Intelligent generation and priority allocation method for gas power plant maintenance work order

Abstract

This invention discloses an intelligent generation and priority allocation method for maintenance work orders in gas-fired power plants, belonging to the field of electronic data technology for gas-fired power plants. The method collects and preprocesses real-time operating data, historical maintenance data, and predictive model output data from the gas-fired power plant to form standardized data to be processed. Based on this data, it intelligently identifies abnormal operating conditions and fault modes of equipment, predicts development trends, and thus determines maintenance needs. It automatically generates maintenance work orders containing equipment numbers, fault descriptions, suggested measures, and required spare parts. Priority influencing factors are extracted and quantified from multiple dimensions such as safety risk, economic loss, fault severity, and urgency, and a priority score for the maintenance work order is obtained through a weighted calculation formula. This invention effectively improves the intelligence level and efficiency of maintenance management in gas-fired power plants, optimizes maintenance resource allocation, reduces operational risks and maintenance costs, and ensures the safe and stable operation of the power plant.

Description

Technical Field

[0001] This invention relates to the field of electronic data in gas-fired power plants, and in particular to a method for intelligent generation and priority allocation of maintenance work orders in gas-fired power plants. Background Technology

[0002] As a vital energy infrastructure, the stable and efficient operation of gas-fired power plants is crucial for national energy supply and economic development. Equipment maintenance is a key aspect of ensuring safe production, reducing operating costs, and extending equipment lifespan. Traditional gas-fired power plant maintenance management methods primarily employ periodic preventative maintenance and corrective maintenance after a failure. While periodic preventative maintenance can mitigate sudden failures to some extent, its strategy, based on fixed times or operating cycles, often fails to adequately consider the actual operating conditions and potential risks of the equipment, potentially leading to resource waste from over-maintenance or potential failures caused by insufficient maintenance. Corrective maintenance after a failure, on the other hand, is reactive and often results in unplanned downtime, causing significant economic losses and the risk of production interruption.

[0003] In terms of generating maintenance work orders, existing technologies mainly rely on manual experience or simple rule triggers. When equipment malfunctions or reaches a preset maintenance cycle, maintenance personnel manually fill out maintenance work orders based on inspection records, equipment alarm information, or preset plans, which are then reviewed and assigned by management personnel. This manually-driven work order generation and management method is inefficient and prone to problems such as information omissions, inaccurate fault descriptions, unreasonable allocation of maintenance tasks, and untimely preparation of spare parts, making it difficult to meet the increasingly complex and sophisticated operational needs of gas-fired power plant equipment.

[0004] Especially in prioritizing maintenance tasks, existing methods typically rely on the experience and judgment of maintenance personnel or managers, or simply prioritize tasks based on the urgency of the faults. This prioritization mechanism generally lacks scientific rigor and intelligence, failing to systematically consider multiple complex factors, such as the potential safety risk level, expected economic losses, the cascading impact of the fault on the overall operational performance of the gas-fired power plant and related equipment, and the urgency of the task. This non-optimal prioritization method easily leads to unreasonable allocation of maintenance resources, potentially causing critical equipment faults to go unaddressed, leading to more serious accidents or unplanned shutdowns, while secondary tasks consume valuable human and time resources, thereby increasing operational risks and maintenance costs, seriously affecting the safe, stable operation and economic benefits of gas-fired power plants. Therefore, existing technologies have significant shortcomings in the intelligent generation and prioritization of maintenance work orders for gas-fired power plants and urgently need improvement. Summary of the Invention

[0005] To address the aforementioned problems in the existing technology, the present invention aims to provide a method for intelligent generation and priority allocation of maintenance work orders for gas-fired power plants, comprising the following steps: Step S1, Data Acquisition and Preprocessing: Acquire and preprocess real-time operation data, historical maintenance data, prediction model output data, and external environmental data of the gas-fired power plant to form standardized data to be processed.

[0006] Step S2: Intelligent identification of maintenance needs. Based on the standardized data to be processed, the maintenance needs of the gas-fired power plant equipment are intelligently identified. The maintenance needs include abnormal operating conditions, fault modes, trend prediction and preventive maintenance needs, and the type of maintenance is determined.

[0007] Step S3: Intelligent generation of maintenance work order content. Based on the maintenance requirements, the content of the maintenance work order is intelligently generated. The content includes matching standard operating procedures and automatically filling in each field of the maintenance work order. Each field includes equipment number, fault description, suggested maintenance measures, estimated completion time, required spare parts, and required tools.

[0008] Step S4, priority influencing factor extraction and quantification, extracting and quantifying priority influencing factors from multiple dimensions, including safety risk factors, economic loss factors, fault severity factors, and urgency factors.

[0009] Step S5: Priority calculation of maintenance work orders. Substitute the safety risk factor score, the economic loss factor score, the fault severity factor score, and the urgency factor score into the priority calculation formula to calculate the priority score of the maintenance work order.

[0010] Step S6: Maintenance work order publication and feedback. The generated maintenance work order and priority score are published to the work order management system and the mobile terminals of relevant personnel, and maintenance result feedback is received to update the data.

[0011] Furthermore, step S1 includes, Step S101: Collect real-time operating data of the gas-fired power plant. The real-time operating data is acquired through the distributed control system of the gas-fired power plant and various sensors. The real-time operating data includes the operating parameters of equipment such as gas turbines, generators, waste heat boilers, steam turbines, and auxiliary systems.

[0012] Step S102: Obtain historical maintenance data of the gas-fired power plant, wherein the historical maintenance data is stored in the database of a computerized maintenance management system or an enterprise resource planning system.

[0013] Step S103: Obtain the prediction model output data and external environment data; perform data preprocessing on the collected real-time operation data, historical maintenance data, prediction model output data and external environment data to form the standardized data to be processed.

[0014] Furthermore, step S2 includes: Step S201: Based on the real-time operating data in the standardized data to be processed, abnormal operating conditions of the gas-fired power plant equipment are detected through preset rules and thresholds.

[0015] Step S202: Using the real-time operating data, historical maintenance data and prediction model output data in the standardized data to be processed, a data analysis model is constructed and applied to identify the equipment's fault modes and predict the development trend of the faults.

[0016] Step S203: Based on the maintenance cycle recommended by the gas-fired power plant equipment manufacturer, operating time, or the historical maintenance data, determine the preventive maintenance requirements for the equipment.

[0017] Step S204: Integrate the abnormal operating condition detection results, fault mode identification and trend prediction results with preventive maintenance and repair requirements, determine the type of maintenance requirement, which includes corrective maintenance, predictive maintenance or preventive maintenance, and identify the equipment components that need to be repaired.

[0018] Furthermore, the construction and application of the data analysis model includes: Based on the real-time operating data, the historical maintenance data, and the output data of the prediction model, the failure risk index of the equipment is calculated; the formula for calculating the failure risk index is: ,in The mapping function is obtained through training on the data. For the aforementioned real-time running data, This refers to the historical maintenance data. Output data for the prediction model; identify the failure mode of the equipment based on the failure risk index, and predict the development trend of the failure.

[0019] Furthermore, step S3 includes, Step S301: Match the standard operating procedure according to the maintenance requirement type, equipment number and failure mode.

[0020] Step S302: Based on the standard operating procedures, the historical maintenance data in the standardized data to be processed, and the expert knowledge base, automatically fill in each field of the maintenance work order.

[0021] Step S303: Based on the nature of the maintenance task, the equipment's historical safety records, and relevant safety procedures, intelligently generate or prompt safety precautions that need to be specially observed during the maintenance process.

[0022] Furthermore, step S4 includes, Step S401: Assess the potential risks of personal injury, environmental pollution, and secondary equipment damage involved in the maintenance work order, and extract and quantify them into safety risk factor scores.

[0023] Step S402: Evaluate the potential power generation loss due to equipment downtime involved in the maintenance work order, the increase in energy consumption caused by the decrease in efficiency of the gas-fired power plant, and the expected maintenance costs, and extract and quantify them into an economic loss factor score.

[0024] Step S403: Assess the impact of the faults involved in the maintenance work order on the overall operational stability of the gas-fired power plant, the risk of chain reactions to other related equipment, and the degree of damage to the faulty components, and extract and quantify them into fault severity factor scores.

[0025] Step S404: Assess the speed of fault development, potential for suddenness, and whether immediate shutdown is required for the fault involved in the maintenance work order, and extract and quantify it into an urgency factor score.

[0026] Furthermore, step S6 includes, Step S601: The maintenance work order and the calculated priority score are pushed to the work order management system, and can also be pushed to the mobile terminals of relevant maintenance personnel via SMS or application messages.

[0027] Step S602: Receive feedback information from the maintenance personnel regarding the maintenance work order submitted through the work order management system or the mobile terminal, and update the gas-fired power plant maintenance data and equipment ledger based on the feedback information.

[0028] Furthermore, the priority calculation formula is as follows: ,in, The final priority score for the aforementioned maintenance work order is determined. The safety risk factor score indicates the degree of potential safety risk that equipment failure or maintenance process may pose to personnel, other equipment, and the environment. The economic loss factor score represents the degree of economic loss that may result from equipment downtime or decreased operating efficiency, including power generation loss, maintenance costs, or fines. The severity factor score represents the degree of impact of equipment failure on the overall operating performance, stability, and other equipment of the gas-fired power plant. The urgency factor is scored to indicate the timeliness of the maintenance task; , , These are the weighting coefficients for the safety risk factor, economic loss factor, fault severity factor, and urgency factor, respectively, and their sum is 1.

[0029] Furthermore, the safety risk factor, the economic loss factor, the fault severity factor, and the urgency factor are all quantified as integer scores from 1 to 5; and the weighting coefficients are set based on the gas-fired power plant's management strategy, safety regulations, or operational objectives.

[0030] Furthermore, when the final priority score is greater than or equal to the first preset threshold, the maintenance work order is assigned to the "immediate processing" level; when the final priority score is less than the first preset threshold but greater than or equal to the second preset threshold, the maintenance work order is assigned to the "emergency scheduling" level.

[0031] When the final priority score is less than the second preset threshold and greater than or equal to the third preset threshold, the maintenance work order is assigned to the "planned scheduling" level; when the final priority score is less than the third preset threshold, the maintenance work order is assigned to the "delayable processing" level.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention enables automatic and accurate generation of maintenance work orders through data-driven intelligent recognition, effectively overcoming the shortcomings of traditional manual methods, such as untimely identification of maintenance needs, incomplete work order information, and low generation efficiency. By integrating real-time operational data, historical maintenance data, and predictive model output, this invention achieves accurate judgment and trend prediction of abnormal operating conditions and fault modes. This makes the generated maintenance work orders more targeted, timely, and standardized, significantly improving the automation level and initial response efficiency of maintenance management, and laying a solid foundation for subsequent maintenance work.

[0033] This invention introduces a multi-dimensional, quantitative priority allocation mechanism, overcoming the shortcomings of existing technologies that lack scientific rigor and rely on experience-based judgment. By systematically extracting and quantifying key influencing factors such as safety risks, economic losses, fault severity, and urgency, and applying a weighted calculation formula to derive an objective priority score, the allocation of maintenance resources becomes more rational and efficient. This helps gas-fired power plant managers prioritize maintenance tasks that have the greatest impact on personnel safety, equipment stability, and economic benefits, thereby minimizing unplanned downtime risks and operating costs, ensuring the continuous, safe, and stable operation of gas-fired power plants, and improving overall maintenance management and plant economic efficiency. Attached Figure Description

[0034] Figure 1 This is an exemplary flowchart of the method of the present invention.

[0035] Figure 2 This is an exemplary flowchart of the data acquisition and preprocessing steps of the present invention.

[0036] Figure 3 This is an exemplary flowchart of the maintenance requirement identification process of the present invention.

[0037] Figure 4 This is an exemplary flowchart of the steps involved in generating the maintenance work order content for this invention.

[0038] Figure 5 This is an exemplary flowchart of the priority influence factor extraction process of the present invention.

[0039] Figure 6 This is an exemplary flowchart of the steps involved in issuing a maintenance work order according to the present invention. Detailed Implementation

[0040] The present invention will be further described below with reference to specific embodiments.

[0041] This embodiment provides a method for intelligent generation and priority allocation of maintenance work orders for gas-fired power plants, such as... Figure 1 The diagram shown is an exemplary flowchart of the method in this embodiment, including the following steps: Step S1, Data Acquisition and Preprocessing: Acquire and preprocess real-time operation data, historical maintenance data, prediction model output data, and external environmental data of the gas-fired power plant to form standardized data to be processed.

[0042] like Figure 2 The above is an exemplary flowchart of the data acquisition and preprocessing steps in step S1 of this embodiment, including: Step S101: Collect real-time operating data of the gas-fired power plant. The real-time operating data is obtained through the distributed control system of the gas-fired power plant and various sensors. The real-time operating data includes the operating parameters of equipment such as gas turbine, generator, waste heat boiler, steam turbine, and auxiliary systems.

[0043] Step S102: Obtain historical maintenance data of the gas-fired power plant. The historical maintenance data is stored in the database of the computerized maintenance management system or enterprise resource planning system.

[0044] Step S103: Obtain the output data of the prediction model and the external environment data; perform data preprocessing on the collected real-time operation data, historical maintenance data, prediction model output data and external environment data to form standardized data to be processed.

[0045] Step S2: Intelligent identification of maintenance needs. Based on the standardized data to be processed, the maintenance needs of the gas-fired power plant equipment are intelligently identified. Maintenance needs include abnormal operating conditions, failure modes, trend prediction and preventive maintenance needs, and the type of maintenance is determined.

[0046] like Figure 3The diagram shown is an exemplary flowchart of step S2, maintenance requirement identification, in this embodiment, which includes: Step S201: Based on the real-time operating data in the standardized data to be processed, abnormal operating conditions of the gas-fired power plant equipment are detected using preset rules and thresholds. In one embodiment, the abnormal operating condition detection is real-time or near real-time. Specifically, the preset rules may include static thresholds set based on equipment design specifications, safe operating procedures, or historical experience. Alternatively, more advanced dynamic thresholds or anomaly detection algorithms may be used, such as control charts based on statistical process control, or machine learning models that learn normal operating modes to identify abnormal patterns deviating from normal behavior.

[0047] Step S202: Using the real-time operating data, historical maintenance data and predictive model output data from the standardized data to be processed, a data analysis model is constructed and applied to identify the equipment's failure modes and predict the development trend of the failures.

[0048] Build and apply data analysis models, including: Based on real-time operational data, historical maintenance data, and predictive model output data, the equipment failure risk index is calculated; the formula for calculating the failure risk index is: ,in The mapping function is obtained through training on the data. To run data in real time, This is historical maintenance data. Output data for the prediction model; identify equipment failure modes based on the failure risk index, and predict the development trend of failures.

[0049] In one embodiment, the mapping function G can be specifically implemented as a multi-input deep neural network capable of handling complex nonlinear relationships from different data sources.

[0050] It can be a vector containing key operating parameters of the current equipment, such as: the amplitude, frequency components, and root mean square value of the gas turbine vibration signal; real-time readings of the temperature of each stage of turbine blades and their deviation from the design value; compressor efficiency indicators; fuel flow and gas consumption rate; and the mean, standard deviation, and rate of change of these parameters over a short time window in the past.

[0051] It can be a vector containing equipment historical maintenance events, failure frequency, mean repair time, mean time between failures, operating hours since the last major overhaul, and statistical characteristics of common failure modes of similar equipment.

[0052] It can be a vector containing features such as the predicted remaining lifespan of the equipment, the probability of failure within a specific future time window, health score, and the slope or acceleration of the degradation curve of key components.

[0053] In practice, the deep neural network G is trained through supervised learning. Training data includes historical equipment operation data, corresponding historical maintenance records, and prediction data pre-generated by other prediction models. The goal of training is to enable the model to learn the mapping relationship from these multi-dimensional input features to the actual failure risk index. For example, historical severe failure events can be defined as corresponding to... The value is high, while the normal operating state corresponds to... The value is low.

[0054] Step S203: Based on the maintenance cycle, operating time, or historical maintenance data recommended by the gas-fired power plant equipment manufacturer, determine the preventive maintenance and repair requirements for the equipment.

[0055] In one embodiment, the determination of preventative maintenance requirements aims to ensure planned maintenance of equipment before failures occur. Specifically: Manufacturer-recommended maintenance cycles: Based on the operating manuals and maintenance specifications provided by the equipment manufacturer, a fixed maintenance plan is set based on time or operating cycles. Operating time: The cumulative operating hours, number of start-stop cycles, or cyclic load counts of specific components of the equipment are monitored, and maintenance requirements are triggered when preset thresholds are reached. Historical maintenance data: Statistical data such as historical average interval between failures and average repair time of similar equipment are analyzed to optimize and adjust the preventative maintenance plan to better reflect actual operating conditions. This information is integrated through a rules engine or time scheduling module to generate regular preventative maintenance tasks.

[0056] Step S204 involves integrating abnormal operating condition detection results, fault mode identification and trend prediction results, and preventive maintenance requirements to determine the type of maintenance requirement. This type may include corrective maintenance, predictive maintenance, or preventive maintenance, and the specific equipment components requiring maintenance are identified. In one embodiment, this integration process is typically implemented through a decision logic or expert system. This system prioritizes and resolves conflicts based on maintenance requirements from different sources.

[0057] Step S3: Intelligent generation of maintenance work order content. Based on maintenance requirements, the content of the maintenance work order is intelligently generated. The content includes matching standard operating procedures and automatically filling in each field of the maintenance work order. Each field includes equipment number, fault description, suggested maintenance measures, estimated completion time, required spare parts, and required tools.

[0058] like Figure 4The diagram shown is an exemplary flowchart of step S3, generating the maintenance work order content, in this embodiment, including: Step S301: Match the standard operating procedure according to the maintenance requirement type, equipment number and failure mode.

[0059] In one embodiment, the matching process is the first step in generating automated work orders. The system matches the maintenance requirement type, specific equipment number (e.g., "GT-01"), and identified fault mode determined in step S2 against a pre-defined standard operating procedure (SOP) knowledge base. Matching can employ rule-based matching, keyword matching, or more advanced natural language processing techniques for semantic matching to ensure the most relevant SOP is found.

[0060] Step S302: Based on standard operating procedures, historical maintenance data in standardized pending data, and an expert knowledge base, the system automatically fills in the various fields of the maintenance work order. In one embodiment, automatic filling is the core function of intelligent work order generation. The system automatically provides suggested values ​​or directly fills in the fields of the maintenance work order based on the standard operating procedures matched in S301, combined with the standardized pending data in S1 and the pre-established expert knowledge base: Equipment Number: Extracted directly from the real-time operating data stream or faulty equipment information identified in S2. Fault Description: Based on the abnormal operating conditions, fault modes, and predicted trends detected in S2, a concise and accurate text description is automatically generated, such as: "The gas turbine exhaust temperature remains consistently high, indicating carbon buildup inside the combustion chamber." Suggested Maintenance Measures: The corresponding maintenance steps and operating guidelines are directly extracted from the matched standard operating procedures. For predictive maintenance, specific inspection or replacement plans can be suggested based on the health status output by the predictive model. For complex faults, the expert knowledge base can provide deeper suggestions based on case-based reasoning. Estimated Completion Time: This can be estimated based on the average completion time of similar tasks in historical maintenance data, or obtained from preset man-hours in standard operating procedures (SOPs), or dynamically calculated considering current resource availability. Required Spare Parts: Based on the matched SOPs and failure modes, the system automatically identifies the required materials from the spare parts list and can interface with the inventory management system to check available inventory. Required Tools: Similarly, the system identifies a list of specialized tools required to complete the maintenance task from the SOPs.

[0061] Step S303: Based on the nature of the maintenance task, the equipment's historical safety records, and relevant safety procedures, intelligently generate or prompt safety precautions that need to be specially observed during the maintenance process.

[0062] Step S4: Extraction and quantification of priority influencing factors. Priority influencing factors are extracted and quantified from multiple dimensions, including safety risk factors, economic loss factors, fault severity factors, and urgency factors.

[0063] like Figure 5 The diagram shown is an exemplary flowchart of step S4, priority influence factor extraction, in this embodiment, including: Step S401: Assess the potential risks of personal injury, environmental pollution, and secondary equipment damage involved in the maintenance work order, and extract and quantify them into safety risk factor scores.

[0064] In one embodiment, the safety risk factor score measures the potential hazards of a maintenance task. A specific quantification method can be the risk matrix approach: assessing the probability of occurrence and severity of consequences for each risk and mapping them to a pre-defined scoring system. For example, the probability of occurrence can be categorized as "extremely low, low, medium, high, extremely high," and the severity of consequences can be categorized as "minor, moderate, severe, catastrophic." The final safety risk factor score can be obtained based on a weighted average of these assessment results or by looking up a table, quantified as an integer score from 1 to 5, where 1 represents extremely low risk and 5 represents extremely high risk. Data sources include: historical accident reports, equipment safety operating procedures, hazard identification lists, and expert assessments.

[0065] Step S402: Assess the potential power generation loss, energy consumption increase caused by the decline in efficiency of the gas-fired power plant, and expected maintenance costs that may result from equipment downtime involved in the maintenance work order, and extract and quantify them into economic loss factor scores.

[0066] Step S403: Assess the impact of the faults involved in the maintenance work order on the overall operational stability of the gas-fired power plant, the risk of cascading effects on other related equipment, and the degree of damage to the faulty components; extract and quantify these as fault severity factor scores. In one embodiment, the fault severity factor score assesses the systemic impact of the fault on the power plant system operation. In one embodiment, the urgency factor score reflects the timeliness requirements of the maintenance task. Step S404: Assess the speed of fault development, potential for suddenness, and whether immediate shutdown is required for the faults involved in the maintenance work order, and extract and quantify them into an urgency factor score.

[0067] Step S5: Priority calculation of maintenance work orders. Substitute the safety risk factor score, economic loss factor score, fault severity factor score, and urgency factor score into the priority calculation formula to calculate the priority score of the maintenance work order.

[0068] The priority calculation formula is: ,in, The final priority score for the maintenance work order; The safety risk factor score indicates the degree of potential safety risk that equipment failure or maintenance process may pose to personnel, other equipment, and the environment. The economic loss factor score indicates the extent of economic losses, such as power generation loss, maintenance costs, or fines, that may result from equipment downtime or decreased operating efficiency. The severity factor score represents the degree of impact of equipment failure on the overall operating performance, stability, and other equipment of the gas-fired power plant. The urgency factor score indicates the timeliness of the maintenance task; , , These are the weighting coefficients for the safety risk factor, economic loss factor, fault severity factor, and urgency factor, respectively, and their sum is 1.

[0069] Safety risk factors, economic loss factors, fault severity factors, and urgency factors are all quantified into integer scores ranging from 1 to 5; and the weighting coefficients are set based on the gas-fired power plant's management strategy, safety regulations, or operational objectives.

[0070] When the final priority score is greater than or equal to the first preset threshold, the maintenance work order will be assigned to the "immediate processing" level; when the final priority score is less than the first preset threshold but greater than or equal to the second preset threshold, the maintenance work order will be assigned to the "emergency scheduling" level.

[0071] When the final priority score is less than the second preset threshold but greater than or equal to the third preset threshold, the maintenance work order will be assigned to the "planned scheduling" level; when the final priority score is less than the third preset threshold, the maintenance work order will be assigned to the "delayable processing" level.

[0072] In one embodiment, the threshold setting is determined based on the power plant's actual operating experience and risk appetite. For example, assume the priority score P ranges from 1 to 5.

[0073] First preset threshold: For example, P>=4.5 indicates extremely high priority, requiring immediate attention, typically corresponding to risks that may lead to personal injury, serious environmental accidents, or significant equipment damage. Second preset threshold: For example, 3.0<=P<4.5 indicates high priority, requiring urgent scheduling, typically corresponding to risks that may lead to equipment downtime, significant economic losses, or short-term worsening of the fault. Third preset threshold: For example, 1.5<=P<3.0 indicates medium priority, which can be included in planned scheduling, typically corresponding to preventative maintenance or faults with minimal short-term impact. Lowest priority: For example, when P<1.5, it indicates low priority, which can be deferred, typically corresponding to minor defects or warnings that do not affect equipment operation. These thresholds are configurable and can be dynamically adjusted based on the power plant's operating strategy, seasonal demand, and unforeseen events.

[0074] Step S6, Maintenance work order release and feedback: The generated maintenance work order and priority score are released to the work order management system and the mobile terminals of relevant personnel, and maintenance result feedback is received to update the data.

[0075] like Figure 6 The diagram shown is an exemplary flowchart of step S6, the issuance of a maintenance work order, in this embodiment, including: Step S601: The maintenance work order and the calculated priority score are pushed to the work order management system, and can also be pushed to the mobile terminals of relevant maintenance personnel via SMS or application messages.

[0076] In one embodiment, the push process is automated. The system synchronizes the maintenance work orders intelligently generated by S3 and the priority scores calculated by S5 to the power plant's existing work order management system in real time via API interface or data integration module. Simultaneously, for work orders of different priorities, the system automatically identifies the maintenance team and personnel responsible for the equipment or area according to preset allocation rules, and sends the newly generated work order information and priority scores to the relevant personnel's mobile terminals via SMS, push notifications from a dedicated mobile application, email, etc. For example, a "handle immediately" work order might trigger an emergency call or a loud alarm, while a "scheduled" work order might only be notified through the app's message list.

[0077] Step S602: Receive feedback information from maintenance personnel submitting maintenance work orders through the work order management system or mobile terminal, and update the gas-fired power plant maintenance data and equipment ledger based on the feedback information. In one embodiment, the feedback mechanism is an important component of the system's self-learning and optimization. After completing a maintenance task, maintenance personnel will submit detailed feedback information through the work order management system or its associated mobile terminal. This feedback information includes, but is not limited to: the completion status of the work order (completed, in progress, delayed), actual start and end times, actual man-hours consumed, the list and quantity of spare parts actually used, the confirmed root cause of the fault, the specific maintenance measures taken, the equipment performance recovery status after maintenance, whether there are any remaining problems, and any additional textual descriptions or photographic evidence.

[0078] The above description is merely an example and illustration of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the scope defined by the invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for intelligent generation and priority allocation of maintenance work orders for gas-fired power plants, characterized by: Includes the following steps, Step S1, Data Acquisition and Preprocessing: Acquire and preprocess real-time operation data, historical maintenance data, predictive model output data, and external environmental data of the gas-fired power plant to form standardized data to be processed; Step S2, intelligent identification of maintenance needs: Based on the standardized data to be processed, intelligently identify the maintenance needs of the gas-fired power plant equipment. The maintenance needs include abnormal operating conditions, failure modes, trend prediction and preventive maintenance needs, and determine the type of maintenance. Step S3: Intelligent generation of maintenance work order content. Based on the maintenance requirements, the content of the maintenance work order is intelligently generated. The content includes matching standard operating procedures and automatically filling in each field of the maintenance work order. Each field includes equipment number, fault description, suggested maintenance measures, estimated completion time, required spare parts, and required tools. Step S4, Priority Influence Factor Extraction and Quantification: Priority influence factors are extracted and quantified from multiple dimensions, including safety risk factors, economic loss factors, fault severity factors, and urgency factors. Step S5: Priority calculation of maintenance work order. Substitute the safety risk factor score, the economic loss factor score, the fault severity factor score, and the urgency factor score into the priority calculation formula to calculate the priority score of the maintenance work order. Step S6: Maintenance work order publication and feedback. The generated maintenance work order and priority score are published to the work order management system and the mobile terminals of relevant personnel, and maintenance result feedback is received to update the data.

2. The intelligent generation and priority allocation method for maintenance work orders in gas-fired power plants according to claim 1, characterized in that: Step S1 includes, Step S101: Collect real-time operating data of the gas-fired power plant. The real-time operating data is obtained through the distributed control system of the gas-fired power plant and various sensors. The real-time operating data includes the operating parameters of equipment such as gas turbine, generator, waste heat boiler, steam turbine, and auxiliary systems. Step S102: Obtain historical maintenance data of the gas-fired power plant, wherein the historical maintenance data is stored in the database of a computerized maintenance management system or an enterprise resource planning system. Step S103: Obtain the prediction model output data and external environment data; The collected real-time operating data, historical maintenance data, prediction model output data, and external environmental data are preprocessed to form the standardized data to be processed.

3. The intelligent generation and priority allocation method for maintenance work orders in gas-fired power plants according to claim 1, characterized in that: Step S2 includes: Step S201: Based on the real-time operating data in the standardized data to be processed, abnormal operating conditions of the gas-fired power plant equipment are detected through preset rules and thresholds. Step S202: Using the real-time operating data, the historical maintenance data, and the prediction model output data in the standardized data to be processed, a data analysis model is constructed and applied to identify the equipment's fault modes and predict the development trend of the faults. Step S203: Based on the maintenance cycle recommended by the gas-fired power plant equipment manufacturer, operating time, or the historical maintenance data, determine the preventive maintenance and repair requirements of the equipment. Step S204: Integrate the abnormal operating condition detection results, fault mode identification and trend prediction results with preventive maintenance and repair requirements, determine the type of maintenance requirement, which includes corrective maintenance, predictive maintenance or preventive maintenance, and identify the equipment components that need to be repaired.

4. The intelligent generation and priority allocation method for maintenance work orders in gas-fired power plants according to claim 3, characterized in that: The construction and application of the data analysis model includes: Based on the real-time operating data, the historical maintenance data, and the output data of the prediction model, the failure risk index of the equipment is calculated; the formula for calculating the failure risk index is: ,in The mapping function is obtained through training on the data. For the aforementioned real-time running data, This refers to the historical maintenance data. Output data for the prediction model; identify the failure mode of the equipment based on the failure risk index, and predict the development trend of the failure.

5. The intelligent generation and priority allocation method for maintenance work orders in gas-fired power plants according to claim 1, characterized in that: Step S3 includes... Step S301: Match the standard operating procedure according to the type of maintenance requirement, equipment number and failure mode; Step S302: Based on the standard operating procedures, the historical maintenance data in the standardized data to be processed, and the expert knowledge base, automatically fill in each field of the maintenance work order; Step S303: Based on the nature of the maintenance task, the equipment's historical safety records, and relevant safety procedures, intelligently generate or prompt safety precautions that need to be specially observed during the maintenance process.

6. The method for intelligent generation and priority allocation of maintenance work orders for gas-fired power plants according to claim 1, characterized in that: Step S4 includes, Step S401: Assess the potential risks of personal injury, environmental pollution, and secondary damage to equipment involved in the maintenance work order, and extract and quantify them into safety risk factor scores; Step S402: Evaluate the potential power generation loss due to equipment downtime involved in the maintenance work order, the increase in energy consumption caused by the decrease in efficiency of the gas-fired power plant, and the expected maintenance costs, and extract and quantify them into an economic loss factor score. Step S403: Assess the impact of the faults involved in the maintenance work order on the overall operational stability of the gas-fired power plant, the risk of chain reactions to other related equipment, and the degree of damage to the faulty components, and extract and quantify them into fault severity factor scores; Step S404: Assess the speed of fault development, potential for suddenness, and whether immediate shutdown is required for the fault involved in the maintenance work order, and extract and quantify it into an urgency factor score.

7. The intelligent generation and priority allocation method for maintenance work orders in gas-fired power plants according to claim 1, characterized in that: Step S6 includes, Step S601: The maintenance work order and the calculated priority score are pushed to the work order management system, and can also be pushed to the mobile terminals of relevant maintenance personnel via SMS or application messages. Step S602: Receive feedback information from the maintenance personnel regarding the maintenance work order submitted through the work order management system or the mobile terminal, and update the gas-fired power plant maintenance data and equipment ledger based on the feedback information.

8. The method for intelligent generation and priority allocation of maintenance work orders for gas-fired power plants according to claim 1, characterized in that: The priority calculation formula is as follows: ,in, The final priority score for the aforementioned maintenance work order is determined. The safety risk factor score indicates the degree of potential safety risk that equipment failure or maintenance process may pose to personnel, other equipment, and the environment. The economic loss factor score represents the degree of economic loss that may result from equipment downtime or decreased operating efficiency, including power generation loss, maintenance costs, or fines. The severity factor score represents the degree of impact of equipment failure on the overall operating performance, stability, and other equipment of the gas-fired power plant. The urgency factor is scored to indicate the timeliness of the maintenance task; , , These are the weighting coefficients for the safety risk factor, economic loss factor, fault severity factor, and urgency factor, respectively, and their sum is 1.

9. The intelligent generation and priority allocation method for maintenance work orders in gas-fired power plants according to claim 8, characterized in that: The safety risk factor, the economic loss factor, the fault severity factor, and the urgency factor are all quantified as integer scores from 1 to 5; and the weighting coefficients are set based on the gas-fired power plant's management strategy, safety regulations, or operational objectives.

10. The method for intelligent generation and priority allocation of maintenance work orders for gas-fired power plants according to claim 8, characterized in that: When the final priority score is greater than or equal to the first preset threshold, the maintenance work order is assigned to the "immediate processing" level; when the final priority score is less than the first preset threshold but greater than or equal to the second preset threshold, the maintenance work order is assigned to the "emergency scheduling" level. When the final priority score is less than the second preset threshold and greater than or equal to the third preset threshold, the maintenance work order is assigned to the "planned scheduling" level; when the final priority score is less than the third preset threshold, the maintenance work order is assigned to the "delayable processing" level.

Images