Power monitoring data quality evaluation system and method

Abstract

The invention relates to an electric power monitoring data quality evaluation system and method, and belongs to the technical field of electric power informatization. According to the system and the method provided by the invention, the metadata model definition module establishes the metadata model; the data extraction module obtains sampling data from the power monitoring system; the evaluation analysis module generates a data quality evaluation result table of each piece of sampling data in the current time period; the database stores a metadata model table, a sampling data table and a quality evaluation result table; the evaluation report module generates an evaluation report for human-computer interface display; the evaluation report shows data integrity, validity and timeliness index value distribution and index volatility distribution of the sampling data objects in the statistical time period in a chart mode. Therefore, the electric power monitoring data quality evaluation system and method which are based on a unified framework, realize electric power monitoring business data sampling and quality evaluation and form a visual evaluation conclusion are provided, and an important decision basis can be provided for a business department to improve the electric power monitoring data quality.

Description

Technical Field

[0001] The present invention relates to the field of electric power information technology, and in particular to an electric power monitoring data quality assessment system and method. Background Art

[0002] The power monitoring system is a basic information platform for power companies, power generation companies, and production companies to ensure power production and supply. Within the business scope of power companies, there are various power monitoring systems according to different business areas. Taking the distribution system of State Grid Shanghai Electric Power Company as an example, it includes SCADA / DSCADA systems for monitoring the main equipment of substations, power and environmental monitoring systems for dynamic environment / flooding / smoke monitoring of substations, medium-voltage cable monitoring systems for distribution cable fault monitoring, and power consumption collection systems for substations and users.

[0003] The data collection of various monitoring systems has its commonality. Generally, data is collected at regular intervals and saved in the historical database of each monitoring system. The collected data has multiple data attributes for different equipment objects. The data objects have upper and lower limits and validity ranges, and have typical time series data characteristics.

[0004] In the actual production process, data quality problems may occur due to various reasons such as channel quality, abnormal network communication, monitoring system software problems, etc. Specific manifestations include short-term / long-term data interruption, abnormal data jump, data storage time lag, etc. These data quality problems will affect the data integrity, effectiveness and timeliness of the monitoring system, and thus affect the development of various data-related businesses.

[0005] At present, power companies lack a unified and effective solution to the above problems, and lack a unified evaluation capability for the data quality of various monitoring systems. In recent years, power companies have continued to strengthen data management, and have clearly stated that data quality management is the basis and guarantee for data application and data value release, covering the entire chain of data creation, storage, transmission, maintenance, and application, and requiring all departments to conduct data quality verification based on data application and professional development needs and relying on data quality governance tools.

[0006] Therefore, how to provide a power monitoring data quality assessment method and system that can form intuitive evaluation conclusions and provide decision-making basis for business departments to improve the quality of power monitoring data has become an urgent problem to be solved in this field. Summary of the invention

[0007] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and to provide a power monitoring data quality assessment system and method based on a unified architecture to implement power monitoring business data sampling and quality assessment, and to form intuitive evaluation conclusions, so as to provide a decision-making basis for business departments to improve the quality of power monitoring data.

[0008] In order to achieve the above-mentioned purpose, the power monitoring data quality assessment system of the present invention has the following structure:

[0009] The database is used to define the metadata model, sampling data, and evaluation result data required for the operation of the system, and store the corresponding metadata model table, sampling data table, and evaluation result table;

[0010] The metadata model definition module is used to provide metadata model editing and definition. Users enter data model records into the database through the human-computer interface and establish metadata models through single record entry or batch data import.

[0011] A data extraction module, for acquiring sampling data at time intervals of hourly minutes or hours through a data interface adapted to the power monitoring system according to the metadata model definition, and saving the data in a sampling data table of the database;

[0012] Evaluation and analysis module, which runs scheduled tasks by hour, day, week, month, and year, and generates a data quality evaluation result table for each sampled data in the current period;

[0013] The evaluation report module is used to generate an evaluation report based on the query conditions preset by the user and return it to the human-computer interface for display;

[0014] A data access interface, used to provide a data access API interface for the metadata model table, the sampling data table and the evaluation result table in the database;

[0015] A human-computer interface provides a graphical human-computer interface for metadata model editing, evaluation report display and query. The evaluation report graphically represents the data integrity, validity and timeliness index numerical distribution and index volatility distribution of the sampled data objects during the statistical period.

[0016] In this power monitoring data quality assessment system,

[0017] The metadata model table is used to define metadata definitions of data measurement points used for data quality assessment by monitoring sampling of the power monitoring system;

[0018] The sampling data table is used to store data records extracted from the power monitoring system;

[0019] The evaluation results are used to store data quality evaluation result records.

[0020] In the power monitoring data quality assessment system, the integrity indicators include:

[0021] Data integrity refers to the proportion of data that is not missing. For each sampling object, the integrity index calculation formula during the statistical period is:

[0022]

[0023] misscnt=totalcnt-cnt

[0024] Where cnt is the number of all data records in the specified statistical period obtained in the sampling database;

[0025] totalcnt is the total number of data records that should be obtained during the statistical period:

[0026]

[0027] T sta is the duration of the statistical period, T sample The sampling interval defined for the target measuring point in the power monitoring system;

[0028] Data integrity standard deviation:

[0029]

[0030] where x i Represents the integrity value of the daily, weekly, monthly, and annual time periods, is the average value of integrity values ​​in the corresponding period, and n is the number of integrity values;

[0031] Data integrity coefficient of variation:

[0032]

[0033] Where CompletnessSD represents the completeness standard deviation within the statistical period. is the average value of the integrity value in the corresponding period;

[0034] The larger the value of data integrity coefficient of variation (CompletnessCV), the worse the stability of the integrity value indicator; the smaller the value, the better the stability of the integrity value indicator.

[0035] In the power monitoring data quality assessment system, the timeliness indicators include:

[0036] Data timeliness is used to evaluate the timeliness of sampled data. It measures the proportion of time series data that arrives on time without delay. The calculation formula is:

[0037]

[0038] Among them, cnt represents the total number of sampled data rows in the statistical period, and latecnt represents the delay count; the latecnt statistical method is that the time difference between the power monitoring system data storage time SystemSampleTime and the device data upload time stamp is greater than the set delay threshold;

[0039] Data timeliness standard deviation:

[0040]

[0041] where x i Represents the time value of the day, week, month, and year time period, is the average value of the timeliness in the corresponding period, and n is the number of timeliness values.

[0042] Data timeliness coefficient of variation:

[0043]

[0044] Among them, TimelinessSD represents the standard deviation of timeliness within the statistical period. It is the average value of timeliness in the corresponding period; the larger the value of the coefficient of variation of data timelinessCV, the worse the stability of the timeliness value indicator, and the smaller the value, the better the stability of the timeliness value indicator.

[0045] In the power monitoring data quality assessment system, the effectiveness indicators include:

[0046] Data validity is used to calculate the validity of time series and measures the proportion of data that meets the constraints. The calculation formula is:

[0047]

[0048] Among them, cnt represents the total number of rows of sampled data in the statistical period, valuecnt represents the count of abnormal data records; the valuecnt statistical method is the number of data values ​​of the power monitoring system that exceed the upper or lower limit range;

[0049] Data validity standard deviation:

[0050]

[0051] where x i Represents the validity value of the day, week, month, and year time periods. is the average of the validity values ​​in the corresponding period, and n is the number of validity values;

[0052] Data validity coefficient of variation:

[0053]

[0054] ValiditySD represents the standard deviation of data validity within the statistical period. It is the average value of the validity value in the corresponding period. The larger the value of the data validity coefficient of variation ValidityCV is, the worse the stability of the validity value indicator is; the smaller it is, the better the stability of the timeliness value indicator is.

[0055] The present invention also provides a method for evaluating the quality of power monitoring data, the method comprising the following steps:

[0056] The user enters the data model record into the database through the human-computer interface, and imports the metadata information into the metadata model definition module by single record entry or batch data import;

[0057] The metadata model definition module provides metadata model editing definition and establishes a metadata model according to the metadata information;

[0058] The data extraction module obtains sampling data at intervals of hours, minutes or hours through a data interface adapted to the power monitoring system according to the metadata model definition, and saves it to the database;

[0059] The database stores a metadata model table according to the obtained metadata information, and stores a sampling data table according to the obtained sampling data;

[0060] The evaluation and analysis module runs scheduled tasks by hour, day, week, month, and year, and generates a data quality evaluation result table for each sampled data in the current period based on the metadata model and the sampled data;

[0061] The database stores the data quality assessment result table;

[0062] The evaluation report module generates an evaluation report based on the query conditions preset by the user and returns it to the human-computer interface for display;

[0063] The evaluation report graphically represents the numerical distribution of data integrity, validity and timeliness indicators of the sampled data objects during the statistical period and the volatility distribution of the indicators.

[0064] The power monitoring data quality assessment system and method of the present invention are adopted. The user enters the data model record into the database and imports the metadata information into the metadata model definition module; the metadata model definition module establishes the metadata model; the data extraction module obtains the sampled data from the power monitoring system and saves it into the database; the assessment and analysis module generates the data quality assessment result table of each sampled data in the current period; the database stores the metadata model table, the sampled data table and the quality assessment result table; the assessment report module generates the assessment report and provides a human-computer interface display; the assessment report graphically represents the data integrity, validity and timeliness index value distribution and the volatility distribution of the index of the sampled data object in the statistical period. Thus, a power monitoring data quality assessment system and method based on a unified architecture is provided to realize the sampling and quality assessment of power monitoring business data and form an intuitive evaluation conclusion, which can provide an important decision-making basis for business departments to improve the quality of power monitoring data. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of the system structure of the power monitoring data quality assessment system of the present invention;

[0066] Figure 2 A schematic diagram of data flow of the power monitoring data quality assessment method of the present invention;

[0067] Figure 3 Schematic diagram of the process flow of the power monitoring data quality assessment method of the present invention. DETAILED DESCRIPTION

[0068] In order to more clearly understand the technical content of the present invention, the following embodiments are given in detail.

[0069] See also Figure 1 FIG. 1 is a schematic diagram of the system structure of the power monitoring data quality assessment system of the present invention.

[0070] In one embodiment, the power monitoring data quality assessment system includes:

[0071] The database is used to define the metadata model, sampling data, and evaluation result data required for the operation of the system, and store the corresponding metadata model table, sampling data table, and evaluation result table;

[0072] The metadata model definition module is used to provide metadata model editing and definition. Users enter data model records into the database through the human-computer interface and establish metadata models through single record entry or batch data import.

[0073] A data extraction module, for acquiring sampling data at time intervals of hourly minutes or hours through a data interface adapted to the power monitoring system according to the metadata model definition, and saving the data in a sampling data table of the database;

[0074] Evaluation and analysis module, which runs scheduled tasks by hour, day, week, month, and year, and generates a data quality evaluation result table for each sampled data in the current period;

[0075] The evaluation report module is used to generate an evaluation report based on the query conditions preset by the user and return it to the human-computer interface for display;

[0076] A data access interface, used to provide a data access API interface for the metadata model table, the sampling data table and the evaluation result table in the database;

[0077] A human-computer interface provides a graphical human-computer interface for metadata model editing, evaluation report display and query. The evaluation report graphically represents the data integrity, validity and timeliness index numerical distribution and index volatility distribution of the sampled data objects during the statistical period.

[0078] In the method corresponding to the system of this embodiment, if Figure 2 As shown in FIG. 1 , it is a data flow diagram of the power monitoring data quality assessment method of the present invention. Figure 3 As shown, the following steps are included:

[0079] The user enters the data model record into the database through the human-computer interface, and imports the metadata information into the metadata model definition module by single record entry or batch data import;

[0080] The metadata model definition module provides metadata model editing definition and establishes a metadata model according to the metadata information;

[0081] The data extraction module obtains the sampled data according to the metadata model definition through the data interface adapted to the power monitoring system at intervals of every hour, minute or hour, such as 1 minute, 5 minutes, 15 minutes, 30 minutes, or 1 hour every hour, and saves it to the database;

[0082] The database stores a metadata model table according to the obtained metadata information, and stores a sampling data table according to the obtained sampling data;

[0083] The evaluation and analysis module runs scheduled tasks by hour, day, week, month, and year, and generates a data quality evaluation result table for each sampled data in the current period based on the metadata model and the sampled data;

[0084] The database stores the data quality assessment result table;

[0085] The evaluation report module generates an evaluation report based on the query conditions preset by the user and returns it to the human-computer interface for display;

[0086] The evaluation report graphically represents the numerical distribution of data integrity, validity and timeliness indicators of the sampled data objects during the statistical period and the volatility distribution of the indicators.

[0087] In a preferred embodiment,

[0088] The metadata model table is used to define metadata definitions of data measurement points used for data quality assessment by monitoring sampling of the power monitoring system;

[0089] The sampling data table is used to store data records extracted from the power monitoring system;

[0090] The evaluation results are used to store data quality evaluation result records.

[0091] In a more preferred embodiment,

[0092] The integrity indicators include:

[0093] Data integrity refers to the proportion of data that is not missing. For each sampling object, the integrity index calculation formula during the statistical period is:

[0094]

[0095] misscnt=totalcnt-cnt

[0096] Where cnt is the number of all data records in the specified statistical period obtained in the sampling database;

[0097] totalcnt is the total number of data records that should be obtained during the statistical period:

[0098]

[0099] T sta is the duration of the statistical period, T sample The sampling interval defined for the target measuring point in the power monitoring system;

[0100] Data integrity standard deviation:

[0101]

[0102] where x i Represents the integrity value of the daily, weekly, monthly, and annual time periods, is the average value of integrity values ​​in the corresponding period, and n is the number of integrity values;

[0103] Data integrity coefficient of variation:

[0104]

[0105] Where CompletnessSD represents the completeness standard deviation within the statistical period. is the average value of the integrity value in the corresponding period;

[0106] The larger the value of data integrity coefficient of variation (CompletnessCV), the worse the stability of the integrity value indicator; the smaller the value, the better the stability of the integrity value indicator.

[0107] The timeliness indicators include:

[0108] Data timeliness is used to evaluate the timeliness of sampled data. It measures the proportion of time series data that arrives on time without delay. The calculation formula is:

[0109]

[0110] Among them, cnt represents the total number of sampled data rows in the statistical period, and latecnt represents the delay count; the latecnt statistical method is that the time difference between the power monitoring system data storage time SystemSampleTime and the device data upload time stamp is greater than the set delay threshold;

[0111] Data timeliness standard deviation:

[0112]

[0113] where x i Represents the time value of the day, week, month, and year time period, is the average value of the timeliness in the corresponding period, and n is the number of timeliness values.

[0114] Data timeliness coefficient of variation:

[0115]

[0116] Among them, TimelinessSD represents the standard deviation of timeliness within the statistical period. It is the average value of timeliness in the corresponding period; the larger the value of the coefficient of variation of data timelinessCV, the worse the stability of the timeliness value indicator, and the smaller the value, the better the stability of the timeliness value indicator.

[0117] The effectiveness indicators include:

[0118] Data validity is used to calculate the validity of time series and measures the proportion of data that meets the constraints. The calculation formula is:

[0119]

[0120] Among them, cnt represents the total number of rows of sampled data in the statistical period, valuecnt represents the count of abnormal data records; the valuecnt statistical method is the number of data values ​​of the power monitoring system that exceed the upper or lower limit range;

[0121] Data validity standard deviation:

[0122]

[0123] where x i Represents the validity value of the day, week, month, and year time periods. is the average of the validity values ​​in the corresponding period, and n is the number of validity values;

[0124] Data validity coefficient of variation:

[0125]

[0126] ValiditySD represents the standard deviation of data validity within the statistical period. It is the average value of the validity value in the corresponding period. The larger the value of the data validity coefficient of variation ValidityCV is, the worse the stability of the validity value indicator is; the smaller it is, the better the stability of the timeliness value indicator is.

[0127] In the actual application of the present invention, the system structure of the power monitoring data quality evaluation system of the present invention is as follows: Figure 1 As shown, the system consists of multiple modules such as database, metadata model definition, data extraction, evaluation analysis, evaluation report, data access interface, human-computer interface, etc.

[0128] (1) Database: defines the metadata model, sampling data, evaluation result data, etc. required for system operation.

[0129] a) Metadata model table, which defines the metadata definition of the data measurement points for data quality assessment of the target power monitoring system monitoring sampling. The data table definition is as follows:

[0130]

[0131] Table 1 Metadata model table b) Sampling data, storing data records extracted from the target power monitoring system.

[0132]

[0133]

[0134] Table 2 Sampling data table c) Evaluation result table, storing data quality evaluation result records.

[0135]

[0136] (2) Metadata model definition

[0137] This module provides metadata model editing and definition. Users can enter data model records into the database through the human-computer interface. Metadata models can be established through single record entry or batch data import.

[0138] (3) Data extraction

[0139] According to the metadata model definition, this module obtains the measurement point data at time intervals of hourly minutes or hours through the data interface adapted by the power monitoring system, and saves it in the sampling data table of the database.

[0140] (4) Evaluation and analysis

[0141] This module runs scheduled tasks on an hourly, daily, weekly, monthly, and yearly basis to generate data quality assessment results for each sampled data in the current period.

[0142] (5) Evaluation report

[0143] This module generates an evaluation report based on the query conditions preset by the user, such as monitoring system, site, communication equipment, measurement point, statistical time period, etc., and returns it to the human-computer interface for display.

[0144] (6)Data access interface

[0145] Provides data access API interface for each data table in the database.

[0146] (7) Human-machine interface

[0147] Provides a graphical human-computer interface for metadata model editing and evaluation report query. The evaluation report graphically represents the distribution of data integrity, validity and timeliness indicators and the distribution of indicator volatility (variation coefficient of each indicator) of the sampled data objects during the statistical period.

[0148] The data flow of the power monitoring data quality assessment method of the present invention is as follows: Figure 2 shown.

[0149] The method for calculating the quality evaluation index of power monitoring data using the present invention includes:

[0150] (1) Integrity indicators

[0151] a) Data Integrity

[0152] Data integrity refers to the proportion of data that is not missing. For each sampling object, the integrity index calculation formula within the statistical period (hour, day, week, month, year) is:

[0153]

[0154] misscnt=totalcnt-cnt

[0155] Where cnt is the number of all data records in the specified statistical period obtained in the sampling database;

[0156] totalcnt is the total number of data records that should be obtained during the statistical period:

[0157] T sta is the duration of the statistical period (unit: seconds), T sample The sampling interval (unit: seconds) defined for the target measurement point in the power monitoring system.

[0158] b) Data integrity standard deviation

[0159]

[0160] where x i Represents the integrity value of the daily, weekly, monthly, and annual time periods, is the average value of integrity values ​​in the corresponding period, and n is the number of integrity values.

[0161] c) Data integrity coefficient of variation

[0162]

[0163] Among them, CompletnessSD represents the completeness standard deviation within the statistical period (day, week, month, year), It is the average value of the integrity value in the corresponding period. The larger the value of the data integrity coefficient of variation, the worse the stability of the integrity value indicator; the smaller the CV, the better the stability of the integrity value indicator.

[0164] (2) Timeliness index

[0165] a) Data timeliness

[0166] Timeliness is used to evaluate the timeliness of sampled data, and measures the proportion of time series data that arrives on time without delay. The calculation formula is:

[0167]

[0168] Among them, cnt represents the total number of rows of sampled data in the statistical period, and latecnt represents the delay count.

[0169] The latent statistical method is that the time difference between the power monitoring system data storage time SystemSampleTime and the device data sending time stamp is greater than the set delay threshold (such as 3 seconds).

[0170] b) Data timeliness standard deviation

[0171]

[0172] where x i Represents the time value of the day, week, month, and year time period, is the average value of the timeliness in the corresponding period, and n is the number of timeliness values.

[0173] c) Data timeliness coefficient of variation

[0174]

[0175] TimelinessSD represents the standard deviation of timeliness within the statistical period (day, week, month, year). It is the average value of timeliness in the corresponding period. The larger the value of the coefficient of variation of the timeliness of the data, the worse the stability of the timeliness value indicator; the smaller the TimelinessCV, the better the stability of the timeliness value indicator.

[0176] (3) Effectiveness indicators

[0177] a) Data validity

[0178] Validity is used to calculate the validity of a time series, and measures the proportion of data that meets the constraints. The calculation formula is:

[0179]

[0180] Among them, cnt represents the total number of rows of sampled data in the statistical period, and valuecnt represents the count of abnormal data records.

[0181] The valuecnt statistical method is the number of power monitoring system data values ​​that exceed the upper or lower limit range.

[0182] b) Data validity standard deviation

[0183]

[0184] where x i Represents the validity value of the day, week, month, and year time periods. is the average of the validity values ​​in the corresponding period, and n is the number of validity values.

[0185] c) Data validity coefficient of variation

[0186]

[0187] ValiditySD represents the standard deviation of data validity within the statistical period (day, week, month, year). It is the average value of the validity value in the corresponding period. The larger the value of the data validity coefficient of variation ValidityCV, the worse the stability of the validity value indicator; the smaller the ValidityCV, the better the stability of the timeliness value indicator.

[0188] The power monitoring data quality assessment system and method of the present invention are adopted. The user enters the data model record into the database and imports the metadata information into the metadata model definition module; the metadata model definition module establishes the metadata model; the data extraction module obtains the sampled data from the power monitoring system and saves it into the database; the assessment and analysis module generates the data quality assessment result table of each sampled data in the current period; the database stores the metadata model table, the sampled data table and the quality assessment result table; the assessment report module generates the assessment report and provides a human-computer interface display; the assessment report graphically represents the data integrity, validity and timeliness index value distribution and the volatility distribution of the index of the sampled data object in the statistical period. Thus, a power monitoring data quality assessment system and method based on a unified architecture is provided to realize the sampling and quality assessment of power monitoring business data and form an intuitive evaluation conclusion, which can provide an important decision-making basis for business departments to improve the quality of power monitoring data.

[0189] In this specification, the present invention has been described with reference to specific embodiments thereof. However, it is apparent that various modifications and variations may be made without departing from the spirit and scope of the present invention. Therefore, the specification and drawings should be regarded as illustrative rather than restrictive.

Claims

1. A power monitoring data quality assessment system, characterized in that: include: The database is used to define the metadata model, sampling data, and evaluation result data required for the operation of the system, and store the corresponding metadata model table, sampling data table, and evaluation result table; The metadata model definition module is used to provide metadata model editing and definition. Users enter data model records into the database through the human-computer interface and establish metadata models through single record entry or batch data import. A data extraction module, for acquiring sampling data at time intervals of hourly minutes or hours through a data interface adapted to the power monitoring system according to the metadata model definition, and saving the data in a sampling data table of the database; Evaluation and analysis module, which runs scheduled tasks by hour, day, week, month, and year, and generates a data quality evaluation result table for each sampled data in the current period; The evaluation report module is used to generate an evaluation report based on the query conditions preset by the user and return it to the human-computer interface for display; A data access interface, used to provide a data access API interface for the metadata model table, the sampling data table and the evaluation result table in the database; A human-computer interface provides a graphical human-computer interface for metadata model editing, evaluation report display and query. The evaluation report graphically represents the data integrity, validity and timeliness index numerical distribution and index volatility distribution of the sampled data objects during the statistical period.

2. The power monitoring data quality assessment system according to claim 1, characterized in that: The metadata model table is used to define metadata definitions of data measurement points used for data quality assessment by monitoring sampling of the power monitoring system; The sampling data table is used to store data records extracted from the power monitoring system; The evaluation results are used to store data quality evaluation result records.

3. The power monitoring data quality assessment system according to claim 1, characterized in that: The integrity indicators include: Data integrity refers to the proportion of data that is not missing. For each sampling object, the integrity index calculation formula during the statistical period is: misscnt=totalcnt-cnt Where cnt is the number of all data records in the specified statistical period obtained in the sampling database; totalcnt is the total number of data records that should be obtained during the statistical period: T sta is the duration of the statistical period, T sample The sampling interval defined for the target measuring point in the power monitoring system; Data integrity standard deviation: Where xi represents the integrity value of the daily, weekly, monthly, and annual time periods. is the average value of integrity values ​​in the corresponding period, and n is the number of integrity values; Data integrity coefficient of variation: Among them, CompletnessSD represents the completeness standard deviation within the statistical period. is the average value of the integrity value in the corresponding period; The larger the value of data integrity coefficient of variation (CompletnessCV), the worse the stability of the integrity value indicator; the smaller the value, the better the stability of the integrity value indicator.

4. The power monitoring data quality assessment system according to claim 1, characterized in that: The timeliness indicators include: Data timeliness is used to evaluate the timeliness of sampled data. It measures the proportion of time series data that arrives on time without delay. The calculation formula is: Among them, cnt represents the total number of sampled data rows in the statistical period, and latecnt represents the delay count; the latecnt statistical method is that the time difference between the power monitoring system data storage time SystemSampleTime and the device data upload time stamp is greater than the set delay threshold; Standard deviation of data timeliness: where x i Represents the time value of the day, week, month, and year time period, is the average value of timeliness in the corresponding period, and n is the number of timeliness values. Data timeliness coefficient of variation: Among them, TimelinessSD represents the standard deviation of timeliness within the statistical period. It is the average value of timeliness in the corresponding period; the larger the value of the coefficient of variation of data timelinessCV, the worse the stability of the timeliness value indicator, and the smaller the value, the better the stability of the timeliness value indicator.

5. The power monitoring data quality assessment system according to claim 1, characterized in that: The effectiveness indicators include: Data validity is used to calculate the validity of time series and measures the proportion of data that meets the constraints. The calculation formula is: Among them, cnt represents the total number of rows of sampled data in the statistical period, valuecnt represents the count of abnormal data records; the valuecnt statistical method is the number of data values ​​of the power monitoring system that exceed the upper or lower limit range; Data validity standard deviation: where x i Represents the validity value of the day, week, month, and year time periods. is the average of the validity values ​​in the corresponding period, and n is the number of validity values; Data validity coefficient of variation: ValiditySD represents the standard deviation of data validity within the statistical period. It is the average value of the validity value in the corresponding period. The larger the value of the data validity coefficient of variation ValidityCV is, the worse the stability of the validity value indicator is; the smaller the value is, the better the stability of the timeliness value indicator is.

6. A method for evaluating the quality of power monitoring data, characterized in that: The method comprises the following steps: The user enters the data model record into the database through the human-computer interface, and imports the metadata information into the metadata model definition module by single record entry or batch data import; The metadata model definition module provides metadata model editing definition and establishes a metadata model according to the metadata information; The data extraction module obtains sampling data at intervals of hours, minutes or hours through a data interface adapted to the power monitoring system according to the metadata model definition, and saves it to the database; The database stores a metadata model table according to the obtained metadata information, and stores a sampling data table according to the obtained sampling data; The evaluation and analysis module runs scheduled tasks by hour, day, week, month, and year, and generates a data quality evaluation result table for each sampled data in the current period based on the metadata model and the sampled data; The database stores the data quality assessment result table; The evaluation report module generates an evaluation report based on the query conditions preset by the user and returns it to the human-computer interface for display; The evaluation report graphically represents the numerical distribution of data integrity, validity and timeliness indicators of the sampled data objects during the statistical period and the volatility distribution of the indicators.

7. The method for evaluating the quality of power monitoring data according to claim 6, characterized in that: The metadata model table is used to define metadata definitions of data measurement points used for data quality assessment by monitoring sampling of the power monitoring system; The sampling data table is used to store data records extracted from the power monitoring system; The evaluation results are used to store data quality evaluation result records.

8. The method for evaluating the quality of power monitoring data according to claim 6, characterized in that: The integrity indicators include: Data integrity refers to the proportion of data that is not missing. For each sampling object, the integrity index calculation formula during the statistical period is: misscnt=totalcnt-cnt Where cnt is the number of all data records in the specified statistical period obtained in the sampling database; totalcnt is the total number of data records that should be obtained during the statistical period: T sta is the duration of the statistical period, T sample The sampling interval defined for the target measuring point in the power monitoring system; Data integrity standard deviation: where x i Represents the integrity value of the daily, weekly, monthly, and annual time periods, is the average value of integrity values ​​in the corresponding period, and n is the number of integrity values; Data integrity coefficient of variation: Among them, CompletnessSD represents the completeness standard deviation within the statistical period. is the average value of the integrity value in the corresponding period; The larger the value of data integrity coefficient of variation (CompletnessCV), the worse the stability of the integrity value indicator; the smaller the value, the better the stability of the integrity value indicator.

9. The method for evaluating the quality of power monitoring data according to claim 6, characterized in that: The timeliness indicators include: Data timeliness is used to evaluate the timeliness of sampled data. It measures the proportion of time series data that arrives on time without delay. The calculation formula is: Among them, cnt represents the total number of sampled data rows in the statistical period, and latecnt represents the delay count; the latecnt statistical method is that the time difference between the power monitoring system data storage time SystemSampleTime and the device data upload time stamp is greater than the set delay threshold; Standard deviation of data timeliness: where x i Represents the time value of the day, week, month, and year time period, is the average value of timeliness in the corresponding period, and n is the number of timeliness values. Data timeliness coefficient of variation: Among them, TimelinessSD represents the standard deviation of timeliness within the statistical period. It is the average value of timeliness in the corresponding period; the larger the value of the coefficient of variation of data timelinessCV, the worse the stability of the timeliness value indicator, and the smaller the value, the better the stability of the timeliness value indicator.

10. The method for evaluating the quality of power monitoring data according to claim 6, characterized in that: The effectiveness indicators include: Data validity is used to calculate the validity of time series and measures the proportion of data that meets the constraints. The calculation formula is: Among them, cnt represents the total number of rows of sampled data in the statistical period, valuecnt represents the count of abnormal data records; the valuecnt statistical method is the number of data values ​​of the power monitoring system that exceed the upper or lower limit range; Data validity standard deviation: where x i Represents the validity value of the day, week, month, and year time periods. is the average of the validity values ​​in the corresponding period, and n is the number of validity values; Data validity coefficient of variation: ValiditySD represents the standard deviation of data validity within the statistical period. It is the average value of the validity value in the corresponding period. The larger the value of the data validity coefficient of variation ValidityCV is, the worse the stability of the validity value indicator is; the smaller the value is, the better the stability of the timeliness value indicator is.

Images