Real-time cost calculation and quotation system for gas turbine unit in spot market

Abstract

The invention discloses a real-time cost calculation and quotation system for a gas turbine unit in a spot market, and the system comprises a real-time cost calculation module which is used for calculating the real-time unit cost according to the actual operation condition of the gas turbine unit; the quotation optimization module is used for starting to calculate a unit quotation strategy by taking a measurement and calculation result as a marginal condition after real-time unit cost calculation is completed, and optimizing the quotation strategy of the gas unit in the spot market by using multiple algorithms; and the market settlement analysis module is used for analyzing the quotation strategies in a simulated market environment after the multiple quotation strategies of the unit are obtained, and evaluating the settlement income and economic benefits of the gas unit, thereby evaluating the effectiveness of the quotation strategies. By measuring and calculating the real-time cost, optimizing the quotation strategy and analyzing the settlement rule, the method helps the gas turbine unit to realize benefit maximization in the electric power spot market, and improves the economical efficiency and stability of power grid operation at the same time.

Description

Technical Field

[0001] The invention relates to the technical field of power market economic analysis and simulation, and in particular to a real-time cost estimation and quotation system for gas generator sets oriented to the spot market. Background Art

[0002] Gas-fired units have gradually become an important part of the power market due to their rapid startup, flexible peak load regulation, and environmental friendliness. However, in the spot market, the competitiveness of gas-fired units is restricted due to their high unit power generation costs. How to reasonably calculate costs, optimize quotations, and evaluate settlement benefits have become key issues in improving the market competitiveness of gas-fired units. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a real-time cost calculation and quotation system for gas turbines in the spot market. By calculating real-time costs, optimizing quotation strategies and analyzing settlement rules, it helps gas turbines to maximize their benefits in the electricity spot market, while improving the economy and stability of power grid operation.

[0004] To achieve the above purpose, the technical solution provided by the present invention is: a real-time cost calculation and quotation system for gas units in the spot market, comprising:

[0005] Real-time cost calculation module, used to calculate the real-time unit cost according to the actual operation of the gas unit;

[0006] The quotation optimization module is used to calculate the unit cost in real time, use the calculated result as the marginal condition, start calculating the unit quotation strategy, and use multiple algorithms to optimize the quotation strategy of the gas unit in the spot market;

[0007] The market settlement analysis module is used to analyze various quotation strategies of the unit after obtaining them in a simulated market environment, evaluate the settlement income and economic benefits of the gas unit, and thus evaluate the effectiveness of the quotation strategy.

[0008] Furthermore, the real-time cost calculation module includes a fixed cost calculation submodule, a variable cost calculation submodule and a startup cost calculation submodule, wherein:

[0009] The fixed cost calculation submodule calculates the unit fixed cost C gd It includes depreciation, financial expenses, employee salaries, material costs and repair costs, as follows:

[0010] The depreciation is calculated by using the comprehensive depreciation rate allocation method, that is, the depreciation of the original value of fixed assets is allocated according to the comprehensive depreciation rate. The calculation formula is as follows:

[0011]

[0012] Where: C zj is the real-time cost per unit of depreciation; OV u is the original value of fixed assets of each unit; R f is the fixed cost allocation ratio; R r is the estimated residual value rate of fixed assets; t is the useful life of the unit; D i This is the depreciation cost of infrastructure construction this month; 0 The amount of electricity used for Internet access on that day;

[0013] The financial expenses refer to the expenses incurred by an enterprise to raise funds for production and operation. The financial expenses are calculated using the trend interpolation method, which is a method of calculating the actual value of the financial expenses of the previous month combined with the forecast value of the financial expenses of this month. The calculation formula is as follows:

[0014]

[0015] Where: C cw is the real-time cost of unit financial expenses; MA cw MB is the financial expenses of the previous month; cw Budget financial expenses for this month; D im The number of days that have occurred this month; D m is the total number of days in this month; D lm is the total number of days in the previous month;

[0016] The employee salary is calculated using the annual budget allocation method, that is, the budget for the year is allocated to each day for calculation. The calculation formula is as follows:

[0017]

[0018] Where: C zgxc The real-time cost of employee salary of the unit; zgxc Budgeted employee salaries for the current year;

[0019] The calculation method of the material cost and repair cost is the same as that of employee salary. It can be calculated by replacing the employee salary budget for the current year in formula (3) with the corresponding cost;

[0020] The unit variable cost C calculated by the variable cost calculation submodule bd It includes gas cost and external power cost, which is allocated based on the real-time data of gas price, external power cost and variable material cost, and is allocated based on the unit grid-connected power consumption. The specific situation is as follows:

[0021] The gas cost is calculated based on the measured energy consumption data of the unit. First, the quadratic function relationship between the total energy consumption of the unit and the power output level is fitted according to the least squares method. Then, the total power supply energy consumption of the unit is calculated, that is, the total power supply energy consumption-the total power generation energy consumption / (1-plant power consumption rate). According to the total power supply energy consumption, the no-load gas consumption of the unit and the marginal gas consumption at different output levels are calculated. Finally, the no-load gas cost and marginal gas cost are calculated in combination with the gas price. The no-load cost caused by non-dispatch reasons is included in the real-time cost. The calculation formula is:

[0022]

[0023] Where: C f is the unit gas cost; P is the power consumption rate; f is the comprehensive gas unit price; b x is the power consumption rate of power generation; ap is the power consumption rate of the plant;

[0024] The purchased power fee refers to the fee paid for purchasing electricity from the power grid to ensure the safe and stable operation of the unit and the fee paid for purchasing electricity or power generation indicators for electricity transfer. It is calculated using the trend extrapolation method. The calculation formula is as follows:

[0025]

[0026] Where: C wg The real-time cost per unit of purchased power; MA wg The actual value of the purchased power cost last month; R v MB is the variable cost allocation ratio; wg This is the budget value of outsourced power costs for this month;

[0027] The startup cost calculation submodule dynamically calculates the unit startup cost C according to the unit startup parameters and loss rate. qd , the calculation formula is as follows:

[0028]

[0029] Where: C qd is the total startup cost; C qdrl Fuel cost for startup; C zs Q is the depreciation fee for unit startup; r The amount of fuel consumed for startup; P r The fuel price consumed to start; UI js I is the unit investment cost; zs R is the loss rate of life due to unit startup; f is the unit capacity allocation ratio; T′ is the monthly allocation ratio.

[0030] Furthermore, the real-time cost calculation module adds the measured unit fixed cost, unit variable cost and unit startup cost to obtain the real-time unit cost C, and the calculation formula is as follows:

[0031] C=C gd +C bd +C qd (8).

[0032] Furthermore, the quotation optimization module includes a quotation strategy submodule based on cost analysis, a quotation optimization submodule based on game theory, and a dynamic optimization submodule based on reinforcement learning, wherein:

[0033] The cost analysis-based quotation strategy submodule sets the quotation based on the unit real-time cost, and its quotation model is:

[0034]

[0035] Where: T is the number of quotation periods per day, t = 1, 2, 3, ... T; M is the number of units operating in the power plant during the day; c ti is the fuel cost per unit of electricity generated by the i-th unit in the t-th period; q ti is the output of the i-th unit in the t-th period; Q t is the power output reported by the power plant during period t; Q net-t is the power demand of the power grid in period t; Q in-max-t is the maximum injected power of the power plant as a node in the power grid during period t; p t is the electricity price reported by the power plant in period t; p t_max is the maximum price limit of the power grid during period t; Δq i is the output increase of the i-th unit in period t; Δq i_max , Δq i_min are the maximum and minimum output increases of the i-th unit in a certain period of time; q i is the output of the i-th unit in period t; G i_max , G i_min are the maximum and minimum output of the i-th unit respectively;

[0036] The game theory-based quotation optimization submodule solves the optimal quotation through Nash equilibrium, including the following steps:

[0037] 1) First set the cost function and quotation function of each unit;

[0038] 2) Taking the cost function and quotation function of the unit as marginal conditions and the marginal income of each unit as the objective function, a Nash equilibrium equation group including the market clearing electricity price and the unit power generation is constructed and solved. At this time, the market clearing process is to solve the following equation group:

[0039]

[0040] Where: p j is the active power generated by the jth unit; u j 、v j is the strategic quotation coefficient of the j-th unit; P T is the system load curve; P 0 is the original load without considering load elasticity; k is the load elasticity coefficient; λ is the final clearing price of the market, whose trajectory is formed by the quotation curves of all power plants; λ j is the bid curve of the jth power plant; the optimal bids of each power plant are related to each other, and finally a mutually acceptable equilibrium solution will be obtained, which is the Nash equilibrium point in game theory;

[0041] The unit parameters obtained at the Nash equilibrium point are the optimal strategy coefficients under equation (10);

[0042] 3) According to the optimal strategy coefficient obtained, adjust the unit's quotation to maximize profit;

[0043] The dynamic optimization submodule based on reinforcement learning, which adaptively adjusts the quotation in combination with real-time market data, includes the following steps:

[0044] 1) In the actual market, the power sales company purchases the grid-connected power of all power generators that have successfully applied for it according to the predicted system load, and then sells it to users in a unified manner; the power trading center is responsible for the operation and management of the power generation market and the dispatching and operation of the power system. It conducts research on the day-ahead market for power trading and requires power generators to submit an available power supply and quotation for the next period every hour; the predicted load adopts the preset scenario of the quotation strategy submodule;

[0045] 2) The bidding function of the generator sets in the system is a linear bidding function. The clearing process sorts the bids of the generator sets from low to high. According to the load demand set in the model scenario, the generators with low bids are given priority to be connected to the grid for power generation. Finally, the unified market clearing price calculated is used for settlement.

[0046] Further, the market settlement analysis module uses the following constraints: power balance constraint, voltage constraint, line capacity constraint, section constraint, maximum and minimum power constraint of generator set, minimum start-up time constraint, ramp constraint and fuel supply constraint;

[0047] The analysis methods used by the market settlement analysis module include:

[0048] ①Descriptive statistical analysis:

[0049] This includes calculating the mean and standard deviation to understand the center position and dispersion of the data; calculating the maximum and minimum values, identifying the extreme values ​​of prices, and understanding the range of price fluctuations;

[0050] ②Trend analysis:

[0051] Conduct time series analysis to observe price trends over time, understand how prices change over a specific period of time, and whether there is a clear upward or downward trend;

[0052] ③Comparative analysis:

[0053] Horizontal comparison: compare the average prices of different quarters or months to analyze the differences in the time dimension;

[0054] Vertical comparison: compare the price changes of the same month in different time periods, for example, compare the price changes of each quarter in 2025;

[0055] ④ANOVA:

[0056] Check whether there are significant differences in prices over different time periods;

[0057] ⑤Regression analysis:

[0058] If you need to predict future prices or analyze the relationship between prices and other factors, you can use regression models for modeling and analysis;

[0059] ⑥Graphical display:

[0060] Using charts to visually display the changing trends and distribution characteristics of price data helps to intuitively understand the data patterns. The charts include line charts, bar charts or box plots.

[0061] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0062] 1. The present invention proposes for the first time a method for completing a quotation for a gas generator set based on the real-time cost of the gas generator set.

[0063] 2. The present invention can efficiently complete the complex process involving unit cost analysis, quotation strategy selection, and market scenario design, saving calculation time.

[0064] 3. The present invention provides different market scenarios when providing gas unit quotation schemes, and different quotation strategies can be selected in different market scenarios.

[0065] 4. The present invention can provide a gas unit quotation scheme that meets the requirements according to the requirements of enterprises and institutions, and has high application value and broad prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1Schematic diagram of the system of the present invention. DETAILED DESCRIPTION

[0067] The present invention is further described in detail below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0068] like Figure 1 As shown, this embodiment discloses a real-time cost calculation and quotation system for gas generator sets in the spot market, including the following functional modules:

[0069] Real-time cost calculation module, used to calculate the real-time unit cost according to the actual operation of the gas unit;

[0070] The quotation optimization module is used to calculate the unit cost in real time, use the calculated result as the marginal condition, start calculating the unit quotation strategy, and use multiple algorithms to optimize the quotation strategy of the gas unit in the spot market;

[0071] The market settlement analysis module is used to analyze various quotation strategies of the unit after obtaining them in a simulated market environment, evaluate the settlement income and economic benefits of the gas unit, and thus evaluate the effectiveness of the quotation strategy.

[0072] Specifically, the real-time cost calculation module includes a fixed cost calculation submodule, a variable cost calculation submodule and a startup cost calculation submodule, wherein:

[0073] The fixed cost calculation submodule calculates the unit fixed cost C gd It includes depreciation, financial expenses, employee salaries, material costs and repair costs, as follows:

[0074] The depreciation is calculated by using the comprehensive depreciation rate allocation method, that is, the depreciation of the original value of fixed assets is allocated according to the comprehensive depreciation rate. The calculation formula is as follows:

[0075]

[0076] Where: C zj is the real-time cost per unit of depreciation; OV u is the original value of fixed assets of each unit; R f is the fixed cost allocation ratio; R r is the estimated residual value rate of fixed assets; t is the useful life of the unit; D i This is the depreciation cost of infrastructure construction this month; 0 The amount of electricity used for Internet access on that day;

[0077] The financial expenses refer to the expenses incurred by an enterprise to raise funds for production and operation. The financial expenses are calculated using the trend interpolation method, which is a method of calculating the actual value of the financial expenses of the previous month combined with the forecast value of the financial expenses of this month. The calculation formula is as follows:

[0078]

[0079] Where: C cw is the real-time cost of unit financial expenses; MA cw MB is the financial expenses of the previous month; cw Budget financial expenses for this month; D im The number of days that have occurred this month; D m is the total number of days in this month; D lm is the total number of days in the previous month;

[0080] The employee salary is calculated using the annual budget allocation method, that is, the budget for the year is allocated to each day for calculation. The calculation formula is as follows:

[0081]

[0082] Where: C zgxc The real-time cost of employee salary of the unit; zgxc Budgeted employee salaries for the current year;

[0083] The calculation method of the material cost and repair cost is the same as that of employee salary. It can be calculated by replacing the employee salary budget for the current year in formula (3) with the corresponding cost;

[0084] The unit variable cost C calculated by the variable cost calculation submodule bd It includes gas cost and external power cost, which is allocated based on the real-time data of gas price, external power cost and variable material cost, and is allocated based on the unit grid-connected power consumption. The specific situation is as follows:

[0085] The gas cost is calculated based on the measured energy consumption data of the unit. First, the quadratic function relationship between the total energy consumption of the unit and the power output level is fitted according to the least squares method. Then, the total power supply energy consumption of the unit is calculated, that is, the total power supply energy consumption-the total power generation energy consumption / (1-plant power consumption rate). According to the total power supply energy consumption, the no-load gas consumption of the unit and the marginal gas consumption at different output levels are calculated. Finally, the no-load gas cost and marginal gas cost are calculated in combination with the gas price. The no-load cost caused by non-dispatch reasons is included in the real-time cost. The calculation formula is:

[0086]

[0087] Where: C f is the unit gas cost; P is the power consumption rate; fis the comprehensive gas unit price; b x is the power consumption rate of power generation; ap is the power consumption rate of the plant;

[0088] The purchased power fee refers to the fee paid for purchasing electricity from the power grid to ensure the safe and stable operation of the unit and the fee paid for purchasing electricity or power generation indicators for electricity transfer. It is calculated using the trend extrapolation method. The calculation formula is as follows:

[0089]

[0090] Where: C wg The real-time cost per unit of purchased power; MA wg The actual value of the purchased power cost last month; R v MB is the variable cost allocation ratio; wg This is the budget value of outsourced power costs for this month;

[0091] The startup cost calculation submodule dynamically calculates the unit startup cost C according to the unit startup parameters and loss rate. qd , the calculation formula is as follows:

[0092]

[0093] Where: C qd is the total startup cost; C qdrl Fuel cost for startup; C zs Q is the depreciation fee for unit startup; r The amount of fuel consumed for startup; P r The fuel price consumed to start; UI js I is the unit investment cost; zs R is the loss rate of life due to unit startup; f is the unit capacity allocation ratio; T′ is the monthly allocation ratio.

[0094] The real-time cost calculation module adds the measured unit fixed cost, unit variable cost and unit startup cost to obtain the real-time unit cost C, and the calculation formula is as follows:

[0095] C=C gd +C bd +C qd (8)

[0096] Real-time cost converts the three types of costs, namely startup cost, variable cost and variable cost, into kilowatt-hours of electricity, with the unit being yuan / MWh.

[0097] Specifically, the quotation optimization module includes a quotation strategy submodule based on cost analysis, a quotation optimization submodule based on game theory, and a dynamic optimization submodule based on reinforcement learning, wherein:

[0098] The cost analysis-based quotation strategy submodule sets the quotation based on the unit real-time cost. Its quotation model is a quotation model based on the variable cost and fixed cost of the power plant. Under the condition of meeting the power flow constraints of the entire network and the power demand restrictions of the power grid, the unit generates as much electricity as possible. Its quotation model is expressed as follows:

[0099]

[0100] Where: T is the number of quotation periods per day, t = 1, 2, 3, ... T; M is the number of units operating in the power plant during the day; c ti is the fuel cost per unit of electricity generated by the i-th unit in the t-th period; q ti is the output of the i-th unit in the t-th period; Q t is the power output reported by the power plant during period t; Q net-t is the power demand of the power grid in period t; Q in-max-t is the maximum injected power of the power plant as a node in the power grid during period t; p t is the electricity price reported by the power plant in period t; p t_max is the maximum price limit of the power grid during period t; Δq i is the output increase of the i-th unit in period t (can be a negative value); Δq i_max , Δq i_min are the maximum and minimum output increases of the i-th unit in a certain period of time; q i is the output of the i-th unit in period t; G i_max , G i_min are the maximum and minimum output of the i-th unit respectively.

[0101] The game theory-based quotation optimization submodule solves the optimal quotation through Nash equilibrium. The clearing power and market electricity price of each unit in this system are determined by preset scenario parameters. At present, the game theory-based quotation optimization submodule includes system load balance constraints, simplifies the safety constraints of the power grid and the start-stop constraints of the unit, and includes the following steps:

[0102] 1) First set the cost function and quotation function of each unit;

[0103] 2) Taking the cost function and quotation function of the unit as marginal conditions and the marginal income of each unit as the objective function, construct and solve the Nash equilibrium equations including the market clearing electricity price and the unit power generation. At this time, the market clearing process is to solve the following equations:

[0104]

[0105] Where: p j is the active power generated by the jth unit; u j 、v jis the strategic quotation coefficient of the j-th unit; P T is the system load curve; P 0 is the original load without considering load elasticity; k is the load elasticity coefficient; λ is the final clearing price of the market, whose trajectory is formed by the quotation curves of all power plants; λ j is the bid curve of each j-th power plant; the optimal bids of each power plant are related to each other, and finally a mutually acceptable equilibrium solution will be obtained, which is the Nash equilibrium point in game theory;

[0106] The unit parameters obtained at the Nash equilibrium point are the optimal strategy coefficients under the parameters of equation group (10).

[0107] 3) According to the optimal strategy coefficient obtained, adjust the unit's quotation to maximize profit;

[0108] The dynamic optimization submodule based on reinforcement learning, which adaptively adjusts the quotation in combination with real-time market data, includes the following steps:

[0109] 1) Based on the electricity market with a single power purchase operation model, multiple sellers (power generators) compete to sell electricity to a single buyer (power sales company). The power sales company purchases the grid-connected electricity of all successfully declared power generators based on the predicted system load, and then sells it to users in a unified manner. The power trading center is responsible for the operation and management of the power generation market and the dispatching and operation of the power system. It conducts research on the day-ahead market for power trading and requires power generators to submit an amount of power available and a quotation for the next period every hour. The predicted load adopts the preset scenario of the quotation strategy submodule.

[0110] 2) The bidding function of the generator sets in the system is a linear bidding function. The clearing process sorts the bids of the generator sets from low to high. According to the load demand set in the model scenario, the generators with low bids are given priority to be connected to the grid for power generation. Finally, the unified market clearing price calculated is used for settlement.

[0111] Specifically, the market settlement analysis module uses the following constraints: power balance constraint, voltage constraint, line capacity constraint, section constraint, maximum and minimum power constraints of generator sets, minimum start-up time constraints, ramp constraints, and fuel supply constraints;

[0112] The analysis methods used by the market settlement analysis module include:

[0113] ①Descriptive statistical analysis:

[0114] This includes calculating the mean and standard deviation to understand the center position and dispersion of the data; calculating the maximum and minimum values, identifying the extreme values ​​of prices, and understanding the range of price fluctuations;

[0115] ②Trend analysis:

[0116] Conduct time series analysis to observe price trends over time, understand how prices change over a specific period of time, and whether there is a clear upward or downward trend;

[0117] ③Comparative analysis:

[0118] Horizontal comparison: compare the average prices of different quarters or months to analyze the differences in the time dimension;

[0119] Vertical comparison: compare the price changes of the same month in different time periods, for example, compare the price changes of each quarter in 2025;

[0120] ④ANOVA:

[0121] Check whether there are significant differences in prices between different time periods (such as different quarters);

[0122] ⑤Regression analysis:

[0123] If you need to predict future prices or analyze the relationship between prices and other factors (such as demand, fuel costs, etc.), you can use regression models for modeling and analysis;

[0124] ⑥Graphical display:

[0125] Using charts to visually display the changing trends and distribution characteristics of price data helps to intuitively understand the data patterns. The charts include line charts, bar charts or box plots.

[0126] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A real-time cost calculation and quotation system for gas generator sets in the spot market, characterized in that: include: Real-time cost calculation module, used to calculate the real-time unit cost according to the actual operation of the gas unit; The quotation optimization module is used to calculate the unit cost in real time, use the calculated result as the marginal condition, start calculating the unit quotation strategy, and use multiple algorithms to optimize the quotation strategy of the gas unit in the spot market; The market settlement analysis module is used to analyze various quotation strategies of the unit after obtaining them in a simulated market environment, evaluate the settlement income and economic benefits of the gas unit, and thus evaluate the effectiveness of the quotation strategy.

2. According to claim 1, a real-time cost calculation and quotation system for gas turbines in the spot market is characterized in that: The real-time cost calculation module includes a fixed cost calculation submodule, a variable cost calculation submodule and a startup cost calculation submodule, wherein: The fixed cost calculation submodule calculates the unit fixed cost C gd It includes depreciation, financial expenses, employee salaries, material costs and repair costs, as follows: The depreciation is calculated by using the comprehensive depreciation rate allocation method, that is, the depreciation of the original value of fixed assets is allocated according to the comprehensive depreciation rate. The calculation formula is as follows: Where: C zj is the real-time cost per unit of depreciation; OV u is the original value of fixed assets of each unit; R f is the fixed cost allocation ratio; R r is the estimated residual value rate of fixed assets; t is the useful life of the unit; D i is the depreciation fee for infrastructure construction this month; E0 is the amount of electricity online on that day; The financial expenses refer to the expenses incurred by an enterprise to raise funds for production and operation. The financial expenses are calculated using the trend interpolation method, which is a method of calculating the actual value of the financial expenses of the previous month combined with the forecast value of the financial expenses of this month. The calculation formula is as follows: Where: C cw is the real-time cost of unit financial expenses; MA cw MB is the financial expenses of the previous month; cw Budget financial expenses for this month; D im The number of days that have occurred this month; D m is the total number of days in this month; D lm is the total number of days in the previous month; The employee salary is calculated using the annual budget allocation method, that is, the budget for the year is allocated to each day for calculation. The calculation formula is as follows: Where: C zgxc The real-time cost of employee salary of the unit; zgxc Budgeted employee salaries for the current year; The calculation method of the material cost and repair cost is the same as that of employee salary. It can be calculated by replacing the employee salary budget for the current year in formula (3) with the corresponding cost; The unit variable cost C calculated by the variable cost calculation submodule bd It includes gas cost and external power cost, which is allocated based on the real-time data of gas price, external power cost and variable material cost, and is allocated based on the unit grid-connected electricity consumption. The specific situation is as follows: The gas cost is calculated based on the measured energy consumption data of the unit. First, the quadratic function relationship between the total energy consumption of the unit and the power output level is fitted according to the least squares method. Then, the total power supply energy consumption of the unit is calculated, that is, the total power supply energy consumption-the total power generation energy consumption / (1-plant power consumption rate). According to the total power supply energy consumption, the no-load gas consumption of the unit and the marginal gas consumption at different output levels are calculated. Finally, the no-load gas cost and marginal gas cost are calculated in combination with the gas price. The no-load cost caused by non-dispatch reasons is included in the real-time cost. The calculation formula is: Where: C f is the unit gas cost; P is the power consumption rate; f is the comprehensive gas unit price; b x is the power consumption rate of power generation; ap is the power consumption rate of the plant; The external power purchase fee refers to the fee paid for purchasing electricity from the power grid to ensure the safe and stable operation of the unit and the fee paid for purchasing electricity or power generation indicators for electricity transfer. It is calculated using the trend extrapolation method. The calculation formula is as follows: Where: C wg The real-time cost per unit of purchased power; MA wg The actual value of the purchased power cost last month; R v MB is the variable cost allocation ratio; wg This is the budget value of outsourced power costs for this month; The startup cost calculation submodule dynamically calculates the unit startup cost C according to the unit startup parameters and loss rate. qd , the calculation formula is as follows: Where: C qd is the total startup cost; C qdrl Fuel cost for startup; C zs Q is the depreciation fee for unit startup; r The amount of fuel consumed for startup; P r The fuel price consumed to start; UI js I is the unit investment cost; zs R is the loss rate of life due to unit startup; f is the unit capacity allocation ratio; T′ is the monthly allocation ratio.

3. A real-time cost calculation and quotation system for gas turbines in the spot market according to claim 2, characterized in that: The real-time cost calculation module adds the measured unit fixed cost, unit variable cost and unit startup cost to obtain the real-time unit cost C, and the calculation formula is as follows: C=C gd +C bd +C qd (8)。 4. The real-time cost calculation and quotation system for gas turbines in the spot market according to claim 1 is characterized in that: The quotation optimization module includes a quotation strategy submodule based on cost analysis, a quotation optimization submodule based on game theory, and a dynamic optimization submodule based on reinforcement learning, wherein: The cost analysis-based quotation strategy submodule sets the quotation based on the unit real-time cost, and its quotation model is: Where: T is the number of quotation periods per day, t = 1, 2, 3, ... T; M is the number of units operating in the power plant during the day; c ti is the fuel cost per unit of electricity generated by the i-th unit in the t-th period; q ti is the output of the i-th unit in the t-th period; Q t is the power output reported by the power plant during period t; Q net-t is the power demand of the power grid in period t; Q in-max-t is the maximum injected power of the power plant as a node in the power grid during period t; p t is the electricity price reported by the power plant in period t; p t_max is the maximum price limit of the power grid during period t; Δq i is the output increase of the i-th unit in period t; Δq i_max , Δq i_min are the maximum and minimum output increases of the i-th unit in a certain period of time; q i is the output of the i-th unit in period t; G i_max , G i_min are the maximum and minimum output of the i-th unit respectively; The game theory-based quotation optimization submodule solves the optimal quotation through Nash equilibrium, including the following steps: 1) First set the cost function and quotation function of each unit; 2) Taking the cost function and quotation function of the unit as marginal conditions and the marginal income of each unit as the objective function, a Nash equilibrium equation group including the market clearing electricity price and the unit power generation is constructed and solved. At this time, the market clearing process is to solve the following equation group: Where: p j is the active power generated by the jth unit; u j 、v j is the strategic quotation coefficient of the j-th unit; P T is the system load curve; P0 is the original load without considering load elasticity; k is the load elasticity coefficient; λ is the final clearing price of the market, whose trajectory is formed by the quotation curves of all power plants; λ j is the bid curve of the jth power plant; the optimal bids of each power plant are related to each other, and finally a mutually acceptable equilibrium solution will be obtained, which is the Nash equilibrium point in game theory; The unit parameters obtained at the Nash equilibrium point are the optimal strategy coefficients under the parameters of equation group (10); 3) According to the optimal strategy coefficient obtained, adjust the unit's quotation to maximize profit; The dynamic optimization submodule based on reinforcement learning, which adaptively adjusts the quotation in combination with real-time market data, includes the following steps: 1) In the actual market, the power sales company purchases the grid-connected power of all power generators that have successfully applied for it according to the predicted system load, and then sells it to users in a unified manner; the power trading center is responsible for the operation and management of the power generation market and the dispatching and operation of the power system. It conducts research on the day-ahead market for power trading and requires power generators to submit an available power supply and quotation for the next period every hour; the predicted load adopts the preset scenario of the quotation strategy submodule; 2) The bidding function of the generator sets in the system is a linear bidding function. The clearing process sorts the bids of the generator sets from low to high. According to the load demand set in the model scenario, the generators with low bids are given priority to be connected to the grid for power generation. Finally, the unified market clearing price calculated is used for settlement.

5. The real-time cost calculation and quotation system for gas turbines in the spot market according to claim 1 is characterized in that: The market settlement analysis module uses the following constraints: power balance constraint, voltage constraint, line capacity constraint, section constraint, maximum and minimum power constraints of generator sets, minimum start-up time constraints, ramp constraints and fuel supply constraints; The analysis methods used by the market settlement analysis module include: ①Descriptive statistical analysis: This includes calculating the mean and standard deviation to understand the center position and dispersion of the data; calculating the maximum and minimum values, identifying the extreme values ​​of prices, and understanding the range of price fluctuations; ②Trend analysis: Conduct time series analysis to observe price trends over time, understand how prices change over a specific period of time, and whether there is a clear upward or downward trend; ③Comparative analysis: Horizontal comparison: compare the average prices of different quarters or months to analyze the differences in the time dimension; Vertical comparison: compare the price changes of the same month in different time periods, for example, compare the price changes of each quarter in 2025; ④ANOVA: Check whether there are significant differences in prices over different time periods; ⑤Regression analysis: If you need to predict future prices or analyze the relationship between prices and other factors, you can use regression models for modeling and analysis; ⑥Graphical display: Using charts to visually display the changing trends and distribution characteristics of price data helps to intuitively understand the data patterns. The charts include line charts, bar charts or box plots.

Images