Intelligent maintenance management system and method for fan equipment

Abstract

The invention belongs to the technical field of fan equipment maintenance management, and relates to an intelligent maintenance management system and method for fan equipment, and the system comprises a data access unit which generates a unified maintenance data set according to a maintenance data exchange standard; the standard library unit is used for storing an operation instruction standard, an equipment fault model library, an acceptance standard, an evaluation standard and a maintenance file package template; the diagnosis decision-making unit is used for outputting diagnosis results containing phenomena, reasons, conclusions and disposal suggestion fields and one-to-one maintenance schemes; the work order management and control unit is used for generating a work order and a maintenance file package and forming a node constraint acquisition process record; the archiving updating unit is used for generating an archiving data set and updating a fault model through consistency verification during closed loop; and the evaluation unit is used for generating an evaluation result after closed loop. According to the technical scheme, the cross-system data connection consistency and evaluation evidence support stability can be improved, the reusability of the maintenance decision basis is enhanced, and the management requirements of maintenance operation on closed-loop treatment and quality management and control are met.

Description

Technical Field

[0001] This invention belongs to the field of wind turbine equipment maintenance management technology, and specifically relates to an intelligent maintenance management system and method for wind turbine equipment. Background Technology

[0002] In existing technologies, the maintenance management of wind turbine equipment typically relies on the collaboration of remote monitoring and diagnostic platforms, defect systems, and maintenance management platforms. This involves collecting operational monitoring data and alarm signals, having maintenance personnel initiate work orders and organize maintenance operations within the platform, and generating records and evaluations upon completion of the work to meet the needs of wind turbine equipment status control, maintenance process tracking, and post-maintenance assessment. However, existing maintenance management methods have some significant shortcomings in terms of cross-system data integration and closed-loop support capabilities.

[0003] In practical applications, although existing maintenance management platforms have certain capabilities in work order flow, online approval, document management, and record archiving, operational monitoring data, alarm signals, defect data, and maintenance process record data are often generated in different systems or different business links. The overall consistency of data object definitions and field expressions is low, and the correlation stability between data in work orders, document packages, and evaluation results is also weak. At the same time, post-maintenance evaluations are mostly formed by post-event summarization, and their correspondence with maintenance process evidence is weak, resulting in less than ideal consistency of evaluation conclusions in historical review and subsequent decision-making.

[0004] Therefore, it is evident that existing technologies often suffer from problems such as low consistency in cross-system data integration during wind turbine equipment maintenance and management, and weak stability in supporting process evidence with closed-loop evaluation conclusions. These are the shortcomings of existing technologies.

[0005] In view of this, it is very necessary for the present invention to provide an intelligent maintenance management system and method for wind turbine equipment to solve the above-mentioned defects in the prior art. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of the existing technology, such as the low level of consistency in cross-system data integration and the weak stability of closed-loop evaluation conclusions in supporting process evidence in wind turbine equipment maintenance management, by providing an intelligent maintenance management system and method for wind turbine equipment to solve the above-mentioned technical problems.

[0007] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, embodiments of this application provide an intelligent maintenance and management system for wind turbine equipment, comprising: The data access unit is used to receive operation monitoring data and alarm signals output by the remote monitoring and diagnostic platform, receive maintenance project data and maintenance process record data entered by the user terminal, receive defect data output by the defect system, and generate a unified maintenance dataset according to the maintenance data exchange standard. The standard library unit is used to store the technical standard database and the maintenance document package template library. The technical standard database includes work instruction standards, equipment failure model library, acceptance standards and evaluation standards. The diagnostic decision unit is used to generate diagnostic results and one-to-one maintenance solutions based on a unified maintenance dataset and technical standard database. The diagnostic results include phenomenon fields, cause fields, conclusion fields, and handling suggestion fields. The work order management unit is used to generate work orders based on one-to-one maintenance plans and perform online approval, generate maintenance document packages associated with work orders based on the maintenance document package template library, form work order node constraints based on work instruction standards, and collect maintenance process record data. The archive update unit is used to perform field mapping and consistency verification on the unified maintenance dataset according to the maintenance data exchange standard when the work order is in a closed loop, generate the archive dataset, and update the equipment fault model in the equipment fault model library based on the defect data and the archive dataset. The evaluation and assessment unit is used to perform post-repair evaluations on work orders in a closed-loop state. The post-repair evaluation includes at least one of installation evaluation, trial operation evaluation, and personnel assessment, and generates assessment results associated with the work order.

[0008] As a preferred option, the maintenance and repair data exchange standard includes equipment identification codes, component identification codes, process identification codes, key point identification codes, evidence type codes, and time indexing rules; The data access unit performs time alignment on operation monitoring data, alarm signals, maintenance project data, maintenance process record data and defect data according to time indexing rules, and performs granular mapping based on equipment identification code, component identification code, process identification code and key point identification code to generate a unified maintenance dataset containing key point identification code, evidence type code and time index field.

[0009] As a preferred option, after generating the diagnostic results and one-to-one repair solutions, the diagnostic decision unit writes the diagnostic results and one-to-one repair solutions into a structured output object. The structured output object includes an evidence citation list and a solution version identifier. The evidence citation list includes evidence time index information corresponding to the time index field. After generating a work order, the work order management unit triggers the automatic upload of structured output objects. The automatic upload records include the upload time field, the work order identifier field, the solution version identifier field, and the evidence reference list.

[0010] As a preferred option, after the work order control unit forms work order node constraints based on the work instruction standards, it configures the set of mandatory fields for nodes, the set of evidence for nodes, and the set of approval fields for nodes. Before the work order is in a closed loop state, the archiving update unit performs integrity and consistency checks on the maintenance process record data and work order node constraints. If the checks fail, a data entry task bound to the work order node is generated. The archiving update unit generates the archived dataset when the data entry task is completed and the work order is in a closed loop state.

[0011] As a preferred embodiment, when the archive update unit generates the archive dataset according to the maintenance and repair data exchange standard, it generates summary values ​​for the field set, evidence reference list and corresponding attachment data in the archive dataset and writes them into the metadata field of the archive dataset. It also associates and stores the summary values ​​with the work order identifier field, scheme version identifier field and node approval field set. When the archive dataset changes, the summary values ​​are updated and the change log is recorded.

[0012] Preferably, the archive update unit configures the model version identifier and model verification rule set for the equipment fault model library, and executes the model verification rule set and generates verification results after updating the equipment fault model. If the verification results do not meet the model verification rule set, the equipment fault model is rolled back to the previous model version identifier. The model verification rule set includes sample playback rules based on the archive dataset and label consistency rules based on defect data.

[0013] As a preferred embodiment, the diagnostic decision unit is also used to generate remaining life parameters based on the life evaluation parameter set in the unified maintenance dataset, write the remaining life parameters into the life field of the diagnostic results, and use the life field as the input field for generating a one-to-one maintenance plan. The life evaluation parameter set includes design life parameters, viscosity ratio parameters, cleanliness coefficient parameters, actual operating time parameters, over-temperature number parameters, acceleration rate constant parameters, and over-rated current number parameters.

[0014] As a preferred option, the evaluation unit generates a set of process quality indicators based on work order node constraints and archived datasets, performs repeatability and anomaly detection on the evidence citation list, and writes the repeatability and anomaly detection results into the rule hit field of the evaluation results. The set of process quality indicators includes key point coverage indicators, evidence consistency indicators, and node time sequence consistency indicators. The node time sequence consistency indicators are calculated based on evidence time index information.

[0015] Secondly, embodiments of this application also provide an intelligent maintenance management method for wind turbine equipment, including: It receives operation monitoring data and alarm signals output from the remote monitoring and diagnostic platform, receives maintenance project data and maintenance process record data entered by the user terminal, receives defect data output by the defect system, and performs time alignment according to the maintenance data exchange standard and maps it according to equipment identification code, component identification code, process identification code and key point identification code to generate a unified maintenance dataset containing key point identification code, evidence type code and time index field. Based on a unified maintenance dataset and a technical standard database, diagnostic results and one-to-one maintenance solutions are generated and written into a structured output object. The diagnostic results include a phenomenon field, a cause field, a conclusion field, and a treatment suggestion field. The structured output object includes an evidence citation list, a solution version identifier, and evidence time index information. Based on the one-to-one maintenance plan, work orders and maintenance document packages associated with the work orders are generated. Work order node constraints are formed according to the work instruction standards, and maintenance process record data is collected. Before the work order is closed, perform integrity and consistency checks on the maintenance process record data and work order node constraints, and generate a data supplementation task; When the data entry task is completed and the work order is closed, an archived dataset is generated, and the equipment fault model library is updated based on the defect data and the archived dataset. Post-repair evaluation is performed on closed-loop work orders and evaluation results are generated. The post-repair evaluation includes at least one of installation evaluation, trial operation evaluation and personnel evaluation.

[0016] As a preferred method, the steps of updating the equipment failure model library based on defect data and archived datasets include: Configure model version identifiers and model validation rule sets for the equipment fault model library; Candidate equipment failure models are generated based on defect data and archived datasets, and the equipment failure models in the equipment failure model library are updated. The model validation rule set is executed and validation results are generated. If the validation results do not meet the model validation rule set, the equipment fault model is rolled back to the previous model version identifier. The model validation rule set includes sample playback rules based on archived datasets and label consistency rules based on defect data.

[0017] As can be seen from the above technical solutions, the present invention has the following advantages: This application provides an intelligent maintenance management system and method for wind turbine equipment. By standardizing the field definitions and mapping objects of operation monitoring, alarms, defects, and maintenance process records using maintenance data exchange standards, a unified maintenance dataset is formed. Upon work order closure, the dataset undergoes field mapping and consistency verification to generate an archived dataset. This achieves alignment, comparability, and traceability of multi-source data across equipment, components, processes, and key points, reducing the impact of cross-system data differences on work order processing and record archiving. By associating diagnostic results with one-to-one maintenance plans, maintenance document packages, work order node constraints, and process records, and generating installation, commissioning, or personnel evaluation results on the closed-loop work order, a structured association and closed-loop solidification of evaluation conclusions and process evidence are achieved. This enhances the evidentiary support of evaluation conclusions and the ability to compare with historical work orders. Furthermore, by updating the equipment fault model based on the archived dataset, continuous consistency between the model and on-site handling results is achieved, improving the stability and reusability of maintenance decision-making basis and meeting the needs of wind turbine equipment maintenance management for data consistency, closed-loop evaluation credibility, and sustainable optimization of operation and maintenance decisions.

[0018] Furthermore, the design principle of this invention is reliable, the structure is simple, and it has a very wide range of application prospects.

[0019] Therefore, it is evident that the present invention has outstanding substantive features and significant progress compared with the prior art, and the beneficial effects of its implementation are also obvious. Attached Figure Description

[0020] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the intelligent maintenance and management system for wind turbine equipment provided by the present invention; Figure 2 This is a flowchart of an intelligent maintenance and management method for wind turbine equipment provided by the present invention.

[0022] The system comprises: 1. Data access unit; 2. Standard library unit; 3. Diagnostic decision-making unit; 4. Work order management unit; 5. Archive update unit; and 6. Evaluation and assessment unit. Detailed Implementation

[0023] Various embodiments of this disclosure are described more fully below with reference to the accompanying drawings. This disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of this disclosure to the specific embodiments disclosed herein, but rather this disclosure should be understood to cover all adjustments, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of this disclosure.

[0024] In the following, the terms “comprising” or “may include”, which may be used in various embodiments of this disclosure, indicate the presence of the disclosed functions, operations, or elements, and do not limit the addition of one or more functions, operations, or elements. Furthermore, as used in various embodiments of this disclosure, the terms “comprising,” “having,” and their cognates are intended only to indicate a particular feature, number, step, operation, element, component, or combination of the foregoing, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing, or the possibility of adding one or more combinations of the foregoing.

[0025] It should be noted that, in various embodiments of this disclosure, the expression "or" or "at least one of A and / or B" includes any combination or all combinations of the words listed simultaneously. For example, the expression "A or B" or "at least one of A and / or B" may include A, may include B, or may include both A and B.

[0026] It should be noted in advance that, in order to facilitate a clear and accurate description of the technical solutions in the embodiments of this application, the following is a brief explanation of some terms and related technologies involved in the embodiments of this application: 1. Inspection and maintenance data exchange standard Maintenance and repair data exchange standards are a set of rules used to constrain how data is described and exchanged between different systems. They typically include data field definitions, coding systems, units and dimensions, data dictionaries, value ranges or enumerations, mandatory field constraints, and interface data formats. These standards are used to ensure semantic consistency of multi-source data during transmission and aggregation and to facilitate subsequent association and processing.

[0027] 2. Time indexing rules and time alignment Time indexing rules define the time stamping method for data records and its usage constraints, such as timestamp precision, time zone reference, sampling period, alignment window, and late data processing standards. Time alignment refers to mapping data from different sampling frequencies or different sources onto the same time axis according to a unified time reference, so as to perform correlation analysis and consistency processing at the same moment or within the same time window.

[0028] 3. Field mapping and consistency verification Field mapping refers to the process of establishing a correspondence between fields in different systems or data structures and fields in a target data model. It typically includes conversion rules for field names, types, units, encoding meanings, and granularity. Consistency verification refers to performing rule checks on the mapped data based on preset constraints, such as the completeness of required fields, the validity of encodings, the consistency of units, the matching of association keys, and the logical consistency across fields, in order to reduce the uncertainty caused by data structure and semantic deviations.

[0029] 4. Summary value and metadata fields A digest value typically refers to a fixed-length hash result calculated from the content of a piece of data or file. It is often used for integrity verification, duplicate identification, and change tracking. Metadata fields are a set of fields used to store descriptive information, such as data version, generation time, source identifier, digest value, association identifier, and audit information, giving data objects the attributes of being manageable and traceable.

[0030] 5. Remaining life parameters RUL (Remaining Useful Life) is a quantitative representation of the time or amount of work that equipment or components are expected to be able to perform stably under given operating conditions. It is usually based on an estimation model built on operating time, stress event counts, degradation indicators and condition monitoring quantities, and outputs a life prediction result that can be used for maintenance decisions. Its essence is to map condition data and degradation patterns into a comparable life measure.

[0031] To address the issues of low data consistency and weak support for maintenance process evidence in wind turbine equipment maintenance management, which arise from the dispersed nature of operational monitoring data, alarm signals, defect data, and maintenance process records across different systems and business stages, and the difficulty in meeting the practical needs of closed-loop governance, historical review comparability, and stability of subsequent maintenance decision-making in wind turbine equipment maintenance management, this application discloses an intelligent maintenance management system and method for wind turbine equipment. By establishing a unified maintenance dataset and archived dataset oriented towards maintenance business objects and performing field mapping and consistency verification at each work order closure point, a unified standard and traceable association of cross-system data are achieved. Simultaneously, by associating diagnostic results with one-to-one maintenance plans, maintenance document packages, work order node constraints, maintenance process records, and post-maintenance evaluation results within the same work order link, the support of evaluation conclusions for process evidence and the stability of work order data association across different stages are enhanced. Furthermore, by updating equipment fault models using archived data, the continuity of the maintenance management closed loop and the stability and reusability of maintenance decision-making basis are further improved.

[0032] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0033] like Figure 1 As shown in the figure, the intelligent maintenance and management system for wind turbine equipment provided in this embodiment includes: Data access unit 1 is used to receive operation monitoring data and alarm signals output by the remote monitoring and diagnostic platform, receive maintenance project data and maintenance process record data entered by the user terminal, receive defect data output by the defect system, and generate a unified maintenance dataset according to the maintenance data exchange standard. Standard Library Unit 2 is used to store the technical standard database and the maintenance document package template library. The technical standard database includes work instruction standards, equipment failure model library, acceptance standards and evaluation standards. Diagnostic decision unit 3 is used to generate diagnostic results and one-to-one maintenance solutions based on a unified maintenance dataset and technical standard database. The diagnostic results include phenomenon fields, cause fields, conclusion fields, and handling suggestion fields. Work order management unit 4 is used to generate work orders based on one-to-one maintenance plans and perform online approval, generate maintenance document packages associated with work orders based on maintenance document package template library, form work order node constraints based on work instruction standards, and collect maintenance process record data. The archive update unit 5 is used to perform field mapping and consistency verification on the unified maintenance dataset according to the maintenance data exchange standard when the work order is in a closed loop state, generate the archive dataset, and update the equipment fault model in the equipment fault model library based on the defect data and the archive dataset. Evaluation and assessment unit 6 is used to perform post-repair evaluation on work orders in a closed-loop state. The post-repair evaluation includes at least one of installation evaluation, trial operation evaluation and personnel assessment, and generates assessment results associated with the work order.

[0034] This embodiment uses data access unit 1 to achieve unified access to operational monitoring data, alarm signals, defect data, and maintenance process record data from remote monitoring and diagnostic platforms, defect systems, and user terminals. It generates a unified maintenance dataset based on maintenance data exchange standards, enabling the convergence and organization of multi-source maintenance-related data under the same data caliber, improving the consistency and traceability of cross-system data. Archive update unit 5 performs field mapping and consistency checks on the unified maintenance dataset in a closed-loop work order state, generating an archive dataset. This ensures that archived data maintains a stable and consistent structured expression at the field level, enhancing the stability of the association between work orders, file packages, process records, and archived data, and improving historical review comparability. Diagnostic decision unit 3 generates a dataset containing phenomena and causes based on the unified maintenance dataset and technical standard database. The diagnostic results in the "Conclusions and Handling Recommendations" fields are used to form a one-to-one maintenance plan. This plan, combined with the maintenance document package and the work instructions standards in the standard library unit 2, forms work order node constraints to guide the collection of maintenance operation process records. This ensures that diagnosis, handling, and process recording form a closed-loop association within the same work order chain, enhancing the support strength of the post-maintenance evaluation in the evaluation and assessment unit 6 for process evidence. Simultaneously, the archiving and updating unit 5 uses defect data and archived datasets to update the equipment fault models in the equipment fault model library, ensuring that the fault models remain consistent with the on-site handling results and providing a more stable basis for subsequent maintenance decisions. Ultimately, this improves the data consistency level, the credibility of closed-loop evaluation, and the reusability of decision-making basis in wind turbine equipment maintenance management, meeting the management needs of closed-loop governance and quality control in wind turbine equipment maintenance operations.

[0035] The following describes in detail the various parts of the above system based on embodiments of this application.

[0036] This system constructs a closed-loop data engineering and closed-loop management mechanism around the business chain of "remote monitoring and diagnostic platform monitoring operating equipment, fault monitoring alarm signals being transmitted to user terminals, platform analyzing and diagnosing faults and providing solutions, end users initiating work orders to handle faults, maintenance teams handling the faults, work orders closing after equipment trial operation and normal operation, data archiving, process termination, and subsequent personnel evaluation and incentives." Specifically, the system abstracts maintenance-related business objects into a unified set of data objects, including operational monitoring data, alarm signals, defect data, maintenance project data, maintenance process record data, maintenance document packages, work instruction standards, diagnostic results, one-to-one maintenance solutions, work orders, archived datasets, equipment fault models, post-repair evaluation results, and assessment results. It achieves object cascading, traceable evidence, and iterative models through unified coding and time indexing rules.

[0037] In this embodiment, regarding data organization, the system constructs a "unified maintenance dataset" as the core carrier for cross-system connectivity. The data granularity is unified to the dimensions of equipment, components, processes, and key points, and monitoring evidence and maintenance evidence are linked by a time index field to achieve consistent expression and traceable association across links. Specifically, the system receives operation monitoring data and alarm signals output by the remote monitoring and diagnostic platform through the data access unit 1, receives maintenance project data and maintenance process record data entered by the user terminal, receives defect data output by the defect system, and generates a unified maintenance dataset according to the maintenance data exchange standard. In this embodiment, the "maintenance data exchange standard" is implemented in the form of a data model and rule set to achieve the execution of unified field definitions, unified coding system, and time alignment and granular mapping.

[0038] In some embodiments of this application, a multi-code unified identification and coding system is adopted around the maintenance and repair data exchange standard to abstract data objects from different sources into a unified coordinate system. Specifically, the maintenance and repair data exchange standard may include equipment identification codes, component identification codes, process identification codes, key point identification codes, evidence type codes, and time indexing rules. Among them, the equipment identification code is used to uniquely identify the wind turbine equipment or a subset of equipment; the component identification code is used to identify component objects such as blades, bearings, gearboxes, and lubrication systems; the process identification code is used to describe process steps such as disassembly and inspection, assembly, alignment, and trial operation; the key point identification code is used to limit the key inspection points that need to be collected and signed off within the process; the evidence type code is used to distinguish evidence forms such as waveforms, photos, tables, and text records; and the time indexing rules are used to normalize multi-source timestamps to a unified time axis.

[0039] Furthermore, the time index can be defined as a discrete index function:

[0040] in, This is the original timestamp. To unify the starting point of the timeline, For time granularity, This retrieves the value for the time index field. For example, for data sampled at non-equal intervals, the original timestamp can first be indexed... Mapping is then performed, followed by alignment and aggregation according to the index window, to ensure that cross-system data can be compared, retrieved, and traced under a unified time index.

[0041] Based on this, data access unit 1 can perform time alignment on operation monitoring data, alarm signals, maintenance project data, maintenance process record data, and defect data according to time indexing rules, and perform granular mapping based on equipment identification codes, component identification codes, process identification codes, and key point identification codes to generate a unified maintenance dataset containing key point identification codes, evidence type codes, and time index fields. Specifically, granular mapping can be achieved through a mapping function:

[0042] Mapping quad-code to uniform entity identifier And attach a record of each piece of evidence. A ternary index is used to achieve structured anchoring of "key point-evidence-time". For example, The value can be taken in minutes or seconds, depending on the sampling rate of the monitoring data and the frequency of maintenance operation records.

[0043] In this embodiment, regarding the standard library and template library, the "standards" and "deliverable templates" for maintenance operations are centrally managed to ensure that the outputs of different suppliers and work teams are aligned, auditable, and statistically verifiable. Specifically, the standard library unit 2 can store a technical standard database and a maintenance document package template library. The technical standard database can include work instruction standards, equipment fault model libraries, acceptance standards, and evaluation standards. Work instruction standards describe process nodes, key requirements, evidence types, and approval rules. Acceptance standards define trial operation criteria and acceptance threshold sets. Evaluation standards define the calculation methods for installation evaluation, trial operation evaluation, and personnel assessment. The equipment fault model library stores model parameters, rule sets, or learning model versions generated for fault location, cause attribution, and handling suggestions. It can be specifically implemented as a combination of rule models, statistical models, or data-driven models.

[0044] In this embodiment, regarding diagnosis and solution generation, the "diagnosis result - one-to-one maintenance solution - evidence reference" is solidified into a structured object to avoid the problems of incalculability and unverifiability caused by outputting only in text form. Specifically, through the diagnosis decision unit 3, diagnosis results and one-to-one maintenance solutions are generated based on a unified maintenance dataset and technical standard database. The diagnosis results may include phenomenon fields, cause fields, conclusion fields, and handling suggestion fields. The phenomenon field is used to describe the external manifestations derived from alarm signals and monitoring feature quantities. The cause field is used to point to a set of candidate causes at the component level or mechanism level. The conclusion field is used to provide fault type determination and confidence information. The handling suggestion field is used to provide a set of handling items bound to procedures and key points.

[0045] In this embodiment of the application, a diagnostic scoring function can be constructed based on feature quantities:

[0046] in, Candidates for fault categories, This refers to the feature vector extracted from the unified maintenance dataset (which may include vibration features, temperature features, current features, oil condition features, etc.). These are the weighting coefficients. The mapping function from features to evidence contribution; when When the classification threshold is exceeded, Write the conclusion field, and generate a list of procedures and key points for the handling suggestion field based on the work instruction standard mapping, thereby forming a one-to-one maintenance plan. For example, Piecewise linear functions or logic functions can be used to correspond to alarm intervals for different feature quantities.

[0047] Furthermore, to link diagnostic output with work order execution, this embodiment introduces a structured output object and automatic record upload to achieve versioning of solutions and evidence, as well as a strong association with work orders. Specifically, after generating diagnostic results and one-to-one repair solutions, the diagnostic decision unit 3 can write the diagnostic results and one-to-one repair solutions into a structured output object. The structured output object includes an evidence citation list and a solution version identifier. The evidence citation list includes evidence time index information corresponding to the time index field, used to list the set of evidence items related to the diagnostic chain. Each item includes at least a key point identifier code, evidence type code, evidence time index information, and evidence storage location index. The solution version identifier is used to identify the version of the diagnostic output, facilitating subsequent review and model verification reproduction.

[0048] Based on this, the work order management unit 4 can trigger the automatic upload of structured output objects after a work order is generated. The automatic upload record includes an upload time field, a work order identifier field, a solution version identifier field, and an evidence citation list. Among them, the upload time field is used to solidify the time when the solution is entered into the database, the work order identifier field is used to achieve a one-to-one binding between the solution and the work order, and the evidence citation list is used to ensure that the diagnostic basis can be replayed and verified according to the evidence time index information during the work order execution phase.

[0049] In this embodiment, regarding work order management, the collection of maintenance document packages, node constraints, and process records can be transformed from "manual self-regulation" to "system constraints," enabling the consistent accumulation of closed-loop data. Specifically, the work order management unit 4 can generate work orders based on a one-to-one maintenance plan and perform online approval, generate maintenance document packages associated with the work orders based on the maintenance document package template library, form work order node constraints based on the work instruction standards, and collect maintenance process record data. Among these, the work order node constraints are used to limit the fields, evidence, and signature information that must be submitted at each node, thereby forming a computable basis for completeness and consistency.

[0050] Furthermore, after the work order control unit 4 forms work order node constraints based on the work instruction standard, it can configure the node mandatory field set, node evidence set, and node approval field set. Among them, the node mandatory field set is used to constrain structured fields (such as process parameters, measurement point readings, operator identification, etc.), the node evidence set is used to constrain evidence type and quantity, and the node approval field set is used to constrain the approval link and approval time index.

[0051] Furthermore, regarding the implementation of archiving and clean-free archiving, this application defines "work order closed loop" as the archiving trigger point and uses pre-archiving verification and supplementary data entry tasks as pre-closed loop constraints to ensure that the archived dataset can be directly used for review and model updates. Specifically, the archiving update unit 5 can perform integrity and consistency checks on the maintenance process record data and work order node constraints before the work order is in a closed loop state. If the checks fail, a data supplementary data entry task bound to the work order node is generated. The archiving update unit 5 generates the archived dataset when the data supplementary data entry task is completed and the work order is in a closed loop state. Among them, the integrity check is used to check whether the set of mandatory fields for the node is complete, whether the set of evidence for the node meets the requirements for evidence type and quantity, and whether the set of signature fields for the node meets the signature link; the consistency check is used to check the degree of consistency between the process record data and the evidence citation list in terms of key point identification encoding, evidence type encoding, and evidence time index information.

[0052] To make consistency quantifiable, this embodiment defines a consistency determination function:

[0053] in, For evidence items, Encode the key points. Encode the type of evidence. For evidence time index information; These are the allowed sets given by the node constraints; when At that time, a data entry task is generated and bound to the corresponding work order node.

[0054] Once a work order reaches a closed-loop state, the archiving update unit 5 can perform field mapping and consistency verification on the unified maintenance dataset according to the maintenance data exchange standard to solidify the historical traceability link. Simultaneously, it generates an archived dataset and updates the equipment fault models in the equipment fault model library based on defect data and the archived dataset. Specifically, field mapping is used to merge operation monitoring, alarms, defects, and maintenance records into the archived field set under a unified field definition.

[0055] Furthermore, to enhance the verifiability and traceability of archived data, this application embodiment introduces a summary value and change log mechanism. Specifically, when generating an archived dataset according to the maintenance and repair data exchange standard, the archive update unit 5 can generate summary values ​​for the field set, evidence citation list, and corresponding attachment data in the archived dataset and write them into the metadata fields of the archived dataset. The summary values ​​are then associated and stored with the work order identifier field, scheme version identifier field, and node approval field set. When the archived dataset changes, the summary values ​​are updated and a change log is recorded. The summary values ​​can be generated using a summary function. Calculations are performed on the field set, the evidence citation list, and the serialized results of the attachment data:

[0056] in, For a collection of fields, For the list of evidence cited, For the attached data set, For deterministic serialization functions, The summary value is used for each change; the change log records the work order identifier field, solution version identifier field, change time index, and old and new summary values ​​corresponding to each change, thereby enabling traceability and verification of the archived dataset. For example, It can be implemented as a summarization algorithm that satisfies collision resistance characteristics. The serialization order can be fixed according to enterprise standards.

[0057] In this embodiment, regarding the continuous updating of the equipment fault model, not only can the model be updated, but the "update-verify-rollback" process is also solidified into an executable set of verification rules to avoid decision instability caused by model drift. Specifically, the archive update unit 5 can configure a model version identifier and a set of model verification rules for the equipment fault model library, and execute the set of model verification rules and generate verification results after updating the equipment fault model. If the verification results do not meet the requirements of the set of model verification rules, the equipment fault model is rolled back to the previous model version identifier. The set of model verification rules includes sample replay rules based on the archived dataset and label consistency rules based on defect data; the model version identifier is used to identify the version of the model parameters or rule set; the sample replay rules are used to reproduce historical samples on the archived dataset and calculate the model output deviation; the label consistency rules are used to use defect data as the label source to check the consistency between the model output category and the defect label. For example, a replay deviation metric can be defined:

[0058] in, To replay the number of samples, Output for candidate models, For defect data labels or archived conclusion labels, Let be the loss function; when When the verification threshold is exceeded or the consistency ratio is lower than the threshold, the verification result does not meet the model verification rule set and triggers a rollback to the previous model version identifier, thereby ensuring the stability of the decision-making basis.

[0059] Furthermore, in terms of lifespan assessment, this application embodiment introduces remaining lifespan parameters and writes them into the diagnostic results as one of the input fields for generating a one-to-one maintenance plan, thereby linking maintenance decisions with health status assessment.

[0060] Specifically, the remaining life parameters can be generated by the diagnostic decision unit 3 based on the set of life evaluation parameters in the unified maintenance dataset, and the remaining life parameters can be written into the life field of the diagnostic results. The life field can also be used as the input field for generating a one-to-one maintenance plan.

[0061] The lifespan evaluation parameter set includes design lifespan parameters, viscosity ratio parameters, cleanliness coefficient parameters, actual operating time parameters, over-temperature count parameters, acceleration rate constant parameters, and over-current count parameters. For example, the remaining lifespan is... In other words, the design life is... Indicates that the viscosity is higher than that of the viscosity. The cleanliness coefficient is expressed as follows: This indicates that the actual running time is based on It indicates that the number of times the temperature exceeds the limit is... This indicates that the acceleration constant is expressed as... This indicates the number of times the rated current is exceeded. express.

[0062] To facilitate engineering implementation, this embodiment presents an implementable accelerated degradation model:

[0063] in, The degradation acceleration term, used to comprehensively reflect the impact of lubricating oil condition and stress events on service life, can be defined as:

[0064] in, These are the weighting coefficients; and To map the viscosity ratio and cleanliness coefficient to a function of dimensionless degradation contribution, a piecewise function can be used to reflect different oil state ranges. For example, a method can be adopted... , The weighting coefficients are determined in conjunction with maintenance strategies, causing the lifespan field to decay at an accelerated rate with increasing over-temperature and over-rated current cycles. When this lifespan field is used in the generation of a one-to-one maintenance plan, it can be used to adjust the process strategies corresponding to key point identification codes, evidence type coding requirements, or key items in trial operation evaluation, thereby forming a treatment plan consistent with the health status. Exemplarily, demonstratively, The number of hours can be obtained from the equipment design data. It can be obtained by centralizing the cumulative running time in the unified maintenance dataset and aligning it with the time index field. and It can be calculated from oil level monitoring or maintenance inspection data. and It can be obtained from alarm signals and event counts in the process log.

[0065] Furthermore, regarding post-repair evaluation and assessment, this application embodiment solidifies the structured association between evaluation results and process evidence, avoiding evaluation that merely remains at the post-event summary level. Specifically, post-repair evaluation can be performed on work orders in a closed-loop state through the evaluation and assessment unit 6, which may include at least one of installation evaluation, trial operation evaluation, and personnel assessment, and generate assessment results associated with the work order.

[0066] Furthermore, to make the evaluation computable, this application embodiment further provides a set of process quality indicators and an evidence anomaly detection mechanism. Specifically, the evaluation unit 6 can generate a set of process quality indicators based on work order node constraints and archived datasets, and perform repeatability and anomaly detection on the evidence citation list, writing the repeatability and anomaly detection results into the rule hit field of the evaluation result. The set of process quality indicators includes key point coverage indicators, evidence consistency indicators, and node temporal consistency indicators, with the node temporal consistency indicator calculated based on evidence time index information. For example, the key point coverage indicator can be defined as:

[0067] in, The set of key points required by node constraints. This is the set of key points in the archived dataset that meet the requirements for evidence and approval; the evidence consistency index can be based on the aforementioned... Statistical analysis shows that the node timing consistency index can be used to determine whether the sequential relationship between nodes meets the timing constraints defined in the work instruction standard through evidence time index information, such as defining the timing violation count:

[0068] in, This is the set of temporal constraint relationships for nodes. These are the corresponding evidence time index information; when When the threshold is exceeded, the anomaly detection result is written to the rule hit field. Repeatability detection can construct similarity metrics for photo-based or waveform-based evidence, recording high-similarity entries to support the objectivity of personnel evaluation.

[0069] In summary, this system organizes operational monitoring data, alarm signals, defect data, and maintenance process records using a standardized maintenance data exchange system. This forms a unified maintenance dataset that can be associated with equipment, components, processes, and key points. At each work order closure point, field mapping and consistency checks are performed to generate archived datasets, achieving consistent expression and closed-loop solidification of cross-system data. By structurally associating diagnostic results with one-to-one maintenance plans, maintenance document packages, work order node constraints, process evidence, and post-maintenance evaluations within the same work order chain, the system ensures stable support of evaluation conclusions for process evidence and comparable review of historical work orders. Furthermore, by driving equipment fault model iteration through archived data, the system achieves continuous consistency and reuse of decision-making basis, improving the reliability and traceability of the maintenance management closed loop, and enhancing the credibility of evaluations and the stability of maintenance decisions.

[0070] Based on the above system structure, the method flow of this embodiment can correspond one-to-one with each unit of the system, forming a closed-loop execution chain from data access, diagnostic decision-making, work order execution, archive updates to evaluation and assessment. At the work order closing point, field mapping and consistency verification ensure that the archived dataset can be directly used for model updates and historical review. At the same time, through the verification and rollback mechanism of model version identifier and model verification rule set, the output stability during the continuous updating of the equipment fault model library is guaranteed. By constructing a life field through the life evaluation parameter set and participating in the generation of one-to-one maintenance plans, the handling suggestions are coupled with the equipment health status, ultimately improving the consistency of cross-system data and the stability of closed-loop evaluation evidence support.

[0071] like Figure 2 As shown, the following is an embodiment of an intelligent maintenance management method for wind turbine equipment provided by this disclosure. This intelligent maintenance management method for wind turbine equipment belongs to the same inventive concept as the intelligent maintenance management system for wind turbine equipment in the above embodiments. For details not described in detail in the embodiments of the intelligent maintenance management method for wind turbine equipment, please refer to the embodiments of the intelligent maintenance management system for wind turbine equipment described above.

[0072] Based on the same concept, another embodiment of this application provides an intelligent maintenance management method for wind turbine equipment, including: Step S1: Receive the operation monitoring data and alarm signals output by the remote monitoring and diagnostic platform, receive the maintenance project data and maintenance process record data entered by the user terminal, receive the defect data output by the defect system, and perform time alignment according to the maintenance data exchange standard and map according to the equipment identification code, component identification code, process identification code and key point identification code to generate a unified maintenance dataset containing key point identification code, evidence type code and time index field. Step S2: Generate diagnostic results and one-to-one repair solutions based on the unified maintenance dataset and technical standard database, and write the diagnostic results and one-to-one repair solutions into a structured output object. The diagnostic results include phenomenon field, cause field, conclusion field and treatment suggestion field. The structured output object includes evidence citation list, solution version identifier and evidence time index information. Step S3: Generate work orders and related maintenance document packages based on the one-to-one maintenance plan, form work order node constraints according to the work instruction standards, and collect maintenance process record data; Step S4: Before the work order is closed, perform integrity and consistency checks on the maintenance process record data and work order node constraints, and generate a data supplementation task; Step S5: When the data entry task is completed and the work order is closed, generate an archived dataset and update the equipment fault model library based on the defect data and the archived dataset; Step S6: Perform a post-repair evaluation on the closed-loop work order and generate the evaluation results. The post-repair evaluation includes at least one of the following: installation evaluation, trial operation evaluation, and personnel evaluation.

[0073] By adopting the above technical solution, and through time alignment and encoding mapping of operation monitoring data, alarm signals, defect data, and maintenance project data and maintenance process record data according to the maintenance data exchange standard, a unified maintenance dataset is formed, which includes key point identification codes, evidence type codes, and time index fields. During the diagnosis phase, the diagnosis results and one-to-one maintenance solutions are written into a structured output object containing evidence reference lists, solution version identifiers, and evidence time index information, realizing the structured connection of cross-system data under the same caliber and the traceability and association of evidence. By forming work order node constraints based on the work instruction standard and collecting process records, combined with the integrity and consistency verification and data supplementation tasks before the closure, the archived dataset generated when the work order is closed has stable and consistent field expressions and process evidence support. After the closure, the equipment fault model library is updated based on the defect data and the archived dataset, post-repair evaluation is performed, and the evaluation results are output, realizing the closed-loop linkage of diagnosis, handling, archiving, model iteration, and evaluation. This can improve the consistency of maintenance management data and the credibility of evaluation, enhance the comparability of historical review, and improve the stability and reusability of maintenance decision-making basis.

[0074] Hereinafter, steps S1 to S6 will be described according to embodiments of this application.

[0075] In step S1, operational monitoring data and alarm signals from the remote monitoring and diagnostic platform, maintenance project data and maintenance process record data from the user terminal, and defect data from the defect system are uniformly accessed and time-aligned and encoded according to the maintenance data exchange standard to generate a unified maintenance dataset. The maintenance data exchange standard constrains equipment identification coding, component identification coding, process identification coding, key point identification coding, evidence type coding, and time indexing rules, ensuring consistency in field scope and object granularity for data from different systems. Time alignment maps data with different sampling frequencies or different generation stages to the same time axis according to a unified time benchmark. Key point identification coding anchors monitoring and maintenance evidence to specific process checkpoints. Evidence type coding distinguishes evidence forms such as waveforms, photos, tables, or text, enabling the unified maintenance dataset to support related queries and retrospective comparisons in subsequent diagnostic, work order execution, archiving, and evaluation stages.

[0076] In step S2, diagnostic results and one-to-one maintenance plans are generated based on a unified maintenance dataset and a technical standard database, and then written into a structured output object. The technical standard database may include work instruction standards, acceptance standards, and evaluation standards, used to constrain the meaning of diagnostic fields, handling items, key point requirements, and post-repair evaluation criteria. Diagnostic results include phenomenon fields, cause fields, conclusion fields, and handling suggestion fields, used to map alarm and monitoring characteristics, defect information, and historical handling evidence into executable diagnostic outputs. One-to-one maintenance plans are generated based on diagnostic conclusions and work instruction standards, and at least include a set of procedures to be performed, a set of key points, and corresponding evidence type requirements. The structured output object includes an evidence citation list, a plan version identifier, and evidence time index information. The evidence citation list lists evidence items related to the diagnostic chain, along with their key point identifier codes and evidence type codes. The plan version identifier identifies the version of the plan generated. The evidence time index information indicates the position of the evidence on a unified timeline, ensuring that the diagnostic output can be stably referenced by work orders and post-event reviews.

[0077] In step S3, a work order and an associated maintenance document package are generated based on the one-to-one maintenance plan. Work order node constraints are formed according to the work instruction standards, and maintenance process record data is collected. The work order carries the handling process and approval workflow, while the maintenance document package carries a set of structured deliverables associated with the work order, covering the recording requirements of the maintenance preparation, implementation, summary, and evaluation stages. Work order node constraints, provided by the work instruction standards, define the mandatory fields, evidence sets, and signature fields for each node, ensuring that maintenance process record data is stored in a structured manner and remains consistent with key point identification codes, evidence type codes, and time index fields. This guarantees that process records can be stably referenced and verified in subsequent archiving and evaluation stages.

[0078] In step S4, before the work order is closed, integrity and consistency checks are performed on the maintenance process record data and work order node constraints, and a data supplementation task is generated. The integrity check verifies whether the required fields, evidence, and signatures in the node constraints are complete. The consistency check verifies whether the process record data matches the evidence reference list in the structured output object, and whether the key point identifier code, evidence type code, and time index field match the unified maintenance dataset. If the checks fail, a data supplementation task is generated and bound to the work order node to guide the supplementation location and content until the closure conditions are met, ensuring that the data quality before closure meets the requirements for archiving and model updates.

[0079] In step S5, an archived dataset is generated upon completion of the data entry task and closure of the work order. The equipment fault model library is then updated based on the defect data and the archived dataset. The archived dataset is organized for historical review, statistical analysis, and decision support. Its field definitions are consistent with the unified maintenance dataset, and key object relationships in the work order chain are solidified through field mapping and consistency checks. The equipment fault model library stores sets of models or rules related to fault identification or handling recommendations. Updates are iteratively performed based on the tag information in the defect data and the handling results, process evidence, and post-repair evaluations in the archived dataset, ensuring the model remains consistent with on-site handling results and providing a stable basis for subsequent diagnosis and solution generation.

[0080] In some embodiments of this application, a model version identifier and a model verification rule set can be configured for the equipment fault model library. Candidate equipment fault models are formed based on defect data and archived datasets, and the equipment fault models in the equipment fault model library are updated. The model verification rule set is executed and verification results are generated. When the verification results do not meet the model verification rule set, the equipment fault model is reverted to the previous model version identifier. The model verification rule set may include sample playback rules based on archived datasets and label consistency rules based on defect data.

[0081] In step S6, a post-repair evaluation is performed on the closed-loop work order, generating evaluation results. The post-repair evaluation includes at least one of installation evaluation, commissioning evaluation, and personnel evaluation. The installation evaluation reflects the quality of the maintenance implementation and compliance with key points; the commissioning evaluation reflects the compliance of the commissioning process with acceptance criteria; and the personnel evaluation reflects the quality of the work process and the standardization of records. The evaluation results are stored in association with the work order and can be combined with work order node constraints, archived datasets, and evidence citation lists to form a traceable evaluation basis, enabling the evaluation conclusions to stably support subsequent review comparisons and maintenance decisions.

[0082] In summary, this method, through time alignment and encoding mapping driven by maintenance data exchange standards, aggregates operation monitoring, alarms, defects, and maintenance process records into a unified maintenance dataset. It also solidifies diagnostic results and one-to-one maintenance solutions into structured output objects containing evidence citation lists and version identifiers. Combined with work order node constraints, pre-loop integrity and consistency verification, supplementary data entry tasks, and closed-loop archiving and model updates, it forms a structured closed loop with traceable evidence. This achieves consistent data caliber across systems and stable expression of archived data, reduces the fluctuations in work order processing and archiving quality caused by cross-system differences, enhances the stability of post-maintenance evaluation support for process evidence, and improves the comparability of historical reviews and the stability and reusability of maintenance decision-making basis.

[0083] It should be noted that, although the embodiments in this application are based on... Figure 1Steps S1 to S6 are described sequentially, but this does not mean that steps S1 to S6 must be performed in a strict order. The reason this embodiment follows this order is... Figure 1 The order in which steps S1 to S6 are described is provided to facilitate understanding of the technical solutions of the embodiments of this application by those skilled in the art. In other words, in the embodiments of this application, the order of steps S1 to S6 can be appropriately adjusted according to actual needs.

[0084] In some embodiments of this application, an intelligent maintenance management system and method for wind turbine equipment is applied to the maintenance business scenario of wind farms. The system operates in conjunction with a remote monitoring and diagnostic platform, a defect system, and user terminals, enabling operational monitoring data, alarm signals, defect data, and maintenance process records to be integrated under the same data caliber, forming an integrated closed loop of diagnosis, work orders, archiving, model updates, and evaluation. For ease of explanation, a remaining life parameter is introduced. Design life parameters Viscosity ratio parameter Cleanliness coefficient parameter Actual running time parameters Overheating frequency parameter Acceleration constant parameter Parameters of the number of times the rated current is exceeded And introduce a time index field Used for unified alignment of multi-source data.

[0085] A complete implementation process may include the following steps: Step 1: The remote monitoring and diagnostic platform continuously outputs operational monitoring data and alarm signals, the defect system outputs defect data, and the user terminal inputs maintenance project data and maintenance process record data. The data access side aggregates the above data, unifies field definitions according to the maintenance data exchange standard, and performs time alignment, binding each record with a time index field. The object mapping is completed according to the equipment identification code, component identification code, process identification code, and key point identification code, forming a field containing key point identification code, evidence type code, and time index field. A unified maintenance dataset enables alarm evidence, defect evidence, and maintenance evidence to be correlated at key points and aligned over time. For example, operational monitoring data can include time-series quantities such as vibration, temperature, and current; alarm signals can correspond to threshold exceedance events and include the event's start and end times for reference. Alignment.

[0086] Step two: The diagnostic decision-making side, based on the unified maintenance dataset, calls the technical standard database to perform diagnostic reasoning and solution generation, outputting diagnostic results and one-to-one maintenance solutions. The diagnostic results are formatted into phenomenon fields, cause fields, conclusion fields, and treatment suggestion fields. Simultaneously, the diagnostic results and one-to-one maintenance solutions are written into a structured output object. The structured output object includes an evidence citation list and a solution version identifier. The evidence citation list includes a time index field. The corresponding evidence time index information enables traceable time positioning and key point positioning of evidence such as waveforms, photos, and tables referenced in the diagnostic chain. For example, the solution version identifier can be generated by combining date and serial number to distinguish the solution differences of the same device in different alarm batches.

[0087] Step 3: Life assessment is performed concurrently with the diagnostic phase. The diagnostic decision-making side extracts the life assessment parameter set from the unified maintenance dataset and generates remaining life parameters. and will The lifespan field of the diagnostic results is written to enable its participation in the generation of a one-to-one maintenance plan. The lifespan assessment parameter set includes... , , , , , , In a feasible computational approach, the degradation acceleration term is first constructed. :

[0088] Then calculate the remaining lifetime parameters:

[0089] in, , , , Preset weighting coefficients are used to unify the contribution scale of different parameters to the degradation intensity. For example, Viscosity can be measured by oil testing. Compared with reference viscosity calculate , It can be obtained by accumulating the events that exceed the relative temperature threshold from the temperature records. It can be obtained by accumulating the over-limit events relative to the rated current threshold from the current record.

[0090] Step four: After receiving the one-to-one maintenance plan, the work order management side generates a work order and performs online approval. Simultaneously, it generates a maintenance document package associated with the work order based on the maintenance document package template library. Then, based on the work instruction standards, it forms work order node constraints, configuring each node with a set of required fields, a set of evidence, and a set of signature fields. During the maintenance implementation, it collects maintenance process record data, linking the process record data with key point identification codes, evidence type codes, and... This consistency allows on-site operational evidence to be structured and preserved according to nodes and key points. For example, a set of node evidence may include attachments corresponding to different evidence types, such as waveform screenshots, tabular records, and on-site photographs.

[0091] Step 5, closed-loop control occurs during the archiving and updating phase: Before the work order is closed, the archiving and updating side performs integrity and consistency checks on the maintenance process record data and work order node constraints. Integrity checks cover the consistency of mandatory field data for nodes, the node evidence set, and the node approval field set. Consistency checks cover the consistency between the evidence time index information in the evidence citation list and the evidence collected on-site. Alignment relationships are established; when verification fails, a data entry task bound to the work order node is generated. Once the data entry task is completed and the work order is closed, field mapping and consistency verification are performed according to the maintenance data exchange standard to generate an archived dataset. This archived data solidifies the association between diagnoses, solutions, process evidence, and evaluation objects in the work order chain using stable field definitions. For example, field mapping can standardize equipment codes from different source systems to the same equipment identification code space.

[0092] Step Six: Model and Evaluation are executed in tandem after the loop is closed: The archiving update side updates the equipment failure model library based on defect data and archived datasets, ensuring consistency between the equipment failure model and on-site handling results. Simultaneously, the evaluation and assessment side performs post-repair evaluations on closed-loop work orders and generates evaluation results associated with the work orders. Post-repair evaluations cover at least one of installation evaluations, commissioning evaluations, and personnel evaluations, ensuring that evaluation conclusions are traceably linked to process evidence within the work order chain and can be used for subsequent work order review and decision support. For example, commissioning evaluations can reference key monitoring evidence from the commissioning phase and correspond it to the evidence time index information in the evidence citation list.

[0093] Through the complete implementation process described above, the system and method utilize maintenance data exchange standards and time index fields. To achieve unified organization of multi-source data and alignment of key-point evidence, the structured output objects solidify the diagnostic evidence chain and solution versions, the work order node constraints and verification and supplementation mechanisms solidify the closed-loop archiving quality, and the archived data drives the continuous updating of equipment fault models and the traceable output of post-repair evaluations. This can improve the consistency of cross-system data integration and the stability of process evidence support, enhance the comparability of historical reviews, and improve the stability and reusability of maintenance decision-making basis.

[0094] It should be understood that the step numbers identified by "Step 1, Step 2" and other similar forms in the above embodiments are only used to distinguish different steps and do not limit the steps to be executed in the order of these numbers. The specific execution order of each step can be adjusted according to its functional requirements and the inherent logic in the actual application scenario. The above step numbers should not be interpreted as a limitation on the implementation process of the embodiments of this application.

[0095] The above-disclosed embodiments are merely preferred embodiments of the present invention, but the present invention is not limited thereto. Any non-creative variations that can be conceived by those skilled in the art, as well as any improvements and modifications made without departing from the principles of the present invention, should fall within the protection scope of the present invention.

Claims

1. An intelligent maintenance and management system for wind turbine equipment, characterized in that, include: The data access unit is used to receive operation monitoring data and alarm signals output by the remote monitoring and diagnostic platform, receive maintenance project data and maintenance process record data entered by the user terminal, receive defect data output by the defect system, and generate a unified maintenance dataset according to the maintenance data exchange standard. The standard library unit is used to store the technical standard database and the maintenance document package template library. The technical standard database includes work instruction standards, equipment failure model library, acceptance standards and evaluation standards. The diagnostic decision unit is used to generate diagnostic results and one-to-one maintenance solutions based on a unified maintenance dataset and a technical standard database. The diagnostic results include phenomenon fields, cause fields, conclusion fields, and treatment suggestion fields. The work order management unit is used to generate work orders based on one-to-one maintenance plans and perform online approval, generate maintenance document packages associated with work orders based on the maintenance document package template library, form work order node constraints based on work instruction standards, and collect maintenance process record data. The archive update unit is used to perform field mapping and consistency verification on the unified maintenance dataset according to the maintenance data exchange standard when the work order is in a closed loop, generate the archive dataset, and update the equipment fault model in the equipment fault model library based on the defect data and the archive dataset. The evaluation and assessment unit is used to perform post-repair evaluations on work orders in a closed-loop state. The post-repair evaluation includes at least one of installation evaluation, trial operation evaluation, and personnel assessment, and generates assessment results associated with the work order.

2. The intelligent maintenance and management system for wind turbine equipment as described in claim 1, characterized in that, The maintenance and repair data exchange standard includes equipment identification codes, component identification codes, process identification codes, key point identification codes, evidence type codes, and time index rules; The data access unit performs time alignment on operation monitoring data, alarm signals, maintenance project data, maintenance process record data and defect data according to time indexing rules, and performs granular mapping based on equipment identification code, component identification code, process identification code and key point identification code to generate a unified maintenance dataset containing key point identification code, evidence type code and time index field.

3. The intelligent maintenance and management system for wind turbine equipment as described in claim 2, characterized in that, After generating diagnostic results and one-to-one repair solutions, the diagnostic decision unit writes the diagnostic results and one-to-one repair solutions into a structured output object. The structured output object includes an evidence citation list and a solution version identifier. The evidence citation list includes evidence time index information corresponding to the time index field. After generating a work order, the work order management unit triggers the automatic upload of structured output objects. The automatically uploaded records include the upload time field, the work order identifier field, the solution version identifier field, and the evidence citation list.

4. The intelligent maintenance and management system for wind turbine equipment as described in claim 3, characterized in that, After the work order control unit forms work order node constraints based on the work instruction standards, it configures the set of mandatory fields for nodes, the set of evidence for nodes, and the set of approval fields for nodes. Before the work order is in a closed loop state, the archiving update unit performs integrity and consistency checks on the maintenance process record data and work order node constraints. If the checks fail, a data entry task bound to the work order node is generated. The archiving update unit generates the archived dataset when the data entry task is completed and the work order is in a closed loop state.

5. The intelligent maintenance and management system for wind turbine equipment as described in claim 4, characterized in that, When the archive update unit generates the archive dataset according to the maintenance and repair data exchange standard, it generates summary values ​​for the field set, evidence reference list and corresponding attachment data in the archive dataset and writes them into the metadata field of the archive dataset. It also associates and stores the summary values ​​with the work order identifier field, scheme version identifier field and node approval field set. When the archive dataset changes, the summary values ​​are updated and the change log is recorded.

6. The intelligent maintenance and management system for wind turbine equipment as described in claim 1, characterized in that, The archive update unit configures the model version identifier and model verification rule set for the equipment fault model library, and executes the model verification rule set and generates verification results after updating the equipment fault model. If the verification results do not meet the model verification rule set, the equipment fault model is rolled back to the previous model version identifier. The model verification rule set includes sample playback rules based on the archive dataset and label consistency rules based on defect data.

7. The intelligent maintenance and management system for wind turbine equipment as described in claim 1, characterized in that, The diagnostic decision unit is also used to generate remaining life parameters based on the life evaluation parameter set in the unified maintenance dataset, write the remaining life parameters into the life field of the diagnostic results, and use the life field as the input field for generating a one-to-one maintenance plan. The life evaluation parameter set includes design life parameters, viscosity ratio parameters, cleanliness coefficient parameters, actual operating time parameters, over-temperature number parameters, acceleration rate constant parameters, and over-rated current number parameters.

8. The intelligent maintenance and management system for wind turbine equipment as described in claim 3, characterized in that, The evaluation unit generates a set of process quality indicators based on work order node constraints and archived datasets, performs repeatability and anomaly detection on the evidence citation list, and writes the repeatability and anomaly detection results into the rule hit field of the evaluation results. The set of process quality indicators includes key point coverage indicators, evidence consistency indicators, and node temporal consistency indicators. The node temporal consistency indicators are calculated based on evidence time index information.

9. A method for intelligent maintenance and management of wind turbine equipment, characterized in that, include: It receives operation monitoring data and alarm signals output from the remote monitoring and diagnostic platform, receives maintenance project data and maintenance process record data entered by the user terminal, receives defect data output by the defect system, and performs time alignment according to the maintenance data exchange standard and maps it according to equipment identification code, component identification code, process identification code and key point identification code to generate a unified maintenance dataset containing key point identification code, evidence type code and time index field. Diagnostic results and one-to-one repair solutions are generated based on a unified maintenance dataset and a technical standard database. The diagnostic results and one-to-one repair solutions are written into a structured output object. The diagnostic results include a phenomenon field, a cause field, a conclusion field, and a treatment suggestion field. The structured output object includes an evidence citation list, a solution version identifier, and evidence time index information. Based on the one-to-one maintenance plan, work orders and maintenance document packages associated with the work orders are generated. Work order node constraints are formed according to the work instruction standards, and maintenance process record data is collected. Before the work order is closed, perform integrity and consistency checks on the maintenance process record data and work order node constraints, and generate a data supplementation task; When the data entry task is completed and the work order is closed, an archived dataset is generated, and the equipment fault model library is updated based on the defect data and the archived dataset. Post-repair evaluation is performed on closed-loop work orders and evaluation results are generated. The post-repair evaluation includes at least one of installation evaluation, trial operation evaluation and personnel evaluation.

10. The intelligent maintenance and management method for wind turbine equipment as described in claim 9, characterized in that, The steps for updating the equipment failure model library based on defect data and archived datasets include: Configure model version identifiers and model validation rule sets for the equipment fault model library; Candidate equipment failure models are generated based on defect data and archived datasets, and the equipment failure models in the equipment failure model library are updated. The model validation rule set is executed and validation results are generated. If the validation results do not meet the model validation rule set, the equipment fault model is reverted to the previous model version identifier. The model validation rule set includes sample playback rules based on archived datasets and label consistency rules based on defect data.

Images