A method and system for early warning and work order scheduling of abnormal operation of a generator set

By performing multi-level threshold comparisons and state transitions on generator set operating parameters, early warning levels and scheduling task contexts are generated. Maintenance work orders are then filtered and pushed, solving the problems of information dispersion and state asynchrony when generator sets are operating abnormally, and achieving synchronization and accuracy of work order scheduling.

CN123070931APending Publication Date: 2026-11-10BEIJING YUNZHI COMMUNICATION TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611285445.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-24
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

When the operating parameters of existing generator sets are continuously abnormal, there is a lack of a unified data receiving object between alarm records and maintenance work orders. This results in information such as warning level, generator set location information, maintenance personnel location update time, and personnel status being scattered, which can easily lead to problems such as disconnection between alarm information and work order scheduling information, and inconsistency between personnel status and work order dispatch results.

Method used

By receiving generator set operating parameters, performing multi-level threshold comparisons to generate abnormal event records, and converting them into early warning levels and scheduling trigger states when the duration of the abnormality reaches a preset time, a scheduling task context containing generator set location information and required skill tags is generated, maintenance personnel are screened, maintenance work orders are generated and pushed, and the work order status is written back based on the status returned by the mobile terminal.

Benefits of technology

When generator set operating parameters are continuously abnormal, the work order scheduling and early warning event status are synchronized, ensuring timely updates of maintenance work orders, generator set monitoring status, and personnel status. This solves the problems of information dispersion and status asynchrony, and improves the accuracy and efficiency of maintenance work orders.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123070931A_ABST
    Figure CN123070931A_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of remote operation and maintenance management of generator sets, and particularly relates to a generator set operation abnormality early warning and work order scheduling method and system. The method comprises the following steps: receiving equipment identification, sampling time, and operation parameters such as water temperature, oil pressure and rotating speed, performing multi-level threshold value comparison according to equipment archives and threshold value rule versions, and generating abnormal event records; generating early warning event records based on abnormal duration and determining scheduling trigger states; generating scheduling task context containing positioning information, service radius and required skill labels when triggering, and screening maintenance personnel to form a candidate personnel set; determining target maintenance personnel according to predicted arrival time and skill matching degree, generating maintenance work orders bound to early warning event records and pushing; and writing back maintenance work order states, unit monitoring states, personnel states and early warning event states according to work order states. The present application effectively improves the state consistency of early warning events and maintenance work order scheduling.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of remote operation and maintenance management technology for generator sets, and in particular to a method and system for early warning of abnormal operation of generator sets and work order scheduling. Background Technology

[0002] In the field of remote operation and maintenance management technology for generator sets, existing monitoring solutions for generator sets typically acquire operating parameters such as water temperature, oil pressure, speed, and battery voltage through data acquisition terminals or unit controllers, and then upload these parameters to a monitoring platform for threshold comparison, alarm display, or fault recording. Some operation and maintenance management systems are also equipped with fault tasks, maintenance information, tool and equipment availability prompts, and maintenance work order functions to record and dispatch occurred faults or alarms.

[0003] In conventional monitoring scenarios, such solutions can complete the collection of operating parameters, abnormal alarms, and work order recording. However, in scenarios where generator set operating parameters are continuously abnormal and require on-site maintenance, there is usually a lack of a unified data receiving object between alarm records and maintenance work orders. Information such as warning level, generator set location information, maintenance personnel location update time, personnel status, and skill tags are scattered in different records or systems, which can easily lead to problems such as disconnection between alarm information and work order scheduling information, and inconsistency between personnel status and work order dispatch results.

[0004] Existing solutions mostly rely on fixed threshold alarms, manual dispatching, or static scheduling for processing. When generator set operating parameters continuously exceed limits, maintenance personnel location changes, personnel status changes, or mobile terminal work order status changes, there is a lack of continuous status transmission between the monitoring link, personnel matching link, and work order status recording link, which can easily lead to asynchrony between maintenance work order status, generator set monitoring status, personnel status, and early warning event status.

[0005] For the joint processing of continuous abnormal generator operating parameters, early warning events, dynamic status of maintenance personnel, and work order status, existing technologies still suffer from common shortcomings, including discontinuous connections between scheduling triggering, personnel screening, work order binding, and status write-back. Therefore, it is necessary to address the problem of generating maintenance work orders and completing status write-back based on early warning event records and scheduling task context under the scenario of generator operating anomaly early warning and maintenance work order scheduling. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method for generator set operation anomaly early warning and work order scheduling, comprising:

[0007] S100: Receive device identifier, sampling time, and generator set operating parameters including water temperature, oil pressure, and speed; perform multi-level threshold comparison according to device file and threshold rule version; and generate anomaly event records including the direction of the boundary crossing and the duration of the anomaly.

[0008] S200. Determine the warning level based on the abnormal event record, generate a warning event record when the duration of the abnormality reaches a preset duration, and determine the scheduling trigger status according to the warning level and scheduling trigger conditions.

[0009] S300: When the scheduling trigger state is triggered, a scheduling task context containing generator set location information, service radius and required skill tags is generated, and maintenance personnel are filtered according to location update time, personnel status, service radius and skill tags to generate a candidate personnel set;

[0010] S400: Based on the expected arrival time and skill matching degree in the candidate personnel set, determine the target maintenance personnel, generate a maintenance work order bound to the warning event record, and push it to the mobile terminal of the target maintenance personnel;

[0011] S500: Receive the work order status returned by the mobile terminal, and write back the maintenance work order status, generator set monitoring status, personnel status, and early warning event status based on the work order status; when the order is rejected, the order is accepted for too long, or the personnel status fails, regenerate the candidate personnel set based on the scheduling task context.

[0012] Furthermore, the generator set operating parameters also include battery voltage; receiving device identification, sampling time, and generator set operating parameters includes: the acquisition terminal reading the generator set operating parameters from the generator set controller at a fixed frequency, and encapsulating the device identification, the sampling time, and the generator set operating parameters into an operating data packet.

[0013] Furthermore, the abnormal event record includes abnormal parameter type, cross-boundary direction, cross-boundary start time, number of consecutive cross-boundary occurrences, abnormal duration, and threshold rule version; when the generator set operating parameters recover to the recovery threshold range, the status of the abnormal event record is updated.

[0014] Furthermore, the warning levels include a first level, a second level, and a third level; the first level corresponds to alarm information, the second level and the third level correspond to a triggering status, and the personnel sorting priority of the third level is higher than that of the second level.

[0015] Furthermore, the scheduling task context also includes an early warning event identifier, equipment model, early warning type, tool and component template, and work order generation template; the tool and component template is read according to the equipment model and early warning type.

[0016] Furthermore, maintenance personnel are filtered based on location update time, personnel status, service radius, and skill tags, including: verifying the online status of maintenance personnel's mobile terminals; filtering maintenance personnel whose location update time is within the validity period of their personnel status, whose personnel status is idle, whose distance from the generator set's location information is less than the service radius, and whose skill tags match the required skill tags.

[0017] Further, the target maintenance personnel are determined based on the expected arrival time and skill matching degree in the candidate personnel set, including: generating the personnel ranking results according to the expected arrival time from small to large, skill matching degree from high to low, historical order confirmation time from small to large, and current task load from small to large, and determining the maintenance personnel ranked first in the personnel ranking results as the target maintenance personnel.

[0018] Furthermore, the maintenance work order includes a warning event identifier, an abnormal parameter snapshot, an abnormal duration, a warning level, generator set location information, a fault code, a suggested maintenance plan, and a list of tools and equipment; the list of tools and equipment is determined from the tools and equipment template based on the equipment model, warning type, and fault code.

[0019] Furthermore, when the work order status is "accepted", the personnel status is updated to "occupied" and the generator set monitoring status is updated to "pending maintenance"; when the work order status is "arrived", the generator set monitoring status is updated to "under maintenance"; when the work order status is "completed" and the generator set operating parameters have recovered to the recovery threshold range, the early warning event record is turned off.

[0020] Furthermore, a generator set operation anomaly early warning and work order scheduling system includes: a data acquisition module, an anomaly determination module, an early warning module, a scheduling matching module, a personnel determination module, a work order generation module, a mobile terminal push module, and a status write-back module; the system is used to implement the method described in any of the above-mentioned embodiments.

[0021] The key innovations of this invention include:

[0022] (1) The generator set operating parameters are compared with the equipment file and threshold rule version at multiple levels to form an abnormal event record containing the direction of the boundary and the duration of the abnormality. When the duration of the abnormality reaches the preset duration, it is converted into an early warning event record containing the early warning level and the dispatch trigger status.

[0023] (2) When the scheduling trigger status is triggered, the early warning event record is organized into a scheduling task context containing generator set location information, service radius and required skill tags, and maintenance personnel are screened based on location update time, personnel status, service radius and skill tags to form a candidate personnel set.

[0024] (3) Based on the expected arrival time and skill matching degree in the candidate personnel set, the target maintenance personnel are determined, a maintenance work order with a binding warning event record is generated, and the maintenance work order status, generator set monitoring status, personnel status and warning event status are written back according to the work order status returned by the mobile terminal.

[0025] The following are its main beneficial effects:

[0026] (1) To address the lack of a unified data receiving object between existing alarm records and maintenance work orders, abnormal event records and early warning event records are used to receive the direction of the boundary crossing, the duration of the abnormality, the early warning level and the scheduling trigger status, so that the continuous abnormal status can be used as a structured input for subsequent work order scheduling.

[0027] (2) To address the problem of dispersed early warning levels, generator set location information and maintenance personnel dynamic status, generator set location information, service radius and required skill tags are collected through scheduling task context, and candidate personnel are formed according to location update time, personnel status, service radius and skill tags, so that the personnel screening results are associated with the on-site maintenance conditions corresponding to the early warning event.

[0028] (3) To address the issue of asynchronous status of maintenance work order, generator set monitoring status, personnel status and early warning event status, the work order execution status can be returned to the generator set monitoring status, personnel status and early warning event status by binding the maintenance work order and mobile terminal work order status back to the early warning event record. When the work order is rejected, the work order is accepted for a timeout or the personnel status is invalid, the scheduling task context is used again to generate the candidate personnel set. Attached Figure Description

[0029] Figure 1 A flowchart illustrating a generator set operation anomaly early warning and work order scheduling method provided in this application embodiment;

[0030] Figure 2 This is a structural block diagram of a generator set operation anomaly early warning and work order scheduling system provided in an embodiment of this application. Detailed Implementation

[0031] Example 1: Refer to Figure 1 This is a flowchart illustrating a generator set operation anomaly early warning and work order scheduling method provided in an embodiment of the present invention. The process may include at least steps S100-S500:

[0032] S100: Receive device identifier, sampling time, and generator set operating parameters including water temperature, oil pressure, and speed; perform multi-level threshold comparison according to device file and threshold rule version; and generate anomaly event records including the direction of the boundary crossing and the duration of the anomaly.

[0033] S200. Determine the warning level based on the abnormal event record, generate a warning event record when the duration of the abnormality reaches a preset duration, and determine the scheduling trigger status according to the warning level and scheduling trigger conditions.

[0034] S300: When the scheduling trigger state is triggered, a scheduling task context containing generator set location information, service radius and required skill tags is generated, and maintenance personnel are filtered according to location update time, personnel status, service radius and skill tags to generate a candidate personnel set;

[0035] S400: Based on the expected arrival time and skill matching degree in the candidate personnel set, determine the target maintenance personnel, generate a maintenance work order bound to the warning event record, and push it to the mobile terminal of the target maintenance personnel;

[0036] S500: Receive the work order status returned by the mobile terminal, and write back the maintenance work order status, generator set monitoring status, personnel status, and early warning event status based on the work order status; when the order is rejected, the order is accepted for too long, or the personnel status fails, regenerate the candidate personnel set based on the scheduling task context.

[0037] S100 receives the device identifier, sampling time, and generator set operating parameters including water temperature, oil pressure, and speed. It performs multi-level threshold comparisons according to the device file and threshold rule version to generate anomaly event records containing the direction of the boundary crossing and the duration of the anomaly.

[0038] In this embodiment, S100 is jointly executed by the data acquisition module and the anomaly determination module. The data acquisition module is deployed on the data access side of the cloud monitoring platform and receives the operation data packets reported by the acquisition terminal. The acquisition terminal is set on the generator set controller side and reads the generator set operation parameters from the generator set controller at a fixed frequency. The operation data packet includes a device identifier, sampling time, and generator set operation parameters. The device identifier is an identification field that distinguishes different generator sets. The sampling time is a time field formed when the acquisition terminal reads the parameter values ​​in the generator set controller. The generator set operation parameters include water temperature, oil pressure, and speed. In one embodiment, the generator set operation parameters also include battery voltage, which is written into the operation data packet along with the water temperature, oil pressure, and speed.

[0039] Specifically, when the acquisition period arrives, the acquisition terminal reads the operating parameter register in the generator controller and binds the read water temperature, oil pressure, speed, and battery voltage with the equipment identifier and sampling time. The acquisition terminal encapsulates the bound fields to form an operating data packet and uploads it to the cloud monitoring platform. The fixed frequency can be configured according to the generator operating scenario, for example, 10 seconds or 30 seconds per acquisition period. In another embodiment, when the generator controller outputs a fault event or alarm event, the acquisition terminal adds an additional operating data packet report outside the current acquisition period. After the operating data packet arrives at the cloud monitoring platform, the data acquisition module establishes a reception record according to the equipment identifier and uses the sampling time as the time sequence basis for the generator set's operating parameters.

[0040] After receiving the operation data packet output by the data acquisition module, the anomaly determination module reads the equipment file and threshold rule version based on the equipment identifier. The equipment file includes the equipment model, rated power, generator controller model, generator set location information, operating status field, and threshold rule version field. The threshold rule version is a multi-level threshold configuration identifier corresponding to the equipment model and operating status. The multi-level thresholds include an upper limit threshold for water temperature, a lower limit threshold for oil pressure, an upper limit threshold for engine speed, and a lower limit threshold for engine speed. In one embodiment, the multi-level thresholds also include a lower limit threshold for battery voltage. The anomaly determination module reads the corresponding threshold configuration according to the threshold rule version, compares each parameter value in the generator set operating parameters with the corresponding multi-level threshold, and generates a parameter status field for each parameter.

[0041] When performing multi-level threshold comparisons, the anomaly detection module processes water temperature, oil pressure, and speed separately according to parameter type. For water temperature, when the water temperature value is greater than the upper limit threshold, the out-of-bounds direction is recorded as an upper out-of-bounds error; for oil pressure, when the oil pressure value is less than the lower limit threshold, the out-of-bounds direction is recorded as a lower out-of-bounds error; for speed, when the speed value is greater than the upper limit threshold, the out-of-bounds direction is recorded as an upper out-of-bounds error, and when the speed value is less than the lower limit threshold, the out-of-bounds direction is recorded as a lower out-of-bounds error. If the same device identifier and the same abnormal parameter type already have an unfinished out-of-bounds state in the previous sampling time, the anomaly detection module keeps the out-of-bounds start time unchanged and writes the time difference between the current sampling time and the out-of-bounds start time into the anomaly duration; if there is no unfinished out-of-bounds state in the previous sampling time, the current sampling time is written into the out-of-bounds start time, and the anomaly duration is accumulated from the current sampling time.

[0042] The abnormal event record is an output object generated by S100, containing device identifier, abnormal parameter type, out-of-bounds direction, out-of-bounds start time, consecutive out-of-bounds count, abnormal duration, sampling time, parameter value, and threshold rule version. The consecutive out-of-bounds count is accumulated based on the number of out-of-bounds occurrences of the same device identifier and the same abnormal parameter type within a continuous sampling period; the abnormal duration is calculated based on the out-of-bounds start time and the current sampling time. The abnormal event record may also contain a snapshot of the original parameters, which records the water temperature, oil pressure, speed, and battery voltage values ​​at the current sampling time. The abnormality determination module writes the abnormal event record into the abnormal event table and provides the abnormal event record to the early warning module in S200.

[0043] When a running data packet lacks a device identifier or sampling time, the data acquisition module marks the running data packet as missing a field and writes it to the access log; this running data packet is not included in the threshold comparison. When a running data packet lacks a parameter value, the anomaly determination module retains the missing parameter status and does not use the missing parameter to generate an anomaly event record; the remaining parameters in the same running data packet are still compared according to the corresponding multi-level thresholds. When the threshold rule version does not match the device file, the anomaly determination module calls the currently effective threshold rule version in the device file and writes the version inconsistency status to the processing record of the running data packet. When the acquisition terminal fails to report running data packets continuously, the data acquisition module generates a communication anomaly status; this communication anomaly status is not written to the anomaly event record as a water temperature, oil pressure, or speed exceeding the limit status.

[0044] In one implementation, when the generator set operating parameters recover to within the recovery threshold range, the anomaly determination module updates the status of the anomaly event record. The recovery threshold range corresponds to multiple threshold levels; for example, the water temperature recovery threshold is lower than the upper water temperature threshold, and the oil pressure recovery threshold is higher than the lower oil pressure threshold. When the parameter value enters the recovery threshold range and remains within it for one sampling cycle, the anomaly determination module marks the anomaly event record status as recovered and records the recovery sampling time. This recovered status serves as a reference field for subsequent steps (S200) to determine whether to generate a warning event record.

[0045] In the on-site implementation scenario, generator set G001 is in operation. At the end of the 10-second sampling cycle, the data acquisition terminal reads water temperature, oil pressure, speed, and battery voltage from the generator set controller. The terminal encapsulates G001, the current sampling time, 92℃ water temperature, 0.18MPa oil pressure, 1500r / min speed, and 24V battery voltage into a data package. The cloud monitoring platform reads the device file and threshold rule version corresponding to G001, comparing the water temperature with multi-level water temperature thresholds, the oil pressure with multi-level oil pressure thresholds, and the speed with multi-level speed thresholds. If the water temperature exceeds the corresponding threshold and the previous sampling cycle was already in a water temperature out-of-bounds state, the anomaly determination module retains the out-of-bounds start time recorded in the previous sampling cycle and accumulates the anomaly duration, forming an anomaly event record for water temperature out-of-bounds. The device identifier, anomaly parameter type, out-of-bounds direction, and anomaly duration in this anomaly event record are entered into S200 for further processing by the early warning module.

[0046] S200. Determine the warning level based on the abnormal event record, generate a warning event record when the duration of the abnormality reaches a preset duration, and determine the scheduling trigger state according to the warning level and scheduling trigger conditions.

[0047] In this embodiment, S200 is executed by the early warning module. The early warning module receives the abnormal event record generated in S100 and uses the abnormal event record as input for calculating the early warning level and generating the early warning event record. The abnormal event record includes device identifier, abnormal parameter type, out-of-bounds direction, out-of-bounds start time, number of consecutive out-of-bounds times, abnormal duration, sampling time, parameter value, and threshold rule version; wherein, the abnormal duration is used to distinguish between short-term fluctuation state and continuous abnormal state, and the threshold rule version is used to read the early warning level rule corresponding to the current device file.

[0048] The early warning module generates an early warning level based on the abnormal parameter type, out-of-bounds direction, abnormal duration, and the threshold range in which the parameter value falls. The early warning levels include a first level, a second level, and a third level; the first level corresponds to alarm information, while the second and third levels correspond to a triggered scheduling status. The early warning module reads the level rules configured in the threshold rule version and matches the parameter values ​​in the abnormal event records with the threshold ranges for each level. When the parameter value is within the first level threshold range and the abnormal duration reaches the first duration configuration, a first-level field is generated; when the parameter value is within the second level threshold range and the abnormal duration reaches the second duration configuration, a second-level field is generated; and when the parameter value is within the third level threshold range and the abnormal duration reaches the third duration configuration, a third-level field is generated. Different abnormal parameter types can correspond to different duration configurations; for example, water temperature and oil pressure correspond to different duration configurations.

[0049] If the duration of the anomaly does not reach the preset duration, the early warning module marks the anomaly event record as a short-term fluctuation state and saves the current anomaly duration and parameter values; no early warning event record is generated for this short-term fluctuation state. The preset duration is configured by the threshold rule version and corresponds to the anomaly parameter type and early warning level. After the anomaly duration reaches the preset duration, the early warning module reads the device identifier, anomaly parameter type, out-of-bounds direction, parameter value, sampling time, and threshold rule version from the anomaly event record and generates an early warning event record. The early warning event record includes the early warning event identifier, device identifier, anomaly parameter type, parameter snapshot, anomaly duration, early warning level, threshold rule version, and scheduling trigger status; the parameter snapshot includes the water temperature, oil pressure, speed, and battery voltage values ​​at the time the early warning event record is generated.

[0050] The early warning module generates a scheduling trigger status based on the early warning level and scheduling trigger conditions. The scheduling trigger conditions are configured by the early warning level rules; in one implementation, the first level corresponds to an alarm message, and the early warning module records the scheduling trigger status as not triggered; the second and third levels correspond to a triggered scheduling trigger status. For third-level early warning event records, the early warning module writes a higher personnel ranking priority into the early warning event record; this personnel ranking priority is invoked when S400 generates the personnel ranking results. For first-level early warning event records, the early warning module generates an alarm message and records it in the early warning event record; this early warning event record does not enter the candidate personnel screening process unless the early warning event is subsequently updated to the second or third level due to an increase in the abnormal duration or the parameter value entering a higher-level threshold range.

[0051] When processing abnormal event records, the early warning module also checks the status field of the abnormal event record. If S100 has already updated the abnormal event record to a recovered state and the abnormal duration has not reached the preset duration, the early warning module archives the abnormal event record as a recovered short-term abnormality. If an abnormal event record has been generated as an early warning event record and then recovered to within the recovery threshold range, the early warning module writes a parameter recovery flag into the corresponding early warning event record. This parameter recovery flag is used by S500 when the processing is completed. If there is already an unclosed early warning event record for the same device identifier and the same abnormal parameter type, the early warning module does not generate a new early warning event record. Instead, it updates the abnormal duration, parameter snapshot, and early warning level in the original early warning event record.

[0052] In one implementation, the early warning module maintains rules for generating early warning event identifiers. These identifiers are generated by combining a device identifier, an abnormal parameter type, and an out-of-bounds start time. Only one early warning event record corresponds to the same device identifier, the same abnormal parameter type, and the same out-of-bounds start time. After an early warning event record is generated, the early warning module writes it into the early warning event table. When the scheduling trigger status in the early warning event table is triggered, the early warning event record is sent to the scheduling matching module in S300. If the device file in the early warning event record is missing, the early warning module retains the early warning event record but marks the scheduling trigger status as requiring a supplementary device file. Once the device file is complete, the scheduling trigger status is regenerated according to the early warning level and scheduling trigger conditions.

[0053] In the field implementation scenario, the water temperature anomaly event log for generator set G001 shows that the water temperature is within the second-level threshold range, exceeding the limit upwards, and the anomaly duration reaches 60 seconds. The early warning module reads the threshold rule version of G001, matches it with the second-level rule for water temperature exceeding the limit, and generates a second-level early warning level. Since the second-level level corresponds to a triggered scheduling status, the early warning module generates an early warning event record, writing the early warning event identifier, G001, the anomaly parameter type as water temperature, parameter snapshot, 60-second anomaly duration, second level, and scheduling trigger status into the record. This early warning event record is sent to S300, and the scheduling matching module in S300 generates a scheduling task context based on this.

[0054] S300: When the scheduling trigger state is triggered, a scheduling task context containing generator set location information, service radius and required skill tags is generated, and maintenance personnel are filtered according to location update time, personnel status, service radius and skill tags to generate a candidate personnel set.

[0055] In this embodiment, S300 is executed by the scheduling matching module. The scheduling matching module receives the early warning event record output by S200 and reads the scheduling trigger status in the early warning event record; when the scheduling trigger status is triggered, the scheduling matching module generates a scheduling task context with the early warning event record as input. The scheduling task context is a data record composed of the early warning event record, generator set positioning information, equipment file, early warning type, service radius, required skill tags, tool and equipment template, and work order generation template; this data record serves as the input for candidate screening in this step and as the input for maintenance work order generation in S400.

[0056] When generating a scheduling task context, the scheduling matching module first reads the device identifier and abnormal parameter type from the warning event record; then, based on the device identifier, it queries the device file to read the generator set location information, device model, and device operating status fields. The generator set location information includes a longitude field, a latitude field, and a location update time. In one implementation, the generator set location information comes from the fixed installation location in the device file; in another implementation, the generator set location information comes from the location record most recently reported by the generator set positioning terminal. If the location update time exceeds the device location validity period, the scheduling matching module marks the scheduling task context as a location pending confirmation state and sends the warning event record to the manual review queue; the location pending confirmation state does not enter the candidate set generation.

[0057] The scheduling and matching module also reads the warning type based on the abnormal parameter type and warning level in the warning event record. For example, water temperature exceeding the limit corresponds to a high water temperature warning type, oil pressure exceeding the limit corresponds to a low oil pressure warning type, speed exceeding the limit corresponds to an overspeed warning type, and speed exceeding the limit corresponds to an underspeed warning type. The scheduling and matching module reads the required skill tag based on the warning type and reads the tool and component template corresponding to the equipment model and warning type; in one embodiment, the high water temperature warning type corresponds to the cooling system maintenance skill tag, the low oil pressure warning type corresponds to the lubrication system maintenance skill tag, and the overspeed or underspeed warning type corresponds to the speed control inspection skill tag. The service radius can be read according to the warning level, with the second level corresponding to the first service radius and the third level corresponding to the second service radius; the service radius of the third level is greater than that of the second level.

[0058] After generating the scheduling task context, the scheduling matching module reads personnel information from the maintenance personnel database. This personnel information includes personnel identifier, maintenance personnel location, location update time, personnel status, skill tags, mobile terminal online status, current task load, and historical order confirmation duration. The scheduling matching module filters maintenance personnel based on location update time, personnel status, service radius, and skill tags, generating a candidate personnel set. Specifically, the scheduling matching module first verifies the maintenance personnel's mobile terminal online status; if the mobile terminal is offline, the maintenance personnel is not included in the candidate personnel set. Subsequently, the scheduling matching module compares the time difference between the location update time and the current scheduling time; only maintenance personnel whose location update time falls within the personnel status validity period proceed to the next screening stage; the personnel status validity period is the time range within which the maintenance personnel's location and status are accepted by the scheduling matching module.

[0059] After location update time verification, the scheduling matching module reads the personnel status field. Maintenance personnel with an "idle" status enter the distance filtering process, while those with statuses of "occupied," "resting," "offline," or "pending confirmation" are not included in the candidate set. Distance filtering is based on the generator set location information and the maintenance personnel's location. The scheduling matching module calculates the spatial distance between the two and compares it with the service radius in the scheduling task context. Maintenance personnel whose spatial distance is less than the service radius enter the skill tag filtering process. Skill tag filtering matches the maintenance personnel's skill tags with the required skill tags in the scheduling task context. Maintenance personnel with matching skill tags are added to the candidate set; those with non-matching skill tags remain in the filtering record but are not selected as target maintenance personnel for the S400.

[0060] The candidate personnel set includes candidate personnel identifier, maintenance personnel location, location update time, personnel status, skill tags, distance from generator set location information, current task load, and historical order confirmation time. After the candidate personnel set is generated, the scheduling matching module binds the candidate personnel set to the scheduling task context, forming a candidate personnel set record. If the candidate personnel set is empty after filtering, the scheduling matching module can call the backup service radius according to the warning level; in one implementation, when the candidate personnel set is empty for a third-level warning event record, the scheduling matching module reads a larger service radius and re-filters the location. If it is still empty after re-filtering, the scheduling matching module updates the warning event record status to pending manual dispatch and writes the filtered record to the scheduling history.

[0061] In the field implementation scenario, the early warning event record of generator set G001 corresponds to the high water temperature early warning type, and the scheduling trigger status is triggered. The scheduling matching module reads the location information and equipment model from the G001 equipment file, reads the cooling system maintenance skill tag corresponding to the high water temperature early warning type, reads the 20km service radius and cooling system tool and component template, and generates a scheduling task context. The scheduling matching module then reads the personnel information of multiple maintenance personnel, first excluding maintenance personnel whose mobile terminals are offline, then excluding maintenance personnel whose location update time exceeds 5 minutes, and then excluding maintenance personnel whose personnel status is not idle; for the remaining maintenance personnel, the distance between them and the G001 location information is calculated, and maintenance personnel whose distance is less than 20km and whose skill tag matches the cooling system maintenance are retained, generating a candidate personnel set. This candidate personnel set and the scheduling task context are entered into S400 for personnel sorting and maintenance work order generation.

[0062] S400. Based on the expected arrival time and skill matching degree of the candidate personnel set, determine the target maintenance personnel, generate a maintenance work order bound to the warning event record, and push it to the mobile terminal of the target maintenance personnel.

[0063] In this embodiment, S400 is jointly executed by the scheduling matching module, the work order generation module, and the mobile terminal push module. The scheduling matching module receives the candidate personnel set and scheduling task context formed in S300, and generates personnel sorting results based on the maintenance personnel location, generator set location information, skill tags, historical order confirmation time, and current task load in the candidate personnel set; the work order generation module generates maintenance work orders based on early warning event records and scheduling task context; the mobile terminal push module pushes the maintenance work orders to the target maintenance personnel's mobile terminals and receives the push status.

[0064] The estimated arrival time is generated based on the location of maintenance personnel in the candidate personnel set and the generator set location information in the scheduling task context. The scheduling matching module can read map services or road network data and generate the estimated arrival time according to the distance between the maintenance personnel's location and the generator set location information and the road traffic status. In implementations without access to road network data, the scheduling matching module generates the estimated arrival time based on spatial distance and a preset driving speed. The skill matching degree is generated by matching the maintenance personnel's skill tags with the required skill tags in the scheduling task context. When the maintenance personnel's skill tags are completely consistent with the required skill tags, the skill matching degree is recorded as the first matching value. When the maintenance personnel's skill tags cover the main tags in the required skill tags but lack auxiliary tags, the skill matching degree is recorded as the second matching value. When the maintenance personnel's skill tags do not match, the maintenance personnel has been excluded in S300 and will not enter the sorting in S400.

[0065] The scheduling and matching module generates personnel sorting results based on estimated arrival time (from shortest to longest), skill matching degree (from highest to lowest), historical order confirmation time (from shortest to longest), and current task load (from shortest to longest). Specifically, the scheduling and matching module first groups the candidate personnel set according to skill matching degree; within the same skill matching degree group, it sorts by estimated arrival time; when estimated arrival times are the same or within the same time interval, it sorts by historical order confirmation time; when historical order confirmation times are the same or within the same time interval, it sorts by current task load. For Level 3 warning event records, the scheduling and matching module calls the personnel sorting priority written in S200, placing the candidate personnel selection process corresponding to Level 3 before Level 2 in the personnel sorting results. After the personnel sorting results are generated, the scheduling and matching module determines the maintenance personnel ranked first as the target maintenance personnel and generates a target maintenance personnel record.

[0066] The work order generation module receives the early warning event record, the scheduling task context, and the target maintenance personnel record, and then generates a maintenance work order. The maintenance work order is bound to the early warning event record by writing an early warning event identifier into the maintenance work order and a maintenance work order identifier into the early warning event record. The maintenance work order includes an early warning event identifier, an abnormal parameter snapshot, an abnormal duration, an early warning level, generator set location information, a fault code, a suggested maintenance plan, and a list of tools and equipment. The abnormal parameter snapshot is derived from the parameter snapshot in the early warning event record; the fault code is read from the fault code table based on the abnormal parameter type, out-of-bounds direction, equipment model, and early warning type; the suggested maintenance plan is read from the work order generation template based on the equipment model, early warning type, and fault code; and the list of tools and equipment is determined from the tools and equipment template based on the equipment model, early warning type, and fault code.

[0067] In one implementation, the tool and spare parts template includes tool name, spare part name, specification field, and quantity field. When the work order generation module reads that the equipment model is a specific diesel generator set model and the warning type is high water temperature, the corresponding tool and spare parts template includes coolant, radiator inspection tools, belt inspection tools, and water pump seals; for the low oil pressure warning type, the tool and spare parts template includes oil pressure testing tools, filter elements, and gaskets. The work order generation module does not directly modify the inventory status of the tool and spare parts library; in implementations that connect to the spare parts library unit, the work order generation module reads the available tool and spare parts records in the spare parts library unit and writes the available status into the tool and spare parts list of the maintenance work order.

[0068] After a maintenance work order is generated, the mobile terminal push module pushes the work order to the target maintenance personnel's mobile terminal. The push content includes the maintenance work order identifier, generator set location information, warning level, abnormal parameter snapshot, suggested maintenance plan, and a list of tools and spare parts. The mobile terminal push module generates a push record for each push, including the maintenance work order identifier, target maintenance personnel identifier, push time, and push status. If the push status is "failed," the mobile terminal push module can re-push within a preset waiting time; if repeated pushes still fail, the mobile terminal push module marks the work order status as "push failed" and provides this status to the S500 for status write-back and regeneration of the candidate personnel set.

[0069] In the on-site implementation scenario, the candidate personnel set generated by S300 includes maintenance personnel A, B, and C. The scheduling and matching module reads the location, skill tags, historical order confirmation time, and current task load of the three personnel. Maintenance personnel A's estimated arrival time is 18 minutes, and their skill tags are completely consistent with those for cooling system maintenance. Maintenance personnel B's estimated arrival time is 12 minutes, and their skill tags cover the main tags for cooling system maintenance. Maintenance personnel C's estimated arrival time is 25 minutes, and their skill tags are completely consistent. The scheduling and matching module generates personnel sorting results based on estimated arrival time, skill matching degree, historical order confirmation time, and current task load, and identifies maintenance personnel A as the target maintenance personnel. The work order generation module generates a maintenance work order based on the G001 warning event record and scheduling task context. The maintenance work order includes the warning event identifier, G001 location information, high water temperature fault code, suggested maintenance solutions for checking the coolant, radiator, fan, and water pump, and a corresponding list of tools and spare parts. The mobile terminal push module pushes the maintenance work order to the mobile terminal of maintenance personnel A; the maintenance work order identifier and push status are entered into S500 as input for writing back the work order status.

[0070] S500: Receive the work order status returned by the mobile terminal, and write back the maintenance work order status, generator set monitoring status, personnel status, and early warning event status based on the work order status; when the order is rejected, the order is accepted for too long, or the personnel status fails, regenerate the candidate personnel set based on the scheduling task context.

[0071] In this embodiment, S500 is jointly executed by the mobile terminal push module and the status write-back module. The mobile terminal push module receives the work order status returned by the target maintenance personnel's mobile terminal in S400 and writes the work order status into the work order status record; the status write-back module updates the maintenance work order status, generator set monitoring status, personnel status, and early warning event status based on the work order status record, generating a status write-back record. The work order status includes order acceptance status, order rejection status, arrival status, completion status, order acceptance timeout status, and personnel status failure status; in one embodiment, it also includes push failure status and unprocessable status.

[0072] The work order status is returned by the mobile terminal according to the maintenance work order identifier. After the target maintenance personnel select to accept the order on the mobile terminal, the mobile terminal push module receives the order acceptance status and the order acceptance time; the status write-back module updates the maintenance work order status to "accepted," the personnel status to "occupied," and the generator set monitoring status to "awaiting maintenance"; at the same time, the status write-back module writes the dispatched status and the target maintenance personnel identifier into the early warning event record. After the target maintenance personnel arrive at the site corresponding to the generator set location information, the mobile terminal returns the arrival status and arrival time; the status write-back module updates the maintenance work order status to "arrived," updates the generator set monitoring status to "under maintenance," and writes the arrival time into the work history record.

[0073] When the mobile terminal returns a completed status, the status write-back module receives the completion time, maintenance result record, and maintenance work order identifier. The status write-back module reads the most recent data packet containing the generator set's operating parameters and compares the water temperature, oil pressure, and speed with the recovery threshold range. When the generator set's operating parameters recover to the recovery threshold range, the status write-back module closes the warning event record, updates the generator set monitoring status to "recovered," updates the personnel status from "occupied" to "idle," and updates the maintenance work order status to "completed." When the mobile terminal returns a completed status but the generator set's operating parameters have not recovered to the recovery threshold range, the status write-back module keeps the warning event record open, updates the maintenance work order status to "pending review," and writes this status to the scheduling history. In one implementation, the status write-back module writes the unrecovered parameter values ​​to a parameter snapshot of the warning event record.

[0074] When the mobile terminal returns to the order rejection status, the status write-back module updates the repair work order status to the rejection status, restores the target repair personnel's status to its pre-rejection state, and writes the rejection status into the status write-back record. Subsequently, the status write-back module calls the scheduling task context bound to the repair work order and notifies the scheduling matching module to regenerate the candidate personnel set based on the scheduling task context. When regenerating the candidate personnel set, the target repair personnel who have already rejected the order are marked as excluded personnel in the re-screening round corresponding to the current warning event record. The scheduling matching module generates a new candidate personnel set according to the location update time, personnel status, service radius, and skill tag screening rules in S300, and S400 re-determines the target repair personnel and generates a push record.

[0075] When the target repair personnel does not return to the order acceptance status within the preset waiting time, the mobile terminal push module generates an order acceptance timeout status. Upon receiving the order acceptance timeout status, the status write-back module updates the repair work order status to the order acceptance timeout status. If the target repair personnel's status is still idle, the status write-back module retains their idle status; if their status has changed to occupied due to other repair work orders, the status write-back module records the personnel status as invalid. For both order acceptance timeout and personnel status invalidation statuses, the status write-back module regenerates the candidate personnel set based on the scheduling task context. When regenerating the candidate personnel set, the scheduling matching module rereads the location update time, personnel status, mobile terminal online status, and current task load from the repair personnel database and generates the candidate personnel set according to the latest personnel information.

[0076] When processing status write-back, the status write-back module maintains the work order status version. Each time a work order status is received from the mobile terminal, the status write-back module reads the current maintenance work order status version. If the work order status version returned by the mobile terminal is earlier than the current maintenance work order status version, the work order status is written to the late arrival record, without changing the current maintenance work order status. If multiple work order statuses are received for the same maintenance work order within a short period, the status write-back module updates the status in the order of acceptance status, arrival status, and completion status. When rejection status and completion status appear simultaneously, the status write-back module uses the earlier valid work order status as the current status and writes the conflicting statuses to the status conflict record.

[0077] In the on-site implementation scenario, after receiving a maintenance work order from G001, maintenance personnel A returns to the order acceptance status on their mobile terminal. The status write-back module updates maintenance personnel A's personnel status to occupied, updates the generator set monitoring status of G001 to pending maintenance, and writes the target maintenance personnel identifier into the warning event record. After maintenance personnel A arrives at the site, they return to the arrival status, and the status write-back module updates the generator set monitoring status of G001 to under maintenance. After maintenance personnel A completes the cooling system check and returns to the completion status, the status write-back module reads the most recent running data packet from G001. If the water temperature enters the recovery threshold range, the status write-back module closes the warning event record, updates the maintenance work order status to the completion status, and updates maintenance personnel A's personnel status to idle. If maintenance personnel A does not return to the order acceptance status within the preset waiting time, the mobile terminal push module generates an order acceptance timeout status, the status write-back module calls the scheduling task context corresponding to the maintenance work order, and causes the scheduling matching module to regenerate the candidate personnel set. The new candidate personnel set enters S400 for target maintenance personnel determination and maintenance work order push.

[0078] Example 2: Figure 2 This diagram illustrates a structural block diagram of a generator set operation anomaly early warning and work order scheduling system according to an embodiment of the present invention. Figure 2As shown, the structure may include:

[0079] The data acquisition module 01 is used to receive equipment identification, sampling time, and generator set operating parameters including water temperature, oil pressure, and speed. Specifically, the data acquisition module receives operating data packets from the acquisition terminal, which contain equipment identification, sampling time, water temperature, oil pressure, and speed. The data acquisition module establishes a receiving record according to the equipment identification and forms a time sequence record for the same generator set according to the sampling time. The data acquisition module verifies the integrity of the fields in the operating data packet; if the equipment identification or sampling time is missing, the operating data packet is recorded as a field missing state; if water temperature, oil pressure, or speed is missing, the missing parameter mark is retained, and the received parameter values ​​are written into the receiving record. The data acquisition module transmits the completed receiving record of equipment identification, sampling time, and generator set operating parameters to the anomaly determination module, so that the anomaly determination module can read the equipment file and threshold rule version.

[0080] Anomaly determination module 02, connected to the data acquisition module, is used to perform multi-level threshold comparisons on the generator set operating parameters according to the equipment file and threshold rule version, generating anomaly event records containing the direction of the boundary violation and the duration of the anomaly. Specifically, the anomaly determination module receives the equipment identifier, sampling time, and generator set operating parameters from the data acquisition module, and reads the corresponding equipment file and threshold rule version based on the equipment identifier. The anomaly determination module calls the multi-level thresholds corresponding to water temperature, oil pressure, and speed according to the threshold rule version, compares the generator set operating parameters item by item, and records the direction of the boundary violation according to the parameter type. The anomaly determination module matches the current sampling time with the previous boundary violation record of the same equipment identifier and the same abnormal parameter type; when there are consecutive boundary violation records, the boundary violation start time is used and the abnormal duration is accumulated; when there are no consecutive boundary violation records, the current sampling time is recorded as the boundary violation start time. The anomaly determination module generates anomaly event records, which contain the equipment identifier, abnormal parameter type, boundary violation direction, boundary violation start time, abnormal duration, and threshold rule version, and transmits the anomaly event records to the early warning module.

[0081] The early warning module 03, connected to the anomaly determination module, is used to determine the early warning level based on the anomaly event records, generate an early warning event record when the anomaly duration reaches a preset duration, and determine the scheduling trigger state according to the early warning level and scheduling trigger conditions. Specifically, the early warning module receives anomaly event records from the anomaly determination module and reads the anomaly parameter type, out-of-bounds direction, anomaly duration, and threshold rule version from the anomaly event records. The early warning module matches the anomaly parameter type, out-of-bounds direction, and anomaly duration with the early warning level rules according to the level rules in the threshold rule version to generate an early warning level. When the anomaly duration reaches the preset duration, the early warning module generates an early warning event record; the early warning event record includes an early warning event identifier, device identifier, anomaly parameter type, anomaly duration, early warning level, and scheduling trigger state. The early warning module generates a scheduling trigger state according to the early warning level and scheduling trigger conditions; when the scheduling trigger state is triggered, the early warning event record is output to the scheduling matching module.

[0082] The scheduling matching module 04, connected to the early warning module, is used to generate a scheduling task context containing generator set location information, service radius, and required skill tags when the scheduling trigger state is triggered. It also filters maintenance personnel based on location update time, personnel status, service radius, and skill tags to generate a candidate personnel set. Specifically, the scheduling matching module receives early warning event records from the early warning module and reads the equipment identifier, abnormal parameter type, early warning level, and scheduling trigger state from the early warning event records. When the scheduling trigger state is triggered, the scheduling matching module reads the generator set location information based on the equipment identifier, the required skill tags based on the abnormal parameter type, and the service radius based on the early warning level to form a scheduling task context. The scheduling matching module reads the maintenance personnel's location update time, personnel status, and skill tags, and performs distance matching between the maintenance personnel's location and the generator set location information. Maintenance personnel whose location update time is within the personnel status validity period, whose personnel status is idle, whose distance is less than the service radius, and whose skill tags match the required skill tags are written into the candidate personnel set. The scheduling and matching module provides the scheduling task context and candidate set to the personnel determination module, and retains the scheduling task context for the status write-back module to call when an order is rejected, an order is accepted and timed out, or a personnel status becomes invalid.

[0083] Personnel determination module 05, connected to the scheduling and matching module, is used to determine target maintenance personnel based on the estimated arrival time and skill matching degree of the candidate personnel set. Specifically, the personnel determination module receives the candidate personnel set and scheduling task context from the scheduling and matching module, and reads the maintenance personnel location, skill tag, personnel status, and location update time from the candidate personnel set. The personnel determination module generates the estimated arrival time based on the maintenance personnel location and generator set positioning information, and generates the skill matching degree based on the matching relationship between the skill tag and the required skill tag. The personnel determination module sorts the candidate personnel set according to the estimated arrival time and skill matching degree to form a target maintenance personnel record. When the candidate personnel set is empty, the personnel determination module returns an empty set status to the scheduling and matching module; when the candidate personnel set contains maintenance personnel, the personnel determination module transmits the target maintenance personnel record to the work order generation module.

[0084] The work order generation module 06, connected to the personnel determination module, is used to generate a maintenance work order bound to the warning event record based on the warning event record and the target maintenance personnel. Specifically, the work order generation module receives the target maintenance personnel record from the personnel determination module and calls the warning event record and scheduling task context corresponding to the target maintenance personnel record. The work order generation module reads the warning event identifier, abnormal parameter type, abnormal duration, and warning level from the warning event record, and reads the generator set location information, service radius, and required skill tags from the scheduling task context. The work order generation module writes the warning event identifier into the maintenance work order and backfills the maintenance work order identifier into the warning event record, forming a binding relationship between the maintenance work order and the warning event record. The maintenance work order includes the warning event identifier, target maintenance personnel, generator set location information, warning level, and abnormal duration. The work order generation module transmits the maintenance work order to the mobile terminal push module.

[0085] The mobile terminal push module 07, connected to the work order generation module, is used to push the repair work order to the mobile terminal of the target repair personnel and receive the work order status returned by the mobile terminal. Specifically, the mobile terminal push module receives the repair work order from the work order generation module and reads the target repair personnel and warning event identifier from the repair work order. The mobile terminal push module sends the repair work order based on the mobile terminal address associated with the target repair personnel and generates a push record; the push record includes the repair work order identifier, target repair personnel, push time, and push status. When the mobile terminal returns the work order status corresponding to "accepted," "rejected," "arrived," "completed," or "timeout," the mobile terminal push module binds the work order status with the repair work order identifier. When the push fails or the acceptance timeout occurs, the mobile terminal push module writes the corresponding status into the work order status record. The mobile terminal push module transmits the work order status record to the status write-back module.

[0086] The status write-back module 08, connected to the mobile terminal push module and the scheduling matching module, is used to write back the maintenance work order status, generator set monitoring status, personnel status, and early warning event status based on the work order status. When a work order is rejected, timed out, or personnel status expires, it regenerates a candidate personnel set based on the scheduling task context. Specifically, the status write-back module receives work order status records from the mobile terminal push module and reads the maintenance work order identifier, target maintenance personnel, and work order status from the work order status records. When the work order status is "accepted," the status write-back module writes the maintenance work order status as "accepted," the personnel status as "occupied," the generator set monitoring status as "pending maintenance," and the early warning event status as "dispatched." When the work order status is "arrived," the status write-back module writes the generator set monitoring status as "under maintenance." When the work order status is "completed," the status write-back module reads the latest record of the generator set operating parameters and, if the generator set operating parameters are within the recovery threshold range, writes the early warning event status as "closed." When the work order status is rejected, timed out, or personnel status expired, the status write-back module calls the scheduling task context retained by the scheduling matching module and instructs the scheduling matching module to regenerate the candidate personnel set according to the location update time, personnel status, service radius, and skill tags.

Claims

1. A method for early warning and work order scheduling of generator set operation anomalies, characterized in that, include: S100: Receive device identifier, sampling time, and generator set operating parameters including water temperature, oil pressure, and speed; perform multi-level threshold comparison according to device file and threshold rule version; and generate anomaly event records including the direction of the boundary crossing and the duration of the anomaly. S200. Determine the warning level based on the abnormal event record, generate a warning event record when the duration of the abnormality reaches a preset duration, and determine the scheduling trigger status according to the warning level and scheduling trigger conditions. S300: When the scheduling trigger state is triggered, a scheduling task context containing generator set location information, service radius and required skill tags is generated, and maintenance personnel are filtered according to location update time, personnel status, service radius and skill tags to generate a candidate personnel set; S400: Based on the expected arrival time and skill matching degree in the candidate personnel set, determine the target maintenance personnel, generate a maintenance work order bound to the warning event record, and push it to the mobile terminal of the target maintenance personnel; S500: Receive the work order status returned by the mobile terminal, and write back the maintenance work order status, generator set monitoring status, personnel status, and early warning event status based on the work order status; when the order is rejected, the order is accepted for too long, or the personnel status fails, regenerate the candidate personnel set based on the scheduling task context.

2. The method according to claim 1, characterized in that, The generator set operating parameters include battery voltage; receiving device identification, sampling time, and generator set operating parameters includes: the acquisition terminal reads the generator set operating parameters from the generator set controller at a fixed frequency, and encapsulates the device identification, the sampling time, and the generator set operating parameters into an operating data packet.

3. The method according to claim 1, characterized in that, The abnormal event record includes the abnormal parameter type, cross-boundary direction, cross-boundary start time, number of consecutive cross-boundary occurrences, abnormal duration, and threshold rule version; when the generator set operating parameters recover to the recovery threshold range, the status of the abnormal event record is updated.

4. The method according to claim 1, characterized in that, The warning levels include a first level, a second level, and a third level; the first level corresponds to alarm information, the second level and the third level correspond to a triggering status, and the personnel sorting priority of the third level is higher than that of the second level.

5. The method according to claim 1, characterized in that, The scheduling task context also includes an early warning event identifier, equipment model, early warning type, tool and component template, and work order generation template; the tool and component template is read according to the equipment model and early warning type.

6. The method according to claim 1, characterized in that, The system filters maintenance personnel based on location update time, personnel status, service radius, and skill tags. This includes verifying the online status of maintenance personnel's mobile terminals and filtering maintenance personnel whose location update time is within the validity period of their personnel status, whose personnel status is idle, whose distance from the generator set's location information is less than the service radius, and whose skill tags match the required skill tags.

7. The method according to claim 1, characterized in that, Determining target maintenance personnel based on the expected arrival time and skill matching degree of the candidate personnel set includes: generating the personnel ranking results by expected arrival time from shortest to longest, skill matching degree from highest to lowest, historical order confirmation time from shortest to longest, and current task load from shortest to longest, and determining the maintenance personnel ranked first in the personnel ranking results as the target maintenance personnel.

8. The method according to claim 1, characterized in that, The maintenance work order includes an early warning event identifier, an abnormal parameter snapshot, an abnormal duration, an early warning level, generator set location information, a fault code, a suggested maintenance plan, and a list of tools and equipment; the list of tools and equipment is determined from the tools and equipment template based on the equipment model, early warning type, and fault code.

9. The method according to claim 1, characterized in that, When the work order status is "accepted", the personnel status is updated to "occupied" and the generator set monitoring status is updated to "pending maintenance"; when the work order status is "arrived", the generator set monitoring status is updated to "under maintenance"; when the work order status is "completed" and the generator set operating parameters have recovered to the recovery threshold range, the early warning event record is turned off.

10. A generator set operation anomaly early warning and work order scheduling system, characterized in that, include: Data acquisition module, anomaly detection module, early warning module, scheduling matching module, personnel identification module, work order generation module, mobile terminal push module, and status write-back module; The system is used to implement the method described in any one of claims 1-9.