Large model system and method for post-evaluation of wind power project, electronic equipment and storage medium

Abstract

The invention provides a wind power project post-evaluation large model system and method, electronic equipment and a storage medium, and the system comprises a data collection module which is used for collecting document data and equipment operation data of a wind power project, and the document data comprises a feasibility research report, a design report and project approval materials of the early stage of the wind power project; the classified storage module is used for identifying the document data through an OCR (Optical Character Recognition) technology, converting the document data into a preset format, and classifying and storing the document data according to an established historical document library; the training module carries out training through the data of the multiple wind power projects based on pre-training, supervised fine tuning, reward modeling and reinforcement learning processes and is used for extracting post-evaluation data of the wind power projects; and the post-evaluation module is used for generating a post-evaluation report of the wind power project according to the post-evaluation data.

Description

Technical Field

[0001] The present invention relates to the field of automation and control, and in particular to a large model system, method, electronic equipment and storage medium for post-evaluation of a wind power project.

[0002] Large model system, method, electronic equipment and storage medium for post-evaluation of wind power projects Background Art

[0003] Post-project evaluation refers to a systematic summary and analysis of the entire process of project establishment, decision-making, implementation and later operation after the project is completed, with the project's entire life cycle as the main line. The evaluation includes the project decision-making process, project implementation process, project operation process, and project social and economic benefits, so as to determine whether the expected project goals have been achieved, summarize experiences and lessons, and provide reference for future projects.

[0004] At present, post-evaluation of wind power projects is mainly based on universal post-evaluation, but this evaluation system is still imperfect. The selection of evaluation indicators is subjective, lacking in systematicity and scientificity; and the blindness of the evaluation method cannot accurately evaluate certain types of projects, resulting in insufficient comprehensiveness and specificity of the research.

[0005] In order to solve the above problems, a large model system, method, electronic device and storage medium for post-evaluation of wind power projects are proposed. Summary of the invention

[0006] In a first aspect of the present disclosure, a large model system for post-evaluation of a wind power project is provided, the system comprising:

[0007] A data collection module is used to collect document data and equipment operation data of the wind power project, wherein the document data includes the feasibility study report, design report, and project establishment materials of the wind power project in the early stage;

[0008] A classification storage module, which recognizes the document data through OCR recognition technology, converts it into a preset format, and classifies and stores the document data according to an established historical document library;

[0009] A training module, which is trained by data of a plurality of wind power projects based on pre-training, supervised fine-tuning, reward modeling and reinforcement learning processes, and is used to extract post-evaluation data of the wind power projects;

[0010] A post-evaluation module generates a post-evaluation report of the wind power project according to the post-evaluation data.

[0011] In combination with the first aspect, the training module is trained with historical data of 30-50 wind power projects to extract post-evaluation data of the wind power projects.

[0012] In combination with the first aspect, the equipment operation data collected by the data collection module includes the power generation of the wind turbine set, the equipment operation status, and the maintenance record.

[0013] In combination with the first aspect, the system further includes a data correction module, which is used to trace and verify the document data of the wind power project and correct it through a preset verification mechanism.

[0014] In combination with the first aspect, the system also includes an evaluation algorithm module, which extracts evaluation indicators of the wind power project based on equipment operation data of the wind power project, and uses Delphi method, paired comparison method and hierarchical analysis method to optimize the weight calculation method of the evaluation indicators.

[0015] In combination with the first aspect, the post-evaluation system module includes six evaluation sub-modules: post-feasibility study evaluation, post-design evaluation, post-project implementation evaluation, post-operation evaluation, post-operation and maintenance evaluation, and post-technical transformation evaluation, which are used to evaluate the wind power project at different stages.

[0016] A second aspect of the present disclosure provides a large model method for post-evaluation of a wind power project, the method comprising the following steps:

[0017] Collect document data and equipment operation data of wind power projects through the data acquisition module, wherein the document data includes feasibility study reports, design reports and project establishment materials of wind power projects;

[0018] Recognize the document data by OCR recognition technology, convert it into document data in a preset format, and classify and store it according to the established historical document library;

[0019] Using the training module, we train on data from multiple wind power projects through pre-training, supervised fine-tuning, reward modeling, and reinforcement learning processes to extract post-evaluation data for wind power projects.

[0020] A post-evaluation module is used to generate a post-evaluation report of the wind power project according to the post-evaluation data.

[0021] In combination with the second aspect, the method further includes tracing and verifying the document data of the wind power project through a data correction module, and correcting the document data according to a preset verification mechanism.

[0022] According to a third aspect of the present disclosure, an electronic device is provided, including:

[0023] one or more processors;

[0024] A storage unit is used to store one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement any of the above-mentioned large-scale model methods for post-evaluation of wind power projects.

[0025] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, it can implement any of the above-mentioned large-scale model methods for post-evaluation of wind power projects.

[0026] The embodiments of the present disclosure aim to solve at least one of the technical problems existing in the prior art, and provide a large model system, method, electronic device and storage medium for post-evaluation of a wind power project. By utilizing a data acquisition module to collect document data and equipment operation data throughout the life cycle of the project, combining OCR recognition technology and a large model training module to classify, store and intelligently analyze the data, a post-evaluation report for the project is generated, thereby ensuring the objectivity of the evaluation process and the accuracy of the data, thereby improving the scientificity and comprehensiveness of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of the structure of a large model system for post-evaluation of a wind power project according to an embodiment of the present disclosure;

[0028] Figure 2 It is a flow chart of a large model method for post-evaluation of a wind power project according to an embodiment of the present disclosure;

[0029] Figure 3 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0030] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the embodiments of the present disclosure.

[0031] like Figure 1 The following is a schematic diagram of the structure of the wind power project post-evaluation large model system according to the embodiment of the present disclosure. It includes:

[0032] The data collection module 110 is used to collect document data and equipment operation data of the wind power project, wherein the document data includes the feasibility study report, design report, and project establishment materials of the wind power project in the early stage;

[0033] The classification storage module 120 recognizes the document data through OCR recognition technology, converts it into document data in a preset format, and classifies and stores the document data according to the established historical document library;

[0034] A training module 130 is used for extracting post-evaluation data of the wind power project by training with data of the plurality of wind power projects based on pre-training, supervised fine-tuning, reward modeling and reinforcement learning processes;

[0035] The post-evaluation module 140 generates a post-evaluation report of the wind power project according to the post-evaluation data.

[0036] Data acquisition module 110: This module is responsible for acquiring different types of data from the entire life cycle of the wind power project. It mainly includes document data and equipment operation data. Document data involves important materials in the early stages of wind power projects, such as feasibility reports (feasibility study reports), design reports, project approval materials, etc. These documents cover the decision-making basis and technical background of wind power projects in planning, design and project approval. Through equipment operation data, the system can obtain the status of wind turbines in real time, such as power generation, equipment operation status and maintenance records, etc. These data are crucial for evaluating the operating efficiency and long-term performance of wind farms.

[0037] Classification storage module 120: The collected data usually contains a large number of unstructured documents, especially feasibility reports and design reports in PDF format, which cannot be used directly for analysis. This module first converts these document data into a preset format that can be read and processed by the computer through OCR (optical character recognition) technology to ensure that the data can be accurately identified and used in the subsequent processing stage. The converted data is classified and stored according to the historical document library established in advance. The role of the historical document library is similar to a reference system, which can help the system quickly locate and classify different document data, making the subsequent evaluation process more systematic and efficient.

[0038] Training module 130: In-depth analysis of wind power project data through a general large model. The training module performs data training based on a process of pre-training, supervised fine-tuning, reward modeling, and reinforcement learning. First, through the pre-training model, the system can obtain basic wind power project-related knowledge; then, with the help of supervised learning, the model is fine-tuned so that it can be optimized for specific wind power project post-evaluation tasks. Through reward modeling and reinforcement learning, the system gradually improves its data analysis capabilities and can extract key information required for post-evaluation from historical data of multiple wind power projects. This information includes the project's technical indicators, financial performance, and the project's environmental and social benefits.

[0039] Post-evaluation module 140: Based on the post-evaluation data extracted by the training module, the post-evaluation module generates a post-evaluation report for the wind power project. The report covers a comprehensive evaluation of the wind power project at multiple stages, and the specific content may include rationality analysis in the planning stage, efficiency evaluation in project implementation, economic benefits and technical performance during operation, and the environmental and social impact of the project. Through intelligent analysis and report generation, the post-evaluation module can provide objective and comprehensive data support, making the later decision-making of the wind power project more scientific and accurate.

[0040] Beneficial effects: Through automated data collection and intelligent analysis, the system realizes the automation of the entire process of post-evaluation of wind power projects, improves the accuracy, objectivity and scientificity of the evaluation, and provides a reliable decision-making basis for the improvement and optimization of the project in the later stage.

[0041] Furthermore, the training module is trained with historical data of 30-50 wind power projects to extract post-evaluation data of the wind power projects.

[0042] The learning and analysis capabilities of the training module are closely related to the quality and quantity of the input data. By selecting historical data from 30-50 wind power projects for training, the system can strike a balance between data size and complexity.

[0043] The 30-50 project data covers different wind farms, which may be located in different geographical areas and are affected by different factors such as climate conditions, equipment selection, investment scale, etc. Such data diversity provides the system with rich reference experience, enabling it to cope with diverse post-evaluation needs of wind power projects.

[0044] The historical data of 30-50 projects is large, which can provide enough samples for the system to train. In the field of machine learning, the larger the sample size, the richer the features learned by the model, so that it can better cope with complex scenarios in different projects and enhance the generalization ability of the model.

[0045] If there is too little project data, the system may not be able to fully learn the complex features of wind power projects, causing the model to deviate or the results to be unstable when facing actual projects. Too much training data may increase the computational burden and prolong the training time of the system. Data from 30-50 projects achieves a reasonable balance between time cost and result quality, ensuring that the system has sufficient learning ability and efficiency.

[0046] The specific working process of the training module:

[0047] Pre-training: In this phase, the training module will use existing historical data to conduct preliminary training on the model to help the system understand the basic laws and characteristics of wind power projects. This includes learning common document structures, common terms, technical parameters, etc. in project data.

[0048] Supervised fine-tuning: Based on pre-training, the system uses the labeled data from historical projects for supervised fine-tuning. With the known evaluation results of these projects, the system can optimize the target indicators and evaluation methods, so that the model can more accurately extract the target information in the wind power project.

[0049] Reward modeling: This process is similar to a feedback loop, where the system judges the accuracy of its predictions by comparing its analysis results with the evaluation results of actual projects. For accurate predictions, the system gives positive feedback, while for inaccurate predictions, the system adjusts the analysis strategy to improve the accuracy of subsequent data extraction and analysis.

[0050] Reinforcement learning: In the reinforcement learning stage, the system further improves its understanding and analysis capabilities of wind power project data through multiple iterations of optimization. Key evaluation data from each project data will be gradually extracted, such as the site selection of wind farms, equipment efficiency, investment returns, operation and maintenance management, etc. These extracted data will eventually be used in subsequent evaluation and analysis.

[0051] Post-evaluation data extracted through training: After learning the training module, the system can accurately extract and analyze multiple key data in the wind power project. The extracted post-evaluation data can include but is not limited to the following categories:

[0052] Technical data: key technical parameters of the project, such as the performance of the wind turbine, power generation, efficiency of the equipment, etc.

[0053] Financial data: financial indicators such as the project's return on investment, construction costs, and operating costs.

[0054] Management data: information such as management efficiency, equipment maintenance status, failure rate, etc. involved in the operation and maintenance process.

[0055] Environmental and social benefit data: Environmental impact analysis of wind power projects, such as reduced carbon emissions and contribution to local social and economic benefits.

[0056] Through in-depth extraction and analysis of these data, the system can generate quantitative and actionable post-evaluation reports for wind power projects, providing a scientific basis for subsequent project improvements.

[0057] Beneficial effects: It ensures that the system can not only process the data of the current project, but also provide a more comprehensive, scientific and objective evaluation based on the experience of multiple historical projects.

[0058] Furthermore, the equipment operation data collected by the data collection module includes the power generation of the wind turbine set, the equipment operation status, and the maintenance record.

[0059] The power generation of wind turbines is one of the key indicators for measuring project benefits and is also the core basis for evaluating the overall operating efficiency of wind farms. Collecting power generation data can help the post-evaluation system achieve the following goals:

[0060] Evaluate equipment efficiency: Power generation reflects the actual output power of the wind turbine. The system can analyze the operating efficiency of the wind turbine based on the comparison between the expected design power generation and the actual power generation. Lower than expected power generation may mean equipment problems, improper site selection or insufficient utilization of wind resources.

[0061] Determine the economic benefits of the project: Power generation is directly related to the economic return of the project. By collecting and analyzing power generation data, the system can calculate the rate of return of the project, evaluate whether the wind farm has achieved the investment expectations, and provide investment reference for future projects.

[0062] Operation status monitoring: The system can identify abnormal operation of equipment through fluctuations in power generation. If the power generation data fluctuates significantly, it may indicate equipment failure, untimely maintenance or changes in the external environment.

[0063] Equipment operating status data: The operating status data of wind turbines includes key performance parameters of the equipment during operation, such as rotation speed, wind speed, temperature, vibration, etc. These data play an important role in evaluating the long-term operating status and technical performance of wind power projects:

[0064] Real-time monitoring: By collecting operating status data, the system can monitor the operating status of the equipment in real time and promptly detect potential faults or abnormal conditions, such as exceeding normal speed or overheating.

[0065] Health assessment: Equipment performance degradation during long-term operation is inevitable. By analyzing the operating status data of wind turbines, the system can evaluate the health status of the equipment and determine whether the equipment needs maintenance or replacement.

[0066] Optimize operation and maintenance strategies: Operation status data can also provide reference for the operation and maintenance of wind farms, helping managers optimize operation and maintenance strategies, maximize the service life of equipment, and improve the overall efficiency of wind farms.

[0067] Maintenance record data: Maintenance record data refers to the maintenance log of the wind turbine during the entire operation cycle, including regular maintenance, fault repair, parts replacement, etc. These data are particularly important for the post-evaluation of wind power projects, specifically in the following aspects:

[0068] Equipment maintenance history: By collecting maintenance records, the system can build equipment maintenance files and provide detailed information on the maintenance status of each wind turbine, the frequency of failures, and the time and type of parts replaced. This information helps predict the remaining life of the equipment and future maintenance needs.

[0069] Maintenance effectiveness evaluation: The system can evaluate the effectiveness of operation and maintenance management based on maintenance records. If the maintenance frequency of the equipment is too high, it means that there are problems with its quality; if the equipment does not require maintenance for a long time, it may indicate that its stability is high.

[0070] Cost analysis: Maintenance records are closely related to equipment operating costs. By analyzing maintenance records, the system can calculate maintenance costs and incorporate them into the overall operating cost assessment of the project to help evaluate the economic benefits of the project.

[0071] Beneficial effects: Through the continuous collection and analysis of equipment operation data, the wind power project post-evaluation system can conduct scientific and comprehensive evaluation of wind power projects from multiple perspectives, including technology, economy and management, and provide a basis for the subsequent construction and operation of wind farms.

[0072] Furthermore, the system also includes a data correction module for tracing and verifying the document data of the wind power project and correcting it through a preset verification mechanism.

[0073] Data traceability refers to tracking the source of document data to ensure that each piece of data can find its original source of generation or record. The document data involved in wind power projects, especially the early feasibility study reports, design reports, project approval materials, etc., often come from different departments and time periods. Therefore, the traceability function can ensure that these data will not be missed or tampered with during the collection process.

[0074] Through the traceability function, the system can mark the generation time, source department and relevant personnel of each document data to ensure that the data used in the subsequent evaluation process is complete, accurate and legal.

[0075] Data verification refers to checking the authenticity and consistency of the collected document data to ensure that the data meets the requirements of wind power project evaluation in terms of content and format. Since the document data involved in wind power projects comes from a wide range of sources and in various formats (such as PDF, Word, Excel, etc.), problems such as data loss, misidentification or format confusion may occur during the collection and conversion process.

[0076] The verification function automatically checks the accuracy, completeness, and format consistency of data through preset inspection rules.

[0077] For example, the system can detect whether the project approval materials contain necessary approval information, whether the design report contains all equipment parameters, whether the power generation data is consistent with the actual operation of the project, etc.

[0078] At the same time, data verification can issue alerts for incomplete or abnormal data and automatically feedback the problem to relevant personnel for manual confirmation or supplementation. This can effectively reduce the deviation of post-evaluation results caused by data errors.

[0079] The preset verification mechanism is one of the core elements of the data correction module. It contains a series of rules and logic for identifying and correcting data errors. The mechanism can dynamically adjust the verification standards according to the characteristics of different wind power projects to ensure that the data correction process is flexible and adaptable.

[0080] The verification mechanism can include multiple levels of inspection standards, such as range checks on numerical data in documents (such as whether the power generation is within a reasonable range), structured checks on text data (such as whether the project approval number is included), and format checks on chart data (such as whether the wind turbine location map is missing, etc.). Through the multi-level verification mechanism, the system can automatically discover potential errors in the data and take corresponding corrective measures.

[0081] Data correction refers to the automatic or semi-automatic correction of errors in document data through a preset verification mechanism. The correction process can adjust the non-compliant data to the correct format or content based on historical data, standard templates or preset rules.

[0082] For simple format errors or incomplete data, the system can automatically supplement or correct them based on the data in the historical document library. For example, if a parameter in a design report is missing, the system can automatically extract the corresponding data from the design documents of similar projects to fill it in.

[0083] For complex data errors or missing data, the system will provide suggested correction solutions and prompt relevant personnel to manually confirm or correct them.

[0084] Beneficial effects: Through the data correction module, the wind power project post-evaluation system can provide an accurate and reliable document data basis for the post-evaluation process, ensuring the scientificity and objectivity of the evaluation results. This not only improves the management level of wind power projects, but also provides strong data support for the optimization and improvement of future projects.

[0085] Furthermore, the system also includes an evaluation algorithm module, which extracts evaluation indicators of the wind power project based on the equipment operation data of the wind power project, and uses the Delphi method, paired comparison method, and hierarchical analysis method to optimize the weight calculation method of the evaluation indicators.

[0086] Functions and roles of the evaluation algorithm module:

[0087] The evaluation algorithm module first extracts the key indicators required for evaluation from the equipment operation data of the wind power project. These equipment operation data include information such as the power generation of wind turbines, equipment operation status, maintenance records, etc. By analyzing these data, the system can identify multiple evaluation indicators related to the project operation effect.

[0088] For example, for power generation, the system can compare the actual power generation of the unit with the design target to evaluate whether the power generation efficiency of the equipment meets the standard. For the equipment operation status, the system can extract key indicators such as failure rate and downtime to evaluate the stability and reliability of the equipment. For maintenance records, the system can analyze the maintenance frequency and maintenance cost to evaluate the operation and maintenance efficiency of the equipment.

[0089] Delphi method optimizes weight calculation:

[0090] The Delphi method is a weight optimization method based on expert feedback, which is often used to evaluate indicators that are difficult to quantify directly. This method invites multiple experts in the field of wind power projects to conduct multiple rounds of anonymous scoring and feedback, and finally summarizes the opinions of each expert to form a relatively consistent evaluation result.

[0091] In the post-evaluation of wind power projects, the weights of some evaluation indicators may be difficult to determine through simple data analysis, such as the social benefits and environmental impact of wind power projects. The weight optimization of these indicators can be achieved through the Delphi method, where multiple experts score their importance based on the specific circumstances of the project and ultimately come up with a more objective weight value.

[0092] Through multiple rounds of feedback mechanisms, the Delphi method can effectively eliminate personal bias and ensure the objectivity and scientific nature of the weight calculation results.

[0093] Paired comparison method optimizes weight calculation:

[0094] Paired comparison method is a method to systematically compare the relative importance of different evaluation indicators, and is often used to determine the priority among multiple indicators. This method compares each evaluation indicator pairwise to determine which indicator is relatively more important, and finally derives the relative weight of each indicator.

[0095] In the post-evaluation of wind power projects, the paired comparison method can help project managers and experts compare one by one in a set of evaluation indicators to ensure that the weight of each indicator reflects its relative importance in the overall evaluation. For example, the power generation efficiency and maintenance frequency of equipment may both be important indicators, but in some projects, the power generation efficiency may be relatively more important. The paired comparison method can systematically determine this relative weight.

[0096] Optimize weight calculation by analytic hierarchy process:

[0097] The Analytic Hierarchy Process (AHP) is a method that decomposes complex problems into multiple levels, analyzes them level by level, and calculates the weights. It establishes a hierarchical structure of evaluation indicators, scores and calculates each indicator level by level, and finally obtains the weight value of each indicator.

[0098] In the post-evaluation of wind power projects, the AHP can divide the evaluation system into multiple levels. For example, the top level can be the comprehensive evaluation results, the middle level can be the economic benefits, environmental benefits, equipment operation status and other categories, and the lower level can be further subdivided into specific evaluation indicators (such as power generation, failure rate, maintenance cost, etc.). Through this hierarchical approach, the AHP can better reflect the relative importance of different indicators in different evaluation dimensions.

[0099] The advantage of the AHP method is that it is logically sound and can intuitively reflect the hierarchical structure of the evaluation system and the relative weight of each indicator. It is suitable for post-evaluation of complex and multi-dimensional wind power projects.

[0100] Due to the large differences in the types and scales of wind power projects, the weights of evaluation indicators for different projects may vary significantly. The evaluation algorithm module can flexibly adjust the weight calculation method according to the specific project characteristics. For example, for projects that focus on environmental protection, the weight of environmental benefits may be relatively high, while for projects that focus on economic returns, the weight of economic benefits may be higher. Through flexible adjustment, the evaluation algorithm module can adapt to the evaluation needs of different wind power projects.

[0101] Beneficial effects: The evaluation algorithm module scientifically calculates the weights of different evaluation indicators by introducing multiple methods such as Delphi method, paired comparison method and hierarchical analysis method. This not only ensures the rationality of weight calculation, but also reduces the influence of human subjective bias, ensuring that the evaluation results have a high degree of credibility.

[0102] Furthermore, the post-evaluation system module includes six evaluation sub-modules: post-feasibility study evaluation, post-design evaluation, post-project implementation evaluation, post-operation evaluation, post-operation and maintenance evaluation, and post-technical transformation evaluation, which are used to evaluate the wind power project at different stages.

[0103] Post-feasibility study evaluation mainly evaluates the feasibility study report of the wind power project. The feasibility study report is a key document before the project starts, which evaluates the project's economic feasibility, technical feasibility, environmental impact and other aspects.

[0104] Through post-feasibility study evaluation, the system can analyze whether the economic and technical conditions of the project meet the original predictions during the actual implementation process, such as whether the expected power generation capacity and return on investment have been achieved. If the results of the project implementation are significantly different from those predicted in the initial feasibility study report, the system will mark the deviation and help project decision makers adjust future project planning and investment strategies.

[0105] Post-design evaluation is a review and assessment of the design phase of a wind power project. The design of a wind power project includes technical solutions, equipment selection, construction planning and other aspects.

[0106] The system will analyze the actual effect and feasibility of the design. For example, the system can verify through operation data whether the designed wind turbine configuration meets the wind resource characteristics of the project site, whether the designed power output matches the grid connection standard, etc. If the design does not meet expectations, the system will analyze the causes of the problems in the design and provide reference for future project design optimization.

[0107] The post-project evaluation assesses the performance of the wind power project during the construction phase. This phase includes equipment procurement, civil construction, installation and commissioning. Various problems may be encountered during the implementation process, such as construction delays and cost overruns.

[0108] Through post-project evaluation, the system can summarize the successful experiences and shortcomings of the implementation process, analyze the deviation between the construction period and the budget, and evaluate the quality and efficiency of equipment installation. For example, after the equipment of a wind farm is installed, the actual power generation does not reach the designed value. The system can trace back to the project implementation link to find out whether the performance decline is caused by deviations in equipment installation or construction planning.

[0109] Post-operation evaluation focuses on the performance of wind power projects after they are put into operation. This includes the power generation, operating efficiency, energy utilization, equipment reliability and stability of wind turbines, etc.

[0110] The system can evaluate the overall operation of wind power projects through long-term tracking and analysis of operation data. For example, by analyzing data such as wind turbine power generation, downtime, and failure rate, the system can determine whether the project is running smoothly and whether the power generation efficiency meets the design expectations. The results of post-operation evaluation can provide experience accumulation for the operation and management of similar projects in the future.

[0111] Post-operation and maintenance evaluation evaluates the maintenance and repair of wind power projects. Wind turbines will inevitably experience equipment aging or failure during long-term operation, and timely and effective maintenance is crucial to ensure the continued operation of the project.

[0112] The system can evaluate the project's operation and maintenance efficiency by collecting data such as maintenance records, fault reports, and repair costs. Post-operation and maintenance evaluation can not only help project managers discover potential equipment problems, but also analyze whether the operation and maintenance costs are reasonable and whether the operation and maintenance strategy needs to be optimized. For example, if some wind turbines have a high failure rate under the same conditions, the system can prompt managers to adjust the maintenance cycle or replace equipment.

[0113] Post-technical transformation evaluation mainly evaluates the effects of technical transformation of wind power projects. With the advancement of technology, wind power projects may upgrade equipment, improve processes or make other technical adjustments to improve power generation efficiency and reduce maintenance costs.

[0114] The post-technical transformation evaluation module will analyze the specific effects of the technological transformation, such as whether the improved equipment has improved power generation efficiency, reduced maintenance costs, and whether the expected technological transformation goals have been achieved. The system can provide a detailed effect analysis of the technological transformation by comparing it with the operating data before the technological transformation, helping decision makers determine whether the technological transformation investment has brought the expected return.

[0115] like Figure 2 , which is a flow chart of a large model method for post-evaluation of a wind power project according to an embodiment of the present disclosure.

[0116] include:

[0117] The method comprises the following steps:

[0118] S201: Collecting document data and equipment operation data of the wind power project through a data acquisition module, wherein the document data includes a feasibility study report, a design report and project establishment materials of the wind power project;

[0119] S202: Recognize the document data by using OCR recognition technology, convert it into document data in a preset format, and classify and store it according to the established historical document library;

[0120] S203: using the training module to perform training based on data of multiple wind power projects through pre-training, supervised fine-tuning, reward modeling and reinforcement learning processes, and extracting post-evaluation data of the wind power projects;

[0121] S204: using a post-evaluation module to generate a post-evaluation report for the wind power project according to the post-evaluation data.

[0122] Collect document data and equipment operation data of wind power projects through the data acquisition module:

[0123] Document data is a key record of the entire life cycle of a wind power project, covering the preliminary planning and design information of the project. It mainly includes three parts:

[0124] Feasibility study report: The feasibility study report of the project evaluates the technical feasibility, economic feasibility and environmental impact of the project, and provides a basis for the project establishment.

[0125] Design report: Detailed design documents for wind power projects, including wind turbine configuration, power generation system design, infrastructure construction, etc.

[0126] Project establishment materials: relevant documents for project approval and establishment, including investment approval, government permits, etc.

[0127] Equipment operation data covers information such as the power generation of wind turbines during the actual operation of the project, the operating status of the equipment, and maintenance and repair records. These data can not only reflect the technical performance of the project, but also reveal the stability and reliability of the equipment in long-term use.

[0128] The document data is identified through OCR recognition technology, converted into document data in a preset format, and classified and stored according to the established historical document library:

[0129] OCR recognition technology (Optical Character Recognition): Through this technology, the system can automatically identify paper or scanned documents of wind power projects and convert them into editable and computer-readable electronic document formats.

[0130] According to the needs of post-evaluation of wind power projects, the system will convert document data into a specific structured format for subsequent analysis and processing. This standardized format ensures data consistency and makes subsequent data classification and analysis more efficient.

[0131] The converted document data will be compared with the historical document library, and the data will be stored in the corresponding categories (such as feasibility study stage, design stage, etc.) through intelligent classification algorithms. This process not only helps the subsequent evaluation work, but also provides data support for the comparative analysis of multiple projects.

[0132] Using the training module, we train on the data of multiple wind power projects through pre-training, supervised fine-tuning, reward modeling and reinforcement learning processes to extract post-evaluation data of wind power projects:

[0133] Training module: aims to train historical wind power project data through machine learning algorithms and automatically extract key data required for post-project evaluation.

[0134] Pre-training: The system is first pre-trained based on a large amount of historical data of wind power projects. The purpose of this step is to allow the model to initially understand the laws and characteristics of wind power project data, so that useful information can be extracted more accurately in subsequent training.

[0135] Supervised fine-tuning: Based on pre-training, the system performs supervised fine-tuning through labeled sample data, so that the model can further learn specific evaluation indicators and data features.

[0136] Reward modeling and reinforcement learning: This process uses a reward mechanism to allow the system to continuously adjust and optimize its evaluation model. By comparing the deviation between the predicted results and the actual results, the system gradually optimizes its data extraction and evaluation capabilities.

[0137] Extracting post-evaluation data: Through training, the system can eventually extract various key data required for post-evaluation of wind power projects. These data may include the project's technical performance, economic benefits, environmental impact, equipment reliability, etc.

[0138] Using the post-evaluation module, a post-evaluation report of the wind power project is generated according to the post-evaluation data:

[0139] Feasibility study evaluation: By comparing the actual operation of the project with the forecast data in the feasibility study report, analyze whether the technical and economic forecasts of the project are accurate and whether the expected results have been achieved.

[0140] Design evaluation: Evaluate the actual implementation effect of the design solution, and whether the designed wind turbine configuration, power generation efficiency, etc. meet the project requirements.

[0141] Implementation evaluation: Evaluate the project's performance during the construction and equipment installation phases to see if there are any problems such as construction delays and cost overruns.

[0142] Operational evaluation: Analyze the power generation, equipment failure rate, maintenance cost, etc. during the project operation to determine the long-term operating benefits of the project.

[0143] Operation and maintenance evaluation: Evaluate the maintenance strategy and cost of the project, and analyze the reliability and maintenance efficiency of equipment operation.

[0144] Technical transformation evaluation: An evaluation after a project has undergone technical transformation, analyzing whether the transformation has effectively improved power generation efficiency or reduced operating costs.

[0145] Application of post-evaluation report: This post-evaluation report not only provides wind power project managers with a complete project evaluation, but also helps them identify problems and potential optimization points in the project. Through data analysis and trend prediction in the report, managers can adopt better design, construction and operation and maintenance strategies in future projects to improve overall benefits.

[0146] Beneficial effects: The wind power project post-evaluation method realizes intelligent and systematic evaluation of wind power projects through a series of modular steps, using data collection, OCR recognition, machine learning training and post-evaluation modules. Through these processes, the system can automatically extract key indicators from a large amount of project data and generate detailed post-evaluation reports. This method not only saves manpower and time, but also greatly improves the accuracy and efficiency of wind power project post-evaluation, providing strong support for project optimization and management.

[0147] The electronic device 300 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 300 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art will appreciate that Figure 3 It is only an example of the electronic device 300 and does not constitute a limitation of the electronic device 300. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0148] The processor 301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0149] The memory 302 may be an internal storage unit of the electronic device 300, for example, a hard disk or memory of the electronic device 300. The memory 302 may also be an external storage device of the electronic device 300, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 300. Further, the memory 302 may also include both an internal storage unit of the electronic device 300 and an external storage device. The memory 302 is used to store the computer program 303 and other programs and data required by the electronic device. The memory 302 may also be used to temporarily store data that has been output or is to be output.

[0150] In the embodiments provided in the present disclosure, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. Multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, which may be electrical, mechanical or other forms.

[0151] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0152] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0153] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present disclosure implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, and the computer program code may be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0154] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should all be included in the protection scope of the present disclosure.

Claims

1. A large model system for post-evaluation of wind power projects, characterized in that: The system comprises: A data collection module is used to collect document data and equipment operation data of the wind power project, wherein the document data includes the feasibility study report, design report, and project establishment materials of the wind power project in the early stage; A classification storage module, which recognizes the document data through OCR recognition technology, converts it into a preset format, and classifies and stores the document data according to an established historical document library; A training module, for extracting post-evaluation data of the wind power project by training with data of a plurality of the wind power projects based on pre-training, supervised fine-tuning, reward modeling and reinforcement learning processes; A post-evaluation module generates a post-evaluation report of the wind power project according to the post-evaluation data.

2. The system according to claim 1, characterized in that The training module is trained by using historical data of 30-50 wind power projects to extract post-evaluation data of the wind power projects.

3. The system according to claim 1, characterized in that The equipment operation data collected by the data collection module includes the power generation of the wind turbine set, the equipment operation status, and the maintenance record.

4. The system according to claim 1, characterized in that The system also includes a data correction module, which is used to trace and verify the document data of the wind power project and correct it through a preset verification mechanism.

5. The system according to claim 1, characterized in that The system also includes an evaluation algorithm module, which extracts evaluation indicators of the wind power project based on equipment operation data of the wind power project, and uses the Delphi method, paired comparison method, and hierarchical analysis method to optimize the weight calculation method of the evaluation indicators.

6. The system according to claim 1, characterized in that The post-evaluation system module includes six evaluation sub-modules: post-feasibility study evaluation, post-design evaluation, post-project implementation evaluation, post-operation evaluation, post-operation and maintenance evaluation, and post-technical transformation evaluation, which are used to evaluate the wind power project at different stages.

7. A large model method for post-evaluation of wind power projects, characterized in that: The method comprises the following steps: Collect document data and equipment operation data of wind power projects through the data acquisition module, wherein the document data includes feasibility study reports, design reports and project establishment materials of wind power projects; Recognize the document data by OCR recognition technology, convert it into document data in a preset format, and classify and store it according to the established historical document library; Using the training module, we train on data from multiple wind power projects through pre-training, supervised fine-tuning, reward modeling, and reinforcement learning processes to extract post-evaluation data for wind power projects. A post-evaluation module is used to generate a post-evaluation report of the wind power project according to the post-evaluation data.

8. The method according to claim 7, characterized in that The method further includes tracing and verifying the document data of the wind power project through a data correction module, and correcting the document data according to a preset verification mechanism.

9. An electronic device, characterized in that: include: one or more processors; A storage unit is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors can implement the large model method for post-evaluation of wind power projects according to any one of claims 7-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it can implement the large model method for post-evaluation of a wind power project according to any one of claims 7-8.

Images