A new energy power station spot transaction strategy simulation and market optimization operation method and system based on principal-agent game

Abstract

The application discloses a new energy power station spot transaction strategy simulation and market optimization operation method and system based on principal-agent game, and belongs to the electrical engineering field.The method adopts a principal-agent double-layer game framework, firstly establishes a spot market transaction strategy optimization model of each new energy power station, secondly constructs a power spot market optimization dispatching model, thirdly equivalently converts the market optimization dispatching model into constraint conditions of the new energy power station optimization problem through KKT conditions, so that the non-convex and nonlinear double-layer optimization model is converted into a "mathematical programming problem with equilibrium constraints" of each new energy power station, and finally, the optimization model of each new energy power station is converted into constraint conditions again, and is formed into an "equilibrium problem with equilibrium constraints" by being simultaneously established.The simulation of the transaction strategies of multiple new energy power stations can be realized, and the market optimization dispatching is realized at the same time, which is used for guiding the core businesses such as dispatching price prediction and market benefit evaluation of the market operator.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of electrical engineering, and more particularly relates to a new energy power station spot transaction strategy simulation and market optimization operation method and system based on principal-agent game. TECHNICAL BACKGROUND

[0002] Since the new round of power reform, China has gradually implemented spot market pilots and promoted renewable energy to participate in spot market transactions, and wind, solar and other new energy and traditional power sources such as thermal power to realize bidding on the same platform. Wind power, photovoltaic and other new energy have the technical and economic characteristics of zero marginal cost, and traditional power market simulation operation methods generally treat them as market boundary conditions for "passive optimization", without considering their transaction strategies and game behaviors in the spot market. On the other hand, existing market force analysis methods and regulatory means are generally limited to thermal power units, and have not considered new energy power stations with almost negligible marginal cost. With the increasing proportion of renewable energy investment installed capacity, the uncertainty on both sides of the system has significantly increased, leading to increased volatility of spot prices, and their transaction strategies and game behaviors will become increasingly important. Therefore, there is an urgent need for a spot transaction strategy simulation and market optimization operation method for new energy power stations. SUMMARY

[0003] In view of the defects of the prior art, the purpose of the present application is to provide a new energy power station spot transaction strategy simulation and market optimization operation method and system based on principal-agent game, aiming to establish a mathematical model to numerically simulate the independent optimal bidding strategies of multiple new energy power stations and the market clearing state under the corresponding conditions, and to provide quantitative support for the clearing price prediction, market benefit evaluation and other core businesses of market operators.

[0004] To achieve the above purpose, the present application provides a new energy power station spot transaction strategy simulation and market optimization operation method based on principal-agent game, comprising the following steps:

[0005] S1, a transaction strategy optimization model of each new energy power station is established with the maximum power generation income of each new energy power station as the target;

[0006] S2, an optimization clearing model of the power spot market is established with the minimum system power supply cost as the target and the power spot market operation constraints as the constraint conditions;

[0007] S3, based on the principal and subordinate game theory, a double game model is established with new energy power station as the upper leader and market operator as the lower follower; further, through Karush-Kuhn-Tucker condition (KKT condition), the market clearing optimization model of the lower layer is equivalent to the constraint condition in the transaction strategy optimization model of each new energy power station of the upper layer, so as to form the "mathematical programs with equilibrium constraints" (MPEC) problem of each new energy power station;

[0008] S4, the MPEC problem of each new energy power station is converted into single-layer nonlinear constraint condition through KKT condition, and is associated to form "equilibrium problems with equilibrium constraints" (EPEC), so as to convert the original complex problem into single-layer optimization model, and can be solved by NLP commercial solver, and finally realize the simulation of new energy power station transaction strategy and market clearing operation simulation, which is used for guiding the clearing price prediction, market benefit evaluation and other core businesses of market operator.

[0009] Further, the step S1 comprises:

[0010] The transaction strategy optimization model of each new energy power station is constructed: the objective function is to maximize the power generation income (= power selling income-power generation cost), and the constraint condition is to ensure the increasing of the seller's offer curve.

[0011] Further, the step S2 comprises:

[0012] The power spot market optimization clearing model is constructed: the objective function is to minimize the system power supply cost (= thermal power unit power generation cost + new energy power station power generation cost + load shedding loss), and the constraint conditions are respectively: ① system power balance constraint; ② thermal power unit output constraint; ③ new energy power station output constraint; ④ load shedding power constraint.

[0013] Further, the step S3 comprises:

[0014] S31, the Lagrange dual function of the market clearing optimization model is established.

[0015] S32, the KKT condition of the market clearing optimization model is established, which specifically comprises: ① original problem constraint; ② first order condition of Lagrange function; ③ complementary relaxation condition transformed from original inequality constraint, and equivalent strong duality theorem equation; ④ non-negativity constraint of dual variable.

[0016] S33, the KKT conditions formed by S32-①~③ are combined into the constraint conditions of the upper optimization problem in S1, and the double-layer master-slave game is converted into a nonlinear single-layer constraint condition, that is, the MPEC problem of each new energy power station.

[0017] Further, the step S4 comprises:

[0018] S41, the MPEC problem of each new energy power station is reconstructed, and a Lagrange dual function of the new energy power station transaction strategy optimization problem is established;

[0019] S42, the KKT conditions of each MEPC problem are combined, the original complex problem is converted into a single-layer nonlinear optimization problem, and a NLP commercial solver is used for solving.

[0020] The application also provides a new energy power station spot transaction strategy simulation and market force analysis method based on master-slave game.

[0021] The computer readable storage medium is used for storing executable instructions.

[0022] The processor is used for reading the executable instructions stored in the computer readable storage medium, and executing the above-mentioned new energy power station spot transaction strategy simulation and market optimization operation method based on master-slave game.

[0023] Overall, the above technical solutions conceived by the application can achieve the following beneficial effects:

[0024] The application adopts the master-slave game architecture, converts the complex double-layer optimization problem into a single-layer optimization problem, realizes the simulation of the new energy power station transaction strategy and the simulation of the market clearing equilibrium state, and further provides a scientific basis for the clearing price prediction, market benefit evaluation and other core businesses of the market operator. Therefore, the application has significant practical value and can better serve the participants and operators of the electricity spot market. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 A flowchart of a new energy power station spot transaction strategy simulation and market optimization operation method based on master-slave game provided by the application.

[0026] Figure 2 The influence of the market force concentration degree of the new energy power station on the market clearing price under different renewable energy penetration rates provided by the embodiment of the application.

[0027] Figure 3 The change distribution trend of the market clearing price with the market force concentration degree of the new energy power station under different renewable energy penetration rates provided by the embodiment of the application.

[0028] Figure 4 The change trend of the new energy power station market transaction strategy (i.e., the bidding curve) under different new energy power station market force concentration degrees provided by the embodiment of the present application;

[0029] Figure 5 The change distribution trend of the market economic surplus with the new energy power station market force concentration degree under different renewable energy penetration rates provided by the embodiment of the present application. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical scheme and advantages of the present application more clear and explicit, the present application is further described in detail below in combination with the drawings and embodiments.

[0031] As shown in the drawings, the new energy power station spot transaction strategy simulation and market optimization operation method based on principal-agent game of the present application comprises the following steps: Figure 1

[0032] S1, a transaction strategy optimization model of each new energy power station is established with the maximum power generation income of each new energy power station as the target. Specifically, the objective function is to maximize the power generation income (= electricity sales income - power generation cost):

[0033]

[0034] The constraint condition is to ensure the increasing nature of the seller bidding curve.

[0035]

[0036] In the formula, Ω R is the set of wind power and photovoltaic units owned by the Rth new energy power station, c w , c pv are the marginal power generation costs of wind power / photovoltaic units, p w，b，t , p pv，b，t are the output powers of the w / pv wind power / photovoltaic unit at the bth segment t moment, and α w，b，t， α pv，b，t are the seller prices of the w / pv wind power / photovoltaic at the bth segment t moment.

[0037] S2, an optimization dispatching model of the electricity spot market is established with the minimum system power supply cost as the target and the power spot market operation constraints as the constraint conditions. Specifically, the objective function is to minimize the system power supply cost (= thermal power unit power generation cost + new energy power station power generation cost + load shedding loss). In the present application, it is assumed that the existing market force supervision means is effective, so the thermal power unit is priced according to its marginal cost:

[0038] ​

[0039] The constraints are respectively:

[0040] ① System power balance constraint;

[0041]

[0042] ② Thermal power unit output constraint;

[0043]

[0044] ③ New energy power station output constraint;

[0045]

[0046] ④ Load shedding power constraint.

[0047]

[0048] In the formula, D t is the electricity demand of all non-market users at time t, is the maximum / minimum declared capacity of thermal power unit g, is the maximum declared capacity of wind power w / solar pv in the b section, Q w，t， α pv，t is the maximum available capacity of wind power w / solar pv at time t, is the maximum capacity of interruptible load. λ t is the Lagrange dual variable of power balance constraint; is the Lagrange dual variable corresponding to each inequality constraint.

[0049] S3, based on the principal-agent game theory, a double game model is established with new energy power station as the upper leader and market operator as the lower follower; further, through Karush-Kuhn-Tucker condition (KKT condition), the market clearing optimization model of the lower layer is equivalent to the constraint condition in the transaction strategy optimization model of each new energy power station of the upper layer, thereby establishing a nonlinear transaction strategy optimization problem of each new energy power station, which is essentially a "mathematical programming with equilibrium constraints" problem (MPEC);

[0050] Specifically, it can be divided into the following three steps.

[0051] S31, the Lagrange dual function of the market clearing optimization model is established.

[0052]

[0053] S32, establish the KKT condition of market clearing optimization model, specifically including:

[0054] ① Original problem constraints: as shown in S2 model.

[0055] ② First order condition of Lagrange function:

[0056]

[0057] ③ Complementary relaxation condition transformed from original inequality constraints, and its equivalent strong duality theorem equation;

[0058] Complementary relaxation condition:

[0059]

[0060] Strong duality theorem equation:

[0061]

[0062] ④ Non-negativity constraint of dual variables.

[0063]

[0064] S33, the KKT conditions formed by S32-①~④ are combined with the constraint conditions of S1 in each new energy power station optimization problem, so that the double-layer master-slave game model of each new energy power station is converted into a nonlinear single-layer constraint condition, that is, the MPEC problem of each new energy power station.

[0065] S4, the MPEC problem of each new energy power station is converted into a constraint condition through KKT condition, and is combined to form "equilibrium problem with equilibrium constraints" (EPEC), so that the original complex problem is converted into a single-layer optimization model, and can be solved by NLP commercial solver, and finally realizes the simulation of new energy power station transaction strategy and market clearing operation simulation, which is used to guide the clearing price prediction, market benefit evaluation and other core businesses of market operator.

[0066] Specifically, it can be divided into the following two steps.

[0067] S41, reconstruct the MPEC problem of each new energy power station, and establish the Lagrange dual function of each new energy power station transaction strategy optimization problem. The reconstructed optimization problem of each new energy power station is as follows:

[0068] S.t.:

[0069]

[0070]

[0071]

[0072]

[0073]

[0074] The Lagrangian dual function is constructed in a similar way as S31, and is not repeated here.

[0075] S42, KKT conditions of each MEPC problem are solved simultaneously to transform the original complex problem into a single-layer nonlinear optimization problem, and NLP commercial solver is used to solve it. The final optimization problem consists of the following parts:

[0076] ① The original problem equality constraints, as shown in step S2.

[0077] ② The Lagrangian first-order conditions of each MPEC problem (i.e. the second-order conditions of the original problem):

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096] 3. First order complementary slackness condition, as shown in step S32-3.

[0097] 4. Second order complementary slackness condition:

[0098]

[0099] 5. Complementary slackness condition between first order and second order dual variables:

[0100]

[0101] The specific implementation steps of the present application will be further described below in combination with a specific test system.

[0102] The test example of the present application sets different two renewable energy penetration rate scenarios of 40%, 50%, 60%, 70% and 80%, and the corresponding power installed capacity under each scenario is shown in Table 1.

[0103] Table 1

[0104]

[0105]

[0106] Among them, the thermal power units include 1 1000MW coal-fired unit, 1 600MW coal-fired unit, 1 300MW coal-fired unit and 1 100MW gas-fired unit.

[0107] The market concentration of new energy power stations under different penetration rate scenarios is set in different scenarios, which are: oligarch market (1 new energy power station), heavy oligarch monopoly market (2-5 new energy power stations), light oligarch monopoly market (6-9 new energy power stations), and competitive market (10 or more new energy power stations). The present example carries out numerical simulation on different market concentration scenarios under different renewable energy penetration rates, and quantitatively compares the market clearing price and its distribution, the new energy power station bidding curve, the market economic surplus transfer and the like, as shown in the accompanying drawings of the specification. Figures 2 to 5

[0108] Those skilled in the art will readily understand that the above description is only the preferred embodiment of the present application, and is not intended to limit the present application, and any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.​

Claims

1. A method for simulating and optimizing the operation of spot trading strategies of new energy power plants based on a principal-agent game, characterized in that, The method comprises the following steps: S1, establishing a transaction strategy optimization model of each new energy power station with the maximum power generation income of each new energy power station as a target; the objective function is: The constraint condition is to ensure the increasing of the seller's offer curve: In the formula, is the set of wind power and photovoltaic units owned by the Rth new energy power station, is the marginal generation cost of the wind power / photovoltaic unit, respectively, is the output power of the w / pvth wind power / photovoltaic unit at the bth section at time t, respectively, is the seller quotation of the w / pvth wind power / photovoltaic unit at the bth section at time t, respectively. S2, establishing a power spot market optimization dispatching model with the minimum system power supply cost as a target and the power spot market operation constraint as a constraint condition; the objective function is: The constraint conditions include: ① system power balance constraint: ② thermal power unit output constraint: ③ new energy power station output constraint: ④ load shedding power constraint: wherein, respectively the maximum / minimum declared capacity of thermal power units g, respectively the maximum declared capacity of wind power w / solar pv in the b-th segment, respectively the maximum available capacity of wind power w / solar pv at time t, is the maximum capacity of interruptible load; respectively the Lagrange dual variable corresponding to each inequality constraint; S3, taking each new energy power station as an upper leader and a market operator as a lower follower, converting the power spot market dispatching optimization model into an MPEC problem in the transaction strategy optimization model of each new energy power station according to the KKT condition and the strong duality theory; S4, converting the MPEC problem into a single-layer nonlinear constraint condition through the KKT condition, combining all single-layer nonlinear constraint conditions, and solving by using an NLP commercial solver to obtain the optimal transaction strategy of the new energy power station and the market optimization dispatching result.

2. The method of claim 1, wherein, The step S3 comprises: S31, establishing a Lagrange dual function of the power spot market optimization dispatch model L : wherein is the electricity demand of all non-marketized users at time t; S32, establishing the KKT condition of the power spot market optimization dispatching model, specifically comprising: ① original problem constraint: ② first-order condition of the Lagrange function: ③ complementary relaxation condition transformed from the original inequality constraint and its equivalent strong duality theorem equation: ④ non-negativity constraint of the dual variable: ； S33, combining the KKT conditions formed by S32-①~③ into the constraint conditions in S1, and converting the power spot market dispatching optimization model into an MPEC problem in the transaction strategy optimization model of each new energy power station.

3. The method of claim 1, wherein, The step S4 comprises: S41, reconstructing the MPEC problem of each new energy power station and establishing the Lagrange dual function of the transaction strategy optimization problem of each new energy power station; S42, combining the KKT conditions of each MEPC problem, converting each MEPC problem into a single-layer nonlinear constraint condition, and using an NLP commercial solver for solving.

4. A method for simulating a new energy power plant spot transaction strategy based on a principal-agent game and analyzing market power, characterized in that a new energy power plant spot transaction strategy simulation and market optimization operation system, It comprises: a computer readable storage medium and a processor; the computer readable storage medium is used for storing executable instructions; the processor is used for reading the executable instructions stored in the computer readable storage medium and executing the new energy power station spot transaction strategy simulation and market optimization operation method based on the master-slave game according to any one of claims 1 to 3.

Images