A day-ahead spot market transaction decision system based on risk scenario driving

Abstract

This invention belongs to the field of power market decision-making technology, specifically relating to a risk-scenario-driven day-ahead spot market trading decision-making system. It includes a risk scenario input module for receiving data on multiple risk scenarios from day-ahead spot market participants within future trading periods. This risk scenario data includes electricity prices, loads, generation, and corresponding scenario probabilities under different scenarios. A net trading volume calculation module calculates the net trading volume of each participant in each trading period and risk scenario based on the electricity prices, loads, generation, and corresponding scenario probabilities under each risk scenario data. A tiered bidding decision module generates segmented bid prices and segmented bid volumes for participants in the day-ahead market. This invention improves the adaptability of bidding schemes to market fluctuations, constructs a tiered bidding and volume optimization model for the day-ahead market, and provides support for market participants to conduct profitable and stable day-ahead trading decisions.

Description

Technical Field

[0001] This invention belongs to the field of power market decision-making technology, specifically relating to a day-ahead spot market trading decision-making system driven by risk scenarios. Background Technology

[0002] With the expansion of new energy installed capacity and the improvement of electricity market trading mechanisms, new energy power plants, electricity sales companies, virtual power plants, and integrated power generation and sales entities are gradually shifting from guaranteed consumption to participating in day-ahead spot market transactions. In the day-ahead market, market participants need to submit their quotations and quantity proposals for future trading periods in advance, the results of which directly affect day-ahead transaction revenue, real-time deviation settlement, and deviation assessment costs. Therefore, developing a reasonable day-ahead reporting strategy under conditions of forecast uncertainty has become a key issue in improving trading revenue and controlling operational risks.

[0003] Current day-ahead trading decision-making methods are mostly based on point forecasts of electricity prices, load, and generation, directly inputting the forecast values ​​into an optimization model to form a single bid. This method is simple to implement and can provide some support when the forecast is relatively accurate. However, in the actual market, factors such as electricity price fluctuations, the randomness of renewable energy output, load changes, and shifts in market supply and demand lead to discrepancies between forecasts and reality. Relying solely on a single forecast result can easily result in over- or under-reported quantities, unreasonable bid ranges, and lost trading opportunities, increased deviation costs, and aggravated revenue volatility. In addition, existing models do not fully utilize risk scenarios; while some methods consider uncertainty, they usually only use simple upper and lower bounds, fixed safety margins, or a few typical scenarios, making it difficult to reflect the joint changes in electricity prices, load, and generation under different risk conditions. Day-ahead trading revenue depends not only on the forecast mean but also on the transaction results, deviation direction, and tail losses under different scenarios. If the model cannot distinguish between ordinary fluctuations and extremely unfavorable scenarios, it is difficult to balance profitability and robustness. In terms of bid and quantity formats, existing methods mostly use single bids or single declared quantities, which are difficult to adapt to trading environments with dual uncertainties in prices and transaction results.

[0004] When market participants pursue the maximization of expected returns, they are prone to making aggressive bidding plans under the expectation of high prices or high output. Once the actual price, load or power generation deviates from the forecast, it may cause a large loss of returns. Especially in the case of extremely low prices, insufficient output or abnormal load fluctuations, the model that simply maximizes expected returns is difficult to effectively constrain tail risks, resulting in the strategy lacking risk defense capabilities. Summary of the Invention

[0005] The purpose of this invention is to provide a risk-scenario driven day-ahead spot market trading decision-making system that can improve the adaptability of the bidding scheme to market fluctuations, construct a tiered quotation and quantity optimization model for the day-ahead market, and provide support for market participants to make stable-profit day-ahead trading decisions.

[0006] The specific technical solution adopted by this invention is as follows:

[0007] A risk scenario-driven day-ahead spot market trading decision system includes a risk scenario input module, which is used to receive multiple risk scenario data of day-ahead spot market participants in the future trading cycle. The risk scenario data includes electricity price, load, power generation and corresponding scenario probabilities under different scenarios.

[0008] The net trading volume calculation module is used to calculate the net trading volume of the trading entity in each trading period and each risk scenario based on the electricity price, load, power generation and corresponding scenario probability under each risk scenario data. The net trading volume calculation module also includes adjustable resource output, which is derived by the system based on the decision of adjustable energy storage resources according to the above risk scenario data.

[0009] The tiered bidding decision module is used to generate segmented bid prices and segmented bid volumes for trading entities in the day-ahead market, forming a tiered bidding and volume reporting scheme.

[0010] The day-ahead transaction simulation module is used to determine whether each bidding segment is traded based on the day-ahead market price under various risk scenario data and the tiered bidding volume scheme, and to calculate the day-ahead transaction volume corresponding to each trading period.

[0011] The real-time deviation settlement module is used to calculate the real-time deviation electricity, deviation settlement revenue, and deviation penalty cost based on the deviation between the daily transaction volume and the net transaction volume.

[0012] The risk-return optimization module is used to solve the day-ahead quotation and volume strategy that satisfies market trading constraints and operational constraints, based on the trading returns under risk scenarios and comprehensively considering day-ahead market returns, real-time deviation settlement returns, deviation penalty costs, and downside risks.

[0013] The results output and interpretation module is used to output the segmented bid price, segmented bid volume, expected return, risk indicators and transaction risk warnings for each trading period, and to interpret at least one of the following for the above segmented bid price, segmented bid volume, expected return, risk indicators and transaction risk warnings: key risk scenarios, transaction probability, deviation direction, risk aversion coefficient and constraint triggering status.

[0014] The scrolling window output module is used to solve the future trading cycle in segments according to the declaration window, and to traverse the trading period and risk scenario in each window to output the quotation and quantity strategy of the current window.

[0015] In a preferred embodiment, the risk scenario data received by the risk scenario input module is obtained from a risk identification model, an error scenario generation model, or a historical similar day scenario generation model;

[0016] The risk scenario data includes day-ahead electricity price scenarios, real-time electricity price scenarios, load scenarios, power generation scenarios, and the probability of each of these scenarios occurring.

[0017] The scenario probability received by the risk scenario input module is used to perform weighted calculation of transaction returns under different risk scenarios, so that the optimization result can reflect the impact of the probability of different risk states on the day-ahead reporting strategy.

[0018] In a preferred embodiment, the net transaction volume calculation module determines the net transaction volume based on the type of the transaction entity;

[0019] When the trading entity is a power generation entity, the net trading volume is determined by both power generation and output of adjustable resources;

[0020] When the trading entity is an electricity seller, the net trading volume is determined by both load demand and adjustable resource output;

[0021] When the trading entity is an integrated power generation and sales entity or a virtual power plant, the net trading volume is determined by the power generation, load demand, and output of adjustable resources.

[0022] In a preferred embodiment, the tiered declaration decision module divides the day-ahead declaration scheme for each trading period into multiple bidding segments, each bidding segment including a declaration price and a declaration electricity volume;

[0023] The segmented bid prices are set to either monotonically increasing or meet a preset price order according to the transaction rules.

[0024] The segmented declared electricity volume is a non-negative value, and the sum of the declared electricity volumes in each segment does not exceed the maximum declared electricity volume of the trading entity in the corresponding time period;

[0025] The tiered application decision module sets price smoothing constraints and quantity smoothing constraints to limit price and electricity differences between adjacent price segments.

[0026] In a preferred embodiment, the day-ahead transaction simulation module determines whether a corresponding segment is traded based on the relationship between the day-ahead market price and the bid prices of each segment under the risk scenario.

[0027] When the day-ahead market price under a risk scenario is not lower than the bid price for a certain segment, the bid volume for that segment is included in the day-ahead transaction volume.

[0028] When the day-ahead market price under a risk scenario is lower than the bid price for a certain segment, the bid volume for that segment will not be included in the day-ahead transaction volume.

[0029] The day-ahead transaction simulation module uses binary variables to represent the transaction status of each price segment under different risk scenarios, and through... The constraints link the segmented transaction status, segmented bid price, and the day-ahead market price, thereby transforming the day-ahead transaction volume calculation process into a solvable mixed integer optimization form.

[0030] In a preferred embodiment, the real-time deviation settlement module compares the day-ahead transaction volume with the net transaction volume under risk scenarios;

[0031] When the former is greater than the latter, an over-reporting deviation occurs;

[0032] When the former is less than the latter, an underreporting bias occurs.

[0033] The deviation settlement benefits or deviation penalty costs corresponding to the over-reporting deviation and under-reporting deviation are calculated separately.

[0034] The real-time deviation settlement module sets different penalty coefficients for over-reporting deviations and under-reporting deviations to reflect the different impacts of different deviation directions on the transaction entity's revenue and performance risk, and uses linear penalties, segmented penalties or quadratic penalties to characterize the deviation cost.

[0035] In a preferred embodiment, the risk-return optimization module is based on maximizing the weighted expected return under multiple risk scenarios, while introducing a conditional value at risk (VAT) index to constrain or penalize the downside risk of returns under adverse scenarios.

[0036] The conditional value at risk metric is used to measure tail loss when returns fall below a pre-set confidence level.

[0037] The risk-return optimization module balances expected returns and downside risks by setting a risk aversion coefficient, thereby obtaining a risk-controllable day-ahead reporting strategy.

[0038] The risk-return optimization module uses a sample average approximation method to handle multiple risk scenarios, incorporating scenario probabilities, scenario returns, and risk constraints into the optimization model, and solving it through mixed-integer linear programming, mixed-integer quadratic programming, or commercial optimization solvers.

[0039] In one preferred embodiment,

[0040] When the trading entity includes energy storage resources, the system also includes an energy storage operation constraint module;

[0041] The energy storage operation constraint module is used to set energy storage charging power constraints, discharging power constraints, state of charge constraints, charging and discharging efficiency constraints, and charging and discharging mutual exclusion constraints.

[0042] When energy storage is in a discharging state, it increases the net amount of electricity available for sale by the trading entity;

[0043] When energy storage is in a charging state, it reduces the net amount of electricity available for sale by the trading entity or increases the demand for electricity purchases.

[0044] The energy storage operation constraint module adjusts the net transaction volume of the trading entity under various risk scenarios based on the energy storage charging and discharging status.

[0045] When the trading entity does not include energy storage resources, the system sets the adjustable resource output to zero or shuts down the energy storage operation constraint module, making the system applicable to power generation entities, electricity sales entities, or ordinary market entities that do not contain energy storage.

[0046] In a preferred embodiment, the day-ahead quotation and volume strategy output by the result output module includes multiple bid prices, corresponding bid volumes, total bid volumes, expected transaction volumes, expected returns, deviation risk level, and conditional value of risk index for each trading period.

[0047] The result output module is also used to mark high-risk trading periods;

[0048] When the probability of extreme risk scenarios, deviation electricity volume, or conditional value of risk index exceeds the preset threshold during a certain trading period, the system outputs a risk warning and prompts the trading entity to reduce the declared electricity volume, adjust the price range, or increase the deviation buffer.

[0049] The interpretation of the output results should at least calculate the contribution of each risk scenario to the objective function, conditional value at risk index, day-ahead transaction volume and real-time deviation volume, identify the key risk scenarios that lead to price increases, price decreases, contraction of transaction volume or increase of transaction volume, and output the transaction probability, deviation direction, risk aversion coefficient, constraint trigger status and the corresponding strategy adjustment reasons.

[0050] A risk-scenario-driven day-ahead spot market trading decision-making method includes the following steps:

[0051] Acquire data on multiple risk scenarios within future trading cycles, including scenario electricity price, scenario load, scenario power generation, and scenario probability;

[0052] The net trading volume of the trading entity is calculated based on the electricity price, load, power generation, and corresponding scenario probability under each risk scenario.

[0053] Set up segmented bidding prices and segmented bidding volumes in the day-ahead market to form a tiered bidding and volume reporting scheme;

[0054] Determine whether each bid segment is traded based on the day-ahead market price under each risk scenario, and calculate the day-ahead traded electricity volume;

[0055] Based on the deviation between the day-to-day transaction volume and the net transaction volume, calculate the real-time deviation volume, deviation settlement revenue, and deviation penalty cost;

[0056] Based on the transaction returns under risk scenarios, a risk-return optimization model is constructed by combining the conditional value at risk (VAT) indicator.

[0057] Solve the risk-return optimization model, output the day-ahead spot market quotation and volume strategy, and provide an explanation;

[0058] The future trading cycle is solved in segments according to the declaration window, and the transaction status, adjustable resource status, energy storage charge status or declaration boundary of the determined window are used as the initial conditions for the next declaration window to output the quotation and quantity strategy within the continuous trading cycle.

[0059] The technical effects achieved by this invention are as follows:

[0060] In this invention, the trading decision system simultaneously considers day-ahead transaction revenue, real-time deviation settlement revenue, deviation penalty cost, and adjustable resource operating cost, avoiding focusing solely on day-ahead transaction results while ignoring losses caused by actual operational deviations. Regarding risk control, the system introduces downside risk constraints to limit revenue losses under extremely unfavorable scenarios, ensuring that the optimization results balance expected returns and risk mitigation.

[0061] In this invention, the system comprehensively considers the constraints of tiered pricing increments, segmented reporting volume constraints, total reporting volume constraints, transaction volume calculation constraints, deviation volume constraints, and operational constraints of adjustable resources such as energy storage, so that the generated trading strategy not only conforms to the day-ahead spot market reporting rules, but also meets the operational boundaries of the market participants themselves.

[0062] This invention can output segmented bid prices and bid volumes for each trading period under risk scenario-driven conditions, providing stable day-ahead trading decision support for new energy power plants, electricity sales companies, virtual power plants, and integrated power generation and sales entities, thereby improving their trading revenue stability and risk control capabilities in uncertain market environments. Attached Figure Description

[0063] Figure 1 This is a schematic diagram of the basic principle structure of this invention;

[0064] Figure 2 This is a schematic diagram of the system's process framework structure in this invention;

[0065] Figure 3 This is a schematic diagram of the extended structure of the system's process framework in this invention;

[0066] Figure 4 This is a schematic diagram showing the return comparison results of the benchmark strategy and the risk scenario-driven strategy under different reporting windows in this invention;

[0067] Figure 5 This is a schematic diagram of the results of the tiered pricing and segmented reporting strategy driven by risk scenarios under a typical reporting window in this invention. Detailed Implementation

[0068] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0069] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0070] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.

[0071] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth.

[0072] In existing day-ahead trading decision-making methods, most of them are based on point forecasts of electricity prices, loads and power generation to form a single-path bidding and quantity reporting scheme. These methods are relatively simple to implement. However, in actual market operation, the market environment is complex. If trading entities only rely on a single forecast result to submit their bids, problems such as over-bidding, under-bidding, and unreasonable bidding ranges are likely to occur, which will lead to lost trading opportunities, increased real-time deviation costs, or aggravated revenue fluctuations.

[0073] Furthermore, existing trading optimization models do not fully utilize risk scenarios. These models cannot distinguish between ordinary fluctuation scenarios and extremely unfavorable scenarios, making it difficult to formulate a bidding plan that balances profitability and robustness. In terms of bidding and quantity reporting formats, existing methods mostly use a single bid or a single declared electricity volume for decision-making, which is difficult to adapt to the trading environment in the day-ahead market where price uncertainty and transaction outcome uncertainty coexist. At the same time, existing methods do not adequately consider downside risk control. When market participants pursue the maximization of expected returns, they may form overly aggressive bidding plans under the expectation of high prices or high output. Once the actual market price, load, or power generation deviates from the forecast results, there may be significant revenue losses, resulting in a lack of risk defense capabilities in the bidding strategy.

[0074] Based on this, this invention constructs a strategy optimization model for the electricity market. The core task of this model is to uniformly optimize the bidding and quantity reporting schemes of market participants in the day-ahead market, given probabilistic information on multiple future scenarios, thereby forming executable trading decisions while balancing profitability and robustness. Unlike methods that directly formulate trading strategies based solely on point forecast results, the optimization model constructed in this invention uses a set of random scenarios as input, incorporating the uncertain changes in future electricity prices, load, and generation into the decision-making process. This ensures that bidding and quantity reporting no longer rely on a single path, but rather involves a comprehensive trade-off across multiple possible market conditions. For details on the principle, please refer to [link to relevant documentation]. Figure 1 .

[0075] Please see the appendix Figures 1-3 As shown, this invention provides a risk-scenario-driven day-ahead spot market trading decision-making system, comprising:

[0076] The risk scenario input module is used to receive data on multiple risk scenarios from day-ahead spot market participants in the future trading cycle. The risk scenario data includes electricity prices, loads, power generation and corresponding scenario probabilities under different scenarios.

[0077] The net trading volume calculation module is used to calculate the net trading volume of the trading entity in each trading period and each risk scenario based on the electricity price, load, power generation and corresponding scenario probability under each risk scenario data. The net trading volume calculation module also includes adjustable resource output, which is derived by the system based on the decision of adjustable energy storage resources according to the above risk scenario data.

[0078] The tiered bidding decision module is used to generate segmented bid prices and segmented bid volumes for trading entities in the day-ahead market, forming a tiered bidding and volume reporting scheme.

[0079] The day-ahead transaction simulation module is used to determine whether each bidding segment is traded based on the day-ahead market price and tiered bidding volume scheme under various risk scenarios, and to calculate the day-ahead transaction volume corresponding to each trading period.

[0080] The real-time deviation settlement module is used to calculate the real-time deviation electricity, deviation settlement revenue, and deviation penalty cost based on the deviation between the day-ahead transaction volume and the net transaction volume.

[0081] The risk-return optimization module is used to solve the day-ahead quotation and volume strategy that satisfies market trading constraints and operational constraints, based on the trading returns under risk scenarios and comprehensively considering day-ahead market returns, real-time deviation settlement returns, deviation penalty costs, and downside risks.

[0082] The results output and interpretation module is used to output the segmented bid price, segmented bid volume, expected return, risk indicators and transaction risk warnings for each trading period, and to interpret at least one of the following for the above segmented bid price, segmented bid volume, expected return, risk indicators and transaction risk warnings: key risk scenarios, transaction probability, deviation direction, risk aversion coefficient and constraint triggering status.

[0083] The scrolling window output module is used to solve the future trading cycle in segments according to the declaration window, and to traverse the trading period and risk scenario in each window to output the quotation and quantity strategy of the current window.

[0084] Please see the appendix Figures 1-3 As shown, the risk scenario input module receives risk scenario data obtained from the risk identification model, the error scenario generation model, or the historical similar day scenario generation model;

[0085] Among them, the scenarios in the risk scenario data include day-ahead electricity price scenarios, real-time electricity price scenarios, load scenarios, power generation scenarios, and the probability of the occurrence of the above scenarios;

[0086] The scenario probability received by the risk scenario input module is used to perform weighted calculations on transaction returns under different risk scenarios, so that the optimization results can reflect the impact of the probability of different risk states on the day-ahead reporting strategy.

[0087] Please see the appendix Figures 1-3 As shown, the net trading volume calculation module determines the net trading volume based on the type of trading entity;

[0088] When the trading entity is a power generation entity, the net trading volume is determined by both power generation and output of adjustable resources;

[0089] When the trading entity is an electricity seller, the net trading volume is determined by both load demand and adjustable resource output;

[0090] When the trading entity is an integrated power generation and sales entity or a virtual power plant, the net trading volume is determined by the power generation, load demand, and output of adjustable resources.

[0091] Please see the appendix Figures 1-3As shown, the tiered declaration decision module divides the day-ahead declaration scheme for each trading period into multiple bidding segments, each of which includes a declaration price and a declaration electricity volume.

[0092] The segmented bid prices are set to either monotonically increasing or meet the preset price order according to the transaction rules.

[0093] The segmented declared electricity volume must be non-negative, and the sum of the declared electricity volumes in each segment must not exceed the maximum declared electricity volume of the trading entity in the corresponding time period;

[0094] The tiered application decision module sets price smoothing constraints and quantity smoothing constraints to limit price and electricity differences between adjacent price segments.

[0095] Please see the appendix Figures 1-3 As shown, the day-ahead transaction simulation module determines whether a transaction should be executed for a given segment based on the relationship between the day-ahead market price and the bid prices for each segment under risk scenarios.

[0096] When the day-ahead market price under a risk scenario is not lower than the bid price for a certain segment, the bid volume for that segment is included in the day-ahead transaction volume.

[0097] When the day-ahead market price under a risk scenario is lower than the bid price for a certain segment, the bid volume for that segment will not be included in the day-ahead transaction volume.

[0098] The current transaction simulation module uses binary variables to represent the transaction status of each price segment under different risk scenarios, and through... The constraints link the segmented transaction status, segmented bid price, and the day-ahead market price, thereby transforming the day-ahead transaction volume calculation process into a solvable mixed integer optimization form.

[0099] Please see the appendix Figures 1-3 As shown, the real-time deviation settlement module compares the day-ahead transaction volume with the net transaction volume under risk scenarios;

[0100] When the former is greater than the latter, an over-reporting deviation occurs;

[0101] When the former is less than the latter, an underreporting bias occurs.

[0102] The deviation settlement benefits or deviation penalty costs corresponding to the over-reporting deviation and under-reporting deviation are calculated separately.

[0103] The real-time deviation settlement module sets different penalty coefficients for over-reporting deviations and under-reporting deviations to reflect the different impacts of different deviation directions on the transaction entity's revenue and performance risk, and uses linear penalties, segmented penalties or quadratic penalties to characterize the deviation cost.

[0104] Please see the appendix Figures 1-3As shown, the risk-return optimization module is based on maximizing the weighted expected return under multiple risk scenarios, while introducing the conditional value at risk index to constrain or penalize the downside risk of returns under adverse scenarios.

[0105] Conditional Value at Risk (VaR) measures tail loss when returns fall below a pre-set confidence level.

[0106] The risk-return optimization module balances expected returns and downside risks by setting a risk aversion coefficient, thereby obtaining a day-ahead reporting strategy with controllable risks.

[0107] The risk-return optimization module uses a sample average approximation method to handle multiple risk scenarios, incorporating scenario probabilities, scenario returns, and risk constraints into the optimization model, and solving it through mixed-integer linear programming, mixed-integer quadratic programming, or commercial optimization solvers.

[0108] The risk-return optimization model considers at least one of the following constraints:

[0109] Segmented pricing increment constraint;

[0110] Non-negative constraints on segmented electricity reporting;

[0111] Total declared electricity volume limit constraint;

[0112] Smoothing constraint for adjacent segment pricing;

[0113] Smoothing constraint for adjacent segmented reporting volumes;

[0114] Constraints for calculating day-to-day transaction volume;

[0115] Real-time deviation power calculation constraints;

[0116] Constraints on energy storage charging and discharging operations;

[0117] Conditional risk value constraint.

[0118] Specifically, when the trading entity includes energy storage resources, the system also includes an energy storage operation constraint module;

[0119] The energy storage operation constraint module is used to set energy storage charging power constraints, discharging power constraints, state of charge constraints, charging and discharging efficiency constraints, and charging and discharging mutual exclusion constraints.

[0120] When energy storage is in a discharging state, it increases the net amount of electricity available for sale by the trading entity;

[0121] When energy storage is in a charging state, it reduces the net amount of electricity available for sale by the trading entity or increases the demand for electricity purchases.

[0122] Among them, the energy storage operation constraint module adjusts the net transaction volume of the trading entity under various risk scenarios based on the energy storage charging and discharging status;

[0123] When the trading entity does not include energy storage resources, the system will set the adjustable resource output to zero or shut down the energy storage operation constraint module, making the system applicable to power generation entities, power sales entities, or ordinary market entities that do not contain energy storage.

[0124] Please see the appendix Figures 1-3 As shown, the day-ahead quotation strategy output by the results output module includes multiple bid prices, corresponding bid volumes, total bid volumes, expected transaction volumes, expected returns, deviation risk level, and conditional value at risk (VAT) indicators for each trading session.

[0125] The results output module is also used to mark high-risk trading periods;

[0126] When the probability of extreme risk scenarios, deviation electricity volume, or conditional value of risk index exceeds the preset threshold during a certain trading period, the system outputs a risk warning and prompts the trading entity to reduce the declared electricity volume, adjust the price range, or increase the deviation buffer.

[0127] The interpretation of the output results should at least calculate the contribution of each risk scenario to the objective function, conditional value at risk index, day-ahead transaction volume and real-time deviation volume, identify the key risk scenarios that lead to price increases, price decreases, contraction of transaction volume or increase of transaction volume, and output the transaction probability, deviation direction, risk aversion coefficient, constraint trigger status and the corresponding strategy adjustment reasons.

[0128] Based on the above system's process framework, and considering that the research object of this invention includes both power generation entities and aggregated market entities such as electricity sales companies and virtual power plants, the model adopts a unified net position modeling approach, defining the net electricity that a market entity can provide to the market in a certain period as:

[0129]

[0130] in, and Representing the scene respectively Next period The amount of electricity generated and consumed. This refers to the flexible adjustment capacity formed by the main body through energy storage, demand response, or adjustable units. When market entities lack internal adjustment capabilities, they can order... .thus, This indicates that the subject is in the time period It has net electricity sales capacity. This indicates that it needs to repurchase or replenish its electricity from the market.

[0131] To facilitate model description, the following optimization model is first established using a net supply entity as an example; for a net purchase entity, the modeling method can be obtained through symbol transformation.

[0132] Let the day-ahead market settlement period be discretized as This refers to a time period. If a 15-minute trading granularity is used, it will cover the complete trading cycle for the following day. To improve the flexibility of the pricing strategy, this invention models each time period using a tiered pricing approach. Let the time period be... Internal settings The first pricing segment, the... The price quoted for the segment is The corresponding declared electricity volume is Then the time period The total number of declarations is:

[0133]

[0134] Since each segment of the bidding process involves market clearing, the actual transaction volume depends not only on the bid volume itself, but also on the market clearing price for the corresponding scenario during that period. .

[0135] If a segmented pricing acceptance rule is adopted, then in the scenario Next period The daily transaction volume can be written as:

[0136]

[0137] in, For indicator functions. When the first... When a segment's bid is lower than or equal to the scenario clearing price, that segment's electricity volume is considered traded; otherwise, it is not traded. Therefore, day-ahead bidding schemes will correspond to different trading results under different price scenarios, which is one of the essential differences between market trading strategy optimization and conventional deterministic scheduling models.

[0138] During the real-time operation phase, if the actual net electricity available for delivery... Compared with the previous day's transaction volume Inconsistencies will result in discrepancies in settlement. Considering that discrepancies in electricity consumption are typically settled according to real-time prices, and that discrepancies are subject to additional assessments or penalties under many market rules, this invention explicitly incorporates the discrepancy cost into the revenue function. Scenario Definition Next period The positive and negative deviations are respectively:

[0139]

[0140] in, This indicates a shortage deviation when the declared transaction volume exceeds the actual deliverable electricity volume. This represents the excess deviation when the actual electricity volume exceeds the previous day's transaction volume. Considering that the larger the deviation, the higher the additional cost usually is, this invention uses a double penalty to characterize the two types of deviations, in order to enhance the strategy's ability to constrain large deviation situations.

[0141] Based on this, the scenario The profit for the entire optimization cycle can be expressed as:

[0142]

[0143] in, Representing a scene Next period Real-time settlement price, and These represent the penalty coefficients for shortage bias and surplus bias, respectively. This represents the adjustment cost of internally flexible resource allocation during that period. The above revenue function consists of four parts: day-ahead market transaction revenue, real-time deviation electricity settlement revenue, deviation penalty cost, and internal adjustment cost. Because... and , All of these come from the scenario generation module in Chapter 3, so this revenue function can uniformly evaluate trading strategies under uncertain environments.

[0144] To balance profitability and risk control, this invention defines the optimization objective as a comprehensive approach of "maximizing expected return and mitigating downside risk." Let the scenario probability be... Then the overall objective function can be written as:

[0145]

[0146] in, This is the risk aversion coefficient. Indicates confidence level Conditional Value at Risk (CVaR) is used to measure both profit and loss. Since market participants are not only concerned with average returns in actual transactions, but also with the lower bound of returns under extremely unfavorable scenarios, introducing CVaR can effectively curb the strategy's over-reliance on a few high-return scenarios, making the solution more in line with the robustness requirements of actual trading decisions.

[0147] To facilitate the solution, the CVaR term is written as an auxiliary variable. Let For Value-at-Risk approximation variables, For the scene Given the excess loss variable, we have

[0148]

[0149] And satisfy the constraints:

[0150]

[0151] Therefore, the objective function can be transformed into

[0152]

[0153] In addition to the above objectives, the model also needs to meet the constraints of transaction declaration and physical operation.

[0154] First, tiered pricing should meet the monotonicity requirement, that is:

[0155]

[0156] Furthermore, the declared quantity for each segment should meet the requirements of non-negative and total quantity limits:

[0157]

[0158]

[0159] in, This represents the maximum net electricity that can be declared during time period t. To avoid unreasonable jumps in declaration schemes between adjacent time periods, a smoothing constraint on declared quantity and price can be further introduced:

[0160]

[0161]

[0162] in, and These represent the upper limit of the reported volume and price change in adjacent time periods, respectively. Such constraints allow for a smoother reporting strategy, preventing overly aggressive pricing behavior from negatively impacting market adaptability and actual execution.

[0163] For internal flexible adjustment variables It must meet its adjustable range constraints:

[0164]

[0165] When market participants include energy storage resources It can be further broken down into charging and discharging quantities, and a state equation can be introduced to represent the evolution of stored energy. For example, let... and These represent the charging and discharging power, To represent the state of charge of the energy storage, we have:

[0166]

[0167]

[0168]

[0169]

[0170] If the research object is an entity that does not include energy storage, then this constraint can be degenerated into In this way, the model can accommodate both pure trading entities and virtual power plants with flexible resources within a unified framework.

[0171] Due to the recent trading volume The problem involves a comparison between tiered pricing and scenario-based electricity pricing, which would require a non-smooth indicator function for direct solution. Therefore, this study introduces a binary variable. Linearization is applied to determine whether a segment has been traded, and the following rules apply:

[0172]

[0173] And establish constraints using the Big-M form:

[0174]

[0175]

[0176] Therefore, the actual transaction volume is written as:

[0177]

[0178] Similarly, to facilitate the representation of positive and negative deviations, auxiliary variables can be introduced that satisfy:

[0179]

[0180]

[0181] After the above transformation, the entire model can be transformed into a mixed integer quadratic programming problem with binary variables and a quadratic penalty term.

[0182] In terms of solution methodology, this invention employs a "scenario-driven + sample average approximation" approach. Specifically, it uses the finite set of scenarios generated in Chapter 3 to approximate the original stochastic process, then transforms the optimization problem into a finite-dimensional deterministic equivalent model for solution. For problems of moderate scale, commercial solvers such as Gurobi and CPLEX can be directly used to solve the mixed-integer quadratic programming model. When the number of scenarios is large or the model is further extended to multi-day rolling optimization or multi-market collaborative optimization, scenario reduction, decomposition coordination, or heuristic algorithms can be combined to improve solution efficiency. Through this approach, the model not only possesses clear economic meaning and physical constraint interpretation but also exhibits good computability.

[0183] Overall, the market strategy optimization model constructed in this chapter is based on the multivariate stochastic scenario generated in the previous chapter. Centered on day-ahead market tiered pricing and volume decision-making, it integrates market transaction mechanisms, deviation settlement rules, internal flexible resource adjustments, and risk-reward trade-offs into a unified optimization framework. Compared to traditional methods that formulate order placement strategies based solely on single pathpoint predictions, this model can simultaneously evaluate potential gains and adverse scenarios under multiple conditions, thereby forming a more adaptive and robust trading plan.

[0184] The present invention will be further described below with reference to embodiments.

[0185] The following embodiments are only used to illustrate the technical solutions of the present invention more clearly, and should not be used to limit the scope of protection of the present invention.

[0186] In this embodiment, market participants obtain forecasts of electricity prices, load, and power generation for future trading days before day-ahead trading. Based on the aforementioned risk identification and scenario generation results, they form multiple risk scenarios with probability weights. Each risk scenario corresponds to a different market price, tradable electricity volume, and deviation risk level, used to describe different market conditions that may occur on future trading days.

[0187] The system first organizes the input data, including day-ahead electricity price scenarios, real-time price scenarios, power generation scenarios, load scenarios, and corresponding probabilities for each time period of the future trading day. Then, based on the power generation, load, and adjustable resource status of market participants, the system calculates the net tradable electricity volume for each scenario. On this basis, the system divides the bidding strategy for each trading period into multiple bid and quantity segments, forming a tiered bidding scheme. Different segments correspond to different transaction probabilities and profit levels, enabling market participants to strive for higher profits when prices are high and reduce deviation exposure when risks are high.

[0188] During the optimization process, the system comprehensively considers day-ahead transaction revenue, real-time deviation settlement, deviation penalty costs, and risk losses to form a revenue-risk synergy optimization objective. At the same time, the system sets constraints such as segmented bidding order, upper limit of declared electricity volume, smoothing of adjacent segments, calculation of transaction volume, and calculation of deviation volume to ensure that the generated bidding and quantity schemes comply with day-ahead market declaration rules and the main body's operating boundaries. If the trading entity includes adjustable resources such as energy storage, charging and discharging power, state of charge, and charging and discharging mutual exclusion constraints are further considered; if it does not include energy storage resources, the corresponding module can be turned off.

[0189] After the solution is completed, the system outputs the segmented bid prices, segmented bid volumes, total bid volumes, expected transaction results, and risk warnings for each time period of the future trading day. Trading entities can directly form a day-ahead market bid plan based on the results, or submit high-risk periods for manual review before submission.

[0190] Implementation results are as follows Figure 3 As shown in the results, the return comparison results indicate that the strategy of this invention achieved higher actual returns than the benchmark strategy in most of the 10 application windows, with the most significant increase in returns in the 8th window. This demonstrates that the risk scenario-driven application strategy can improve the ability to obtain returns during some high-yield opportunity periods.

[0191] Meanwhile, the returns in some windows were lower than the benchmark strategy, indicating that risk constraints may have led to a certain degree of conservatism in some periods, but overall the positive return increase outweighed the negative return decrease.

[0192] Figure 4 The strategy optimization results for the 8th declaration window are given. In terms of price declaration, the three-segment quotation forms a hierarchical structure around the benchmark price. In terms of electricity declaration, the total declaration volume is adjusted according to the risk status, and the declaration volume is actively reduced during periods of greater volatility.

[0193] Therefore, it can be seen that the present invention can dynamically adjust the pricing and reporting scheme according to the risk scenario, so that the day-ahead reporting strategy has both the ability to capture profits and the ability to defend against risks.

[0194] This application also provides a risk-scenario-driven day-ahead spot market trading decision-making method, including the following steps:

[0195] Acquire data on multiple risk scenarios within the future trading cycle. The risk scenario data includes scenario electricity price, scenario load, scenario power generation, and scenario probability.

[0196] The net trading volume of the trading entity is calculated based on the electricity price, load, power generation, and corresponding scenario probability under each risk scenario.

[0197] Set up segmented bidding prices and segmented bidding volumes in the day-ahead market to form a tiered bidding and volume reporting scheme;

[0198] Determine whether each bid segment is traded based on the day-ahead market price under each risk scenario, and calculate the day-ahead traded electricity volume;

[0199] Based on the deviation between the day-to-day transaction volume and the net transaction volume, calculate the real-time deviation volume, deviation settlement revenue, and deviation penalty cost;

[0200] Based on the transaction returns under risk scenarios, a risk-return optimization model is constructed by combining the conditional value at risk (VAT) indicator.

[0201] Solve the risk-return optimization model, output and interpret the day-ahead spot market quotation and volume strategy. The interpretation module receives the optimal segmented bid price, segmented bid volume, transaction state variables, scenario return, deviation volume, CVaR auxiliary variables, and constraint dual information or constraint relaxation state from the risk-return optimization module, and generates strategy interpretation results for each trading period t and quotation segment k. The strategy interpretation results include: First, according to scenario probabilities... and scene benefits First, calculate the contribution of each risk scenario to the expected return; second, according to... and confidence level First, identify the extremely unfavorable scenarios that enter the tail set of CVaR; second, statistically analyze the transaction frequency of each segment in the risk scenario set Ω to form the segmented transaction probability; third, based on... and Fifth, determine the main deviation direction of over- or under-reporting; and record whether constraints such as total quantity limit, price monotonicity, smoothing constraint, adjustable resource upper and lower limits, and energy storage charge status upper and lower limits are triggered.

[0202] In one implementation, the strategy interpretation module defines a key risk scenario as one that simultaneously meets any of the following conditions: the scenario return is lower than a preset quantile, and the CVaR auxiliary variable... If the value is greater than zero, the deviation penalty cost exceeds a threshold, the deviation between the day-ahead transaction volume and the net transaction volume exceeds a threshold, or the marginal impact of the scenario on the objective function exceeds a preset proportion, the system outputs its day-ahead price, real-time price, load, generation, net transaction volume, transaction segment, deviation volume, and revenue decomposition for the identified key risk scenarios. This allows traders to determine whether a price increase in a certain period is driven by a high-price but low-transaction-probability scenario, whether a contraction in volume is triggered by downside risk or energy storage constraints, or whether a price reduction is used to improve the certainty of transactions in a low-price scenario.

[0203] The strategy explanation module can also generate sensitivity interpretations based on the solution results under different risk aversion coefficients β. When β increases, if the total declared volume decreases, the volume in the low-price segment increases, or the volume in the high-price segment decreases during a certain period, the system outputs the correspondence between this change and tail risk control. When β decreases, if the declaration strategy shifts to increasing expected returns, the system outputs the strategy change caused by the increased probability of a transaction or the increased returns in the high-price scenario. Thus, the system not only outputs the final bid volume value, but also outputs verifiable reasons corresponding to the risk scenario, return composition, and constraint status.

[0204] The future trading cycle is solved in segments according to the declaration window, and the transaction status, adjustable resource status, energy storage charge status or declaration boundary of the determined window are used as the initial conditions for the next declaration window to output the quotation and quantity strategy within the continuous trading cycle.

[0205] The scrolling window output module divides the complete trading period T into several windows. Where H is the window length, Let this be the rolling step size. Within the m-th window, the system establishes a local optimization model only for the trading periods and risk scenarios covered by the window, and can retain a few periods at the end of the window as a look-ahead interval to reduce policy mutations at the window boundaries. After solving, the system fixes the current rolling step size. The system implements a quotation and quantity reporting strategy within the window and transmits the adjustable resource output, energy storage status of charge, cumulative deviation, declared power boundary, and smoothing constraint status of adjacent time periods at the end of the window to the next window.

[0206] During the rolling solution process, when new day-ahead electricity price forecasts, real-time electricity price forecasts, load forecasts, power generation forecasts, scenario probabilities, or market price limit parameters are updated, the rolling window optimization module regenerates or corrects the risk scenario set and resolves the subsequent windows that have not yet been fixed using the latest scenario data. For periods that have entered the application freeze period or have already been submitted to the market, the system maintains its bidding and quantity reporting strategy unchanged; for periods that are not frozen, the system adjusts the segmented bidding, segmented quantity reporting, and risk buffer based on the updated risk-return optimization results.

[0207] The rolling window optimization module can also be used in conjunction with a sample average approximation mechanism. For each window, the system extracts a representative subset of scenarios from a large-scale scenario library for optimization, and maintains the statistical characteristics of price, load, power generation, and deviation risk through scenario weights. When the proportion of high-risk scenarios within the window increases or the CVaR index exceeds a threshold, the system increases the number of scenario samples in that window or shortens the rolling step size to improve the accuracy of risk identification. Through the above processing, the long-cycle, multi-scenario, and multi-segmented mixed integer optimization problem is decomposed into multiple continuously updatable local problems, thereby improving computational efficiency and engineering deployability.

[0208] Furthermore, the above method employs a tiered reporting strategy for day-ahead spot market quotations and quantities.

[0209] The tiered bidding strategy includes multiple bid prices and multiple bid volumes within the same trading session, enabling trading entities to achieve different transaction results under different market price scenarios.

[0210] The objectives of the risk-return optimization model include maximizing the weighted expected return under multiple risk scenarios and controlling the impact of conditional value of risk on the objective function through the risk aversion coefficient.

[0211] Based on the systems and methods mentioned above, they can be applied to at least one or more computer-readable storage media, servers, or cloud computing platforms, which have processors storing computer programs and are automatically executed by corresponding system functions and method flows to achieve automated operation of system functions and method flows.

[0212] This invention introduces multiple risk scenarios and their probabilities related to electricity price, load, and power generation. Instead of relying solely on a single forecast value for bidding, it uniformly evaluates the transaction results, deviation electricity volume, and transaction revenue of different bidding and quantity schemes under multiple risk scenarios. By constructing a tiered bidding strategy, the system enables market participants to achieve differentiated transaction results at different price levels, improving the adaptability of bidding schemes to market fluctuations. It constructs a tiered bidding and quantity optimization model for the day-ahead market, comprehensively considering day-ahead transaction revenue, real-time deviation settlement, deviation penalty costs, and downside risk constraints, and outputs the optimal bidding price and bidding electricity volume for each time period, providing support for market participants to make profitable and stable day-ahead trading decisions.

[0213] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.

Claims

1. A risk-scenario-driven day-ahead spot market trading decision-making system, characterized in that: include The risk scenario input module is used to receive multiple risk scenario data from day-ahead spot market participants in the future trading cycle. The risk scenario data includes electricity price, load, power generation and corresponding scenario probability under different scenarios. The net trading volume calculation module is used to calculate the net trading volume of the trading entity in each trading period and each risk scenario based on the electricity price, load, power generation and corresponding scenario probability under each risk scenario data. The net trading volume calculation module also includes adjustable resource output, which is derived by the system based on the decision of adjustable energy storage resources according to the above risk scenario data. The tiered bidding decision module is used to generate segmented bid prices and segmented bid volumes for trading entities in the day-ahead market, forming a tiered bidding and volume reporting scheme. The day-ahead transaction simulation module is used to determine whether each bidding segment is traded based on the day-ahead market price under various risk scenario data and the tiered bidding volume scheme, and to calculate the day-ahead transaction volume corresponding to each trading period. The real-time deviation settlement module is used to calculate the real-time deviation electricity, deviation settlement revenue, and deviation penalty cost based on the deviation between the daily transaction volume and the net transaction volume. The risk-return optimization module is used to solve the day-ahead quotation and volume strategy that satisfies market trading constraints and operational constraints, based on the trading returns under risk scenarios and comprehensively considering day-ahead market returns, real-time deviation settlement returns, deviation penalty costs, and downside risks. The results output and interpretation module is used to output the segmented bid price, segmented bid volume, expected return, risk indicators and transaction risk warnings for each trading period, and to interpret at least one of the following for the above segmented bid price, segmented bid volume, expected return, risk indicators and transaction risk warnings: key risk scenarios, transaction probability, deviation direction, risk aversion coefficient and constraint triggering status. The scrolling window output module is used to solve the future trading cycle in segments according to the declaration window, and to traverse the trading period and risk scenario in each window to output the quotation and quantity strategy of the current window.

2. The day-ahead spot market trading decision-making system based on risk scenario driving according to claim 1, characterized in that: The risk scenario input module receives risk scenario data obtained from a risk identification model, an error scenario generation model, or a historical similar day scenario generation model. The risk scenario data includes day-ahead electricity price scenarios, real-time electricity price scenarios, load scenarios, power generation scenarios, and the probability of each of these scenarios occurring. The scenario probability received by the risk scenario input module is used to perform weighted calculation of transaction returns under different risk scenarios, so that the optimization result can reflect the impact of the probability of different risk states on the day-ahead reporting strategy.

3. The day-ahead spot market trading decision-making system based on risk scenario driving according to claim 1, characterized in that: The net transaction volume calculation module determines the net transaction volume based on the type of transaction entity. When the trading entity is a power generation entity, the net trading volume is determined by both power generation and output of adjustable resources; When the trading entity is an electricity seller, the net trading volume is determined by both load demand and adjustable resource output; When the trading entity is an integrated power generation and sales entity or a virtual power plant, the net trading volume is determined by the power generation, load demand, and output of adjustable resources.

4. The day-ahead spot market trading decision-making system based on risk scenario driving according to claim 1, characterized in that: The tiered declaration decision module divides the day-ahead declaration scheme for each trading period into multiple bidding segments, each of which includes a declaration price and a declaration electricity volume. The segmented bid prices are set to either monotonically increasing or meet a preset price order according to the transaction rules. The segmented declared electricity volume is a non-negative value, and the sum of the declared electricity volumes in each segment does not exceed the maximum declared electricity volume of the trading entity in the corresponding time period; The tiered application decision module sets price smoothing constraints and quantity smoothing constraints to limit price and electricity differences between adjacent price segments.

5. The day-ahead spot market trading decision-making system based on risk scenario driving according to claim 1, characterized in that: The day-ahead transaction simulation module determines whether a corresponding segment is traded based on the relationship between the day-ahead market price and the bid prices of each segment under the risk scenario. When the day-ahead market price under a risk scenario is not lower than the bid price for a certain segment, the bid volume for that segment is included in the day-ahead transaction volume. When the day-ahead market price under a risk scenario is lower than the bid price for a certain segment, the bid volume for that segment will not be included in the day-ahead transaction volume. The day-ahead transaction simulation module uses binary variables to represent the transaction status of each price segment under different risk scenarios, and through... The constraints link the segmented transaction status, segmented bid price, and the day-ahead market price, thereby transforming the day-ahead transaction volume calculation process into a solvable mixed integer optimization form.

6. The day-ahead spot market trading decision-making system based on risk scenario driving according to claim 1, characterized in that: The real-time deviation settlement module compares the day-ahead transaction volume with the net transaction volume under risk scenarios. When the former is greater than the latter, an over-reporting deviation occurs; When the former is less than the latter, an underreporting bias occurs. And calculate the deviation settlement benefit or deviation penalty cost corresponding to the over-reporting deviation and under-reporting deviation respectively; The real-time deviation settlement module sets different penalty coefficients for over-reporting deviations and under-reporting deviations to reflect the different impacts of different deviation directions on the transaction entity's revenue and performance risk, and uses linear penalties, segmented penalties or quadratic penalties to characterize the deviation cost.

7. The day-ahead spot market trading decision system based on risk scenario driving according to claim 1, characterized in that: The risk-return optimization module is based on maximizing the weighted expected return under multiple risk scenarios, and at the same time introduces the conditional value of risk index to constrain or penalize the downside risk of returns under adverse scenarios. The conditional value at risk metric is used to measure tail loss when returns fall below a pre-set confidence level. The risk-return optimization module balances expected returns and downside risks by setting a risk aversion coefficient, thereby obtaining a risk-controllable day-ahead reporting strategy. The risk-return optimization module uses a sample average approximation method to handle multiple risk scenarios, incorporating scenario probabilities, scenario returns, and risk constraints into the optimization model, and solving it through mixed-integer linear programming, mixed-integer quadratic programming, or commercial optimization solvers.

8. The day-ahead spot market trading decision system based on risk scenario driving according to claim 1, characterized in that: When the trading entity includes energy storage resources, the system also includes an energy storage operation constraint module; The energy storage operation constraint module is used to set energy storage charging power constraints, discharging power constraints, state of charge constraints, charging and discharging efficiency constraints, and charging and discharging mutual exclusion constraints. When energy storage is in a discharging state, it increases the net amount of electricity available for sale by the trading entity; When energy storage is in a charging state, it reduces the net amount of electricity available for sale by the trading entity or increases the demand for electricity purchases. The energy storage operation constraint module adjusts the net transaction volume of the trading entity under various risk scenarios based on the energy storage charging and discharging status. When the trading entity does not include energy storage resources, the system sets the adjustable resource output to zero or shuts down the energy storage operation constraint module, making the system applicable to power generation entities, electricity sales entities, or ordinary market entities that do not contain energy storage.

9. The day-ahead spot market trading decision-making system based on risk scenario driving according to claim 1, characterized in that: The day-ahead quotation and volume strategy output by the result output module includes multiple bid prices, corresponding bid volumes, total bid volumes, expected transaction volumes, expected returns, deviation risk level, and conditional value of risk index for each trading period. The result output module is also used to mark high-risk trading periods; When the probability of extreme risk scenarios, deviation electricity volume, or conditional value of risk index exceeds the preset threshold during a certain trading period, the system outputs a risk warning and prompts the trading entity to reduce the declared electricity volume, adjust the price range, or increase the deviation buffer. The interpretation of the output results should at least calculate the contribution of each risk scenario to the objective function, conditional value at risk index, day-ahead transaction volume and real-time deviation volume, identify the key risk scenarios that lead to price increases, price decreases, contraction of transaction volume or increase of transaction volume, and output the transaction probability, deviation direction, risk aversion coefficient, constraint trigger status and the corresponding strategy adjustment reasons.

10. A risk-scenario-driven day-ahead spot market trading decision-making method, applied to the risk-scenario-driven day-ahead spot market trading decision-making system described in claims 1-9, characterized in that, Includes the following steps: Acquire data on multiple risk scenarios within future trading cycles, including scenario electricity price, scenario load, scenario power generation, and scenario probability; The net trading volume of the trading entity is calculated based on the electricity price, load, power generation, and corresponding scenario probability under each risk scenario. Set up segmented bidding prices and segmented bidding volumes in the day-ahead market to form a tiered bidding and volume reporting scheme; Determine whether each bid segment is traded based on the day-ahead market price under each risk scenario, and calculate the day-ahead traded electricity volume; Based on the deviation between the day-to-day transaction volume and the net transaction volume, calculate the real-time deviation volume, deviation settlement revenue, and deviation penalty cost; Based on the transaction returns under risk scenarios, a risk-return optimization model is constructed by combining the conditional value at risk (VAT) indicator. Solve the risk-return optimization model, output the day-ahead spot market quotation and volume strategy, and provide an explanation; The future trading cycle is solved in segments according to the declaration window, and the transaction status, adjustable resource status, energy storage charge status or declaration boundary of the determined window are used as the initial conditions for the next declaration window to output the quotation and quantity strategy within the continuous trading cycle.

Images