Method and system for constructing medium-and-long-term electricity purchasing and selling combined transaction model of power grid enterprise

Abstract

The invention provides a method and system for constructing a medium-and-long-term electricity purchasing and selling combined transaction model for a power grid enterprise, and the method comprises the steps: carrying out the modeling of the uncertain electricity purchasing cost of an electricity purchasing side for the scenes of different types of power supplies at the electricity purchasing side, and obtaining the uncertain factors of the electricity purchasing side; the method comprises the following steps: analyzing power demands of various types of users on a power selling side, and modeling the power purchase quantity and load of a power grid enterprise agent power purchase user on the power selling side and the uncertain total power consumption to obtain uncertain factors of the power selling side; according to the uncertainty factors of the electricity purchasing side and the electricity selling side, risk control index evaluation is carried out based on the conditional value-at-risk CVaR, and annual and monthly conditional value-at-risk is obtained; based on the annual condition value-at-risk, establishing a combined transaction model for determining monthly electric quantity by annual transaction considering an annual electricity purchase cost expected value and risk management and control; and based on the monthly conditional value-at-risk, establishing a combined transaction model for determining the daily electric quantity by monthly transaction in consideration of the expected value of the annual electricity purchase cost and risk management and control.

Description

Technical Field

[0001] The present invention relates to the technical field of transaction behavior modeling of power grid enterprises, and in particular to a method and system for constructing a medium- and long-term power purchase and sales combination transaction model of a power grid enterprise, and also to a corresponding computer terminal and a computer-readable storage medium. Background Art

[0002] The "double high" access to the grid with high penetration rate of renewable energy and high proportion of power electronic equipment has profoundly changed the composition and operation characteristics of the traditional power grid, thereby promoting the construction of a new power system and its evolution to a comprehensive smart energy system. At the same time, the medium- and long-term power market operation and spot market pilot promoted by the new power reform, as well as the process of building a national unified power market, continue to accelerate, prompting traditional power grid companies to focus on the continuous optimization of power transmission and distribution business, and in accordance with policy requirements, to act as agents of current non-market chemical commercial users (or establish independent power sales companies) to expand the purchase and sale of power business, thereby transforming into a comprehensive service provider covering transmission and distribution and purchase and sale of power.

[0003] As the hub of the new power system and an important participant in the power market, power grid companies are faced with key issues and practical challenges such as high-proportion access and consumption of renewable energy, sufficient and reliable power system, safe and economical operation of the power grid, and responsibility for ensuring power supply for energy supply and demand. New business forms such as establishing independent power sales companies, assuming minimum power supply for market users and acting as agents for non-market users provide important opportunities for improvement of the traditional profit model centered on the collection of transmission and distribution electricity prices.

[0004] Existing research on power purchase and sale is mostly focused on the "power sales company model" for profit (power grid enterprises set up independent power sales companies and undertake the minimum power supply for market users, which is equivalent to the power sales company model), and rarely involves the optimization of transaction strategy combinations and risk control of the "agent power purchase model" under the constraints of power supply guarantee of power grid enterprises. Further theoretical exploration and empirical research are necessary. At present, no description or report of similar technology to the present invention has been found, and similar information at home and abroad has not been collected. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method and system for constructing a medium- and long-term power purchase and sale combination transaction model for power grid enterprises, and also provides a corresponding computer terminal and computer-readable storage medium.

[0006] According to a first aspect of the present invention, a method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise is provided, comprising:

[0007] For scenarios with different types of power sources on the power purchasing side, the uncertain power purchasing cost on the power purchasing side is modeled to obtain the uncertainty factors on the power purchasing side;

[0008] Analyze the electricity demand of multiple types of users on the power sales side, model the power purchase volume and load of power purchase users on behalf of power grid enterprises on the power sales side, and the uncertain total power consumption, and obtain the uncertainty factors on the power sales side;

[0009] According to the uncertainty factors on the power purchasing and selling sides, the risk control index is evaluated based on the conditional risk value CVaR to obtain the annual and monthly conditional risk value;

[0010] Based on the annual conditional value at risk, a combined trading model for monthly electricity consumption is established, which considers the expected value of the annual electricity purchase cost and risk control.

[0011] Based on the monthly conditional value at risk, a combined trading model for daily electricity consumption determined by monthly transactions is established, which takes into account the expected value of the annual electricity purchase cost and risk control.

[0012] Preferably, the modeling of the uncertain electricity purchase cost on the electricity purchase side for the scenario of different types of power sources on the electricity purchase side includes:

[0013] Clarify the uncertainty of power sources of various transaction products on the power purchasing side and generate uncertain scenarios of power sources of various transaction products;

[0014] For the uncertain scenarios of power sources of various trading varieties, the purchase cost of electricity for local preferential power generation and inter-provincial planned power generation is calculated. and Conduct annual modeling:

[0015]

[0016] in, Respectively represent the monthly electricity consumption of local wind power, photovoltaic power and gas turbine in the mth month, Respectively represent the corresponding prices;

[0017]

[0018] in, They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively. is the corresponding price;

[0019] For the uncertain scenarios of power sources of various trading varieties, the purchase cost of electricity for local preferential power generation and inter-provincial planned power generation is calculated. and To perform monthly modeling:

[0020]

[0021] in, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. is the corresponding price, T is the number of discrete time periods;

[0022]

[0023] in, They represent the planned electricity of inter-provincial coal-fired power, hydropower and nuclear power in period t, For the corresponding price.

[0024] Preferably, the step of clarifying the uncertainty of power sources of various transaction products on the power purchasing side and generating uncertain scenarios of power sources of various transaction products includes:

[0025] The various types of power sources traded on the power purchasing side include: wind power, photovoltaic power generation and inter-provincial power purchase; among which:

[0026] The wind power mentioned above takes into account the uncertain impact of wind power fluctuations on electricity prices and electricity purchase costs;

[0027] The photovoltaic power generation mentioned above takes into account the uncertain impact of weather changes on electricity prices and electricity purchase costs;

[0028] The cross-provincial power purchase mentioned above takes into account the price fluctuations of power markets in different regions, transmission costs, and the uncertain impact of policy subsidies on electricity prices and power purchase costs;

[0029] Statistical analysis methods are used to quantitatively analyze the historical data of the output of various types of power sources on the power purchasing side to obtain the fluctuations of uncertain variables, and the Monte Carlo sampling method is used to generate uncertain scenarios for various types of power sources.

[0030] Preferably, the analysis of the electricity demand of multiple types of users on the electricity sales side, modeling of the power purchase amount and load of the power purchase users on behalf of the power grid enterprises on the electricity sales side and the uncertain total electricity consumption, and obtaining the uncertainty factors on the electricity sales side include:

[0031] A combined prediction model for the amount of electricity purchased and the load of electricity purchased by a power grid enterprise as an agent is provided, wherein the combined prediction model comprises:

[0032] Multiple regression analysis model, which uses historical data to establish a multiple regression model, in which electricity demand is used as the dependent variable, and gross domestic product, industrial proportion, population size, urbanization rate and energy conservation and emission reduction index are used as independent variables. The expression of the model is:

[0033] Y=β0+β1X1+β2X2+...,+β n X n +ε (5)

[0034] Among them, Y is the dependent variable, X1, X2,,,X n are independent variables, β0,β1,,,β nis the coefficient, ε is the error term;

[0035] Time series analysis model, which is used to analyze the historical time series data of electricity demand to capture seasonal, trend and cyclical factors, and then predict future electricity demand. The expression of this model is:

[0036] (1-φ1L-φ2L 2 -...-φ p L p )(1-L) d Y t =(1-θ1L-θ2L 2 -...θ q L q )ε t (6)

[0037] Where L is the lag operator, d is the difference order, φ and θ are the autoregressive and moving average coefficients, p and q are the corresponding orders, ε t is the error term.

[0038] Economic transmission analysis model, which is used to analyze the impact of macroeconomic factors on electricity demand. The model links macroeconomic indicators with electricity demand and analyzes the potential impact of policy changes on electricity demand. The expression of the model is:

[0039] E t =α+β1·GDP t +β2·IPI t +β3·IR t +γ·Z t +ε t

[0040] Among them, E t is the electricity demand, α, β1, β2, β3, γ are model coefficients, GDP t is the gross domestic product, IPI t is the industrial production index, IR t is the policy interest rate, Z t for other relevant macroeconomic factors;

[0041] Residential air conditioning model, which uses the degree-day method to estimate the electricity demand for residential air conditioning. Based on the relationship between temperature and air conditioning electricity consumption, the expression of the model is:

[0042] AC load =α(TM-TM base )×D (7)

[0043] Among them, AC loadis the air conditioning load, α is the power sensitivity coefficient per unit temperature change, TM is the average temperature, TM base is the reference temperature for air conditioning startup, and D is the number of degree days;

[0044] The prediction model takes the prediction results of the above four models as input, adopts weighted average or ensemble learning method, considers the performance of different models in different time periods, and obtains the power purchase amount and load of the power grid enterprise's proxy power purchase users;

[0045] Before making an annual electricity purchase decision, the annual and monthly electricity consumption of the three categories of users, namely agriculture, residents and industrial and commercial agents, is predicted to obtain the total monthly electricity consumption of these three categories of users. for:

[0046]

[0047] in, Respectively represent the electricity consumption of agricultural, residential, and agent industrial and commercial users in the mth month;

[0048] Before making monthly power purchase decisions, the time-of-use power consumption of the three types of users, namely agriculture, residents and agent industry and commerce, is predicted to obtain the total power demand of these three types of users in period t. for:

[0049]

[0050] in, They represent the electricity demand of agriculture, residents, and industrial and commercial users in period t respectively.

[0051] Preferably, the risk control indicator assessment is performed based on the conditional value at risk (CVaR) according to the uncertainty factors on the power purchasing side and the power selling side, including:

[0052] Based on the uncertainty factors on the power purchase and sales sides, the Monte Carlo sampling method is used to generate power purchase and sales uncertainty scenarios. According to the defined probability distribution, random sampling is performed to generate uncertain data including power purchase and sales demand, supply and price under multiple scenarios. Based on the input of each scenario, key output indicators including power purchase cost, power sales revenue and profit are calculated. The obtained uncertain data and key output indicators are summarized into a power purchase and sales scenario database, recording the key indicator data under each scenario.

[0053] Based on the key indicator data, the conditional value at risk (CVaR) is used to measure the annual power purchase risk of power grid enterprises, characterize the average excess cost of power grid enterprises, and use the Monte Carlo sampling method to obtain the annual conditional value at risk under the confidence level β. for:

[0054]

[0055] Where N is the number of Monte Carlo samples of variable ω, C Y (·) represents the annual electricity purchase cost in each scenario, Q represents the purchase quantity decision, ω n represents the nth sample of Monte Carlo sampling, VaR represents the value at risk;

[0056] Based on the key indicator data, the conditional value at risk (CVaR) is used to measure the monthly power purchase risk of power grid enterprises, characterize the average excess cost of power grid enterprises, and use the Monte Carlo sampling method to obtain the monthly conditional value at risk under the confidence level β. for:

[0057]

[0058] Among them, C d (·) represents the monthly electricity purchase cost in each scenario.

[0059] Preferably, the annual conditional risk value is used to establish a combined transaction model for determining monthly electricity volume based on annual transaction expectations and risk control, including:

[0060] Calculate the annual market-based electricity purchase cost for:

[0061]

[0062] in, Respectively represent the monthly electricity volume of the local and inter-provincial annual medium- and long-term markets, is the corresponding price;

[0063] Calculate monthly market-based electricity purchase costs for:

[0064]

[0065] in, represents the monthly medium- and long-term market contract electricity in the mth month, is the corresponding electricity price;

[0066] Calculate inter-provincial spot market transaction costs, including:

[0067] The spot trading volume of the power grid company in the mth month for:

[0068]

[0069] in, is the total monthly electricity consumption of each type of user, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively;

[0070] When making annual electricity purchase decisions, use the average price of the inter-provincial spot market each month. Cost of conducting transactions Calculation:

[0071]

[0072] Combining equations (1), (2), (12), (13) and (15), we can obtain the total annual electricity purchase cost C of the power grid enterprise: Y for:

[0073]

[0074] The annual electricity purchase cost of the power grid enterprise is expressed as a function of the decision variable Q and the random factor ω C Y (Q,ω), and express the expected value of the cost as Among them, Q is ω is

[0075] For the decision variables of the annual model, the upper and lower limits of electricity trading are set based on the historical power purchase plans, power generation capacity and related policies of the power grid companies, including:

[0076]

[0077]

[0078] Considering the expected value and risk of electricity purchase cost, the objective function of the annual decision-making of power grid enterprises is obtained as follows:

[0079]

[0080] in, is the expected value of annual electricity purchase cost, ρ is the risk aversion factor, the larger the value, the higher the risk aversion of the power grid enterprise; is the annual conditional value at risk;

[0081] By combining equations (17) to (23), we can construct a combined trading model in which the monthly electricity consumption is determined by annual transactions, taking into account the expected value of the annual electricity purchase cost and risk control.

[0082] Preferably, the monthly conditional risk value is used to establish a combined transaction model for determining daily electricity quantity based on monthly transaction considerations of the expected value of the annual electricity purchase cost and risk control, including:

[0083] Calculate the annual market-based electricity purchase cost for:

[0084]

[0085] in, They represent the electricity consumption in the local and inter-provincial annual medium- and long-term markets during period t, respectively, and T is the number of discrete time periods;

[0086] Calculate monthly market-based electricity purchase costs for:

[0087]

[0088] in, represents the monthly medium- and long-term market contract electricity in period t, is the monthly medium and long-term market price;

[0089] Calculate inter-provincial spot market transaction costs, including:

[0090] The spot trading volume of power grid enterprises in period t It is expressed as:

[0091]

[0092] in, represents the total power demand of various users in period t, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively;

[0093] set up represents the price of the inter-provincial spot market during period t, and the daily cost of inter-provincial spot transactions is expressed as:

[0094]

[0095] Combining formulas (3, (4), (24), (25), and (27), we can obtain the monthly electricity purchase cost C of the power grid company. d for:

[0096]

[0097] The monthly electricity purchase cost of the power grid company is expressed as a function of the decision variable Q and the quantity and price random factor ω C d (Q,ω), and express the expected value of the cost as Among them, Q is ω is

[0098] For local gas turbines, inter-provincial coal-fired power and inter-provincial nuclear power formed according to annual transactions, as well as power purchase contracts formed according to local and inter-provincial annual medium- and long-term markets, the sum of the time-of-use electricity in all periods of the day is set to be equal to the monthly electricity in the contract divided evenly into daily electricity. At the same time, the output of nuclear power in each period remains unchanged, and the upper and lower limits of electricity transaction physical constraints are set as follows:

[0099]

[0100] Considering the expected and risk of electricity purchase cost, the objective function of the monthly decision-making of power grid enterprises is obtained as follows:

[0101]

[0102] in, is the mathematical expectation of the monthly electricity purchase cost, ρ is the risk aversion factor, the larger the value, the higher the risk aversion of the power grid enterprise; is the monthly conditional value at risk;

[0103] By combining equations (29) to (35), we can construct a combined trading model for daily electricity consumption determined by monthly transactions, which takes into account the expected value of annual electricity purchase cost and risk control.

[0104] According to a second aspect of the present invention, a system for constructing a medium- and long-term power purchase and sale combination transaction model for a power grid enterprise is provided, comprising:

[0105] The uncertainty modeling module on the power purchasing side is used to model the uncertain power purchasing cost on the power purchasing side for scenarios with different types of power sources on the power purchasing side, and obtain the uncertainty factors on the power purchasing side;

[0106] The uncertainty modeling module on the power sales side is used to analyze the power demand of multiple types of users on the power sales side, model the power purchase volume and load of the power purchase users on behalf of the power grid enterprises on the power sales side, and the uncertain total power consumption, so as to obtain the uncertainty factors on the power sales side;

[0107] Conditional risk assessment module, which is used to evaluate risk control indicators based on the conditional risk value CVaR according to the uncertainty factors on the power purchasing side and the power selling side, and obtain the annual and monthly conditional risk value;

[0108] Annual combination transaction modeling module, which is based on the annual conditional risk value and establishes a combination transaction model for the monthly electricity consumption determined by annual transactions taking into account the expected value of the annual electricity purchase cost and risk control;

[0109] Monthly combination transaction modeling module, based on the monthly conditional risk value, establishes a combination transaction model for daily electricity consumption determined by monthly transactions that takes into account the expected value of the annual electricity purchase cost and risk control.

[0110] According to a third aspect of the present invention, there is provided a computer terminal comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, can be used to execute any one of the methods described above in the present invention, or to run the system described above in the present invention.

[0111] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can be used to execute any of the methods described above in the present invention, or to run the system described above in the present invention.

[0112] Due to the adoption of the above technical solution, the present invention has at least one of the following beneficial effects compared with the prior art:

[0113] Uncertainty modeling on the power purchasing side: For different types of power sources such as wind power, photovoltaic power, and inter-provincial power purchases, the uncertainty cost of the power purchasing side is modeled using statistical analysis and Monte Carlo sampling. This can effectively quantify the fluctuation characteristics of various types of power sources, thereby accurately reflecting the potential uncertainty of power purchasing costs and generating a variety of scenarios for subsequent risk analysis and decision optimization. It solves the cost uncertainty problem on the power purchasing side caused by natural factors, cross-regional scheduling, and electricity price fluctuations, and improves the robustness of power purchasing decisions.

[0114] Multi-model demand forecasting on the power sales side: Multiple models such as multiple regression analysis, time series model, economic transmission model, degree-day method, etc. are used to make a combined forecast of the power demand on the power sales side. By integrating the forecast results of different models, the power demand of different users can be captured more accurately, especially the detailed forecast of the demand of residential, agricultural, industrial and commercial users, which improves the accuracy and robustness of demand forecasting, solves the problem of uncertainty in power demand caused by seasonality, economic fluctuations, etc. on the power sales side, and helps power grid companies better plan power sales strategies.

[0115] Risk control assessment based on conditional value at risk (CVaR): CVaR is used to conduct risk assessment on power purchase and sales scenarios, and the annual and monthly conditional value at risk are calculated separately. The CVaR method can capture extreme risk situations in power purchase costs, help power grid companies control the risk of exceeding budgets, and improve the safety of annual and monthly decisions. It solves the problem that traditional mean models cannot identify extreme risks, allowing power grid companies to avoid major economic losses while controlling the expected value of power purchase costs.

[0116] Annual and monthly combined transaction model: Establish an annual and monthly combined transaction model, combine the expected value of electricity purchase cost with risk aversion factors, and use it to determine the allocation of monthly and daily electricity. This multi-level transaction optimization method enables power grid companies to make detailed decisions from the perspective of the whole year to the monthly and daily levels, ensuring a balance between cost and risk in both the long and short term. It solves the problem of power grid companies' difficulty in reasonably allocating monthly and daily electricity in medium- and long-term electricity purchase and sales transactions, and improves the flexibility and responsiveness of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0118] Figure 1 The present invention is a flowchart of a method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise in a preferred embodiment of the present invention.

[0119] Figure 2 This is a schematic diagram of the components of a system for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0120] The following is a detailed description of the embodiments of the present invention: This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given. It should be pointed out that for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

[0121] An embodiment of the present invention provides a method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise. The method takes into account the uncertainty of power purchase and sales and considers risk control factors as well as medium- and long-term annual transactions and monthly transactions, thereby realizing the medium- and long-term power purchase and sales optimization of the power grid enterprise based on combination transactions.

[0122] Specifically, Figure 1 As shown, the method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise provided in this embodiment may include the following operations:

[0123] S1, for the scenarios of different types of power sources on the power purchasing side, the uncertain power purchasing cost on the power purchasing side is modeled to obtain the uncertainty factors on the power purchasing side;

[0124] S2, analyze the electricity demand of multiple types of users on the power sales side, model the power purchase volume and load of the power grid enterprise on behalf of the power sales user and the uncertain total power consumption, and obtain the uncertainty factors on the power sales side;

[0125] S3, according to the uncertainty factors on the power purchasing side and the power selling side, the risk control index is evaluated based on the conditional risk value CVaR to obtain the annual and monthly conditional risk value;

[0126] S4, based on the annual conditional risk value, establish a combined trading model for the monthly electricity consumption determined by annual transactions that considers the expected value of the annual electricity purchase cost and risk control;

[0127] S5, based on the monthly conditional value at risk, establish a combined trading model for daily electricity consumption determined by monthly transactions that considers the expected value of the annual electricity purchase cost and risk control.

[0128] The technical solution provided by the above embodiment of the present invention is further described in detail below in conjunction with the preferred implementation manner.

[0129] like Figure 2 As shown, in this preferred implementation, the method for constructing a medium- and long-term power purchase and sale combination transaction model for a power grid enterprise provided in the above embodiment includes the following steps:

[0130] S1, considering the characteristics of the structure and output of different types of power sources on the power purchasing side, including wind power, photovoltaic power, biomass, and inter-provincial power, the uncertain power purchasing cost on the power purchasing side is modeled to obtain the uncertainty factors on the power purchasing side;

[0131] S2, considering the electricity consumption characteristics, electricity consumption behaviors and other electricity demand of multiple types of users on the electricity sales side, modeling the power purchase quantity and load of the power grid enterprise on behalf of the power purchase users on the electricity sales side and the uncertain total electricity consumption, and obtaining the uncertainty factors on the electricity sales side;

[0132] S3, according to the above-mentioned uncertainties on both sides of electricity purchase and sales, the risk control index is evaluated based on the conditional risk value CVaR to obtain the annual and monthly conditional risk value;

[0133] S4, based on the annual conditional risk value, combined with the power grid company's own historical power purchase plan, power generation capacity and related policies and other comprehensive factors to determine the upper and lower limits of power trading physical constraints, to establish an annual transaction model that considers the expected value of the annual power purchase cost and risk control to determine the monthly power volume.

[0134] S5, based on the monthly conditional risk value, according to the local preferential gas turbines, inter-provincial planned coal power, inter-provincial planned nuclear power formed by annual transactions, and the local and inter-provincial annual medium- and long-term market formed power purchase contracts, with the expected value of monthly power purchase cost and risk control as the goal, the monthly electricity consumption is evenly divided into daily electricity consumption, and a combined trading model for daily electricity consumption determined by monthly transactions that considers the expected value of annual power purchase cost and risk control is established.

[0135] In some preferred implementations, the above S1, for scenarios with different types of power sources on the power purchasing side, models the uncertain power purchasing cost on the power purchasing side, and may further include the following operations:

[0136] S11, analyze the characteristics of the structure and output of various types of power sources on the power purchasing side, and clarify their characteristics in terms of uncertainty:

[0137] Wind power: It is greatly affected by natural conditions and is intermittent and unstable, so it is necessary to consider the uncertainty of wind power fluctuations on electricity prices and electricity purchase costs. Photovoltaic power generation: It is also unstable, but it is affected by sunshine intensity and seasonality, and the fluctuation of electricity purchase costs is also closely related to weather changes. Inter-provincial power purchase: It involves cross-regional power dispatching, which may involve price fluctuations in different regional power markets, transmission costs, and policy subsidies.

[0138] Through quantitative analysis of different power supply characteristics (such as volatility, peak output, etc.), a clear parameter basis is provided for subsequent cost modeling. Statistical analysis methods can be used to analyze historical data to predict possible future power supply instability and its potential impact on electricity purchase costs. In the present invention, the uncertainty variable fluctuates: the monthly power generation of wind power, photovoltaic power, and hydropower fluctuates by a certain amplitude based on historical data, and the Monte Carlo sampling method is used to generate uncertain scenarios.

[0139] S12, calculate the electricity purchase cost of local preferential power generation and inter-provincial planned power in the annual model under each scenario:

[0140]

[0141] in, Respectively represent the monthly electricity consumption of local wind power, photovoltaic power and gas turbine in the mth month, Respectively represent the corresponding prices;

[0142]

[0143] in, They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively. For the corresponding price.

[0144] Calculate the electricity purchase costs of local power generation and inter-provincial planned electricity in the monthly model under each scenario:

[0145]

[0146] in, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. is the corresponding price;

[0147]

[0148] in, They represent the planned electricity of inter-provincial coal-fired power, hydropower and nuclear power in period t, is the corresponding price, and T is the number of discrete time periods.

[0149] In some preferred implementations, the above S11, using the Monte Carlo sampling method to generate uncertain scenes, may further include the following operations:

[0150] Statistical analysis methods are used to quantitatively analyze the historical data of the output of various types of power sources on the power purchasing side to obtain the fluctuations of uncertain variables, and the Monte Carlo sampling method is used to generate uncertain scenarios for various types of power sources.

[0151] In some preferred implementations, the above S2, analyzing the electricity demand of multiple types of users on the electricity sales side, modeling the amount of electricity purchased and the load as well as the uncertain total electricity consumption of the power grid enterprise acting as the agent for the electricity purchase user on the electricity sales side, may further include the following operations:

[0152] S21, a combined forecasting model using four models of power purchase quantity and load of power purchasers on behalf of power grid enterprises, including:

[0153] Multiple regression analysis model: Use historical data (including electricity demand, GDP, industry proportion, population size, urbanization rate and energy conservation and emission reduction index, etc., for regression) to build a multiple regression model, where electricity demand is the dependent variable, and GDP, industry proportion, population size, urbanization rate and energy conservation and emission reduction index are independent variables. The model can be expressed as:

[0154] Y=β0+β1X1+β2X2+...,+β n X n +ε (5)

[0155] Among them, Y is the dependent variable, X1, X2,,,X n are independent variables, β0,β1,,,β n is the coefficient, ε is the error term;

[0156] Time series analysis model (such as ARIMA): Analyze the historical time series data of electricity demand to capture seasonality, trend and cyclical factors. Use the ARIMA model to predict future electricity demand. The model can be expressed as:

[0157] (1-φ1L-φ2L 2 -...-φ p L p )(1-L) d Yt =(1-θ1L-θ2L 2 -...θ q L q )ε t (6)

[0158] Where L is the lag operator, d is the difference order, φ and θ are the autoregressive and moving average coefficients, p and q are the corresponding orders, ε t is the error term.

[0159] Economic transmission analysis: Analyze the impact of macroeconomic factors such as economic growth and industrial structure adjustment on electricity demand. Establish an economic transmission model, link macroeconomic indicators with electricity demand, and analyze the potential impact of policy changes on electricity demand. The expression of the model is:

[0160] E t =α+β1·GDP t +β2·IPI t +β3·IR t +γ·Z t +ε t

[0161] Among them, E t is the electricity demand, α, β1, β2, β3, γ are model coefficients, GDP t is the gross domestic product, IPI t is the industrial production index, IR t is the policy interest rate, Z t Other relevant macroeconomic factors.

[0162] Residential air conditioning method (degree-day method): The degree-day method is used to estimate the residential air conditioning electricity demand. This method is based on the relationship between temperature and air conditioning electricity consumption. The model can be expressed as:

[0163] AC load =α(TM-TM base )×D (7)

[0164] Among them, AC load is the air conditioning load, α is the power sensitivity coefficient per unit temperature change, TM is the average temperature, TM base is the reference temperature for air conditioning startup, and D is the number of degree days.

[0165] Prediction model: The prediction results of the above models are used as input to build a combined model, which is then used through weighted average or ensemble learning methods to improve prediction accuracy and stability. The combined model can consider the performance of different models in different time periods, giving higher weights to models with better performance, and obtain the power purchase amount and load of power grid companies acting as agents for power purchasing users.

[0166] S22: Grid companies need to make annual electricity consumption and monthly electricity consumption forecasts for agriculture, residents, and commercial and industrial users before making annual electricity purchase decisions. The total monthly electricity consumption of the three types of users can be expressed as:

[0167]

[0168] in, Respectively represent the electricity consumption of agriculture, residents, and industrial and commercial users in the mth month.

[0169] Before making monthly power purchase decisions, the time-of-use power consumption of agriculture, residents and industrial and commercial agents is further predicted. The power demand of the three types of users in period t is expressed as:

[0170]

[0171] in, They represent the electricity demand of agriculture, residents, and industrial and commercial users in period t respectively.

[0172] In some preferred implementations, the above S3, based on the uncertainty factors on the power purchasing side and the power selling side, performs risk control index evaluation based on the conditional value at risk CVaR to obtain the annual and monthly conditional value at risk, and may further include the following operations:

[0173] S31, using Monte Carlo sampling to generate uncertain scenarios for power purchase and sales, randomly sampling from each input variable (i.e., uncertainty factors on the power purchase and sales side) according to the defined probability distribution, to generate uncertain data such as power purchase and sales demand, supply, price, etc. under multiple scenarios. Simulate a large number (thousands to tens of thousands) of scenarios and explore different possible future states through random sampling. According to the input of each scenario, calculate key output indicators such as power purchase cost, power sales revenue, and profit. Summarize the simulation results (i.e., uncertain data and key output indicators) into a power purchase and sales scenario database, record the key financial indicators (such as profit or loss) under each scenario, and provide basic data for subsequent risk analysis.

[0174] S32, using CVaR to measure the annual power purchase risk of power grid enterprises, characterizing the average excess cost of power grid enterprises, using Monte Carlo sampling, the annual CVaR calculation expression under the confidence level β is obtained as follows:

[0175]

[0176] Where N is the number of Monte Carlo samples of variable ω, C Y (·) represents the annual electricity purchase cost in each scenario, Q represents the purchase quantity decision, ω n represents the nth sample of Monte Carlo sampling, and VaR represents value at risk.

[0177] To measure the monthly power purchase risk of power grid enterprises and characterize the average excess cost of power grid enterprises, the monthly CVaR calculation expression under the confidence level β is obtained by Monte Carlo sampling:

[0178]

[0179] Among them, C d (·) represents the monthly electricity purchase cost in each scenario.

[0180] In some preferred implementations, the above S4, based on the annual conditional risk value, establishes a combined transaction model for determining monthly electricity volume by annual transaction considering the expected value of annual electricity purchase cost and risk control, and may further include the following operations:

[0181] S41, calculate the annual market-based electricity purchase cost:

[0182]

[0183] in, Respectively represent the monthly electricity volume of the local and inter-provincial annual medium- and long-term markets, For the corresponding price.

[0184] Calculate the monthly market-based electricity purchase cost:

[0185]

[0186] in, represents the monthly medium- and long-term market contract electricity in the mth month, is the corresponding electricity price.

[0187] Calculate inter-provincial spot market transaction costs:

[0188] The spot trading volume of the power grid company in the mth month can be expressed as:

[0189] The spot trading volume of the power grid company in the mth month for:

[0190]

[0191] in, is the total monthly electricity consumption of each type of user, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively.

[0192] Since it is difficult to accurately predict the time-based fluctuations in electricity consumption and electricity prices when making annual power purchase decisions, the average price of the inter-provincial spot market each month is used. Calculate transaction costs:

[0193]

[0194] Combining formulas (1), (2), (12), (13), and (15), we can obtain the total annual electricity purchase cost of power grid enterprises:

[0195]

[0196] S42, combining the annual power purchase cost formula, trading power constraints, and expectation and risk comprehensive decision-making objectives to obtain the final annual combination trading model, including:

[0197] The annual electricity purchase cost of the power grid enterprise is expressed as the decision variable Q (respectively ) and the random factor ω Function C Y (Q,ω), and express the expected value of the cost as

[0198] For the decision variables of the annual model, there are physical constraints on the upper and lower limits of electricity transactions, which are usually determined based on the grid company's own historical electricity purchase plans, power generation capacity and related policies.

[0199]

[0200] Considering the expected value and risk of power purchase cost, the objective function of Shanghai Power Grid's annual decision-making is:

[0201]

[0202] in, is the expected value of annual electricity purchase cost, ρ is the risk aversion factor, the larger the value, the higher the risk aversion of the power grid enterprise; is the annual conditional value at risk;

[0203] Finally, the combined equations (17) to (23) are used to construct a combined trading model in which the monthly electricity consumption is determined by annual transactions, taking into account the expected value of the annual electricity purchase cost and risk control.

[0204] In some preferred implementations, the above S5, based on the monthly conditional risk value, establishes a combined transaction model for determining daily electricity volume for monthly transactions that considers the expected value of the annual electricity purchase cost and risk control, and may further include the following operations:

[0205] S51, calculate the annual market-based electricity purchase cost:

[0206]

[0207] in, They represent the electricity volume in period t of the local and inter-provincial annual medium- and long-term markets respectively, and T is the number of discrete time periods.

[0208] Calculate the monthly market-based electricity purchase cost:

[0209]

[0210] in, represents the monthly medium- and long-term market contract electricity in period t, It is the monthly medium and long-term market price.

[0211] Calculate inter-provincial spot market transaction costs:

[0212] The spot trading electricity volume of the power grid company in period t is expressed as:

[0213]

[0214] in, represents the total power demand of various users in period t, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively;

[0215] use represents the price of the inter-provincial spot market during period t, and the daily cost of inter-provincial spot transactions is expressed as:

[0216]

[0217] Combining formulas (3, (4), (24), (25), and (27), we can obtain the monthly electricity purchase cost of power grid enterprises as follows:

[0218]

[0219] S52, combining the monthly electricity purchase cost formula, transaction power constraints, and expectation and risk comprehensive decision-making objectives to obtain the final annual portfolio transaction model, including:

[0220] The monthly electricity purchase cost of the power grid enterprise is expressed as the decision variable Q (respectively ) and the quantity-price stochastic factor ω Function C d (Q,ω), and express the expected value of the cost as

[0221] For local gas turbines, inter-provincial coal power, inter-provincial nuclear power formed by annual transactions, and power purchase contracts formed in the local and inter-provincial annual medium- and long-term markets, it is necessary to meet the requirement that the sum of the time-sharing electricity in all periods of the day is equal to the monthly electricity in the contract divided evenly into daily electricity. In addition, due to its own power generation characteristics, it is believed that the output of nuclear power in each period remains unchanged.

[0222]

[0223] Considering the expected and risk of electricity purchase cost, the objective function of the monthly decision-making of Shanghai Power Grid Company is:

[0224]

[0225] in, is the mathematical expectation of the monthly electricity purchase cost, ρ is the risk aversion factor, the larger the value, the higher the risk aversion of the power grid enterprise; is the monthly conditional value at risk.

[0226] Finally, the combined equations (29) to (35) are used to construct a combined trading model for daily electricity consumption determined by monthly transactions, which takes into account the expected value of the annual electricity purchase cost and risk control.

[0227] Based on the same inventive concept, an embodiment of the present invention also provides a system for constructing a medium- and long-term power purchase and sales combination transaction model for power grid enterprises.

[0228] Specifically, Figure 2 As shown, the system for constructing a medium- and long-term power purchase and sale combination transaction model for a power grid enterprise provided in this embodiment may include the following modules:

[0229] The uncertainty modeling module on the power purchasing side is used to model the uncertain power purchasing cost on the power purchasing side for scenarios with different types of power sources on the power purchasing side, and obtain the uncertainty factors on the power purchasing side;

[0230] The uncertainty modeling module on the power sales side is used to analyze the power demand of multiple types of users on the power sales side, model the power purchase volume and load of the power purchase users on behalf of the power grid enterprises on the power sales side, and the uncertain total power consumption, so as to obtain the uncertainty factors on the power sales side;

[0231] Conditional risk assessment module, which is used to evaluate risk control indicators based on the conditional risk value CVaR according to the uncertainty factors on the power purchasing side and the power selling side, and obtain the annual and monthly conditional risk value;

[0232] Annual combination transaction modeling module, which is based on the annual conditional risk value and establishes a combination transaction model for the monthly electricity consumption determined by annual transactions taking into account the expected value of the annual electricity purchase cost and risk control;

[0233] Monthly combination transaction modeling module, based on the monthly conditional risk value, establishes a combination transaction model for daily electricity consumption determined by monthly transactions that takes into account the expected value of the annual electricity purchase cost and risk control.

[0234] It should be noted that the steps in the method provided by the present invention can be implemented by using corresponding modules, devices, units, etc. in the system. Those skilled in the art can refer to the technical solution of the method to realize the composition of the system, that is, the embodiments in the method can be understood as preferred examples for constructing the system, which will not be elaborated here.

[0235] An embodiment of the present invention further provides a computer terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the processor can be used to execute any method of the above-mentioned embodiments of the present invention, or to run any system of the above-mentioned embodiments of the present invention.

[0236] Optionally, the memory is used to store programs; the memory may include volatile memory (English: volatile memory), such as random-access memory (English: random-access memory, abbreviated: RAM), such as static random-access memory (English: static random-access memory, abbreviated: SRAM), double data rate synchronous dynamic random access memory (English: Double Data Rate Synchronous Dynamic Random Access Memory, abbreviated: DDR SDRAM), etc.; the memory may also include non-volatile memory (English: non-volatile memory), such as flash memory (English: flash memory). The memory is used to store computer programs (such as applications, functional modules, etc. that implement the above method), computer instructions, etc., and the above computer programs, computer instructions, etc. can be partitioned and stored in one or more memories. And the above computer programs, computer instructions, data, etc. can be called by the processor.

[0237] The above-mentioned computer programs, computer instructions, etc. may be stored in one or more memories in partitions, and the above-mentioned computer programs, computer instructions, data, etc. may be called by a processor.

[0238] The processor is used to execute the computer program stored in the memory to implement the various steps of the method or various modules of the system involved in the above embodiments. For details, please refer to the relevant descriptions in the above method and system embodiments.

[0239] The processor and the memory may be independent structures or integrated structures. When the processor and the memory are independent structures, the memory and the processor may be coupled and connected via a bus.

[0240] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it can be used to execute any method of the above embodiments of the present invention, or to run any system of the above embodiments of the present invention.

[0241] Among them, computer-readable media include computer storage media and communication media, wherein the communication media include any media that facilitates the transmission of computer programs from one place to another. The storage medium can be any available medium that can be accessed by a general or special-purpose computer. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a user device. Of course, the processor and the storage medium can also be present in a communication device as discrete components.

[0242] The method and system for constructing a medium- and long-term power purchase and sales combination transaction model for power grid enterprises provided in the above embodiments of the present invention can comprehensively identify and evaluate the impact of uncertainty in power purchase and sales by modeling various uncertain factors on the power purchase and sales sides, including the uncertainty of power supply from different power sources and the uncertainty of power demand from users. Risk assessment is performed based on the conditional value at risk (CVaR) method, and the annual and monthly power purchase risk values ​​are calculated respectively, which helps power grid enterprises control risks and improve stability and economy when making power purchase and sales decisions. An annual and monthly combined transaction model is established to optimize power transaction decisions by balancing the expected value and risk of power purchase costs. The annual model is used to determine the monthly power allocation, and the monthly model is refined to the daily power allocation to improve the level of refinement of transaction decisions. By calculating and combining the annual and monthly power purchase costs, and comprehensively considering multiple factors such as market-based power purchase costs and spot market transaction costs, power grid enterprises can achieve a balance between cost control and risk avoidance and improve overall benefits.

[0243] All matters not covered in the above embodiments of the present invention are well known in the art.

[0244] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art may make various modifications or variations within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. A method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise, characterized in that: include: For scenarios with different types of power sources on the power purchasing side, the uncertain power purchasing cost on the power purchasing side is modeled to obtain the uncertainty factors on the power purchasing side; Analyze the electricity demand of multiple types of users on the power sales side, model the power purchase volume and load of power purchase users on behalf of power grid enterprises on the power sales side, and the uncertain total power consumption, and obtain the uncertainty factors on the power sales side; According to the uncertainty factors on the power purchasing and selling sides, the risk control index is evaluated based on the conditional risk value CVaR to obtain the annual and monthly conditional risk value; Based on the annual conditional value at risk, a combined trading model for monthly electricity consumption is established, which considers the expected value of the annual electricity purchase cost and risk control. Based on the monthly conditional value at risk, a combined trading model for daily electricity consumption determined by monthly transactions is established, which takes into account the expected value of the annual electricity purchase cost and risk control.

2. The method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise according to claim 1 is characterized in that: The uncertain electricity purchasing cost of the electricity purchasing side is modeled for the scenarios of different types of power sources on the electricity purchasing side, including: Clarify the uncertainty of power sources of various transaction products on the power purchasing side and generate uncertain scenarios of power sources of various transaction products; For the uncertain scenarios of power sources of various trading varieties, the purchase cost of electricity for local preferential power generation and inter-provincial planned power generation is calculated. and Conduct annual modeling: in, Respectively represent the monthly electricity consumption of local wind power, photovoltaic power and gas turbine in the mth month, Respectively represent the corresponding prices; in, They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively. is the corresponding price; For the uncertain scenarios of power sources of various trading varieties, the purchase cost of electricity for local preferential power generation and inter-provincial planned power generation is calculated. and To perform monthly modeling: in, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. is the corresponding price, T is the number of discrete time periods; in, They represent the planned electricity of inter-provincial coal-fired power, hydropower and nuclear power in period t, For the corresponding price.

3. The method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise according to claim 2 is characterized in that: The uncertainties of the power sources of various transaction products on the power purchasing side are clarified, and uncertain scenarios of the power sources of various transaction products are generated, including: The various types of power sources traded on the power purchasing side include: wind power, photovoltaic power generation and inter-provincial power purchase; among which: The wind power mentioned above takes into account the uncertain impact of wind power fluctuations on electricity prices and electricity purchase costs; The photovoltaic power generation mentioned above takes into account the uncertain impact of weather changes on electricity prices and electricity purchase costs; The cross-provincial power purchase mentioned above takes into account the price fluctuations of power markets in different regions, transmission costs, and the uncertain impact of policy subsidies on electricity prices and power purchase costs; Statistical analysis methods are used to quantitatively analyze the historical data of the output of various types of power sources on the power purchasing side to obtain the fluctuations of uncertain variables, and the Monte Carlo sampling method is used to generate uncertain scenarios for various types of power sources.

4. The method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise according to claim 1 is characterized in that: The analysis of the electricity demand of multiple types of users on the power sales side models the power purchase quantity and load of the power purchase users on behalf of the power grid enterprises on the power sales side and the uncertain total power consumption, and obtains the uncertainty factors on the power sales side, including: A combined prediction model for the amount of electricity purchased and the load of electricity purchased by a power grid enterprise as an agent is provided, wherein the combined prediction model comprises: Multiple regression analysis model, which uses historical data to establish a multiple regression model, in which electricity demand is used as the dependent variable, and gross domestic product, industrial proportion, population size, urbanization rate and energy conservation and emission reduction index are used as independent variables. The expression of the model is: Y=β0+β1X1+β2X2+...,+β n X n +e (5) Among them, Y is the dependent variable, X1, X2,,,X n are independent variables, β0,β1,,,β n is the coefficient, ε is the error term; Time series analysis model, which is used to analyze the historical time series data of electricity demand to capture seasonal, trend and cyclical factors, and then predict future electricity demand. The expression of this model is: (1-φ1L-φ2L 2 -...-f p L p (1-L) d Y t =(1-θ1L-θ2L 2 -...i q L q )e t (6) Where L is the lag operator, d is the difference order, φ and θ are the autoregressive and moving average coefficients, p and q are the corresponding orders, ε t is the error term; Economic transmission analysis model, which is used to analyze the impact of macroeconomic factors on electricity demand. The model links macroeconomic indicators with electricity demand and analyzes the potential impact of policy changes on electricity demand. The expression of the model is: E t =α+β1·GDP t +β2·IPI t +β3·IR t +γ·Z t +e t Among them, E t is the electricity demand, α, β1, β2, β3, γ are model coefficients, GDP t is the gross domestic product, IPI t is the industrial production index, IR t is the policy interest rate, Z t for other relevant macroeconomic factors; Residential air conditioning model, which uses the degree-day method to estimate the residential air conditioning electricity demand. Based on the relationship between temperature and air conditioning electricity consumption, the expression of the model is: AC load =α(TM-TM base )×D(7) Among them, AC load is the air conditioning load, α is the power sensitivity coefficient per unit temperature change, TM is the average temperature, TM base is the reference temperature for air conditioning startup, and D is the number of degree days; The prediction model takes the prediction results of the above four models as input, adopts weighted average or ensemble learning method, considers the performance of different models in different time periods, and obtains the power purchase amount and load of the power grid enterprise's proxy power purchase users; Before making an annual electricity purchase decision, the annual and monthly electricity consumption of the three categories of users, namely agriculture, residents and industrial and commercial agents, is predicted to obtain the total monthly electricity consumption of these three categories of users. for: in, Respectively represent the electricity consumption of agricultural, residential, and agent industrial and commercial users in the mth month; Before making monthly power purchase decisions, the time-of-use power consumption of the three types of users, namely agriculture, residents and agent industry and commerce, is predicted to obtain the total power demand of these three types of users in period t. for: in, They represent the electricity demand of agriculture, residents, and industrial and commercial users in period t respectively.

5. The method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise according to claim 1 is characterized in that: According to the uncertainty factors on the power purchasing and selling sides, the risk control index assessment is carried out based on the conditional value at risk CVaR, including: Based on the uncertainty factors on the power purchase and sales sides, the Monte Carlo sampling method is used to generate power purchase and sales uncertainty scenarios. According to the defined probability distribution, random sampling is performed to generate uncertain data including power purchase and sales demand, supply and price under multiple scenarios. Based on the input of each scenario, key output indicators including power purchase cost, power sales revenue and profit are calculated. The obtained uncertain data and key output indicators are summarized into a power purchase and sales scenario database, recording the key indicator data under each scenario. Based on the key indicator data, the conditional value at risk (CVaR) is used to measure the annual power purchase risk of power grid enterprises, characterize the average excess cost of power grid enterprises, and use the Monte Carlo sampling method to obtain the annual conditional value at risk under the confidence level β. for: Where N is the number of Monte Carlo samples of variable ω, C Y (·) represents the annual electricity purchase cost in each scenario, Q represents the purchase quantity decision, ω n represents the nth sample of Monte Carlo sampling, VaR represents the value at risk; Based on the key indicator data, the conditional value at risk (CVaR) is used to measure the monthly power purchase risk of power grid enterprises, characterize the average excess cost of power grid enterprises, and use the Monte Carlo sampling method to obtain the monthly conditional value at risk under the confidence level β. for: Among them, C d (·) represents the monthly electricity purchase cost in each scenario.

6. The method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise according to claim 1, characterized in that: The combined transaction model for determining monthly electricity volume based on the annual conditional risk value and annual transaction considering the expected value of annual electricity purchase cost and risk control is established, including: Calculate the annual market-based electricity purchase cost for: in, Respectively represent the monthly electricity volume of the local and inter-provincial annual medium- and long-term markets, is the corresponding price; Calculate monthly market-based electricity purchase costs for: in, represents the monthly medium- and long-term market contract electricity in the mth month, is the corresponding electricity price; Calculate inter-provincial spot market transaction costs, including: The spot trading volume of the power grid company in the mth month for: in, is the total monthly electricity consumption of each type of user, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively; When making annual electricity purchase decisions, use the average price of the inter-provincial spot market each month. Cost of conducting transactions Calculation: Combining equations (1), (2), (12), (13) and (15), we can obtain the total annual electricity purchase cost C of the power grid enterprise: Y for: The annual electricity purchase cost of the power grid enterprise is expressed as a function of the decision variable Q and the random factor ω C Y (Q,ω), and express the expected value of the cost as Among them, Q is ω is For the decision variables of the annual model, the upper and lower limits of electricity trading are set based on the historical power purchase plans, power generation capacity and related policies of the power grid companies, including: Considering the expected value and risk of electricity purchase cost, the objective function of the annual decision-making of power grid enterprises is obtained as follows: in, is the expected value of annual electricity purchase cost, ρ is the risk aversion factor, the larger the value, the higher the risk aversion of the power grid enterprise; is the annual conditional value at risk; By combining equations (17) to (23), we can construct a combined trading model in which the monthly electricity consumption is determined by annual transactions, taking into account the expected value of the annual electricity purchase cost and risk control.

7. The method for constructing a medium- and long-term power purchase and sales combination transaction model for a power grid enterprise according to claim 1, characterized in that: Based on the monthly conditional risk value, a combined trading model for determining daily electricity volume based on monthly trading that considers the expected value of the annual electricity purchase cost and risk control is established, including: Calculate the annual market-based electricity purchase cost for: in, They represent the electricity consumption in the local and inter-provincial annual medium- and long-term markets during period t, respectively, and T is the number of discrete time periods; Calculate monthly market-based electricity purchase costs for: in, represents the monthly medium- and long-term market contract electricity in period t, is the monthly medium and long-term market price; Calculate inter-provincial spot market transaction costs, including: The spot trading volume of power grid enterprises in period t It is expressed as: in, represents the total power demand of each type of user in time period t, They represent the power of local wind power, photovoltaic power and gas turbine in period t respectively. They represent the monthly electricity of inter-provincial coal-fired power, hydropower and nuclear power in the mth month respectively; set up represents the price of the inter-provincial spot market during period t, and the daily cost of inter-provincial spot transactions is expressed as: Combining formulas (3, (4), (24), (25), and (27), we can obtain the monthly electricity purchase cost C of the power grid company. d for: The monthly electricity purchase cost of the power grid company is expressed as a function of the decision variable Q and the quantity and price random factor ω C d (Q,ω), and express the expected value of the cost as Among them, Q is ω is For local gas turbines, inter-provincial coal-fired power and inter-provincial nuclear power formed according to annual transactions, as well as power purchase contracts formed according to local and inter-provincial annual medium- and long-term markets, the sum of the time-of-use electricity in all periods of the day is set to be equal to the monthly electricity in the contract divided evenly into daily electricity. At the same time, the output of nuclear power in each period remains unchanged, and the upper and lower limits of electricity transaction physical constraints are set as follows: Considering the expected and risk of electricity purchase cost, the objective function of the monthly decision of the power grid enterprise is: in, is the mathematical expectation of the monthly electricity purchase cost, ρ is the risk aversion factor, the larger the value, the higher the risk aversion of the power grid enterprise; is the monthly conditional value at risk; By combining equations (29) to (35), we can construct a combined trading model for daily electricity consumption determined by monthly transactions, which takes into account the expected value of the annual electricity purchase cost and risk control.

8. A system for constructing a medium- and long-term power purchase and sales combination transaction model for power grid enterprises, characterized in that: include: The uncertainty modeling module on the power purchasing side is used to model the uncertain power purchasing cost on the power purchasing side for scenarios with different types of power sources on the power purchasing side, and obtain the uncertainty factors on the power purchasing side; The uncertainty modeling module on the power sales side is used to analyze the power demand of multiple types of users on the power sales side, model the power purchase volume and load of the power purchase users on behalf of the power grid enterprises on the power sales side, and the uncertain total power consumption, so as to obtain the uncertainty factors on the power sales side; Conditional risk assessment module, which is used to evaluate risk control indicators based on the conditional risk value CVaR according to the uncertainty factors on the power purchasing side and the power selling side, and obtain the annual and monthly conditional risk value; Annual combination transaction modeling module, which is based on the annual conditional risk value and establishes a combination transaction model for monthly electricity consumption determined by annual transactions taking into account the expected value of the annual electricity purchase cost and risk control; Monthly combination transaction modeling module, based on the monthly conditional risk value, establishes a combination transaction model for daily electricity consumption determined by monthly transactions that takes into account the expected value of the annual electricity purchase cost and risk control.

9. A computer terminal comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it can be used to perform the method described in any one of claims 1 to 7, or to run the system described in claim 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it can be used to perform the method described in any one of claims 1 to 7, or to run the system described in claim 8.

Images