Cooperative transaction strategy generation method for multi-market joint optimization modeling of thermal power enterprises

Abstract

The application discloses a kind of collaborative transaction strategy generation methods of thermal power enterprise multi-market joint optimization modeling, it is related to electric power market optimization trading technical field, method includes: the preprocessing of multiple source data;Price prediction model is constructed;Scenario analysis model is constructed to generate price scenario and simulate transaction result;Multiple market declaration power, output, price are used as joint decision variable, unit output, offer, compliance deviation are used as constraint condition, and optimization model with single day comprehensive income maximization as target is constructed;Solving model obtains multi-market collaborative transaction strategy.The application can improve the overall income of thermal power enterprise under the condition of multi-market parallel operation and the strategy robustness.

Description

Technical Field

[0001] This invention relates to the field of power market optimization trading technology, and in particular to a collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises. Background Technology

[0002] In a market-based electricity market, thermal power plants typically need to participate in multiple trading categories simultaneously, including long-term, medium-term, and short-term markets, as well as day-ahead markets. They must comprehensively consider long-term contracted volume, short-term market volume adjustment strategies, and day-ahead market bidding strategies within the transaction submission timeline to maximize overall profitability. Under multi-market parallel operation, various markets are closely linked in terms of contracted volume, deviation assessment, and revenue structure. Thermal power unit trading decisions are characterized by multiple variables, constraints, and cross-market coupling. Currently, thermal power plants primarily adopt a market-specific, phased approach to trading decisions under multi-market conditions: long-term contracted volume is often broken down based on annual or monthly plans; short-term market submissions and adjustment ratios rely on manual experience or fixed-ratio rules; and day-ahead market submission strategies are usually based on unit marginal costs, empirical rules, or single-price forecasts. While these methods can meet basic trading needs to some extent, they struggle to coordinate different trading behaviors under multi-market parallel operation, resulting in limited overall revenue optimization capabilities.

[0003] Current power generation companies' trading decision-making methods, while making some progress in single-market optimization, price forecasting, or local decision support in a multi-market parallel electricity market environment, still have significant shortcomings in terms of overall comprehensiveness, robustness, and engineering feasibility, making it difficult to meet the actual decision-making needs of thermal power companies in complex trading environments. Existing technologies generally adopt independent decision-making methods based on different markets and stages, making it difficult to characterize the inherent coupling relationships between multi-market trading behaviors. This results in insufficient overall return optimization capabilities, easily achieving locally optimal results in one market while sacrificing overall trading returns. Existing methods are mostly based on single predicted prices or deterministic assumptions for modeling, lacking a systematic characterization of the impact of price uncertainty. When actual market prices deviate from predicted values, trading strategies generated based on deterministic assumptions often fail to maintain stable bidding performance and return levels, leading to insufficient adaptability and reliability of the strategies in actual operation. Some multi-market optimization methods have high computational complexity, involving high-dimensional decision variables and numerous scenario calculations, making it difficult to complete calculations and output clear bidding schemes within the actual trading declaration window, thus limiting their application value in the actual trading decision-making of power generation companies.

[0004] For the reasons mentioned above, existing technologies are insufficient to provide thermal power companies with a collaborative transaction decision-making method that is comprehensive, robust, and can be executed within the application period, under conditions of multiple markets operating in parallel, significant price uncertainty, and strict transaction timing requirements. Summary of the Invention

[0005] The technical problem this invention aims to solve is to address the shortcomings of existing technologies, specifically the collaborative trading decision-making problem faced by thermal power plants operating in multiple markets. This problem stems from insufficient overall revenue optimization capabilities due to independent decision-making in each market, poor strategy robustness caused by the assumption of a single price forecast, and the difficulty in meeting the time limit requirements for application by complex models. Specifically, this invention provides a collaborative trading strategy generation method for thermal power plants using multi-market joint optimization modeling, as detailed below: 1) In a first aspect, the present invention provides a method for generating collaborative trading strategies for multi-market joint optimization modeling of thermal power enterprises, the specific technical solution of which is as follows: S1 performs unified preprocessing on environmental data, market data, and transaction entity status data to obtain a structured dataset; S2, Based on the structured dataset, construct a multi-market price prediction model for the target trading day; perform price prediction based on the multi-market price prediction model to obtain the price prediction result; S3. Based on the price prediction results, construct a clearing simulation model based on price scenario analysis; based on the clearing simulation model, perform profit prediction to obtain multiple price scenarios and their corresponding simulated transaction results; S4 uses short-term, medium-term and long-term market bid electricity, day-ahead market segmented bid output, and day-ahead market segmented bid price as joint decision variables, and thermal power unit output constraints, price constraints, and performance deviation constraints as constraints. Based on multiple price scenarios and their corresponding simulated transaction results, the objective function for optimizing the multi-market joint daily comprehensive revenue under multiple price scenarios is determined, resulting in a multi-market joint optimization model with the goal of maximizing daily comprehensive revenue. S5, Solve the multi-market joint optimization model to obtain the multi-market collaborative trading strategy for the target trading day.

[0006] The beneficial effects of the collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises provided by this invention are as follows: By constructing a multi-market joint optimization model using short-term, medium-term, and long-term market electricity declarations, day-ahead market segmented power output declarations, and declaration prices as joint decision variables, collaborative decision-making for multi-market trading behaviors is achieved. This avoids the overall revenue loss problem caused by independent optimization in each market and improves the overall revenue optimization capability under the condition of parallel operation of multiple markets. By generating multiple price scenarios and their corresponding simulated transaction results based on price scenario analysis, and incorporating the simulated transaction results of each scenario into the daily comprehensive revenue optimization objective function, the same declaration strategy maintains an overall revenue advantage under all price scenarios, effectively characterizing the impact of price uncertainty and enhancing the robustness of the trading strategy in a price volatility environment. The constructed multi-market joint optimization model meets the solution complexity requirements and can calculate the multi-market collaborative trading strategy for the target trading day within the actual trading declaration time limit, and output clear short-term, medium-term, and long-term market electricity declaration schemes and day-ahead market segmented price-output declaration curves, thus achieving engineering feasibility.

[0007] Based on the above solution, the present invention can be further improved as follows.

[0008] Furthermore, the environmental data includes: meteorological data; The market data includes: long-term medium- and long-term contract electricity volume data, long-term medium- and long-term contract electricity price data, short-term medium- and long-term market historical transaction price data, day-ahead market historical clearing price data, day-ahead market historical transaction electricity volume data, historical system load data, and historical renewable energy output data. The transaction entity status data includes: historical thermal power unit operating data and historical fuel cost data; The unified preprocessing includes: abnormal data removal, time scale alignment, and unit parameter processing; The unit parameter processing includes: extracting the minimum output parameters, maximum output parameters, and ramp rate parameters of the thermal power units from the historical thermal power unit operation data, and extracting the upper and lower limits of market prices from the market data.

[0009] Furthermore, the multi-market price forecasting model includes a day-ahead market price forecasting model and short-term, medium-term, and long-term market price forecasting models; The day-ahead market price forecasting model is based on the day-ahead market historical clearing price data, day-ahead market historical transaction volume data, historical system load data, and historical renewable energy output data, and outputs the expected value of the forecasted electricity price and uncertainty parameters for each period of the target trading day. The short-term, medium-term, and long-term market price forecasting model is based on historical transaction price data of the short-term, medium-term, and long-term markets and the state of market supply and demand, and outputs the predicted transaction price of the short-term, medium-term, and long-term markets on the target trading day.

[0010] Furthermore, multiple price scenarios are generated based on the expected value of the predicted electricity price and uncertainty parameters in the price forecast results; The simulated transaction results for each price scenario are determined based on the simulated transaction volume under that price scenario; the simulated transaction volume includes the winning bid volume in the daytime market under that price scenario and the transaction volume in the short-term and medium-term markets under that price scenario.

[0011] 2) In a second aspect, the present invention also provides a collaborative trading strategy generation system for multi-market joint optimization modeling of thermal power enterprises. The specific technical solution is as follows, including: a data processing module, a price prediction module, a scenario simulation module, a model building module, and a strategy solving module; The data processing module is used to perform unified preprocessing on environmental data, market data, and transaction entity status data to obtain a structured dataset. The price prediction module is used to construct a multi-market price prediction model for the target trading day based on the structured dataset; and to perform price prediction based on the multi-market price prediction model to obtain the price prediction result. The scenario simulation module is used to construct a clearing simulation model based on price scenario analysis based on the price prediction results; and to perform profit prediction based on the clearing simulation model to obtain multiple price scenarios and their corresponding simulated transaction results. The model building module is used to take the short-term, medium- and long-term market declared electricity volume, day-ahead market segmented declared output, and day-ahead market segmented declared price as joint decision variables, and take thermal power unit output constraints, price constraints, and performance deviation constraints as constraints. Based on multiple price scenarios and their corresponding simulated transaction results, the objective function for optimizing the multi-market joint daily comprehensive revenue under multiple price scenarios is determined, resulting in a multi-market joint optimization model with the goal of maximizing daily comprehensive revenue. The strategy solving module is used to solve the multi-market joint optimization model to obtain the multi-market collaborative trading strategy for the target trading day.

[0012] Based on the above solution, the present invention can be further improved as follows.

[0013] Furthermore, the environmental data includes: meteorological data; The market data includes: long-term medium- and long-term contract electricity volume data, long-term medium- and long-term contract electricity price data, short-term medium- and long-term market historical transaction price data, day-ahead market historical clearing price data, day-ahead market historical transaction electricity volume data, historical system load data, and historical renewable energy output data. The transaction entity status data includes: historical thermal power unit operating data and historical fuel cost data; The unified preprocessing includes: abnormal data removal, time scale alignment, and unit parameter processing; The unit parameter processing includes: extracting the minimum output parameters, maximum output parameters, and ramp rate parameters of the thermal power units from the historical thermal power unit operation data, and extracting the upper and lower limits of market prices from the market data.

[0014] Furthermore, the multi-market price forecasting model includes a day-ahead market price forecasting model and short-term, medium-term, and long-term market price forecasting models; The day-ahead market price forecasting model is based on the day-ahead market historical clearing price data, day-ahead market historical transaction volume data, historical system load data, and historical renewable energy output data, and outputs the expected value of the forecasted electricity price and uncertainty parameters for each period of the target trading day. The short-term, medium-term, and long-term market price forecasting model is based on historical transaction price data of the short-term, medium-term, and long-term markets and the state of market supply and demand, and outputs the predicted transaction price of the short-term, medium-term, and long-term markets on the target trading day.

[0015] Furthermore, multiple price scenarios are generated based on the expected value of the predicted electricity price and uncertainty parameters in the price forecast results; The simulated transaction results for each price scenario are determined based on the simulated transaction volume under that price scenario; the simulated transaction volume includes the winning bid volume in the daytime market under that price scenario and the transaction volume in the short-term and medium-term markets under that price scenario.

[0016] 3) In a third aspect, the present invention also provides a computer device, the computer device including a processor coupled to a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor to enable the computer device to implement any of the above methods.

[0017] 4) In a fourth aspect, the present invention also provides a computer-readable storage medium storing at least one computer program, which is loaded and executed by a processor to enable a computer to implement any of the above methods.

[0018] It should be noted that the beneficial effects of the technical solutions of the second to fourth aspects of the present invention and their corresponding possible implementations can be found in the above description of the technical effects of the first aspect and its corresponding possible implementations, and will not be repeated here. Attached Figure Description

[0019] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1This is a flowchart illustrating the steps of a collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises according to an embodiment of the present invention. Figure 2 This is a schematic diagram of a computer device structure according to an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0021] like Figure 1 As shown in the figure, a collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises according to an embodiment of the present invention includes the following steps: S1 performs unified preprocessing on environmental data, market data, and transaction entity status data to obtain a structured dataset; S2, Based on a structured dataset, construct a multi-market price prediction model for the target trading day; perform price prediction based on the multi-market price prediction model to obtain the price prediction results; S3. Based on the price forecast results, construct a clearing simulation model based on price scenario analysis; based on the clearing simulation model, perform profit forecasting to obtain multiple price scenarios and their corresponding simulated transaction results; S4 uses short-term, medium-term and long-term market bid electricity, day-ahead market segmented bid output, and day-ahead market segmented bid price as joint decision variables, and thermal power unit output constraints, price constraints, and performance deviation constraints as constraints. Based on multiple price scenarios and their corresponding simulated transaction results, the objective function for optimizing the multi-market joint daily comprehensive revenue under multiple price scenarios is determined, resulting in a multi-market joint optimization model with the goal of maximizing daily comprehensive revenue. S5 solves the multi-market joint optimization model to obtain the multi-market collaborative trading strategy for the target trading day.

[0022] The beneficial effects of the collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises provided by this invention are as follows: By constructing a multi-market joint optimization model using short-term, medium-term, and long-term market electricity declarations, day-ahead market segmented power output declarations, and declaration prices as joint decision variables, collaborative decision-making for multi-market trading behaviors is achieved. This avoids the overall revenue loss problem caused by independent optimization in each market and improves the overall revenue optimization capability under the condition of parallel operation of multiple markets. By generating multiple price scenarios and their corresponding simulated transaction results based on price scenario analysis, and incorporating the simulated transaction results of each scenario into the daily comprehensive revenue optimization objective function, the same declaration strategy maintains an overall revenue advantage under all price scenarios, effectively characterizing the impact of price uncertainty and enhancing the robustness of the trading strategy in a price volatility environment. The constructed multi-market joint optimization model meets the solution complexity requirements and can calculate the multi-market collaborative trading strategy for the target trading day within the actual trading declaration time limit, and output clear short-term, medium-term, and long-term market electricity declaration schemes and day-ahead market segmented price-output declaration curves, thus achieving engineering feasibility.

[0023] It should be noted that, for ease of understanding, the technical terms used in this solution will be explained one by one, and will not be repeated hereafter: The day-ahead market refers to the electricity spot market where market participants submit their declaration information to the trading center one day before the trading date, and the market determines the winning generating units and output plans through a clearing algorithm.

[0024] Long-term market: refers to a medium- to long-term power trading market with a time span of one year or one month.

[0025] Short-term and medium-term markets: These refer to short-term and medium-term electricity trading markets with a time span of several days, allowing power generation companies to flexibly adjust the electricity volume of medium- and long-term contracts before the target trading date.

[0026] Environmental data refers to external environmental information that affects the operation of the power system and electricity market transactions. In this scheme, environmental data includes meteorological data, which is used to support the construction of multi-market price forecasting models. Meteorological data refers to historical observation data reflecting the state of the atmospheric environment, including but not limited to parameters such as temperature, humidity, and wind speed.

[0027] Market data refers to historical operational and contractual data directly related to electricity market transactions. In this scheme, market data includes long-term contract electricity volume data, long-term contract electricity price data, short-term historical transaction price data, day-ahead market historical clearing price data, day-ahead market historical transaction volume data, historical system load data, and historical renewable energy output data. Market price upper and lower limits can be extracted from this data for the construction and training of multi-market price prediction models, constraint setting, and revenue calculation.

[0028] Transaction entity status data: refers to historical data reflecting the operating characteristics of thermal power units and fuel costs. In this scheme, the transaction entity status data includes historical thermal power unit operating data and historical fuel cost data. Among them, the historical thermal power unit operating data includes the minimum output parameters, maximum output parameters, and ramp-up rate parameters of the thermal power units, which are used to construct output constraints for thermal power units and calculate power generation costs.

[0029] Unified preprocessing refers to the standardized processing of raw data from different sources, with varying time scales and formats, involving cleaning, alignment, and organization. In this solution, unified preprocessing includes outlier removal, time scale alignment, and unit parameter organization, used to eliminate data noise, unify the time base, and extract key parameters.

[0030] Structured datasets refer to data sets that have undergone unified preprocessing and are organized according to a unified time index and standard format. In this solution, the structured dataset includes processed environmental data, market data, and trading entity status data, serving as the input basis for the multi-market price prediction model.

[0031] Target trading day: refers to the specific date on which thermal power companies submit market transaction declarations. In this scheme, the target trading day is the time frame for declarations in the short-term, medium-term, and long-term markets, as well as the day-ahead market. This scheme is used to generate the optimal declaration strategy applicable to this target trading day.

[0032] Multi-market price forecasting model: refers to a mathematical model built based on structured datasets to predict electricity market prices on a target trading day. In this scheme, the multi-market price forecasting model includes a day-ahead market price forecasting model and short-term and medium-term market price forecasting models, which respectively predict the expected value of spot market electricity prices on the target trading day and short-term and medium-term market transaction prices, and output the price forecast results.

[0033] Price forecast results: These refer to the forecast results output by the multi-market price forecasting model. In this scheme, the price forecast results include the expected value of the predicted electricity price for each period of the target trading day, as well as uncertainty parameters, and also the predicted transaction prices of the short-term, medium-term, and long-term markets on the target trading day.

[0034] Price scenario analysis: This refers to an analytical method that characterizes the uncertainty of price predictions by constructing multiple possible price realization scenarios based on the predicted electricity price. In this approach, price scenario analysis generates multiple price scenarios based on the expected value of the predicted electricity price and uncertainty parameters, using random sampling or interval discretization to evaluate the performance of trading strategies under different electricity price environments.

[0035] Clearing simulation model: This refers to a computational model that simulates market transaction clearing results based on price scenarios and bidding strategies. In this solution, the clearing simulation model is constructed based on price scenario analysis. It determines the winning bid status of day-ahead market segmented quotations and the transaction volume in the short-term, medium-term, and long-term markets under each price scenario through clearing rules, and outputs simulated transaction results corresponding to multiple price scenarios.

[0036] Price Scenario: Refers to a possible electricity price realization state within the target trading day. In this plan, each price scenario includes a specific electricity price level for each time period of the day-ahead market and a specific transaction price for the short-term, medium-term, and long-term markets.

[0037] Simulated transaction results: These refer to the daily comprehensive return of the trading strategy under different price scenarios. In this scheme, the simulated transaction results are calculated by substituting the simulated transaction volume under each price scenario into the daily return model, reflecting the return fluctuation characteristics of the strategy under price uncertainty. The daily return model refers to the mathematical model used to calculate the comprehensive return of thermal power units within the target trading day.

[0038] Short-term and medium-term market declared electricity volume: refers to the adjusted value of electricity volume declared by thermal power enterprises in the short-term and medium-term markets. In this scheme, the short-term and medium-term market declared electricity volume is one of the joint decision variables and has an adjustment relationship with the long-term and medium-term contracted electricity volume, which is used to achieve flexible adjustment of the medium- and long-term contracted electricity volume.

[0039] Day-ahead market segmented power output: refers to the power output level corresponding to each segment of the price submitted by thermal power units in the day-ahead market. In this scheme, day-ahead market segmented power output is one of the joint decision variables, which must meet the power output constraints of thermal power units and constitute the power output component of the day-ahead market segmented price-power output declaration curve.

[0040] Day-ahead market segmented bid prices: These refer to the price levels corresponding to each bid segment submitted by thermal power units in the day-ahead market. In this scheme, the day-ahead market segmented bid prices are one of the joint decision variables, and must meet the bidding constraints, constituting a price component of the day-ahead market segmented price-output bid curve.

[0041] Joint decision variables: These refer to multiple decision variables that need to be solved simultaneously in a multi-market joint optimization model. In this scheme, the joint decision variables include the electricity volume submitted in the short-term, medium-term, and long-term markets, the power output submitted in the day-ahead market segments, and the price submitted in the day-ahead market segments. These three variables are determined collaboratively under the same objective function to achieve joint optimization of multi-market trading behavior.

[0042] Thermal power unit output constraints: These refer to the technical limitations that thermal power units must meet during physical operation. In this plan, thermal power unit output constraints include minimum output limits, maximum output limits, and ramp-up rate limits, derived from the minimum output parameters, maximum output parameters, and ramp-up rate parameters, to ensure that the declared output is within the physically feasible range of the unit.

[0043] Bidding constraints refer to the restrictions imposed by electricity market rules on bidding behavior. In this scheme, bidding constraints include bidding monotonicity constraints and price upper and lower limit constraints. The bidding monotonicity constraint requires that the segmented bidding price increases with the declared output, while the price upper and lower limit constraints require that the bidding price be within the preset market price upper and lower limits, which are extracted from market data.

[0044] Performance deviation constraint: refers to the restriction on the allowable deviation between the electricity volume of medium- and long-term contracts and the electricity volume traded in the day-ahead market. In this scheme, the performance deviation constraint requires that the deviation between the electricity volume traded in the day-ahead market and the electricity volume of long-term medium- and long-term contracts be within the allowable range, and is derived from the electricity volume data of long-term medium- and long-term contracts.

[0045] Constraints: These refer to the equality or inequality restrictions that the decision variables in the optimization model must satisfy. In this scheme, the constraints consist of thermal power unit output constraints, pricing constraints, and performance deviation constraints, which are used to ensure that the generated trading strategy meets the physical operating conditions of the units, market rule requirements, and contract performance accuracy requirements.

[0046] The multi-market joint daily comprehensive return optimization objective function refers to a mathematical expression aimed at maximizing the daily comprehensive return under multiple price scenarios. In this scheme, the multi-market joint daily comprehensive return optimization objective function is constructed by weighting and summing the daily comprehensive returns of simulated transactions under each price scenario according to preset weights. The daily comprehensive return of simulated transactions under a single price scenario includes long-term medium- and long-term contract returns, short-term medium- and long-term market adjustment returns, day-ahead market returns, power generation costs, and deviation assessment costs.

[0047] Maximizing daily comprehensive revenue: This refers to the optimization objective of maximizing the comprehensive revenue obtained by thermal power units within a target trading day. In this scheme, maximizing daily comprehensive revenue is achieved by solving a multi-market joint optimization model, avoiding the local optimum problem caused by independent decision-making in different markets, and achieving global revenue optimization under multi-market conditions.

[0048] Multi-market joint optimization model: This refers to a mathematical optimization model that uniformly considers the trading behaviors of multiple electricity markets. In this scheme, the multi-market joint optimization model is constructed using a mixed-integer linear programming method. It takes the short-term, medium-term, and long-term market bid electricity volume, day-ahead market segmented bid output, and day-ahead market segmented bid price as joint decision variables, and thermal power unit output constraints, bidding constraints, and performance deviation constraints as constraints. The optimization objective is to maximize the comprehensive daily revenue, which is used to generate the optimal collaborative trading strategy.

[0049] Multi-market coordinated trading strategy: refers to a trading application scheme that can be coordinated and executed across multiple electricity markets. In this scheme, the multi-market coordinated trading strategy includes the electricity application scheme for short-term, medium-term and long-term markets, as well as the segmented price-output application curve for the day-ahead market. The segmented price-output application curve is a monotonically increasing segmented price curve applicable to all periods of the target trading day.

[0050] Long-term contract electricity volume data: refers to the electricity volume agreed upon in the annual or monthly medium-term market by thermal power enterprises, broken down to the target transaction date. It is the benchmark data for performance deviation constraints and is used to calculate deviation assessment costs.

[0051] Long-term contract electricity price data: refers to the price data corresponding to the electricity volume of long-term contracts, which is used to calculate the revenue of long-term contracts.

[0052] Historical transaction price data for short-term, medium-term, and long-term markets: refers to the actual transaction price records formed in the short-term, medium-term, and long-term markets during historical periods, used to predict the transaction price of the short-term, medium-term, and long-term markets on the target trading day.

[0053] Previous-day market historical cleared electricity price data: refers to the electricity price records formed by the market clearing mechanism in the previous-day market during historical trading periods. It is used to train the previous-day market price prediction model and output the expected value of the predicted electricity price.

[0054] Previous-day market historical transaction volume data: refers to the actual transaction volume records of the previous-day market during historical trading periods. It serves as auxiliary input data for the previous-day market price forecasting model to improve the accuracy of electricity price forecasting.

[0055] Historical system load data refers to the total electricity load record of the power system during historical periods. In this scheme, historical system load data serves as one of the quantitative representations of market supply and demand, and is used as input for short-term, medium-term, and long-term market price forecasting models, as well as day-ahead market price forecasting models.

[0056] Historical renewable energy output data: refers to the actual power generation records of renewable energy generator units such as wind power and solar power in historical periods.

[0057] Historical fuel cost data: refers to the fuel cost records of thermal power units during historical periods, used to calculate the marginal cost of the unit and as the basis for calculating the power generation cost.

[0058] Outlier removal: refers to the process of identifying and removing erroneous values, missing values, and noisy data through data cleaning methods. Statistical detection methods or rule-based judgment methods can be used to ensure data quality.

[0059] Time scale alignment: This refers to unifying data from different sources and with different time resolutions to the same time base and time granularity. In this solution, variables such as long-term contract electricity volume, short-term market adjustment electricity volume, and day-ahead market transaction electricity volume are aligned according to a unified time index to ensure consistency of each variable within the same time period.

[0060] Unit parameter processing: This refers to the process of extracting the minimum output parameters, maximum output parameters, and ramp rate parameters of thermal power units from historical thermal power unit operating data, and extracting the upper and lower limits of market prices from market data, so as to provide parameter support for the construction of constraints.

[0061] Minimum output parameters of thermal power units: refers to the minimum active power limit that a thermal power unit is allowed to output under stable operating conditions, which serves as the lower limit for the output constraints of thermal power units.

[0062] Maximum output parameters of thermal power units: refers to the maximum active power limit that thermal power units are allowed to output under rated operating conditions, which serves as the upper limit basis for the output constraints of thermal power units.

[0063] The ramp rate parameter of thermal power units refers to the maximum allowable change in output of a thermal power unit per unit time, which is used to construct the ramp limit in the output constraint of thermal power units.

[0064] Market price upper and lower limits: refers to the minimum and maximum range of fluctuations allowed in the bid price as stipulated by the electricity market rules. It is derived from market rule information in market data and serves as the parameter basis for the upper and lower limits of price constraints in the bidding constraints.

[0065] Uncertainty parameter: This refers to a quantitative indicator describing the extent to which the predicted electricity price may deviate from the expected value. In this scheme, the uncertainty parameter is output by the day-ahead market price forecasting model and is used to characterize the distribution characteristics of the forecast error. It is a key parameter for generating multiple price scenarios.

[0066] Market supply and demand status: refers to the relative relationship between supply capacity and demand level in the electricity market. In this scheme, market supply and demand status is comprehensively characterized by historical system load data, historical renewable energy output data, and historical thermal power unit operation data, serving as input features for short-term, medium-term, and long-term market price forecasting models.

[0067] Transaction price forecast: refers to the expected transaction price on the target trading day output by the short-term, medium-term and long-term market price forecasting model, which is used to calculate the short-term, medium-term and long-term market adjustment returns.

[0068] Expected electricity price forecast: This refers to the most likely electricity price level for each time period on the target trading day, as output by the day-ahead market price forecasting model. In this scheme, the expected electricity price forecast is the point forecast result, which is the center value for constructing the price scenario.

[0069] Simulated trading volume: refers to the market trading results calculated by the clearing simulation model based on price scenarios and bidding strategies. In this scheme, the simulated trading volume includes the winning bid volume in the day-ahead market under various price scenarios and the trading volume in the short-term, medium-term, and long-term markets under various price scenarios, which are then substituted into the daily profit model to obtain the simulated trading results.

[0070] Winning bid volume: refers to the volume of electricity traded in the day-ahead market where the bid price is no higher than the predicted electricity price under the corresponding price scenario. In this scheme, the winning bid volume is determined by comparing the bid prices in the day-ahead market segments with the predicted electricity price in the price scenario, serving as the basis for calculating day-ahead market revenue.

[0071] Transaction volume: refers to the adjusted value of the electricity volume declared and actually traded by thermal power plants in the short-term, medium-term, and long-term markets. In this scheme, the transaction volume is equal to the declared electricity volume in the short-term, medium-term, and long-term markets by default, and is used to calculate the adjustment revenue in the short-term, medium-term, and long-term markets.

[0072] In another embodiment of this solution, S1 is specifically implemented as follows: Meteorological data is collected as environmental data; long-term contract electricity volume data, long-term contract electricity price data, short-term and long-term market historical transaction price data, day-ahead market historical clearing price data, day-ahead market historical transaction volume data, historical system load data, and historical renewable energy output data are collected as market data; historical thermal power unit operation data and historical fuel cost data are collected as trading entity status data.

[0073] Outlier removal is performed on the data using statistical detection methods or rule-based judgment methods to identify and remove outliers, missing values, and noisy data in environmental, market, and trading entity status data. Time scale alignment is performed on the data, aligning environmental, market, and trading entity status data from different sources and with different time resolutions to the same time period and granularity using a unified time index. Unit parameter processing is performed on the data, extracting minimum output parameters, maximum output parameters, and ramp-up rate parameters of thermal power units from historical thermal power unit operating data, and extracting the upper and lower limits of market prices from market data.

[0074] After the above processing, we obtain processed environmental data, market data, and transaction entity status data. All processed data form a structured dataset.

[0075] In another embodiment of this solution, S2 is specifically implemented as follows: Based on structured datasets, two sub-models were constructed: a day-ahead market price forecasting model and a short-term and medium-to-long-term market price forecasting model.

[0076] The process of constructing the day-ahead market price forecasting model is as follows: using historical clearing electricity price data, historical transaction volume data, historical system load data, and historical renewable energy output data as input features, a statistical model or machine learning method is used for training, and the expected value of the forecasted electricity price and uncertainty parameters for each period of the target trading day are output. The expected value of the forecasted electricity price represents the most likely electricity price level, and the uncertainty parameters represent the distribution range of the forecast error.

[0077] The process of constructing a short-term, medium-term, and long-term market price forecasting model is as follows: using historical transaction price data of the short-term, medium-term, and long-term markets as the basic input, and combining it with the market supply and demand status comprehensively represented by historical system load data, historical new energy output data, and historical thermal power unit operation data, the model is trained using time series analysis or machine learning methods, and outputs the predicted transaction price value of the short-term, medium-term, and long-term markets for the target trading day. The predicted transaction price value is a single value or a probability distribution.

[0078] Price forecasts are made based on the day-ahead market price forecasting model and the short-term and medium-term market price forecasting model, respectively, and the price forecast results are obtained. The price forecast results include the expected value of the predicted electricity price for each period of the target trading day and the uncertainty parameters, as well as the predicted value of the transaction price in the short-term and medium-term markets on the target trading day.

[0079] In another embodiment of this solution, S3 is specifically implemented as follows: To characterize the impact of forecast errors on transaction results, a clearing simulation model based on price scenario analysis is constructed using the expected value of the forecasted electricity price and uncertainty parameters from the price forecast results. The specific process is as follows: Multiple price scenarios are determined: Samples are drawn from the probability distribution defined by the uncertainty parameters using a random sampling method, or the range of uncertainty parameters is divided into several discrete intervals using an interval discretization method. Each sample or each discrete interval corresponds to a price scenario, thus obtaining multiple price scenarios. Each price scenario includes the specific electricity price level for each time period in the day-ahead market (i.e., the predicted electricity price, denoted as...). (and specific transaction prices in the short, medium and long term markets.)

[0080] The simulated trading volume is calculated by using clearing simulation rules to determine the winning bid volume in the day-ahead market segment where the bid price is no higher than the predicted electricity price. The trading volume in the short-term, medium-term, and long-term markets is then calculated. The winning bid volume and the trading volume together constitute the simulated trading volume. Details are as follows: Determining the winning bid volume: In the day-ahead market, an integer decision variable is introduced to indicate whether the unit wins the bid for each bidding segment under different price scenarios. A value of 1 indicates that the corresponding segment has won the bid under the corresponding price scenario, and a value of 0 indicates that it has not won the bid. When the bid price for a certain segment is not higher than the predicted electricity price under the corresponding price scenario, that segment is considered to have won the bid, and its corresponding bid volume is included in the transaction volume. For example, assuming the market adopts a marginal clearing mechanism, for the segmented bids submitted by thermal power units in each time period t of the day-ahead period... If the declared price for this segment The predicted electricity price will not exceed a certain price scenario s. That is, satisfying: If the bid price is higher than the predicted electricity price under that scenario, the bid price for that segment is considered acceptable, and the corresponding bid electricity volume is included in the winning bid electricity volume of the thermal power unit. If the bid price is higher than the predicted electricity price under that scenario, the segment is not accepted. The total winning bid electricity volume of the thermal power unit in time period t is calculated by summing the winning bid electricity volumes of the corresponding accepted bid segments. This refers to the declared output of thermal power units in the j-th segment of the t-th period in the day-ahead market. This refers to the bid price of thermal power units in the j-th segment of the t-th time period in the day-ahead market.

[0081] Determining the transaction volume: In the short, medium, and long term, the market defaults to the volume declared by thermal power plants. Equal to its short-term, medium-term, and long-term market transaction volume This involves adopting a simplified mechanism where orders are executed immediately upon submission, directly treating short-term and medium-term market orders as executed volumes for subsequent calculations. Specifically, the short-term and medium-term market orders... As one of the joint decision variables, it participates in the optimization process. Its optimal value is extracted from the optimal solution after the solution is completed in step S5. In this step, the short-term, medium-term and long-term market declaration electricity is simulated using the candidate value or preset initial value provided in step S4.

[0082] Finally, the simulated transaction volume under each price scenario is substituted into the daily revenue model to calculate the daily comprehensive revenue under that price scenario. The daily revenue model, as part of constructing the multi-market joint daily comprehensive revenue optimization objective function, is described in detail in step S402.

[0083] In another embodiment of this solution, S4 is specifically implemented as follows: S401 uses short-term, medium-term, and long-term market-declared electricity volume, day-ahead market segmented power output, and day-ahead market segmented price as joint decision variables, and solves for all three types of variables simultaneously in a multi-market joint optimization model.

[0084] In the short- to medium-term market, this scheme does not precisely model the market-side transaction or clearing mechanism. Instead, it constructs a short- to medium-term market declaration adjustment strategy from the perspective of generating companies' declaration strategies. Based on the relative deviation between the short- to medium-term market forecast price and the day-ahead market forecast price, a continuous decision variable is introduced to describe the electricity declaration by thermal power companies within the target trading day, based on the difference between the long-term contract electricity volume and the day-ahead market transaction electricity volume. A continuous decision variable refers to a decision variable that can take any real value within a continuous interval in the multi-market joint optimization model. Under a certain price scenario s, the adjustment strategy for short- to medium-term market declared electricity volume is defined as follows: ; in, This represents the short-term, medium-term, and long-term market-declared electricity volume during the t-th period of the target trading day m+2 under price scenario s, and is a continuous decision variable; This represents the day-ahead market forecast price for the t-th time period of the target trading day m+2, which serves as the benchmark price in the adjustment strategy. This indicates that, under price scenario s, the medium- and long-term contracted electricity volume in the t-th period of the target trading day m+2 is derived from the decomposition value of the medium- and long-term contracted electricity volume over a long period, and serves as the adjustment benchmark. This represents the short-term to medium-term market forecast price for the target trading day m+2 at time t under price scenario s, used to compare with the previous day's market forecast price to calculate the relative deviation; K1 and K2 represent the deviation coefficient thresholds between the short-term to medium-term market price and the previous day's market forecast price, used to divide three different price deviation ranges, with K1 being the lower threshold and K2 being the upper threshold; R1 and R2 represent the adjustment coefficients for the declared electricity volume of thermal power enterprises in the short-term to medium-term market, used to proportionally scale the declared electricity volume within different price deviation ranges, with R1 corresponding to the low deviation range and R2 corresponding to the medium deviation range. In one embodiment, the short-term to medium-term market defaults to the declared electricity volume of thermal power enterprises being equal to their short-term to medium-term market transaction volume (the process of determining the transaction volume in step S3), or an approximate correction is made by introducing a transaction ratio coefficient on this basis. It represents the relative deviation rate of short-term, medium-term, and long-term market forecast prices from the current day's market forecast prices.

[0085] Explanation of the rules and logic for adjusting short-term, medium-term, and long-term market reported electricity volumes: Based on the relative deviation rate between the short-term, medium-term, and long-term market forecast prices and the day-ahead market forecast prices, the adjustment strategy is divided into three intervals. When the relative deviation rate is less than K1, between K1 and K2, or greater than or equal to K2, the corresponding reported electricity volume coefficients R1, R2, or 1.0 are used to scale the benchmark adjustment amount, thereby realizing the function of dynamically adjusting the reported electricity volume according to the degree of price deviation.

[0086] In the day-ahead market, the day-ahead market bidding strategy for thermal power units consists of a price curve (i.e., the WP price curve) composed of segmented power output and segmented price bids across multiple time periods throughout the day. Thermal power units submit a uniform, monotonically increasing segmented price curve in the day-ahead market, which applies to all time periods of the day. The power output bid constraints are: the minimum power output parameter of the thermal power unit is less than the first segment's bid output; the first segment's bid output is less than the second segment's bid output; the previous bid output is less than the next bid price; and all segmented bid outputs are less than the thermal power unit's maximum power output parameter. This is represented as follows: Constraints on reported output: ; in, These are the minimum output parameters for thermal power units. These are the maximum output parameters of the thermal power unit. The power output is subject to segmented reporting in the market before the day, and this constraint ensures that the reported power output is within the physical operating capacity of the unit. At the same time, the range of power output variation between adjacent time periods is limited by the thermal power unit ramp rate parameter.

[0087] The pricing constraints are: the lower limit of the market price must be lower than the first segment's bid price, the previous bid price must be lower than the next bid price, and all segment bid prices must be lower than the upper limit of the market price. This is expressed as follows: Price constraints as of today: ; in, as well as These are the upper and lower limits of market prices. This represents the segmented market bid prices for the previous day. It also implies a monotonicity constraint on bids, meaning that the segmented bid prices increase with the bid volume, and the upper and lower limits of the market price are derived from the market price range extracted by S1 from the market data.

[0088] The advantage of this approach is that it integrates the trading behavior of thermal power units in the short-term, medium-term, and day-ahead markets into a unified decision-making framework for modeling. It uses the electricity volume declared in the short-term, medium-term, and long-term markets, the output declared in segments in the day-ahead market, and the price declared in segments in the day-ahead market as joint decision-making variables. It also achieves the inherent coupling between different markets through thermal power unit output constraints, price constraints, and performance deviation constraints. This avoids the local optima problem caused by independent decision-making in different markets and stages in existing technologies, and achieves overall revenue optimization under multi-market conditions.

[0089] S402 constructs a daily comprehensive revenue model for thermal power units that simultaneously covers long-term, short-term, and day-ahead markets, mapping multi-market trading behaviors to a single optimization objective function. Within the target trading day, the day is divided into multiple time periods. Under any price scenario, the daily comprehensive revenue of the thermal power unit in each time period is composed of long-term contract revenue, short-term market adjustment revenue, day-ahead market revenue, power generation costs, and deviation assessment costs. Based on the above definitions, the daily comprehensive revenue of the thermal power unit under different price scenarios is obtained. Then, by uniformly aggregating the daily comprehensive revenue under each price scenario using a weighted summation method, a cross-scenario daily comprehensive revenue index is constructed, and maximizing this comprehensive revenue index is used as the optimization objective. The thermal power unit uses a multi-market joint daily comprehensive revenue optimization objective function, where the daily comprehensive revenue R is represented as follows: Where maxR represents maximizing the daily comprehensive return R, which is the optimization objective; t is the time period index, n is the total number of time periods; s is the price scenario index, and S is the total number of price scenarios; This represents the weight of the s-th price scenario, reflecting the probability of each price scenario occurring; This represents the long-term contract electricity price for the t-th time period; This represents the long-term contract electricity volume during time period t. This represents the short-term, medium-term, and long-term market transaction prices in the t-th time period under the s-th price scenario; This represents the short-term, medium-term, and long-term market transaction volume during the t-th time period under the s-th price scenario. This represents the day-ahead market transaction price for time period t; This represents the day-ahead market transaction volume in time period t under the s-th price scenario; This represents the medium- and long-term contracted electricity volume in time period t under the s-th price scenario; This represents the power generation cost of the thermal power unit in time period t; This represents the assessment cost arising from the deviation between medium- and long-term electricity volume and day-ahead market transaction volume under the s-th price scenario. This is the performance deviation constraint, which incurs assessment or penalty costs for the deviation between medium- and long-term electricity volume and day-ahead electricity volume, indirectly constraining the deviation range. At the same time, the market clearing constraint also reflects the performance deviation control requirements. The market clearing constraint indicates that the day-ahead market transaction volume should be greater than the medium- and long-term contract volume, ensuring that the deviation between the day-ahead market transaction volume and the long-term medium- and long-term performance volume is within a reasonable range, and guaranteeing that the day-ahead market has sufficient volume space to cover the medium- and long-term performance needs.

[0090] The advantage of this approach is that by introducing price scenario analysis, the uncertainty of price prediction can be characterized. Under the constraint of unified decision variables, a daily comprehensive return index spanning multiple price scenarios can be constructed. The optimization objective is to maximize the comprehensive return across these scenarios. This allows the generated trading strategy to maintain an overall return advantage across the entire set of price scenarios, rather than being optimal only for a single predicted price scenario, thereby improving the robustness of the trading strategy under price volatility conditions.

[0091] By using the aforementioned joint decision variables, constraints, and objective function, the various benefits and costs generated by the multi-market trading activities of thermal power units within the target trading day (m+2) are uniformly mapped into a single mathematical expression, forming a multi-market joint optimization model with the goal of maximizing the comprehensive daily benefit.

[0092] In another embodiment of this solution, S5 is specifically implemented as follows: A mixed-integer linear programming solver is used to solve a multi-market joint optimization model. The solution process meets the transaction submission deadline requirements and can complete the calculation before the submission deadline. A mixed-integer linear programming solver refers to a mathematical optimization computational tool used to solve mixed-integer linear programming problems, and is existing technology.

[0093] The optimal solution obtained includes the optimal short-term, medium-term, and long-term market electricity demand values, as well as the optimal day-ahead market segmented output and price values. This optimal solution is converted into an executable trading strategy format, resulting in a multi-market coordinated trading strategy for the target trading day, which guides the actual market demand of thermal power units during the target trading day. Specifically, the multi-market coordinated trading strategy for the target trading day includes the electricity demand scheme for the short-term, medium-term, and long-term markets, which can be a single value or segmented values; and the segmented price-output demand curve for the day-ahead market (i.e., the WP quotation curve). The WP quotation curve is a monotonically increasing piecewise linear function, containing multiple output demand intervals and corresponding price levels, applicable to all periods of the target trading day.

[0094] Furthermore, environmental data includes: meteorological data; Market data includes: long-term medium- and long-term contract electricity volume data, long-term medium- and long-term contract electricity price data, short-term medium- and long-term historical transaction price data, day-ahead market historical clearing price data, day-ahead market historical transaction electricity volume data, historical system load data, and historical renewable energy output data. The transaction entity status data includes: historical thermal power unit operating data and historical fuel cost data; Unified preprocessing includes: outlier removal, time scale alignment, and unit parameter processing; The unit parameter compilation includes: extracting the minimum output parameters, maximum output parameters, and ramp rate parameters of thermal power units from historical thermal power unit operating data, and extracting the upper and lower limits of market prices from market data.

[0095] Furthermore, the multi-market price forecasting model includes day-ahead market price forecasting models as well as short-term and medium-to-long-term market price forecasting models; The day-ahead market price forecasting model is based on the day-ahead market historical clearing price data, day-ahead market historical transaction volume data, historical system load data, and historical renewable energy output data. It outputs the expected value of the forecasted electricity price for each period of the target trading day and uncertainty parameters. The short-term, medium-term, and long-term market price forecasting model is based on historical transaction price data and market supply and demand status in the short-term, medium-term, and long-term markets, and outputs the predicted transaction prices of the short-term, medium-term, and long-term markets on the target trading day.

[0096] Furthermore, multiple price scenarios are generated based on the expected value of the predicted electricity price and uncertainty parameters in the price forecast results; The simulated transaction results for each price scenario are determined based on the simulated transaction volume under that price scenario; the simulated transaction volume includes the winning bid volume in the daytime market under that price scenario and the transaction volume in the short-term and medium-term markets under that price scenario.

[0097] Beneficial effects: Because thermal power units operate in multiple markets simultaneously, their trading decisions exhibit significant coupling relationships in terms of contracted electricity volume, deviation assessment, and revenue structure. Adopting an independent decision-making approach for each market can easily lead to locally optimal results in a single market, thereby compromising overall profitability. This solution constructs a multi-market joint optimization model, simultaneously considering trading behaviors in long-term, short-term, and day-ahead markets within a unified decision-making framework. This allows various trading decisions to be collaboratively determined under the same optimization objective function, thereby generating a superior trading strategy from an overall profitability perspective. Furthermore, considering the significant price fluctuations and unavoidable forecasting errors in the electricity market, this solution introduces price scenario analysis and constructs a cross-scenario expected return optimization objective function. This ensures that the same trading strategy maintains an overall profitability advantage across multiple price scenarios, effectively reducing the decision-making risk associated with a single forecast price assumption and enhancing the robustness of the trading strategy in complex market environments.

[0098] In the above embodiments, although the steps are numbered S1, S2, etc., they are only specific embodiments given by the present invention. Those skilled in the art can adjust the execution order of S1, S2, etc. according to the actual situation, and these situations are also within the protection scope of the present invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0099] Furthermore, the acquisition process of the data involved in this application follows the principles of legality, legitimacy, and necessity. Based on obtaining the explicit authorization and consent of the user, only the minimum necessary information required to achieve the purpose is collected, and data security protection obligations are fulfilled in accordance with the law.

[0100] This invention also provides a collaborative trading strategy generation system for multi-market joint optimization modeling of thermal power enterprises. The specific technical solution is as follows: a data processing module, a price prediction module, a scenario simulation module, a model building module, and a strategy solving module. The data processing module is used to perform unified preprocessing of environmental data, market data, and transaction entity status data to obtain a structured dataset; The price prediction module is used to build a multi-market price prediction model for a target trading day based on a structured dataset; and to perform price predictions based on the multi-market price prediction model to obtain the price prediction results. The scenario simulation module is used to construct a clearing simulation model based on price scenario analysis based on price forecast results; and to perform profit forecasting based on the clearing simulation model to obtain multiple price scenarios and their corresponding simulated transaction results. The model building module is used to take the short-term, medium- and long-term market bid electricity, day-ahead market segmented bid output, and day-ahead market segmented bid price as joint decision variables, and take thermal power unit output constraints, bidding constraints, and performance deviation constraints as constraints. Based on multiple price scenarios and their corresponding simulated transaction results, the objective function for optimizing the multi-market joint daily comprehensive revenue under multiple price scenarios is determined, resulting in a multi-market joint optimization model with the goal of maximizing daily comprehensive revenue. The strategy solving module is used to solve the multi-market joint optimization model to obtain the multi-market collaborative trading strategy for the target trading day.

[0101] It should be noted that the beneficial effects of the collaborative trading strategy generation system for multi-market joint optimization modeling of thermal power enterprises provided in the above embodiments are the same as those of the collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises, and will not be repeated here. Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiments, and will not be repeated here.

[0102] like Figure 2As shown, an embodiment of the present invention provides a computer device 300, which includes a processor 320 coupled to a memory 310. The memory 310 stores at least one computer program 330, which is loaded and executed by the processor 320 to enable the computer device 300 to implement any of the above-described methods. Specifically: The computer device 300 can vary considerably due to differences in configuration or performance. It may include one or more processors 320 (Central Processing Units, CPUs) and one or more memories 310. The memories 310 store at least one computer program 330, which is loaded and executed by the processors 320 to enable the computer device 300 to implement the collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises provided in the above embodiment. Of course, the computer device 300 may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input / output. It may also include other components for implementing device functions, which will not be elaborated upon here.

[0103] An embodiment of the present invention provides a computer-readable storage medium storing at least one computer program, which is loaded and executed by a processor to enable a computer to implement any of the above-described methods.

[0104] Alternatively, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0105] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform any of the above-described collaborative trading strategy generation methods for multi-market joint optimization modeling of thermal power enterprises.

[0106] It should be noted that the terms "first," "second," etc., used in the specification of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown in the figures or description.

[0107] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this disclosure can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0108] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0109] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for generating collaborative trading strategies for multi-market joint optimization modeling of thermal power enterprises, characterized in that, include: S1 performs unified preprocessing on environmental data, market data, and transaction entity status data to obtain a structured dataset; S2, Based on the structured dataset, construct a multi-market price prediction model for the target trading day; perform price prediction based on the multi-market price prediction model to obtain the price prediction result; S3. Based on the price prediction results, construct a clearing simulation model based on price scenario analysis; based on the clearing simulation model, perform profit prediction to obtain multiple price scenarios and their corresponding simulated transaction results; S4 uses short-term, medium-term and long-term market bid electricity, day-ahead market segmented bid output, and day-ahead market segmented bid price as joint decision variables, and thermal power unit output constraints, price constraints, and performance deviation constraints as constraints. Based on multiple price scenarios and their corresponding simulated transaction results, the objective function for optimizing the multi-market joint daily comprehensive revenue under multiple price scenarios is determined, resulting in a multi-market joint optimization model with the goal of maximizing daily comprehensive revenue. S5, Solve the multi-market joint optimization model to obtain the multi-market collaborative trading strategy for the target trading day.

2. The collaborative transaction strategy generation method for multi-market joint optimization modeling of thermal power enterprises according to claim 1, characterized in that, The environmental data includes: meteorological data; The market data includes: long-term medium- and long-term contract electricity volume data, long-term medium- and long-term contract electricity price data, short-term medium- and long-term market historical transaction price data, day-ahead market historical clearing price data, day-ahead market historical transaction electricity volume data, historical system load data, and historical renewable energy output data. The transaction entity status data includes: historical thermal power unit operating data and historical fuel cost data; The unified preprocessing includes: abnormal data removal, time scale alignment, and unit parameter processing; The unit parameter processing includes: extracting the minimum output parameters, maximum output parameters, and ramp rate parameters of the thermal power units from the historical thermal power unit operation data, and extracting the upper and lower limits of market prices from the market data.

3. The collaborative transaction strategy generation method for multi-market joint optimization modeling of thermal power enterprises according to claim 1, characterized in that, The multi-market price forecasting model includes a day-ahead market price forecasting model and short-term, medium-term, and long-term market price forecasting models. The day-ahead market price forecasting model is based on the day-ahead market historical clearing price data, day-ahead market historical transaction volume data, historical system load data, and historical renewable energy output data, and outputs the expected value of the forecasted electricity price and uncertainty parameters for each period of the target trading day. The short-term, medium-term, and long-term market price forecasting model is based on historical transaction price data of the short-term, medium-term, and long-term markets and the state of market supply and demand, and outputs the predicted transaction price of the short-term, medium-term, and long-term markets on the target trading day.

4. The collaborative transaction strategy generation method for multi-market joint optimization modeling of thermal power enterprises according to claim 1, characterized in that, Multiple price scenarios are generated based on the expected value of the predicted electricity price and uncertainty parameters in the price forecast results; The simulated transaction results for each price scenario are determined based on the simulated transaction volume under that price scenario; the simulated transaction volume includes the winning bid volume in the daytime market under that price scenario and the transaction volume in the short-term and medium-term markets under that price scenario.

5. A collaborative trading strategy generation system for multi-market joint optimization modeling of thermal power enterprises, characterized in that, include: The system includes a data processing module, a price prediction module, a scenario simulation module, a model building module, and a strategy solving module. The data processing module is used to perform unified preprocessing on environmental data, market data, and transaction entity status data to obtain a structured dataset. The price prediction module is used to construct a multi-market price prediction model for the target trading day based on the structured dataset; and to perform price prediction based on the multi-market price prediction model to obtain the price prediction result. The scenario simulation module is used to construct a clearing simulation model based on price scenario analysis based on the price prediction results; and to perform profit prediction based on the clearing simulation model to obtain multiple price scenarios and their corresponding simulated transaction results. The model building module is used to take the short-term, medium- and long-term market declared electricity volume, day-ahead market segmented declared output, and day-ahead market segmented declared price as joint decision variables, and take thermal power unit output constraints, price constraints, and performance deviation constraints as constraints. Based on multiple price scenarios and their corresponding simulated transaction results, the objective function for optimizing the multi-market joint daily comprehensive revenue under multiple price scenarios is determined, resulting in a multi-market joint optimization model with the goal of maximizing daily comprehensive revenue. The strategy solving module is used to solve the multi-market joint optimization model to obtain the multi-market collaborative trading strategy for the target trading day.

6. The collaborative transaction strategy generation system for multi-market joint optimization modeling of thermal power enterprises according to claim 5, characterized in that, The environmental data includes: meteorological data; The market data includes: long-term medium- and long-term contract electricity volume data, long-term medium- and long-term contract electricity price data, short-term medium- and long-term market historical transaction price data, day-ahead market historical clearing price data, day-ahead market historical transaction electricity volume data, historical system load data, and historical renewable energy output data. The transaction entity status data includes: historical thermal power unit operating data and historical fuel cost data; The unified preprocessing includes: abnormal data removal, time scale alignment, and unit parameter processing; The unit parameter processing includes: extracting the minimum output parameters, maximum output parameters, and ramp rate parameters of the thermal power units from the historical thermal power unit operation data, and extracting the upper and lower limits of market prices from the market data.

7. The collaborative transaction strategy generation system for multi-market joint optimization modeling of thermal power enterprises according to claim 5, characterized in that, The multi-market price forecasting model includes a day-ahead market price forecasting model and short-term, medium-term, and long-term market price forecasting models. The day-ahead market price forecasting model is based on the day-ahead market historical clearing price data, day-ahead market historical transaction volume data, historical system load data, and historical renewable energy output data, and outputs the expected value of the forecasted electricity price and uncertainty parameters for each period of the target trading day. The short-term, medium-term, and long-term market price forecasting model is based on historical transaction price data of the short-term, medium-term, and long-term markets and the state of market supply and demand, and outputs the predicted transaction price of the short-term, medium-term, and long-term markets on the target trading day.

8. The collaborative transaction strategy generation system for multi-market joint optimization modeling of thermal power enterprises according to claim 5, characterized in that, Multiple price scenarios are generated based on the expected value of the predicted electricity price and uncertainty parameters in the price forecast results; The simulated transaction results for each price scenario are determined based on the simulated transaction volume under that price scenario; the simulated transaction volume includes the winning bid volume in the daytime market under that price scenario and the transaction volume in the short-term and medium-term markets under that price scenario.

9. A computer device, characterized in that, The computer device includes a processor coupled to a memory, the memory storing at least one computer program, which is loaded and executed by the processor to enable the computer device to implement a collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to enable the computer to implement a collaborative trading strategy generation method for multi-market joint optimization modeling of thermal power enterprises as described in any one of claims 1 to 4.

Images