New energy power generation equipment fault maintenance intelligent scheduling system and method

Through the intelligent scheduling system for troubleshooting of new energy power generation equipment, using technologies such as Markov decision-making model and Weibull-gamma distribution model, the uneven level of maintenance teams and difficult to control during equipment maintenance of new energy stations is solved. The problem has achieved accurate prediction of equipment failures and optimal maintenance decisions, reduced operation and maintenance costs and improved equipment reliability and operation efficiency.

CN118941273AInactive Publication Date: 2024-11-12HUADIAN ELECTRIC POWER SCI INST CO LTD +1

Patent Information

Application Number
CN202411043219.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

During the maintenance of new energy station equipment, there are problems such as uneven maintenance team level, difficult to implement supervision of maintenance process, and difficult to control the quality of maintenance, resulting in a decrease in equipment reliability, an increase in maintenance costs, and an increase in operating risks.

Method used

An intelligent scheduling system for fault maintenance of new energy power generation equipment was designed, including maintenance decision-making module, fault maintenance strategy library, task planning and scheduling module, and optimization module, using Markov decision-making model, Weibull-gamma distribution model, AI technology, big data analysis and multi-source data fusion achieve accurate prediction of equipment failures and optimal overhaul decision-making.

Benefits of technology

Through intelligent maintenance strategies and decision-making models, over-inspection and under-inspection are reduced, operation and maintenance costs are reduced, equipment reliability and operation efficiency are improved, operation and maintenance efficiency and power output are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941273A_ABST
    Figure CN118941273A_ABST
Patent Text Reader

Abstract

The invention particularly relates to an intelligent scheduling system and method for fault maintenance of new energy power generation equipment, and the system comprises a maintenance decision module, a fault maintenance strategy library, a task planning and scheduling module, and an optimization module. The fault maintenance strategy library is based on big data and is fused with a fault tree and think tank mechanism information of fan design to form a final expert knowledge base, and the task planning and scheduling module is used for intelligently scheduling defect work orders, special maintenance work orders and maintenance work orders of a new energy station, quantitatively analyzing maintenance tasks to be executed, and determining whether the maintenance tasks are to be executed or not. And the optimization module is used for sensing influence factors of field maintenance in multiple dimensions, determining maintenance task execution priorities by means of an intelligent algorithm, and optimizing a new energy equipment fault maintenance scheme library and an algorithm model.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of new energy station operation and maintenance, and specifically relates to a new energy power generation equipment fault inspection and maintenance intelligent scheduling system and method. Background Art

[0002] With the continuous increase in the installed capacity of new energy, the problems faced in the maintenance of new energy station equipment are becoming increasingly prominent, such as: uneven residual levels of the maintenance team, difficulty in implementing supervision of the maintenance process, and difficulty in controlling the quality of maintenance. Based on solving the problems faced in the maintenance process of new energy station equipment, meeting the actual needs of wind and solar station maintenance management, focusing on business functions such as maintenance strategy optimization and maintenance result evaluation, and relying on multi-source data analysis, we will carry out the construction of an intelligent maintenance system.

[0003] This section builds a maintenance model that is more in line with new energy equipment, grasps the equipment's operating status and reliability in real time, maintains or restores the equipment's inherent reliability level at the lowest cost, reduces its operating and maintenance costs, and provides a basis for formulating the best maintenance strategy. Establish an intelligent maintenance organization system that supports company groups, regional companies, power plant companies, and teams, and units at all levels carry out intelligent maintenance work based on their responsibilities.

[0004] At present, domestic new energy equipment, mainly wind power and photovoltaic power, still adopts the traditional maintenance method of post-maintenance and preventive maintenance, which will inevitably lead to "over-maintenance and under-maintenance", resulting in reduced equipment reliability, increased maintenance costs, and increased operating risks. Therefore, it is urgent to build a maintenance model that is more in line with new energy equipment, grasp the equipment operation status and reliability status in real time, maintain or restore the inherent reliability level of the equipment at the lowest cost, so as to reduce its operation and maintenance costs, and provide a basic basis for formulating the optimal maintenance strategy. Establish an intelligent maintenance organization system to support the company headquarters (Electrical Science Institute), regional companies, power plant companies and teams, and units at all levels carry out intelligent maintenance work according to the responsibilities of intelligent maintenance work. The project is based on the real-time production and operation data assets of all stations and regions, and conducts targeted research on typical defects and difficult faults, combining the project's innovations and advanced features in multi-level system linkage, low-code model R&D management, and data asset management. Summary of the invention

[0005] The purpose of the present invention is to provide a new energy power generation equipment fault inspection and maintenance intelligent scheduling system and method to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] The intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment includes a maintenance decision module, a fault inspection and maintenance strategy library, a task planning and scheduling module, and an optimization module. The maintenance decision module is used to determine the optimal maintenance decision for equipment components under various operating conditions and the next maintenance, and intelligently push the equipment maintenance strategy. The fault inspection and maintenance strategy library is based on big data and uses AI technology to mine from the industry's stock work order system, and is integrated with the fault tree and think tank mechanism information of the wind turbine design to form the final expert knowledge base. The task planning and scheduling module is used to intelligently schedule defect work orders, special maintenance work orders, and maintenance work orders for new energy stations, and quantitatively analyze these maintenance tasks to be executed, perceive the influencing factors of on-site maintenance in multiple dimensions, and determine the execution priority of maintenance tasks with the help of intelligent algorithms. The optimization module is used to optimize the fault inspection and maintenance solution library and algorithm model of new energy equipment.

[0008] Furthermore, the maintenance decision module is based on the Markov decision model. The shape parameters and scale parameters of the component failure distribution are estimated by the least squares method, and the Weibull-Gamma distribution model of the component failure is established to obtain the reliability and failure curve of the component. The component operating status is divided based on the component reliability curve. At the same time, according to the change of component reliability before and after maintenance, the maintenance cost of the component under various maintenance decisions is calculated, and the insulation degradation curve of the electrical equipment is obtained through the uniform distribution model. On this basis, relying on the Markov maintenance decision model and the strategy iteration algorithm, the optimal maintenance decision of the equipment components under each operating state and the next maintenance are determined, and the equipment maintenance strategy is intelligently pushed to achieve accurate maintenance of the equipment or system, reduce maintenance costs, and avoid excessive and insufficient maintenance. The maintenance decision module also uses the constructed intelligent knowledge base tool to uniformly collect, process and apply the knowledge materials of operation and maintenance. By least squares method The process of estimating the shape and scale parameters of the component failure distribution and then building the Weibull-Gamma distribution model: There are n

[0009] t1, t2, ..., t n

[0010] Failure time data point;

[0011] The Weibull-Gamma distribution is determined by two parameters: shape parameter α and scale parameter β. Transformation,

[0012] ln(t i )

[0013] Logarithmize the failure time data to obtain

[0014] In the logarithmic coordinate system, the cumulative distribution function of the Weibull distribution can be expressed as ln(-ln(1-F(t)))=αln (t)-αln(β), where F(t) is the cumulative distribution function; for (ln(t i ), ln(-ln(1-F(t i ))))) Linear regression of data points Normalize to get the slope and intercept;

[0015] Parameter estimation: The slope of the linear regression is the estimate of the shape parameter α, and the negative value of the intercept is the estimate of ln(β). Calculate the value, and get β = e -截距 .

[0016] Furthermore, the fault repair strategy library includes a defect data dictionary maintenance module, a repair plan library maintenance module, an operation instruction module, and a repair process card;

[0017] The defect data dictionary maintenance module is used to maintain the basic definition information of defects, unify the descriptions of the same defect by business personnel, and provide standard words for system users;

[0018] The maintenance module of the maintenance plan library is used to cluster the facts and concepts formed after knowledge management, provide a theoretical basis and decision-making basis for equipment maintenance, enable maintenance personnel to quickly identify and troubleshoot equipment failures, share failure handling experience, and associate failure cases with the equipment structure tree in equipment management to classify and manage all failures;

[0019] The work instruction module is used to formulate safety and quality control measures and implement them during the operation process;

[0020] The maintenance process card is compiled by integrating the troubleshooting before maintenance, maintenance ticket processing and confirmation, spare parts confirmation, disassembly and inspection process, technical requirements for each part, maintenance progress, maintenance acceptance, and test run performance acceptance.

[0021] Furthermore, the functional design of the defect data dictionary maintenance module includes:

[0022] Defect level: used to establish or modify the defect type of the unit, which is divided into Class I defect, Class II defect, Class III defect, Class IV defect, or into emergency defect and general defect, and set different processing period requirements for different types of defects;

[0023] Defect content: If the defects that may occur in a certain type of equipment are relatively regular, define the defect content and define the possible defects into the system. Once a defect is found, enter the defect content through quick selection;

[0024] Equipment classification: Obtain equipment classification and equipment information in defect management from the equipment ledger system. Each defect information of the equipment is recorded in the equipment ledger system for comparative analysis of the equipment operation status and equipment quality.

[0025] Furthermore, the functional design of the maintenance scheme library maintenance module includes:

[0026] Maintenance of maintenance plan library: Based on experience and common equipment failures, a maintenance plan library for wind farms and photovoltaic stations is established. All fault cases are associated with the equipment structure tree in equipment management. The maintenance plan library records each type of equipment at any time, including: fault type, fault name, equipment manufacturer, fault phenomenon, number of fault occurrences, fault cause, fault handling plan, and fault analysis report. It supports the uploading of fault phenomenon pictures and fault analysis reports. When on-site personnel find equipment failures, they can quickly search for similar failures in the maintenance plan library based on the manufacturer, fault type, and fault phenomenon of the equipment. By consulting the fault phenomenon, handling plan, and fault analysis report of related fault cases and comparing the actual equipment failure conditions, they can quickly identify the cause of the equipment failure and provide a reference plan for solving actual equipment failures. The failure cases include: photovoltaic component failure, wind turbine failure, power transmission failure, inverter failure, and control system failure.

[0027] At the same time, the system can retrieve information on troubleshooting processes, preventive measures, and emergency response plans to ensure that problems can be solved quickly and effectively when they occur;

[0028] There are two ways to enter fault cases in the maintenance plan library. The first is to summarize and refine major equipment failure cases that occur in the industry and enter them into the system regularly. The second is to enter detailed event records, processing procedures, and result analysis into the maintenance plan library after major failures of on-site equipment are handled. The maintenance plan generated in this way needs to be reviewed and approved before it can take effect.

[0029] The modules involved in the maintenance plan library include: maintenance plan information, maintenance plan attachment information, maintenance plan spare parts information, maintenance plan tool information, maintenance plan operation instruction information, and maintenance plan process card information;

[0030] Fault case search:

[0031] When special maintenance occurs, the corresponding maintenance plan needs to be quickly retrieved from the system. The system has set up quick search conditions to facilitate maintenance personnel to quickly locate from the maintenance plan library. The set search conditions include: applicable equipment type, applicable equipment model, applicable equipment location, and applicable defect type.

[0032] Furthermore, the task planning and scheduling module includes:

[0033] Data fusion and analysis submodule: This module is responsible for integrating multi-source online and offline data, including unit SCADA operation data, equipment operating years, historical maintenance records, and historical defect records. It uses the coupling analysis method to determine the influence weights of key factors and analyzes their impact on the accuracy of intelligent scheduling. It supports transfer learning technology for model iteration and dynamic optimization to meet the needs of different scenarios at sea and on land.

[0034] Comprehensive scheduling model construction submodule: Based on the regional data platform, it integrates equipment, software and information data to build a unified and coordinated intelligent scheduling system, realize automatic sorting and selection of operation and inspection task priorities, optimize work processes, improve work efficiency, and support flexible adjustment of scheduling results on site;

[0035] Defect task management module: specializes in handling defect work orders of new energy stations, including the receipt, classification, priority setting and related data analysis of work orders, to ensure that defect problems are dispatched and resolved in a timely and effective manner;

[0036] Maintenance plan management module: responsible for the intelligent scheduling of special maintenance work orders and regular maintenance work orders, sorting and recommending tasks based on task urgency and importance parameters, supporting the generation, adjustment and tracking of plans, and ensuring that maintenance activities are carried out efficiently and orderly;

[0037] Defect statistical analysis module: conducts in-depth analysis of historical defect data, identifies failure modes, predicts potential problems, provides data support for preventive maintenance, and provides a quantitative basis for scheduling decisions;

[0038] Task execution module: realizes automatic scheduling and execution tracking of tasks, supports real-time monitoring of task progress, dynamic allocation of resources and handling of abnormal situations, and ensures that tasks proceed smoothly as planned;

[0039] Maintenance evaluation module: After the task is completed, the execution data is collected and the maintenance effect is evaluated, including time cost, material consumption, and power generation recovery, providing feedback for future task optimization and strategy adjustment;

[0040] Resource and Constraint Management Module: Integrates intelligent maintenance resource management and operating environment constraint monitoring functions to ensure reasonable resource allocation, safe and compliant operations, and dynamically adjusts scheduling strategies based on real-time environmental data;

[0041] Decision support and intelligent recommendation module: Based on the decision factor library, combined with real-time data and historical experience, it provides intelligent suggestions and warnings for operation and maintenance decisions, and enhances the flexibility and adaptability of scheduling;

[0042] System integration and expansion module: Provides standard interfaces and protocols to ensure seamless integration of the system with the external environment, supports horizontal and vertical expansion, meets the customized needs of sites of different sizes, and ensures data security and privacy protection.

[0043] Furthermore, the intelligent scheduling method for fault inspection and maintenance of new energy power generation equipment includes the following steps:

[0044] S1. Build an intelligent scheduling platform for task dispatching: integrate multi-source online and offline data such as unit SCADA operation data, equipment operation years, historical maintenance records, and historical defect records, couple and analyze the weight of key factors affecting work type, unit location, climate environment, transportation, personnel skill level, communication conditions, spare parts, and spot trading price information, analyze the impact of multi-parameter and multi-factor change trends on the accuracy of intelligent scheduling, apply transfer learning, conduct research on indicator iteration and model dynamic optimization algorithms, and train different scheduling models for new energy applications in different scenarios at sea and on land;

[0045] S2. Through functional integration, data integration and business integration of existing systems, various separation equipment, software and information data elements are integrated into a unified and coordinated intelligent scheduling system based on the regional data platform, a comprehensive scheduling model is constructed, and the priority of operation and inspection tasks is automatically sorted and selected to optimize the workflow;

[0046] S3. Intelligent scheduling of defect work orders, special maintenance work orders, and maintenance work orders for new energy stations, with the multi-dimensional goals of minimizing power loss and maximizing maintenance efficiency, quantitative analysis of maintenance tasks to be executed, multi-dimensional perception of the influencing factors of on-site maintenance, and determination of maintenance task execution priority with the help of intelligent algorithms;

[0047] S4. Automatically arrange the execution order of operation and maintenance tasks according to the set rules and priorities, and automatically adjust task arrangements and priorities based on historical data and trend forecasts.

[0048] Furthermore, the defect work order content in S3 includes the work order serial number, area name, station name, equipment name, equipment number, defect location, defect type, defect source / discoverer, discovery time, status, and defect level;

[0049] The contents of a maintenance work order include the work order serial number, work order number, work order name, work order type, person who dispatches the order, dispatch time, work order status, two-ticket approval information of the work order, spare parts, operating instructions, and a maintenance process card; the contents of a special maintenance work order include the work order serial number, special maintenance work order number, work order name, work order type, person who dispatches the order, dispatch time, work order status, two-ticket approval information of the work order, spare parts, operating instructions, and a maintenance process card.

[0050] The beneficial effects of this application are specifically reflected in the following aspects:

[0051] 1. Improve operation and maintenance efficiency and accuracy: Through maintenance strategies based on artificial intelligence models and big data intelligent decision-making models, the accuracy of equipment failure prediction and optimization of maintenance decisions are achieved. Using the Markov decision process and Weibull-Gamma distribution model, the reliability and failure curve of the equipment are modeled, which effectively guides the maintenance timing of the equipment, reduces over-maintenance and under-maintenance, reduces operation and maintenance costs, and improves the operating efficiency and power output of the equipment.

[0052] 2. Knowledge management and sharing optimization: The intelligent knowledge base tool built uniformly collects, processes, and applies various types of operation and maintenance knowledge, including unstructured, semi-structured, and structured knowledge, greatly improving the utilization and dissemination efficiency of knowledge. Operation and maintenance personnel can quickly acquire fault handling experience, improve problem solving speed, promote cross-regional knowledge sharing, and enhance the overall operation and maintenance capabilities of the enterprise.

[0053] 3. Standardization and intelligence of fault repair strategies: The establishment of a fault repair strategy library, combined with big data and AI technology, provides a comprehensive and accurate fault solution for direct-drive wind turbines and photovoltaic power generation equipment, covering the entire process from fault identification to processing. Through the standardized maintenance solution library, operation instructions, and maintenance process cards, the maintenance process is standardized and intelligent, the accuracy and efficiency of fault handling are improved, and human errors are reduced.

[0054] 4. Intelligent scheduling and resource optimization: The multi-objective optimization of intelligent new energy station operation and maintenance task planning and scheduling strategy realizes automatic priority sorting and intelligent scheduling of tasks by comprehensively considering multi-source data and multi-factor influences, optimizes the workflow, and improves work efficiency. The intelligent scheduling system is highly flexible and adaptable, and can adjust strategies according to different scenarios and needs, ensuring the efficient execution of operation and maintenance tasks and the maximum utilization of resources.

[0055] 5. Decision support and risk management: Through decision factor management and restricted condition management, the system can intelligently identify operating condition restrictions, adjust maintenance strategies in a timely manner, reduce operating risks, and ensure operating safety. At the same time, dynamic maintenance and management of decision-making influencing factors enable the system to continuously optimize and adapt to complex and changing operation and maintenance environments.

[0056] 6. Continuous optimization of fault knowledge and algorithm models: Through in-depth analysis of historical fault data and the application of machine learning technology, the fault inspection and repair solution library and algorithm models are continuously optimized, so that the system can more accurately predict and respond to future equipment failures, promoting the continuous progress and innovation of operation and maintenance technology.

[0057] To sum up, the technical solution of this application has significantly improved the intelligence level of new energy station operation and maintenance through a series of intelligent and data-driven methods, enhanced the reliability of equipment, optimized resource allocation, improved operation and maintenance efficiency, and reduced operation and maintenance costs, which is of great significance to promoting the efficient and sustainable development of the new energy industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is the architecture diagram of the intelligent maintenance business in the present invention.

[0059] Figure 1 The symbols in are interpreted as follows:

[0060] A—remote monitoring; B—tiered diagnosis; C—operation management; D—two-ticket system; E—technical supervision; F—training system; G—benchmarking evaluation; H—equipment asset management; I—intelligent inspection; J—intelligent maintenance; K—safety management; L—spare parts; M—others.

[0061] 1—The remote monitoring module inputs the intelligent maintenance module, and the main content is defect registration;

[0062] 2—The hierarchical diagnosis module inputs the intelligent maintenance module, and the main content is defect registration;

[0063] 3—Technical supervision module inputs intelligent maintenance module, the main content of which is defect registration;

[0064] 4—The equipment asset management module inputs the intelligent maintenance module, which mainly contains the basic data of the maintenance equipment;

[0065] 5—Intelligent inspection module inputs intelligent maintenance module, the main content is defect registration;

[0066] 6—Safety management module input intelligent maintenance module, mainly used for safety management of the whole maintenance process;

[0067] 7—The spare parts module inputs the intelligent maintenance module, which is mainly used for the management of tools and required spare parts during the maintenance process;

[0068] 8—The power prediction module inputs the intelligent maintenance module, and its main contents include meteorological information, which is used for maintenance work arrangement;

[0069] 9—The power market module inputs the intelligent maintenance module, and its main contents include electricity price transaction information, which is used for maintenance work arrangement;

[0070] 10—Mobile APP (H5) module input intelligent maintenance module, mainly used for on-site maintenance work;

[0071] 11—Intelligent maintenance module outputs operation management module, the main content of which is work order information;

[0072] 12—Intelligent maintenance module outputs technical supervision module, the main content of which is the result of regular maintenance application, which is used for partial supervision plan closed loop;

[0073] 13—The intelligent maintenance module outputs the equipment asset management module, which mainly includes the work packages and documents generated during the maintenance period, which are used to generate defect and maintenance records;

[0074] 14—The intelligent maintenance module outputs the spare parts module, the main content of which is the spare parts usage of the maintenance results, which is used for demand forecasting.

[0075] ①—Regular inspection plan is issued and planned tasks are executed;

[0076] Data collection management unit②—Issuance of defect elimination tasks and execution of on-site defect elimination tasks;

[0077] ③—The maintenance plan library generates maintenance plans according to the task type and guides the execution of on-site work;

[0078] ④—After the defect elimination task is completed and accepted, the defect elimination information is entered into the defect statistics analysis;

[0079] ⑤—After the maintenance task is completed, a maintenance evaluation is carried out. DETAILED DESCRIPTION

[0080] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0081] This application is applicable to the application of wind turbines and photovoltaic equipment for the new energy smart production management platform. This application includes: maintenance strategies based on artificial intelligence models, intelligent decision-making models based on big data; general fault maintenance strategies for direct-drive wind turbines and photovoltaic power generation equipment; research on multi-objective optimization of intelligent new energy station operation and maintenance task planning and scheduling strategies; optimization of new energy equipment fault maintenance solution library and algorithm model; deployment, optimization and debugging of related algorithms and functions, which is applicable to the application and construction of the new energy smart production management platform.

[0082] The business logic relationship diagram of the intelligent maintenance module of this application and other modules and the logical relationship diagram between the sub-function modules within the intelligent maintenance are as follows: Figure 1 As shown;

[0083] Specifically, it includes a connected data acquisition management module, an intelligent maintenance module, and a background maintenance module. The data acquisition management module includes a remote monitoring module, a hierarchical diagnosis module, a technical supervision module, an equipment asset management module, an intelligent inspection module, a safety management module, a spare parts module, a power prediction module, a power market module, and a mobile APP module. The intelligent maintenance module includes a defect task management module, a maintenance plan management module, a maintenance plan library module, a defect statistical analysis module, a task execution module, and a maintenance evaluation module. The background maintenance module includes an operation management module, a technical supervision module, and a spare parts module;

[0084] The remote monitoring module, hierarchical diagnosis module, technical supervision module and intelligent inspection module of the data acquisition management module are respectively input into the intelligent maintenance module for defect registration;

[0085] The equipment asset management module inputs the intelligent maintenance module to inspect the basic data of the equipment;

[0086] The safety management module is input into the intelligent maintenance module for safety management of the entire maintenance process

[0087] The spare parts module is input into the intelligent maintenance module for the management of tools and required spare parts during the maintenance process;

[0088] The power prediction module is input into the intelligent maintenance module, and the power prediction module includes meteorological information for maintenance work arrangement;

[0089] The power market module inputs the intelligent maintenance module, and the power market module includes power price transaction information for maintenance work arrangement;

[0090] The mobile APP module is input into the intelligent maintenance module for use in on-site maintenance work;

[0091] The intelligent maintenance module outputs the operation management module, which includes work order information;

[0092] The intelligent maintenance module outputs the technical supervision module, which includes the results of regular maintenance applications and is used to close the loop of partial supervision plans;

[0093] The intelligent maintenance module outputs the equipment asset management module, which includes the work packages and documents generated during the maintenance period, and is used to generate defect and maintenance records;

[0094] The intelligent maintenance module outputs the spare parts module, which includes the spare parts usage with maintenance results for demand forecasting;

[0095] The defect task management module of the intelligent maintenance module is connected to the task execution module for issuing scheduled inspection plans and executing planned tasks;

[0096] The maintenance plan management module is connected to the task execution module for issuing fault elimination tasks and executing fault elimination tasks on site;

[0097] The maintenance plan library module is connected to the task execution module, and is used to generate maintenance plans according to the task type and guide the execution of on-site work;

[0098] The task execution module is connected to the defect statistics analysis module, and is used to input the defect elimination information into the defect statistics analysis after the defect elimination task is completed and accepted;

[0099] The task execution module is connected to the maintenance evaluation module to perform maintenance evaluation after the maintenance task is completed.

[0100] This application conducts research on intelligent maintenance strategy library and decision-making model for direct-drive wind turbines and photovoltaic equipment in new energy stations. Intelligent maintenance decision-making focuses on providing maintenance decision-making capabilities for efficient, accurate and economical fault elimination. That is, when a defect occurs in the equipment, how to accurately assess the impact of the defect and accurately analyze the details of the defect. After the defect registration is completed, the system generates a defect work order based on the defect registration content. The defect work order field content includes: relevant information when the defect is logged in, automatically generates the required tools and spare parts list, work instructions, and maintenance process cards according to the defect type, and links to the two-ticket system. A set of intelligent maintenance work modes to guide operators to conduct safe and high-quality fault elimination processes.

[0101] The intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment provided by the present invention comprises a maintenance decision module, a fault inspection and maintenance strategy library, a task planning and scheduling module, and an optimization module.

[0102] The maintenance decision-making module is a maintenance strategy based on artificial intelligence models and an intelligent decision-making model based on big data; the fault maintenance strategy library includes a general fault maintenance strategy library for direct-drive wind turbines and photovoltaic power generation equipment; the task planning and scheduling module is a multi-objective optimized intelligent new energy station operation and maintenance task planning and scheduling strategy; the optimization module is a new energy equipment fault maintenance solution library and algorithm model optimization.

[0103] Specifically, this application mainly includes the following functional modules:

[0104] 1. Maintenance decision module - maintenance strategy based on artificial intelligence model, intelligent decision model based on big data

[0105] Markov decision process (MDP) is an important concept in artificial intelligence and one of the theoretical foundations of reinforcement learning. Based on the Markov decision model, the shape parameters and scale parameters of the component failure distribution are estimated by the least squares method, and the Weibull-Gamma distribution model of component failure is established to obtain the reliability and failure curve of the component. The operating status of the component is divided based on the reliability curve of the component. At the same time, according to the change of component reliability before and after maintenance, the maintenance cost of the component under various maintenance decisions is calculated. The insulation degradation curve of electrical equipment is obtained by developing a uniform distribution model. On this basis, relying on the Markov maintenance decision model and the strategy iteration algorithm, the optimal maintenance decision of the equipment components under each operating state and the next maintenance are determined, and the equipment maintenance strategy is intelligently pushed, so as to finally achieve accurate maintenance of equipment or system, reduce maintenance cost, and avoid excessive maintenance and under-maintenance.

[0106] The shape parameters and scale parameters of component failure distribution are estimated by the least squares method, and then the Weibull-Gaussian The process of the Ma distribution model:

[0107] Data on component failure times (in hours) representing the time to first failure of different components The goal is to estimate the parameters of the Weibull-Gamma distribution from this data in order to understand component reliability and predict Future failure.

[0108] Data preparation: First, we have n failure time data points t1, t2, ..., t n .

[0109] The Weibull-Gamma distribution is a lifespan model that is determined by two parameters: the shape parameter and The probability density function (PDF) of the Weibull distribution is:

[0110]

[0111] Where t≥0,α>0,β>0.

[0112] We log-transform the failure data because the cumulative distribution function (CDF) of the Weibull distribution is in the logarithmic coordinates. This makes it easier for us to estimate parameters using the least squares method.

[0113] Logarithmic transformation: Logarithmize the failure time data to obtain ln(t i ) . Fitting line: In logarithmic coordinate system, Weibull The CDF of the Möller distribution can be expressed as ln(-ln(1-F(t)))=αln(t)-αln(β), where F(t) is the cumulative distribution function. (ln(t i ), ln(-In(1-F(t i ))))Perform linear regression on the data points and get the slope and intercept.

[0114] Parameter estimation: The slope of the linear regression is the estimate of the shape parameter α, and the negative value of the intercept is the estimate of ln(β). Calculate the value, and get β = e -截距 .

[0115] Specific mathematical process Assume that we have n logarithmically transformed failure time data points ln(t1), ln(t2), ..., ln (t n ), fitting this These points are connected to the straight line y = αx + ln(-ln(1-F)), where y = ln(-ln(1-F(t))) and x = ln(t).

[0116] The goal of the least squares method is to minimize the residual sum of squares, that is:

[0117] To simplify the calculation, we know that F(t i ) corresponds to each t i Accumulation Distribution function value. Then, we solve α to minimize the above error.

[0118] In practice, computing α and β requires iterative or numerical solutions such as gradient descent or direct use of the α and β values ​​provided by statistical packages. The optimization function.

[0119] Once α and β are estimated, we can use these parameters to build component reliability curves and failure curves. Divide the operating status and calculate the maintenance cost under different maintenance decisions. For example, by comparing the expected Failure cost and maintenance cost, find the optimal maintenance time point.

[0120] Through the construction of intelligent knowledge base tools, the knowledge materials rich in operation and maintenance are uniformly collected, processed and applied, such as:

[0121] Unstructured knowledge: analysis reports, work instructions, case collections

[0122] Semi-structured knowledge: SOP

[0123] Structured Knowledge: Frequently Asked Questions (FAQ)

[0124] 2. Fault troubleshooting strategy library - general fault troubleshooting strategy library for direct-drive wind turbines and photovoltaic power generation equipment

[0125] The fault inspection strategy library is based on big data, using AI technology to mine from the industry's stock work order system, and is integrated with the fault tree, think tank and other mechanism information of the fan design to form the final expert knowledge base. The knowledge base must contain 23 types of entity information, including fan model, fault name, fault number, primary cause of fault, secondary cause of fault, fault solution, tools, consumables, preventive measures, reference work instructions, reference analysis reports, expert summary experience, etc.

[0126] The maintenance knowledge base mainly includes defect data dictionary, maintenance plan library, operation instructions, and maintenance process card functional modules, and a dedicated person is required to perform regular maintenance on the maintenance knowledge base.

[0127] Defect Data Dictionary

[0128] Functional description:

[0129] The defect data dictionary maintenance module provided by the system can maintain the basic definition information of defects, which is conducive to the application and standardization of basic information by system maintenance personnel. When basic information needs to be changed, the changes to the global application module can take effect immediately. At the same time, creating a defect data dictionary can unify the description of the same defect by business personnel, provide standard words for system users, and improve the professional terminology and daily work efficiency of business personnel.

[0130] Functional design:

[0131] (1) Defect level

[0132] It is used to establish or modify the defect types of this unit. For example, they can be divided into Class I defects, Class II defects, Class III defects, Class IV defects, or emergency defects and general defects. Different processing deadlines can be set for different types of defects.

[0133] (2) Defect content

[0134] If the defects that may occur in a certain type of equipment are relatively regular, the defect content can be defined. That is, the possible defects are defined in the system, and once a defect is found, the defect content can be entered through quick selection.

[0135] (3) Equipment classification

[0136] Equipment classification and equipment information in defect management should be obtained from the equipment inventory system. Each defect information of the equipment will also be recorded in the equipment inventory system to facilitate comparative analysis of the equipment operation status and equipment quality.

[0137] Maintenance solution library

[0138] Functional description:

[0139] (1) Maintenance plan library maintenance

[0140] The maintenance solution library is a knowledge cluster of facts and concepts formed through comprehensive and organized knowledge management. It is the theoretical basis for equipment maintenance and the basis for intelligent decision-making. Rich maintenance solutions enable maintenance team members in the area to quickly identify and troubleshoot equipment failures, and share fault handling experience within the group.

[0141] Based on experience and common equipment failures, a maintenance solution library for wind farms and photovoltaic stations is established, and all failure cases are associated with the equipment structure tree in equipment management to facilitate the classification management of all failures.

[0142] The maintenance plan library records each type of equipment at any time, including:

[0143] Fault type, fault name, equipment manufacturer, fault phenomenon, number of fault occurrences, fault cause, fault handling plan, fault analysis report, support for uploading fault phenomenon pictures and fault analysis reports.

[0144] (2) Fault case retrieval

[0145] When on-site personnel discover equipment failure, they can quickly search for similar failures in the maintenance solution library based on the equipment manufacturer, failure type, and failure phenomenon. By consulting the failure phenomenon, treatment plan, and failure analysis report of related failure cases and comparing with the actual equipment failure situation, they can quickly identify the cause of the equipment failure and provide a reference solution for solving the actual equipment failure.

[0146] Failure cases include: photovoltaic module failure, wind turbine failure, power transmission failure, inverter failure, control system failure, etc.

[0147] At the same time, the system can retrieve information such as troubleshooting processes, preventive measures, emergency response plans, etc. to ensure that problems can be solved quickly and effectively when failures occur, thereby ensuring the stable operation of the new energy power generation system.

[0148] Functional design:

[0149] (1) Maintenance plan library maintenance

[0150] There are two ways to enter fault cases into the maintenance plan library. The first is to summarize and refine major equipment failure cases that occur in the industry and enter them into the system regularly. The second is to enter detailed event records, processing procedures, and result analysis into the maintenance plan library after major failures of on-site equipment are handled. The maintenance plan generated in this way needs to be reviewed and approved before it can take effect.

[0151] The modules involved in the maintenance solution library include:

[0152] Maintenance plan information: Please check the page elements for specific field information.

[0153] Maintenance plan attachment information: Please check the page elements for specific field information.

[0154] Maintenance plan spare parts information: Please check the page elements for specific field information.

[0155] Maintenance plan tool information: Please check the page elements for specific field information.

[0156] Maintenance plan work instruction information: Please check the page elements for specific field information.

[0157] Maintenance plan process card information: Please check the page elements for specific field information.

[0158] (2) Fault case retrieval

[0159] When special maintenance occurs, the corresponding maintenance plan needs to be quickly retrieved from the system. The system has set up quick retrieval conditions to facilitate maintenance personnel to quickly locate from the maintenance plan library.

[0160] The search conditions include: applicable equipment type, applicable equipment model, applicable equipment location, and applicable defect type.

[0161] Work Instructions

[0162] Functional description:

[0163] Standardized management of on-site operations in the process of power production is one of the effective ways to identify dangerous sources, control their dangers and avoid accidents. Standardized operation management of power production is to analyze the danger points, center on the needs of personal safety, equipment safety, process and quality control of the operation application, formulate corresponding safety and quality control measures and implement them in the operation process based on the relevant provisions such as production safety regulations, safety management system, anti-accident measures, equipment maintenance process guidelines and construction and acceptance applications, and take the organization, technology and safety management of the operation process as the center to analyze the danger points and take the organization, technology and safety management of the operation process as the center to formulate corresponding safety and quality control measures.

[0164] Functional design:

[0165] (1) Custom maintenance

[0166] The system provides a custom maintenance function for work instructions. The work instruction maintenance content includes:

[0167] Maintenance of basic information of work instructions: version number, drafter, instruction name, department, purpose, and scope.

[0168] Reference file information maintenance: record, upload and download reference files in work instructions.

[0169] Maintenance of preparatory work information: preparatory work content and responsible unit.

[0170] Safety measures information maintenance: hazard point analysis, hazard measures.

[0171] Personnel requirement information maintenance: job types, personnel responsibilities, qualification requirements, and number of personnel.

[0172] Spare parts information maintenance: material name, specification model, unit, quantity.

[0173] Tool information maintenance: tool name, specification, unit, and quantity.

[0174] Process card information maintenance: You can select from the maintenance process card list.

[0175] Involving work information maintenance: two votes.

[0176] (2) Review of work instructions

[0177] After the drafter of the work instruction has finished editing, it needs to be submitted for review. It cannot take effect before passing the review, and the maintenance personnel cannot search and match the work instruction in their daily work. Only after it is reviewed and approved by a dedicated person can it be applied and recommended in the system.

[0178] (3) Automatic maintenance recommendation

[0179] When maintaining the work instructions, you need to set the applicable scope, such as equipment model and specific maintenance parts. When the system performs maintenance operations such as fault elimination and special maintenance, the system can recommend applicable work instructions based on the maintenance equipment. In order to increase the intelligent recommendation of maintenance work instructions, the applicable scope can be customized to increase the recommendation judgment conditions.

[0180] After the maintenance personnel obtain the list of maintenance work instructions automatically recommended by the system, they can preview the instructions and match the work instructions that meet the requirements according to the actual work.

[0181] (4) Import / Export of Work Instructions

[0182] Since the work instructions involve a lot of maintenance content, the system provides a work instruction import function to facilitate the daily management and maintenance of the document.

[0183] Maintenance process card

[0184] Functional description:

[0185] In order to strengthen the supervision of the equipment maintenance process and strictly control the quality of key points, we rely on the technical parameters of the equipment drawings and combine the actual maintenance situation on site to integrate the pre-maintenance problem investigation, maintenance ticket processing and confirmation, spare parts confirmation, disassembly and inspection process, technical requirements of each part, maintenance progress, maintenance acceptance, test run performance acceptance and other contents into a maintenance process card. The application of maintenance process cards not only puts an end to the previous maintenance mode of checking drawings and handwriting records during on-site maintenance, but also puts forward higher standards and requirements for each detail of the entire maintenance process. It is an important measure to improve equipment maintenance management.

[0186] Functional design:

[0187] (1) Custom maintenance

[0188] The system provides a custom maintenance function for the maintenance process card. The maintenance content of the maintenance process card includes:

[0189] Maintenance of basic information of maintenance process card: process card number, process card name, equipment name, equipment component, department, specialty, team, and added person.

[0190] Maintenance of maintenance process card contents: operation classification, operation sequence number, operation process contents and process standards, and execution confirmation.

[0191] (2) Review of maintenance process card

[0192] After the maintenance process card is edited, it needs to be submitted for review. It cannot take effect before it passes the review. The maintenance personnel cannot search and match the maintenance process card in their daily work. It can only be applied in the system after it is reviewed and approved by a dedicated person.

[0193] (3) Automatic maintenance recommendation

[0194] When maintaining the maintenance procedure card, the applicable equipment and equipment components must be clearly identified. When the system performs maintenance operations such as fault elimination and special maintenance, the system can recommend maintenance procedure cards based on the maintenance equipment.

[0195] After the maintenance personnel obtain the list of maintenance process cards automatically recommended by the system, they can preview the maintenance process cards and match the maintenance process cards that meet the requirements according to the actual work.

[0196] (4) Import / Export of maintenance process cards

[0197] Since the maintenance process card involves more maintenance content, the system provides the maintenance process card import function, and also provides export and modification to facilitate daily management and maintenance of documents.

[0198] The fault knowledge base must provide more than 100 fault knowledge (for example, one "anemometer working abnormally" fault knowledge must include: fault name, fault number, solution (including: solutions for multiple fault causes), cause analysis, model, wind turbine capacity, trigger conditions and other information). The maintenance strategy library should at least include a corresponding number of defect modes of direct-drive units and photovoltaic power generation system components, and must cover at least 80% of the frequent and major fault modes of mainstream units; as well as the maintenance strategy library corresponding to the warning of direct-drive units and photovoltaic power generation systems.

[0199] The fault knowledge base requirements are as follows:

[0200]

[0201]

[0202] 3. Task planning and scheduling module - multi-objective optimization of intelligent new energy station operation and maintenance task planning and scheduling strategy

[0203] Build an intelligent scheduling system for task dispatching. Integrate multi-source online and offline data such as unit SCADA operation data, equipment operating years, historical maintenance records, historical defect records, etc., couple and analyze the weights of key factors such as work type, unit location, climate environment, transportation, personnel skill level, communication conditions, spare parts, and spot trading price information, analyze the impact of multi-parameter and multi-factor change trends on the accuracy of intelligent scheduling, apply transfer learning technology, carry out research on indicator iteration and model dynamic optimization algorithms, and at the same time, train different scheduling models for new energy applications in different scenarios such as offshore and land (plains and mountains).

[0204] By integrating the functions, data and business of the existing systems, various separation equipment, software, information data and other elements are integrated into a unified and coordinated intelligent scheduling system based on the regional data platform, a comprehensive scheduling model is constructed, and the priority of operation and maintenance tasks is automatically sorted and selected to optimize the workflow and improve work efficiency. At the same time, the scheduling results pushed by the platform can be modified on site based on actual conditions.

[0205] Smart scheduling is the core intelligent application of smart maintenance, responsible for the intelligent scheduling of defect work orders, special maintenance work orders, and maintenance work orders for new energy stations. With the multi-dimensional goals of minimizing power loss and maximizing maintenance efficiency, it quantitatively analyzes these maintenance tasks to be executed, perceives the influencing factors of on-site maintenance in multiple dimensions, and uses intelligent algorithms to determine the priority of maintenance task execution, thereby increasing power generation and improving maintenance efficiency.

[0206] Smart scheduling should have the ability to automate and automatically arrange the execution order of maintenance tasks according to the set rules and priorities. At the same time, it should also have intelligent features and be able to automatically adjust task scheduling and priorities based on historical data and trend forecasts to achieve better maintenance efficiency and quality.

[0207] Smart scheduling should be flexible and scalable to adapt to different scenarios and needs. It should support multiple task types, priorities, scheduling strategies, and be able to be customized and expanded as needed.

[0208] Smart scheduling should have high reliability and stability, be able to reliably perform tasks under different circumstances, and be able to handle abnormal situations. It should have self-healing and fault-tolerant mechanisms to ensure the normal operation of the system.

[0209] Smart scheduling should be efficient and real-time, able to quickly execute tasks and respond to events. It should be able to handle a large number of tasks and events, and be able to make decisions and adjustments in a short time.

[0210] Smart scheduling should be integrated and open, and be able to integrate and interact with other systems or modules. It should support standard interfaces and protocols so that it can be seamlessly integrated with other systems.

[0211] 3.1 Intelligent Scheduling of Maintenance Tasks

[0212] The intelligent scheduling system needs to support scheduling of maintenance tasks for onshore wind power, onshore photovoltaic, offshore wind power and other scenarios, and also needs to meet the following technical requirements:

[0213] Multiple resource adaptation: In order to meet the maintenance needs of different scenarios, the system should support multiple resource adaptation, including different types of wind turbines, power transmission and transformation equipment, photovoltaic power generation equipment and other new energy station equipment. In addition, it is also necessary to support the scheduling of maintenance tasks for other auxiliary equipment to ensure the stability and reliability of the entire system.

[0214] Task priority management: For different maintenance tasks, the system should have a task priority management function. Based on factors such as the urgency and importance of the task, the background algorithm should recommend and sort the tasks to achieve more efficient task execution. This helps improve operational efficiency and reduce operation and maintenance costs.

[0215] Visual display: In order to facilitate the operation and maintenance personnel to understand the task execution status, the system should have a visual display monitoring function. Through intuitive interfaces and charts, it should be able to display maintenance task work orders, maintenance task importance, station distribution, wind speed, task distribution, severe weather, personnel distribution, warehouse distribution and other information. In addition, it should also be able to display station weather information in real time on the map to help maintenance personnel better understand the on-site weather conditions.

[0216] Intelligent recommendations and suggestions: In addition to scheduling and managing tasks, the system should also be able to provide intelligent recommendations and suggestions to maintenance personnel based on historical data and current status. For example, before a task is executed, the system can remind maintenance personnel to pay attention to relevant equipment parameters, weather conditions and other information; after the task is executed, the system can analyze and evaluate the maintenance results and provide references and suggestions for subsequent maintenance plans.

[0217] Integration and scalability: In order to meet the needs of different users and adapt to the ever-changing environment, the system should have good integration and scalability. It should be able to integrate with other systems to achieve data sharing and interaction; at the same time, it should support horizontal and vertical expansion to adapt to operation and maintenance tasks of different scales and types. This helps to improve the flexibility and maintainability of the system.

[0218] Security and privacy protection: The intelligent operation and maintenance task scheduling system involves a large amount of sensitive information and important data, so it should meet security and privacy protection requirements. A strict security access control mechanism should be set up to ensure the security and stability of the system.

[0219] User-friendly interface: To facilitate user use, the system should have a concise and easy-to-operate user interface. The interface should clearly display various functions and operations, and provide necessary help and prompt information. This helps improve user work efficiency and user experience.

[0220] Customizable according to customer needs: In response to different user needs and usage scenarios, customized development of interfaces, functions and operating processes should be supported according to user needs, which will help improve the adaptability and ease of use of the system.

[0221] 3.2 Maintenance task connection

[0222] Supports maintenance task data obtained from the platform.

[0223] Maintenance tasks include but are not limited to defect work orders, special maintenance work orders, maintenance work orders and other maintenance tasks.

[0224] The contents of the defect work order include the work order serial number, area name, site name, equipment name, equipment number, defect location, defect type, defect source / discoverer, discovery time, status, defect level and other information.

[0225] The contents of the maintenance work order include the work order serial number, work order number, work order name, work order type, person who dispatched the order, dispatch time, work order status, two-ticket approval information of the work order, spare parts, operation instructions, maintenance process card and other information.

[0226] The content of a special maintenance work order includes the work order serial number, special maintenance work order number, work order name, work order type, person who dispatches the order, dispatch time, work order status, two-ticket approval information of the work order, spare parts, operation instructions, maintenance process card and other information.

[0227] 3.3 Maintenance resource management

[0228] It can obtain information such as weather, power trading, location of tasks, personnel resources and skills from systems and modules such as intelligent maintenance systems, wind power prediction systems, power trading, equipment management, and data bases.

[0229] The data quality requirements for maintenance resource docking shall be implemented in accordance with the company's overall data quality application. The data storage architecture shall comply with the company's master data architecture design and requirements.

[0230] 3.4 Restricted Condition Management

[0231] It provides management of restricted operating personnel, restricted operating wind speed, and restricted operating window. When a maintenance task triggers a restricted condition, it can clearly indicate the restricted condition triggered by the current task.

[0232] It can make intelligent judgments on whether the maintenance tasks to be performed on the day are restricted, and feed back the judgment results to the intelligent maintenance system.

[0233] 3.5 Decision Factor Management

[0234] It can provide maintenance and management of decision-making influencing factors, including the number of standard operating personnel, personnel skills, wind speed restrictions, extreme weather, operating time, distance, occurrence time, wind speed, failure frequency, window period, power loss, task expiration, etc. (decision-making factors cover remote monitoring, hierarchical diagnosis, technical supervision, ledger system, intelligent inspection, safety management, spare parts, electricity market, power forecasting, mobile APP modules, etc.).

[0235] The environmental conditions of offshore wind farms have a great impact on operation and maintenance tasks, so the intelligent scheduling algorithm should comprehensively consider environmental factors such as wind speed, wave height, and weather, and dynamically adjust according to actual conditions. The scheduling algorithm should be able to improve operation and maintenance efficiency based on task requirements and resource availability.

[0236] 3.6 Scheduling Strategy Management

[0237] It can configure and manage the scheduling strategies for various scenarios such as onshore wind power, onshore photovoltaic power, and offshore wind power. Different strategies can be configured for different stations to adapt to the actual management characteristics of the station. After the strategy configuration of the station is successful, the latest strategy can be used to make decisions during the next scheduling.

[0238] 4. Optimization module - new energy equipment fault inspection and repair solution library and algorithm model optimization

[0239] Collect, organize and analyze historical fault data of new energy equipment, including fault type, occurrence time, impact range, etc., and identify the patterns and trends of fault occurrence through data mining and machine learning technology. Use AI-based knowledge engineering technology to automatically batch process work order knowledge, document types, drawings and images, etc. to generate new knowledge. After expert verification, knowledge application and maintenance strategy optimization and update can be directly performed.

[0240] This application proposes the minimum technical requirements, does not stipulate all technical details, and does not fully quote the provisions of relevant standards and regulations. During implementation, it should ensure the provision of high-quality products and corresponding services that meet industry standards and the requirements of this application and have complete functions and excellent performance. At the same time, it must meet the national mandatory standards and application requirements for safety, environmental protection, etc. If the standards used in this application are inconsistent with the standards implemented during implementation, the higher standards shall be implemented.

[0241] From scheme design to project implementation, this system requires strict compliance with the "Twenty-Five Key Requirements for Preventing Power Production Accidents (2023 Edition)".

[0242] In order to prevent hackers and malicious codes from attacking the power monitoring system and causing power system accidents, and to ensure the safe and stable operation of various monitoring systems for power generation and transmission and transformation equipment, this system must strictly abide by the principles of the "Electric Power Monitoring System Safety Protection Regulations" (National Energy Safety

[2015] No. 36) from scheme design to project implementation.

[0243] This application is mainly implemented with reference to the following standards, but is not limited to them: high-quality products and related services that meet the standards listed in this application should be provided during implementation, and the requirements of relevant national mandatory standards must be met. The standards used must be the latest versions. If there are conflicts between these standards, they should be implemented in accordance with the terms of the highest standard or the standards agreed upon by both parties. If standards other than those specified in this technical agreement are selected during implementation, proof that the replacement standard is equivalent to or better than the standard specified in the technical agreement must be submitted and confirmed by the purchaser.

[0244] Software Standards

[0245] Information processing - data flow chart, program flow chart, system flow chart,

[0246] GB 1526

[0247] GB 4943 Safety of information technology equipment

[0248] GB 4943 Safety of information technology equipment (including electrical business equipment) GB 8566 Application for computer software development

[0249] GB 8567 Application for the preparation of computer software product development documents GB 9385 Application for the preparation of computer software requirements specification

[0250] GB 18030 Information technology - Chinese coded character set for information exchange GB / T 11457 Software engineering terminology

[0251] GB / T 12505 Application for Computer Software Configuration Management Plan

[0252] GB / T 13702 Computer software classification and code

[0253] GB / T 15532 Computer software unit testing

[0254] GB / T 15629.3 Regulations of the People's Republic of China on Computer Information Security ProtectionTIA / EIATSB-67 Unshielded Twisted Pair UTP End-to-End System Function Test StandardISO / IEC 11801 International Standards Organization Structured Cabling Standard

[0255] DL / Z 981 Power system control and its communication data and communication security State Electricity Regulatory Commission Order No. 5 Provisions on the Safety Protection of Power Secondary Systems China Huadian Corporation New Energy Remote Diagnosis Platform System Safety Design Manual, etc.

[0260] Power Standards

[0261] Twenty-five key requirements for preventing power production accidents

[0262] GB 26860 Electricity Safety Work Regulations (Electrical Parts of Power Plants and Substations)

[0263] Electrical installation engineering panel, cabinet and secondary circuit wiring construction and inspection GB 50171

[0264] Receive applications

[0265] GB 50174 Application for Design of Computer Room

[0266] GB / T 2423 Environmental testing for electrical and electronic products

[0267] GB / T 2887 General application for computer sites

[0268] GB / T 9813 Microcomputer General Application

[0269] GB / T 13729 Telecontrol terminal equipment

[0270] GB / T 13730 Regional power grid dispatching automation system

[0271] GB / T 15532 Computer software unit testing

[0272] GB / T 17626 Electromagnetic compatibility test and measurement technology

[0273] GB / T 25385 Operation and maintenance requirements for wind turbine generator sets

[0274] GB / T 32128 Offshore wind farm operation and maintenance regulations

[0275] NB / T 10595 Technical Guidelines for Intelligent Maintenance of Wind Farms

[0276] NB / T 10918 Technical Guidelines for Smart Wind Farms

[0277] DL / T 393 Condition maintenance test procedures for power transmission and transformation equipment

[0278] DL / T 572 Power Transformer Operation Procedures

[0279] DL / T 573 Guidelines for Maintenance of Power Transformers

[0280] DL / T 596 Preventive test procedures for power equipment

[0281] DL / T 664 Application for infrared diagnosis of live equipment

[0282] DL / T 666 Wind farm operation regulations

[0283] DL / T 722 Guidelines for analysis and determination of dissolved gases in transformer oil

[0284] DL / T 796 Safety regulations for wind farms

[0285] DL / T 797 Wind farm maintenance regulations

[0286] DL / T 1684 Guidelines for condition inspection and maintenance of oil-immersed transformers (reactors)

[0287] T / IEIA 0003 Maintenance procedures for electrical equipment in photovoltaic power stations

[0288] Regulations on Safety Protection of Power Secondary Systems

[0289] China Huadian Corporation Limited Power Generation Enterprises Technical Transformation Management Measures

[0290] China Huadian Corporation Hydropower and New Energy Maintenance Management Measures (Trial)

[0291] The intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment of the present invention mainly includes: ① maintenance strategy based on artificial intelligence model and intelligent decision-making model based on big data; ② general fault inspection and maintenance strategy library for direct-drive wind turbines and photovoltaic power generation equipment (including high-incidence faults, major faults, and regular maintenance strategies; mainly including: case library, fault library, knowledge base, etc.); ③ research on multi-objective optimization of intelligent new energy station operation and maintenance task planning and scheduling strategy; ④ new energy equipment fault inspection and maintenance solution library and algorithm model optimization; ⑤ deployment, optimization and debugging of related algorithms and functions.

[0292] During implementation, the technical capability requirements for implementation personnel include:

[0293] (1) Have basic technical capabilities and organizational system for algorithm and software system development;

[0294] (2) Have the ability to develop microservice architecture application systems;

[0295] (3) Have the ability to use and integrate time series databases;

[0296] (4) Have the ability to develop visualization-related applications;

[0297] (5) Have the ability to develop applications related to big data systems.

[0298] In summary, this application has the following beneficial effects:

[0299] 1. Improve operation and maintenance efficiency and accuracy: Through maintenance strategies based on artificial intelligence models and big data intelligent decision-making models, the accuracy of equipment failure prediction and optimization of maintenance decisions are achieved. Using the Markov decision process and Weibull-Gamma distribution model, the reliability and failure curve of the equipment are modeled, which effectively guides the maintenance timing of the equipment, reduces over-maintenance and under-maintenance, reduces operation and maintenance costs, and improves the operating efficiency and power output of the equipment.

[0300] 2. Knowledge management and sharing optimization: The intelligent knowledge base tool built uniformly collects, processes, and applies various types of operation and maintenance knowledge, including unstructured, semi-structured, and structured knowledge, greatly improving the utilization and dissemination efficiency of knowledge. Operation and maintenance personnel can quickly acquire fault handling experience, improve problem solving speed, promote cross-regional knowledge sharing, and enhance the overall operation and maintenance capabilities of the enterprise.

[0301] 3. Standardization and intelligence of fault repair strategies: The establishment of a fault repair strategy library, combined with big data and AI technology, provides a comprehensive and accurate fault solution for direct-drive wind turbines and photovoltaic power generation equipment, covering the entire process from fault identification to processing. Through the standardized maintenance solution library, operation instructions, and maintenance process cards, the maintenance process is standardized and intelligent, the accuracy and efficiency of fault handling are improved, and human errors are reduced.

[0302] 4. Intelligent scheduling and resource optimization: The multi-objective optimization of intelligent new energy station operation and maintenance task planning and scheduling strategy realizes automatic priority sorting and intelligent scheduling of tasks by comprehensively considering multi-source data and multi-factor influences, optimizes the workflow, and improves work efficiency. The intelligent scheduling system is highly flexible and adaptable, and can adjust strategies according to different scenarios and needs, ensuring the efficient execution of operation and maintenance tasks and the maximum utilization of resources.

[0303] 5. Decision support and risk management: Through decision factor management and restricted condition management, the system can intelligently identify operating condition restrictions, adjust maintenance strategies in a timely manner, reduce operating risks, and ensure operating safety. At the same time, dynamic maintenance and management of decision-making influencing factors enable the system to continuously optimize and adapt to complex and changing operation and maintenance environments.

[0304] 6. Continuous optimization of fault knowledge and algorithm models: Through in-depth analysis of historical fault data and the application of machine learning technology, the fault inspection and repair solution library and algorithm models are continuously optimized, so that the system can more accurately predict and respond to future equipment failures, promoting the continuous progress and innovation of operation and maintenance technology.

[0305] To sum up, the technical solution of this application has significantly improved the intelligence level of new energy station operation and maintenance through a series of intelligent and data-driven methods, enhanced the reliability of equipment, optimized resource allocation, improved operation and maintenance efficiency, and reduced operation and maintenance costs, which is of great significance to promoting the efficient and sustainable development of the new energy industry.

Claims

1. Intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment, characterized by: It includes a maintenance decision module, a fault maintenance strategy library, a task planning and scheduling module, and an optimization module. The maintenance decision module is used to determine the optimal maintenance decision for equipment components under various operating conditions and intelligently push equipment maintenance strategies for the next maintenance. The fault maintenance strategy library is generated from the industry's existing work order system based on big data, and is integrated with the fault tree and think tank mechanism information of the wind turbine design to form the final expert knowledge base. The task planning and scheduling module is used to intelligently schedule defect work orders, special maintenance work orders, and maintenance work orders for new energy sites, and quantitatively analyze these maintenance tasks to be executed, perceive the influencing factors of on-site maintenance in multiple dimensions, and determine the execution priority of maintenance tasks. The optimization module is used for the new energy equipment fault maintenance solution library and algorithm model optimization.

2. The intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment according to claim 1 is characterized in that: The maintenance decision module is based on the Markov decision model. The shape parameters and scale parameters of the component failure distribution are estimated by the least squares method, and the Weibull-Gamma distribution model of the component failure is established to obtain the reliability and failure curve of the component. The component operation status is divided based on the component reliability curve. At the same time, according to the change of component reliability before and after maintenance, the maintenance cost of the component under various maintenance decisions is calculated, and the insulation degradation curve of the electrical equipment is obtained through the uniform distribution model. On this basis, relying on the Markov maintenance decision model and the strategy iteration algorithm, the optimal maintenance decision of the equipment components under each operating state and the intelligent push of the equipment maintenance strategy at the next maintenance are determined to achieve accurate maintenance of the equipment or system. The maintenance decision module also uses the constructed intelligent knowledge base tool to uniformly collect, process and apply the knowledge materials of operation and maintenance; The shape parameters and scale parameters of component failure distribution are estimated by the least squares method, and then the Weibull-Gamma distribution is established. Modeling process: There are n failure time data points t1, t2, ..., t n ; The Weibull-Gamma distribution is determined by two parameters: shape parameter α and scale parameter β. In(t i ); In the logarithmic coordinate system, the cumulative distribution function of the Weibull distribution can be expressed as In(-In(1-F(t)))=αIn(t)-αIn(β), where F(t) is the cumulative distribution function; (In(t i ),In(-In(1-F(t i )))) Linear regression is performed on the data points to obtain to the slope and intercept; Parameter estimation: The slope of the linear regression is the estimated value of the shape parameter α, and the negative value of the intercept is the estimated value of ln(β). Thus we get β = e -截距 .

3. The intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment according to claim 1 is characterized in that: The fault repair strategy library includes a defect data dictionary maintenance module, a repair plan library maintenance module, an operation instruction module, and a repair process card; The defect data dictionary maintenance module is used to maintain the basic definition information of defects, unify the descriptions of the same defect by business personnel, and provide standard words for system users; The maintenance module of the maintenance plan library is used to cluster the facts and concepts formed after knowledge management, provide a theoretical basis and decision-making basis for equipment maintenance, enable maintenance personnel to quickly identify and troubleshoot equipment failures, share failure handling experience, and associate failure cases with the equipment structure tree in equipment management to classify and manage all failures; The work instruction module is used to formulate safety and quality control measures and implement them during the operation process; The maintenance process card is compiled by integrating the troubleshooting before maintenance, maintenance ticket processing and confirmation, spare parts confirmation, disassembly and inspection process, technical requirements for each part, maintenance progress, maintenance acceptance, and test run performance acceptance.

4. The intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment according to claim 3 is characterized in that: The functional design of the defect data dictionary maintenance module includes: Defect level: used to establish or modify the defect type of the unit, which is divided into Class I defect, Class II defect, Class III defect, Class IV defect, or into emergency defect and general defect, and set different processing period requirements for different types of defects; Defect content: If the defects that may occur in a certain type of equipment are relatively regular, define the defect content and define the possible defects into the system. Once a defect is found, enter the defect content through quick selection; Equipment classification: Obtain equipment classification and equipment information in defect management from the equipment ledger system. Each defect information of the equipment is recorded in the equipment ledger system for comparative analysis of the equipment operation status and equipment quality.

5. The intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment according to claim 3 is characterized in that: The functional design of the maintenance module of the maintenance scheme library includes: Maintenance of maintenance plan library: Based on experience and common equipment failures, a maintenance plan library for wind farms and photovoltaic stations is established. All failure cases are associated with the equipment structure tree in equipment management. The maintenance plan library records each type of equipment at any time, including: failure type, failure name, equipment manufacturer, failure phenomenon, number of failures, failure cause, failure treatment plan, and failure analysis report. It supports the uploading of failure phenomenon pictures and failure analysis reports. When on-site personnel find equipment failures, they can quickly search for similar failures in the maintenance plan library based on the manufacturer, failure type, and failure phenomenon of the equipment. By consulting the failure phenomenon, treatment plan, and failure analysis report of related failure cases and comparing them with the actual equipment failures, the cause of the equipment failure can be quickly identified. The failure cases include: photovoltaic component failure, wind turbine failure, power transmission failure, inverter failure, and control system failure. The maintenance plan library includes: maintenance plan information, maintenance plan attachment information, maintenance plan spare parts information, maintenance plan tool information, maintenance plan operation instruction information, and maintenance plan process card information.

6. The intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment according to claim 1 is characterized in that: The task planning and scheduling module includes: Data fusion and analysis submodule: This module is responsible for integrating multi-source online and offline data, including unit SCADA operation data, equipment operating years, historical maintenance records, and historical defect records. It uses the coupling analysis method to determine the influence weights of key factors and analyzes their impact on the accuracy of intelligent scheduling. It supports transfer learning technology for model iteration and dynamic optimization to meet the needs of different scenarios at sea and on land. Comprehensive scheduling model construction submodule: Based on the regional data platform, it integrates equipment, software and information data to build a unified and coordinated intelligent scheduling system, realize automatic sorting and selection of operation and inspection task priorities, optimize work processes, improve work efficiency, and support flexible adjustment of scheduling results on site; Defect task management module: specializes in handling defect work orders of new energy stations, including the receipt, classification, priority setting and related data analysis of work orders, to ensure that defect problems are dispatched and resolved in a timely and effective manner; Maintenance plan management module: responsible for the intelligent scheduling of special maintenance work orders and regular maintenance work orders, sorting and recommending tasks based on task urgency and importance parameters, supporting the generation, adjustment and tracking of plans, and ensuring that maintenance activities are carried out efficiently and orderly; Defect statistical analysis module: conducts in-depth analysis of historical defect data, identifies failure modes, predicts potential problems, provides data support for preventive maintenance, and provides a quantitative basis for scheduling decisions; Task execution module: realizes automatic scheduling and execution tracking of tasks, supports real-time monitoring of task progress, dynamic allocation of resources and handling of abnormal situations, and ensures that tasks proceed smoothly as planned; Maintenance evaluation module: After the task is completed, the execution data is collected and the maintenance effect is evaluated, including time cost, material consumption, and power generation recovery, providing feedback for future task optimization and strategy adjustment; Resource and Constraint Management Module: Integrates intelligent maintenance resource management and operating environment constraint monitoring functions to ensure reasonable resource allocation, safe and compliant operations, and dynamically adjusts scheduling strategies based on real-time environmental data; Decision support and intelligent recommendation module: Based on the decision factor library, combined with real-time data and historical experience, it provides intelligent suggestions and warnings for operation and maintenance decisions, and enhances the flexibility and adaptability of scheduling; System integration and expansion module: Provides standard interfaces and protocols to ensure seamless integration of the system with the external environment, supports horizontal and vertical expansion, meets the customized needs of sites of different sizes, and ensures data security and privacy protection.

7. An intelligent scheduling method using the intelligent scheduling system for fault inspection and maintenance of new energy power generation equipment according to any one of claims 1 to 6, characterized in that: Includes steps: S1. Build an intelligent scheduling platform for task dispatching: integrate multi-source online and offline data such as unit SCADA operation data, equipment operation years, historical maintenance records, and historical defect records, couple and analyze the weight of key factors affecting work type, unit location, climate environment, transportation, personnel skill level, communication conditions, spare parts, and spot trading price information, analyze the impact of multi-parameter and multi-factor change trends on the accuracy of intelligent scheduling, apply transfer learning, conduct research on indicator iteration and model dynamic optimization algorithms, and train different scheduling models for new energy applications in different scenarios at sea and on land; S2. Through functional integration, data integration and business integration of existing systems, various separation equipment, software and information data elements are integrated into a unified and coordinated intelligent scheduling system based on the regional data platform, a comprehensive scheduling model is constructed, and the priority of operation and inspection tasks is automatically sorted and selected to optimize the workflow; S3. Intelligent scheduling of defect work orders, special maintenance work orders, and maintenance work orders for new energy stations, with the multi-dimensional goals of minimizing power loss and maximizing maintenance efficiency, quantitative analysis of maintenance tasks to be executed, multi-dimensional perception of the influencing factors of on-site maintenance, and determination of maintenance task execution priority with the help of intelligent algorithms; S4. Automatically arrange the execution order of operation and maintenance tasks according to the set rules and priorities, and automatically adjust task arrangements and priorities based on historical data and trend forecasts.

8. The intelligent scheduling method for fault inspection and maintenance of new energy power generation equipment according to claim 7 is characterized in that: The defect work order content in S3 includes the work order serial number, area name, station name, equipment name, equipment number, defect location, defect type, defect source / discoverer, discovery time, status, and defect level; The contents of a maintenance work order include the work order serial number, work order number, work order name, work order type, person who dispatches the order, dispatch time, work order status, two-ticket approval information of the work order, spare parts, operating instructions, and a maintenance process card; the contents of a special maintenance work order include the work order serial number, special maintenance work order number, work order name, work order type, person who dispatches the order, dispatch time, work order status, two-ticket approval information of the work order, spare parts, operating instructions, and a maintenance process card.

Citation Information

Patent Citations

  • Fully mechanized coal mining face fault judgment method based on big data technology

    CN112801313A

  • Full-automatic running train intelligent operation and maintenance management platform and method

    CN113554193A

  • Power transformation maintenance plan intelligent scheduling strategy method based on machine learning

    CN115689177A

  • Production management system for intelligent supervision

    CN116596177A

  • Offshore wind turbine generator maintenance planning system and method based on genetic algorithm

    CN117291572A

Cited By

  • Client management method and system based on artificial intelligence

    CN119671508A

  • Intelligent scheduling management system and method for power grid communication maintenance plan

    CN119784361A

  • Control method for production equipment of photovoltaic round wire copper strip

    CN120447494A

  • Maintenance decision system and method based on big data driving

    CN120509754A

  • Industrial knowledge query method and device, equipment and medium

    CN120849554A