Retail market reference price adjustment method based on electric power rule file content

Abstract

The invention discloses a retail market reference price adjustment method based on electric power rule file content, and belongs to the technical field of electric power marketing. An electric power transaction file and an electric power rule file are analyzed, and the content related to the electricity price in the electric power rule file is extracted and analyzed according to the electric power transaction file; the reference electricity price calculation rule is obtained by adjusting the reference electricity price calculation rule based on the analysis result, and the retail market reference electricity price is obtained through calculation according to the reference electricity price calculation rule, so that the problem that the electricity price making reasonable degree is low due to the fact that the electricity trading market is difficult to adjust the market rule in real time according to the updated content of the electricity rule file is solved; the electricity price formulation is more in line with market and user requirements, the electricity purchasing experience and the use satisfaction degree of the user are improved, and the speed of adjusting the electricity price in response to the electric power rule file is increased at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of power marketing, and in particular to a method for adjusting a retail market reference price based on the content of a power rule file. Background Art

[0002] The existing comprehensive reference price for the retail electricity market is calculated only in accordance with existing market rules by weighting historical price data. In judging the comprehensive reference price for the retail market, on the one hand, the data basis is not comprehensive enough and the impact of the spot market price level is not considered; on the other hand, comprehensive factors are not considered comprehensively, and content identification and analysis are not performed on various relevant electricity rule documents and other text files, and the content that may affect electricity prices is not considered.

[0003] Chinese Patent, Publication Number: CN112382006A, Publication Date: February 19, 2021, discloses a method for adjusting electricity prices, which comprises: receiving electricity price adjustment information from a marketing system, the electricity price adjustment information comprising user information and first electricity price information; obtaining the meter information of the corresponding electric energy meter according to the user information; checking and verifying the meter information, and adding the electric energy meter that has passed the verification to a list waiting for issuance; issuing the first electricity price information to the electric energy meter in the list waiting for issuance. Although there is a certain analysis of the relevant electric power rule files for the electricity price adjustment, the electricity price is adjusted by the analyzed content of the electric power rule files, which is mainly through manual identification, and the workload is relatively heavy, and it is impossible to automatically identify the content of the electric power rule files and adjust the electricity price. Summary of the invention

[0004] The present invention aims at the problem that the existing electricity price is mainly adjusted based on the set calculation rules, and it is difficult to adjust the calculation rules in real time according to the changes in the content of the electricity rule file, resulting in a low degree of rationality in electricity price formulation. The present invention provides a retail market reference price adjustment method based on the content of the electricity rule file, by analyzing the electricity trading file and the electricity rule file, extracting and analyzing the content related to the electricity price in the electricity rule file according to the electricity trading file, adjusting the benchmark electricity price calculation rules based on the analysis results to obtain the reference electricity price calculation rules, and calculating the retail market reference electricity price according to the reference electricity price calculation rules. The problem that the electricity trading market is difficult to adjust the market rules in real time according to the updated content of the electricity rule file, resulting in a low degree of rationality in electricity price formulation, makes the electricity price formulation more in line with the market and user needs, improves the user's electricity purchasing experience and usage satisfaction, and at the same time improves the speed at which the electricity price is adjusted in response to the electricity rule file.

[0005] In a first aspect, a technical solution provided in an embodiment of the present invention is: a method for adjusting a retail market reference price based on the content of a power rule file, comprising the following steps: S1. Obtain power transaction files and power rule files in real time through the content supervision mechanism, and obtain price feature data and power consumption feature data in power transaction files and power price associated data in power rule files through text recognition algorithms; S2, extracting the electricity price related data using the price feature data as the identification condition to obtain the target feature entry; S3. Perform correlation analysis on the target feature terms and price feature data to obtain the target management weight, and adjust the benchmark electricity price calculation rule according to the target management weight to obtain the reference electricity price calculation rule; S4. The reference electricity price calculation rule obtains the reference electricity price for the retail market in response to the electricity consumption characteristic data and the price characteristic data.

[0006] In this scheme, by obtaining power trading documents on the power trading platform, analyzing the price level of the power market through the power trading documents, and constructing benchmark electricity price calculation rules, the comprehensive situation of the power market is reflected more accurately. In order to reflect the comprehensive situation of the power market more accurately and make the adjusted electricity price more in line with user and market needs, the content of the power rule file is analyzed and the entries that have an impact on the electricity price are extracted. By further analyzing these entries, the benchmark electricity price calculation rules are further adjusted to obtain the reference electricity price calculation rules. The reference electricity price for the retail market calculated according to the reference electricity price calculation rules not only saves the traditional tedious steps of manually interpreting and extracting the content of the power rule file before adjusting the electricity price according to the power rule file, and formulating the adjustment rules, but also improves the speed and efficiency of electricity price updating, and also comprehensively analyzes the overall situation of the power market, so that the adjusted electricity price is more in line with market and user needs, making the electricity price more reasonable and improving user satisfaction.

[0007] Preferably, in S1, obtaining the electricity price associated data in the electricity rule file by using a text recognition algorithm includes the following steps: The electricity price setting data is obtained by extracting the electricity price setting content from the electricity rule file using the power generation cost and preferential regulations as identification conditions; Taking the supply and demand relationship as the identification condition, the electricity price classification content of the power rule file is extracted to obtain the electricity price classification data; Taking the operation status of the regulated object as the identification condition, the electricity price regulation content is extracted from the power rule file to obtain the electricity price regulation data; The electricity price setting data, electricity price classification data and electricity price supervision data are normalized to obtain electricity price related data.

[0008] In this scheme, the contents related to electricity prices in the electricity rules and documents can be roughly divided into three categories, including electricity price formulation contents, electricity price classification contents and electricity price supervision contents. Among them, electricity prices are formulated according to power generation costs and preferential regulations, and electricity prices are classified according to supply and demand relationships. According to the operating conditions of the regulated objects, supervision and management are mainly carried out in terms of the implementation of electricity price management regulations, the accuracy of electricity fee settlement, and the implementation of electricity price preferential policies. This is used to prevent unreasonable electricity price adjustments from affecting normal transactions between users and power generation companies. By screening electricity price contents and removing irrelevant content, the efficiency of screening electricity price-related contents in electricity rules and documents can be greatly improved, thereby ensuring the timeliness of electricity price updates and helping to keep electricity prices in the best state at all times.

[0009] Preferably, in S2, extracting the electricity price associated data using the price feature data as the identification condition to obtain the target feature entry includes the following steps: The annual transaction data, monthly transaction data and spot market transaction data in the price feature data are used to perform feature analysis on the electricity price related data to obtain the target feature terms.

[0010] In this solution, although the contents of electricity price associated data have an impact on electricity prices, it is unclear which part of the electricity price is affected. Therefore, in order to associate which part of the electricity price is affected in the electricity price associated data, the annual transaction data, monthly transaction data and spot market transaction data in the price feature data are used to perform feature analysis on the electricity price associated data, and the obtained target feature terms are associated with the corresponding electricity prices they affect, so as to facilitate the subsequent adjustment of the benchmark electricity price calculation rules.

[0011] Preferably, in S3, performing correlation analysis on the target feature terms and the price feature data to obtain the target management weight includes the following steps: Perform feature analysis on price feature data to obtain market price feature vectors, build a text content classification model, use the market price feature vectors and target feature terms as inputs to the text content classification model, and filter out key feature terms based on the correlation between the target feature terms and the market price feature vectors. The target management weight is obtained based on the impact of key characteristic terms on the comprehensive benchmark electricity price in the market.

[0012] In this solution, the main user of the text content classification model further screens the feature terms related to electricity prices in the power rule file, so that the text content analysis is more targeted to improve the calculation rate. Before building the text content classification model, the text content of the power rule file is cleaned, segmented, and stop words are removed to obtain a data set suitable for model training, and then the neural network model is used to extract features from the text data. The TextCNN model can be used to train the text content classification model. In the TextCNN model, different n-gram features are extracted through the convolution layer, and then the extracted feature sequence is encoded using the RNN; then word vector training is performed: in the TextCNN model, pre-trained word vectors are used to train the neural network model TextCNN to identify the key feature terms that affect the electricity market price. The key feature terms can be used to judge the degree of influence on the electricity price in the electricity market, which provides a data basis for the subsequent adjustment of the benchmark electricity price calculation rules based on the key feature terms.

[0013] Preferably, characteristic analysis is performed on the price characteristic data to obtain a market price characteristic vector, comprising the following steps: performing annual transaction characteristic analysis on the price characteristic data based on the annual transaction electricity price during period t to obtain an annual transaction average price characteristic vector; performing monthly transaction characteristic analysis on the price characteristic data based on the monthly transaction electricity price during period t to obtain a monthly transaction average price characteristic vector; performing spot transaction characteristic analysis on the price characteristic data based on the spot market electricity price during period t to obtain a spot transaction average price characteristic vector; and normalizing the annual transaction average price characteristic vector, the monthly transaction average price characteristic vector and the spot transaction average price characteristic vector to obtain a market price characteristic vector.

[0014] In this scheme, the electricity price in the power market is adjusted mainly by adjusting the annual transaction time-sharing average price, the monthly transaction time-sharing average price and the spot market time-sharing average price to correct the overall electricity price in the power market. Therefore, through the data characteristics of the annual, monthly and spot market transaction electricity prices of the power market, the power transaction price data and the user's electricity consumption data are feature analyzed to obtain feature vectors, which are used to match the feature terms subsequently extracted from the power rule file, find the feature terms corresponding to the adjustment of the time-sharing average price, and classify them, so as to adjust the corresponding time-sharing average price according to the corresponding feature terms to achieve the purpose of adjusting the overall electricity price, so that the extracted feature terms are closely related to the power market, and the useless content with little impact on the electricity price is further removed, so that the electricity price is more in line with market demand.

[0015] Preferably, obtaining the target management weight based on the influence of the key feature terms on the comprehensive benchmark electricity price in the market includes the following steps: Construct a market price characteristic index evaluation system based on the composition of market price characteristic vectors; Describe the key feature terms to obtain feature factors, standardize each feature factor to obtain the feature value of each feature factor, and determine the weight of each feature factor based on the feature value of each feature factor and the overall feature value of the overall feature factor; The weight of each characteristic factor is substituted into the market price characteristic index evaluation system to obtain the target management weight, which includes annual transaction weight, monthly transaction weight and spot transaction weight.

[0016] In this scheme, in order to obtain the proportion of key feature terms that have an impact on the annual, monthly and time-of-use average prices of the electricity market and the electricity spot market, feature analysis is performed on the key terms, and weight calculations are performed on them according to the corresponding features to obtain the proportion of the corresponding affected time-of-use electricity prices. The greater the proportion, the greater the degree of impact, and the smaller the proportion, the smaller the degree of impact. According to the degree of impact on the corresponding time-of-use electricity price, the market comprehensive benchmark electricity price is adjusted to obtain the retail market reference electricity price, so that the electricity price can be updated in time according to the updated content in the electricity rule file, thereby ensuring the real-time nature of the electricity price.

[0017] Preferably, in S3, adjusting the benchmark electricity price calculation rule according to the target management weight to obtain the reference electricity price calculation rule includes the following steps: The initial management weight in the benchmark electricity price calculation rule is replaced with the target management weight to obtain the reference electricity price calculation rule.

[0018] In this scheme, since the benchmark electricity price calculation rules for the electricity market have been formulated according to the historical electricity rule files before the electricity price is adjusted, when the new electricity rule files are released, the text content of the current electricity rule files is extracted and analyzed. After the analysis, the initial annual transaction weights, initial monthly transaction weights and initial spot transaction weights generated according to the historical electricity rule files are replaced, so that the current electricity market electricity price is updated according to the text content of the newly released electricity rule files, thereby improving the response rate of electricity price updates and making electricity prices more in line with market demand.

[0019] Preferably, in S4, the reference electricity price calculation rule obtains the retail market reference electricity price in response to the electricity consumption characteristic data and the price characteristic data, comprising the following steps: Based on the price characteristic data, the day-ahead market price and the real-time market price of time period t on day d are obtained, and the day-ahead market price and the real-time market price are weighted and summed based on the total power consumption of the user in the power consumption characteristic data to obtain the time-sharing average price of the power spot market; based on the price characteristic data, the annual transaction price and the monthly transaction price are obtained, and based on the time-sharing average price of the power spot market, the annual transaction time-sharing average price and the monthly transaction time-sharing average price of time period t are obtained; Obtain the market time-sharing reference price based on the annual transaction time-sharing average price and monthly transaction time-sharing average price of period t and the target management weight; The retail market reference electricity price is obtained based on the market time-of-use reference price and the user's electricity consumption in period t.

[0020] In this scheme, since it is necessary to analyze the comprehensive situation of the electricity market before calculating the reference electricity price in the retail market, it is necessary to first obtain the time-of-use average price of the electricity spot market, which is obtained by weighted summing the day-ahead market price and the real-time market price based on the user's total electricity consumption, and then calculating the annual transaction time-of-use average price and the monthly transaction time-of-use average price according to the time-of-use average price of the electricity spot market. Then, the market time-of-use reference price is calculated and the retail market reference electricity price is calculated according to the user's electricity consumption and management weights. Through layer-by-layer analysis of the electricity spot market and the annual and monthly time-of-use electricity prices, the electricity market price level is evaluated based on a huge data base, so that the obtained retail market reference electricity price is consistent with the electricity market price level, so that the electricity price remains within a normal range.

[0021] Preferably, the specific workflow of the content supervision mechanism includes the following steps: The update time of the power rule file is detected in real time. When the content of the power rule file is updated, the updated power rule file is used as the target power rule file, and the content that is different between the target power rule file and the historical power rule file is extracted.

[0022] In this scheme, when the power rule file is updated, the content that changes will not be much compared with the historical power rule file updated last time, so in order to improve the speed of power rule file analysis, it is only necessary to compare the contents of the target power rule file and the historical power rule file, extract the changed content and analyze the content, and the other unchanged content can directly extract the historical analysis results of the historical power rule file, so as to improve the speed of power rule file analysis and thus improve the response efficiency of electricity price updates.

[0023] Preferably, the day-ahead market price is obtained based on the day-ahead market electricity quantity of the user in the d-day period t and the unified settlement point price of the day-ahead market in the d-day period t; The real-time market price is obtained based on the actual electricity consumption of the user in the d-day period t and the unified settlement point price of the real-time market d-day period t.

[0024] In this scheme, when dividing the time periods, time period t starts from 0 o'clock and is divided into 48 time periods every half hour. The time period division is more detailed, which helps to prevent the problem of large fluctuations in time-of-use electricity prices and low reliability of time-of-use average prices. When calculating the day-ahead market price and the real-time market price, the unified settlement point price of time period t on the market day d is used to obtain it, which helps to collect data more regularly. The collected time-of-use electricity prices can correspond to each other according to the time periods, which helps to reduce the steps of data preprocessing and thus improve calculation efficiency.

[0025] Beneficial effects of the present invention: (1) The present invention obtains power trading files on a power trading platform with a content supervision mechanism, compares the contents of a target power trading file with those of a historical power trading file, extracts different contents, analyzes the extracted contents, and directly extracts historical analysis results of the historical power rule file for other unchanged contents, thereby improving the rate of analyzing the power rule file and thus improving the response efficiency of power price updates; (2) The present invention analyzes the content of the power rule file, extracts the terms that have an impact on the electricity price, and further analyzes these terms to further adjust the benchmark electricity price calculation rules to obtain the reference electricity price calculation rules. The reference electricity price for the retail market is calculated according to the reference electricity price calculation rules. This not only eliminates the tedious steps of manually interpreting and extracting the content of the power rule file and formulating adjustment rules before adjusting the electricity price according to the power rule file, thereby improving the speed and efficiency of electricity price updates, but also comprehensively analyzes the overall situation of the power market, so that the adjusted electricity price is more in line with market and user needs, making the electricity price more reasonable and improving user satisfaction.

[0026] The above invention content is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are only for the purpose of illustrating preferred embodiments and are not to be considered as limiting the present invention. Also, the same reference symbols are used throughout the drawings to represent the same parts.

[0028] Figure 1 The present invention is a flow chart of a method for adjusting a retail market reference price based on the content of an electricity rule file. DETAILED DESCRIPTION

[0029] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0030] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations (or steps) therein can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the drawings; the process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0031] Example: Figure 1 As shown, in order to solve the problem that the existing electricity price is mainly adjusted based on the set calculation rules, it is difficult to adjust the calculation rules in real time according to the changes in the content of the electricity rule file, resulting in a low degree of rationality in electricity price formulation, this embodiment provides a method for adjusting the retail market reference price based on the content of the electricity rule file, including the following steps: S1: Obtain power transaction files and power rule files in real time through the content supervision mechanism, and obtain price feature data and power consumption feature data in power transaction files and electricity price associated data in power rule files through text recognition algorithms.

[0032] In this embodiment, the electricity price associated data in the electricity rule file is obtained by using a text recognition algorithm, including the following steps: extracting the electricity price setting content of the electricity rule file using the power generation cost and preferential regulations as recognition conditions to obtain the electricity price setting data; Taking the supply and demand relationship as the identification condition, the electricity price classification content of the power rule file is extracted to obtain the electricity price classification data; Taking the operation status of the regulated object as the identification condition, the electricity price regulation content is extracted from the power rule file to obtain the electricity price regulation data; The electricity price setting data, electricity price classification data and electricity price supervision data are normalized to obtain electricity price related data.

[0033] The contents related to electricity prices in the electricity rule file of this embodiment can be roughly divided into three categories, including electricity price formulation contents, electricity price classification contents and electricity price supervision contents, among which electricity prices are formulated according to power generation costs and preferential regulations, and electricity prices are classified according to supply and demand relationships. According to the operating conditions of the supervised objects, supervision and management are mainly carried out in terms of the implementation of electricity price management regulations, the accuracy of electricity fee settlement, and the implementation of electricity price preferential policies, so as to prevent unreasonable electricity price adjustments from affecting normal transactions between users and power generation companies. By screening the electricity price contents and removing irrelevant contents, the efficiency of screening the electricity price-related contents in the electricity rule files can be greatly improved, thereby ensuring the timeliness of electricity price updates and helping to keep electricity prices in the best state at all times.

[0034] S2: Using price feature data as identification conditions, extract electricity price related data to obtain target feature terms.

[0035] In this embodiment, the price feature data is used as the identification condition to extract the electricity price associated data to obtain the target feature entry, including the following steps: The annual transaction data, monthly transaction data and spot market transaction data in the price feature data are used to perform feature analysis on the electricity price related data to obtain the target feature terms.

[0036] Although the contents of the electricity price associated data in this embodiment have an impact on the electricity price, it is unclear which part of the electricity price is affected. Therefore, in order to associate which part of the electricity price is specifically affected in the electricity price associated data, the electricity price associated data is subjected to feature analysis based on the annual transaction data, monthly transaction data and spot market transaction data in the price feature data, and the obtained target feature terms are associated with the corresponding electricity prices they affect, so as to facilitate the subsequent adjustment of the benchmark electricity price calculation rules.

[0037] S3: Perform correlation analysis on the target feature terms and price feature data to obtain the target management weight, and adjust the benchmark electricity price calculation rule according to the target management weight to obtain the reference electricity price calculation rule.

[0038] In this embodiment, the correlation analysis of the target feature terms and the price feature data is performed to obtain the target management weight, including the following steps: Perform feature analysis on price feature data to obtain market price feature vectors, build a text content classification model, use the market price feature vectors and target feature terms as inputs to the text content classification model, and filter out key feature terms based on the correlation between the target feature terms and the market price feature vectors. The target management weight is obtained based on the impact of key characteristic terms on the comprehensive benchmark electricity price in the market.

[0039] The text content classification model of this embodiment is mainly used by users to further screen the feature terms related to electricity prices in the power rule file, so that the text content analysis is more targeted to improve the calculation rate. Before building the text content classification model, the text content of the power rule file is cleaned, segmented, and stop words are removed to obtain a data set suitable for model training, and then a neural network model is used to extract features from the text data. The TextCNN model can be used to train the text content classification model. In the TextCNN model, different n-gram features are extracted through the convolution layer, and then the extracted feature sequence is encoded using the RNN; then word vector training is performed: in the TextCNN model, pre-trained word vectors are used to train the neural network model TextCNN to identify key feature terms that affect the electricity market price. The key feature terms can be used to judge the degree of influence on the electricity price in the electricity market, which provides a data basis for the subsequent adjustment of the benchmark electricity price calculation rules based on the key feature terms.

[0040] When extracting and analyzing the text content of the power rule file, the TextCNN network, as a trained neural network, can automatically extract features of different lengths from the power rule file text through convolution kernels of different sizes, thereby capturing local and global semantic information. The training and prediction speed of TextCNN is fast, which is suitable for large-scale text data such as power rule files. The structure is more intuitive, easy to understand and implement. At the same time, due to the characteristics of parameter sharing and sparse interaction, the computational complexity is effectively reduced, thereby increasing the computational rate, thereby improving the text recognition efficiency and making the electricity price update more timely.

[0041] In this embodiment, characteristic analysis is performed on price characteristic data to obtain a market price characteristic vector, including the following steps: performing annual transaction characteristic analysis on the price characteristic data based on the annual transaction electricity price during period t to obtain an annual transaction average price characteristic vector; performing monthly transaction characteristic analysis on the price characteristic data based on the monthly transaction electricity price during period t to obtain a monthly transaction average price characteristic vector; performing spot transaction characteristic analysis on the price characteristic data based on the spot market electricity price during period t to obtain a spot transaction average price characteristic vector; and normalizing the annual transaction average price characteristic vector, the monthly transaction average price characteristic vector, and the spot transaction average price characteristic vector to obtain a market price characteristic vector.

[0042] In this embodiment, the electricity price in the electricity market is adjusted mainly by adjusting the annual transaction time-sharing average price, the monthly transaction time-sharing average price and the spot market time-sharing average price, so as to correct the overall electricity price in the electricity market. Therefore, through the data characteristics of the annual, monthly and spot market transaction electricity prices in the electricity market, the characteristics of the electricity transaction price data and the user's electricity consumption data are analyzed to obtain a feature vector, which is used to match the feature terms subsequently extracted from the electricity rule file, find the feature terms corresponding to the adjustment of the time-sharing average price, and classify them, so as to adjust the corresponding time-sharing average price according to the corresponding feature terms to achieve the purpose of adjusting the overall electricity price, so that the extracted feature terms are closely related to the electricity market, and the useless content with less impact on the electricity price is further removed, so that the electricity price is more in line with market demand.

[0043] In this embodiment, obtaining the target management weight based on the influence of the key feature terms on the comprehensive benchmark electricity price in the market includes the following steps: Construct a market price characteristic index evaluation system based on the composition of market price characteristic vectors; Describe the key feature terms to obtain feature factors, standardize each feature factor to obtain the feature value of each feature factor, and determine the weight of each feature factor based on the feature value of each feature factor and the overall feature value of the overall feature factor; The weight of each characteristic factor is substituted into the market price characteristic index evaluation system to obtain the target management weight, which includes annual transaction weight, monthly transaction weight and spot transaction weight.

[0044] In order to obtain the proportion of key feature terms that have an impact on the annual, monthly and time-of-use average prices of the electricity market and the electricity spot market, this embodiment performs feature analysis on the key terms and calculates their weights according to the corresponding features to obtain the proportion of the corresponding affected time-of-use electricity prices. The greater the proportion, the greater the degree of impact, and the smaller the proportion, the smaller the degree of impact. The comprehensive market benchmark electricity price is adjusted according to the degree of impact on the corresponding time-of-use electricity price to obtain the retail market reference electricity price, so that the electricity price can be updated in time according to the updated content in the electricity rule file, thereby ensuring the real-time nature of the electricity price.

[0045] In this embodiment, the base electricity price calculation rule is adjusted according to the target management weight to obtain the reference electricity price calculation rule, including the following steps: The initial management weight in the benchmark electricity price calculation rule is replaced with the target management weight to obtain the reference electricity price calculation rule.

[0046] In this embodiment, since the benchmark electricity price calculation rules for the electricity market have been formulated according to the historical electricity rule file before the electricity price is adjusted, when the new electricity rule file is released, the text content of the current electricity rule file is extracted and analyzed. After the analysis, the initial annual transaction weight, initial monthly transaction weight and initial spot transaction weight generated according to the historical electricity rule file are replaced, so that the current electricity market electricity price is updated according to the text content of the newly released electricity rule file, thereby improving the response rate of electricity price updates and making the electricity price more in line with market demand.

[0047] S4: The reference electricity price calculation rule obtains the retail market reference electricity price in response to the electricity consumption characteristic data and the price characteristic data.

[0048] In this embodiment, the reference electricity price calculation rule obtains the retail market reference electricity price in response to the electricity consumption characteristic data and the price characteristic data, including the following steps: Based on the price characteristic data, the day-ahead market price and real-time market price of time period t on day d are obtained. The day-ahead market price and real-time market price are weighted and summed based on the total power consumption of the users in the power consumption characteristic data to obtain the time-sharing average price of the power spot market. The formula is as follows: P 现货,t =Σ d [Q 日前,d,t ×P 日前,d,t +(Q 实际,d,t -Q 日前,d,t )×P 实时,d,t ] / Q 实际,t Among them, Q 实际,d,t is the actual electricity consumption of all users directly participating in market transactions during period t on day d, Q 日前,d,t is the day-ahead market electricity consumption of all users directly participating in market transactions in period t on day d, P 日前,d,t is the unified settlement point price of the day-ahead market during period t on day d, P 实时,d,t It is the unified settlement point price of the real-time market on day d during period t. 实际,t The total actual electricity consumption of all users directly participating in market transactions in time period t on each day of the month is calculated as follows: Q 实际,t =Σ d Q 实际,d,t ; Based on the price feature data, the annual transaction price and monthly transaction price are obtained, and based on the time-sharing average price of the electricity spot market, the annual transaction time-sharing average price and monthly transaction time-sharing average price of period t are obtained. The calculation formula is as follows: P 年度,t =P 年度 ×P 现货,t ×Σ t Q实际,t / Σ t (Q 实际,t ×P 现货,t ) P 月度,t =P 月度 ×P 现货,t ×Σ t Q 实际,t / Σ t (Q 实际,t ×P 现货,t ) Among them, P 年度 It is the overall average price of annual transactions, which refers to the weighted average price of various annual (monthly) medium- and long-term transactions organized by the provincial trading platforms.

[0049] The market time-sharing reference price is obtained based on the annual transaction time-sharing average price and monthly transaction time-sharing average price of time period t and the target management weight. The time-sharing reference price is calculated on a rolling basis on a natural month basis, and 48 time-sharing reference prices are formed every half hour, which are the weighted averages of the annual transaction time-sharing average price, monthly transaction time-sharing average price and the day-ahead market time-sharing average price of the corresponding time period. The relevant weights are determined by the relevant work notification on market-oriented transactions, and the calculation formula is: P 套餐,t =k 年度 ×P 年度,t +k 月度 ×P 月度,t +k 现货 ×P 现货,t Among them, P 年度,t , P 月度,t , P 现货,t are the annual transaction time-sharing average price, monthly transaction time-sharing average price and spot market time-sharing average price in period t, respectively. 年度 , k 月度 , k 现货 are their weights respectively.

[0050] The retail market reference electricity price is obtained based on the market time-of-use reference price and the user's electricity consumption in period t. The formula is as follows: P 套餐，i =Σ t (Q 用户,i,t ×P 套餐,t ) / Σ t Q 用户,i,t Among them, the time period t starts from 0 o'clock and is divided into 48 time periods per half hour in a day. 用户,i,t The sum of the electricity consumption of retail user i in time period t on each day of the month is calculated as follows: Q 用户,i,t =Σ d Q 用户,d,t Among them, Q 用户,i,d,t is the electricity consumption of retail user i in period t on day d.

[0051] In this embodiment, since it is necessary to analyze the comprehensive situation of the electricity market before calculating the reference electricity price for the retail market, it is necessary to first obtain the time-of-use average price of the electricity spot market, which is obtained by weighted summing the day-ahead market price and the real-time market price based on the user's total electricity consumption, and then calculating the annual transaction time-of-use average price and the monthly transaction time-of-use average price according to the time-of-use average price of the electricity spot market, and then calculating the market time-of-use reference price and calculating the retail market reference electricity price according to the user's electricity consumption and management weights. By analyzing the electricity spot market and the annual and monthly time-of-use electricity prices layer by layer, the electricity market price level is evaluated based on a huge data base, so that the obtained retail market reference electricity price is consistent with the electricity market price level, so that the electricity price remains within a normal range.

[0052] This embodiment obtains power transaction files on the power transaction platform, analyzes the price level of the power market through the power transaction files, and constructs benchmark electricity price calculation rules, which more accurately reflects the comprehensive situation of the power market. In order to reflect the comprehensive situation of the power market more accurately and make the adjusted electricity price more in line with user and market needs, the content of the power rule file is analyzed to extract the terms that have an impact on the electricity price. The benchmark electricity price calculation rules are further adjusted by further analyzing these terms to obtain the reference electricity price calculation rules. The reference electricity price for the retail market is calculated according to the reference electricity price calculation rules, which not only saves the tedious steps of manually interpreting and extracting the content of the power rule file before adjusting the electricity price according to the power rule file and formulating the adjustment rules, but also improves the speed and efficiency of electricity price updating, and also comprehensively analyzes the overall situation of the power market, so that the adjusted electricity price is more in line with market and user needs, makes the electricity price more reasonable, and improves user satisfaction.

[0053] It can be seen from the above embodiments that at least the following substantial effects are achieved: (1) The present invention obtains power transaction files on a power transaction platform with a content supervision mechanism, compares the contents of a target power transaction file with those of a historical power transaction file, extracts different contents, analyzes the extracted contents, and directly extracts historical analysis results of the historical power rule file for other unchanged contents, thereby improving the speed of analyzing the power rule file and thus improving the response efficiency of power price updates; (2) The present invention analyzes the content of the power rule file, extracts the terms that have an impact on the electricity price, and further analyzes these terms to further adjust the benchmark electricity price calculation rules to obtain the reference electricity price calculation rules. The reference electricity price for the retail market is calculated according to the reference electricity price calculation rules. This not only eliminates the tedious steps of manually interpreting and extracting the content of the power rule file and formulating adjustment rules before adjusting the electricity price according to the power rule file, thereby improving the speed and efficiency of electricity price updates, but also comprehensively analyzes the overall situation of the power market, so that the adjusted electricity price is more in line with market and user needs, making the electricity price more reasonable and improving user satisfaction.

[0054] The specific implementation described above is a preferred implementation of the retail market reference price adjustment method based on the content of the electricity rule file of the present invention, and is not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation. All equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A method for adjusting a retail market reference price based on the content of an electricity rule file, characterized in that: The following steps are involved: S1. Obtain power transaction files and power rule files in real time through the content supervision mechanism, and obtain price feature data and power consumption feature data in power transaction files and power price associated data in power rule files through text recognition algorithms; S2, extracting the electricity price related data using the price feature data as the identification condition to obtain the target feature entry; S3. Perform correlation analysis on the target feature terms and price feature data to obtain the target management weight, and adjust the benchmark electricity price calculation rule according to the target management weight to obtain the reference electricity price calculation rule; S4. The reference electricity price calculation rule obtains the reference electricity price for the retail market in response to the electricity consumption characteristic data and the price characteristic data.

2. A method for adjusting the retail market reference price based on the content of the power rule file according to claim 1, characterized in that: In S1, the electricity price associated data in the electricity rule file is obtained by using a text recognition algorithm, including the following steps: The electricity price setting data is obtained by extracting the electricity price setting content from the electricity rule file using the power generation cost and preferential regulations as identification conditions; Taking the supply and demand relationship as the identification condition, the electricity price classification content of the electricity rule file is extracted to obtain the electricity price classification data; Taking the operation status of the regulated object as the identification condition, the electricity price regulation content is extracted from the power rule file to obtain the electricity price regulation data; The electricity price setting data, electricity price classification data and electricity price supervision data are normalized to obtain electricity price related data.

3. A method for adjusting the retail market reference price based on the content of the power rule file according to claim 1, characterized in that: In S2, the price feature data is used as the identification condition to extract the electricity price associated data to obtain the target feature entry, including the following steps: The annual transaction data, monthly transaction data and spot market transaction data in the price feature data are used to perform feature analysis on the electricity price related data to obtain the target feature terms.

4. A method for adjusting the retail market reference price based on the content of the power rule file according to claim 1, characterized in that: In S3, the target feature terms and the price feature data are subjected to correlation analysis to obtain the target management weight, including the following steps: characteristic analysis is performed on the price feature data to obtain a market price feature vector, a text content classification model is constructed, the market price feature vector and the target feature terms are used as inputs of the text content classification model, and the electricity price feature terms are screened based on the correlation between the target feature terms and the market price feature vector to output key feature terms; The target management weight is obtained based on the impact of key characteristic terms on the comprehensive benchmark electricity price in the market.

5. A method for adjusting the retail market reference price based on the content of the power rule file according to claim 4, characterized in that: Performing feature analysis on price feature data to obtain a market price feature vector includes the following steps: Based on the annual transaction electricity price in period t, the price characteristic data is subjected to annual transaction characteristic analysis to obtain the annual transaction average price characteristic vector; based on the monthly transaction electricity price in period t, the price characteristic data is subjected to monthly transaction characteristic analysis to obtain the monthly transaction average price characteristic vector; based on the spot market electricity price in period t, the price characteristic data is subjected to spot transaction characteristic analysis to obtain the spot transaction average price characteristic vector; the annual transaction average price characteristic vector, the monthly transaction average price characteristic vector and the spot transaction average price characteristic vector are normalized to obtain the market price characteristic vector.

6. A method for adjusting the retail market reference price based on the content of the power rule file according to claim 4, characterized in that: Obtaining the target management weight based on the impact of key feature terms on the comprehensive benchmark electricity price in the market includes the following steps: Construct a market price characteristic index evaluation system based on the composition of market price characteristic vectors; Describe the key feature terms to obtain feature factors, standardize each feature factor to obtain the feature value of each feature factor, and determine the weight of each feature factor based on the feature value of each feature factor and the overall feature value of the overall feature factor; The weight of each characteristic factor is substituted into the market price characteristic index evaluation system to obtain the target management weight, which includes annual transaction weight, monthly transaction weight and spot transaction weight.

7. A method for adjusting the retail market reference price based on the content of the power rule file according to claim 1, characterized in that: In S3, the benchmark electricity price calculation rule is adjusted according to the target management weight to obtain the reference electricity price calculation rule, including the following steps: The initial management weight in the benchmark electricity price calculation rule is replaced with the target management weight to obtain the reference electricity price calculation rule.

8. A method for adjusting a retail market reference price based on the content of an electricity rule file according to claim 1, characterized in that: In S4, the reference electricity price calculation rule obtains the retail market reference electricity price in response to the electricity consumption characteristic data and the price characteristic data, including the following steps: Based on the price characteristic data, the day-ahead market price and the real-time market price of time period t on day d are obtained, and the day-ahead market price and the real-time market price are weighted and summed based on the total power consumption of the user in the power consumption characteristic data to obtain the time-sharing average price of the power spot market; Obtain the annual transaction price and monthly transaction price based on the price feature data, and obtain the annual transaction time-sharing average price and monthly transaction time-sharing average price for period t based on the time-sharing average price of the electricity spot market; Obtain the market time-sharing reference price based on the annual transaction time-sharing average price and monthly transaction time-sharing average price of period t and the target management weight; The retail market reference electricity price is obtained based on the market time-of-use reference price and the user's electricity consumption in period t.

9. A method for adjusting a retail market reference price based on the content of an electricity rule file according to claim 1, characterized in that: The specific workflow of the content supervision mechanism includes the following steps: The update time of the power rule file is detected in real time. When the content of the power rule file is updated, the updated power rule file is used as the target power rule file, and the content that is different between the target power rule file and the historical power rule file is extracted.

10. A method for adjusting the retail market reference price based on the content of the power rule file according to claim 8, characterized in that: The day-ahead market price is obtained based on the day-ahead market electricity quantity of the user in the d-day period t and the unified settlement point price of the day-ahead market in the d-day period t; The real-time market price is obtained based on the actual electricity consumption of the user in the d-day period t and the unified settlement point price of the real-time market d-day period t.

Images