Business data intelligent processing method and system for power transaction auxiliary decision making

Abstract

The invention discloses a business data intelligent processing method and system for power transaction auxiliary decision making. The method comprises the following steps: a data acquisition step: business data acquisition and data processing template matching; a data processing step: data preprocessing and target data file generation; a page display step: receiving a login instruction and displaying business data; a work order creating step: operation triggering and work order obtaining; and a data writing step: extracting the data and writing the data into the work order. According to the invention, intelligent processing of the business data of the power transaction auxiliary decision-making system can be realized, the automation degree and the data processing efficiency of the business process are greatly improved, the normalization, the safety and the accuracy of the data are guaranteed, the processing cost is effectively reduced, and powerful technical support is provided for healthy development and efficient operation of the power transaction market. The method has important practical value for promoting efficient, safe and stable operation of the electricity market.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for intelligently processing business data for auxiliary decision-making in power trading. Background Art

[0002] The power transaction decision-making support system plays an increasingly critical role in optimizing resource allocation and improving market efficiency. As the core hub connecting energy flow, information flow and value flow, the power transaction decision-making support system has a direct impact on the efficiency of market resource allocation and the safety of power grid operation.

[0003] In the process of implementing the technical solution of the present invention, the inventor of this patent discovered that the current electricity trading platform has the problem of insufficient data processing efficiency, which seriously restricts the healthy development of the market. The existing technology specifically has the following technical problems.

[0004] 1. Dilemma of operational standardization: lack of dual-track system of process standardization and automation; At present, the processing of power trading data generally suffers from the stubborn problem of "too many manual intervention nodes". Taking a provincial power trading center as an example, its day-ahead market clearing process involves more than ten business departments and more than twenty types of data sources, with a high proportion of manual verification links, resulting in a long daily settlement cycle. This operating mode is not only prone to inconsistent data calibers, but also induces systemic risks in extreme market scenarios. In a certain year, a city in East China suffered a large fluctuation in the auction price of spare capacity due to manual correction errors in load forecast data. In short, irregular operations make it difficult to ensure the consistency and accuracy of data processing results.

[0005] 2. Data security protection: Traditional technical architecture faces new attack challenges; As the power market becomes more open, the value of data assets has risen significantly. Monitoring data from a power grid company shows that the number of cyber attacks on power trading systems has increased year by year in recent years, with unauthorized access to API interfaces and lateral penetration attacks accounting for a large proportion. An International Energy Agency research report pointed out that attackers can construct false quotation strategies by stealing real-time load data, resulting in high returns from market manipulation. In short, low security increases the risk of sensitive data leakage.

[0006] 3. Cost control problem: structural contradiction between computing resources and economic benefits; The processing of power trading data shows a significant "rigid growth in computing power demand" feature. Taking the participation of virtual power plants in the spot market as an example, each entity needs to process the day-ahead quotation data of more than 2,000 nodes in real time, and the computing power cost of traditional server clusters is high. What's more serious is that most data processing tasks involve repeated calculations. For example, a regional market in East China needs to repeatedly verify the available capacity of cross-provincial transmission channels many times a day, resulting in a large amount of invalid calculations. This directly leads to an increase in the cost of participation of market entities-the average transaction cost of small and medium-sized power sales companies has increased, seriously weakening market vitality. In short, high processing costs directly affect the overall economic efficiency of power trading.

[0007] In summary, the existing data processing methods for power trading business cannot meet the actual application needs. Summary of the invention

[0008] Based on the problem that the existing power trading business data processing method cannot meet the actual application needs, there is an urgent need to develop a more advanced and efficient intelligent processing method and system. To this end, the present invention provides a business data intelligent processing method and system for power trading auxiliary decision-making.

[0009] In a first aspect, the present invention provides a method for intelligently processing business data for auxiliary decision-making in power trading, comprising: Data acquisition step: collecting business data related to power trading decision support, retrieving and matching data processing templates corresponding to the business data, wherein the data processing templates predefine the format, processing rules and verification standards of the business data to ensure the consistency and accuracy of the business data; Data processing steps: according to the requirements of the data processing template, the collected business data is automatically filled into the data processing template, and the Z score value of each electricity price data in the electricity price data set is first calculated. If the absolute value of the Z score value exceeds the preset threshold, it is determined to be an abnormal electricity value, and the lower limit of the reasonable electricity price range determined by the market situation is used to automatically correct the abnormal electricity value, so as to clean the electricity price data based on the Z score, and then the historical data is converted into value equivalent according to the inflation rate to make the inter-period data comparable, and then the data is verified; after the filling is completed, a target data file is generated, and the target data file contains processed standardized business data, which is convenient for subsequent analysis and decision-making; Page display step: after receiving the user's login instruction to log in to the power trading decision-making support system, the business data display page is loaded and displayed in the preset menu area of ​​the power trading decision-making support system to intuitively display the business data for the convenience of user understanding and operation; wherein, the dynamic dashboard is used to provide intuitive data display and real-time early warning notification functions; New work order step: In the business data display page, obtain the operation that triggers the new business work order process, automatically create a new business work order, and record the detailed information in the business work order for easy tracking and management; Data writing step: extract the required business data from the target data file according to the instructions of the data processing template, and write the extracted business data into the business work order to ensure that the work order content is complete and accurate; Data mining steps: mining valuable information and trends from business data through association rule analysis, cluster analysis and trend prediction analysis; Dynamic permission management steps: Different data views and operation options are displayed according to the user's role and permission level. The permission level corresponding to the administrator role can access all data, the permission level corresponding to the operator role can access 70% of the data, and the permission level corresponding to the auditor role can only access audit-related data; Among them, the real-time early warning notification function of the dynamic dashboard and the data mining steps based on cluster analysis are used to identify trading patterns, providing multi-dimensional decision support for auxiliary decision-making in power trading; by combining Z-score-based data cleaning operations with dynamic authority management, the conflict between data security and decision-making efficiency can be addressed.

[0010] Optionally, the method further includes: a data backup step: backing up the collected business data to the cloud.

[0011] Optionally, in the page display step, the distribution of transaction volume in different time periods is displayed by a transaction volume change graph.

[0012] Optionally, the method further includes: a secondary data verification step: performing secondary data verification before writing the business data into the business work order.

[0013] Optionally, the method further includes: a work order flow tracking step: displaying the processing status and historical flow records of the business work order.

[0014] Optionally, the method further includes: a logging step: automatically recording the historical records of business data processing and user operations.

[0015] Optionally, the method further includes: a security audit step: recording and monitoring business data processing activities.

[0016] In a second aspect, the present invention provides a business data intelligent processing system for auxiliary decision-making in power trading, comprising: A data acquisition module is used to collect business data related to power trading auxiliary decision-making, retrieve and match data processing templates corresponding to the business data, and the data processing templates predefine the format, processing rules and verification standards of the business data to ensure the consistency and accuracy of the business data; The data processing module is used to automatically fill the collected business data into the data processing template according to the requirements of the data processing template, first calculate the Z score value of each electricity price data in the electricity price data set, and if the absolute value of the Z score value exceeds the preset threshold, it is determined to be an abnormal electricity value, and the lower limit of the reasonable electricity price range determined by the market situation is used to automatically correct the abnormal electricity value, so as to clean the electricity price data based on the Z score, and then perform value equivalent conversion of historical data according to the inflation rate to make the inter-period data comparable, and then perform data verification; after the filling is completed, a target data file is generated, and the target data file contains processed standardized business data, which is convenient for subsequent analysis and decision-making; The page display module is used to load and display the business data display page in the preset menu area of ​​the power trading auxiliary decision-making system after receiving the user's login instruction to log in to the power trading auxiliary decision-making system, so as to intuitively display the business data and facilitate the user's understanding and operation; wherein, the dynamic dashboard is used to provide intuitive data display and real-time early warning notification functions; The work order creation module is used to obtain the operation that triggers the process of creating a new business work order in the business data display page, automatically create a new business work order, and record the detailed information in the business work order for easy tracking and management; The data writing module is used to extract the required business data from the target data file according to the instructions of the data processing template, and write the extracted business data into the business work order to ensure that the work order content is complete and accurate; Data mining module, used to mine valuable information and trends from business data through association rule analysis, cluster analysis and trend prediction analysis; The dynamic authority management module is used to display different data views and operation options according to the user's role and authority level. The authority level corresponding to the administrator role can access all data, the authority level corresponding to the operator role can access 70% of the data, and the authority level corresponding to the auditor role can only access audit-related data; Among them, the real-time early warning notification function of the dynamic dashboard and the data mining module based on cluster analysis are used to identify trading patterns, providing multi-dimensional decision support for auxiliary decision-making in power trading; by combining Z-score-based data cleaning operations with dynamic authority management, the conflict between data security and decision-making efficiency can be addressed.

[0017] One or more technical solutions provided by the present invention have at least the following technical effects or advantages: The present invention can realize the intelligent processing of business data of the power trading auxiliary decision-making system, greatly improve the automation degree of business processes and data processing efficiency, ensure the standardization, security and accuracy of data, effectively reduce processing costs, and provide strong technical support for the healthy development and efficient operation of the power trading market. It has important strategic significance and practical value for promoting the efficient, safe and stable operation of the power market. Specifically, the beneficial effects of the present invention include at least the following 7 points.

[0018] 1. By introducing intelligent processing procedures, ensure that the processing of business data follows unified specifications and standards, including not only the unification of data formats, but also the consistency of data processing logic, thereby greatly reducing data errors and deviations caused by non-standard operations and improving the data management level and business execution standards of the entire power trading market.

[0019] 2. By reducing the number of manual intervention nodes, the overall process is automated, and the real-time collection and processing of business data is realized, which significantly improves the data processing efficiency and shortens the data processing cycle. For example, the portability of user operations is increased to 90%, and the speed of business work order creation is increased by at least 65%. It can quickly respond to user needs and market changes, provide real-time data support for power trading decisions, and ensure the timeliness of trading decisions.

[0020] 3. The data processing template of the present invention will perform data verification when automatically filling the collected business data into the data processing template, which can enhance data security and effectively prevent data leakage and illegal tampering.

[0021] 4. Reduce human intervention through automated processes, significantly reduce human involvement in data processing, and reduce rework and correction costs caused by human operational errors; intelligent processing enables more efficient use of data processing resources, reduces unnecessary waste of resources, optimizes resource allocation, improves cost-effectiveness, and thus reduces overall operating costs.

[0022] 5. Through the data cleaning operation based on Z score, the abnormal value of electricity price is identified and automatically corrected, which can detect and correct the abnormal value of electricity price in real time and reduce data noise; the historical data is converted into equivalent value according to the inflation rate to make the inter-period data comparable, and the data of different time spans can be compared under the same value standard to improve the reliability of data processing; the data after cleaning and conversion is verified to improve data security; in short, the data value is maximized through the data preprocessing operation of cleaning, conversion and verification. Through the real-time early warning notification function of the dynamic dashboard and the data mining step based on cluster analysis to identify the transaction mode, multi-dimensional decision support is provided for the auxiliary decision of power trading, making the auxiliary decision of power trading more scientific and reliable. By combining the data cleaning operation based on Z score with the dynamic management of permissions, the conflict between data security and decision efficiency is dealt with, and the data environment of auxiliary decision of power trading is optimized.

[0023] 6. The processing flow of business data has been optimized, intermediate links have been reduced, and data flow has become smoother, thereby significantly improving overall work efficiency.

[0024] 7. Enhance the scalability and adaptability of the system to meet the changing power market demands and regulatory requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is an architecture diagram of a business data intelligent processing system for auxiliary decision-making in power trading in one embodiment of the present invention; Figure 2 A simplified flow chart of a method for intelligently processing business data for auxiliary decision-making in power trading in one embodiment of the present invention; Figure 3 The present invention is a detailed flow chart of a method for intelligently processing business data for auxiliary decision-making in power trading in one embodiment of the present invention. DETAILED DESCRIPTION

[0026] The present invention provides a method and system for intelligently processing business data for auxiliary decision-making in power trading, which solves the problem that existing methods for processing business data in power trading cannot meet actual application requirements.

[0027] In order to clearly and in detail describe the intelligent processing method and system of business data for auxiliary decision-making in power trading of the present invention, the following will explain it from four major sections, namely: system composition and use section, method implementation section, function expansion section and effect verification section.

[0028] 1. System composition and purpose; like Figure 1As shown, a business data intelligent processing system for auxiliary decision-making in power trading of the present invention includes: a data acquisition module, a data processing module, a page display module, a work order creation module and a data writing module.

[0029] 1.1 The data acquisition module is responsible for business data collection and data processing template matching, including data interface, data collection unit, template matching unit and template library.

[0030] The data interface is used to connect with multiple data sources of business data. The data acquisition unit is used to collect business data related to power trading decision-making assistance. The template matching unit is used to retrieve and match data processing templates corresponding to business data. The data processing template predefines the format, processing rules and verification standards of business data to ensure the consistency and accuracy of business data. The template library is used to store data processing templates.

[0031] 1.2 The data processing module is responsible for data preprocessing and target data file generation, including data filling unit, data preprocessing unit, data verification unit and data output unit.

[0032] The data filling unit is used to automatically fill the collected business data into the data processing template according to the requirements of the data processing template. The data preprocessing unit and the data verification unit are used to perform data cleaning, data conversion and data verification through preset algorithms and logics. The data output unit is used to generate a target data file after filling. The target data file contains processed standardized business data for subsequent analysis and decision-making.

[0033] 1.3 The page display module is responsible for receiving login instructions and displaying business data, including user interface components, interaction control unit, page loading unit and display control unit.

[0034] Among them, the user interface component is used to provide a user operation interface, allowing the user to interact through the user interface component, such as filtering, sorting, viewing detailed data, etc. The interactive control unit is used to receive the user's login instruction to log in to the power trading auxiliary decision-making system, and control the user interaction, such as generating filtering instructions, sorting instructions, viewing detailed data instructions, etc. The page loading unit is used to load the business data display page in the preset menu area of ​​the power trading auxiliary decision-making system after receiving the user's login instruction to log in to the power trading auxiliary decision-making system. The display control unit is used to display the business data display page, and through the intuitive display of business data, ensure that the user can effectively view and analyze the data, which is convenient for the user to understand and operate.

[0035] 1.4 The new work order module is responsible for operation triggering and work order acquisition, including the work order triggering unit, work order acquisition unit, and work order management unit.

[0036] The work order triggering unit is used to obtain the operation that triggers the new business work order process in the business data display page. The work order acquisition unit is used to automatically create a new business work order in response to triggering the new business work order process, and provide a target location for subsequent data writing. The work order management unit is used to record detailed information in the business work order for easy tracking and management.

[0037] 1.5 The data writing module is responsible for data extraction and data writing work orders, including data extraction unit and data entry unit.

[0038] The data extraction unit is used to extract the required business data from the target data file according to the instructions of the data processing template. The data entry unit is used to write the extracted business data into the business work order, complete the final recording and storage of the data, ensure the completeness and accuracy of the work order content, and provide reliable data support for power trading decision-making assistance.

[0039] 2. Method implementation; Below, please refer to Figure 2 The present invention provides an intelligent processing method for business data for auxiliary decision-making in power trading, including a data acquisition step, a data processing step, a page display step, a work order creation step and a data writing step.

[0040] 2.1 The data acquisition step includes the business data collection sub-step and the data processing template matching sub-step.

[0041] 2.1.1 Business data collection sub-step, specifically including: collecting business data related to power trading auxiliary decision-making.

[0042] Among them, the business data related to power trading auxiliary decision-making is collected, specifically: real-time or periodic collection of business data related to power trading auxiliary decision-making from multiple data sources, multiple data sources include: power market trading system, power dispatch center dispatch system, power generation enterprise production management system, power sales company customer service system. Of course, in practical applications, the acquisition cycle of real-time business data collection can be set according to specific needs, such as every 3 minutes, every 6 minutes, etc.; the time node for regular business data collection can also be set according to specific needs, such as 3 o'clock in the morning, 5 o'clock in the morning, etc.; specific data sources can also be set according to specific needs, such as the power market trading system and the power sales company's customer service system, the power market trading system, the power dispatch center dispatch system, and the power generation enterprise's production management system. The present invention does not limit the acquisition cycle, regular collection time node, and data source combination method.

[0043] The business data related to power transaction auxiliary decision-making includes at least one of the following data: transaction volume, electricity price, market supply and demand, transaction price, transaction information, transaction time, transaction type, and power grid operation status. Of course, in actual applications, a combination of two or more data can be set according to specific needs, such as a first combination of transaction volume and transaction price, and a second combination of transaction volume, transaction time, and power grid operation status, which is not limited by the present invention.

[0044] When collecting business data related to power trading decision-making assistance in real time, the business data collection sub-step is implemented in the specific process, for example: in the data acquisition module, a real-time data synchronization mechanism is implemented, and the data interface of the data acquisition module is connected to the API interface of the power market trading system. The data collection unit obtains the latest transaction data in real time to ensure the timeliness of the data. For example, the data collection unit automatically requests the latest transaction data through the API interface every 5 minutes. According to statistics, 1,000 transaction records were synchronized in the most recent trading day, involving a total of 500GWh of electricity.

[0045] When regularly collecting business data related to power trading decision-making assistance, the business data collection sub-step is implemented in a specific process. For example, if a certain power market generates 1,000 transaction records every day, the data collection unit collects data from the API interface of the power market's trading system on a regular basis every day. Specifically, the data collection unit starts at 1 a.m. every day, and sends a request to the API interface of the power market's trading system using the POST request method through the HTTPS protocol to obtain the latest transaction records, so as to capture data during the time period when the power market data is less updated at night, and finally collects 1,000 transaction records from the API interface of the power market's trading system.

[0046] 2.1.2 The data processing template matching sub-step specifically includes: retrieving and matching the data processing template corresponding to the business data. The data processing template predefines the format, processing rules and verification standards of the business data to ensure the consistency and accuracy of the business data.

[0047] In the data processing module, different data processing templates are designed according to the transaction types of business data, such as spot transactions, futures transactions, peak-shaving transactions, etc., to meet the processing requirements of different types of business data. Among them, the data processing templates corresponding to the transaction types at least include data processing template A corresponding to spot transactions, data processing template B corresponding to futures transactions, and data processing template C corresponding to peak-shaving transactions.

[0048] a. Data processing template A corresponding to spot transactions focuses on processing real-time transaction data, including electricity volume, price and transaction time. According to statistics, data processing template A processes 300 spot transactions in a quarter, and the average processing time of each transaction is shortened by 50%.

[0049] b. Data processing template B corresponding to futures trading includes fields such as future delivery month, contract price, and contract quantity. According to statistics, in one quarter, data processing template B processed 200 futures contracts with an accuracy rate of 99.8%.

[0050] c. The data processing template C corresponding to the peak-shaving transaction includes spare capacity, peak-shaving price, etc. According to statistics, during the peak period of peak-shaving transactions, the data processing template C helps process an additional 50 transactions, effectively supporting the stable operation of the power grid.

[0051] Of course, in actual applications, data processing templates may be defined based on other different characteristics of business data, such as buyer type, transaction level, etc., and the present invention does not limit this.

[0052] The template library includes data processing templates for predefined transaction records, including field formats, field processing rules, field validation standards, etc., which are used to standardize how business data should be formatted, processed and displayed, including field formats, field processing rules, and field validation standards.

[0053] Field format: The data processing template defines a clear field format for each transaction record. For example, the transaction time field format is "YYYY-MM-DD HH:MM:SS" and the electricity price field format is "floating point number, retain two decimal places".

[0054] Field processing rules: The data processing template contains field processing rules. For example, the electricity price field needs to be multiplied by a conversion factor to obtain the standard electricity price.

[0055] Field validation criteria: The data processing template also defines field validation criteria, for example: the transaction power field must be greater than 0 MWh and less than 10,000 MWh.

[0056] Continuing with the previous embodiment, the data processing template matching sub-step is implemented in the specific process. For example, the business data returned by the API interface is in JSON format, which contains 1,000 transaction records, and each transaction record contains 20 fields, such as transaction time, transaction volume, electricity price, transaction information, etc. After the data interface receives the business data in JSON format, it first starts the matching process, parses the business data in JSON format into a data structure in memory, and then compares it with the data processing template in the template library one by one; during the matching process, if it is found that the field of a record does not conform to the format or rules defined by the data processing template, the system will mark it as abnormal data and record the abnormal information. After matching, a total of 990 transaction records fully meet the requirements of the data processing template, and 10 transaction records are marked due to format errors or data anomalies, which require manual review or correction.

[0057] 2.2 The data processing step includes the data preprocessing sub-step and the target data file generation sub-step.

[0058] 2.2.1 Data preprocessing sub-steps, specifically including: according to the requirements of the data processing template, automatically fill the collected business data into the data processing template, and perform data cleaning, data conversion and data verification through preset algorithms and logic. The collected raw data is preliminarily cleaned, converted and verified to ensure data quality.

[0059] Data cleaning refers to automatic correction operations such as removing null values ​​and correcting obvious errors on business data. The automatic correction process includes data verification, identification of unreasonable values, recording and reporting of anomalies, determination of correction plan, implementation of correction, and verification of correction results.

[0060] Data verification stage: Verify the collected business data to check whether there are abnormal values, missing values, incorrect formats or logically unreasonable data.

[0061] Identify unreasonable values: Identify unreasonable values ​​in business data by setting thresholds, data ranges, logical rules and other standards. For example, the price of electricity cannot be negative, and the amount of electricity cannot exceed the capacity of a certain power generation facility.

[0062] Record and report abnormality stage: record the unreasonable values ​​identified and generate an abnormality report to facilitate subsequent analysis and correction.

[0063] Correction plan determination stage: According to the specific circumstances of abnormal data, formulate a correction plan. The correction plans include the following: If it is an obvious input error, it can be directly corrected to the correct value; if it is a missing value, it can be estimated and filled based on historical data or industry average; if it is an abnormal value, but it may reflect special circumstances, such as market fluctuations, further analysis is needed to decide whether to correct it.

[0064] Implementation and correction phase: Correct unreasonable values ​​according to the determined correction plan.

[0065] Verification and correction result stage: After the correction is completed, the data is verified again to ensure that the corrected values ​​are reasonable and meet the data specifications.

[0066] In practical applications, automatic correction includes electricity price correction, electricity quantity correction, etc.

[0067] Example of electricity price revision: The electricity price in the original data is -10 yuan / kWh. The electricity price should be a real number greater than 0. -10 yuan / kWh is an unreasonable value. By checking relevant market information and historical data, a reasonable electricity price range is determined according to market conditions, such as 0.5 yuan / kWh to 1.5 yuan / kWh. Finally, -10 yuan / kWh is revised to a reasonable lower limit of 0.5 yuan / kWh.

[0068] Example of electricity revision: The electricity recorded in the original data is 5000MWh. Assuming that the maximum capacity of the power generation facility is 1000MWh, 5000MWh is an unreasonable value. By checking the capacity limit of the power generation facility and the historical power generation data, 5000MWh is finally revised to the maximum capacity of the power generation facility, 1000MWh.

[0069] In the data processing module, a data cleaning operation based on the Z score can be used to automatically identify and correct abnormal data in the business data to improve the accuracy of the data. The data cleaning operation based on the Z score specifically includes: first calculating the Z score value of each electricity price data in the electricity price data set. If the absolute value of the Z score value exceeds the preset threshold, it is determined to be an abnormal electricity value, and the lower limit of the reasonable electricity price range determined by the market situation is used to automatically correct the abnormal electricity value. Assume that the data preprocessing unit uses a data cleaning operation based on the Z score to clean the electricity price data in 1000 transaction records to identify abnormal data in the electricity price data and remove abnormal data. The formula is as follows: Z=(X-μ) / σ, where X is the observed value of the electricity price data, μ is the average value of the electricity price data set, and σ is the standard deviation of the electricity price data set. If |Z|>2, X is considered to be abnormal data. Of course, in practical applications, different abnormal data judgment conditions can be set according to specific needs, for example: |Z|>3, etc., and the present invention is not limited. Assuming that the mean of the electricity price data set is 50 yuan / MWh, the standard deviation is 5 yuan / MWh, and the electricity price of a transaction record is 85 yuan / MWh, then the electricity price Z score is: Z=(85-50) / 5=7. Since 7>2, this transaction record is marked as abnormal. Among the 1,000 transaction records, by calculating the electricity price Z score of each transaction record, it is found that the absolute value of the electricity price Z score of 20 transaction records is greater than 2, and these transaction records are marked as abnormal data. By using the data cleaning operation based on the Z score to detect abnormal electricity prices in real time, it is not only possible to remove abnormal data, but also to self-correct the processing logic, such as automatically modifying the abnormal value of -10 yuan / kWh to 0.5 yuan / kWh, so as to automatically correct the abnormal data to a reasonable value, improve the accuracy and reliability of the overall data, and provide accurate data support for auxiliary decision-making in power trading.

[0070] Data conversion includes: converting power transaction data provided by different data sources into a unified format to facilitate subsequent analysis and decision-making; converting price, electricity and other information in historical data into equivalent values ​​under current market rules to maintain data comparability; and encoding and converting the identification information of both parties to ensure the consistency of transaction records. Data conversion helps ensure the consistency, accuracy and availability of business data in the power transaction decision-making support system.

[0071] a. Convert the power transaction data provided by different data sources into a unified format, unify the data formats of different data sources, and facilitate further intelligent processing and analysis of business data in the power transaction decision-making support system. In the specific implementation, for example: there are two different power transaction data sources, namely data source A and data source B. The power transaction data format provided by data source A is as follows: the date format is dd / mm / yyyy, the unit of electricity is kilowatt-hour (kWh), and the price unit is yuan / kilowatt-hour (CNY / kWh). The power transaction data format provided by data source B is as follows: the date format is yyyy-mm-dd, the unit of electricity is megawatt-hour (MWh), and the price unit is yuan / megawatt-hour (CNY / MWh). Convert the data of data source A to the format of data source B.

[0072] Assume that the data source A has the following data: date 15 / 03 / 2025, electricity 3000kWh, electricity price 0.8CNY / kWh. You need to convert the date format from 15 / 03 / 2025 to 2025-03-15; convert the electricity from kilowatt-hour (kWh) to megawatt-hour (MWh), because 1MWh=1000kWh, so 3000kWh=3MWh; convert the price from yuan / kilowatt-hour (CNY / kWh) to yuan / megawatt-hour (CNY / MWh), because 1MWh=1000kWh, so 0.8CNY / kWh=0.0008CNY / MWh.

[0073] Converted data: Date 2025-03-15, electricity volume 3MWh, electricity price 0.0008CNY / MWh.

[0074] b. Convert the price, electricity and other information in the historical data into equivalent values ​​under the current market rules, such as inflation adjustment, to ensure that the historical data can be accurately analyzed and compared in the current market environment. In the specific implementation process, for example: the historical electricity price is 500 yuan / MWh, the historical data is recorded in 2019, and the inflation rate for the six years from 2019 to 2025 is 3% per year. The historical electricity price needs to be converted into an equivalent value under the market rules of 2025, which can solve the problem of comparability of inter-period data.

[0075] The historical electricity price is 500 yuan / MWh. The electricity price needs to be converted to reflect inflation. The inflation adjustment is 500 yuan / MWh×(1+3%)^6. The converted electricity price is 500 yuan / MWh×(1.03)^6≈500 yuan / MWh×1.194052≈597 yuan / MWh.

[0076] The converted electricity price is 597 yuan / MWh.

[0077] c. Encoding and converting the identification information of both parties to the transaction ensures the consistency of transaction records between different systems, thereby maintaining the consistency and accuracy of transaction records in the new power transaction decision support system.

[0078] In the specific implementation process, for example, there are two power trading parties, namely power generation enterprise A and power user B, and they have different identification information in different systems: the identification information of power generation enterprise A is HeaderCode_A1 in system A, and the identification information of power user B is UserCode_B2 in system B. Among them, the encoding conversion process includes the stage of establishing a mapping relationship, the stage of applying the mapping relationship to perform encoding conversion, and the stage of post-conversion transaction recording.

[0079] Establishing mapping relationship phase: Establish a mapping table to match the identification information of the old system with the identification information of the new system, such as HeaderCode_A1 in System A corresponds to GeneratorCode_A in the new system, and UserCode_B2 in System B corresponds to ConsumerCode_B in the new system.

[0080] Apply the mapping relationship to the encoding conversion stage: Use the mapping table to convert the identification information in the original data into the new system code. Look up the mapping table and find that the new system code corresponding to HeaderCode_A1 is GeneratorCode_A, and the new system code corresponding to UserCode_B2 is ConsumerCode_B.

[0081] Transaction record stage after conversion: power generation enterprise A is identified as GeneratorCode_A, and power user B is identified as ConsumerCode_B.

[0082] Data validation refers to the validation of data according to preset validation rules. Preset validation rules include data type check, value range check, logical relationship check, etc. The data validation process can use the following formula to check the consistency of business data: Let D be the original data set, V(D) be the validation function, if V(D)=True, the data passes the validation; otherwise, the data is marked as requiring further inspection. For example, the formula for verifying the transaction power can be V(power)=power≥0.

[0083] Continuing with the previous embodiment, in the specific implementation process, for example: among 1000 transaction records, 20 abnormal transactions are found, and the remaining 980 transaction records will be subject to data verification, and the next step will be processed after the verification is correct. Through the preset verification rule verification, all 980 transaction records pass, and no new erroneous data is found. The formula for calculating the verification pass rate is as follows: Verification pass rate = (number of data items that pass the verification / total number of data items that are verified) × 100%, substituting the data into the calculation, the verification pass rate = (980 / 980) × 100% = 100%.

[0084] 2.2.2 The target data file generation sub-step specifically includes: after filling, generating the target data file, the target data file contains processed standardized business data, which is convenient for subsequent analysis and decision-making. Among them, standardized business data refers to structured and formatted business data, which can be Excel tables, CSC files or other formats of data files.

[0085] Continuing with the previous embodiment, in the specific implementation process of the target data file generation sub-step, for example, the data output unit outputs the 980 valid transaction records after cleaning, conversion and verification into a target data file in CSV format. The generated target data file contains 980 transaction records, each of which contains the cleaned, converted and verified electricity price data and other related fields.

[0086] 2.3 The page display step includes a login instruction receiving sub-step and a business data display sub-step.

[0087] 2.3.1 The login instruction receiving sub-step specifically includes: receiving a login instruction from a user to log in to the power trading decision-making support system. The interactive control unit is used to receive a login instruction from a user to log in to the power trading decision-making support system.

[0088] 2.3.2 The business data display sub-step specifically includes: after receiving a login instruction from a user to log into the power trading decision-making support system, loading and displaying a business data display page in a preset menu area of ​​the power trading decision-making support system to intuitively display business data for easy understanding and operation by the user.

[0089] Among them, the business data display page specifically displays business data in a graphical or tabular form. It is the front-end interface for users to review and operate data, and provides a visual display of data. The page loading unit is used to load the business data display page in the preset menu area of ​​the power trading decision-making support system after receiving the user's login instruction to log in to the power trading decision-making support system. The display control unit is used to display the business data display page, and through the intuitive display of business data, ensure that users can effectively view and analyze data, which is convenient for users to understand and operate. The operation details of the page display module and the display control module in the page loading time and chart display are as follows.

[0090] Page loading time: Assume that the average loading time of each business data display page is 2 seconds, and the total loading time of 5 business data display pages is 10 seconds.

[0091] Chart display: Assuming that each business data display page displays 10 charts, 5 business data display a total of 50 charts. Each chart is dynamically generated based on business data to ensure the real-time and accuracy of information.

[0092] In the specific implementation process, for example, assuming that there are 5 business data display pages in the system preset menu area, and each business data display page contains an average of 10 charts. After the user logs in to the system, the page loading unit loads 5 business data display pages, displaying a total of 50 charts.

[0093] 2.4 The work order creation step includes the operation triggering sub-step and the work order acquisition sub-step.

[0094] 2.4.1 The operation triggering sub-step specifically includes: obtaining an operation that triggers the creation of a new business work order process in the business data display page. Among them, triggering the creation of a new business work order process specifically includes the following two triggering methods.

[0095] The first method is to trigger the process of creating a new business work order in response to the user's operation of creating a new work order. In the specific implementation process, for example, in the business data display page, the user clicks a button on a chart, and the work order triggering unit triggers the process of creating a new business work order.

[0096] The second type: in response to intelligent operation, automatically trigger the new business work order process. In the specific implementation process, for example, when entering the business data display page, the work order triggering unit automatically triggers the new business work order process.

[0097] 2.4.2 The work order acquisition sub-step specifically includes: in response to triggering a new business work order process, automatically creating a new business work order, and recording detailed information in the business work order for easy tracking and management.

[0098] Among them, in the new work order creation module, different types of work orders are automatically created according to preset business rules through the automation rule engine.

[0099] Continuing with the previous embodiment, in the specific implementation process of the work order acquisition sub-step, for example: the work order acquisition unit responds to the operation of triggering the new business work order process, creates a new business work order, and presents it to the user for further editing and confirmation, and the work order management unit records the detailed information to the business work order. The following describes the page loading time, user operation response time, and work order management process respectively.

[0100] Page loading time: After the user logs in to the system, assuming that the page loading unit loads 5 business data display pages within 5 seconds, the business data display page loading efficiency is 1 page per second (5 pages / 5 seconds = 1 page / second).

[0101] User operation response time: After the user clicks a button on the chart, the average time for the work order acquisition unit to respond and start the work order creation process is 1 second. Specific calculation example: It takes an average of 5 seconds for the user to complete the work order creation from clicking the chart button to the work order creation completion, and the work order creation speed is 1 work order every 5 seconds.

[0102] Work order management process: The work order management unit automatically fills in the preset fields of the business work order, such as date, time, user information, etc., based on the user's operations, and allows users to add specific descriptions or notes.

[0103] 2.5 The data writing step includes the data extraction sub-step and the data writing work order sub-step.

[0104] 2.5.1 The data extraction sub-step specifically includes: extracting the required business data from the target data file according to the instruction of the data processing template. The data extraction sub-step is completed by the data extraction unit.

[0105] 2.5.2 The data writing work order sub-step specifically includes: writing the extracted business data into the business work order to ensure that the work order content is complete and accurate, and to provide reliable data support for power trading auxiliary decision-making. Among them, the data entry unit completes the data writing work order sub-step.

[0106] 3. Functional expansion; 3.1 Data backup; In order to prevent data loss, in the data acquisition module, the collected business data is regularly backed up to the cloud through the data backup mechanism.

[0107] In the specific implementation process, for example: assuming that the data backup mechanism performs data backup once a day, the daily backup data volume V is gigabytes, and the daily backup data volume V should satisfy V>0; assuming that the cloud storage space is S gigabytes, it must satisfy S≥V×number of backup days; assuming the backup is 30 days, then S≥30V; if the daily backup data volume V=100GB, then the cloud requires at least 3000GB (30 days×100GB / day) of storage space.

[0108] 3.2 Data mining; In order to help analysts extract valuable information and trends from large amounts of business data, a data mining tool is integrated into the data processing module. The data mining tool has the functions of association rule analysis, clustering analysis, and trend prediction analysis.

[0109] Association rule analysis: helps analysts discover correlations in transaction data, such as which types of transactions tend to occur together.

[0110] Cluster analysis: Grouping large amounts of trading data allows analysts to identify different market behavior patterns. Cluster analysis can identify trading patterns, such as peak trading concentration periods.

[0111] Trend forecasting: Use statistical methods such as time series analysis to predict future market trends.

[0112] According to statistics, data mining tools discovered three main trading patterns in a recent analysis, which helped analysts predict changes in market demand in the next month and thus guide trading strategies.

[0113] 3.3 Dynamic management of permissions; In order to display different data views and operation options according to the user's role and permission level, a dynamic user permission management function is added to the page display module. By setting strict role permissions and access control, it is ensured that only authorized personnel can access and process sensitive business data, further improving data security.

[0114] In the specific implementation process, for example, assume that there are three user roles: administrator R_admin, operator R_operator, and auditor R_auditor. The authority levels of the user roles are 3, 2, and 1 respectively. The formula for the user authority management function is described as follows: If the user role is administrator R_admin, all data D_all will be displayed; If the user role is operator R_operator, partial data D_partial is displayed, such as D_partial=D_all×70%; If the user role is auditor R_auditor, he can only view audit-related data D_audit, such as D_audit=D_all×30%.

[0115] 3.4 Dynamic dashboard and trading volume change chart; In the page display module, a dynamic dashboard and transaction volume change chart are provided to display the transaction status of the power market in real time, including price fluctuations, transaction volume changes, etc.

[0116] The dynamic dashboard uses advanced visualization technology to provide intuitive data display. For example, the price fluctuation chart on the dynamic dashboard uses time series data to display the fluctuation of electricity prices in the past 24 hours, so that users can clearly see the trend of electricity prices rising from 100 yuan / MWh to 150 yuan / MWh. The dynamic dashboard also provides real-time early warning notification function. When the market fluctuates abnormally, the dynamic dashboard will immediately warn users. According to statistics: in a certain month, the dynamic dashboard issued a total of 10 market fluctuation warnings to help users adjust their trading strategies in a timely manner.

[0117] The trading volume change chart shows the distribution of trading volume in different time periods. For example, the trading volume change chart shows that the trading volume reached its peak between 9 and 10 a.m., with a total transaction volume of 200MWh, providing an important reference for market participants to formulate trading strategies.

[0118] 3.5 Secondary data verification; In order to ensure the integrity and consistency of the data, a secondary data verification function is developed in the data writing module to perform secondary data verification before the business data is written into the business work order. The secondary data verification includes format check, consistency check, and range verification.

[0119] In the specific implementation process, for example: assuming that the transaction data received by the system includes electricity Q and price P, if a transaction record shows Q=200MWh, P=50 yuan / MWh, the electricity Q and price P are format checked, consistent checked and range verified.

[0120] Format check: The quantity Q must be a real number greater than 0, and the price P must be a real number greater than or equal to 0; Consistency check: If the transaction type is buy, the price P should not be lower than the market minimum price L, that is, P ≥ L; Range verification: The power quantity Q should be within a reasonable range, for example, Q∈[10MWh,1000MWh].

[0121] 3.6 Work order flow tracking; In order to allow users to view the processing status and historical flow records of business work orders in real time, a work order flow tracking function is implemented in the new work order module. This function allows users to view real-time status updates, historical flow records, notification reminders and other messages.

[0122] Real-time status update: Users can check the current processing status of business work orders at any time, such as "New", "Processing", "Completed", etc.

[0123] Historical flow records: Users can view the entire flow process of a business work order from creation to completion, including the time, processor, and notes for each processing step.

[0124] Notification reminder: When the status of a business work order changes, the system will automatically notify relevant users.

[0125] For example, the entire flow process of a business work order from creation to completion is recorded in 10 steps, and the processing time and processor of each step are recorded in detail to ensure the transparency and traceability of work order processing.

[0126] 3.7 Logging; To facilitate tracking and auditing, the present invention also includes an integrated logging function for automatically recording the history of business data processing and user operations.

[0127] Among them, log records include operation timestamp, user information, operation type, operation details, data snapshot and other information.

[0128] Operation timestamp: records the exact time of each user operation.

[0129] User information: records the user ID and role that performs user operations.

[0130] Operation type: records the nature of user operations, such as data synchronization, data modification, work order creation, etc.

[0131] Operation details: record the specific content of the user's operation, such as which data fields were modified and which work orders were created.

[0132] Data snapshot: Save key data snapshots before and after user operations to facilitate comparison and auditing.

[0133] For example, in a certain month, the logging function records 1,000 operation records, including 500 data synchronizations, 200 data modifications, and 300 work order creations.

[0134] 3.8 Security audit; In order to have a complete security audit function, the present invention records and monitors business data processing activities, and once an abnormality is found, it can respond and take measures in a timely manner.

[0135] 4. Evidence of effect; By implementing the intelligent processing method of business data for auxiliary decision-making in power trading of the present invention, the accuracy and efficiency of business data processing are greatly improved, and the user experience improvement in user operation convenience and business work order creation speed is demonstrated as shown below.

[0136] User operation convenience has been improved to 90%: This data is supported by a user satisfaction survey, and the user satisfaction improvement ratio = (satisfaction after upgrade - satisfaction before upgrade) / satisfaction before upgrade × 100%. Specifically, before the system upgrade, based on a survey of 100 users, user satisfaction was 60%; after the upgrade, based on the same number of user surveys, user satisfaction increased to 90%.

[0137] The speed of creating business tickets has increased by at least 65%: This data is measured by comparing the time it takes to create business tickets before and after the upgrade. The speed increase ratio of work order creation = (work order creation time before the upgrade - work order creation time after the upgrade) / work order creation time before the upgrade × 100%. Specifically, it took an average of 15 seconds to create a work order before the system upgrade, but only 5 seconds after the upgrade. The speed increase ratio of business ticket creation is (15-5) / 15≈66.67%.

[0138] Please refer to the following Figure 3 , clearly describing the processing logic of the intelligent processing method of business data for auxiliary decision-making in power trading of the present invention.

[0139] After the user logs in to the system, authentication is first performed. If the authentication is successful, the data acquisition step is triggered; otherwise, the login interface is returned. In the data acquisition step, business data and matching data processing modules are collected in sequence. Next, the data processing step is executed to automatically fill in the template, clean / convert / verify the data, and generate the target data file in sequence. Then the display page is loaded. In the page display process, visual charts are rendered and real-time warnings are displayed in sequence. Then determine whether the user triggers a work order. If so, enter the work order processing step; otherwise, maintain monitoring status. In the work order processing process, create a new business work order, write the target data, and record the operation log in sequence. Finally, the work order processing results and monitoring results are aggregated to support decision analysis.

[0140] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for intelligent processing of business data for auxiliary decision-making in power trading, characterized in that: include: Data acquisition step: collecting business data related to power trading decision support, retrieving and matching data processing templates corresponding to the business data, wherein the data processing templates predefine the format, processing rules and verification standards of the business data to ensure the consistency and accuracy of the business data; Data processing steps: according to the requirements of the data processing template, the collected business data is automatically filled into the data processing template, and the Z score value of each electricity price data in the electricity price data set is first calculated. If the absolute value of the Z score value exceeds the preset threshold, it is determined to be an abnormal electricity value, and the lower limit of the reasonable electricity price range determined by the market situation is used to automatically correct the abnormal electricity value, so as to clean the electricity price data based on the Z score, and then the historical data is converted into value equivalent according to the inflation rate to make the inter-period data comparable, and then the data is verified; after the filling is completed, a target data file is generated, and the target data file contains processed standardized business data, which is convenient for subsequent analysis and decision-making; Page display step: after receiving the user's login instruction to log in to the power trading decision-making support system, the business data display page is loaded and displayed in the preset menu area of ​​the power trading decision-making support system to intuitively display the business data for the convenience of user understanding and operation; wherein, the dynamic dashboard is used to provide intuitive data display and real-time early warning notification functions; New work order step: In the business data display page, obtain the operation that triggers the new business work order process, automatically create a new business work order, and record the detailed information in the business work order for easy tracking and management; Data writing step: extract the required business data from the target data file according to the instructions of the data processing template, and write the extracted business data into the business work order to ensure that the work order content is complete and accurate; Data mining steps: mining valuable information and trends from business data through association rule analysis, cluster analysis and trend prediction analysis; Dynamic permission management steps: Different data views and operation options are displayed according to the user's role and permission level. The permission level corresponding to the administrator role can access all data, the permission level corresponding to the operator role can access 70% of the data, and the permission level corresponding to the auditor role can only access audit-related data; Among them, the real-time early warning notification function of the dynamic dashboard and the data mining steps based on cluster analysis are used to identify trading patterns, providing multi-dimensional decision support for auxiliary decision-making in power trading; by combining Z-score-based data cleaning operations with dynamic authority management, the conflict between data security and decision-making efficiency can be addressed.

2. The method according to claim 1, characterized in that The method further comprises: Data backup steps: Back up the collected business data to the cloud.

3. The method according to claim 1, characterized in that In the page display step, the transaction volume distribution in different time periods is displayed through a transaction volume change chart.

4. The method according to claim 1, characterized in that The method further comprises: Secondary data verification step: Secondary data verification is performed before business data is written into the business work order.

5. The method according to claim 1, characterized in that The method further comprises: Work order flow tracking steps: Display the processing status and historical flow records of business work orders.

6. The method according to claim 1, characterized in that The method further comprises: Logging step: Automatically record the history of business data processing and user operations.

7. The method according to claim 1, characterized in that The method further comprises: Security audit steps: Record and monitor business data processing activities.

8. An intelligent processing system for business data used for auxiliary decision-making in power trading, characterized in that: include: A data acquisition module is used to collect business data related to power trading auxiliary decision-making, retrieve and match data processing templates corresponding to the business data, and the data processing templates predefine the format, processing rules and verification standards of the business data to ensure the consistency and accuracy of the business data; The data processing module is used to automatically fill the collected business data into the data processing template according to the requirements of the data processing template, first calculate the Z score value of each electricity price data in the electricity price data set, and if the absolute value of the Z score value exceeds the preset threshold, it is determined to be an abnormal electricity value, and the lower limit of the reasonable electricity price range determined by the market situation is used to automatically correct the abnormal electricity value, so as to clean the electricity price data based on the Z score, and then perform value equivalent conversion of historical data according to the inflation rate to make the inter-period data comparable, and then perform data verification; after the filling is completed, a target data file is generated, and the target data file contains processed standardized business data, which is convenient for subsequent analysis and decision-making; The page display module is used to load and display the business data display page in the preset menu area of ​​the power trading auxiliary decision-making system after receiving the user's login instruction to log in to the power trading auxiliary decision-making system, so as to intuitively display the business data and facilitate the user's understanding and operation; wherein, the dynamic dashboard is used to provide intuitive data display and real-time early warning notification functions; The work order creation module is used to obtain the operation that triggers the process of creating a new business work order in the business data display page, automatically create a new business work order, and record the detailed information in the business work order for easy tracking and management; The data writing module is used to extract the required business data from the target data file according to the instructions of the data processing template, and write the extracted business data into the business work order to ensure that the work order content is complete and accurate; Data mining module, used to mine valuable information and trends from business data through association rule analysis, cluster analysis and trend prediction analysis; The dynamic authority management module is used to display different data views and operation options according to the user's role and authority level. The authority level corresponding to the administrator role can access all data, the authority level corresponding to the operator role can access 70% of the data, and the authority level corresponding to the auditor role can only access audit-related data; Among them, the real-time early warning notification function of the dynamic dashboard and the data mining module based on cluster analysis are used to identify trading patterns, providing multi-dimensional decision support for auxiliary decision-making in power trading; by combining Z-score-based data cleaning operations with dynamic authority management, the conflict between data security and decision-making efficiency can be addressed.

Images