Power failure data analysis method and device, electronic equipment and storage medium

Abstract

The invention provides a power failure data analysis method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining power failure initial data; performing compliance inspection on the power failure initial data, and performing integration cleaning on the inspected data to obtain first data; performing quality inspection and correction on the first data according to a set rule to obtain second data; inputting the second data into a preset analysis model, and analyzing the power failure range, the power failure reason and the power failure loss to obtain an analysis result; and outputting the analysis result.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a power outage data analysis method, device, electronic device and storage medium. Background Art

[0002] With the development of social economy and the improvement of people's living standards, electricity has become an indispensable and important energy source in modern society. Power outages not only affect people's normal life and work, but may also cause significant economic losses to industrial production and commercial operations. Therefore, the power industry pays more and more attention to the reliability of power supply and is committed to reducing the occurrence of power outages and reducing the risks and losses of power outages.

[0003] However, due to the complexity of the grid coverage area and the users involved, electricity consumption data is usually scattered in different systems and in different formats, which brings challenges to data integration. At the same time, due to equipment failures, human operational errors and other reasons, data may be distorted or missing, affecting the accuracy of analysis results. Ultimately, the current power outage data analysis has problems such as low overall efficiency and poor accuracy, which affects the user experience. Summary of the invention

[0004] In view of this, the present application proposes a power outage data analysis method, device, electronic device and storage medium to solve or partially solve the above-mentioned problems.

[0005] Based on the above purpose, the present application provides a power outage data analysis method, including:

[0006] Get the initial data of power outage;

[0007] Performing a compliance check on the initial power outage data, and integrating and cleaning the checked data to obtain first data;

[0008] Performing quality inspection and correction on the first data according to a set rule to obtain second data;

[0009] Inputting the second data into a preset analysis model, analyzing the power outage scope, power outage cause and power outage loss, and obtaining an analysis result;

[0010] The analysis result is outputted.

[0011] In some exemplary embodiments, the obtaining of power outage initial data includes:

[0012] Basic data acquisition is performed through ports adapted to external terminals;

[0013] The basic data is sorted and traced according to the set requirements, the association relationship between the data items in the basic data is established, and the data items are graded and classified according to the association relationship, so as to generate the initial power outage data.

[0014] In some exemplary embodiments, the step of integrating and cleaning the checked data includes:

[0015] Performing unit conversion and encoding conversion on the checked data;

[0016] Determine the fields with the same meaning in the checked data, and merge the fields with the same meaning.

[0017] In some exemplary embodiments, the step of integrating and cleaning the checked data includes:

[0018] The checked data is subjected to non-empty check, primary key duplication cleaning, illegal code cleaning, illegal value cleaning, data format check and record number check.

[0019] In some exemplary embodiments, after obtaining the first data, the method further includes:

[0020] The first data is verified and a data directory is constructed according to the first data.

[0021] In some exemplary embodiments, the performing quality inspection and correction on the first data according to a set rule includes:

[0022] Performing a data timeliness check, uniqueness check, consistency check, integrity check, and rationality check on the first data to identify problematic data therein;

[0023] The problematic data is traced back and compared with historical data to correct the problematic data.

[0024] In some exemplary embodiments, inputting the second data into a preset analysis model to analyze the power outage scope, power outage cause, and power outage loss includes:

[0025] Construct power outage scope analysis model, power outage cause analysis model and power outage loss analysis model based on random forest algorithm and clustering algorithm;

[0026] The second data is analyzed using the power outage range analysis model, the power outage cause analysis model and the power outage loss analysis model.

[0027] Based on the same concept, the present application also provides a power outage data analysis device, comprising:

[0028] The first module is used to obtain the initial data of power outage;

[0029] The second module is used to perform compliance check on the initial power outage data, integrate and clean the checked data, and obtain first data;

[0030] A third module is used to perform quality inspection and correction on the first data according to a set rule to obtain second data;

[0031] The fourth module is used to input the second data into a preset analysis model, analyze the power outage scope, power outage cause and power outage loss, and obtain analysis results;

[0032] The fifth module is used to output the analysis result.

[0033] Based on the same concept, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the methods described above when executing the program.

[0034] Based on the same concept, the present application also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to implement any of the methods described above.

[0035] As can be seen from the above, the present application provides a method, device, electronic device and storage medium for analyzing power outage data, the method comprising: obtaining initial power outage data; performing compliance check on the initial power outage data, integrating and cleaning the checked data to obtain first data; performing quality check and correction on the first data according to the set rules to obtain second data; inputting the second data into a preset analysis model, analyzing the power outage range, power outage cause and power outage loss to obtain analysis results; and outputting the analysis results. The present application first performs a compliance check on the acquired initial power outage data to determine whether the data is filled in according to the set value range or requirements, and then integrates the data into a unified unit and code, and cleans the unreasonable data in the data, so that each data in the first data is filled in in compliance, and the format is unified and there is no repeated data. After that, the first data is quality checked, that is, rationality checked, to determine whether the fluctuation range of the data is within a reasonable range, and the unreasonable data is adjusted and corrected to obtain the second data, and finally the corresponding model is used for power outage analysis. In this way, the format of power outage data can be quickly integrated and the rationality of data can be adjusted, thereby improving the overall efficiency and accuracy of power outage data analysis and enhancing the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 A flowchart of an exemplary method provided in an embodiment of the present application.

[0038] Figure 2 A schematic diagram of the structure of an exemplary device provided in an embodiment of the present application.

[0039] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of this specification more clear, this specification is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0041] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements, objects or method steps appearing before the word cover the elements, objects or method steps listed after the word and their equivalents, without excluding other elements, objects or method steps. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0042] As mentioned in the background technology section, power outage loss analysis involves various types of data, including power grid operation data, user electricity consumption data, socio-economic data, etc. These data are usually scattered in different systems and in different formats, which brings challenges to data integration. Power outage loss analysis has extremely high requirements for data accuracy, but in actual situations, due to equipment failures, human operational errors, etc., data may be distorted or missing, affecting the accuracy of the analysis results. Power outage events are sudden, requiring data to be updated and fed back in real time in order to timely analyze power outage losses and conduct risk assessments. However, it is difficult to obtain and process real-time data.

[0043] At the same time, power outage risk involves multiple factors, such as weather, equipment aging, human operational errors, etc. These factors may interact with each other, increasing the uncertainty of risk assessment. Due to the complexity and randomness of power outage events, risk prediction and quantification are very difficult. Current risk assessment methods are mainly based on experience and expert judgment, lacking scientific quantification and prediction methods. Risk assessment is a subjective process, and the experience and knowledge of the assessor will affect the results. In addition, risk assessment is often based on the current system and environment, but as technology and environment change, the assessment results may become outdated or inaccurate.

[0044] In combination with the above-mentioned actual situation, an embodiment of the present application provides a power outage data analysis solution. The present application first performs a compliance check on the acquired initial power outage data to determine whether the data is filled in according to the set value range or requirements, and then integrates the data into a unified unit and code, and cleans the unreasonable data in the data, so that each data in the first data obtained is filled in in compliance, and the format is unified and there is no duplicate data. Afterwards, a quality check, that is, a rationality check, is performed on the first data to determine whether the fluctuation range of the data is within a reasonable range, and the unreasonable ones are adjusted and corrected to obtain the second data, and finally the corresponding model is used for power outage analysis. In this way, the format of the power outage data can be quickly integrated and the data rationality can be adjusted, thereby improving the overall efficiency and accuracy of the power outage data analysis and improving the user experience.

[0045] Figure 1 A flow chart of an exemplary method provided in an embodiment of the present application is shown.

[0046] like Figure 1 As shown, the power outage data analysis method exemplarily proposed in the embodiment of the present application includes the following steps.

[0047] Step 102, obtaining initial power outage data.

[0048] In this step, the initial power outage data may be all data related to the user's daily electricity consumption, where the user may be an individual or an enterprise, etc. Since the power outage itself is an emergency that occurs during daily electricity use, the daily electricity consumption data itself will also record the corresponding data during the power outage, and the corresponding situation of the power outage can be analyzed based on this.

[0049] In some embodiments, through data access, the power quality is monitored online, and the relevant power consumption information is collected, the product management system (PMS), the telephone platform business support system and other systems are analyzed, and it is found that when the power quality is monitored online, the power distribution network will store the relevant data of the power outage event in the distribution network area, and then the corresponding data interface can be designed to obtain these basic data. After that, for these basic data, the data can be combed and traced according to the set rules, so as to establish the association relationship between the data, and thus the data can be graded and classified, and finally the data after classification is used as the initial power outage data. That is, in some embodiments, the acquisition of the initial power outage data includes: acquiring the basic data through the port adapted to the external terminal; combing and tracing the basic data according to the set requirements, establishing the association relationship between the data items in the basic data, and grading and classifying according to the association relationship, so as to generate the initial power outage data.

[0050] In more specific application scenarios, based on the more than 100 commonly used data items of various types that have been investigated, based on the results of differentiated comparison, and the business system data such as the procurement system, telephone platform system, product management system, power grid GIS (Geographic Information System) actually connected to the data center, according to the system docking and data center acquisition method, database association, interface acquisition and other operations are performed, and the system is connected to the power outage analysis and monitoring system, and the data tables of various business systems are initially investigated.

[0051] Collect and summarize the data requirements of the telephone platform, collect and sort out the data sharing requirements; sort out and analyze the existing data requirements table based on data application, data sharing requirements and business rules; trace the source of data, trace the source business system, data table, corresponding fields for the data items in the data requirements table, and clarify the correlation and matching relationship between data items; classify and classify the data for analysis and preparation, communicate with the data production department, and conduct data classification and analysis for the business. At the same time, coordinate the construction manufacturer to carry out the preliminary preparation work for data access.

[0052] Collect and summarize data requirements for product management systems, collect and sort out data sharing requirements; sort out and analyze existing data requirements tables based on data applications, data sharing requirements and business rules; trace the source of data, trace the source business system, data table, corresponding fields for data items in the data requirements table, and clarify the correlation and matching relationship between data items; classify and classify data for analysis and preparation, communicate with data production departments, and conduct data classification and analysis for business. At the same time, coordinate the construction manufacturer to carry out preliminary preparations for data access.

[0053] Collect and summarize the data requirements of the power grid GIS, collect and sort out the data sharing requirements; sort out and analyze the existing data requirements table based on data application, data sharing requirements and business rules; trace the source of data, trace the source business system, data table, corresponding fields for the data items in the data requirements table, and clarify the correlation and matching relationship between data items; classify and classify the data for analysis and preparation, communicate with the data production department, and conduct data classification and analysis for the business. At the same time, coordinate the construction manufacturer to carry out the preliminary preparation work for data access.

[0054] After that, preliminary data retrieval can be carried out. According to the scenario function requirements and data traceability results, the specific data requirements involved in the scenario can be sorted out, and data can be retrieved on the digital capability open platform according to the data requirement list. Data can be retrieved through the data map module of the data platform (such as Dataworks, etc.), and the data table corresponding to the required field can be screened. Data can be retrieved from the model data display page of the digital capability open platform. For data that cannot be directly retrieved, the required data can be retrieved through the State Grid Cloud Data Map page and data access can be performed. After that, further data application can be carried out, that is, after retrieving the required data, data use application can be made through the digital capability open platform to obtain the data required for scenario analysis.

[0055] Step 104, performing a compliance check on the initial power outage data, integrating and cleaning the checked data to obtain first data.

[0056] In this step, the compliance check is to examine whether each data in the initial power outage data meets the basic requirements of the data item. For example, for electricity consumption data, its reasonable value should be a number, and it should be a number of 0 and above. If a negative number or other characters appear in the corresponding data item, it can be considered non-compliant. After that, after completing the compliance check, the data can be integrated and cleaned. It can mainly integrate inconsistent units and inconsistent codes. The integration standards and rules can be set in advance to integrate the data in the corresponding fields. Finally, the data segments that appear in the set situation are cleaned and removed, such as the removal of some illegal codes, the removal of duplicate fields, etc., so that the first data can be obtained after integration and cleaning.

[0057] In some embodiments, the specific integration process may be to unify the unit conversion and encoding escape of the checked data; determine the fields with the same meaning in the checked data, and merge the fields with the same meaning. Afterwards, the cleaning process may include non-empty check, primary key duplication cleaning, illegal code cleaning, illegal value cleaning, data format check and record number check, that is, at least the above cleaning process may be performed on the checked data.

[0058] In more specific application scenarios, this step can be based on multiple system data such as electricity consumption information collection system, product management system, power grid GIS system, power supply service system, etc., according to the unified data model physical model standard (data table), to perform data cleaning, correlation matching, data conversion, data verification and other operations to form standardized data, and connect it to the customer's power outage monitoring and analysis scenario database.

[0059] First, we can combine the data demand sorting and traceability verification results, and according to the data item requirements, associated matching relationships, and business data traceability verification results of multiple hot wide tables, based on the data middle platform, carry out the design of historical stock data and incremental data extraction interface of the unified data model of hot wide tables, configure data processing calculation logic, manually calculate and extract the detailed data of the wide tables in adjacent years, perform preliminary verification on the calculated detailed data, ensure that the key field values ​​are not empty and available, and then configure the incremental data calculation tasks, including electricity application information, prepaid electricity fee information, and operating energy meter information. Information, on-site investigation and evidence collection information, customer electricity prices and charges, electricity theft case information, electricity charges receivable information, electricity usage in breach of contract information, marketing topology information, low-voltage area asset group, energy meter, terminal power outage / power-on event, daily measurement point current curve, daily frozen energy at the measurement point, daily measurement point voltage curve, power outage information, regional power outage information, electricity consumption - metering cabinets, distribution transformers, energy meter power outages, power supply units, marketing areas, line area relationships, energy meter metering point relationships, metering points, process instance information, transformer area relationships, telephone platform work orders and other information tables.

[0060] Afterwards, based on the data center, you can design the interface for extracting historical stock data and incremental data from the application result table according to the data item requirements, business logic, and business data traceability verification results of the application result table involved in multiple applications such as indicator dashboards and frequent power outages. Configure the data processing and calculation logic, manually calculate and extract the statistical analysis result data of adjacent years, perform preliminary verification on the calculated result data, ensure that the key field values ​​are not empty and available, configure the incremental data calculation tasks, and build the data calculation logic for multiple result data tables involved in multiple applications.

[0061] Third, for data integration and conversion: the data in the application result table involving multiple applications such as hot spot wide tables, indicator dashboards, frequent power outages, etc. can be sorted out and refined according to professional application needs. Statistical analysis dimension fields include dimension codes, dimension levels, dimension names, hot spot wide tables (result tables), business definitions, and professional fields. Dimensions can be classified and divided according to organizational structures, professional fields, and application scenarios, and statistical analysis dimension field data can be converted and integrated, including business data coding escape, classified data statistical summary, unit code correspondence, etc.

[0062] Fourth, for data cleaning and conversion: for data in application result tables involving multiple applications such as hotspot wide tables, indicator dashboards, frequent power outages, etc., formulate data cleaning and conversion rules based on professional application requirements, clean invalid detailed data and statistical result data in the data table, and further clean data with obviously unreasonable values. Focus on data desensitization for sensitive fields such as user names, user addresses, and contact information.

[0063] Based on the data middle platform, combined with the source data tracing results, business logic combing results and data correlation matching relationship results, for multiple commonly used wide table data needs in scenarios such as orderly electricity consumption analysis, power load characteristics analysis, short-term forecast of key industry electricity, electricity substitution analysis, change electricity consumption analysis, line loss analysis, electricity theft and breach of contract analysis, and peak and valley time-of-use electricity price sensitivity analysis, a unified data cleaning rule is determined for this part of the data.

[0064] Specific data cleaning rules may include: non-empty check, primary key duplicate cleaning, illegal code cleaning, illegal value cleaning, data format check and record number check. Among them, non-empty check: when the field is required to be non-empty, the field data needs to be checked. Primary key duplicate cleaning: after the same type of data in multiple business systems is cleaned, it is necessary to check it when it is saved in a unified manner to ensure the uniqueness of the primary key. Illegal code and illegal value cleaning: illegal code problems include illegal code, inconsistency between code and data standards, etc. Illegal value problems include value errors, format errors, extra characters, garbled characters, etc., which need to be checked and corrected according to specific circumstances. Data format check: the accuracy of the attribute value in the table is measured by checking whether its format is correct, such as time format, currency format, extra characters, garbled characters. Record number check: refers to the total number of data check between related data in each system or the fluctuation check of daily data volume in the data table.

[0065] Specifically, for missing value cleaning, missing values ​​are the most common data problem, and there are many ways to deal with missing values. You can follow the following steps: 1) Determine the range of missing values, calculate the missing value ratio for each field, and then formulate strategies according to the missing ratio and field importance. 2) Remove unnecessary fields: After the data is backed up, it can be deleted directly. 3) Fill in missing content: Some missing values ​​can be filled in. The methods can be: fill in missing values ​​with business knowledge or experience speculation; fill in missing values ​​with the calculation results of the same indicator (mean, median, mode, etc.); fill in missing values ​​with the calculation results of different indicators. 4) Re-obtain data: If some indicators are very important and have a high missing rate, you need to consult with the data collector or business personnel to see if there are other channels to obtain relevant data.

[0066] For format content cleaning, 1) the display formats of time, date, value, full-width and half-width are inconsistent. This problem is usually related to the input end, and it may also be encountered when integrating data from multiple sources. Just process it into a consistent format. 2) There are characters that should not exist in the data. Some content may only include a part of the characters. In this case, it is necessary to use a semi-automatic verification and semi-manual method to find out possible problems and remove unnecessary characters. 3) The data does not match the content of the field. For example, the name has the gender written, the ID number has the mobile phone number written, etc., all of which belong to this type of problem. However, the particularity of this problem is that it cannot be simply handled by deletion, because the cause may be manual filling errors, it may also be that the front end has not been verified, and it may be that some or all of the columns are not aligned when importing data, so the problem type is identified in detail.

[0067] For logical error cleaning, the following methods are used for logical problem data processing: 1) Understand the underlying logical rules of the data, adopt logical reasoning methods, and directly remove some data that can be found to have problems using simple logical reasoning. 2) For unimportant and unreasonable data, it is advisable to directly delete them, and for important and unreasonable values, it is advisable to conduct manual intervention or introduce more data sources for association identification. 3) Correct the contradictory content through the method of mutual verification between fields, such as judging which field provides more reliable information based on the data source of the field, and removing or reconstructing unreliable fields. 4) For complex logical data problems, consult to understand the reasons for the generation of the data, and process it according to the negotiated cleaning and processing rules.

[0068] After data cleaning is completed, the data in the analysis scenario can be further sorted out in combination with professional application requirements and statistical analysis dimension fields can be extracted, and the dimensions can be classified and divided according to organizational structure, professional field, and application scenario. Specifically, (1) According to data requirements, statistical analysis dimension fields are extracted from the aspects of dimension coding, dimension level, dimension name, data set, business definition, and professional field, and classified and divided according to organizational structure, professional field, and application scenario. Statistical dimensions can be: by year, by quarter, by month, by day, by city level, by county unit, by power supply station, by substation, etc. Statistical types can be: sum, count, year-on-year, month-on-month, etc. (2) For the statistical analysis fields of the data set, conversion and fusion work can be carried out according to the statistical dimensions. The statistical fields are mainly numerical fields.

[0069] Afterwards, in some embodiments, data verification can be performed on the first data and a data directory can be constructed based on the first data. Specifically, preliminary verification is performed on the availability of detailed data and statistical summary data in the hotspot wide table and application result data table, and preliminary comparison is performed between the front-end data of the business system and the data in the wide table and the result table. Problem verification and logic tuning are performed for the data inconsistencies found. Focusing on the application requirements of detailed data and statistical summary data in the hotspot wide table and application result data table, the access timeliness of the data tables involved in the procurement system, telephone platform system, product management system, power grid GIS, power supply service, etc. in the data middle platform is verified, and the consistency with the source end system is verified. Business experts and data personnel are organized to jointly verify the availability and validity of the data, and preliminary comparison is performed between the front-end data of the business system and the data in the wide table and the result table. Problem verification and logic tuning are performed for the data inconsistencies found. (1) Preliminary verification. The application requirement proposer is organized to conduct preliminary verification on the availability of detailed data and statistical summary data in the data table, and preliminary comparison is performed between the front-end data of the business system and the data in the result table. Problem verification and logic tuning are performed for the data inconsistencies found. (2) Secondary verification. Focusing on the application requirements of detailed data and statistical summary data in the data table, complete the access timeliness of the data tables involved in the marketing business application system, procurement system and other systems in the data middle platform, and verify the consistency with the source system. Organize business experts and data personnel to jointly verify the availability and validity of the data, conduct a preliminary comparison between the front-end data of the business system and the data in the result table, and conduct problem verification and logic optimization for any data inconsistencies found.

[0070] Afterwards, in the data catalog construction phase, the sorted data source table, hotspot wide table, and application result table are used as inputs, and hierarchical classification is carried out according to the application professional field, the unit submitting the requirements, the application scenario, and the source business system. The parent-child correspondence of the function menus at each level of the data resource catalog, as well as the correspondence between the last-level function menu and the data table are sorted out to build a hotspot wide table and application result table data resource catalog system. Furthermore, based on the sorted data source table, hotspot wide table, and application result table data resource catalog system, with the specific data items and business data in the hotspot wide table as the object, in conjunction with the hotspot wide table requirement proposer and the specific user, the hotspot wide table name and the application result table are uniquely defined to form business terms, improve the readability and authority of the data, and facilitate business personnel and data personnel to quickly find the required data. Based on the results of the data inventory, the Chinese name of the database table is used as the object, and after completing the object deduplication operation and function path merging, the unique definition of the database table name is formulated to form business terms. Taking the specific data items and business data in the hotspot wide table as the object, in conjunction with the hotspot wide table demand proposers and specific users, formulate the unique definition of the hotspot wide table name and application result table, form business terms, improve the readability and authority of the data, and facilitate business personnel and data personnel to quickly find the required data. Finally, the sorted data source table, hotspot wide table, application result table keyword tags, business terms, etc. are associated with data objects such as data tables and fields in the data resource directory system, and keyword tags, business terms, etc. are mounted on the objects to complete the initialization of the basic data of the hotspot wide table data directory and form a standardized reference to database objects.

[0071] Step 106: Perform quality check and correction on the first data according to set rules to obtain second data.

[0072] In this step, quality inspection and correction can be understood as checking the rationality of the data in the first data, such as the change range and change rules need to meet certain requirements, and the data that does not meet the requirements can be corrected with the help of historical data or context data to obtain the second data. The specific inspection items to be carried out during the quality inspection can be specifically set according to the specific scenario.

[0073] In some embodiments, the data scope of the first data needs to be determined first. Specifically, the data governance scope of the source table, hotspot wide table, and application result table can be sorted out and clarified, mainly including the historical data scope, field verification scope, data attribution unit scope, hotspot wide table related business system scope, application result table related hotspot wide table data scope, business system data table and field scope, etc. Afterwards, for the detailed data of the key verification and governance in the source table, hotspot wide table, and application result table (for example, the business detailed data can be sampled for detailed data within one month, and 10% of the data can be sampled for ledger data), key fields, etc., organize data verification personnel to carry out data quality problem analysis and sorting work for hotspot wide tables and application result tables, including data timeliness, data uniqueness, consistency of data volume, consistency of data value with business system, integrity of field value, rationality of numerical business, etc. According to the sorted data quality problem types, combined with the data in the data middle platform detailed data table, analyze the root causes of the sorted data quality problems, and determine the data quality governance technical routes for wide table data and application result table data respectively. For the problem data that has been checked, its source can be determined by tracing the source, and the problem data can be corrected by using the source, or corrected by combining historical data and normal data before and after the problem data. That is, in some embodiments, the quality inspection and correction of the first data according to the set rules include: performing a data timeliness check, uniqueness check, consistency check, integrity check and rationality check on the first data to determine the problem data therein; tracing the problem data and comparing it with historical data to correct the problem data. Among them, the consistency of data volume and the consistency of data value with the business system can be regarded as consistency checks.

[0074] In more specific application scenarios, for example, for the analysis and sorting of data timeliness issues, we can specifically compare and analyze the latest data of the extracted source table, hotspot wide table, and application result table in the past month with the latest existing data in the corresponding marketing, equipment, regulation and other business systems, and find out the problem that the latest data exists in the business system but the hotspot wide table data and the application result table data are missing. For this type of data timeliness issue, we trace the source layer by layer according to the data processing and calculation logic, locate the problem link where the missing data cannot be calculated and stored in the hotspot wide table or application result table in time, and analyze the cause of the problem with the data middle platform project team. Finally, we form data timeliness problem verification rules, cause analysis conclusions, and data problem governance methods and strategies.

[0075] For the analysis and sorting of data uniqueness problems, we can screen out the data from the extracted source tables, hotspot wide tables, and application result tables for the past month to see if there are problem data tables with consistent statistical caliber but inconsistent data values, redundant data, etc. Based on the calculation logic of the hotspot wide table data and the application result table data, combined with the data from the marketing, equipment, regulation and other business systems, we can conduct root cause analysis of inconsistent data values ​​and redundant data and locate the problem links. Finally, we can form data uniqueness problem verification rules, cause analysis conclusions, and data problem governance methods and strategies.

[0076] For the analysis and sorting of data volume consistency issues, the data volume of the source table, hot spot wide table, and application result table in the past month can be counted, and the corresponding data volume statistical results can be counted in the corresponding marketing, equipment, regulation and other business systems according to the statistical period. The two are compared and analyzed to screen out hot spot wide tables or application result tables with inconsistent data volumes. For the hot spot wide tables or application result tables with inconsistent magnitudes found, first, investigate whether there are differences between the data calculation logic of the business system background and the calculation logic of the hot spot wide table or application result table, and optimize the logic with differences; second, jointly with the data middle platform project team, trace the data with differences according to the data access, processing and calculation process, locate the problem link, and conduct root cause analysis of data volume consistency issues. Finally, the verification rules, cause analysis conclusions, and data problem governance methods and strategies for source tables, hot spot wide tables, and application result tables are formed.

[0077] For the analysis and sorting of numerical consistency issues, we can screen out key business value fields in source tables, hotspot wide tables, and application result tables, such as: monthly electricity, daily electricity, monthly electricity bills, power outage start (end) time, power outage duration, total number of equipment ledgers, etc., about 480. Compare the key field values ​​of the data sampled in the wide table and application result table in the past month to see if they are consistent with the data values ​​in the marketing, equipment, regulation and other business systems, and locate inconsistent data and field values. For the hotspot wide table or application result table where the key field value data is inconsistent with the business system data value, first, investigate whether the statistical caliber of the field value in the business system, the data processing and calculation logic are different from the statistical caliber and calculation logic of the hotspot wide table or application result table, and optimize the statistical caliber and calculation logic with differences; second, jointly with the data middle platform project team, trace the key data field values ​​with differences according to the data access, processing and calculation process, locate the problem link, and conduct root cause analysis of data value inconsistency. Finally, the key business field value problem verification rules, cause analysis conclusions and data problem governance methods and strategies for source tables, hotspot wide tables, and application result tables are formed.

[0078] For the analysis and sorting of field integrity issues, we can target multiple fields in the source table, hot spot wide table, and application result table, which can specifically include two types of basic information fields and business information fields. Based on the vacant field data in the spot-checked data of the past month, compare it with the field values ​​in the corresponding marketing, equipment, regulation and other business systems to sort out problem data such as incomplete data fields in the wide table or application result table and the business system, complete business system data fields but incomplete wide table or application result table fields, incomplete application result table fields but complete wide table and business system data fields. For the hotspot wide tables or application result tables involved in the discovered problematic data, first, investigate whether the statistical caliber of the field values ​​in the business system, the data processing and calculation logic, etc. are different from the statistical caliber and calculation logic of the hotspot wide tables or application result tables, and optimize the statistical caliber and calculation logic that are different; second, jointly with the data middle platform project team, trace the data with incomplete fields according to the data access, processing and calculation process, locate the problem link, and conduct root cause analysis of inconsistent data values; third, jointly with business application personnel, conduct business data problem tracing and root cause analysis of data with incomplete data fields in the business system, hotspot wide tables, and application result tables. Finally, form verification rules for incomplete problems in multiple fields of source tables, hotspot wide tables, and application result tables, cause analysis conclusions, and data problem governance methods and strategies.

[0079] For the analysis and sorting of numerical rationality issues, we can work with data demanders to conduct data value rationality analysis from the perspective of actual business applications for the key business value fields in the source table, hotspot wide table, and application result table for the past month, and screen out the hotspot wide table or application result table data whose data values ​​seriously do not conform to the actual business logic. For this type of problematic data, first, analyze whether there are problems with the statistical caliber and calculation logic of the hotspot wide table or application result table, and optimize the problematic statistical caliber and calculation logic; second, work with the data middle platform project team to trace the problematic data according to the data access, processing and calculation process, locate the problem link, and conduct root cause analysis of unreasonable data value issues. Finally, more than 480 key business field value rationality verification rules and cause analysis conclusions were formed for source tables, hotspot wide tables, and application result tables.

[0080] Afterwards, we can optimize and improve the rules based on the specific causes and cause analysis conclusions of the various types of data quality issues involved in the sorted source tables, hot wide tables, and application result tables, build a data quality governance rule system based on hot wide tables and application result tables, create and maintain a data quality rule library, and use the data middle-office components to deploy data quality verification rule scripts and scheduled verification tasks.

[0081] Furthermore, after completing the above-mentioned quality inspection, another quality verification can be conducted. Specifically, according to the sorted out multi-category data quality verification rules, data quality verification can be carried out on the business history data of the adjacent years in the source table, hot spot wide table, and application result table and the full amount of ledger data, and the data verification results can be output. Based on the verification results, a multi-dimensional comprehensive evaluation system for data quality is established using the analytic hierarchy process, which can comprehensively analyze and evaluate the data quality verification results in multiple dimensions, multiple levels, and in combination with expert theory quantification and stereotyping. Among them: 1) Classification of data verification results, the verification result data sets of the business history data of the adjacent years in the source table, hot spot wide table, and application result table and the full amount of ledger data can be divided into multiple sub-factor sets to form a matrix. 2) Analysis of the impact weight of problem data, the analytic hierarchy process can be used to determine the weights. The specific approach can be to invite business experts to compare multiple evaluation factors in pairs and form a corresponding judgment matrix according to the degree of importance. Then calculate the maximum eigenvalue and eigenvector of the judgment matrix, and use the consistency index, random consistency index and consistency ratio to perform consistency test. If the test passes, the normalized eigenvector is the weight vector; if it fails, it means that the differences in the considerations of the expert group members are too large, and the judgment matrix needs to be reconstructed. 3) Quantitative evaluation of data quality. An evaluation set can be introduced to quantitatively evaluate the data quality of business history data and full ledger data in adjacent years in the source table, hot spot wide table, and application result table. The evaluation takes into account the universality and easy distinction, and is divided into four levels: excellent, good, medium, and poor. Taking the degree of missing data integrity as an example, the value range is [0,1]. If the missing rate is below 0.5, it can be judged as "excellent"; if the missing rate is between 0.6 and 0.75, it can be judged as "good"; if the missing rate is between 0.75 and 0.85, it can be judged as "medium"; if the missing rate is between 0.85 and 1, it can be judged as "poor".

[0082] Afterwards, based on the data verification results, data verification personnel and data application personnel can be organized to confirm data quality issues, further clarify which are data quality issues and which are not, form a list of data quality issues in hot wide tables and application result tables that need to be rectified, and compile a data quality verification report for in-depth application of the municipal data center.

[0083] Finally, formulate a data problem rectification plan for the list of data quality problems in the manually confirmed source tables, hot spot wide tables, and application result tables. According to the data responsibility management mechanism, formulate data quality problem rectification plans for hot spot wide tables and application result tables according to different professions and systems. For the implementation of specific rectification plans, the data center operation and maintenance personnel and business system data responsible persons can be organized according to the data problem rectification plan to carry out data problem rectification work for the list of data quality problems in the detailed data tables of the source tables, hot spot wide tables, and application result tables, including the rectification of source data, circulation data, and system data quality problems, and track the rectification of data problems.

[0084] Step 108, inputting the second data into a preset analysis model, analyzing the power outage scope, power outage cause and power outage loss, and obtaining analysis results.

[0085] In this step, after the second data is determined, it can be input into the corresponding analysis model to analyze the power outage scope, power outage cause and power outage loss to form an analysis result. Specifically, the random forest algorithm and clustering algorithm can be applied to analyze the power outage cause. Among them, the power outage repair reports of different power supply companies can be associated. The increase in the number of repair reports has led to an increase in user complaints against power supply companies, so this application is of great significance in improving customer service satisfaction.

[0086] In some embodiments, the constructed analysis model may include an analysis model for studying the scope of power outages. By building a model through data analysis and calculating the power outages in different areas, the number of power outages, frequency, cause distribution, robustness index, etc. in the area can be counted, providing a comprehensive and intuitive data analysis of the power outage area, improving the efficiency of power outage management, and laying the foundation for improving quality and efficiency.

[0087] Recall and precision are the best tools to measure the accuracy of the prediction model. Similarly, the prediction model can also use these two concepts to judge its accuracy. By continuously adjusting the sampling method and ratio, the accuracy of the model can be continuously improved. Assume T p To predict the correct number of outage-sensitive customers, F p is the number of outage-sensitive customers predicted incorrectly, F n To predict the number of wrong outage insensitive customers, T n is the number of customers who are not sensitive to power outages. The precision rate Z refers to the ratio of the number of correct predictions to the number of customers who are sensitive to power outages, that is, The recall rate R refers to the ratio of the correct number of model predictions to the actual number of power outage sensitive customers, that is, Since there may be inconsistencies between the precision rate Z and the recall rate R, it is necessary to consider the two comprehensively. The most common comprehensive consideration method is to use the F measure, which is the weighted harmonic mean of the recall rate and the precision rate, that is, Where a is the weight coefficient.

[0088] In some embodiments, the constructed analysis model may also include a model for analyzing the causes of power outages. Power outages in the distribution network are mainly divided into two categories: pre-scheduled power outages and fault power outages. By analyzing the causes of power outages within the scope of the power outage, a data model is established, including pre-scheduled power outages (planned power outages, temporary power outages, power restrictions) analysis and fault power outages (internal fault power outages and external fault power outages). Combined with the ledger data to find correlations, predict the possibility of power outages, avoid fault power outages as much as possible, and reduce the losses caused by power outages.

[0089] The cause analysis model of power outage can be converted into a 0-1 integer programming problem, that is, the diagnostic hypothesis D must satisfy the condition that its corresponding E(D) value is the largest among all M+ non-redundant covering solutions.

[0090] In some embodiments, the constructed analysis model may also include studying the power outage loss analysis model. Analyze the number of users within the power outage range and the user's electricity consumption pattern, and estimate the economic losses and grid load losses of each power outage through algorithms; analyze the number of users within the power outage range, load level and user importance, and estimate the adverse social effects and negative public opinion effects that may be caused by the power outage. The formula for calculating the direct economic losses of the power supply department caused by the power outage is:

[0091] CD=∑ e＝1,2,3,4 S e Q e t

[0092] Among them, e=1, 2, 3, 4 represent agricultural, industrial, commercial and residential users in the power outage area respectively; S e is the electricity price for category e users, in yuan / kWh; Q e is the loss load of the e-th category user, in MW; t is the duration of the power outage, in hours.

[0093] After that, for the study of power outage loss assessment model:

[0094] We conducted field surveys on typical electricity users of different business scales and types in different regions, analyzed their load characteristics, power outage characteristics and production process flow, and combined them with the power outage loss survey data to obtain a specific function expression. The two complement each other and not only achieve the establishment of a specific power outage loss model for electricity users, but also overcome the shortcoming of simply adopting the user survey method that is not convincing.

[0095] 1) User loss assessment model for primary industry.

[0096] We conducted field research on plantations, nursery and flower bases, livestock farms and aquatic product breeding areas in different regions and of different operating scales. Based on their electricity consumption characteristics and power outage loss characteristics, we performed a fitting analysis of a large amount of power outage loss data and concluded that the relationship curve between the economic losses in agriculture, forestry, animal husbandry and fishery and the duration of power outages approximately satisfies the "S" type. Therefore, we can use the logistic function to establish a loss assessment model.

[0097] 2) Loss assessment model for typical users in the tertiary industry.

[0098] The main focus can be on typical users in the tertiary industry whose economic losses can be quantified after a power outage, including commerce, accommodation and catering, entertainment, finance, etc. In addition, television broadcasting, medical care, etc., power outages cause serious social impacts, and it is difficult to quantitatively assess the economic losses. Large shopping malls, star-rated hotels, entertainment venues, banks, etc. have short-term power outages during business hours, and the losses are relatively small. As the power outage continues, the emergency power supply loses its function, the number of customers decreases, and the turnover loss increases. At the same time, the refrigerated items in the refrigerators of shopping malls and hotels gradually deteriorate. When the power outage time is close to the business hours, the business is stopped and the loss value tends to stabilize. In summary, the user unit load loss first increases with the duration and then tends to stabilize. The evaluation model approximately satisfies the following formula:

[0099] CY 3 =M[1-e -(t / λ) ]

[0100] Among them, CY 3 is the economic loss of users in the tertiary industry, in ten thousand yuan; M is the stable value of loss, in ten thousand yuan; λ is the loss time constant.

[0101] 3) Model for assessing power outage losses in urban and rural residents.

[0102] The power outage losses of residents include the loss of ineffective electrical appliances, food spoilage caused by long-term power outages (such as 4 hours), and the value of other substitutes consumed due to power outages. Urban and rural residents in different regions and different economic development levels will suffer less losses from short-term power outages; long-term power outages will lead to the purchase of other forms of substitutes and the gradual spoilage of refrigerated food in refrigerators. The power outage loss assessment model approximately satisfies a two-part piecewise linear relationship.

[0103] Study the process of estimating power outage losses for power users:

[0104] 1) Design a user power outage loss survey form. Select typical users as the survey subjects, and the questionnaire content focuses on the user unit name, annual electricity consumption, annual peak load, annual GDP, power outage loss composition and original data of power outage losses at specific times. The survey involves economic loss estimation issues, which are the most valuable and inevitable in the research.

[0105] 2) Analysis and processing of power outage loss data. The original data of power outage loss is normalized by the annual peak load value, and then the unit load loss value of typical users is combined with the annual electricity consumption and the annual GDP in the industry to which they belong, and the unit load loss value of the industry is obtained by weighted average.

[0106] 3) Determine the coefficients to be determined in the power outage loss assessment model.

[0107] According to the unit load loss values ​​of each industry determined in the power outage loss data analysis and processing in step 2), the specific parameters are obtained by curve fitting using the MATLAB program.

[0108] 4) According to step 3), determine the power outage loss assessment model with the coefficients to be determined, calculate the unit load economic loss value of each industry at any time, and calculate the direct economic loss of power users under the power outage based on the respective loss load amounts.

[0109] Finally, regarding the study of the calculation method of indirect economic losses: indirect economic losses include losses due to delayed production materials and losses due to detention of production personnel.

[0110] That is, in some embodiments, the second data is input into a preset analysis model to analyze the power outage scope, power outage cause and power outage loss, including: constructing a power outage scope analysis model, a power outage cause analysis model and a power outage loss analysis model based on a random forest algorithm and a clustering algorithm; and using the power outage scope analysis model, the power outage cause analysis model and the power outage loss analysis model to analyze the second data.

[0111] Step 110: output the analysis result.

[0112] In this step, the analysis results may be output, for example, they may be displayed on a corresponding device to give the operator corresponding feedback. Of course, in other embodiments, the output method of the analysis results may not be limited to output display, but may also be used to store, display, use or reprocess the analysis results. According to different application scenarios and implementation needs, the specific output method of the analysis results can be flexibly selected.

[0113] Specifically, for example, in an application scenario where the method of this embodiment is executed on a single device, the analysis results can be directly output in a displayed manner on the display component (display, projector, etc.) of the current device, so that the operator of the current device can directly see the content of the analysis results from the display component.

[0114] For another example, for an application scenario where the method of this embodiment is executed on a system composed of multiple devices, the analysis results can be sent to other preset devices as receivers in the system, that is, synchronization terminals, through any data communication method (wired connection, NFC, Bluetooth, wifi, cellular mobile network, etc.), so that the synchronization terminal can perform subsequent processing on them. Optionally, the synchronization terminal can be a preset server, which is generally set up in the cloud as a data processing and storage center, which can store and distribute the analysis results; wherein the recipients of the distribution are terminal devices, and the holders or operators of these terminal devices can be power grid managers, supervisors, maintenance and sales personnel, etc.

[0115] For another example, in an application scenario where the method of this embodiment is executed on a system composed of multiple devices, the analysis results can be sent directly to a preset terminal device through any data communication method, and the terminal device can be one or more of the ones listed in the preceding paragraphs.

[0116] In more specific application scenarios, for the deployment and launch of power outage scenarios: a unified coordination network can be built to deploy power outage monitoring and analysis scenarios, achieve multi-level data penetration of the system, open up the network data level, and promote application. For the output of power outage scenarios: the GIS map can integrate the longitude and latitude coordinates of the substation area, and perform heat map data visualization based on the substation power outage data. The GIS map boundaries and substation location information of each unit can be configured separately. Customers can drill down the geographic location level through the GIS map control to support customers to conduct macro-control of the power outage distribution in the substation area. (1) Power outage range analysis: By displaying the power outage situation in different areas, the number of power outages in the area can be counted, and a comprehensive and intuitive display of the power outage area can be provided to customers, thereby improving the efficiency of power outage management. (2) Analysis of the causes of power outages. Power outages in the distribution network can be divided into two categories: pre-arranged power outages and fault power outages. Pre-arranged power outages include: planned power outages (planned maintenance power outages, construction power outages, user-application power outages), temporary power outages (temporary maintenance power outages, temporary construction power outages, user-application power outages), and power restrictions; fault power outages include: internal fault power outages and external fault power outages. The causes of power outages within the power outage range are displayed on a large screen with clear prompts to reduce the delay in handling caused by fault power outages and reduce the losses caused by power outages. (3) Analysis of power outage losses. Display the number of users within the power outage range and their electricity usage patterns, display the economic losses and grid load losses for each power outage; display the number of users within the power outage range, load levels, and user importance, and estimate the adverse social effects and negative public opinion effects that may be caused by the power outage.

[0117] It can be seen from the above embodiments that the embodiment of the present application provides a method for analyzing power outage data, which includes: obtaining initial power outage data; performing compliance check on the initial power outage data, integrating and cleaning the checked data to obtain first data; performing quality check and correction on the first data according to the set rules to obtain second data; inputting the second data into a preset analysis model, analyzing the power outage range, power outage cause and power outage loss, and obtaining analysis results; outputting the analysis results. The present application first performs a compliance check on the acquired initial power outage data to determine whether the data is filled in according to the set value range or requirements, and then integrates the data into a unified unit and code, and cleans the unreasonable data in the data, so that each data in the first data is filled in in compliance, and the format is unified and there is no duplicate data. After that, the first data is quality checked, that is, rationality checked, to determine whether the fluctuation range of the data is within a reasonable range, and the unreasonable data is adjusted and corrected to obtain the second data, and finally the corresponding model is used for power outage analysis. In this way, the format of power outage data can be quickly integrated and the rationality of data can be adjusted, thereby improving the overall efficiency and accuracy of power outage data analysis and enhancing the user experience.

[0118] In a specific scenario, the method of the embodiment of the present application, by investigating the source of power outage analysis data, completes the collection and aggregation of power consumption data needs, collects and sorts out data sharing needs; sorts out and analyzes the existing data demand table based on data application, data sharing demand content and business rules; traces the source of data items in the data demand table, traces the source business system, data table, corresponding fields, and clarifies the correlation and matching relationship between data items; analyzes and prepares data by classification and classification, communicates with the data production department, and conducts data classification and analysis for the business. At the same time, coordinate the construction manufacturer to carry out preliminary preparations for data access. Carry out functional and prototype design work to confirm the display form.

[0119] It can be seen that driven by big data technology, the power industry has begun to use massive data to analyze power outage risks and power outage losses. By mining and analyzing historical power outage data, grid structure data, power equipment status data, etc., the main types and causes of power outages can be identified, and corresponding preventive measures and response strategies can be formulated. In addition, big data technology can also realize real-time monitoring and early warning of power outages, and improve the emergency response capabilities of power supply companies.

[0120] Finally, the method of the embodiment of the present application improves the intelligence level of business monitoring and operation analysis by constructing a wide power outage monitoring table, conducting power outage data management, realizing autonomous configuration of power outage parameters, visual analysis of power outage stations, multi-level penetration of power outage data, and linking customer power outages with telephone platform work order analysis, thereby ensuring efficient loading, accurate acquisition and flexible application of data.

[0121] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of the embodiment of the present application can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.

[0122] It should be noted that the above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0123] Based on the same concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a power outage data analysis device.

[0124] refer to Figure 2 , the power outage data analysis device comprises:

[0125] The first module 210 is used to obtain initial power outage data.

[0126] The second module 220 is used to perform a compliance check on the initial power outage data, and integrate and clean the checked data to obtain first data.

[0127] The third module 230 is used to perform quality inspection and correction on the first data according to a set rule to obtain second data.

[0128] The fourth module 240 is used to input the second data into a preset analysis model, analyze the power outage scope, power outage cause and power outage loss, and obtain analysis results.

[0129] The fifth module 250 is used to output the analysis result.

[0130] In some exemplary embodiments, the first module 210 is further configured to:

[0131] Basic data acquisition is performed through ports adapted to external terminals;

[0132] The basic data is sorted and traced according to the set requirements, the association relationship between the data items in the basic data is established, and the data items are graded and classified according to the association relationship, so as to generate the initial power outage data.

[0133] In some exemplary embodiments, the second module 220 is further configured to:

[0134] Performing unit conversion and encoding conversion on the checked data;

[0135] Determine the fields with the same meaning in the checked data, and merge the fields with the same meaning.

[0136] In some exemplary embodiments, the second module 220 is further configured to:

[0137] The checked data is subjected to non-empty check, primary key duplication cleaning, illegal code cleaning, illegal value cleaning, data format check and record number check.

[0138] In some exemplary embodiments, the second module 220 is further configured to:

[0139] The first data is verified and a data directory is constructed according to the first data.

[0140] In some exemplary embodiments, the third module 230 is further configured to:

[0141] Performing a data timeliness check, uniqueness check, consistency check, integrity check, and rationality check on the first data to identify problematic data therein;

[0142] The problematic data is traced back and compared with historical data to correct the problematic data.

[0143] In some exemplary embodiments, the fourth module 240 is further configured to:

[0144] Construct power outage scope analysis model, power outage cause analysis model and power outage loss analysis model based on random forest algorithm and clustering algorithm;

[0145] The second data is analyzed using the power outage range analysis model, the power outage cause analysis model and the power outage loss analysis model.

[0146] For the convenience of description, the above devices are described in terms of functions divided into various modules. Of course, when implementing the embodiments of the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0147] The device of the above embodiment is used to implement the corresponding power outage data analysis method in the above embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0148] Based on the same concept, corresponding to any of the above-mentioned embodiments, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the power outage data analysis method as described in any of the above embodiments is implemented.

[0149] Figure 3 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0150] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0151] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0152] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0153] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0154] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0155] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0156] The electronic device of the above embodiment is used to implement the corresponding power outage data analysis method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0157] Based on the same concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the power outage data analysis method described in any of the above embodiments.

[0158] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0159] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the power outage data analysis method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0160] Based on the same concept, corresponding to any of the above-mentioned embodiments, the present application also provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer so that the computer and / or the processor execute the power outage data analysis method. Corresponding to the execution subject corresponding to each step in each embodiment of the power outage data analysis method, the processor that executes the corresponding step can belong to the corresponding execution subject.

[0161] The computer program product of the above embodiment is used to enable the computer and / or the processor to execute the power outage data analysis method described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0162] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0163] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0164] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0165] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.

Claims

1. A method for analyzing power outage data, characterized in that: include: Get the initial data of power outage; Performing a compliance check on the initial power outage data, and integrating and cleaning the checked data to obtain first data; Performing quality inspection and correction on the first data according to a set rule to obtain second data; Inputting the second data into a preset analysis model, analyzing the power outage scope, power outage cause and power outage loss, and obtaining an analysis result; The analysis result is outputted.

2. The method according to claim 1, characterized in that The obtaining of initial power outage data includes: Basic data acquisition is performed through ports adapted to external terminals; The basic data is sorted and traced according to the set requirements, the association relationship between the data items in the basic data is established, and the data items are graded and classified according to the association relationship, so as to generate the initial power outage data.

3. The method according to claim 1, characterized in that The integration and cleaning of the checked data includes: Performing unit conversion and encoding conversion on the checked data; Determine the fields with the same meaning in the checked data, and merge the fields with the same meaning.

4. The method according to claim 1, characterized in that: The integration and cleaning of the checked data includes: The checked data is subjected to non-empty check, primary key duplication cleaning, illegal code cleaning, illegal value cleaning, data format check and record number check.

5. The method according to claim 1, characterized in that After obtaining the first data, the method further includes: The first data is verified and a data directory is constructed according to the first data.

6. The method according to claim 1, characterized in that The performing quality inspection and correction on the first data according to the set rules includes: Performing a data timeliness check, uniqueness check, consistency check, integrity check, and rationality check on the first data to identify problematic data therein; The problematic data is traced back and compared with historical data to correct the problematic data.

7. The method according to claim 1, characterized in that The inputting the second data into a preset analysis model to analyze the power outage scope, power outage cause and power outage loss includes: Construct power outage scope analysis model, power outage cause analysis model and power outage loss analysis model based on random forest algorithm and clustering algorithm; The second data is analyzed using the power outage range analysis model, the power outage cause analysis model and the power outage loss analysis model.

8. A power outage data analysis device, characterized in that: include: The first module is used to obtain the initial data of power outage; The second module is used to perform compliance check on the initial power outage data, integrate and clean the checked data, and obtain first data; A third module is used to perform quality inspection and correction on the first data according to a set rule to obtain second data; The fourth module is used to input the second data into a preset analysis model, analyze the power outage scope, power outage cause and power outage loss, and obtain analysis results; The fifth module is used to output the analysis result.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to implement the method according to any one of claims 1 to 7.

Images