A method and system for evaluating operation index data of a thermal power unit

Abstract

The present application relates to a kind of operation index data evaluation method and system of thermal power generating unit, it is related to data analysis field, obtains the component service life list of the first number thermal power generating unit of target model, operating environment monitoring information and operation instruction time sequence information;When component service life list is inconsistent with component reference service life list, or operating environment monitoring information is inconsistent with reference operating environment, with component reference service life list, reference operating environment and operation instruction time sequence information as constraint, based on historical sample library, retrieve the operation index record data set satisfying target model;Mode analysis is carried out to operation index record data set, obtains operation index reference state, corrects operation index monitoring state, obtains operation index update state, executes operation index data evaluation, solves the technical problem that thermal power generating unit examination index data is not accurate, only considers the limitation of evaluation result caused by operation difference.

Description

Technical Field

[0001] This invention relates to the field of data analysis, and in particular to a method and system for evaluating the operating indicators of thermal power units. Background Technology

[0002] Operational assessments of thermal power units can accurately characterize the operational skills of assessors, thereby motivating operators to improve their skills. This assessment mechanism is of great significance for improving the operational efficiency and safety of thermal power units. However, in actual operation, the assessment indicators of thermal power units are often not solely influenced by differences in the operations of assessors. The performance indicators of thermal power units involve multiple complex factors, including but not limited to the service life of equipment components, changes in actual operating environment conditions, and the timing of execution of operating instructions. These factors are interrelated and work together to affect the unit's operating status, thus influencing the final assessment indicators. Currently, the assessment process for most thermal power units still focuses primarily on differences in the operations of assessors, neglecting the influence of other key factors. This approach often results in assessment results that fail to comprehensively and accurately reflect the actual operating status and performance level of the unit. Therefore, assessment methods that rely solely on operational differences have significant limitations in evaluating the operational indicator data of thermal power units, and cannot ensure the accuracy and reliability of the data. Summary of the Invention

[0003] This invention addresses the technical problem in the prior art where the performance indicators of thermal power units are inaccurate and the evaluation results are limited due to only considering operational differences. It provides a method and system for evaluating the performance indicators of thermal power units to solve this problem.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0005] In a first aspect, the present invention provides a method for evaluating the operational performance data of thermal power units. The method includes: obtaining a list of component service durations, operating environment monitoring information, and operation command timing information for the first thermal power unit of a target model; when the list of component service durations is inconsistent with the reference list of component service durations, or the operating environment monitoring information is inconsistent with the reference operating environment, using the reference list of component service durations, the reference operating environment, and the operation command timing information as constraints, and based on a historical sample database, retrieving a set of operational performance record data that meets the requirements of the target model; performing mode analysis on the set of operational performance record data to obtain a reference state of operational performance; correcting the monitoring state of operational performance to obtain an updated state of operational performance; and performing an operational performance data evaluation.

[0006] Secondly, the present invention provides an operational indicator data evaluation system for thermal power units. The system includes: an information acquisition module, used to obtain a component service duration list, operating environment monitoring information, and operation command timing information for the first thermal power unit of a target model; a data retrieval module, used to retrieve a set of operational indicator records that meet the target model based on a historical sample database when the component service duration list is inconsistent with the component baseline service duration list, or the operating environment monitoring information is inconsistent with the baseline operating environment, using the component baseline service duration list, the baseline operating environment, and the operation command timing information as constraints; and a data evaluation module, used to perform mode analysis on the set of operational indicator records to obtain a reference state of operational indicators, correct the monitoring state of operational indicators, obtain an updated state of operational indicators, and perform operational indicator data evaluation.

[0007] The beneficial effects of this invention are: by comprehensively considering the service life of thermal power unit components, operating environment monitoring information and operation command timing information, and by performing precise retrieval and data correction based on historical sample database, a more accurate and comprehensive evaluation of thermal power unit operating index data is achieved. Attached Figure Description

[0008] Figure 1 This is a flowchart illustrating a method for evaluating the operating performance data of thermal power units provided by the present invention.

[0009] Figure 2 This is a schematic diagram of the structure of an evaluation system for operating index data of thermal power units provided by the present invention.

[0010] Explanation of reference numerals in the attached diagram: Information acquisition module 11, data retrieval module 12, data evaluation module 13. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0013] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0014] Example 1:

[0015] like Figure 1 As shown in the figure, this embodiment of the invention provides a method for evaluating the operating index data of thermal power units, the method comprising:

[0016] S10: Obtain the component service life list, operating environment monitoring information, and operation command timing information for the first thermal power unit of the target model.

[0017] S20: When the component service duration list is inconsistent with the component reference service duration list, or the operating environment monitoring information is inconsistent with the reference operating environment, the component reference service duration list, the reference operating environment, and the operation command timing information are used as constraints, and a set of operating indicator records that meet the target model is retrieved based on the historical sample library.

[0018] S30: Perform mode analysis on the set of recorded operation indicators to obtain the reference status of operation indicators, correct the monitoring status of operation indicators, obtain the updated status of operation indicators, and perform operation indicator data evaluation.

[0019] For example, to evaluate the operational performance data of the first thermal power unit of a target model, it is first necessary to comprehensively collect and organize the key information of the unit. Determining the model and designation of the target thermal power unit is fundamental to ensuring the accuracy of subsequent analysis. For this specific thermal power unit, the service life list of its components is collected. This step involves checking and recording the service time of various key components within the unit, such as turbines, generators, and boiler components, since installation. Component service life is one of the important indicators for assessing the overall health of the unit and predicting potential failures. Next, environmental monitoring information of the thermal power unit is collected, including but not limited to key parameters such as temperature, pressure, humidity, vibration levels, and emission concentrations. These environmental parameters are crucial for understanding the stability, safety, and compliance with environmental standards of the unit's current operating environment. Real-time monitoring of these environmental variables allows for the timely detection and response to anomalies that may adversely affect unit operation. Furthermore, it is necessary to collect the timing information of the thermal power unit's operating commands, which involves all control commands received by the unit over a past period and their execution sequence. Operational command timing information plays a crucial role in understanding the unit's operating status within a specific time period, analyzing the effectiveness of operational strategies, and identifying potential operational errors or inappropriate behaviors. In summary, obtaining the component service life list, operating environment monitoring information, and operational command timing information for the first thermal power unit of the target model lays the foundation for subsequent evaluation of operational indicator data. This information will serve as input data for in-depth analysis of the unit's operating status, identification of potential problems, and proposal of corresponding improvement measures.

[0020] Furthermore, when faced with discrepancies between the component service life list and the baseline list for thermal power units, or between the operating environment monitoring information and the baseline environment, a series of measures are needed to ensure the accuracy and effectiveness of the operational indicator data evaluation. When a difference is identified between the component service life list of the target thermal power unit and the preset baseline component service life list, or when the operating environment monitoring information does not match the baseline operating environment, this difference may stem from various factors, such as component aging, replacement, maintenance, or changes in the operating environment due to weather, season, geographical location, etc. To address this difference, the baseline component service life list, the baseline operating environment, and the timing information of operating commands are used as constraints. This baseline information represents the unit's operating status under ideal or standard conditions and is the basis for subsequent analysis and evaluation. Next, a database containing a large number of historical samples—the historical sample database—is used to retrieve a set of operational indicator records matching the target model. This database contains operational indicator data from similar units in the past under different service lives, operating environments, and operating commands, and is crucial for comparative analysis and evaluation. During the retrieval process, data in the historical sample database is filtered and matched based on constraints such as the component baseline service duration list, baseline operating environment, and operating command timing information. By calculating the similarity or distance between service duration, operating environment, and operating commands, the set of historical samples most closely resembling the target unit's condition is identified. These historical sample sets provide operational indicator data for the target unit under similar conditions, enabling a more accurate assessment of the target unit's operating status and performance. Analysis and comparison of this data can identify potential problems or deficiencies in the unit's operation, and corresponding improvement measures and optimization suggestions can be proposed accordingly.

[0021] Finally, mode analysis is performed on the retrieved operational indicator record dataset. Mode analysis is a statistical method that identifies the most frequently occurring value in a dataset, known as the mode, which can be considered the most representative value of the dataset. In this scenario, mode analysis on the operational indicator record dataset yields a reference state for the operational indicators. This reference state represents a typical or expected performance of the thermal power unit's operational indicators under current conditions. This reference state is then used to correct the actual operational indicator monitoring status. Since thermal power units in actual operation may be affected by various factors, the monitored operational indicator status may exhibit certain deviations or fluctuations. Therefore, by comparing the monitored status with the reference state, these deviations can be identified, and the monitored status can be corrected accordingly, resulting in a more accurate and realistic updated operational indicator status. Subsequently, operational indicator data evaluation is performed based on this updated operational indicator status. The purpose of this step is to quantitatively assess the operational status of the thermal power unit to understand its performance under current conditions. By comparing the updated operating indicators with preset evaluation standards or historical data, a comprehensive evaluation result can be obtained for the thermal power unit in terms of operating efficiency, stability, energy consumption, etc. This evaluation result can provide an important reference for the operation and maintenance management of the thermal power unit, promptly identify and resolve potential problems, and ensure the safe and efficient operation of the unit.

[0022] In a preferred embodiment, when the component service duration list is inconsistent with the component baseline service duration list, or the operating environment monitoring information is inconsistent with the baseline operating environment, the process includes: obtaining the component service duration list set and the operating environment monitoring information set of the assessment personnel; performing cluster analysis on the component service duration list set to obtain a multi-cluster component service duration list, and taking the centroid component service duration list of the largest cluster as the component baseline service duration list; performing cluster analysis on the operating environment monitoring information set to obtain multi-cluster operating environment monitoring information, and taking the centroid operating environment monitoring information of the largest cluster as the baseline operating environment.

[0023] Specifically, when discrepancies are found between the component service duration list of the target thermal power unit and the preset component baseline service duration list, or when the operating environment monitoring information does not match the baseline operating environment, these baseline information needs to be updated or adjusted. To this end, a set of component service duration lists and a set of operating environment monitoring information provided by assessment personnel are collected. These sets contain data from multiple assessment personnel or different time periods, providing a rich sample. Next, cluster analysis is performed on the component service duration list set. Cluster analysis is an unsupervised learning method that groups similar objects into a single class, ensuring that objects within the same class are similar to each other, while objects in different classes are different from each other. By clustering the component service duration list set, multiple clusters are obtained, each representing a specific service duration pattern. To determine the most representative component baseline service duration list, the centroid of the largest cluster is selected as the new component baseline service duration list. Here, the centroid refers to the average value of all points within the cluster; it represents the central position of the cluster and is therefore representative. Similarly, cluster analysis is also performed on the operating environment monitoring information set. Through clustering, multiple clusters of operating environment monitoring information are obtained, each representing a specific operating environment pattern. To determine the most representative baseline operating environment, the centroid of the largest cluster was selected as the new baseline operating environment. This resulted in an updated list of component baseline service durations and baseline operating environments. This method allows for dynamic adjustment of the component baseline service duration list and baseline operating environments based on current conditions, making them more closely reflect actual operating conditions. This helps to more accurately assess the operating status of thermal power units, promptly identify and resolve problems, and ensure the safe and efficient operation of the units.

[0024] In a preferred embodiment, cluster analysis is performed on the component service duration list set to obtain a multi-cluster component service duration list. The service duration list of the centroid component of the largest cluster is set as the component baseline service duration list. This includes: performing a weighted distribution on the component model set to obtain the component model weighted distribution result; calculating pairwise weighted Euclidean distances on the component service duration list set based on the component model weighted distribution result to obtain a set of Euclidean distribution distances; and performing cluster analysis on the component service duration list set based on the Euclidean distribution distance threshold and the set of Euclidean distribution distances to obtain the multi-cluster component service duration list.

[0025] Optionally, the system faces multiple sets of component service duration lists provided by evaluators. These lists may differ due to various factors such as component model, operating environment, and operating habits. To more accurately reflect the characteristics of different component models in terms of service duration, a weighted distribution is applied to the component model set. The weighted distribution reflects the importance or representativeness of different component models within the overall system, which helps to give appropriate attention to different models in subsequent cluster analysis. Next, pairwise weighted Euclidean distance is calculated for the component service duration lists based on the component model weighted distribution results. Weighted Euclidean distance is a distance metric that considers the weights of different dimensions (in this case, component models). By calculating the weighted Euclidean distance between each pair of service duration lists, a set of Euclidean distances is obtained, which reflects the similarity or difference between the lists. Then, a Euclidean distance threshold is set to determine when two service duration lists are considered sufficiently close to be grouped into the same cluster. The selection of the threshold needs to be based on domain knowledge and the specific circumstances of the data to ensure the rationality and effectiveness of the clustering results. Finally, cluster analysis is performed on the component service duration list set by combining the Euclidean distance set and a set threshold. The cluster analysis process groups similar service duration lists into one category until each list belongs to a specific cluster. During this process, special attention is paid to the centroid of the largest cluster, which is the average of all service duration lists within that cluster. This centroid represents the most typical service duration pattern in that cluster and is therefore set as the new baseline service duration list for the component. This method not only allows for dynamic adjustment of the baseline service duration list based on current conditions, making it closer to actual operating conditions, but also takes into account the characteristic differences in service duration among different component models, thereby improving the accuracy and effectiveness of the cluster analysis.

[0026] In a preferred embodiment, weighting the component model set to obtain the component model weight distribution result includes: sending the component model set and the target model to several importance evaluation parties that do not communicate with each other to obtain several sets of importance score values; statistically analyzing the several sets of importance score values ​​to obtain the sum of importance score values ​​for the first model up to the sum of importance score values ​​for the Nth model, and the sum of importance score values; iterating through the sum of importance score values ​​for the first model up to the sum of importance score values ​​for the Nth model, and comparing each sum with the sum of importance score values ​​to obtain the component model weight distribution result.

[0027] In detail, when determining the baseline service life of thermal power unit components, a method based on multi-party evaluation is adopted to determine the weight distribution of component models in order to fully consider the differences in importance among different component models. First, a set of various component models is already available, representing the various components that may be used in a thermal power unit. Simultaneously, the target models are identified, i.e., the component models currently of concern or requiring evaluation. Next, to obtain importance evaluations of different component models, the set of component models (including the target models) is sent to several importance evaluation parties. These evaluation parties are independent and do not share information to ensure the independence and objectivity of the evaluation. Each evaluation party scores the importance of each component model based on its own expertise and experience. Then, the scores returned by these evaluation parties are collected and statistically analyzed. For each component model, the sum of all obtained scores is calculated, thus obtaining the sum of the importance scores for the first model, the second model, and so on, up to the Nth model. Simultaneously, the sum of all component model scores is also calculated, i.e., the sum of the importance scores. After obtaining these sums, a comprehensive calculation is performed. For each component model, the sum of its importance scores is compared to the sum of all other importance scores. This ratio reflects the importance or weight of that model among all component models. By traversing all models, a set of component model weight distribution results was obtained. This set of weight distribution results not only considers the actual usage of different component models in thermal power units but also incorporates the professional knowledge and experience of multiple evaluation parties, thus possessing high accuracy and reliability. These weights can be used to optimize the calculation process in subsequent cluster analysis or baseline service life determination, making the results more consistent with reality. In summary, this method can comprehensively consider multiple factors to determine a reasonable component model weight distribution, providing strong support for subsequent thermal power unit component evaluation and management.

[0028] In a preferred embodiment, the process of calculating the importance score values ​​of the plurality of sets to obtain the sum of importance scores for the first model up to the sum of importance scores for the Nth model, and the sum of importance score values, includes: calculating the importance score values ​​of the plurality of sets to obtain the set of importance scores for the first model up to the set of importance scores for the Nth model; traversing the set of importance scores for the first model up to the set of importance scores for the Nth model and removing outliers to obtain the set of importance scores for the first model up to the set of importance scores for the Nth model; summing the set of importance scores for the first model up to the set of importance scores for the Nth model to obtain the sum of importance score values; and traversing the set of importance scores for the first model up to the set of importance scores for the Nth model and summing them separately to obtain the sum of importance scores for the first model up to the sum of importance scores for the Nth model.

[0029] Preferably, several importance evaluation parties are collected to assign scores to each component model. These scores form several datasets, each containing importance scores for all components from model 1 to model N. Preliminary statistical analysis is then performed on these scores to obtain an importance score set for each model. This means that for model 1, there is a set containing scores from all evaluation parties; for model 2, there is also such a set; and so on, up to model N. This yields a complete dataset from the importance score set for model 1 to model N. However, in actual data, there may be outliers—scores that differ significantly from the majority of scores. These outliers may be due to misjudgments by evaluation parties, data entry errors, or other reasons. To eliminate the impact of these outliers on subsequent analysis, outlier removal is performed by iterating through the importance score set for each model. Outlier identification can be based on statistical methods, such as setting a reasonable threshold, and considering scores exceeding this threshold as outliers and removing them. After this step, a more accurate and reliable score set is obtained, namely, the importance scores from model 1 to model N. Then, the remaining importance scores for each model set are summed to obtain the total importance score for each model. Simultaneously, the importance scores for all models are summed to obtain the total importance score value. These two sums will be used to subsequently calculate the weight distribution of component models. Finally, the total importance score for each model is iterated to obtain complete data from the total importance score of the first model to the total importance score of the Nth model. This data will serve as the basis for subsequent calculations of the component model weight distribution. This process allows for more accurate statistical analysis and processing of importance scores from multiple evaluation parties, providing strong support for subsequently determining the weight distribution of component models.

[0030] In a preferred embodiment, based on the component reference service duration list, the reference operating environment, and the operation command timing information as constraints, and using a historical sample database, a set of operation indicator records that meet the target model is retrieved. This includes configuring service duration tolerance Euclidean distance, operating environment tolerance Euclidean distance, and operation sequence tolerance distance, wherein the operation sequence distance is equal to the ratio of the number of out-of-order to the number of in-order operation commands. Based on the service duration tolerance Euclidean distance, the operating environment tolerance Euclidean distance, and the operation sequence tolerance distance, and using the component reference service duration list, the reference operating environment, and the operation command timing information as a reference, a first retrieval constraint is constructed. From the historical sample database, a set of operation indicator records that meet the first retrieval constraint and the target model is retrieved.

[0031] Next, the three core constraints for retrieval are defined: the list of baseline service durations of components, the baseline operating environment, and the timing information of operating instructions. These conditions constitute the basic framework for the retrieved data. To enable flexible retrieval within the historical sample database, three tolerance distances are configured: service duration tolerance Euclidean distance, operating environment tolerance Euclidean distance, and operating sequence tolerance distance. The service duration tolerance Euclidean distance measures the difference between the service duration of components in the historical samples and the baseline service duration; the operating environment tolerance Euclidean distance assesses the similarity between the operating environment of the historical samples and the baseline operating environment; and the operating sequence tolerance distance is understood as the ratio of the number of out-of-order to the number of in-order misalignments in the instruction sequence, i.e., the misalignment rate, which reflects the degree of matching between the operating instruction sequence in the historical samples and the baseline operating instruction sequence. With these tolerance distances, the first retrieval constraint is constructed. This constraint, based on the list of baseline service durations of components, the baseline operating environment, and the timing information of operating instructions, combined with the above three tolerance distances, forms a multi-dimensional retrieval condition. This condition ensures that the retrieved data not only matches the target model but also closely resembles the baseline conditions in terms of service duration, operating environment, and operational sequence. Finally, this first retrieval constraint is used to retrieve a set of operational indicator records that meet the conditions from the historical sample database. The data in these sets not only contains information about the target model but also strictly conforms to the set constraints, thus providing accurate and reliable operational indicator data.

[0032] In a preferred embodiment, retrieving the set of operational indicator records that satisfy the first retrieval constraint and the target model from the historical sample library includes: retrieving a first sample set that satisfies the first retrieval constraint and the target model from the historical sample library; traversing the first sample set based on the service duration tolerance Euclidean distance, the operating environment tolerance Euclidean distance, and the operation sequence tolerance distance, and retrieving a second sample set that satisfies the target model from the historical sample library; and statistically analyzing the set of operational indicator records of the first sample set and the second sample set.

[0033] Specifically, to improve the accuracy and efficiency of the retrieval, a retrieval strategy based on multiple constraints and a secondary index is adopted. The retrieval is performed from a database containing historical samples of operational performance data for different component models under various service durations, operating environments, and operating commands. The goal is to retrieve a set of operational performance record data that matches the target model and meets specific baseline service duration, baseline operating environment, and operating command timing information. To achieve this goal, a first retrieval constraint is set, based on the component's baseline service duration list, baseline operating environment, and operating command timing information, combined with service duration tolerance Euclidean distance, operating environment tolerance Euclidean distance, and operating sequence tolerance distance. Next, a first-level retrieval is performed, which retrieves a first set of samples from the historical sample database that meets the first retrieval constraint and the target model. This step is a preliminary screening; filtering by the first retrieval constraint allows for the rapid location of a batch of samples similar to the target model and baseline conditions from a massive amount of historical samples. However, since the first retrieval constraint may have a certain tolerance range, the first sample set may contain some samples that do not fully meet the conditions. To further improve retrieval accuracy, a second-level retrieval was performed. This involved traversing each sample in the first sample set based on the Euclidean distance for service duration, operational environment, and operational sequence tolerance. Then, a second sample set was retrieved from the historical sample database that met the target model requirements and was closer to the currently traversed sample in these distance metrics. This step essentially refined and optimized the first sample set, increasing the retrieval volume (the second-level retrieval) to improve the accuracy of the results. Finally, the operational indicator records in both the first and second sample sets were statistically analyzed to obtain a final set of operational indicator records that met all conditions. This set not only contained information about the target model but also strictly adhered to the set benchmark conditions and tolerance ranges. This two-level indexing retrieval strategy improves the accuracy and reliability of retrieval results while maintaining retrieval efficiency. This strategy is particularly suitable for handling large-scale historical sample databases, effectively reducing unnecessary retrieval and computational overhead and improving overall retrieval performance.

[0034] The method for evaluating the operating index data of thermal power units provided in this embodiment of the invention has at least the following technical effects:

[0035] 1. The system can automatically determine the baseline service duration list and baseline operating environment of components based on the component service duration list and operating environment monitoring information set provided by the assessment personnel through cluster analysis. This adaptive approach avoids the subjectivity of manually setting baselines and improves the accuracy and applicability of the baselines. Furthermore, by weighting component models, the system further enhances the rationality of the baseline determination, making the baselines more reflective of actual conditions.

[0036] 2. When retrieving the data set of operational performance records that meet the target model, Euclidean distances for service duration, operating environment, and operation sequence were used as constraints. This multi-dimensional constraint retrieval technique can more accurately match historical samples, improving the accuracy and relevance of the retrieval results. By configuring reasonable tolerance distances, the retrieval scope can be effectively expanded and the diversity of retrieval results increased while ensuring retrieval efficiency.

[0037] 3. After obtaining the first sample set through preliminary retrieval, a second sample set that meets the target model is retrieved from the historical sample database based on the same constraints. This two-stage sample retrieval technique further refines the retrieval results, improving the accuracy and reliability of the data. Simultaneously, by performing mode analysis on the operational indicator record data set, a reference state for the operational indicators is obtained, and the monitoring state of the operational indicators is corrected to obtain the updated state. This data correction technique eliminates outliers and errors in the monitoring data, improving the accuracy and credibility of the data evaluation.

[0038] Example 2:

[0039] like Figure 2 As shown, based on the same inventive concept as the method for evaluating the operating index data of a thermal power unit provided in Embodiment 1, this embodiment of the invention also provides a system for evaluating the operating index data of a thermal power unit, the system comprising:

[0040] The information acquisition module 11 is used to obtain the component service life list, operating environment monitoring information and operation command timing information of the first thermal power unit of the target model.

[0041] The data retrieval module 12 is used to retrieve a set of operational indicator records that meet the target model based on a historical sample library when the component service duration list is inconsistent with the component reference service duration list, or the operating environment monitoring information is inconsistent with the reference operating environment, using the component reference service duration list, the reference operating environment, and the operation command timing information as constraints.

[0042] The data evaluation module 13 is used to perform mode analysis on the set of recorded operation indicators, obtain the reference status of operation indicators, correct the monitoring status of operation indicators, obtain the update status of operation indicators, and perform operation indicator data evaluation.

[0043] Furthermore, the data retrieval module 12 is also used to perform the following steps:

[0044] Obtain a set of component service duration lists and a set of operating environment monitoring information for the personnel being assessed; perform cluster analysis on the component service duration lists to obtain multi-cluster component service duration lists, and take the centroid component service duration list of the largest cluster as the baseline component service duration list; perform cluster analysis on the operating environment monitoring information set to obtain multi-cluster operating environment monitoring information, and take the centroid operating environment monitoring information of the largest cluster as the baseline operating environment.

[0045] Furthermore, the data retrieval module 12 is also used to perform the following steps:

[0046] The component model set is weighted and distributed to obtain the component model weight distribution result; based on the component model weight distribution result, the component service duration list set is weighted and Euclidean distance is calculated pairwise to obtain the Euclidean distribution distance set; based on the Euclidean distribution distance threshold and combined with the Euclidean distribution distance set, the component service duration list set is clustered to obtain the multi-cluster component service duration list.

[0047] Furthermore, the data retrieval module 12 is also used to perform the following steps:

[0048] The component model set and the target model are sent to several importance evaluation parties that do not communicate with each other to obtain several sets of importance scores; the several sets of importance scores are counted to obtain the sum of importance scores for the first model up to the sum of importance scores for the Nth model, and the sum of importance scores; the sum of importance scores for the first model up to the sum of importance scores for the Nth model is iterated and compared with the sum of importance scores to obtain the weight distribution result of the component model.

[0049] Furthermore, the data retrieval module 12 is also used to perform the following steps:

[0050] The importance score values ​​of the aforementioned sets are statistically analyzed to obtain the first model importance score set up to the Nth model importance score set; outliers are removed from the first model importance score set up to the Nth model importance score set to obtain the first model set importance score up to the Nth model set importance score set; the first model set importance score up to the Nth model set importance score set are summed to obtain the total importance score value; the first model set importance score up to the Nth model set importance score set are summed separately to obtain the total importance score value up to the Nth model set importance score set.

[0051] Furthermore, the data retrieval module 12 is also used to perform the following steps:

[0052] Configure service duration tolerance Euclidean distance, operating environment tolerance Euclidean distance, and operation sequence tolerance distance, wherein the operation sequence distance is equal to the ratio of the number of out-of-order to the number of in-order operation instructions; based on the service duration tolerance Euclidean distance, the operating environment tolerance Euclidean distance, and the operation sequence tolerance distance, and using the component baseline service duration list, the baseline operating environment, and the operation instruction timing information as a benchmark, construct a first retrieval constraint; retrieve the set of operation indicator record data that satisfies the first retrieval constraint and the target model from the historical sample database.

[0053] Furthermore, the data retrieval module 12 is also used to perform the following steps:

[0054] From the historical sample library, a first sample set that satisfies the first retrieval constraint and the target model is retrieved; based on the service duration fault tolerance Euclidean distance, the operating environment fault tolerance Euclidean distance, and the operation sequence fault tolerance distance, the first sample set is traversed and a second sample set that satisfies the target model is retrieved from the historical sample library; the operating indicator record data sets of the first sample set and the second sample set are statistically analyzed.

[0055] Through the foregoing detailed description of a method for evaluating the operating index data of a thermal power unit, those skilled in the art can clearly understand the operating index data evaluation system of a thermal power unit in this embodiment. As the system disclosed in the embodiment corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to the method section.

[0056] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating the operating performance data of thermal power units, characterized in that, The method includes: Obtain the component service life list, operating environment monitoring information, and operation command timing information of the first thermal power unit of the target model; When the component service duration list is inconsistent with the component baseline service duration list, or when the operating environment monitoring information is inconsistent with the baseline operating environment, the component baseline service duration list, the baseline operating environment, and the operation command timing information are used as constraints, and a set of operating indicator records that meet the target model is retrieved based on the historical sample library. A mode analysis is performed on the set of recorded operational indicators to obtain a reference status for the operational indicators. The monitoring status of the operational indicators is then corrected to obtain an updated status for the operational indicators. Finally, an evaluation of the operational indicator data is performed. When the component service life list is inconsistent with the component baseline service life list, or when the operating environment monitoring information is inconsistent with the baseline operating environment, including: Obtain a list of component service durations and a set of operating environment monitoring information for the personnel being assessed; Cluster analysis is performed on the set of component service duration lists to obtain multi-cluster component service duration lists. The centroid component service duration list of the largest cluster is taken and set as the reference service duration list of the components. Cluster analysis is performed on the set of operating environment monitoring information to obtain multi-cluster operating environment monitoring information. The centroid operating environment monitoring information of the largest cluster is taken as the benchmark operating environment. Constrained by the component reference service duration list, the reference operating environment, and the operation command timing information, and based on the historical sample database, a set of operational indicator records that meet the target model is retrieved, including: Configure service duration fault tolerance Euclidean distance, operating environment fault tolerance Euclidean distance, and operation sequence fault tolerance distance, wherein the operation sequence distance is equal to the ratio of the number of out-of-order operations to the number of in-order operations. Based on the service duration tolerance Euclidean distance, the operating environment tolerance Euclidean distance, and the operation sequence tolerance distance, and using the component baseline service duration list, the baseline operating environment, and the operation instruction timing information as a benchmark, a first retrieval constraint is constructed. From the historical sample database, retrieve the set of operational indicator records that satisfy the first retrieval constraint and the target model.

2. The method as described in claim 1, characterized in that, Cluster analysis is performed on the set of component service duration lists to obtain multi-cluster component service duration lists. The centroid component service duration list of the largest cluster is selected and set as the baseline component service duration list, including: Perform a weighted distribution on the component model set to obtain the component model weighted distribution results; Based on the component model weight distribution results, pairwise weighted Euclidean distance calculations are performed on the component service duration list set to obtain the Euclidean distribution distance set. Based on the Euclidean distribution distance threshold and combined with the Euclidean distribution distance set, cluster analysis is performed on the component service duration list set to obtain the multi-cluster component service duration list.

3. The method as described in claim 2, characterized in that, The component model set is weighted and distributed to obtain the component model weight distribution results, including: The component model set and the target model are sent to several importance evaluation parties that do not communicate with each other to obtain several sets of importance score values; Statistically calculate the importance score values ​​of the aforementioned groups to obtain the total importance score of the first model up to the total importance score of the Nth model, and the total importance score value; The sum of importance scores for the first model is iterated up to the sum of importance scores for the Nth model, and each sum is compared with the sum of importance scores to obtain the weight distribution result of the component model.

4. The method as described in claim 3, characterized in that, Statistically calculate the importance score values ​​of the aforementioned groups to obtain the total importance score of the first model up to the total importance score of the Nth model, and the total importance score value, including: By statistically analyzing the aforementioned importance score values, we obtain the first model importance score set up to the Nth model importance score set. Traverse the first model importance score set up to the Nth model importance score set and remove outliers to obtain the first model set importance score up to the Nth model set importance score; The importance scores of the first model set are summed up up to the importance scores of the Nth model set to obtain the total importance score value; Iterate through the importance scores of the first model up to the importance scores of the Nth model, sum them up to obtain the total importance scores of the first model up to the total importance scores of the Nth model.

5. The method as described in claim 1, characterized in that, From the historical sample database, retrieve a set of operational indicator records that satisfy the first retrieval constraint and the target model, including: From the historical sample database, retrieve a first sample set that satisfies the first retrieval constraint and the target model; By utilizing the smaller service life tolerance Euclidean distance, the operating environment tolerance Euclidean distance, and the operation sequence tolerance distance, the first sample set is traversed to retrieve the second sample set that meets the target model. The data set of the operational indicators recorded in the first sample set and the second sample set are statistically analyzed.

6. A system for evaluating the operational performance data of thermal power units, characterized in that, The system is used to implement the method for evaluating the operating index data of a thermal power unit according to any one of claims 1-5, the system comprising: The information acquisition module is used to obtain the component service life list, operating environment monitoring information, and operation command timing information of the first thermal power unit of the target model; The data retrieval module is used to retrieve a set of operational indicator records that meet the target model based on a historical sample library when the component service duration list is inconsistent with the component reference service duration list, or the operating environment monitoring information is inconsistent with the reference operating environment, using the component reference service duration list, the reference operating environment, and the operation command timing information as constraints. The data evaluation module is used to perform mode analysis on the set of recorded operational indicators, obtain the reference status of operational indicators, correct the monitoring status of operational indicators, obtain the update status of operational indicators, and perform operational indicator data evaluation.

Images