A method for constructing power system evolution model based on multi-agent interaction

Abstract

The present invention provides a method for constructing an electric power system evolution model based on multi-agent interaction, which belongs to the field of electric power system planning; it solves the problem of how to realize the reasonable allocation of traditional thermal power units and new energy units in a new electric power system during low-carbon transformation; it comprises the following steps: obtaining the planning boundary conditions, data of thermal power units and new energy units required for evaluating the regional electric power system; constructing an electric power system evolution model based on multi-agent interaction composed of a subject operation model and operation constraints according to the obtained data; the subject operation model takes into account policy makers and power generators, and establishes subject behavior models of policy makers and power generators respectively, the subject behavior model of the policy makers includes a carbon emission policy model and a deep peak regulation compensation policy model, and the subject behavior model of the power generator includes a profit factor calculation model; the present invention is applied to a new electric power system.

Description

Technical Field

[0001] The invention provides a method for constructing an electric power system evolution model based on multi-agent interaction, belonging to the technical field of electric power system planning. Background Art

[0002] The establishment of a new power system has brought about major changes in the structure and form of my country's power system. The power industry's carbon dioxide emissions account for a large proportion of total carbon emissions, and coal-fired units have hindered my country's progress in achieving low-carbon development. However, coal-fired power generation also plays a vital role in supporting renewable energy to flexibly adjust peak loads and ensure the stable and safe operation of the power system.

[0003] Recent power market reforms have optimized the power supply structure and improved the efficiency and competitiveness of the power industry. The support provided by power market reforms to various types of power generation resources varies. Reforms can provide more opportunities for the development of clean energy and accelerate the adjustment of the energy structure. At the same time, the market share of coal-fired power generation is constrained by low-carbon policies and is also severely squeezed by clean energy. The power system cannot lack the regulation and safety support capabilities of coal-fired units in the short term, but in the new power system, coal-fired power generation is in a very difficult situation.

[0004] Therefore, it is necessary to establish a reasonable incentive mechanism to ensure that coal-fired power has certain benefits, ensure that the power system has the ability to ensure supply and peak load regulation, and maintain the safety and stability of the power system. Reasonable incentive policy design requires evaluating the interaction between policy design and market behavior and quantifying the substitutability between different resources to achieve the optimal combination and support the low-carbon transformation of renewable energy systems. Summary of the invention

[0005] In order to solve the problem of how to achieve reasonable allocation of traditional thermal power units and new energy units in the new power system during the low-carbon transformation, the present invention proposes a method for constructing a power system evolution model based on multi-agent interaction.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for constructing a power system evolution model based on multi-agent interaction, comprising the following steps:

[0007] S1: Obtain the planning boundary conditions, thermal power units and renewable energy units data required for evaluating the regional power system;

[0008] S2: constructing a power system evolution model based on multi-agent interaction consisting of an agent operation model and operation constraints according to the data obtained in step S1;

[0009] The subject operation model takes into account the policy makers and power generators, and establishes subject behavior models of the policy makers and power generators respectively. The subject behavior model of the policy makers includes a carbon emission policy model and a deep peak load compensation policy model, and the subject behavior model of the power generator includes a profit factor calculation model;

[0010] The carbon emission policy model is used to determine the carbon emission cost factor for the next year based on the carbon emission intensity of the current year and the carbon emission intensity targets for the next few years;

[0011] The deep peak load regulation compensation policy model is used to adjust the influencing factors of thermal power units participating in deep peak load regulation compensation;

[0012] The profit factor calculation model is used to evaluate the profitability of different investment plans of power generators based on the profit factors;

[0013] The operation constraints include unit operation constraints and energy storage constraints.

[0014] The planning boundary conditions in step S1 include the load forecast results and load characteristic curves of each node in the power system of the evaluation area for 365×24 hours based on the consideration of electricity consumers, as well as the capacity planning of existing units.

[0015] The data of the thermal power unit in step S1 include the thermal power unit power generation cost factor, the thermal power unit investment cost factor, the carbon emission cost factor, the thermal power unit output upper and lower limits, and the operating characteristics of the thermal power unit output upper and lower limits;

[0016] The data of new energy units include the investment cost factors of new energy units, the upper and lower limits of new energy unit output, the probabilistic scenario prediction results of the output of wind power, hydropower, and photovoltaic units in each period of the next day, and the operating characteristics of the upper and lower limits of new energy unit output.

[0017] According to the carbon emission intensity of the current year and the carbon emission intensity target for the next five years, the carbon emission cost factor for the next year is determined. The expression of the carbon emission cost factor is as follows:

[0018] ;

[0019] In the above formula: represents the carbon emission cost factor in year y+1, represents the carbon emission cost factor adjustment coefficient, represents the carbon emission cost factor in year y, represents the carbon emission cost factor predicted by policymakers in year y+1, represents the actual carbon emission intensity in year y, Represents the carbon emission intensity target for year y+5.

[0020] The expression of the deep peak load compensation policy model is as follows:

[0021] ;

[0022] ;

[0023] ;

[0024] In the above formula: R CG Compensation income from providing deep peak load regulation for thermal power units, G M is the collection of thermal power units, T To calculate the cycle, and Respectively represent thermal power units gm The minimum output of conventional peak-shaving and the minimum output of oil-free deep peak-shaving, for t Thermal power unit gm of efforts, k 1 and k 2 are the unit power compensation prices for oil-free deep peak regulation and oil-based deep peak regulation of thermal power units, k 1 and k 2The unit is RMB / MWh; k t It is the total compensation price for the non-oil deep peak regulation during the oil deep peak regulation period; k is the unit power compensation price at different peak load regulation stages, , Thermal power units gm Maximum and minimum output.

[0025] The profit factor includes power generation profit factor and policy subsidy factor. After deducting the unit operation cost factor of the power generator, the investment cost factor of the thermal power unit, the retirement cost factor of the thermal power unit and the environmental cost factor, the profit factor calculation model F The expression is as follows:

[0026] F = R EE + R CG - C INV - C OM - C RET - C E ;

[0027] In the above formula: R EEThe profit factor of power generation is calculated by multiplying the unit output by the economic parameter of electricity quantity at that time; R CG The compensation income obtained from providing deep peak load regulation for thermal power units is the policy subsidy factor; C INV is the investment cost factor of thermal power units; C OM is the unit operating cost factor of the generator; C RET is the cost factor for decommissioning thermal power units; C E It is an environmental cost factor.

[0028] R EE The calculation formula is as follows:

[0029] ;

[0030] In the above formula: G is a collection of generator sets, T To calculate the cycle, ep t for t The power economy parameters at the moment, for t Time Generator g contribution;

[0031] C INV The calculation formula is as follows:

[0032] ;

[0033] ;

[0034] In the above formula: It is a generator set g The unit annual investment cost factor is It is a generator set g of new installed capacity; is the bank interest rate, Y g Representing the generator set g service life, Representing the generator set g Unit capacity investment cost factor;

[0035] C OM The calculation formula is as follows:

[0036] ;

[0037] In the above formula: and Generator sets g Unit variable operation cost factor and unit fuel cost factor;

[0038] C RET The calculation formula is as follows:

[0039] ;

[0040] ;

[0041] ;

[0042] In the above formula: For generator sets g The retirement cost factor in year y, For generator sets g Equipment removal costs, For generator sets g Recovery income of the net value of the equipment; For generator sets g The retention renovation cost factor in year y, Describes a binary variable representing the generator set g The retirement decision of the unit is: a value of 0 means that the unit has not been retired and can continue to be expanded, while a value of 1 means that the unit has been retired; For generator sets g The capacity that needs to be retired in year y, For generator sets g The maximum retired capacity in year y, Retain renovation cost factor for the unit;

[0043] C E The calculation formula is as follows:

[0044] ;

[0045] In the above formula: For thermal power units gm CO2 intensity of the power source; Represents the carbon emission cost factor in year y.

[0046] The expression of the unit operation constraint is as follows:

[0047] ;

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] In the above formula: for t Time Generator g of efforts, Indicates the generator set g The maximum output of wind power and solar power generation, the maximum output power at each time point is the product of the unit available output power value predicted at that time and the installed capacity decision variable; It is a thermal power unit gm exist t The operating status at the moment is a 0-1 variable. When the unit is running, the variable is 1, otherwise it is 0; and Thermal power units gm Minimum continuous start time and minimum continuous stop time; RD gm and RU gm Thermal power units gm The downhill and uphill rates.

[0053] The expression of the energy storage constraint is as follows:

[0054] ;

[0055] ;

[0056] ;

[0057] ;

[0058] In the above formula: and Respectively t Energy storage at all times s The charging and discharging states of are both 0-1 variables; and They are t Energy storage at all times s Charging efficiency and discharging efficiency; for t Energy storage at all times s The available capacity of the remaining power; G ess A collection of energy storage units; For energy storage s The capacity, fort Energy storage at all times s The charging power, for t Energy storage at all times s The discharge power.

[0059] The beneficial effects of the present invention compared to the prior art are as follows: the present invention first obtains the planning boundary conditions required for evaluating the regional power system, and then establishes an evolution model that takes into account multiple power system entities based on the planning boundary conditions and solves the transformation paths of energy and power sectors in different regions within the calculation period. The present invention takes policy makers, power generators, and electricity consumers as the main subjects of the model research, including thermal power, hydropower, wind power, photovoltaic and other types of power generation units, and examines the impact of low-carbon policies and deep peak-shaving compensation policies on the orderly retirement of coal-fired power generation units and new energy investment, which will help the new power system achieve a safe and stable low-carbon transformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The present invention will be further described below in conjunction with the accompanying drawings:

[0061] Figure 1 It is a structural schematic diagram of the model of the present invention;

[0062] Figure 2 Schematic diagram of the operation of the model of the present invention. DETAILED DESCRIPTION

[0063] like Figure 1 to Figure 2 As shown, the present invention provides a method for constructing an electric power system evolution model based on multi-agent interaction, taking policy makers, power generators, and electricity consumers as the model research subjects, including multiple types of power generating units such as thermal power, hydropower, wind power, and photovoltaic power, and examines the impact of low-carbon policies and deep peak-shaving compensation policies on the orderly retirement of thermal power units and new energy (renewable energy) investment.

[0064] The construction steps of the present invention are as follows:

[0065] S1: Obtain the data of planning boundary conditions, thermal power units (conventional units) and new energy units required to evaluate the regional power system.

[0066] The planning boundary conditions include the load forecast results and load characteristic curves of each node in the regional power system for 365×24 hours, and the capacity planning of existing units. The load forecast results and load characteristic curves are the boundary conditions for electricity consumers.

[0067] The data of thermal power units include the power generation cost factor of thermal power units, the investment cost factor of thermal power units, the carbon emission cost factor, the upper and lower limits of thermal power unit output, and the operating characteristics of the upper and lower limits of thermal power unit output.

[0068] The data of new energy units include the investment cost factors of new energy units, the upper and lower limits of new energy unit output, the probabilistic scenario prediction results of the output of wind power, hydropower, and photovoltaic units in each period of the next day, and the operating characteristics of the upper and lower limits of new energy unit output.

[0069] S2: Construct a power system evolution model based on multi-agent interaction, which is composed of the subject operation model and operation constraints, according to the data obtained in step S1.

[0070] The subject operation model takes into account policy makers and power generators, and establishes the subject behavior models of policy makers and power generators respectively.

[0071] Policymakers have two goals: achieving carbon emission targets and maintaining supply security targets, which requires policymakers to adjust carbon emission policies and deep peak load regulation compensation policies.

[0072] First, a carbon emission policy model is established. Policymakers set annual carbon emission intensity targets, which are determined by the expected level of carbon emission reduction technology and the carbon emission requirements of the year. At the end of each year, policymakers receive information on the total amount of carbon dioxide emissions for the year from power generators and adjust the carbon emission cost factor to achieve the established carbon emission intensity target. The new carbon emission cost factor is passed on to power generators, which will affect their future operating cost factors and investment decisions.

[0073] If the carbon emission intensity in year y exceeds the carbon emission intensity target for the next five years, the carbon emission cost factor for the next year will be set to the larger value of the current carbon emission cost factor or the forecast parameter increased by 10%. Otherwise, the carbon emission cost factor for the second year will be the larger value between the current carbon dioxide price or the carbon tax. The specific formula is as follows:

[0074] (1);

[0075] In the above formula: represents the carbon emission cost factor in year y+1, represents the carbon emission cost factor adjustment coefficient, represents the carbon emission cost factor in year y, represents the carbon emission cost factor predicted by policymakers in year y+1, represents the actual carbon emission intensity in year y, Represents the carbon emission intensity target for year y+5.

[0076] In order to ensure supply security, policymakers have also established a deep peak-shaving compensation policy model for maintaining supply security goals. The deep peak-shaving compensation policy model can adjust the unit price of deep peak-shaving compensation to achieve a more effective mechanism for thermal power units to meet system adjustments. According to different adjustment capabilities and cost factor characteristics, the operation of thermal power units can be divided into three stages: conventional peak-shaving, deep peak-shaving without oil, and deep peak-shaving with oil. The calculation method of deep peak-shaving compensation is as follows:

[0077] (2);

[0078] (3);

[0079] (4);

[0080] In the above formula: R CG Compensation income from providing deep peak load regulation for thermal power units, G M is the collection of thermal power units, T To calculate the cycle, and Respectively represent thermal power units gm The minimum output of conventional peak-shaving and the minimum output of oil-free deep peak-shaving, for t Thermal power unit gm of efforts, k 1 and k 2 are the unit power compensation prices for oil-free deep peak regulation and oil-based deep peak regulation of thermal power units, respectively, and the unit of this value is RMB / MWh; k t is the total compensation price for the non-oil deep peak-shaving part during the oil deep peak-shaving period. Formula (4) provides two levels of floating quotation ranges for the peak-shaving auxiliary services of thermal power units. k is the unit power compensation price at different peak load regulation stages, , Thermal power units gm Maximum and minimum output.

[0081] Power generators have one goal: to make a profit. The present invention considers multiple types of power generators, including power generation companies and energy storage companies, each of which has a different portfolio of units. These power generators evaluate the profitability of their assets by participating in the electricity market and decide to retire unprofitable plants or invest in new plants. Each power generator has its own portfolio of power generation resources, including thermal power units, new energy units and energy storage units.

[0082] At the end of each year, generators evaluate the profitability of different investment options, such as the type and capacity of new power plants, to determine the most profitable option. To evaluate the profitability of investment options, generators consider various investment scenarios. Investment decisions are determined by calculating the return on investment for a range of power plant sizes, including the minimum generation capacity and maximum annual construction rate for each technology for each generator.

[0083] These power generators evaluate the profitability of different investment plans every year based on their profit factors. The profit factors of these power generators include power generation profit factors and policy subsidy factors, excluding the unit operation cost factors, thermal power unit investment cost factors, thermal power unit retirement cost factors and environmental cost factors of the power generators. Therefore, the main behavior model of the power generator constructed is a profit factor calculation model, and the expression of the profit factor calculation model is as follows:

[0084] F = R EE + R CG - C INV - C OM - C RET - C E (5);

[0085] In the above formula: R EE It is the profit factor of power generation, which is calculated by multiplying the unit output by the economic parameter of electricity quantity at that time. R CG The compensation income obtained from providing deep peak load regulation for thermal power units is the policy subsidy factor. The specific calculation method is shown in formulas (2)-(4). C INV It is the investment cost factor of thermal power unit. C OM is the unit operating cost factor of the generator. C RET is the cost factor for decommissioning thermal power units. C E It is an environmental cost factor.

[0086] in: R EE The calculation formula is as follows:

[0087] (6);

[0088] In the above formula: G is a collection of generator sets, T To calculate the cycle, ept for t The power economy parameters at the moment, for t Time Generator g of effort.

[0089] C INV The calculation formula is as follows:

[0090] (7);

[0091] (8);

[0092] In the above formula: It is a generator set g The unit annual investment cost factor is It is a generator set g of new installed capacity; is the bank interest rate, which is set to 7% in this embodiment; Y g Representing the generator set g service life, Representing the generator set g Unit capacity investment cost factor.

[0093] C OM The calculation formula is as follows:

[0094] (9);

[0095] In the above formula: and Generator sets g The unit variable operation cost factor and unit fuel cost factor.

[0096] C RET The calculation formula is as follows:

[0097] (10);

[0098] (11);

[0099] (12);

[0100] In the above formula: For generator sets g The retirement cost factor in year y is determined by the difference between the unit equipment removal cost factor and the equipment net value recovery income, as shown in formula (11): For generator sets gThe cost of equipment removal, For generator sets g Recovery income of the net value of the equipment; For generator sets g The retention renovation cost factor in year y is shown in formula (12). Describes a binary variable (0 or 1) representing the generator set g The decommissioning decision of the unit is indicated by a value of 0, indicating that the unit has not been decommissioned and can continue to be expanded, while a value of 1 indicates that the unit has been decommissioned. For generator sets g The capacity that needs to be retired in year y, For generator sets g The maximum retired capacity in year y, Keep the renovation cost factor for the unit.

[0101] C E The calculation formula is as follows:

[0102] (13);

[0103] In the above formula: For thermal power units gm CO2 intensity of the power source; Represents the carbon emission cost factor in year y.

[0104] The operating constraints of the model include:

[0105] Condition 1: Unit operation constraints, the expression is as follows:

[0106] (14);

[0107] (15);

[0108] (16);

[0109] (17);

[0110] (18);

[0111] In the above formula: Indicates the generator set g The maximum output of wind power and solar power generation, the maximum output power at each time point is the product of the unit available output power value predicted at that time and the installed capacity decision variable; It is a thermal power unit gm exist tThe operating status at the moment is a 0-1 variable. When the unit is running, the variable is 1, otherwise it is 0; and Thermal power units gm Minimum continuous start time and minimum continuous stop time; RD gm and RU gm Thermal power units gm The downhill and uphill rates.

[0112] Wherein, formula (14) represents the output constraint of the generating set (including thermal power set and new energy set). Formula (15) and formula (16) represent the minimum continuous operation time and minimum continuous shutdown time constraints of the thermal power set respectively. Formula (17) and formula (18) represent the ramp constraints of the thermal power set.

[0113] Condition 2: Energy storage constraint, the expression is as follows:

[0114] (19);

[0115] (20);

[0116] (twenty one);

[0117] (twenty two);

[0118] In the above formula: and Respectively t Energy storage at all times s The charging and discharging states of are both 0-1 variables; and They are t Energy storage at all times s Charging efficiency and discharging efficiency; for t Energy storage at all times s The available capacity of the remaining power; G ess A collection of energy storage units; For energy storage s The capacity, for t Energy storage at all times s The charging power, for t Energy storage at all times s The discharge power.

[0119] Wherein, formula (19) describes the relationship between the charging and discharging states of energy storage. Formula (20) describes the discharge rate limit of energy storage, and formula (21) shows that the discharge power of energy storage is limited by the current SoC state. Formula (22) represents the energy storage balance constraint in adjacent time periods.

[0120] At the end of each year, power generation entities estimate the profitability of various investment plans (including the type and capacity of new power plants) and decide to invest in the most profitable options. To determine which investment plan is the most profitable, power generation companies should consider a number of factors. When determining the investment capacity, the return on investment is estimated for a range of power plant capacities, from the smallest power plant size to the company's largest annual power generation construction rate.

[0121] At the same time, when assessing the feasibility of new thermal and renewable energy power plants, the specifications of new facilities are determined in the following way: at the end of each year, the power generator selects all plants with a positive net present value of investment. NPV The calculation formula is as follows:

[0122] (twenty three);

[0123] In the above formula: Y For the unit life, Cash y is the cash flow of the power plant, It is the bank interest rate.

[0124] The reason for choosing a five-year period is that the future profitability factor has less impact on the NPV of investment than the near-term profitability factor due to the impact of future bank interest rates. Next, calculate the return on investment for each plant with an investment NPV greater than zero. ROI The calculation formula is as follows:

[0125] (twenty four);

[0126] In the above formula: R y is the total revenue during the useful life of the unit, C y is the total cost during the unit's service life, C I The cost of investing in a generating unit for a power plant.

[0127] In each category (e.g. coal, etc.), the plant with the highest ROI is selected. If the plant's ROI is above 0.5, the investment is considered ideal. Plants with ROI above 0.5 are assumed to be ideal investments. By calculating the ROI for a range of plant sizes between the minimum and maximum allowed production capacities, the optimal production capacity size of the plant can be estimated. The new plant is then added to the construction queue.

[0128] The operation steps of the model of the present invention are described below with reference to the accompanying drawings. The specific steps are as follows:

[0129] 1) Operates on both annual and hourly time scales. On the annual time scale, decisions related to capacity planning and policy making can be made, while on the hourly time scale, operational aspects of the power system can be modeled. Figure 2 The model's input data parameters and output results, including power plant profitability, CO2 emissions, installed capacity, and power generation, are shown.

[0130] 2) Combine short-term system operation with long-term investment by calculating cost factors, carbon emissions, and profit factors for each power plant every hour. This information is then used to determine the annual CO2 emissions, net present value of investment, and rate of return on investment for each power generation technology in the power system.

[0131] 3) Annual carbon dioxide emissions will be used to calculate the carbon emission cost factor for the next year (calculated using formula (1)). The net present value of investment (calculated using formula (23)) and the rate of return on investment (calculated using formula (24)) are determined based on the profit factor and cost factor obtained from the short-term model operation of each power plant. This information will be used to guide long-term investment.

[0132] 4) The specific loop iteration steps of the model are as follows:

[0133] 4.1) Load the parameters of electricity consumers, generators and decision makers and input them into the evolution model;

[0134] 4.2) Carry out annual cycle:

[0135] Update policy parameters: including carbon dioxide prices and thermal power deep peak load compensation policies;

[0136] Evaluate the profitability of the unit;

[0137] Eliminate unprofitable units according to the standard of net present value of investment < 0;

[0138] Perform hourly cycle (8760 hours): first calculate the load demand at that time, and then dispatch power plants in the following order to meet the load demand: new energy units, energy storage units and thermal power units. The one-day-ahead market and the one-day-ahead market are not considered.

[0139] 4.3) Output the annual change of unit capacity, update the policy parameters according to the rules, input the obtained unit capacity and policy parameters into the model, and return to step 4.2).

[0140] In summary, the power system evolution model based on multi-agent interaction constructed by the present invention comprehensively considers the influence of multiple types of power system subjects and multiple policies such as policy makers, power generators, and electricity consumers, and considers that multiple types of power generation units such as thermal power, hydropower, wind power, and photovoltaic power take their own profit factors as the main evolution goal under the influence of formulated policies, and conducts capacity investment / reduction in combination with operating constraints. The planning and operation results are fed back to influence the market environment and policy makers to form an iterative evolution. The planning boundary conditions of the evaluation area are input and calculated using the annual rolling evolution method to obtain the evolution result of the power system in the evaluation area.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a power system evolution model based on multi-agent interaction, characterized in that: The following steps are involved: S1: Obtain the planning boundary conditions, thermal power units and renewable energy units data required for evaluating the regional power system; S2: constructing a power system evolution model based on multi-agent interaction consisting of an agent operation model and operation constraints according to the data obtained in step S1; The subject operation model takes into account the policy makers and power generators, and establishes subject behavior models of the policy makers and power generators respectively. The subject behavior model of the policy makers includes a carbon emission policy model and a deep peak load compensation policy model, and the subject behavior model of the power generator includes a profit factor calculation model; The carbon emission policy model is used to determine the carbon emission cost factor for the next year based on the carbon emission intensity of the current year and the carbon emission intensity target for the next five years; According to the carbon emission intensity of the current year and the carbon emission intensity target for the next five years, the carbon emission cost factor for the next year is determined. The expression of the carbon emission cost factor is as follows: ； In the above formula: represents the carbon emission cost factor in year y+1, represents the carbon emission cost factor adjustment coefficient, represents the carbon emission cost factor in year y, represents the carbon emission cost factor predicted by policymakers in year y+1, represents the actual carbon emission intensity in year y, represents the carbon emission intensity target for year y+5; The deep peak load regulation compensation policy model is used to adjust the influencing factors of thermal power units participating in deep peak load regulation compensation; The expression of the deep peak load compensation policy model is as follows: ； ； ； In the above formula: R CG Compensation income from providing deep peak load regulation for thermal power units, G M is the collection of thermal power units, T To calculate the cycle, and Respectively represent thermal power units gm The minimum output of conventional peak-shaving and the minimum output of oil-free deep peak-shaving, for t Thermal power unit gm of efforts, k 1 and k 2 are the unit power compensation prices for oil-free deep peak regulation and oil-based deep peak regulation of thermal power units, k 1 and k 2The unit is RMB / MWh; k t It is the total compensation price for the non-oil deep peak regulation during the oil deep peak regulation period; k is the unit power compensation price at different peak load regulation stages, , Thermal power units gm Maximum and minimum output; The profit factor calculation model is used to evaluate the profitability of different investment plans of power generators based on the profit factors; The profit factor includes power generation profit factor and policy subsidy factor. After deducting the unit operation cost factor of the power generator, the investment cost factor of the thermal power unit, the retirement cost factor of the thermal power unit and the environmental cost factor, the profit factor calculation model F The expression is as follows: F = R EE + R CG - C INV - C OM - C RET - C E ； In the above formula: R EE The profit factor of power generation is calculated by multiplying the unit output by the economic parameter of electricity quantity at that time; R CG The compensation income obtained from providing deep peak load regulation for thermal power units is the policy subsidy factor; C INV is the investment cost factor of thermal power units; C OM is the unit operating cost factor of the generator; C RET is the cost factor for decommissioning thermal power units; C E It is the environmental cost factor; R EE The calculation formula is as follows: ； In the above formula: G is a collection of generator sets, T To calculate the cycle, ep t for t The power economy parameters at the moment, for t Time Generator g contribution; C INV The calculation formula is as follows: ； ； In the above formula: It is a generator set g The unit annual investment cost factor is It is a generator set g of new installed capacity; is the bank interest rate, Y g Representing the generator set g service life, Representing the generator set g Unit capacity investment cost factor; C OM The calculation formula is as follows: ； In the above formula: and Generator sets g Unit variable operation cost factor and unit fuel cost factor; C RET The calculation formula is as follows: ； ； ； In the above formula: For generator sets g The retirement cost factor in year y, For generator sets g Equipment removal costs, For generator sets g Recovery income of the net value of the equipment; For generator sets g The retention renovation cost factor in year y, Describes a binary variable representing the generator set g The retirement decision of the unit is: a value of 0 means that the unit has not been retired and can continue to be expanded, while a value of 1 means that the unit has been retired; For generator sets g The capacity that needs to be retired in year y, For generator sets g The maximum retired capacity in year y, Retain renovation cost factor for the unit; C E The calculation formula is as follows: ； In the above formula: For thermal power units gm CO2 intensity of the power source; represents the carbon emission cost factor in year y; The operation constraints include unit operation constraints and energy storage constraints.

2. The method for constructing a power system evolution model based on multi-agent interaction according to claim 1, characterized in that: The planning boundary conditions in step S1 include the load forecast results and load characteristic curves of each node in the power system of the evaluation area for 365×24 hours based on the consideration of electricity consumers, as well as the capacity planning of existing units.

3. The method for constructing a power system evolution model based on multi-agent interaction according to claim 1, characterized in that: The data of the thermal power unit in step S1 include the thermal power unit power generation cost factor, the thermal power unit investment cost factor, the carbon emission cost factor, the thermal power unit output upper and lower limits, and the operating characteristics of the thermal power unit output upper and lower limits; The data of new energy units include the investment cost factors of new energy units, the upper and lower limits of new energy unit output, the probabilistic scenario prediction results of the output of wind power, hydropower, and photovoltaic units in each period of the next day, and the operating characteristics of the upper and lower limits of new energy unit output.

4. The method for constructing a power system evolution model based on multi-agent interaction according to claim 1 is characterized in that: The expression of the unit operation constraint is as follows: ； ； ； ； ； In the above formula: for t Time Generator g of efforts, Indicates the generator set g The maximum output of wind power and solar power generation, the maximum output power at each time point is the product of the unit available output power value predicted at that time and the installed capacity decision variable; It is a thermal power unit gm exist t The operating status at the moment is a 0-1 variable. When the unit is running, the variable is 1, otherwise it is 0; and Thermal power units gm Minimum continuous start time and minimum continuous stop time; RD gm and RU gm Thermal power units gm The downhill and uphill rates.

5. The method for constructing a power system evolution model based on multi-agent interaction according to claim 1, characterized in that: The expression of the energy storage constraint is as follows: ； ； ； ； In the above formula: and Respectively t Energy storage at all times s The charging and discharging states of are both 0-1 variables; and They are t Energy storage at all times s Charging efficiency and discharging efficiency; for t Energy storage at all times s The available capacity of the remaining power; G ess A collection of energy storage units; For energy storage s The capacity, for t Energy storage at all times s The charging power, for t Energy storage at all times s The discharge power.

Images