A new energy power station equipment operation and maintenance work order closed loop method

Abstract

This invention discloses a closed-loop method for operation and maintenance work orders in new energy power plant equipment, specifically relating to the field of new energy power plant operation and maintenance management technology. It addresses the problems in existing new energy power plant operation and maintenance work order processing, such as the tendency to repeatedly create work orders when the same type of anomaly on the same equipment is triggered consecutively or recurs intermittently, and the lack of retesting constraints after on-site handling. This leads to work orders being closed directly based solely on manual receipts, and subsequent work order creation lacking state boundary criteria. By establishing an equipment anomaly boundary rule table, collecting equipment operation alarm data and equipment operating condition data, and generating equipment anomaly evidence pieces, this invention matches the equipment anomaly evidence pieces with existing work order state boundaries to determine the work order acceptance status. This achieves the effect of inheriting existing work order state boundaries when the same type of anomaly on the same equipment is triggered consecutively or recurs intermittently, avoiding the creation of duplicate work orders, and maintaining the continuity of the anomaly handling chain.

Description

Technical Field

[0001] This invention relates to the field of operation and maintenance management technology for new energy power plants, specifically a closed-loop method for operation and maintenance work orders for new energy power plant equipment. Background Technology

[0002] New energy power plants typically include various types of field equipment such as photovoltaic inverters, transformer substations, combiner boxes, energy storage converters, meteorological data acquisition units, and station-side communication equipment. During power plant operation, the station-side monitoring system, centralized control platform, or regional operation and maintenance platform continuously collects information such as equipment operation alarms, operating conditions, communication status, power rationing status, and maintenance plans. Alarms requiring manual handling are then transferred to the operation and maintenance work order system. Existing operation and maintenance work order systems generally generate and dispatch work orders according to equipment type, alarm level, site location, operation and maintenance team, or area of ​​responsibility. After receiving the order, field personnel arrive at the equipment location to complete operations such as inspection, repair, replacement, reset, or confirmation. They then submit a field execution receipt via a handheld terminal or work order page, and the work order system records the processing result, supporting documents, and work order status. This method can meet the needs of general alarm transfer, manual work order dispatch, handling record keeping, and progress management, and is a common work order processing method in the daily operation and maintenance of new energy power plants.

[0003] However, alarm statuses of equipment in new energy power plants do not always appear in a single, isolated manner. Influenced by factors such as light fluctuations, grid connection status, power rationing commands, equipment temperature rise, branch contact status, communication link stability, and maintenance plans, similar anomalies in the same equipment may persist continuously across multiple data collection cycles, or reappear after a short recovery. For example, a low string current alarm for the same inverter may persist continuously in adjacent data collection cycles, excessive oil temperature in a transformer substation may recur intermittently under load, and communication interruptions may be triggered multiple times when the equipment should be online. Existing work order processing procedures typically use a single alarm record pushed by the monitoring system or a manually created work order as the basic processing object, focusing on alarm reception, work order dispatch, personnel handling, and work order closure. There is a lack of clear data criteria for determining whether similar anomalies in the same equipment still fall within the same handling boundary across different data collection cycles, different alarm segments, and different work order statuses.

[0004] In the aforementioned scenarios, if the system only triggers new work orders based on new alarms, it's easy for the same type of anomaly on the same device to be duplicated while existing work orders are still in the pending, executing, pending retesting, or pending verification stages. Conversely, if the system only closes work orders based on manual confirmations, it may prematurely terminate work orders when similar alarms recur after processing, communication fails during the retesting phase, or the equipment is under maintenance and shut down. Especially in centralized operation and maintenance scenarios for new energy power plants, a regional operation and maintenance center often manages multiple sites and a large number of devices simultaneously. Duplicate work orders increase the burden of dispatching and verification, while closure records lacking retesting constraints can affect the judgment of whether the same anomaly should continue to be considered within the existing work order status boundary in the next cycle. Therefore, in operation and maintenance scenarios where similar anomalies on the same device in new energy power plants are continuously triggered or recur intermittently, it is urgent to solve the problem of how to determine whether the current anomaly continues to belong to the existing work order status boundary, and how to uniformly handle the closed, pending verification, or secondary dispatch status based on retesting results after on-site handling. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a closed-loop method for operation and maintenance work orders of new energy power plant equipment, thereby solving the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a closed-loop method for operation and maintenance work orders of new energy power plant equipment, comprising:

[0007] S1. The operation and maintenance management server establishes a device anomaly boundary rule table, recording the device number, anomaly type, valid rule version, valid operating conditions, alarm duration window, merging time interval, retest window, and cooling shutdown window;

[0008] S2. The operation and maintenance management server collects the equipment operation alarm data and equipment status data of the current period, merges the same type of anomalies of the same equipment according to the equipment anomaly boundary rule table, and generates equipment anomaly evidence pieces.

[0009] S3. The operation and maintenance management server reads the work order status boundary ledger, matches the equipment anomaly evidence pieces with the existing work order status boundary, and determines the work order acceptance status.

[0010] S4. The operation and maintenance management server generates a work order policy based on the work order acceptance status. The work order policy includes the accepted work order number, the retest window, the prohibition of direct closure flag, and necessary feedback fields.

[0011] S5. The operation and maintenance management server receives the field execution receipt, matches the field execution receipt with the necessary feedback fields in the work order policy, and after the match is successful, the corresponding work order is transferred to the pending retest status.

[0012] S6. The operation and maintenance management server reads the processed equipment operation alarm data in the retest window, sets the work order to one of the following states according to the reproduction of similar alarms: closed, pending review, or secondary dispatch, and writes it back to the work order status boundary ledger to form the work order status boundary version for the next cycle of work order creation judgment.

[0013] Furthermore, S1 includes:

[0014] The operation and maintenance management server converts the time field to a unified time base for the site.

[0015] Organize the equipment anomaly boundary rule table in the following order: equipment number, anomaly type, effective operating condition, alarm duration window, merging time interval, retest window, and cooling shutdown window;

[0016] Each rule record generates a valid rule version and effective time, and is recorded using the same device number, the same exception type, and the valid rule version as unique keys.

[0017] Furthermore, S2 includes:

[0018] The operation and maintenance management server uses a left-closed, right-open approach to process the collected equipment operation alarm data and equipment status data.

[0019] Convert the alarm start time and alarm recovery time to a unified time base at the station.

[0020] For continuous alarms across cycles, retain the original alarm start time and read the valid rule version in the device anomaly boundary rule table according to the alarm start time to determine the anomaly type.

[0021] Furthermore, S2 also includes:

[0022] The operation and maintenance management server uses the same device number, the same anomaly type, and the same valid rule version as the merging criteria;

[0023] Confirm the valid alarm range based on the alarm duration window;

[0024] Adjacent valid alarm intervals are merged based on the merging time interval to form equipment anomaly evidence pieces, and equipment anomaly evidence piece numbers are generated according to equipment number, anomaly type, valid rule version, and evidence start time.

[0025] Furthermore, S3 includes:

[0026] The operation and maintenance management server reads the work order status boundary ledger according to the device anomaly evidence piece number;

[0027] Filter existing work order status boundaries by equipment number and anomaly type;

[0028] Based on the current work order status, and by comparing the evidence start time and evidence end time with the boundary start time, boundary end time, retest window, closing cooldown window, and status update time, matching results are generated for boundary acceptance, boundary continuation acceptance, acceptance pending retest, cooling-down acceptance, secondary dispatch acceptance, and newly created acceptance.

[0029] When two or more matching results are generated for the same equipment anomaly evidence, a unique work order acceptance status is determined in the following order: acceptance pending retest, acceptance for secondary dispatch, acceptance within the boundary, acceptance for boundary continuation, acceptance for cooling, and acceptance for new creation.

[0030] Furthermore, S4 includes:

[0031] The operation and maintenance management server determines the work order number based on the work order acceptance status.

[0032] Read the retest window corresponding to the equipment number, anomaly type and valid rule version in the equipment anomaly boundary rule table, and use the retest window if the existing work order status boundary to be retested has a retest window recorded.

[0033] The policy will include a prohibition on direct closure flag, necessary feedback fields, and policy version number in the work order policy. The necessary feedback fields include on-site execution completion time, disposal object, disposal action, on-site confirmation result, and on-site credential identifier.

[0034] The equipment malfunction evidence piece number is used as the sole submission criterion for accounting purposes.

[0035] Furthermore, S5 includes:

[0036] The operation and maintenance management server reads the work order policy according to the work order number, the equipment abnormality evidence piece number, and the policy version number;

[0037] Match the on-site execution receipt with the necessary feedback fields in the work order strategy in the following order: on-site execution completion time, disposal object, disposal action, on-site confirmation result, and on-site voucher identifier.

[0038] Once the match is successful, it will be confirmed that the flag cannot be closed directly, and the accepted work orders whose current status is either pending execution or in execution will be transferred to the pending retest status.

[0039] Furthermore, S6 includes:

[0040] The operation and maintenance management server reads the retest window in the work order policy and uses the on-site execution completion time in the on-site execution receipt as the starting point of the retest window.

[0041] Accumulate the observable duration during periods when there is no downtime, no maintenance, effective communication, and similar alarms can be read;

[0042] Read the processed equipment operation alarm data according to the equipment number, anomaly type and valid rule version that are consistent with the equipment anomaly evidence film, and form a similar alarm reproduction situation.

[0043] Furthermore, S6 also includes:

[0044] The operation and maintenance management server determines one of the following states based on the reproduction of similar alarms: shutdown, pending review, or secondary dispatch.

[0045] The final status is determined in the order of secondary dispatch status, pending review status, and closed status;

[0046] The final state is written back to the work order state boundary ledger to form the work order state boundary version for the next cycle's work order creation judgment.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. By establishing a device anomaly boundary rule table, collecting device operation alarm data and device operating condition data, and generating device anomaly evidence pieces, the device anomaly evidence pieces are then matched with existing work order status boundaries to determine the work order acceptance status. This achieves the effect of inheriting existing work order status boundaries, avoiding duplicate work order creation, and maintaining the continuity of the anomaly handling chain when the same type of anomaly is triggered continuously or reproduced at intervals on the same device.

[0049] 2. By writing the work order number, retest window, prohibition of direct closure, and necessary feedback fields into the work order strategy, and after the on-site execution receipt matching is successful, the corresponding work order is transferred to the retesting state. Then, based on the reproduction of similar alarms in the retest window, the work order is set to the closed state, the pending review state, or the secondary dispatch state and written back to the work order status boundary ledger. This achieves the effect of preventing work orders from being directly closed by manual recording and enabling the processed retest results to inversely constrain the next cycle of work order creation judgment. Attached Figure Description

[0050] Figure 1 A schematic diagram of the overall process for the closed-loop method of operation and maintenance work orders for new energy power plant equipment;

[0051] Figure 2 A diagram illustrating the creation of the equipment anomaly boundary rule table and the recording of valid rule versions.

[0052] Figure 3 This diagram illustrates the merging of alarm data for the current period and the generation of evidence of equipment malfunctions.

[0053] Figure 4 A schematic diagram for determining the acceptance status of work order status boundary ledgers when matching equipment anomaly evidence pieces;

[0054] Figure 5A diagram illustrating the generation of work order strategies and the writing of necessary feedback fields;

[0055] Figure 6 A schematic diagram illustrating the on-site execution receipt matching and the transition to the pending retest status;

[0056] Figure 7 This is a schematic diagram for retest status determination and work order status boundary version write-back. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Example: Combined with Appendix Figures 1-7 This embodiment provides a closed-loop method for operation and maintenance work orders of new energy power plant equipment, including:

[0059] S1. The operation and maintenance management server establishes a device anomaly boundary rule table, recording the device number, anomaly type, valid rule version, valid operating condition, alarm duration window, merging time interval, retest window, and cooling-off window. The specific implementation is as follows:

[0060] In this embodiment, the operation and maintenance management server is a work order closed-loop processing device deployed at the new energy power plant station or regional operation and maintenance center. It is used to perform tasks such as establishing equipment anomaly boundary rule tables, generating equipment anomaly evidence pieces, determining work order acceptance status, generating work order strategies, matching on-site execution receipts, and writing back retest status. The operation and maintenance management server establishes a unified time base at the station. This unified time base is the unified time standard used for time conversion, interval judgment, and accounting of equipment operation alarm data, equipment operating data, work order status records, on-site execution receipts, and retest read records within the same new energy power plant. When the same record source has both original collection time and server receiving time, the original collection time is prioritized and then converted by the operation and maintenance management server before accounting. Records pending verification are used to store records where there are conflicts or where the equipment number, anomaly type, time field, or version field cannot participate in the current judgment. Records pending supplementation are used to store cases where fields are missing and allow supplementation of source record identifiers to continue forming valid records. Neither records pending verification nor records pending supplementation participate in the formation of the current valid status. After supplementation or verification is completed, the process re-enters the corresponding step according to the unique submission standard.

[0061] The new energy power plant operation and maintenance management server establishes the equipment anomaly boundary rule table during the rule filing period. The rule filing period is the compilation period before the current rule version is put into use. The execution entity is the station-side operation and maintenance management server or the regional operation and maintenance center rule maintenance server. On-site sources include equipment ledgers, alarm dictionaries, operating condition records, maintenance plan records, historical work order records, and manufacturer maintenance manuals.

[0062] The equipment anomaly boundary rule table only records the equipment number, anomaly type, effective operating conditions, alarm duration window, merging time interval, retest window, and cooling shutdown window. It does not record subsequent equipment anomaly evidence pieces and work order acceptance status in advance in this table.

[0063] The order of processing is fixed as follows: first confirm the device number, then confirm the anomaly type, then confirm the valid operating condition, and then determine the alarm duration window, merging time interval, retest window, and cooling shutdown window; if the previous field is not confirmed, the next field will not be processed.

[0064] The equipment number is taken from the unique identification code in the equipment ledger and corresponds to the asset records of inverters, transformer substations, combiner boxes, energy storage converters, and meteorological acquisition units on site. The record is based on the unique number in the equipment ledger that is in operation. When the same equipment has multiple system numbers, the number bound to the operation and maintenance management server in the equipment ledger shall prevail. The point number inside the monitoring system is only used for source trace and does not replace the equipment number.

[0065] If the equipment number is missing, duplicated, the out-of-service status is not confirmed, or the parent equipment relationship is missing, it should not be automatically filled in, and the current rule record should be transferred to the pending record entry.

[0066] The anomaly type is taken from the alarm dictionary and the alarm category name that matches the device type to which the device number belongs. When forming the anomaly type, the device type is read first, then the alarm code and alarm name are read, and finally alarms with the same operation and maintenance handling criteria under the same device type are grouped into the same anomaly type. Alarms across device types are not merged into the same anomaly type. If the same alarm code has different meanings under different device types, rule records are formed separately.

[0067] When the anomaly type lacks an alarm dictionary source, the alarm code corresponds to multiple handling standards, or the alarm level is inconsistent with the device type, the current rule record will be transferred to pending verification.

[0068] Effective operating conditions are combinations of operating states that allow the current anomaly type to enter subsequent alarm merging. They are derived from operating condition records and maintenance plan records. When compiling these, the maintenance plan coverage period is first removed, then the downtime maintenance period is removed, and then the communication status is checked to see if the anomaly type can be observed. Finally, the operating condition boundaries that are allowed to take effect are recorded. For example, low string current corresponds to grid-connected power generation, no power restriction, and effective communication; excessive oil temperature in the transformer corresponds to load operation and effective temperature measurement channel; and communication interruption corresponds to the equipment being online and the station network not being in a station-wide interruption.

[0069] Effective operating conditions must not be replaced by descriptions based on human experience; they must be recorded as readable operating condition field names. If the source of an operating condition field cannot be confirmed, it should be moved to the stage of pending data entry.

[0070] The alarm duration window is the minimum duration, in minutes, from the first valid alarm time to the point where subsequent equipment anomaly evidence can be formed under the same equipment number, the same anomaly type, and the same valid operating conditions. The starting point is the valid alarm start time confirmed by the alarm dictionary, and the ending point is the time when continuous alarms are maintained for this duration. When compiling historical records, maintenance and debugging, manual testing, station-wide communication interruption, and equipment downtime periods are excluded. Instantaneous recovery alarms are not included in the formation of this field.

[0071] The merging time interval is the maximum interval between two valid alarms under the same equipment number and the same anomaly type, which is allowed to be classified into the same work order status boundary in subsequent stages. The unit is minutes. The starting point is the recovery time of the previous valid alarm and the ending point is the start time of the next valid alarm. When sorting, only duplicate alarm records that are manually confirmed to belong to the same on-site handling object in historical work orders are used. Records with different equipment numbers, different anomaly types, and different maintenance batches are not used.

[0072] The retest window is the observation period for subsequent steps to read equipment operation alarm data after the on-site execution receipt is completed. The unit is hours, and the starting point is fixed at the time of completion of the on-site execution receipt. If the equipment is in a state of shutdown, maintenance, or communication failure during this period, the rule table only records this period and does not include it in the observable duration. Subsequent steps calculate the retest window according to the cumulative observable duration and do not determine the work order closure in this step.

[0073] The "Close Cooling Window" is the retention period for reading the existing work order status boundary when the same device number and the same anomaly type alarms again after the work order is closed. The unit is hours, the starting point is fixed at the work order closure time, and the ending point is the time when the window expires. Historical closed work orders that have not formed a retest record are not used as the basis for determining this field.

[0074] The above time fields are uniformly converted to minutes or hours for accounting purposes. The time base is the time of the operation and maintenance management server, and the source system time is only retained as the source record.

[0075] After the fields are organized, the server performs an integrity check in the following order: device number, anomaly type, effective operating condition, alarm duration window, merging time interval, retest window, and cooling shutdown window. The device number and anomaly type should not be automatically filled in. The effective operating condition can only be selected from existing operating condition records. The time field can be filled in by maintenance personnel based on the manufacturer's maintenance manual or historical work order records, but the source record identifier must be written when filling in the field.

[0076] When field conflicts exist, the order of priority is determined as follows: equipment ledger over monitoring point table, alarm dictionary over manual notes, maintenance plan over operating condition record, and manufacturer's maintenance manual over unclosed historical work order; fields that still cannot be determined will not generate valid rule records.

[0077] Each rule record generates a rule version number and an effective time. Only one valid rule version is allowed for the same device number and the same anomaly type within the same effective time. After a new version is submitted, the old version enters the historical retention state and is neither deleted nor overwritten. When the next step reads the rule, the valid rule version whose alarm start time falls within the current period's device operation alarm data shall prevail.

[0078] Rule posting uses the same device number, exception type, and rule version number as unique keys. When the same unique key is submitted repeatedly, the existing posting result is returned and no duplicate record is generated. Rules in the same batch are submitted after being sorted by device number and exception type. When duplicate unique keys are found in the batch, the record of the first completed verification is retained, and subsequent duplicate records are entered into the duplicate submission record.

[0079] Rule records are stored in the rule base of the operation and maintenance management server in the form of relational tables, configuration tables, or rule files with versions. The minimum fields include device number, anomaly type, effective operating condition, alarm duration window, merging time interval, retest window, shutdown cooling window, rule version number, effective time, source record identifier, organizer account, and submission time.

[0080] The source records of personnel names, mobile phone numbers, and location trajectories only retain account identifiers and job identifiers, and are not written into the device anomaly boundary rule table.

[0081] After the rule base is submitted, the next step is to read the corresponding device number, exception type, and valid rule version according to the device number, alarm code, and alarm start time in the current period's device operation alarm data, and then perform subsequent merging according to the rule version.

[0082] Preferably, when a photovoltaic power station establishes a rule for low inverter string current, the equipment number is taken from the inverter ledger code, the anomaly type is mapped from the alarm code to low string current, the effective operating conditions are limited to grid-connected power generation, non-limited power generation, and effective communication, the alarm duration window is set to 10 minutes, the merging time interval is set to 30 minutes, the retest window is set to 2 hours, and the cooling shutdown window is set to 24 hours; when the rule is checked, more than 20 rules are checked in each batch, the equipment number is consistent with the ledger, the anomaly type is consistent with the alarm dictionary at no less than 99%, the time field unit is consistent at 100%, and if a key field is found to be missing, the rule will not be included in the effective version.

[0083] Alternatively, the equipment anomaly boundary rule table can be maintained by the station server and synchronized to the regional operation and maintenance center, or it can be uniformly maintained by the regional operation and maintenance center and then distributed to the station server. As long as the equipment number, anomaly type, effective operating condition, alarm duration window, merging time interval, retest window, shutdown cooling window, rule version number, effective time and source record identifier are retained after synchronization, and the effective rule version read in the next step is consistent, they all belong to the same implementation method.

[0084] S2. The operation and maintenance management server collects equipment operation alarm data and equipment status data for the current period, merges similar anomalies of the same equipment according to the equipment anomaly boundary rule table, and generates equipment anomaly evidence pieces. The specific implementation is as follows:

[0085] The operation and maintenance management server collects equipment operation alarm data and equipment status data according to the current cycle. The current cycle adopts a left-closed and right-open sorting interval. The starting point is the end time of the previous cycle and the ending time is the end time of the current cycle. Records that arrive at the end time are included in the next cycle to avoid duplicate collection in adjacent cycles. This action is triggered by the station-side operation and maintenance management server according to the unified time base of the station. The field sources include power plant monitoring alarm records, equipment operation status records, communication status records, limited emission records, and maintenance plan records.

[0086] Equipment operation alarm data comes from station-side monitoring alarm records, and at least the equipment number, alarm code, alarm name, alarm start time, alarm recovery time, alarm level, and source record identifier are recorded. The equipment number follows the equipment number standard in the equipment anomaly boundary rule table. The alarm start time and alarm recovery time are converted to the station-side unified time base. Records with alarm start time earlier than the start of the current cycle and alarm recovery time that is empty or later than the end of the current cycle are collected as cross-cycle continuous alarms, and their original alarm start time remains unchanged. This cycle only records the continuous coverage from the start of this cycle to the end of this cycle.

[0087] Equipment operating condition data comes from equipment operation status records, communication status records, limited transmission records, and maintenance plan records. At a minimum, the equipment number, operating condition field name, operating condition start time, operating condition end time, operating condition value, and source record identifier must be recorded. The operating condition field name must correspond to a valid operating condition in the equipment abnormal boundary rule table. Records that do not correspond will not be included in the current cycle's data processing.

[0088] The server first assigns the equipment operation alarm data and equipment status data to the same field equipment according to the equipment number, then reads the valid rule version in the equipment anomaly boundary rule table according to the alarm start time, and then determines the alarm code as the anomaly type according to the valid rule version; the anomalies of the same type shall be classified according to the anomaly type determined by the valid rule version, and not according to the similarity of alarm name text.

[0089] Records with missing device numbers, missing alarm codes, missing alarm start times, alarm codes that cannot be mapped to anomaly types, or alarm start times that do not match a valid rule version are written to the pending verification records and are not included in the current period's merging.

[0090] Time alignment is based on the time of the operation and maintenance management server, and the start and end times of equipment operation alarm data and equipment status data are uniformly converted to the same time axis. When the same equipment number, the same alarm code, and the same alarm start time come from multiple sources, the station monitoring alarm record is used first, and the edge acquisition record only retains the source record identifier. If the station monitoring alarm record is missing the alarm recovery time and the edge acquisition record has the same alarm recovery time, the alarm recovery time can be added and the source of the addition can be recorded.

[0091] The noise reduction order is fixed as follows: first, remove manual test records; then, remove maintenance plan coverage records; then, remove records caused by station-wide communication interruptions; and finally, filter according to the valid operating conditions in the equipment anomaly boundary rule table. When the alarm occurrence interval falls into the valid operating condition for only part of the time, only the overlapping part is extracted for merging. The starting point of the overlapping part is the later of the alarm occurrence interval starting point and the valid operating condition starting point, and the ending point of the overlapping part is the earlier of the alarm occurrence interval ending point and the valid operating condition ending point.

[0092] Automatic completion is only applicable to cases where alarm recovery time is missing and the alarm remains unrecovered within the current period. The completion value is the end time of the current period and is marked as the provisional end of the current period. The device number, alarm code, alarm start time, operating condition field name, and operating condition value must not be automatically completed. If any of these are missing, the processing of that record will be stopped.

[0093] When merging similar anomalies of the same device, the server uses the same device number, the same anomaly type, and the same valid rule version as a fixed merging criterion. The alarms are arranged in order of their starting points after being extracted. First, the valid alarm interval is determined based on the alarm duration window. The valid alarm interval is the alarm interval of the same type of anomaly of the same device that is continuously maintained under effective operating conditions and reaches the alarm duration window. The duration is calculated from the start to the end of the continuous interval. Intervals that do not reach the alarm duration window do not form a piece of evidence of device anomaly.

[0094] The interval between adjacent valid alarm intervals is calculated from the end of the previous valid alarm interval to the start of the next valid alarm interval. If the interval does not exceed the merging time interval, it is included in the same equipment anomaly evidence piece; if it exceeds the merging time interval, it is formed into a new equipment anomaly evidence piece.

[0095] The equipment anomaly evidence piece is a merged record generated in this step. It records the evidence range of the same type of anomaly of the same equipment after reaching the requirements of the equipment anomaly boundary rule table in the current period. It includes at least the equipment anomaly evidence piece number, equipment number, anomaly type, valid rule version, evidence start time, evidence end time, source record identifier participating in the merge, operating condition matching result, alarm duration, and merge formation time.

[0096] The equipment anomaly evidence fragment number is generated in a fixed order of equipment number, anomaly type, valid rule version, and evidence start time; the evidence start time is taken as the starting point of the first valid alarm interval participating in the merging, the evidence end time is taken as the ending point of the last valid alarm interval participating in the merging, and the alarm duration is the sum of the durations of the valid alarm intervals participating in the merging, without including the interval between two valid alarm intervals in the alarm duration.

[0097] Equipment anomaly evidence pieces are stored in the equipment anomaly evidence piece record table on the operation and maintenance management server in the form of record rows. When writing, the equipment number, anomaly type, valid rule version, and evidence start time are used as the unique submission criteria. If the same criteria are submitted repeatedly, the existing equipment anomaly evidence piece number is returned and no record is generated again. If some records in the same batch cannot match the valid operating conditions, only those records are transferred to the pending verification record, and the remaining records continue to be merged.

[0098] After the accounting is completed, the next step is to read the corresponding equipment anomaly evidence piece according to the equipment anomaly evidence piece number, equipment number, anomaly type, and valid rule version, without directly reading the original alarm records that have been removed in this step.

[0099] The safety boundary is limited to the on-site records accessed in the current step. Personnel accounts, inspection locations, and contact information are not included in the equipment anomaly evidence file, and the source record identifier is only retained to the internal number of the traceable original record.

[0100] Preferably, the acquisition trigger interval of a photovoltaic power station is set to 5 minutes, the inverter equipment number is INV-03-02, the alarm code is mapped to low string current, the alarm duration window in the effective rule version is 10 minutes, and the merging time interval is 30 minutes; the equipment has a low string current alarm at 10:00 and continues until 10:12. During this period, the operating conditions are grid-connected power generation, non-limited power generation, and effective communication. The server keeps the original alarm start time unchanged in adjacent acquisition cycles and confirms that the alarm duration window has been reached at 10:10, forming a piece of equipment abnormality evidence. The evidence start time is 10:00 and the evidence end time is 10:12; the same equipment with the same abnormal type has an alarm again at 10:35 and continues until 10:48. The interval between the end of the previous effective alarm interval and the start of the next effective alarm interval is 23 minutes, which is included in the same piece of equipment abnormality evidence. The alarm duration is recorded as 12 minutes plus 13 minutes, which is 25 minutes.

[0101] No fewer than 20 pieces of equipment anomaly evidence should be randomly checked in each batch. The consistency rate between the equipment number and the equipment anomaly boundary rule table should be 100%, and the consistency rate between the anomaly type and the valid rule version should be no less than 99%. The evidence start time and evidence end time can be traced and confirmed by the current cycle collection record or the cross-cycle continuous alarm record. The original alarm start time of the cross-cycle continuous alarm should be retained, and the provisional end time of the current cycle should be recorded when the alarm has not been recovered.

[0102] Alternatively, equipment operation alarm data and equipment status data can be pushed from the station-side monitoring server to the operation and maintenance management server, or the operation and maintenance management server can actively read them according to the current cycle. As long as the collection interval, equipment number, anomaly type, valid rule version, alarm duration window, and merging time interval are consistent, and the equipment anomaly evidence pieces corresponding to the same type of anomaly of the same equipment can be generated, they all belong to the same implementation method.

[0103] S3. The operation and maintenance management server reads the work order status boundary ledger, matches the device anomaly evidence fragments with the existing work order status boundaries, and determines the work order acceptance status. The specific implementation is as follows:

[0104] After the device anomaly evidence piece is written, the operation and maintenance management server reads the work order status boundary ledger. The reading action is triggered by the station-side operation and maintenance management server according to the device anomaly evidence piece number. The on-site source is the work order record that has been written, the work order status change record, the retest status record, and the closure record. The effective period is from the time the device anomaly evidence piece is formed until the next step of reading the work order acceptance status.

[0105] The work order status boundary ledger is a collection of records that store existing work order status boundaries according to the work order number. Its on-site meaning is the time boundary and status boundary that the same equipment and the same type of anomaly have occupied in the work order process. It records at least the work order number, equipment number, anomaly type, valid rule version, boundary start time, boundary end time, current work order status, retest window, closed cooling window, status update time, and ledger version number. Among them, the equipment number and anomaly type follow the definition in the equipment anomaly evidence piece. The boundary start time is the evidence start time of the first equipment anomaly evidence piece received for this work order. The boundary end time is the evidence end time of the most recent equipment anomaly evidence piece received for this work order. The current work order status is the latest recorded status in the work order status change record.

[0106] The time interval determination adopts the standard of including the start time and excluding the end time; when the start time of the evidence of equipment abnormality is equal to the end time of a certain time interval, it is not considered to fall into that time interval, but will enter the subsequent interval determination.

[0107] When the server reads the work order status boundary ledger, it first filters the same field equipment by the equipment number in the equipment anomaly evidence piece, then filters the same type of anomaly by anomaly type, and then compares the existing work order status boundaries by the evidence start time and evidence end time. The existing work order status boundaries only include records whose current work order status is pending execution, in execution, pending retesting, pending review, secondary dispatch, or closed and still within the closure cooling window. Records whose current work order status is voided, tested, revoked, or closed and whose closure cooling window has expired are not included in the matching.

[0108] Ledger records with missing equipment number, anomaly type, boundary start time, or boundary end time are transferred to pending verification records and are not used as matching criteria. When the retest window is missing, it will not participate in the pending retest acceptance judgment, and when the cooling closure window is missing, it will not participate in the cooling acceptance judgment, but the ledger record can still be compared according to the acceptance within the boundary and the acceptance of the boundary continuation.

[0109] Existing work order status boundaries with the same valid rule version are given priority for matching; records with different valid rule versions but the same device number and exception type, and whose current work order status is still matchable, are not directly excluded. Only the valid rule version of the evidence piece and the valid rule version of the ledger are specified in the work order acceptance status record. If there are candidate records with the same valid rule version, the candidate records with different valid rule versions are not used.

[0110] When matching equipment anomaly evidence with existing work order status boundaries, first determine whether the evidence start time falls within the interval between the boundary start time and the boundary end time. If it does, it is determined as an acceptance within the boundary. If the evidence start time is earlier than the boundary start time and the evidence end time is later than the boundary start time, it is considered to overlap with the existing work order status boundary and is also determined as an acceptance within the boundary. If the evidence start time is later than or equal to the boundary end time, calculate the interval between the boundary end time and the evidence start time. If this interval does not exceed the merged time interval in the equipment anomaly boundary rule table, it is determined as a boundary continuation acceptance.

[0111] If no boundary acceptance or boundary continuation acceptance is formed, then determine whether the evidence start time falls within the retest window recorded by the existing work order status boundary. If it does, it is determined as acceptance pending retest. If it does not fall within the retest window, then determine whether the evidence start time falls within the closed cooling window. If it does, it is determined as cooling acceptance. If the current work order status is secondary dispatch and the evidence start time is later than the status update time, it is determined as secondary dispatch acceptance.

[0112] If the equipment number and anomaly type are the same but do not meet any of the above time boundaries, it is determined as a newly accepted work order; if there are no candidate records with the same equipment number and anomaly type in the work order status boundary ledger, it is also determined as a newly accepted work order.

[0113] If the same device abnormal evidence piece matches multiple existing work order status boundaries simultaneously, the server determines a unique work order acceptance status in the following order: acceptance pending retest, acceptance for secondary dispatch, acceptance within the boundary, acceptance for boundary continuation, acceptance for cooling down, and acceptance for new creation. If there are still multiple candidate records under the same work order acceptance status, the record with the same valid rule version, the later status update time, the boundary deadline time being closer to the evidence start time, and the earlier work order number generation time is selected in sequence. Other candidate records are only retained for matching traces.

[0114] After determining the work order acceptance status, the server writes the device anomaly evidence piece number, the matched work order number, the matching type, the matching basis time, the current work order status, the work order acceptance status, the valid rule version of the evidence piece, the valid rule version of the ledger, and the matching formation time into the work order acceptance status record. The work order acceptance status record uses the device anomaly evidence piece number as the unique submission criterion. If the same device anomaly evidence piece number is submitted repeatedly, the existing work order acceptance status is returned, and no duplicate record is formed.

[0115] When matching evidence of equipment anomalies within the same batch, the matching is performed sequentially according to the start time of the evidence. If an evidence of equipment anomalies is included in the pending verification record due to the missing key fields in the ledger, the matching of evidence of equipment anomalies will not be interrupted.

[0116] Both the reading of the work order status boundary ledger and the writing of the work order acceptance status retain the ledger version number, reading time, system account, number of candidate boundaries, and the final matched work order number. Old ledger records are not overwritten. Records involving on-site personnel names, contact information, and location trajectories are not written into the work order acceptance status record; only the account identifier and job identifier are retained.

[0117] After the accounting is completed, the next step is to read the work order acceptance status according to the equipment abnormality evidence piece number, and then continue to form a work order strategy based on the work order acceptance status.

[0118] Preferably, in a photovoltaic power station, the equipment anomaly evidence piece number is INV-03-02-string current low-V3-10:00, corresponding to the equipment number INV-03-02, the anomaly type is string current low, the valid rule version is V3, the evidence start time is 10:00, and the evidence end time is 10:48; in the work order status boundary ledger, there is an existing work order W20260706001 with a boundary start time of 09:40 and a boundary end time of 09:55. The current work order status is in execution, the valid rule version is V3, and the merging interval in the equipment anomaly boundary rule table is 30 minutes. Since there are 5 minutes between 10:00 and 09:55, the server determines that the work order acceptance status of this equipment anomaly evidence piece is boundary continuation acceptance, and writes the work order number W20260706001 into the work order acceptance status record.

[0119] For each batch of work orders, at least 20 records of acceptance status are randomly checked, with 100% consistency of equipment number, no less than 99% consistency of abnormality type, and 100% traceability of matching time relationship.

[0120] Alternatively, the work order status boundary ledger can be stored on the station-side operation and maintenance management server or on the regional operation and maintenance center server. As long as the device number, anomaly type, valid rule version, evidence start time, and evidence end time can be obtained according to the device anomaly evidence piece number when reading, and the unique work order acceptance status can be determined according to the same matching order, they all belong to the same implementation method.

[0121] S4. The operation and maintenance management server generates a work order policy based on the work order acceptance status. The work order policy includes the accepted work order number, the retest window, a flag prohibiting direct closure, and necessary feedback fields. The specific implementation is as follows:

[0122] After the work order acceptance status is recorded, the operation and maintenance management server generates a work order policy based on the work order acceptance status. The generation action is triggered by the station-side operation and maintenance management server according to the equipment anomaly evidence piece number. The on-site location corresponds to the operation and maintenance management server of the power plant central control room or regional operation and maintenance center. The effective period is from the time the work order acceptance status is formed until the next step receives the on-site execution receipt. The work order policy is the policy record formed in this step, which records the acceptance interface, necessary feedback interface, and closure constraints of the determined work order acceptance status in the subsequent work order flow. It includes at least the equipment anomaly evidence piece number, equipment number, anomaly type, effective rule version, work order acceptance status, accepted work order number, retest window, prohibition of direct closure flag, necessary feedback fields, policy formation time, and policy version number.

[0123] The server first reads the device anomaly evidence piece number, device number, anomaly type, valid rule version, work order acceptance status, and matched work order number from the work order acceptance status record. Then, it reads the retest window of the corresponding valid rule version from the device anomaly boundary rule table. If the work order acceptance status record lacks the device anomaly evidence piece number, device number, anomaly type, valid rule version, or work order acceptance status, no work order policy is generated, and the record is written to the pending verification record.

[0124] When generating work orders based on their acceptance status, the work order number matched in the work order acceptance status record is used as the acceptance work order number for acceptance within the boundary, acceptance continuing at the boundary, acceptance pending retesting, acceptance during cooling, and acceptance for secondary dispatch. If a matched work order number is missing, the record is written to the pending verification record. When a new acceptance does not match an existing work order number, the operation and maintenance management server generates a new acceptance work order number. The new acceptance work order number consists of the power plant number, generation date, and daily sequence number in a fixed order. The daily sequence number increments according to the new acceptance records already logged in the operation and maintenance management server, and canceled work order numbers are not reused.

[0125] The work order number remains unchanged in the work order policy and is not replaced by the device number, alarm code, or manual remarks. If the same device abnormal evidence piece number has already been formed into the work order policy and the generation action is triggered again, the server will return the existing work order policy and will not regenerate the work order number or increment the policy version number.

[0126] When a retest window is written into the work order policy, the retest window used for boundary acceptance, boundary continuation acceptance, cooling acceptance, secondary dispatch acceptance, and new acceptance is consistent with the equipment number, exception type, and valid rule version in the equipment exception boundary rule table. If a retest window already exists in the work order status boundary ledger corresponding to the acceptance to be retested, the retest window in the work order status boundary ledger is used, and the source of the retest window is recorded in the work order policy. If no retest window is recorded in the work order status boundary ledger corresponding to the acceptance to be retested, the retest window in the equipment exception boundary rule table is used.

[0127] The retest window only records the duration, starting caliber, and source rule version. This step does not determine whether the retest is completed. The starting caliber is fixed to the completion time of the subsequent on-site execution receipt.

[0128] When a retest window is missing, a valid rule version cannot be matched, or the source of the retest window cannot be traced, the record is written to the pending record and does not form a valid work order strategy.

[0129] When the "Prohibit Direct Closure" flag is written into the work order policy, its value is fixed as "Prohibit Direct Closure," and no value is set as "Allow Direct Closure." After the subsequent on-site execution receipt is completed, this flag restricts the corresponding work order from entering the closed state solely based on manual recording. Necessary feedback fields are written along with the work order policy. These fields must include at least the on-site execution completion time, the object of the action, the action taken, the on-site confirmation result, and the on-site voucher identifier. The object of the action is determined based on the equipment number and the type of anomaly; the action taken is determined based on the on-site handling criteria corresponding to the anomaly type; the valid value for the on-site confirmation result is fixed as "On-site handling completed"; and the on-site voucher identifier corresponds to the photo number, barcode number, or attachment number. Necessary feedback fields cannot be added or deleted by on-site personnel when submitting the on-site execution receipt; they remain unchanged under the same policy version.

[0130] When a work order is accepted as a new work order, the closing flag must not be written directly along with the new work order number. When a work order is accepted as a work order within the boundary, a work order continuing at the boundary, a work order awaiting retesting, a work order in cooling-off period, or a work order being dispatched for the second time, the closing flag must not be written directly into the current work order strategy record corresponding to the original work order number. Existing work order strategy records are retained as historical records and are not overwritten.

[0131] The strategy version number is determined according to the order of the strategy records of the work orders that have been settled under the same work order number. The first version is recorded as version 1. When a new work order strategy is formed from a new piece of equipment anomaly evidence, the version number is incremented by 1 based on the existing maximum sequence number. No new version is generated when the same piece of equipment anomaly evidence number is submitted repeatedly.

[0132] Before the work order policy is recorded, the server checks the completeness of the fields in the following order: device anomaly evidence piece number, work order acceptance status, accepted work order number, retest window, prohibition of direct closure flag, necessary feedback fields, and policy version number. The accepted work order number, prohibition of direct closure flag, necessary feedback fields, and policy version number must not be manually supplemented. The retest window can only be obtained from the device anomaly boundary rule table or the work order status boundary ledger.

[0133] Work order policies are stored in the work order policy record table of the operation and maintenance management server in the form of record rows. The device abnormality evidence piece number is used as the unique submission criterion. When submitted repeatedly, the existing work order policy is returned. Work order policies in the same batch are generated in the order of the work order acceptance status. When a single record enters the pending verification or pending supplementation stage, the generation of other records is not interrupted.

[0134] When writing work order policies, the following information is retained: equipment anomaly evidence piece number, source work order acceptance status record number, valid rule version, retest window source, policy formation time, and system account. Content involving on-site personnel names, contact information, and location trajectories is not written into the work order policy.

[0135] After the accounting is completed, the next step is to read the work order strategy according to the work order number and the equipment abnormality evidence piece number, and read the necessary feedback fields, retest window and prohibition of direct closure mark as status constraints after receiving the on-site execution receipt.

[0136] Preferably, in a photovoltaic power station, the equipment anomaly evidence piece number is INV-03-02-string current low-V3-10:00, the work order acceptance status is boundary continuation acceptance, the matched work order number is W20260706001, the retest window in the equipment anomaly boundary rule table is 2 hours, when the server generates the work order strategy, it writes the accepted work order number as W20260706001, writes the retest window as 2 hours, writes the prohibition of direct closure flag as prohibition of direct closure, and forms a second version of the work order strategy under this accepted work order number; when another equipment anomaly evidence piece is determined to be a newly accepted piece, the server generates a new accepted work order number W20260706002 according to the power station number, generation date, and daily sequence number, and writes it into the 2-hour retest window and the prohibition of direct closure flag.

[0137] For each batch of work orders, at least 20 work orders must be randomly checked, the traceability rate of the work order number must be 100%, the consistency rate between the retest window and the value source must be no less than 99%, and the direct closure mark writing rate must be 100%.

[0138] Alternatively, the work order policy can be stored on the station-side operation and maintenance management server or the regional operation and maintenance center server. As long as the work order policy can be generated based on the work order acceptance status, and the work order policy includes the accepted work order number, retest window, prohibition of direct closure flag, and necessary feedback fields, and subsequent steps can read the work order policy according to the accepted work order number and the equipment abnormality evidence piece number, it belongs to the same implementation method.

[0139] S5. The operation and maintenance management server receives the on-site execution receipt, matches it against the necessary feedback fields in the work order policy, and if the match is successful, the corresponding work order is transferred to the pending retest status. The specific implementation is as follows:

[0140] After a work order policy is formed, the operation and maintenance management server receives on-site execution receipts. The receiving action is triggered by the station-side operation and maintenance management server according to the accepted work order number, equipment anomaly evidence piece number, and policy version number. On-site sources include receipt records generated by the operation and maintenance personnel's handheld terminal, the station-side work order terminal, and the work order page of the regional operation and maintenance center. The effective period is from when the on-site personnel complete the on-site handling of the corresponding work order until the work order enters the waiting-for-retest state.

[0141] The on-site execution receipt is a record of on-site handling completion submitted by on-site personnel for the assigned work order number. It includes at least the assigned work order number, equipment anomaly evidence piece number, strategy version number, on-site execution completion time, executor account, job identifier, handling object, handling action, on-site confirmation result, on-site voucher identifier, and receipt submission time. The assigned work order number, equipment anomaly evidence piece number, and strategy version number must be consistent with the work order strategy. The on-site execution completion time uses the unified time base of the station and must not be earlier than the work order strategy formation time or later than the receipt submission time. The executor account and job identifier come from the work order system login record. The handling object comes from the equipment number in the equipment ledger or the name of the component under that equipment number. The handling action comes from the action name recorded in the current strategy version of the work order strategy. The on-site confirmation result is fixed as "on-site handling completed" when the match is successful. The on-site voucher identifier corresponds to the photo number, barcode number, or attachment number already saved in the attachment library.

[0142] After receiving the on-site execution receipt, the server first reads the corresponding work order policy according to the work order number, and then confirms that the receipt belongs to the current work order policy according to the equipment anomaly evidence piece number and policy version number. If the work order number is missing, the equipment anomaly evidence piece number is missing, the policy version number is missing, or the three cannot match the same work order policy at the same time, the on-site execution receipt will not be entered into field matching and written to the pending verification record.

[0143] The necessary feedback fields in the work order strategy are the field names and values ​​that are already recorded in the current strategy version and must be filled in on-site execution receipts. They must include at least the on-site execution completion time, the object to be handled, the handling action, the on-site confirmation result, and the on-site voucher identifier. Necessary feedback fields are read along with the strategy version and cannot be added or deleted by on-site personnel when submitting receipts. For example, the object to be handled for low string current is the inverter's sub-string or terminal block, the handling action is one of wiring tightening, fuse replacement, or branch circuit re-inspection, and the on-site voucher identifier is the photo number or equipment barcode number.

[0144] When matching the necessary feedback fields in the work order policy with the on-site execution receipt, the server confirms each item in the order of on-site execution completion time, disposal object, disposal action, on-site confirmation result, and on-site credential identifier. If the on-site execution completion time is missing, the receipt submission time cannot be used as a substitute. The disposal object must be consistent with the device number associated with the work order policy or the component under that device number. The disposal action must fall within the range of action names recorded in the current policy version. The on-site confirmation result must be that the on-site disposal has been completed. The on-site credential identifier must be able to be associated with the on-site photos, device scanning records, or on-site attachments already saved in the attachment library.

[0145] If the on-site execution receipt lacks necessary feedback fields, the object of disposal is inconsistent with the equipment ledger, the disposal action is not within the scope of the current policy version, the on-site confirmation result is not that the on-site disposal has been completed, or the on-site voucher identifier cannot be traced, no matching result will be generated. The receipt will be written to the pending record, and the corresponding work order will remain in its original state. If the on-site confirmation result is that the on-site disposal cannot be completed or that supplementary materials are required, only the receipt record will be saved, and the corresponding work order will not be transferred to the pending retesting state.

[0146] When a field execution receipt is submitted repeatedly, the criteria for determining duplicates are the work order number, equipment anomaly evidence piece number, strategy version number, field execution completion time, and field voucher identifier. Duplicate records will be returned to the existing matching results, while duplicates will be placed in a pending retest status. Receipts with different equipment anomaly evidence piece numbers under the same work order number will be matched according to the order of receipt submission. When a single receipt is pending verification or supplementary entry, the matching of other receipts will not be interrupted.

[0147] After a successful match, the server reads the "prohibit direct closure" flag from the work order policy, confirms its value as "prohibit direct closure," and then reads the current work order status corresponding to the accepted work order number. If the current work order status is "cancelled," "revoked," or "closed," the status is not transferred and written to the "pending verification" record. If the current work order status is "pending execution" or "in execution," the corresponding work order is transferred to the "pending retest" status. If the current work order status is already "pending retest," the existing "pending retest" status record is returned, and a new "pending retest" status is not written again.

[0148] The corresponding work order is the work order number pointed to by the on-site execution receipt. The status pending retest is the status value in the work order status change record. The status formation time is the on-site execution completion time, and the status accounting time is the server write time. The retest window follows the retest window in the work order strategy. This step only completes the status transition, does not determine whether the retest is qualified, and does not generate a closed status.

[0149] When the system enters the "pending retest" state, the server writes a status change record. The minimum record content includes the work order number, equipment anomaly evidence piece number, policy version number, on-site execution receipt number, "pending retest" status, status formation time, status recording time, retest window, "prohibit direct closure" flag, and system account. Old work order status records are retained as historical records and are not overwritten.

[0150] Information such as the names, mobile phone numbers, and location trajectories of on-site personnel will not be recorded in the pending retest status record; only the personnel account and job identifier will be retained. On-site voucher documents will be stored in the attachment library, and the pending retest status record will only retain the on-site voucher identifier.

[0151] After the accounting is completed, the next step is to read the status to be retested according to the work order number and the equipment abnormality evidence piece number, and read the retest window as the time period for reading equipment operation alarm data during the retesting stage.

[0152] Preferably, in a photovoltaic power station, the work order number is W20260706001, the equipment abnormality evidence piece number is INV-03-02-string current low-V3-10:00, the strategy version number is version 2, the retest window in the work order strategy is 2 hours, the prohibition of direct closure is marked as prohibition of direct closure, the on-site personnel submit the on-site execution receipt at 10:58, the object of the disposal is the 5th string terminal under INV-03-02, the disposal action is recorded as wiring tightening, the on-site confirmation result is recorded as on-site disposal completed, the on-site certificate identification corresponds to 2 photos and 1 scan record; after the server matches each item and passes, W20260706001 is transferred to the waiting retest state at 10:58, and the retest window is recorded as 2 hours.

[0153] No fewer than 20 on-site execution receipts should be randomly checked for each batch, the ratio of work order number matching the work order strategy should be 100%, the completeness rate of necessary feedback fields should be no less than 99%, the reading rate of the prohibition of direct closure mark should be 100%, and the traceability rate of the formation time of the pending retest status should be 100%.

[0154] Alternatively, the on-site execution receipt can be submitted by a handheld terminal or a station-end work order terminal. As long as the work order policy can be read according to the accepted work order number, equipment abnormality evidence piece number and policy version number, the on-site execution receipt can be matched with the necessary feedback fields in the work order policy, and the corresponding work order can be transferred to the pending retest status after the match is successful, they all belong to the same implementation method.

[0155] S6. The operation and maintenance management server reads the processed equipment operation alarm data within the retest window, and sets the work order to one of the following states based on the recurrence of similar alarms: closed, pending review, or secondary dispatch. This information is then written back to the work order status boundary ledger to form the work order status boundary version for the next cycle's work order creation judgment. Specifically, the implementation is as follows:

[0156] After the corresponding work order is transferred to the retesting state, the operation and maintenance management server reads the processed equipment operation alarm data in the retesting window. The reading action is triggered by the station-side operation and maintenance management server according to the accepted work order number, equipment abnormal evidence piece number and policy version number. The on-site sources are power plant monitoring alarm records, equipment operation status records, communication status records, limited transmission records and maintenance plan records. The effective period is from the on-site execution completion time to the end of the retesting window.

[0157] The retest window adopts the retest window already written in the work order strategy. The starting point is fixed as the on-site execution completion time in the on-site execution receipt, and the end point of the retest window is the moment when the cumulative observable duration reaches the retest window duration recorded in the work order strategy. The observable duration only counts the time period when the equipment is not shut down, not under maintenance, and has effective communication and can read the same type of alarm. The time periods when the equipment is shut down, under maintenance, or has ineffective communication are not included in the observable duration. The start time, end time, and source record identifier of the time period are written into the retest reading record.

[0158] If the duration of continuous unobservable events from the time the on-site execution is completed reaches the retest window duration recorded in the work order strategy, then the reading will no longer be extended, and the corresponding work order will be set to the pending review status.

[0159] The processed equipment operation alarm data consists of station-side monitoring alarm records corresponding to the same equipment number after the retest status is formed. It includes at least the equipment number, alarm code, alarm name, alarm start time, alarm recovery time, alarm level, and source record identifier. The server reads the equipment anomaly boundary rule table according to the equipment number, anomaly type, and valid rule version in the equipment anomaly evidence piece, maps the alarm code in the retest window to the anomaly type, and retains only alarm records with the same equipment number, the same anomaly type, and the same valid rule version as the same type of alarm.

[0160] When reading data, the time of the operation and maintenance management server is used as the reference. The alarm start time, alarm recovery time, retest window start time, and retest window end time are converted to the same time axis. If the alarm start time is earlier than the retest window start time and the alarm recovery time is later than the retest window start time, only the overlapping part within the retest window is extracted. If the alarm start time is equal to or later than the retest window end time, it is not included in this reading.

[0161] Device number, alarm code, and alarm start time must not be automatically filled in. If any are missing, the alarm record will be written to the pending verification record. If the alarm recovery time is missing and the alarm remains unrecovered within the retest window, the end of the retest window will be used as the provisional end time for this reading, and the unrecovered mark will be retained.

[0162] The server sets the work order to closed, pending review, or re-dispatch status based on the reproduction of similar alarms; similar alarm reproduction is confirmed according to the same device number, the same anomaly type, and the same valid rule version, and not based on the similarity of alarm names.

[0163] If no similar alarms are read within the cumulative observable duration of the retest window, and there are no pending verification records, unrecovered markers, or valid rule versions that cannot be matched, the work order will be set to the closed state. The time when the closed state is formed is taken as the end of the retest window. Downtime, maintenance, and communication invalid periods that have been removed from the observable duration will not be used as a basis for preventing the formation of the closed state.

[0164] If a similar alarm occurs within the cumulative observable duration of the retest window, but each continuous duration does not reach the alarm duration window in the equipment anomaly boundary rule table, or if there are pending verification records, unrecovered markers, or valid rule versions that cannot be matched during the retest reading process, the work order will be set to pending review status. The time for the pending review status to be formed is the end of the retest window. If the continuous unobservable duration from the time of completion of on-site execution reaches the retest window duration recorded in the work order strategy, the work order will also be set to pending review status.

[0165] If the same type of alarm is continuously maintained within the retest window and reaches the alarm duration window in the equipment anomaly boundary rule table, or if the same type of alarm forms equipment anomaly evidence again within the retest window, the work order will be set to secondary dispatch status. The time of formation of the secondary dispatch status is the moment when the same type of alarm is continuously maintained to reach the alarm duration window.

[0166] Only one of the three states—closed, pending review, and secondary dispatch—is allowed. The server determines the final state in the order of secondary dispatch, pending review, and closed. Once the secondary dispatch state has been determined, it will not be set to pending review or closed.

[0167] When a status is recorded, the server writes the work order number, the device anomaly evidence piece number, the policy version number, the retest window, the start point of the retest window, the end point of the retest window, the reproduction status of similar alarms, the final status, the status formation time, the status recording time, and the source record identifier; old status records are retained as historical records and are not overwritten.

[0168] The server then writes the final status back to the work order status boundary ledger, forming the work order status boundary version for the next cycle of work order creation judgment; this work order status boundary version includes at least the work order number, device number, exception type, valid rule version, boundary start time, boundary end time, current work order status, retest window, closed cooling window, status update time, and ledger version number.

[0169] When writing back, the original boundary start time is used; when the final state is closed, the boundary deadline is the end point of the retest window, the current work order status is written as closed, the cooling window is written according to the cooling window in the equipment abnormality boundary rule table, and the start point of the cooling window is the state formation time; when the final state is pending review, the boundary deadline is the end point of the retest window, the current work order status is written as pending review, and the cooling window remains inactive; when the final state is secondary dispatch, the boundary deadline is the time when the same type of alarm is continuously maintained to reach the alarm duration window, the current work order status is written as secondary dispatch, and the cooling window remains inactive.

[0170] The ledger version number is incremented by 1 according to the maximum sequential number of the ledger entries already recorded under the same work order number. When the same work order number, equipment anomaly evidence piece number, strategy version number, retest window start point, and retest window end point are repeatedly written back, the existing work order status boundary version is returned, and the increment is not repeated.

[0171] When reading alarms with the current device number and anomaly type in the next cycle, first read the work order status boundary version, and then perform the work order creation judgment in the next cycle.

[0172] Information such as the names, mobile phone numbers, and location trajectories of on-site personnel will not be written into the work order status boundary ledger; only the system account and source record identifier will be retained.

[0173] Preferably, in a photovoltaic power station, the work order number is W20260706001, the equipment anomaly evidence number is INV-03-02-string current low-V3-10:00, the on-site execution completion time is 10:58, and the retest window is 2 hours. The server reads the equipment operation alarm data within the cumulative 2-hour observable time after 10:58. If no similar low string current alarm occurs during this period, the work order is set to closed and the 24-hour closed cooling window is written back. If low string current occurs again between 11:40 and 11:55 and reaches the 10-minute alarm duration window, the work order is set to secondary dispatch status at 11:50. If a similar alarm occurs between 11:20 and 11:25 but does not reach 10 minutes, the work order is set to pending review status. If the continuous unobservable time from 10:58 reaches 2 hours, the work order is also set to pending review status.

[0174] No fewer than 20 status records are randomly checked for each batch of retests, the traceability rate of similar alarm reproduction is 100%, the uniqueness rate of the final status is 100%, and the success rate of writing back the work order status boundary version is no less than 99%.

[0175] Alternatively, the processed equipment operation alarm data can be pushed by the station monitoring server or actively read by the operation and maintenance management server at the end of the retest window. As long as the processed equipment operation alarm data can be read in the retest window, the work order can be set to closed, pending review, or secondary dispatch status according to the reproduction of similar alarms, and written back to the work order status boundary ledger to form the work order status boundary version for the next cycle of work order creation judgment, all of these belong to the same implementation method.

[0176] In the operating scenario shown in this embodiment: the station-side operation and maintenance management server of a photovoltaic power station performs closed-loop management of operation and maintenance work orders for the low string current anomaly of inverter INV-03-02.

[0177] When creating the rule, the server reads INV-03-02 from the equipment ledger as the equipment number, reads the alarm code and alarm name corresponding to low string current from the alarm dictionary, determines the anomaly type as low string current, and limits the effective operating conditions to grid-connected power generation, non-limited power generation, and effective communication. Based on the manufacturer's maintenance manual and historical work order records, the alarm duration window is set to 10 minutes, the merging time interval is set to 30 minutes, the retest window is set to 2 hours, and the cooling shutdown window is set to 24 hours, forming the effective rule version V3 and writing it into the equipment anomaly boundary rule table.

[0178] During operation, the station-side operation and maintenance management server reads equipment operation alarm data and equipment status data at a 5-minute trigger interval. INV-03-02 issued a low string current alarm at 10:00 and continued until 10:12. During this period, the operating status was grid-connected power generation, non-limited power generation, and effective communication. The server kept the original alarm start time of 10:00 unchanged in adjacent collection cycles and confirmed at 10:10 that the alarm of the same type had reached the 10-minute alarm duration window, forming an equipment anomaly evidence piece. The evidence start time was 10:00 and the evidence end time was 10:12. The same equipment issued another low string current alarm at 10:35 and continued until 10:48. The interval between the end of the previous valid alarm interval 10:12 and the start of the next valid alarm interval 10:35 was 23 minutes, which did not exceed the 30-minute merging time interval. The server classified the two valid alarm intervals into the same equipment anomaly evidence piece. The equipment anomaly evidence piece number was generated according to the equipment number, anomaly type, valid rule version, and evidence start time.

[0179] Subsequently, the server reads the work order status boundary ledger and finds that the existing work order W20260706001 corresponds to device number INV-03-02, with the anomaly type being low string current and the valid rule version being V3. The boundary start time is 09:40, the boundary end time is 09:55, and the current work order status is "in execution". Since the evidence start time of the device anomaly evidence piece (10:00) is 5 minutes away from the boundary end time (09:55), which does not exceed the merging time interval, the server determines that the work order acceptance status of the device anomaly evidence piece is "boundary continuation acceptance" and writes the matched work order number W20260706001 into the work order acceptance status record.

[0180] Afterwards, the server generates a work order policy based on the work order acceptance status, writes the accepted work order number as W20260706001, writes the retest window as 2 hours, writes the prohibition of direct closure flag as prohibition of direct closure, and writes the necessary feedback fields, including the on-site execution completion time, the object to be handled, the handling action, the on-site confirmation result, and the on-site voucher identifier. At the same time, a second version of the work order policy is formed under this accepted work order number.

[0181] After arriving at the INV-03-02 handling point, the on-site maintenance personnel tightened the wiring of the 5th series terminal block and submitted an on-site execution receipt via handheld terminal at 10:58. The on-site execution receipt included the following information: work order number W20260706001, equipment abnormality evidence piece number INV-03-02-string current low-V3-10:00, strategy version number 2, on-site execution completion time 10:58, handling object INV-03-02 subordinate 5th series terminal block, handling action wiring tightening, on-site confirmation result on-site handling completed, on-site certificate identification corresponding to 2 photos and 1 equipment barcode scan record.

[0182] After receiving the on-site execution receipt, the server reads the work order policy according to the work order number, equipment anomaly evidence piece number, and policy version number, and matches the on-site execution completion time, disposal object, disposal action, on-site confirmation result, and on-site certificate identifier item by item. After the match is successful, the server reads the prohibition on direct closure flag, confirms that its value is prohibition on direct closure, and then transfers W20260706001 to the pending retest status. The time of status formation is recorded as 10:58.

[0183] After entering the retesting phase, the server reads and processes the equipment operation alarm data for a cumulative 2-hour observable period starting at 10:58. During this period, INV-03-02 remains non-shutdown, non-maintenance, and communication is effective. Within the cumulative observable period, the server does not read any alarms of the same type as INV-03-02, with the same abnormality type of low string current and valid rule version V3. There are also no pending verification records, unrecovered markers, or situations where the valid rule version cannot be matched. Therefore, at the end of the retesting window, W20260706001 is set to the closed state, and the final state is written back to the work order status boundary ledger, forming a new work order status boundary version. In this version, the current work order status is written as closed, the boundary deadline is taken as the end of the retesting window, and the cooling window is closed according to the equipment abnormality boundary rule table, which is written for 24 hours.

[0184] When the server reads the low string current alarm from INV-03-02 again in the next cycle, it first reads the status boundary version of the work order. If the new alarm falls into the cooling shutdown window, it continues to judge according to the existing work order status boundary and does not directly generate a duplicate new work order.

[0185] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0186] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0187] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0188] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0189] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0190] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0191] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0192] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0193] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0194] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A closed-loop method for operation and maintenance work orders of new energy power plant equipment, characterized in that, include: S1. The operation and maintenance management server establishes a device anomaly boundary rule table, recording the device number, anomaly type, valid rule version, valid operating conditions, alarm duration window, merging time interval, retest window, and cooling shutdown window; S2. The operation and maintenance management server collects the equipment operation alarm data and equipment status data of the current period, merges the same type of anomalies of the same equipment according to the equipment anomaly boundary rule table, and generates equipment anomaly evidence pieces. S3. The operation and maintenance management server reads the work order status boundary ledger, matches the equipment anomaly evidence pieces with the existing work order status boundary, and determines the work order acceptance status. S4. The operation and maintenance management server generates a work order policy based on the work order acceptance status. The work order policy includes the accepted work order number, the retest window, the prohibition of direct closure flag, and necessary feedback fields. S5. The operation and maintenance management server receives the field execution receipt, matches the field execution receipt with the necessary feedback fields in the work order policy, and after the match is successful, the corresponding work order is transferred to the pending retest status. S6. The operation and maintenance management server reads the processed equipment operation alarm data in the retest window, sets the work order to one of the following states according to the reproduction of similar alarms: closed, pending review, or secondary dispatch, and writes it back to the work order status boundary ledger to form the work order status boundary version for the next cycle of work order creation judgment.

2. The closed-loop method for operation and maintenance work orders of new energy power plant equipment according to claim 1, characterized in that, S1 includes: The operation and maintenance management server converts the time field to a unified time base for the site. Organize the equipment anomaly boundary rule table in the following order: equipment number, anomaly type, effective operating condition, alarm duration window, merging time interval, retest window, and cooling shutdown window; Each rule record generates a valid rule version and effective time, and is recorded using the same device number, the same exception type, and the valid rule version as unique keys.

3. The closed-loop method for operation and maintenance work orders of new energy power plant equipment according to claim 1, characterized in that, S2 include: The operation and maintenance management server uses a left-closed, right-open approach to process the collected equipment operation alarm data and equipment status data. Convert the alarm start time and alarm recovery time to a unified time base at the station. For continuous alarms across cycles, retain the original alarm start time and read the valid rule version in the device anomaly boundary rule table according to the alarm start time to determine the anomaly type.

4. The closed-loop method for operation and maintenance work orders of new energy power plant equipment according to claim 3, characterized in that, S2 also includes: The operation and maintenance management server uses the same device number, the same anomaly type, and the same valid rule version as the merging criteria; Confirm the valid alarm range based on the alarm duration window; Adjacent valid alarm intervals are merged based on the merging time interval to form equipment anomaly evidence pieces, and equipment anomaly evidence piece numbers are generated according to equipment number, anomaly type, valid rule version, and evidence start time.

5. A closed-loop method for operation and maintenance work orders of new energy power plant equipment according to claim 4, characterized in that, S3 include: The operation and maintenance management server reads the work order status boundary ledger according to the device anomaly evidence piece number; Filter existing work order status boundaries by equipment number and anomaly type; Based on the current work order status, and by comparing the evidence start time and evidence end time with the boundary start time, boundary end time, retest window, closing cooldown window, and status update time, matching results are generated for boundary acceptance, boundary continuation acceptance, acceptance pending retest, cooling-down acceptance, secondary dispatch acceptance, and newly created acceptance. When two or more matching results are generated for the same equipment anomaly evidence, a unique work order acceptance status is determined in the following order: acceptance pending retest, acceptance for secondary dispatch, acceptance within the boundary, acceptance for boundary continuation, acceptance for cooling, and acceptance for new creation.

6. A closed-loop method for operation and maintenance work orders of new energy power plant equipment according to claim 5, characterized in that, S4 include: The operation and maintenance management server determines the work order number based on the work order acceptance status. Read the retest window corresponding to the equipment number, anomaly type and valid rule version in the equipment anomaly boundary rule table, and use the retest window if the existing work order status boundary to be retested has a retest window recorded. The policy will include a prohibition on direct closure flag, necessary feedback fields, and policy version number in the work order policy. The necessary feedback fields include on-site execution completion time, disposal object, disposal action, on-site confirmation result, and on-site credential identifier. The equipment malfunction evidence piece number is used as the sole submission criterion for accounting purposes.

7. A closed-loop method for operation and maintenance work orders of new energy power plant equipment according to claim 6, characterized in that, S5 include: The operation and maintenance management server reads the work order policy according to the work order number, the equipment abnormality evidence piece number, and the policy version number; Match the on-site execution receipt with the necessary feedback fields in the work order strategy in the following order: on-site execution completion time, disposal object, disposal action, on-site confirmation result, and on-site voucher identifier. Once the match is successful, it will be confirmed that the flag cannot be closed directly, and the accepted work orders whose current status is either pending execution or in execution will be transferred to the pending retest status.

8. A closed-loop method for operation and maintenance work orders of new energy power plant equipment according to claim 7, characterized in that, S6 include: The operation and maintenance management server reads the retest window in the work order policy and uses the on-site execution completion time in the on-site execution receipt as the starting point of the retest window. Accumulate the observable duration during periods when there is no downtime, no maintenance, effective communication, and similar alarms can be read; Read the processed equipment operation alarm data according to the equipment number, anomaly type and valid rule version that are consistent with the equipment anomaly evidence film, and form a similar alarm reproduction situation.

9. A closed-loop method for operation and maintenance work orders of new energy power plant equipment according to claim 8, characterized in that, S6 also includes: The operation and maintenance management server determines one of the following states based on the reproduction of similar alarms: shutdown, pending review, or secondary dispatch. The final status is determined in the order of secondary dispatch status, pending review status, and closed status; The final state is written back to the work order state boundary ledger to form the work order state boundary version for the next cycle's work order creation judgment.

Images