Power-assisted transaction decision support method and system

Abstract

The invention provides an electric power auxiliary transaction decision support method and system, and particularly relates to the technical field of electric power transaction management, and the method comprises the steps: carrying out the electricity price prediction and supply and demand prediction of various types of electric power markets, and obtaining a market transaction structure prediction vector of each type of electric power market, the market transaction structure prediction vector comprises a transaction electricity price and a market electricity supply-demand ratio; according to the transaction characteristics of the various types of power markets, power generation cost prediction and power generation capacity prediction are carried out on the station, station transaction structure prediction vectors of the station under the condition that the station participates in the various types of power markets are obtained, and the station transaction structure prediction vectors comprise cost electricity prices and power generation capacity; and based on the market transaction structure prediction vector and the station transaction structure prediction vector, calculating the unit marginal contribution of the station under the condition that the station participates in various types of power markets. According to the scheme, the accuracy and rationality of station quotation can be improved, and losses caused by quotation errors are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power transaction management, and in particular to a power auxiliary transaction decision support method and system. Background Art

[0002] Due to the high complexity, long-term and iterative nature of the construction of the power trading system, high-quality market trading decisions are a huge challenge facing all participants. Electricity is a special commodity that cannot be stored in large quantities. Its price changes cyclically due to various factors. The trading, delivery, production and use of electricity all occur simultaneously at the moment of electricity consumption. Its trading strictly follows the constraints of power system dispatching and operation, which requires both commodity trading logic and proficiency in power system operation, and has extremely high technical barriers.

[0003] At present, the stations are facing the situation of participating in multiple power markets at the same time, including the medium- and long-term power market, the power spot market, the power auxiliary service market and other types of markets. Due to the lack of in-depth understanding of some stations for participating in multiple types of power markets at the same time, they have made mistakes in quotation, provided wrong signals to the power grid and the trading center, and suffered losses. Therefore, it is necessary to establish an auxiliary decision-making system for power trading that participates in multiple power markets to assist station personnel in making price declarations and making decisions on market operations. Summary of the invention

[0004] The present invention provides an electric power auxiliary transaction decision support method and system, which can improve the accuracy and rationality of site quotations and reduce losses caused by quotation errors.

[0005] In a first aspect, the present invention provides a method for supporting decision making in power assisted trading, the method comprising:

[0006] Performing electricity price forecasting and supply and demand forecasting for each type of electricity market, and obtaining a market transaction structure forecasting vector for each type of electricity market, wherein the market transaction structure forecasting vector includes a transaction electricity price and a market electricity supply and demand ratio;

[0007] According to the transaction characteristics of various types of power markets, the power generation cost and power generation of the station are predicted, and the station transaction structure prediction vector is obtained when the station participates in various types of power markets. The station transaction structure prediction vector includes the cost electricity price and the power generation;

[0008] Based on the market transaction structure prediction vector and the station transaction structure prediction vector, calculating the unit marginal contribution of the station when participating in various types of power markets;

[0009] For each electricity market, the unit marginal contribution and the market electricity supply and demand ratio are used as traversal conditions, and the pre-built site transaction record database is traversed to screen out the optimal transaction records, extract the electricity transaction volume corresponding to the optimal transaction record, and use the electricity transaction volume and transaction electricity price as the trading strategy for the site to participate in this type of electricity market.

[0010] Optionally, provide supply and demand forecasts for various types of electricity markets, including:

[0011] Collecting a set of supply and demand influencing parameters within a time window to be predicted, wherein the set of supply and demand influencing parameters includes meteorological quantitative parameters, holiday quantitative parameters, and policy quantitative parameters;

[0012] Taking the time window to be predicted as the extraction condition, extract the basic power supply and basic power demand from the pre-built supply-demand ratio benchmark database;

[0013] Calculate the ratio of basic power supply to basic power demand to obtain a benchmark supply-demand ratio corresponding to the time window to be predicted;

[0014] Conduct a comprehensive supply-demand ratio impact analysis on each influencing parameter in the supply-demand influencing parameter set to obtain the supply-demand impact rate;

[0015] Based on the supply and demand impact rate and the benchmark supply and demand ratio, the market electricity supply and demand ratio of the electricity market within the time window to be predicted is calculated.

[0016] Optionally, electricity price forecasts are performed for various types of electricity markets, including:

[0017] Collect statistics on historical transaction data within a set period of time in the power market, filter out transaction records with the same power supply and demand ratio in the market, and mark all the filtered transaction records as analog transaction records;

[0018] Extract features from all analog transaction records to obtain historical transaction electricity prices and historical electricity surplus and shortage values ​​in each analog transaction record;

[0019] Conduct correlation analysis on historical transaction electricity prices and historical electricity surplus and shortage values ​​to obtain the electricity price surplus and shortage correlation matrix;

[0020] Calculate the difference between the basic power supply and the basic power demand to obtain the predicted power surplus or shortage value;

[0021] The historical transaction electricity price with the strongest correlation with the predicted electricity surplus or shortage value is extracted from the electricity price surplus or shortage correlation matrix as the transaction electricity price in the power market within the time window to be predicted.

[0022] Optionally, the unit marginal contribution and the market power supply-demand ratio are used as traversal conditions to traverse the pre-built site transaction record database, including:

[0023] Obtain the unit marginal contribution of the station when participating in various types of power markets and the market power supply-demand ratio in the market transaction structure prediction vector;

[0024] In the station transaction record database, find the record closest to the currently predicted unit marginal contribution; find the record closest to the currently predicted market power supply and demand ratio; on the basis of satisfying the two conditions of unit marginal contribution and market power supply and demand ratio, select the record with the highest total profit as the optimal transaction record;

[0025] The corresponding power transaction volume and transaction price are extracted from the selected optimal transaction records, and the extracted power transaction volume and transaction price are used as the trading strategy for the site to participate in this type of power market.

[0026] Optionally, a preset comprehensive evaluation mechanism is used to score the transaction records in the station transaction record database respectively, and the best transaction record is determined according to the scoring results, including:

[0027] For each transaction record, calculate the proximity of the unit marginal contribution; calculate the proximity of the market power supply and demand ratio; extract the total profit of the transaction record;

[0028] Normalize the proximity of unit marginal contribution, the proximity of market power supply-demand ratio, and total profit;

[0029] The standardized unit marginal contribution proximity, the market electricity supply-demand ratio proximity and the total profit are weighted to obtain a comprehensive score for electricity trading.

[0030] Optionally, the comprehensive score of power trading is calculated by the following formula:

[0031]

[0032] Among them, G i represents the comprehensive score of the power transaction of the ith power transaction record, t i represents the transaction price of the ith electricity transaction record, c i represents the cost price of the i-th electricity transaction record, g i represents the power generation of the ith power transaction record, s i represents the market power supply-demand ratio of the i-th power transaction record, p i represents the total profit of the ith electricity transaction record, represents the predicted transaction electricity price, represents the predicted cost electricity price, represents the predicted cost electricity price, represents the predicted market electricity supply-demand ratio, ∈ is a positive coefficient used to avoid division by zero error, w1, w2 and w3 are the weights of unit marginal contribution proximity, supply-demand ratio proximity and total profit respectively; f1, f2 and f3 are the standardized functions of unit marginal contribution proximity, supply-demand ratio proximity and total profit respectively.

[0033] On the other hand, the present application also provides an electric power auxiliary transaction decision support system, the system comprising:

[0034] The market prediction module is used to predict the electricity price and supply and demand of various types of electricity markets to obtain the market transaction structure prediction vector of each type of electricity market; the market transaction structure prediction vector includes the transaction electricity price and the market electricity supply and demand ratio;

[0035] The station prediction module is used to predict the power generation cost and power generation of the station according to the transaction characteristics of various types of power markets, and generate the station transaction structure prediction vector when the station participates in various types of power markets; the station transaction structure prediction vector includes the cost electricity price and the expected power generation;

[0036] The profit calculation module is used to calculate the unit marginal contribution of the station when participating in various types of power markets based on the market transaction structure prediction vector and the station transaction structure prediction vector;

[0037] The strategy screening module is used to traverse the pre-built station transaction record database for each power market, using the unit marginal contribution and the market power supply and demand ratio as traversal conditions, and screen out the optimal transaction record; and extract the corresponding power transaction volume from the optimal transaction record, and combine it with the current transaction electricity price to form a trading strategy for the station to participate in this type of power market;

[0038] The decision output module is used to display the trading strategies generated by the strategy screening module to the station personnel, and assist the station personnel in making price declarations and market operation decisions.

[0039] In a third aspect, an embodiment of the present invention further provides a computing device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method described in any first aspect of this specification is implemented.

[0040] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, enables the computer to execute the method described in any first aspect of this specification.

[0041] In a fifth aspect, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the method described in any first aspect of this specification are implemented.

[0042] The embodiment of the present invention provides a method and system for supporting decision-making in power-assisted trading. The method constructs a comprehensive market trading structure and a prediction vector of a station trading structure by predicting electricity prices and supply and demand for various types of power markets, as well as predicting the power generation cost and power generation of the station itself, which helps the station to grasp the market dynamics more accurately and provide a solid data basis for formulating trading strategies. By calculating the unit marginal contribution, the profitability of the station in different power markets can be accurately evaluated, which helps the station to give priority to trading in markets with stronger profitability, thereby optimizing resource allocation and improving overall economic benefits. By using a pre-built station transaction record database, the transaction records that best match the current market conditions are traversed and screened, and trading strategies are formulated accordingly, which can improve the accuracy and rationality of the station's quotation and reduce losses caused by quotation errors. Taking into account the complexity and variability of the power market, the transaction record database can be continuously learned and updated to gradually adapt to market changes, improve the flexibility and adaptability of the decision-making system, and effectively support the station's participation in trading decisions in multiple types of power markets in the long term. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 It is a flow chart of a power auxiliary transaction decision support method provided by an embodiment of the present invention;

[0045] Figure 2 is a hardware architecture diagram of a computing device provided by an embodiment of the present invention;

[0046] Figure 3 It is a structural schematic diagram of an electric power auxiliary transaction decision support system provided by an embodiment of the present invention.

[0047] Figure 4 It is a table schematic diagram of a station transaction record database provided by an embodiment of the present invention;

[0048] Figure 5 It is a schematic diagram of implementing a query using SQL in step 402 provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in 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.

[0050] The specific implementation of the concept of this application is described below.

[0051] Please refer to Figure 1 The embodiment of the present invention provides a method for supporting power auxiliary trading decision, the method comprising:

[0052] Step 100, forecasting electricity prices and supply and demand for various types of electricity markets, and obtaining a market transaction structure forecast vector for each type of electricity market, where the market transaction structure forecast vector includes a transaction electricity price and a market electricity supply and demand ratio;

[0053] Step 200, according to the transaction characteristics of various types of power markets, the power generation cost and power generation of the station are predicted, and the station transaction structure prediction vector is obtained when the station participates in various types of power markets. The station transaction structure prediction vector includes the cost electricity price and the power generation;

[0054] Step 300, based on the market transaction structure prediction vector and the station transaction structure prediction vector, calculate the unit marginal contribution of the station when participating in various types of power markets; the unit marginal contribution is the unit profit between the transaction price and the cost price;

[0055] Step 400, for each power market, with the unit marginal contribution and the market power supply-demand ratio as traversal conditions, traverse the pre-built site transaction record database, screen out the transaction records that are closest to the unit marginal contribution and the market power supply-demand ratio and have the highest total profit, extract the power transaction volume corresponding to the transaction record, and use the power transaction volume and transaction electricity price as the trading strategy for the site to participate in this type of power market.

[0056] In this embodiment, by predicting electricity prices and supply and demand for various types of power markets, as well as predicting the power generation cost and power generation of the station itself, a comprehensive market transaction structure and station transaction structure prediction vector are constructed, which helps the station to grasp the market dynamics more accurately and provide a solid data basis for formulating trading strategies; by calculating the unit marginal contribution, the profitability of the station in different power markets can be accurately evaluated, which helps the station to give priority to trading in markets with stronger profitability, thereby optimizing resource allocation and improving overall economic benefits; by using the pre-built station transaction record database, the transaction records that best match the current market conditions are traversed and screened, and trading strategies are formulated accordingly, which can improve the accuracy and rationality of the station's quotation and reduce losses caused by quotation errors; considering the complexity and variability of the power market, it is also possible to gradually adapt to market changes by continuously learning and updating the transaction record database, improve the flexibility and adaptability of the decision-making system, and effectively support the station's participation in trading decisions in multiple types of power markets in the long term.

[0057] Described below Figure 1 How the various steps are performed.

[0058] In step 100, the transaction electricity price and market electricity supply-demand ratio of each type of electricity market are mainly predicted to provide data support for the subsequent steps of the power transaction auxiliary decision support method.

[0059] The power supply and demand ratio prediction methods of the power market include:

[0060] Step 111, collecting a set of supply and demand influencing parameters within the time window to be predicted, the supply and demand influencing parameter set includes meteorological quantitative parameters, holiday quantitative parameters and policy quantitative parameters; wherein meteorological quantitative parameters include temperature, humidity, wind speed, etc., meteorological quantitative parameters directly affect electricity demand, especially residential and commercial electricity consumption, for example, high temperature weather will increase the use of air conditioners, thereby pushing up electricity demand; during holidays, electricity consumption behavior will change, such as industrial electricity consumption may decrease, while residential electricity consumption may increase, by quantifying the changes caused by holidays, electricity demand can be predicted more accurately; policy quantitative parameters include electricity price adjustment, energy conservation and emission reduction policies, etc., which will also have an important impact on electricity supply and demand, and policy changes will lead to the encouragement or restriction of certain electricity consumption behaviors, thereby affecting the supply and demand relationship in the electricity market; through a data collection system or a public data source, obtain the specific values ​​or quantitative indicators of various supply and demand influencing parameters within the time window to be predicted;

[0061] Step 112, taking the time window to be predicted as the extraction condition, extracting the basic power supply and basic power demand from a pre-built supply-demand ratio benchmark database; the supply-demand ratio benchmark database contains historical power supply and power demand data, and is classified and stored by time window (such as day, week, month, etc.); using database query technology, retrieve the basic power supply and basic power demand data matching the time window to be predicted from the supply-demand ratio benchmark database;

[0062] Step 113, calculating the ratio of the basic power supply and the basic power demand to obtain a benchmark supply-demand ratio corresponding to the time window to be predicted; the benchmark supply-demand ratio reflects the supply-demand relationship of the power market in the absence of external interference;

[0063] Step 114, conduct a comprehensive supply-demand ratio impact analysis on each influencing parameter in the supply-demand influencing parameter set to obtain a supply-demand impact rate; the specific implementation is as follows: analyze the impact of factors such as temperature and humidity on the supply-demand ratio, and calculate the impact degree through historical data and statistical methods; analyze the impact of holidays on power supply and demand, and calculate the impact degree during holidays through historical data; analyze the impact of policy changes on the supply-demand ratio, and calculate the impact degree of the policy through historical data before and after the implementation of the policy; use expert scoring method, hierarchical analysis method and other methods to assign weights to each of the above parameters; associate parameter values ​​with supply-demand impact rates through mathematical models (such as linear regression, neural network, etc.); comprehensively consider the impact of various parameters and calculate the comprehensive supply-demand impact rate;

[0064] Step 115, based on the supply and demand impact rate and the benchmark supply and demand ratio, calculate the electricity supply and demand ratio of the electricity market within the time window to be predicted; based on the benchmark supply and demand ratio obtained in step 113 and the supply and demand impact rate calculated in step 114, calculate the electricity supply and demand ratio of the electricity market within the time window to be predicted; electricity supply and demand ratio = benchmark supply and demand ratio * (1 + supply and demand impact rate).

[0065] More specifically, the transaction price forecasting methods for the power market include:

[0066] Step 121, statistics are collected on historical transaction data within a set time of the power market, transaction records with the same power supply-demand ratio as that in step 115 are screened out, and all screened transaction records are marked as analog transaction records; first, a reasonable time range is determined for retrieving transaction data from the historical database; all transaction records within this time range are traversed, and the power supply-demand ratio of each record is compared with the power supply-demand ratio predicted in step 115; if the two are the same or very close (determined according to a preset threshold), the record is marked as an analog transaction record;

[0067] Step 122, extracting features from all analog transaction records to obtain historical transaction prices and historical electricity surplus / deficit values ​​in each analog transaction record; transaction prices correspond to electricity surplus / deficit values ​​one by one, and electricity surplus / deficit values ​​are the difference between power supply and power demand;

[0068] Step 123, performing correlation analysis on historical transaction electricity prices and historical electricity surplus or shortage values ​​to obtain an electricity price surplus or shortage correlation matrix; using statistical methods (such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) to analyze the correlation between transaction electricity prices and electricity surplus or shortage values; and establishing an electricity price surplus or shortage correlation matrix based on the correlation analysis results, wherein each element in the matrix represents the degree of correlation of transaction electricity prices corresponding to a specific electricity surplus or shortage value;

[0069] Step 124, performing a difference calculation on the basic power supply and basic power demand obtained in step 112 to obtain a predicted power surplus or shortage value;

[0070] Step 125, extracting the historical transaction electricity price with the strongest correlation with the predicted electricity surplus or shortage value from the electricity price surplus or shortage correlation matrix as the transaction electricity price in the power market within the time window to be predicted.

[0071] The electricity supply-demand ratio prediction method can more accurately reflect the actual supply and demand situation of the electricity market by comprehensively considering multiple influencing parameters (such as weather, holidays, policies, etc.) and making corrections based on the benchmark supply-demand ratio; the transaction electricity price prediction method uses the correlation between electricity prices and electricity surplus and shortage values ​​in historical transaction data, combined with the predicted electricity surplus and shortage values ​​to predict future transaction electricity prices, considering the impact of market supply and demand balance on electricity prices, and improving the accuracy of electricity price predictions; accurate electricity supply-demand ratio and transaction electricity price predictions provide important decision-making basis for station personnel; power generation plans can be adjusted and quotation strategies can be optimized according to the prediction results to maximize profits and reduce risks; auxiliary decision-making methods can provide scientific transaction strategy recommendations for stations based on these prediction results, helping stations to better participate in electricity market competition.

[0072] The power supply-demand ratio and transaction power price obtained from the same power market forecast are integrated to obtain the market transaction structure forecast vector of the power market.

[0073] In step 200, the specific implementation of step 200 involves predicting the power generation cost and power generation of the site in different power markets. In the prediction process, it is first necessary to understand the transaction cycle characteristics of each type of power market; for example:

[0074] The medium- and long-term electricity market: usually involves quarterly, annual or even longer-term transactions, with relatively mild price fluctuations, but is greatly affected by policy, macroeconomic and other factors;

[0075] Electricity spot market: The transaction cycle is short, which may be measured in hours or days. The price fluctuates frequently and violently, and is affected by real-time factors such as supply and demand, weather conditions, etc.

[0076] Power ancillary service market: may involve various services such as frequency regulation, peak regulation, and standby. The transaction cycles and rules vary, but are usually closely related to the immediate operating conditions of the power system.

[0077] Secondly, analyze the power generation cost structure of the station, which includes:

[0078] Fixed costs: such as equipment depreciation, employee salaries, insurance costs, etc. These costs are relatively stable in the short term and do not fluctuate significantly with changes in market trading cycles;

[0079] Variable costs: such as fuel costs (coal, natural gas, water, etc.), consumables costs in operation and maintenance costs, etc. These costs are greatly affected by market price fluctuations, especially in the spot market, where fuel prices may change rapidly with market supply and demand.

[0080] Specifically, for the prediction of power generation costs, it is necessary to establish a corresponding power generation cost prediction model based on the cycle characteristics of different markets. The specific steps are as follows:

[0081] Step 211, medium- and long-term market cost forecast: Analyze long-term contracts and forecast future fixed costs; forecast fuel costs based on historical trends in fuel prices and market supply and demand; forecast maintenance costs based on equipment maintenance plans and historical maintenance costs; seasonal factors have a significant impact on fuel costs (such as coal, natural gas, etc.), which need to be considered in the forecast;

[0082] Step 212, spot market cost forecast: forecast fuel costs based on real-time fuel prices and market supply and demand; forecast maintenance costs based on real-time equipment operating status and maintenance plans; consider the impact of emergencies on costs;

[0083] Step 213, ancillary service market cost forecast: forecast the ancillary service cost based on the type of ancillary service provided; forecast the additional cost required to provide the ancillary service based on the operating status and maintenance plan of the equipment;

[0084] Step 214, select a suitable forecasting model according to the market cycle characteristics and cost structure; for the medium and long-term market, the weighted average method or regression analysis method can be used; for the spot market and ancillary service market, machine learning algorithms such as random forests and support vector machines can be used to improve the accuracy and flexibility of the forecast; use historical data to train the forecasting model to ensure that the model can accurately capture the market cycle characteristics and cost change patterns; evaluate the forecasting performance of the model through methods such as cross-validation, and adjust the model parameters to optimize the forecasting results.

[0085] Through the above steps, the power generation cost can be systematically predicted based on the transaction characteristics of various types of power markets, providing reliable data support for the subsequent formulation of trading strategies.

[0086] In the above implementation process, data acquisition systems (such as SCADA systems) and data processing tools (such as Python and R) are used to collect, clean and process data; statistical analysis software (such as SPSS and SAN) and machine learning platforms (such as TensorFlow and PyTorch) are used to build and train prediction models; model evaluation indicators (such as RMSE and MAE) and optimization algorithms (such as genetic algorithms and particle swarm optimization) are used to evaluate and optimize the models; through the above steps and technical means, the power generation cost of the site can be predicted more accurately according to the transaction characteristics of various types of power markets, especially the characteristics of the cycle length, providing a scientific basis for subsequent trading decisions.

[0087] More specifically, power generation needs to be predicted based on the transaction characteristics of various types of power markets. Different types of power markets (such as medium- and long-term power markets, power spot markets, power ancillary service markets, etc.) have different transaction cycles and characteristics, which will affect the power generation of the site. The methods for predicting power generation based on the above characteristics include:

[0088] Step 221, data collection: collect the power generation data of the station over the past period of time, including power generation, power generation efficiency, equipment status, etc. in different time periods; collect historical transaction data of various types of power markets, including transaction electricity prices, transaction volumes, market supply-demand ratios, etc., to understand market trends and cyclical changes; for new energy power stations that rely on weather conditions (such as wind power and photovoltaics), it is also necessary to collect meteorological data, including wind speed, light intensity, temperature, etc., to evaluate the impact of weather factors on power generation;

[0089] Step 222, equipment performance analysis: Based on the equipment's rated power, operating efficiency and other parameters, evaluate the equipment's power generation capacity under different operating conditions; for the mid- to long-term market, the equipment's performance is relatively stable, and a simple linear model can be used; for the spot market and ancillary service market, the equipment's operating status may change frequently, and a more complex model needs to be used;

[0090] Step 223, meteorological factor analysis: meteorological conditions have a significant impact on power generation, especially for renewable energy (such as wind power and photovoltaics); different meteorological data analysis methods are used according to the cycle characteristics of different markets. For the medium and long-term markets, long-term trend analysis and seasonal analysis methods are used, combined with historical meteorological data and climate forecast data; for the spot market and ancillary service market, due to the short transaction cycle and rapid changes in meteorological conditions, short-term prediction models such as time series analysis and neural networks can be used;

[0091] Step 224, market demand analysis: changes in market demand will also affect power generation; different market demand forecasting methods are used according to the cyclical characteristics of different markets. For the medium and long-term markets, regression analysis, economic models and other methods are used in combination with macroeconomic data and historical power demand data; for the spot market and ancillary service market, due to the large fluctuations in market demand, real-time data analysis and machine learning algorithms such as support vector machines and random forests can be used;

[0092] Step 225, constructing a power generation forecast model: for different power markets, combining equipment performance, meteorological factors and market demand to establish a comprehensive power generation forecast model; using historical data to verify the power generation forecast model and evaluate the accuracy and reliability of the model; adjusting model parameters and optimizing model performance based on the verification results; cross-validation, rolling window validation and other methods may be used.

[0093] The power generation and cost electricity price predicted in the same power market are integrated to obtain the site transaction structure prediction vector corresponding to the power market.

[0094] In step 300, calculating the unit marginal contribution of the station when participating in various types of power markets is a key link in the power auxiliary transaction decision support method. As an important indicator for measuring transaction profitability, the unit marginal contribution directly reflects the additional profit that the station can obtain per unit of power generation under specific market conditions. The following is a detailed process of the specific implementation of step 300:

[0095] The market transaction structure prediction vector of each type of power market obtained from step 100 includes the transaction electricity price and the market electricity supply-demand ratio; the site transaction structure prediction vector of the site obtained from step 200 when participating in each type of power market includes the cost electricity price and power generation; the transaction electricity price is obtained from the market transaction structure prediction vector, and the cost electricity price is obtained from the site transaction structure prediction vector; for each type of power market, the unit marginal contribution is calculated using the formula: unit marginal contribution = transaction electricity price - cost electricity price; the unit marginal contribution reflects the profit that can be brought by each unit of power generation.

[0096] In step 400, the key task of implementing the power transaction auxiliary decision-making system is to select the optimal transaction strategy from the pre-built site transaction record database according to the unit marginal contribution and the market power supply and demand ratio; the site transaction record database can effectively store and manage the historical transaction data of the site, and facilitate subsequent screening and analysis; the specific structure includes the transaction unique identifier, market type, transaction date and time, transaction electricity price, cost electricity price, transaction electricity, supply and demand ratio and other related metadata, and the site transaction record database is as follows: Figure 4As shown, in order to improve query efficiency, you can set appropriate indexes in the database; common index fields include:

[0097] TransactionID: primary key index, used to quickly locate specific transaction records;

[0098] MarketType: used to quickly filter transaction records of a specific market type;

[0099] TransactionDate and TransactionTime: used to filter transaction records by time and date;

[0100] SupplyDemandRatio and Profit: used to filter by supply-demand ratio and profit.

[0101] To perform traversal and screening in step 400, the following operations may be performed:

[0102] Step 401, the unit marginal contribution of the station in participating in various types of power markets obtained in step 300, and the market power supply-demand ratio in the market transaction structure prediction vector obtained in step 100;

[0103] Step 402: In the station transaction record database, find the record closest to the currently predicted unit marginal contribution; find the record closest to the currently predicted market power supply and demand ratio; on the basis of satisfying the above two conditions, select the record with the highest total profit as the optimal transaction record. The SQL query traversal process is as follows: Figure 5 As shown;

[0104] Step 403, extracting the corresponding power transaction volume and transaction price from the selected optimal transaction records, and using the extracted power transaction volume and transaction price as the transaction strategy for the site to participate in the power market of this type;

[0105] Step 404, output the finalized trading strategy to the station personnel in a clear and understandable manner, including specific information such as the setting of trading electricity prices and the allocation of electricity trading volume; the station personnel perform price declaration and market operations based on the output trading strategy.

[0106] In the above steps, by utilizing the historical transaction data in the station transaction record database, step 400 can provide the station with a more scientific and accurate transaction strategy based on data analysis and pattern recognition, reduce the subjectivity and uncertainty of human judgment, and make the decision more objective and reliable; in the face of the complexity and variability of the electricity market, step 400 can dynamically adjust the transaction strategy according to the latest market forecast results and transaction records; by optimizing the setting of electricity transaction volume and transaction electricity price, step 400 helps the station to achieve reasonable allocation and efficient utilization of resources; by automatically screening and extracting the best transaction records, it simplifies the complex calculations and manual judgment links in the traditional decision-making process, which not only improves the efficiency of decision-making, but also reduces labor costs and time costs.

[0107] In a preferred embodiment, in step 402, during the traversal process, it is necessary to simultaneously consider the degree of proximity of unit marginal contribution, the degree of proximity of market power supply and demand ratio, and the level of total profit; by setting a comprehensive scoring mechanism, for example, weighted summing the degree of proximity of unit marginal contribution (such as the inverse of the absolute difference), the degree of proximity of market power supply and demand ratio (such as the inverse of the absolute difference) and the total profit, the comprehensive score of power trading of each record is obtained, and then, the record with the highest comprehensive score of power trading is selected as the optimal trading record, so that the optimal trading strategy can be effectively screened out from the historical trading records, and strong decision-making support can be provided for the site to participate in various power markets; the calculation formula of the comprehensive score of power trading is as follows:

[0108]

[0109] Among them, G i represents the comprehensive score of the power transaction of the ith power transaction record, t i represents the transaction price of the ith electricity transaction record, c i represents the cost price of the i-th electricity transaction record, g i represents the power generation of the ith power transaction record, s i represents the market power supply-demand ratio of the i-th power transaction record, p i represents the total profit of the ith electricity transaction record, represents the predicted transaction electricity price, represents the predicted cost electricity price, represents the predicted cost electricity price, represents the predicted market electricity supply-demand ratio, ∈ is a positive coefficient used to avoid division by zero error, w1, w2 and w3 are the weights of unit marginal contribution proximity, supply-demand ratio proximity and total profit respectively; f1, f2 and f3 are the standardized functions of unit marginal contribution proximity, supply-demand ratio proximity and total profit respectively;

[0110] Through the above-mentioned comprehensive scoring mechanism, the best transaction records can be quickly screened out from the station transaction record database, thereby providing strong decision-making support for the station's participation in various electricity markets. It not only considers the unit marginal contribution and the degree of closeness of the market electricity supply and demand ratio, but also combines the total profit to ensure the comprehensiveness and scientificity of the comprehensive score.

[0111] like Figure 2 , Figure 3 As shown, the embodiment of the present invention provides a power auxiliary trading decision support system. The system embodiment can be implemented by software, or by hardware or a combination of software and hardware. From the hardware level, Figure 2 As shown, it is a hardware architecture diagram of a computing device where a power auxiliary trading decision support system provided by an embodiment of the present invention is located, except Figure 2 In addition to the processor, memory, network interface, and non-volatile memory shown, the computing device in the embodiment where the system is located may also generally include other hardware, such as a forwarding chip responsible for processing messages, etc. Taking software implementation as an example, Figure 3 As shown, as a system in a logical sense, the CPU of the computing device in which it is located reads the corresponding computer program in the non-volatile memory into the memory and runs it. This embodiment provides an electric power auxiliary transaction decision support system, which includes:

[0112] The market forecasting module is used to forecast electricity prices and supply and demand for various types of electricity markets (such as medium- and long-term electricity markets, electricity spot markets, and electricity auxiliary service markets) to obtain the market transaction structure forecast vector for each type of electricity market; this vector includes the transaction electricity price and the market electricity supply and demand ratio, providing a data basis for subsequent decision-making;

[0113] The station prediction module is used to predict the power generation cost and power generation of the station according to the transaction characteristics of various types of power markets, and generate a station transaction structure prediction vector when the station participates in various types of power markets; this vector contains the cost electricity price and the expected power generation, reflecting the potential cost and capacity of the station in different markets;

[0114] The profit calculation module is used to calculate the unit marginal contribution of the station when participating in various types of power markets based on the market transaction structure prediction vector and the station transaction structure prediction vector; the unit marginal contribution is defined as the unit profit between the transaction price and the cost price, which is a key indicator to measure market profitability;

[0115] The strategy screening module is used to search the pre-built and continuously updated station transaction record database for each power market, using the unit marginal contribution and the market power supply-demand ratio as traversal conditions, and screen out the transaction records that are closest to the current unit marginal contribution and market power supply-demand ratio and have the highest historical total profit; and extract the corresponding power transaction volume from the screened transaction records, and combine the current transaction electricity price to form a specific transaction strategy for the station to participate in this type of power market;

[0116] The decision output module is used to present the trading strategy (including power trading volume and trading electricity price) generated by the strategy screening module to the station personnel in a clear and easy-to-understand manner, assisting the station personnel in making the final price declaration and market operation decision.

[0117] The various variations and specific embodiments of the aforementioned power-assisted trading decision support method are also applicable to the power-assisted trading decision support system of this embodiment. Through the aforementioned detailed description of the power-assisted trading decision support method, those skilled in the art can clearly understand the implementation method of the power-assisted trading decision support system of this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0118] An embodiment of the present invention further provides a computing device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, an electricity assisted trading decision support method in any embodiment of the present invention is implemented.

[0119] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the processor executes an electricity-assisted trading decision support method in any embodiment of the present invention.

[0120] An embodiment of the present application also provides a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium, and the processor executes the computer program, so that the computer device executes an electricity-assisted trading decision support method described in any of the above embodiments.

[0121] Specifically, a system or device equipped with a storage medium can be provided, on which software program code that implements the functions of any of the above-mentioned embodiments is stored, and a computer (or CPU or MPU) of the system or device can be enabled to read and execute the program code stored in the storage medium.

[0122] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0123] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0124] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling the operating system on the computer to complete part or all of the actual operations based on the instructions of the program code.

[0125] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or expansion module is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0126] It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical factors in the process, method, article or device including the elements.

[0127] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for supporting power auxiliary trading decisions, characterized in that: The method comprises: Performing electricity price forecasting and supply and demand forecasting for each type of electricity market, and obtaining a market transaction structure forecasting vector for each type of electricity market, wherein the market transaction structure forecasting vector includes a transaction electricity price and a market electricity supply and demand ratio; According to the transaction characteristics of various types of power markets, the power generation cost and power generation of the station are predicted, and the station transaction structure prediction vector is obtained when the station participates in various types of power markets. The station transaction structure prediction vector includes the cost electricity price and the power generation; Based on the market transaction structure prediction vector and the station transaction structure prediction vector, calculating the unit marginal contribution of the station when participating in various types of power markets; For each electricity market, the unit marginal contribution and the market electricity supply and demand ratio are used as traversal conditions, and the pre-built site transaction record database is traversed to screen out the optimal transaction records, extract the electricity transaction volume corresponding to the optimal transaction record, and use the electricity transaction volume and transaction electricity price as the trading strategy for the site to participate in this type of electricity market.

2. The power-assisted trading decision support method according to claim 1, characterized in that: Supply and demand forecasts for all types of electricity markets, including: Collecting a set of supply and demand influencing parameters within a time window to be predicted, wherein the set of supply and demand influencing parameters includes meteorological quantitative parameters, holiday quantitative parameters, and policy quantitative parameters; Taking the time window to be predicted as the extraction condition, extract the basic power supply and basic power demand from the pre-built supply-demand ratio benchmark database; Calculate the ratio of basic power supply to basic power demand to obtain a benchmark supply-demand ratio corresponding to the time window to be predicted; Conduct a comprehensive supply-demand ratio impact analysis on each influencing parameter in the supply-demand influencing parameter set to obtain the supply-demand impact rate; Based on the supply and demand impact rate and the benchmark supply and demand ratio, the market electricity supply and demand ratio of the electricity market within the time window to be predicted is calculated.

3. The power-assisted trading decision support method according to claim 2, characterized in that: Electricity price forecasts for various types of electricity markets, including: Collect statistics on historical transaction data within a set period of time in the power market, filter out transaction records with the same power supply and demand ratio in the market, and mark all the filtered transaction records as analog transaction records; Extract features from all analog transaction records to obtain historical transaction electricity prices and historical electricity surplus and shortage values ​​in each analog transaction record; Conduct correlation analysis on historical transaction electricity prices and historical electricity surplus and shortage values ​​to obtain the electricity price surplus and shortage correlation matrix; Calculate the difference between the basic power supply and the basic power demand to obtain the predicted power surplus or shortage value; The historical transaction electricity price with the strongest correlation with the predicted electricity surplus or shortage value is extracted from the electricity price surplus or shortage correlation matrix as the transaction electricity price in the power market within the time window to be predicted.

4. The power-assisted trading decision support method according to claim 3, characterized in that: Using unit marginal contribution and market power supply-demand ratio as traversal conditions, traverse the pre-built site transaction record database, including: Obtain the unit marginal contribution of the station when participating in various types of power markets and the market power supply-demand ratio in the market transaction structure prediction vector; In the station transaction record database, find the record closest to the currently predicted unit marginal contribution; find the record closest to the currently predicted market power supply and demand ratio; on the basis of satisfying the two conditions of unit marginal contribution and market power supply and demand ratio, select the record with the highest total profit as the optimal transaction record; The corresponding power transaction volume and transaction price are extracted from the selected optimal transaction records, and the extracted power transaction volume and transaction price are used as the trading strategy for the site to participate in this type of power market.

5. The power-assisted trading decision support method according to claim 3, characterized in that: Using the preset comprehensive evaluation mechanism, the transaction records in the station transaction record database are scored respectively, and the best transaction record is determined based on the scoring results, including: For each transaction record, calculate the proximity of the unit marginal contribution; calculate the proximity of the market power supply and demand ratio; extract the total profit of the transaction record; Normalize the proximity of unit marginal contribution, the proximity of market power supply-demand ratio, and total profit; The standardized unit marginal contribution proximity, the market electricity supply-demand ratio proximity and the total profit are weighted to obtain a comprehensive score for electricity trading.

6. The method according to claim 5, characterized in that The comprehensive score of power trading is calculated by the following formula: Among them, G i represents the comprehensive score of the power transaction of the ith power transaction record, t i represents the transaction price of the ith electricity transaction record, c i represents the cost price of the i-th electricity transaction record, g i represents the power generation of the ith power transaction record, s i represents the market power supply-demand ratio of the i-th power transaction record, p i represents the total profit of the ith electricity transaction record, represents the predicted transaction electricity price, represents the predicted cost electricity price, represents the predicted cost electricity price, represents the predicted market electricity supply-demand ratio, ∈ is a positive coefficient used to avoid division by zero error, w1, w2 and w3 are the weights of unit marginal contribution proximity, supply-demand ratio proximity and total profit respectively; f1, f2 and f3 are the standardized functions of unit marginal contribution proximity, supply-demand ratio proximity and total profit respectively.

7. An electric power auxiliary transaction decision support system, characterized in that: The system comprises: The market prediction module is used to predict the electricity price and supply and demand of various types of electricity markets to obtain the market transaction structure prediction vector of each type of electricity market; the market transaction structure prediction vector includes the transaction electricity price and the market electricity supply and demand ratio; The station prediction module is used to predict the power generation cost and power generation of the station according to the transaction characteristics of various types of power markets, and generate the station transaction structure prediction vector when the station participates in various types of power markets; the station transaction structure prediction vector includes the cost electricity price and the expected power generation; The profit calculation module is used to calculate the unit marginal contribution of the station when participating in various types of power markets based on the market transaction structure prediction vector and the station transaction structure prediction vector; The strategy screening module is used to traverse the pre-built station transaction record database for each power market, using the unit marginal contribution and the market power supply and demand ratio as traversal conditions, and screen out the optimal transaction record; and extract the corresponding power transaction volume from the optimal transaction record, and combine it with the current transaction electricity price to form a trading strategy for the station to participate in this type of power market; The decision output module is used to display the trading strategies generated by the strategy screening module to the station personnel, and assist the station personnel in making price declarations and market operation decisions.

8. A computing device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 6.

Images