown; the "shutdown to start-up" time period refers to the time period between the start time of shutdown and the start time of startup.

[0018] Furthermore, the first mode corresponds to the vast majority of normal standby scenarios. The unit enters the standby state after a normal shutdown and continues to be put into operation after the standby state ends. The second mode corresponds to scenarios with planned maintenance. The unit first enters the standby state after a normal shutdown, which can be determined by analyzing the shutdown time and the start time of the maintenance plan, and then enters the maintenance state as planned, returns to the standby state after the maintenance is completed, and finally is put into operation. The third mode corresponds to abnormal shutdown scenarios. If there is an unplanned shutdown, planned maintenance is usually arranged after the event to find out the cause of the unplanned shutdown, and then the unit returns to standby and is put into operation.

[0019] Preferably, the marked data is corrected, including reviewing the data marked with R based on the real-time data of the grid-connected state, and the review method is: Taking the time node corresponding to the data marked with R as the starting point, find the first grid-connected status record in the past or future M time. If the grid-connected status is grid-connected, the digital expression is 1, then modify the mark R to S, otherwise it remains unchanged; When executing the above review method, the data marked as R needs to meet the following conditions: ; in, For the current time point The active power, The minimum operating threshold of the unit determined in step S2; for The grid-connected status of the unit at the moment; is the minimum time that satisfies the following conditions: ; If no data is retrieved within M time, the original identification is maintained.

[0020] Furthermore, the identification method of the present invention is regarded as supervised learning, and the reliability events historically reported in the reliability system in the data set are used as reference for the results to set and adjust the relevant thresholds in the identification method of the present invention; conversely, the reliability event list generated by the identification method of the present invention is also used to check and fill in the gaps in the reported results in the reliability system.

[0021] The beneficial effects of the method for identifying reliability events of hydropower equipment based on production monitoring data provided by the present invention are as follows: 1. The present invention uses power system operation data and power equipment monitoring data to carry out analysis and evaluation, promotes the construction of a power reliability management system based on real-time data, breaks the traditional post-management mode of reliability events, conducts timely overall monitoring and management of production conditions, and realizes automated collection, intelligent analysis, traceable management, and transparent supervision of reliability data, laying a foundation for reliability data to assist in supporting operation and maintenance, equipment selection, construction and renovation, and other work.

[0022] 2. The recognition method proposed in the present invention efficiently uses real-time data of equipment detection, and effectively improves the data processing speed without losing accuracy by processing minute-level data and using original second-level data.

[0023] 3. The identification method proposed in the present invention proposes an application method for different frequency time series data, and effectively solves the problem caused by the instability of a single data source by integrating and using multiple measurement points.

[0024] 4. The identification results of the identification method proposed in the present invention can support the management needs of different levels. For example, in the process of historical data identification, several data that meet the characteristics of abnormal shutdown are found. For power generation companies, such data is the focus of management attention. In the previous manual data reporting process, such situations will not be reported and are easily overlooked. On the other hand, the identification results provide a sufficient basis for manual verification work, avoiding the characteristics of untimely reporting and easy omissions in traditional reporting methods. The event results that have been manually verified can better meet the requirements of China Electricity Council for reliability management. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is a schematic diagram of the process framework of the present invention; Figure 2 It is a schematic diagram of three switching modes of reliability events in the present invention. DETAILED DESCRIPTION

[0026] Embodiment 1: like Figure 1 As shown, a method for identifying reliability events of hydropower equipment based on production monitoring data includes: S1, construct a data set for reliability event identification, which includes equipment monitoring data, maintenance records of hydro-generator units, and reliability events reported in the reliability system; S2, performing data preprocessing and data labeling on the data in the data set to obtain labeled data; S3, generating a start and stop record of the hydro-generator set according to the marked data; S4, determining the shutdown category based on the start and stop records of the hydro-generator set; S5, generating a reliability event list of the hydro-generator set based on the event switching mode according to the start-stop record of the hydro-generator set and the maintenance record of the hydro-generator set in S1-S4; S6, compare the generated list of reliability events of the hydro-turbine generator set with the reported records in the reliability system and improve the rules.

[0027] Preferably, in step S1, the equipment monitoring data includes power data of the hydro-generator set and equipment status data; Power data includes active power and reactive power. A power measurement point is set for each hydro-turbine generator set. The power data has a frequency of seconds during the operation of the hydro-turbine generator set. During shutdown, each data is recorded every 20 minutes.

[0028] The equipment status refers to the status of the hydro-turbine generator set, including eight equipment statuses: shutdown status, idling status, no-load status, grid-connected status, transition status, water system operation status, maintenance status, and unknown status. One equipment status measurement point is set for each status, and eight equipment status measurement points are set for each hydro-turbine generator set. When the unit is in this state, the corresponding measurement point value is 1, otherwise it is 0. The data frequency is not synchronized with the power measurement points, and is about 10-20 minutes per item.

[0029] Preferably, in step S1, the maintenance record of the hydro-turbine generator set includes a maintenance application form, which includes the equipment name, maintenance level, work content, adjustment order time and completion report time; the maintenance application form belongs to power production management data, and the declared maintenance plan takes effect after approval by the dispatcher.

[0030] Preferably, the reliability events reported in the reliability system include: Reliability event types include operation events and shutdown events; Operation events consist of full-output operation and reduced-output operation; shutdown events consist of standby, planned shutdown, unplanned shutdown and deactivation events; the basis for reporting reliability events is the start-up and shutdown records and maintenance application forms in the power production management system.

[0031] Furthermore, reliability events are reported regularly every month by a dedicated person; operation events consist of full-output operation and reduced-output operation. Full-output operation events are generally not reported, and are only reported manually when no other events occur in the unit that month. The required information for event reporting includes the unit number, event start and end time, and event status; reduced-output operation events cannot be technically determined at present, and this status event has not yet been reported in the reliability system.

[0032] Furthermore, shutdown events are composed of standby, planned shutdown, unplanned shutdown and deactivation events. The reference basis is the start and shutdown records and maintenance application forms in the power production management system. Unplanned shutdown events have no system data support and need to be reported in the reliability system after manual judgment.

[0033] Preferably, data preprocessing and data labeling are performed on the data in the data set, and the labeled data obtained includes: Process the data in the dataset into minute-level data, and use marking methods to confirm the minute node where the startup or shutdown operation is located; Set the minimum threshold for the operation of the hydro-generator set Based on minute-level data, the active power of the hydro-generator unit in operation is greater than the threshold. The data is recorded as the operating state and marked as S. The active power of the hydro-generator set in the operating state is less than the threshold The data record is in outage state, marked as R.

[0034] Furthermore, the selection of the minimum operating threshold can be set according to the unit capacity and experience, or by using a grid search to compare the output result corresponding to the threshold with the reported result.

[0035] Preferably, in step S3, generating a start and stop record of a hydro-generator set according to the marked data includes: According to the data mark in step S2, the start time and the end time of shutdown within the generation cycle are generated, and the generation rule is: The first recorded time node of N consecutive data marked as R is the shutdown start time of the hydro-generator set; The first recorded time node of N consecutive data marked as S is the start time of the hydro-generator set; Perform data correction on the marked data.

[0036] Preferably, in step S4, determining the shutdown category based on the start and stop records of the hydro-generator set includes: On the basis of the shutdown termination time identified in step S3, it is determined whether the shutdown process is a normal shutdown or an abnormal shutdown based on the characteristics of the original data of the active power of the hydro-generator set, wherein the original data is the data collected in step S1 with a frequency of seconds; the specific determination method is: S301, setting the benchmark change range and interval duration threshold ; S302, taking the shutdown end time as the reference time point, t minutes ahead of the reference time point to form a time period T, and cyclically calculating the time nodes within the time period T Active power data and the previous time node Active power data The range of change And the interval duration , for the change range and interval duration Execution judgment: If the change Exceeding the baseline change , and the interval time Not less than the interval duration threshold , then the shutdown is judged to be an abnormal shutdown due to data loss. The calculation formula is as follows: ; If the change Exceeding the baseline change , and the interval time Less than the interval duration threshold , then the shutdown is judged as a general abnormal shutdown, and the calculation formula is as follows; ; If the change No more than the benchmark change , then the shutdown is judged to be a normal shutdown, and the calculation formula is as follows: ; S303, set the cycle stop condition: When the following conditions are met for the first time, the loop is stopped and the time node of the data is reassigned to the shutdown start time of the hydro-generator set; ; in The current time node The active power, The minimum operating threshold of the hydro-generator set set in step S2; is the range of change of the set benchmark; The loop stops after all the data in period T has been circulated; If no data is retrieved within the T period, it will be directly judged as an abnormal shutdown due to data loss.

[0037] Furthermore, when marking data and determining the downtime category, the data used is the original data at the second level, but not all the second-level data in the T period is used. Instead, part of the second-level data in a fixed interval before and after each data node of the minute-level data is used. This ensures the use of second-level data to achieve higher accuracy and avoids the problem of slow processing speed caused by processing a large amount of second-level data.

[0038] Preferably, the reference variation range is the difference between the active power data of two adjacent time nodes during a normal shutdown process; the data of the normal shutdown process is obtained through the planned shutdown history data in the shutdown events in the data set.

[0039] Preferably, in step S5, according to the start-stop record of the hydro-generator set and the maintenance record of the hydro-generator set in S1 to S4, the method for generating a reliability event list of the hydro-generator set based on the event switching mode includes: S501, based on historical data and unit operating conditions, summarizes three common event switching modes, including: First mode: operation, standby, operation mode; Second mode: operation, standby, planned outage, standby, operation mode; The third mode: operation, unplanned outage, planned outage, standby, and operation mode; S502, generating a reliability event list of the hydro-generator set according to the start-up and shutdown records of the hydro-generator set and the maintenance records of the hydro-generator set in S1-S4 by following the steps below: a. All time periods from “start-up to shutdown” are considered as operation events; b. For the "shutdown to restart" time period and the shutdown type is normal shutdown, if there is no maintenance record in this time period, it is judged as a backup event; if there is a maintenance record in this time period, the reliability events in this time period are backup, planned shutdown and backup in order; c. For the "shutdown to restart" time period and the shutdown type is a general abnormal shutdown, search for maintenance records within this time period. The reliability events within this time period are unplanned shutdown, planned shutdown, and standby in order; if no maintenance records are found, it means that the maintenance records are missing, and the maintenance records are added to the data set; d. For the "shutdown to restart" time period and the shutdown type is abnormal shutdown due to data loss, if there is a maintenance record in this time period, the reliability events in this time period are unplanned shutdown due to lack of data, planned shutdown and standby in order; if there is no maintenance record in this time period, the reliability event in this time period is unplanned shutdown due to lack of data; Among them, the "start-up to shutdown" time period refers to the time period between the start time of startup and the start time of shutdown; the "shutdown to start-up" time period refers to the time period between the start time of shutdown and the start time of startup.

[0040] Furthermore, the first mode corresponds to the vast majority of normal standby scenarios. The unit enters the standby state after a normal shutdown and continues to be put into operation after the standby state ends. The second mode corresponds to scenarios with planned maintenance. The unit first enters the standby state after a normal shutdown, which can be determined by analyzing the shutdown time and the start time of the maintenance plan, and then enters the maintenance state as planned, returns to the standby state after the maintenance is completed, and finally is put into operation. The third mode corresponds to abnormal shutdown scenarios. If there is an unplanned shutdown, planned maintenance is usually arranged after the event to find out the cause of the unplanned shutdown, and then the unit returns to standby and is put into operation.

[0041] Preferably, the marked data is corrected, including reviewing the data marked with R based on the real-time data of the grid-connected state, and the review method is: Taking the time node corresponding to the data marked with R as the starting point, find the first grid-connected status record in the past or future M time. If the grid-connected status is grid-connected, the digital expression is 1, then modify the mark R to S, otherwise it remains unchanged; When executing the above review method, the data marked as R needs to meet the following conditions: ; in, For the current time point The active power, The minimum operating threshold of the unit determined in step S2; for The grid-connected status of the unit at the moment; is the minimum time that satisfies the following conditions: ; If no data is retrieved within M time, the original identification is maintained.

[0042] Furthermore, the identification method of the present invention is regarded as supervised learning, and the reliability events historically reported in the reliability system in the data set are used as reference for the results to set and adjust the relevant thresholds in the identification method of the present invention; conversely, the reliability event list generated by the identification method of the present invention is also used to check and fill in the gaps in the reported results in the reliability system.

[0043] Embodiment 2: In the data preprocessing and data marking steps, this embodiment takes Unit 16 of a certain station as an example, with an installed capacity of 700MW, and preliminarily sets the minimum threshold of active power in the operating state to 20MW. The processed and marked sample data are shown in Table 1 below:

[0044] Table 1: Preprocessed and labeled example data; In the step of generating the start and stop record of the unit, the example data provided in this embodiment is shown in Table 2:

[0045] Table 2: Start-stop record data; In the shutdown category determination step, the shutdown time finally determined by this embodiment is 12:02. In this step, searching forward 30 minutes, the active power of 6.1229 million kW at 11:53:39 drops to 0.012 million kW at the next moment 11:53:40. The change value exceeds the set threshold, and the time interval between the two points is 1 second, which meets the determination rules for general abnormal shutdowns. This embodiment takes the data of Unit 14 on January 6, 2024 as an example, and the example data of abnormal shutdown due to data missing is shown in Table 3:

[0046] Table 3: Example data of abnormal downtime due to missing data; In Table 3, the determined downtime is 21:27. In this step, searching forward 30 minutes, it can be found that the active power of 6,053,400 kW at 21:08:29 drops to -3,100 kW at the next moment 21:27:23. The change value exceeds the set threshold, but the time interval between the two points is about 20 minutes, which may be an abnormality or interruption in the data transmission process. Therefore, it is marked as an abnormal downtime due to data missing and is submitted for further manual review.

[0047] Embodiment three: Step S5 of this embodiment provides example data: According to the start-stop records and maintenance records of the hydro-generator sets in S1-S4, in the step of generating a reliability event list of the hydro-generator sets based on the event switching mode, the generated sample data is shown in Table 4 below:

[0048] Table 4: Reliability event list example data; From Table 4, we can see the generated reliability event list data of the hydro-generator set.

[0049] Embodiment 4: This embodiment provides data for modifying marked data, including data marked as R based on real-time data review of the grid-connected state, as shown in Table 5 below. Taking the data of Unit 21 around 16:47-16:55 on March 28, 2023 as an example, the sample data is as follows:

[0050] Table 5: Active power and identification, and grid-connected status data of the units within the same time range; Match the data marked as R from 16:55 to 16:57. The most recent past state timestamp corresponding to 16:55 is 16:54, at which time the grid-connected state is 0, so the mark remains R; the most recent past state timestamp corresponding to 16:56 is 16:54, at which time the grid-connected state is 0, so the mark remains R; the most recent past state timestamp corresponding to 16:57 is 16:57, at which time the grid-connected state is 1, so the correction mark is S. The corrected data is shown in Table 6 below:

[0051] Table 6: Data after review and correction; As shown in Table 7 below, the above method was used to analyze the data of 32 units in a certain station from 0:00 on January 1, 2023 to 0:00 on March 31, 2024, involving more than 979 million measurement point data.

[0052] As shown in Table 7 below, a total of 2,687 standby events, 9 unplanned outages and 48 unplanned outages were identified; the reliability management system recorded 2,750 standby events and 1 unplanned outage; the unplanned outage events that were the focus of reliability management and reported were accurately identified, and the outage events with abnormal characteristics in the unit operating conditions were monitored. The overall event identification results were consistent with the reporting results at a rate of 97.7%.

[0053] For shutdown events with abnormal characteristics in the unit operating conditions, the causes are analyzed from two aspects: data characteristics and on-site investigation, mainly including data missing, data anomalies, unit debugging and unit experiments.

[0054]

[0055] Table 7: Reliability event identification results; According to the user's reporting habits, the reliability system generally reports full-power operation events only when no other events occur in the month, and splits the events across months. From the results, it can be seen that the reliability event results identified based on the above steps are highly consistent with the reported data.

Claims

1. A method for identifying reliability events of hydropower equipment based on production monitoring data, characterized in that: include: S1, construct a data set for reliability event identification, which includes equipment monitoring data, maintenance records of hydro-generator units, and historical reliability events reported in the reliability system; S2, performing data preprocessing and data labeling on the equipment monitoring data in the data set to obtain labeled data; S3, generating a start and stop record of a hydro-generator set according to the marked data; S4, determining the shutdown category based on the start and stop records of the hydro-generator set; S5, generating a reliability event list of the hydro-generator set based on the event switching mode according to the start-stop record of the hydro-generator set and the maintenance record of the hydro-generator set obtained in S1-S4; S6, compare the generated reliability event list of the hydro-generator set with the historical reporting records in the reliability system to verify the results and optimize the rules.

2. A method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 1, characterized in that: In step S1, the equipment monitoring data includes the power data of the hydro-generator set and the equipment status data; Power data includes active power and reactive power. Each hydro-generator set has a power measurement point. During the operation of the hydro-generator set, the power data has a frequency of seconds. The equipment status refers to the status of the hydro-turbine generator set, including eight equipment statuses: shutdown status, idling status, no-load status, grid-connected status, transition status, water system operation status, maintenance status, and unknown status. One equipment status measurement point is set for each type of status, and eight equipment status measurement points are set for each hydro-turbine generator set.

3. A method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 2, characterized in that: In step S1, the maintenance record of the hydro-generator set includes a maintenance application form, and the maintenance application form includes the equipment name, maintenance level, work content, adjustment order time and completion report time.

4. A method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 3, characterized in that: In step S1, the historical reliability events reported in the reliability system include: Reliability event types include operation events and shutdown events; Operation events consist of full-output operation and reduced-output operation; shutdown events consist of standby, planned shutdown, unplanned shutdown and deactivation events; the basis for reporting reliability events is the start-up and shutdown records and maintenance application forms in the power production management system.

5. A method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 4, characterized in that: In step S2, data preprocessing and data labeling are performed on the data in the data set, and the labeled data obtained includes: Process the data in the dataset into minute-level data, and use marking methods to confirm the minute node where the startup or shutdown operation is located; Set the minimum threshold for the operation of the hydro-generator set Based on minute-level data, the active power of the hydro-generator unit in operation is greater than the threshold. The data is recorded as the operating state and marked as S. The active power of the hydro-generator set in the operating state is less than the threshold The data record is in outage state, marked as R.

6. A method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 5, characterized in that: In step S3, generating a start and stop record of a hydro-generator set according to the marked data includes: According to the data mark in step S2, the start time and the end time of shutdown within the generation cycle are generated, and the generation rule is: The first recorded time node of N consecutive data marked as R is the shutdown start time of the hydro-generator set; The first recorded time node of N consecutive data marked as S is the start time of the hydro-generator set; Perform data correction on the marked data.

7. A method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 6, characterized in that: In step S4, determining the shutdown type based on the start and stop records of the hydro-generator set includes: On the basis of the shutdown termination time identified in step S3, the shutdown process is judged to be a normal shutdown or an abnormal shutdown based on the characteristics of the original second-level data of the active power of the hydro-generator set, wherein the original data is the data collected in step S1 with a frequency of seconds; the specific judgment method is: S301, setting the benchmark change range and interval duration threshold ; S302, taking the shutdown end time as the reference time point, t minutes ahead of the reference time point to form a time period T, and cyclically calculating the time nodes within the time period T Active power data and the previous time node Active power data The change range And the interval duration , for the change range and interval duration Execution judgment: If the change Exceeding the baseline change , and the interval time Not less than the interval duration threshold , then the shutdown is judged to be an abnormal shutdown due to data loss. The calculation formula is as follows: ； If the change Exceeding the baseline change , and the interval time Less than the interval duration threshold , then the shutdown is judged as a general abnormal shutdown, and the calculation formula is as follows; ； If the change No more than the benchmark change , then the shutdown is judged to be a normal shutdown, and the calculation formula is as follows: ； S303, set the cycle stop condition: When the following conditions are met for the first time, the loop is stopped and the time node of the data is reassigned to the shutdown start time of the hydro-generator set; ； in The current time node The active power, The minimum operating threshold of the hydro-generator set set in step S2; is the range of change of the set benchmark; Stop the loop after all the data in period T has been circulated; If no data is retrieved within the T period, it will be directly judged as an abnormal shutdown due to data loss.

8. A method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 7, characterized in that: The benchmark change range is the difference between the active power data of two adjacent time nodes during the normal shutdown process; the data of the normal shutdown process is determined by the planned shutdown history data in the shutdown events in the data set.

9. A method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 8, characterized in that: In step S5, according to the start-stop record of the hydro-generator set and the maintenance record of the hydro-generator set in S1-S4, the method for generating a reliability event list of the hydro-generator set based on the event switching mode includes: S501, based on historical data and unit operating conditions, summarizes three common event switching modes, including: First mode: operation, standby, operation mode; Second mode: operation, standby, planned outage, standby, operation mode; The third mode: operation, unplanned outage, planned outage, standby, and operation mode; S502, according to the start-up and shutdown records of the hydro-generator set and the maintenance records of the hydro-generator set in S1-S4, a reliability event list of the hydro-generator set is generated by following the steps below: a. All time periods from "start-up to shutdown" are considered as operation events; b. For the "shutdown to restart" time period and the shutdown type is normal shutdown, if there is no maintenance record in this time period, it is judged as a backup event; if there is a maintenance record in this time period, the reliability events in this time period are backup, planned shutdown and backup in order; c. For the "shutdown to restart" time period and the shutdown type is general abnormal shutdown, search for maintenance records in this time period. The reliability events in this time period are unplanned shutdown, planned shutdown and standby in order; if no maintenance record is found, it means that the maintenance record is missing, and the maintenance record is added to the data set; d. For the "shutdown to restart" time period and the shutdown type is abnormal shutdown due to data loss, if there is a maintenance record in this time period, the reliability events in this time period are unplanned shutdown due to lack of data, planned shutdown and standby in order; if there is no maintenance record in this time period, the reliability event in this time period is unplanned shutdown due to lack of data; The "start-up to shutdown" time period refers to the time period between the start time of startup and the start time of shutdown; the "shutdown to start-up" time period refers to the time period between the start time of shutdown and the start time of startup.

10. The method for identifying reliability events of hydropower equipment based on production monitoring data according to claim 6, characterized in that: The marked data is corrected, including real-time data review of the data marked with R based on the grid-connected status. The review method is: Taking the time node corresponding to the data marked with R as the starting point, find the first grid-connected status record in the past or future M time. If the grid-connected status is grid-connected, the digital expression is 1, then modify the mark R to S, otherwise it remains unchanged; When executing the above review method, the data marked as R needs to meet the following conditions: ； in, The current time node The active power, The minimum operating threshold of the unit determined in step S2; for The grid-connected status of the unit at the moment; is the minimum time that satisfies the following conditions: ； If no data is retrieved within M time, the original identification is maintained.

Images