Real-time evaluation method and system for power production data quality

Abstract

The invention discloses a real-time evaluation method and system for power production data quality, and belongs to the technical field of power generation industry data security. Classified data quality measurement points and overall data quality measurement points are created in a real-time database according to the types of measurement points in an I / O (input / output) list of a power production monitoring system; an evaluation rule is established on the basis of five items of data including effective measuring points, normal data quality information, non-out-of-range data, normal multi-selection measuring point data and timely data refreshing, the quality of the power production data is evaluated in real time according to the types of the measuring points, and the production data quality is comprehensively evaluated from the aspects of effectiveness, accuracy and timeliness of the data. Abnormal conditions in the production process are accurately positioned, the use value of data is guaranteed, and the requirements of power enterprises for big data analysis and data security management are met.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of data security in the power generation industry, and particularly relates to a real-time evaluation method and system for power production data quality. BACKGROUND

[0002] Under the background of the current big data era, the power industry is accelerating the digital transformation, and big data analysis technology is increasingly applied in the fields of generator unit state monitoring, energy efficiency optimization, fault warning, etc. However, the power production environment has the characteristics of strong coupling, multi-variable and non-linearity. The data is abnormal due to problems such as equipment aging, sensor drift and communication interruption during production. In addition, the influence of special working conditions such as maintenance shutdown and operation intervention makes the real-time data have quality problems such as missing, out-of-range and time scale misplacement. At present, power enterprises generally use DCS and other monitoring systems to collect massive production data. However, due to the lack of unified data quality evaluation standard, there are problems such as non-uniform data standards and single quality evaluation dimension among different production systems. The traditional post-inspection method cannot meet the real-time requirement, and simple data checking rules cannot fully reflect the core indicators such as effectiveness, accuracy and timeliness of the measuring point. This lagging and one-sided nature of data quality control not only leads to distorted big data analysis results, but also may cause safety hazards such as false alarms and misoperations. The existing technology has not established a multi-dimensional real-time evaluation system covering the whole process of power production, and it is urgent to develop an intelligent evaluation scheme that can dynamically track data quality changes, automatically identify abnormal periods and support graded warning. SUMMARY

[0003] The purpose of the present application is to provide a real-time evaluation method and system for power production data quality to overcome the technical problem that the evaluation of data quality by various departments in the prior art is not standardized, resulting in a large amount of data that does not meet the quality requirements.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present application is as follows: A real-time evaluation method for power production data quality, comprising the following steps: S1, creating classified data quality measuring points and overall data quality measuring points in the real-time database according to the measuring point types in the I / O list of the power production monitoring system; S2, establishing evaluation rules based on five data items of measuring point effectiveness, normal data quality information, data not out-of-range, normal multi-selected one measuring point data and timely data refresh; S3, collecting real-time data based on classified data quality measuring points, overall data quality measuring points and evaluation rules at a set frequency, using a simple ratio method to calculate the scores of classified data quality measuring points in real time, and using a weighted method to calculate the scores of overall data quality measuring points according to the number of classified measuring points; S4 removes the evaluation results of downtime periods from the overall data quality measurement point score, and finally generates a production data quality report after statistically analyzing the extreme values ​​and the mean.

[0005] Preferably, the measurement points are organized according to the I / O list of the power production monitoring system, including their range, quantity, and type. They are then further divided into categories based on production equipment, power production process subsystems, and operation monitoring screens. Classified quality assessment measurement points and overall quality assessment measurement points are then created in the real-time database.

[0006] Preferably, the "based on valid measuring points" specifically refers to valid measuring points in the power production monitoring system, not abandoned measuring points, and having a fixed data source; the "normal data quality information" specifically refers to the normal signal quality information inherent in the real-time data of the power production monitoring system; the "data not exceeding the range" specifically refers to the real-time data of the measuring point being within the range; the "normal data from multiple selected measuring points" specifically refers to the deviation of the data from other measuring points among multiple analog monitoring measuring points at the same location not exceeding a set threshold; and the "timely data refresh" specifically refers to the data of the analog measuring points changing with the time scale, and the deviation between the inherent time scale and the system clock not exceeding a set time threshold.

[0007] Preferably, the evaluation rules specifically include the validity, accuracy, and timeliness of the indicators. When both the validity of the measurement points and the normality of the data quality information are valid, the validity indicator is satisfied. When both the accuracy indicator and the data from multiple selected measurement points are normal are satisfied, the validity indicator is satisfied. When the data is updated in a timely manner, the timeliness indicator is satisfied.

[0008] Preferably, the weights of the three evaluation indicators, effectiveness, accuracy, and timeliness, are 0.3, 0.4, and 0.3, respectively.

[0009] Preferably, the power production monitoring system specifically targets the DCS system. The measurement point types of the DCS system include four categories: AI, AO, DI, and DO. It is further divided according to the dimensions of production equipment, power production process subsystem, and operation monitoring screen. According to production equipment, it can be divided into coal mill, coal feeder, primary air fan, secondary air fan, circulating water pump, and feed water pump. According to production process subsystem or operation monitoring screen, it can be divided into pulverizing system, boiler steam-water system, desulfurization system, denitrification system, and circulating water system.

[0010] A real-time evaluation system for power production data quality includes a measurement point data processing module, a rule establishment module, a measurement point scoring module, and an evaluation module. The measurement point data processing module creates categorized data quality measurement points and overall data quality measurement points in the real-time database based on the measurement point types in the I / O list of the power production monitoring system. The rule-setting module establishes evaluation rules based on five data criteria: valid measurement points, normal data quality information, data within the range, normal data from multiple selected measurement points, and timely data updates. The measurement point scoring module collects real-time data at a set frequency based on the classification data quality measurement points, the overall data quality measurement points, and the evaluation rules. It uses a simple ratio method to calculate the classification data quality measurement point score in real time and a weighted method to calculate the overall data quality measurement point score based on the number of classification measurement points. The evaluation module removes the evaluation results of downtime periods from the overall data quality measurement point scores, and finally generates a production data quality report after statistically analyzing the extreme values ​​and the mean.

[0011] Preferably, the measurement points are organized according to the I / O list of the power production monitoring system, including their range, quantity, and type. They are then further divided into categories based on production equipment, power production process subsystems, and operation monitoring screens. Classified quality assessment measurement points and overall quality assessment measurement points are then created in the real-time database.

[0012] Preferably, the "based on valid measuring points" specifically refers to valid measuring points in the power production monitoring system, not abandoned measuring points, and having a fixed data source; the "normal data quality information" specifically refers to the normal signal quality information inherent in the real-time data of the power production monitoring system; the "data not exceeding the range" specifically refers to the real-time data of the measuring point being within the range; the "normal data from multiple selected measuring points" specifically refers to the deviation of the data from other measuring points among multiple analog monitoring measuring points at the same location not exceeding a set threshold; and the "timely data refresh" specifically refers to the data of the analog measuring points changing with the time scale, and the deviation between the inherent time scale and the system clock not exceeding a set time threshold.

[0013] Preferably, the evaluation rules specifically include the validity, accuracy, and timeliness of the indicators. When both the validity of the measurement points and the normality of the data quality information are valid, the validity indicator is satisfied. When both the accuracy indicator and the data from multiple selected measurement points are normal are satisfied, the validity indicator is satisfied. When the data is updated in a timely manner, the timeliness indicator is satisfied.

[0014] Compared with the prior art, the present invention has the following beneficial technical effects: This invention provides a real-time assessment method for the quality of power production data. Based on the types of measurement points in the I / O list of the power production monitoring system, categorized data quality measurement points and overall data quality measurement points are created in a real-time database. Assessment rules are established based on five data criteria: measurement point validity, normal data quality information, data within range, normal data from multiple selected measurement points, and timely data updates. The quality of power production data is assessed in real-time according to the type of measurement point. This comprehensive evaluation of production data quality, considering its validity, accuracy, and timeliness, allows for accurate identification of anomalies in the production process, ensures the usability of the data, and meets the needs of power companies for big data analysis and data security management. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the real-time evaluation method for power production data quality in an embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram of the process structure of the DCS system production system in an embodiment of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] In existing technologies, the power industry generally suffers from inconsistent data quality assessment standards, and differences in data specifications between various production systems lead to low efficiency in quality monitoring. Traditional methods rely on manual sampling and offline analysis, which cannot capture data deviations caused by equipment anomalies or sensor failures in real time. In a thermal power plant operation monitoring scenario, due to the lack of a dynamic evaluation mechanism, the pressure measuring points of the pulverizing system continuously generated out-of-range data during equipment maintenance, but the outliers were not identified and removed in time, resulting in distorted subsequent energy efficiency analysis results.

[0020] To address these challenges, the R&D team identified three core difficulties in data quality assessment: real-time performance, multi-dimensionality, and dynamic weight allocation. By analyzing the data flow characteristics of the DCS system, they proposed combining measurement point classification management with a rule engine to establish a hierarchical assessment model. To mitigate data interference during downtime, a time-window-based anomaly removal algorithm was designed to ensure that the assessment results reflect the true operational status. like Figure 1 As shown, this invention provides a real-time evaluation method for the quality of power production data, specifically including the following steps: S1. Based on the measurement point types in the I / O list of the power production monitoring system, create classified data quality measurement points and overall data quality measurement points in the real-time database. S2 establishes evaluation rules based on five data items: valid measurement points, normal data quality information, data within the range, normal data from multiple selected measurement points, and timely data updates. S3 collects real-time data at a set frequency based on the classification data quality measurement points, the overall data quality measurement points and the evaluation rules. It uses the simple ratio method to calculate the classification data quality measurement point score in real time and uses the weighted method to calculate the overall data quality measurement point score based on the number of classification measurement points. S4 removes the evaluation results of downtime periods from the overall data quality measurement point score, and finally generates a production data quality report after statistically analyzing the extreme values ​​and the mean.

[0021] In a specific embodiment of this application, the range, quantity, and type of measurement points are organized according to the I / O list of the power production monitoring system, and secondary division is carried out according to the dimensions of production equipment, power production process subsystem, and operation monitoring screen. Classified quality assessment measurement points and overall quality assessment measurement points are created in the real-time database.

[0022] In a specific embodiment of this application, the evaluation rules based on five data points—valid measurement points, normal data quality information, data within the measurement range, normal data from multiple selected measurement points, and timely data updates—specifically include: The "valid measurement point" specifically refers to the power production monitoring system's measurement points being valid, not abandoned, and having a fixed data source; the "normal data quality information" specifically refers to the power production monitoring system's real-time data having normal inherent quality information; the "data not exceeding the range" specifically refers to the measurement point's real-time data being within its range; the "normal data from multiple selected measurement points" specifically refers to the deviation of a single measurement point's data from other measurement point data at the same location not exceeding a set threshold, which is set to 5% in this specific embodiment; the "timely data refresh" specifically refers to the analog measurement point's data changing with the time scale, with the inherent time scale deviating from the system clock by no more than a set time threshold, which is 5 seconds.

[0023] The evaluation rules specifically include the effectiveness, accuracy, and timeliness of the indicators. The effectiveness indicator is met when both the measurement point validity and the normal data quality information are valid. The effectiveness indicator is met when both the data not exceeding the range and the data from multiple selected measurement points being normal are accurate. The timeliness indicator is met when the data is updated in a timely manner.

[0024] In a specific embodiment of this application, a power production monitoring system is evaluated as effective when the monitoring point is valid, not an abandoned monitoring point, has a fixed data source, and the signal quality information inherent in the real-time data of the power production monitoring system is normal. When the real-time data of the monitoring point is within its range, and the data from multiple selected monitoring points is normal (specifically, among multiple analog monitoring points at the same location, the deviation of a single monitoring point's data from other monitoring point data does not exceed a set threshold), the evaluation is considered high accuracy. When the data of the analog monitoring point changes with time scale, and the deviation between its inherent time scale and the system clock does not exceed a set time threshold, the evaluation is considered timely. The weights of the three evaluation indicators—effectiveness, accuracy, and timeliness—are 0.3, 0.4, and 0.3, respectively.

[0025] Specifically, after the real-time database periodically collects measurement point data, the validity verification module first filters out abandoned measurement points and abnormal data from the source. The accuracy assessment unit performs range verification and redundant measurement point consistency checks on the measurement points that pass the validity test. The timeliness judgment module compares the data timescale with the system clock and marks data with excessive delay. The classification scoring module calculates the percentage of valid measurement points that meet the three indicators. For example, if 80% of the measurement points in the water pump category meet the validity, 75% meet the accuracy, and 90% meet the timeliness in a certain period, then the classification score is calculated with weights of 0.3, 0.4, and 0.3 respectively. The overall scoring module weights the scores according to the proportion of measurement points in each category and finally generates a quality report that includes extreme value and mean indicators.

[0026] The validity weight of 0.3 indicates the relative importance of this indicator in the overall data quality assessment. This weight can be determined using expert scoring or the analytic hierarchy process (AHP). Adjusting the weight reflects the differences in the impact of different indicators on data quality. The accuracy weight of 0.4 indicates a higher priority for this indicator in the assessment system. This can be quantified through statistical analysis of historical data anomalies. A higher weight is beneficial for monitoring key process parameters. The timeliness weight of 0.3 reflects the fundamental role of this indicator in the assessment. This can be determined through time series analysis to assess the impact of data delays on the system. A moderate weight ensures basic monitoring needs are met while avoiding over-sensitivity.

[0027] Specifically, in real-time calculation of the overall data quality measurement point score, the effectiveness index is achieved by detecting the status of measurement points and data quality identifiers; the accuracy index is achieved by range verification and redundant measurement point comparison; and the timeliness index is achieved by time-stamp verification. The three indicators are weighted and summed with weight coefficients of 0.3, 0.4, and 0.3 respectively to form a comprehensive quality score. The weight allocation scheme quantifies the impact of different quality dimensions, ensuring that data anomalies in key process links such as the boiler steam-water system receive greater attention, while avoiding excessive impact from data fluctuations in auxiliary equipment such as the coal feeder on the overall assessment. By establishing a weight allocation mechanism based on quantitative analysis, links prone to data exceeding the range, such as the pulverizing system, receive more accurate monitoring, while optimizing the data delay detection sensitivity of time-sensitive systems such as the desulfurization system. This application effectively solves the problem of monitoring focus shift caused by unreasonable weight settings in traditional assessment methods, enabling timely identification of data accuracy defects in key equipment such as the circulating water system, while ensuring reasonable evaluation of the data refresh status of the operation monitoring screen, providing a reliable weight calculation benchmark for filtering abnormal data during unit start-up and shutdown.

[0028] This application shortens the evaluation cycle to minutes by leveraging the millisecond-level data capture capability of a real-time database. Existing single-dimensional evaluation methods ignore the differences in data characteristics among different equipment types. This application dynamically adjusts the classification weights, giving higher weight to the accuracy of turbine vibration data than to temperature data, thus better meeting actual operational needs.

[0029] Through the above technical solutions, this application realizes the transformation of data quality assessment from static sampling to dynamic monitoring, effectively identifying real-time faults such as sensor drift and communication interruption. The classification and scoring mechanism accurately locates problematic equipment subsystems, the weighted algorithm objectively reflects the overall data health, and the time-period filtering function eliminates interference from non-operating data, providing reliable data support for production decisions. The present invention provides a real-time evaluation method for the quality of power production data, which is applicable to various power production monitoring systems such as DCS, NCS, and power billing systems, including primary monitoring points and secondary calculation points.

[0030] In a specific embodiment of the present invention, the quality of data after the creation of measurement points can be evaluated by supplementing historical data.

[0031] This application further proposes to organize the scope, quantity, and type of measurement points according to the I / O list of the power production monitoring system, and to perform secondary division according to the dimensions of production equipment, power production process subsystems, and operation monitoring screens, and to create classified quality assessment measurement points and overall quality assessment measurement points in the real-time database.

[0032] The I / O inventory, which organizes the scope, quantity, and type of measurement points, refers to extracting measurement point information from the input / output configuration files of the power production monitoring system. This includes information such as the equipment areas covered by the measurement points, the total number of points, and the data types. Specifically, this can be achieved by parsing the configuration files of the DCS system or the configuration database of the SCADA system, ensuring that all valid measurement points are included in the evaluation scope.

[0033] The production equipment dimension refers to classifying measurement points according to the physical equipment they are associated with in the production process, such as independent equipment units like coal mills and water pumps. This can be achieved through the association between equipment codes and measurement point tags, facilitating data quality monitoring for specific equipment.

[0034] The power generation process subsystem dimension refers to classifying measuring points according to their functional modules within the power generation process, such as pulverizing systems and desulfurization systems. Specifically, this can be done by categorizing measuring points according to their control logic group or process flow layer, which is beneficial for assessing data reliability at the subsystem level. The operation monitoring screen dimension refers to grouping measuring points according to their display location on the user interface, such as boiler monitoring screens and turbine monitoring screens. This can be achieved by binding measuring points to HMI screen elements, facilitating quick location of the corresponding user interface for abnormal data.

[0035] Specifically, during the real-time database creation process, the currently valid measurement points are first selected based on the I / O inventory, excluding those marked as obsolete. Then, the measurement points are categorized according to production equipment, such as coal mill-related measurement points and feedwater pump-related measurement points; according to process subsystems, they are classified into pulverizing system measurement points and desulfurization system measurement points; and simultaneously, they are assigned to corresponding operation interface groups based on the operation monitoring screen. The set of measurement points at each classification level is defined as the classification quality assessment measurement points, while the set of all measurement points serves as the overall quality assessment measurement points. For example, the classification measurement points for primary air blower equipment may include parameters such as blower speed and bearing temperature, while the classification measurement points for the pulverizing system may include parameters such as pulverized coal flow rate and coal mill current. This application solves the problems of incomplete assessment coverage and low anomaly location efficiency caused by the single classification dimension of measurement points in existing technologies. By using multi-dimensional secondary segmentation, the mapping relationship between measuring points and equipment, subsystems and operating interfaces can be accurately established. For example, in the event of a fan failure, the data quality of the relevant measuring points of the equipment, the measuring points of the boiler steam-water subsystem, and the corresponding monitoring screen measuring points can be analyzed simultaneously, thereby improving the accuracy of anomaly diagnosis and the comprehensiveness of the evaluation results.

[0036] In a specific embodiment of the present invention, based on the measurement point types in the I / O list of the power production monitoring system, the power production monitoring system specifically targets the DCS system. The measurement point types of the DCS system specifically include four categories: AI, AO, DI, and DO. These are further divided according to the dimensions of production equipment, power production process subsystems, and operation monitoring screens. Based on production equipment, they can be divided into coal mills, coal feeders, primary air fans, secondary air fans, circulating water pumps, and feedwater pumps. Based on the production process subsystems or operation monitoring screens, they can be divided into pulverizing systems, boiler steam-water systems, desulfurization systems, denitrification systems, and circulating water systems, such as... Figure 2 As shown, classification quality assessment points and overall quality assessment points are created in the real-time database.

[0037] The various statistical reports generated by this method can be used for performance evaluation of power production and maintenance personnel, which helps to improve the quality of power production data and enhance the maintenance level of production equipment and the operation and maintenance level of DCS system.

[0038] This invention addresses anomalies in the power production process and problems in the operation and maintenance of power production monitoring systems. It performs real-time evaluation of the quality of power production data according to the type of measurement point, comprehensively assessing the quality of production data in terms of validity, accuracy, and timeliness. It accurately locates anomalies in the production process, ensures the usability of the data, and meets the needs of power companies for big data analysis and data security management.

[0039] Valid measurement points refer to measurement points that are not marked as abandoned and have a stable data source. This can be achieved by verifying the measurement point status identifier and the stability of the communication link, thus eliminating the impact of invalid measurement points on the evaluation results. Normal data quality information means that the integrity and consistency markers carried by real-time data are in a normal state. This can be achieved by parsing the quality flag bits in the data packet, thus identifying data damaged during transmission. Data within range means that the collected values ​​are within the preset sensor range. This can be achieved by comparing real-time values ​​with the upper and lower limits of the range stored in the database, thus detecting sensor malfunctions or abnormal operating conditions. Normal data from multiple redundant measurement points means that the numerical differences between multiple redundant measurement points at the same monitoring location are within the allowable range. This can be achieved by calculating the standard deviation or judging the maximum deviation threshold, thus identifying misjudgments caused by anomalies in a single measurement point. Timely data refresh means that the synchronization error between the data timestamp and the system clock is controllable. This can be achieved through timestamp comparison and clock synchronization protocols, thus ensuring that the timeliness of the data meets the real-time evaluation requirements.

[0040] Specifically, during the operation of the power production monitoring system, invalid data sources can be filtered out by verifying the validity of measuring points in real time, for example, marking abandoned measuring points as not participating in the scoring. The data quality information parsing module continuously monitors message quality indicators and automatically excludes abnormal data when communication interruption or data verification failure is detected. The range verification module compares real-time data with range data in the equipment technical parameter database, and immediately triggers an alarm if an over-limit is found. For multiple-select measuring points located in critical positions such as boiler steam-water systems, the deviation calculation module periodically calculates the data differences between measuring points in the same group, and marks an abnormality when the deviation of a measuring point exceeds a set threshold (e.g., the allowable deviation of the circulating water pump pressure monitoring point is 2% of the range). The time synchronization module maintains system clock synchronization through the NTP protocol and performs downweighting on data with time scale deviations exceeding a set time threshold (e.g., 5 seconds).

[0041] The examples described in this invention are only for illustrating the invention and do not provide a detailed description of all the technical solutions. Substitutions or modifications made by those skilled in the art to some of the technical implementation processes do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the embodiments of this invention.

Claims

1. A real-time evaluation method for the quality of power production data, characterized in that, Includes the following steps: S1. Based on the measurement point types in the I / O list of the power production monitoring system, create classified data quality measurement points and overall data quality measurement points in the real-time database. S2 establishes evaluation rules based on five data items: valid measurement points, normal data quality information, data within the range, normal data from multiple selected measurement points, and timely data updates. S3 collects real-time data at a set frequency based on the classification data quality measurement points, the overall data quality measurement points and the evaluation rules. It uses the simple ratio method to calculate the classification data quality measurement point score in real time and uses the weighted method to calculate the overall data quality measurement point score based on the number of classification measurement points. S4 removes the evaluation results of downtime periods from the overall data quality measurement point score, and finally generates a production data quality report after statistically analyzing the extreme values ​​and the mean.

2. The real-time evaluation method for power production data quality according to claim 1, characterized in that, Based on the I / O register of the power production monitoring system, the scope, quantity, and type of measurement points are organized, and secondary classification is carried out according to the dimensions of production equipment, power production process subsystems, and operation monitoring screens. Classified quality assessment measurement points and overall quality assessment measurement points are created in the real-time database.

3. The real-time evaluation method for power production data quality according to claim 2, characterized in that, The "valid measurement point" specifically refers to the power production monitoring system's measurement points being valid, not abandoned, and having a fixed data source; the "normal data quality information" specifically refers to the power production monitoring system's real-time data having normal inherent quality information; the "data within range" specifically refers to the measurement point's real-time data being within its range; the "normal data from multiple selected measurement points" specifically refers to the deviation of a single measurement point's data from other measurement points' data not exceeding a set threshold among multiple analog monitoring measurement points at the same location; and the "timely data refresh" specifically refers to the analog measurement point's data changing with the time scale, and the deviation between its inherent time scale and the system clock not exceeding a set time threshold.

4. The real-time evaluation method for power production data quality according to claim 1, characterized in that, The evaluation rules specifically include the effectiveness, accuracy, and timeliness of the indicators. The effectiveness indicator is met when both the measurement point validity and the normal data quality information are valid. The effectiveness indicator is met when both the data not exceeding the range and the data from multiple selected measurement points being normal are accurate. The timeliness indicator is met when the data is updated in a timely manner.

5. The real-time evaluation method for power production data quality according to claim 4, characterized in that, The weights for the three evaluation indicators, effectiveness, accuracy, and timeliness, are 0.3, 0.4, and 0.3, respectively.

6. The real-time evaluation method for power production data quality according to claim 1, characterized in that, The power production monitoring system specifically targets the DCS system. The measurement point types of the DCS system include four categories: AI, AO, DI, and DO. It is further divided according to the dimensions of production equipment, power production process subsystems, and operation monitoring screens. According to production equipment, it can be divided into coal mill, coal feeder, primary air fan, secondary air fan, circulating water pump, and feed water pump. According to production process subsystems or operation monitoring screens, it can be divided into pulverizing system, boiler steam-water system, desulfurization system, denitrification system, and circulating water system.

7. A real-time evaluation system for the quality of power production data, characterized in that, It includes a data processing module, a rule establishment module, a data scoring module, and an evaluation module. The measurement point data processing module creates categorized data quality measurement points and overall data quality measurement points in the real-time database based on the measurement point types in the I / O list of the power production monitoring system. The rule-setting module establishes evaluation rules based on five data criteria: valid measurement points, normal data quality information, data within the range, normal data from multiple selected measurement points, and timely data updates. The measurement point scoring module collects real-time data at a set frequency based on the classification data quality measurement points, the overall data quality measurement points, and the evaluation rules. It uses a simple ratio method to calculate the classification data quality measurement point score in real time and a weighted method to calculate the overall data quality measurement point score based on the number of classification measurement points. The evaluation module removes the evaluation results of downtime periods from the overall data quality measurement point scores, and finally generates a production data quality report after statistically analyzing the extreme values ​​and the mean.

8. The real-time evaluation system for power production data quality according to claim 1, characterized in that, Based on the I / O register of the power production monitoring system, the scope, quantity, and type of measurement points are organized, and secondary classification is carried out according to the dimensions of production equipment, power production process subsystems, and operation monitoring screens. Classified quality assessment measurement points and overall quality assessment measurement points are created in the real-time database.

9. A real-time evaluation system for power production data quality according to claim 8, characterized in that, The "valid measurement point" specifically refers to the power production monitoring system's measurement points being valid, not abandoned, and having a fixed data source; the "normal data quality information" specifically refers to the power production monitoring system's real-time data having normal inherent quality information; the "data within range" specifically refers to the measurement point's real-time data being within its range; the "normal data from multiple selected measurement points" specifically refers to the deviation of a single measurement point's data from other measurement points' data not exceeding a set threshold among multiple analog monitoring measurement points at the same location; and the "timely data refresh" specifically refers to the analog measurement point's data changing with the time scale, and the deviation between its inherent time scale and the system clock not exceeding a set time threshold.

10. A real-time evaluation system for power production data quality according to claim 7, characterized in that, The evaluation rules specifically include the effectiveness, accuracy, and timeliness of the indicators. The effectiveness indicator is met when both the measurement point validity and the normal data quality information are valid. The effectiveness indicator is met when both the data not exceeding the range and the data from multiple selected measurement points being normal are accurate. The timeliness indicator is met when the data is updated in a timely manner.

Images