Electricity-carbon collaborative transaction simulation system and method based on adaptive multiple subjects

Abstract

The invention discloses an electricity-carbon collaborative transaction simulation system and method based on adaptive multiple subjects. The system comprises an acquisition module, an adjustment module, an updating module and a simulation module. The acquisition module is used for acquiring a first quotation of a power generation enterprise in a power bilateral negotiation market, a second quotation in a power centralized bidding market, a third quotation in a first-level carbon market, a fourth quotation in a second-level bulk trading carbon market, a fifth quotation in a second-level listed trading carbon market and a transaction average price of a CCER market; the adjusting module is used for adjusting the first quotation and the fourth quotation respectively; the updating module is used for updating the second quotation and the third quotation; the simulation module is used for carrying out analog simulation on the transaction; and the quotation is dynamically adjusted, so that all transactions of planned quantity reporting can be completed as much as possible. By constructing a strategy updating model, multiple associations are generated among the electricity market, the carbon market and the CCER market on the price formation mechanism level.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity-carbon trading simulation, and in particular to an electricity-carbon collaborative trading simulation system and method based on adaptive multi-agents. Background Art

[0002] With the market-oriented reform of my country's electricity market and the improvement and optimization of the trading and settlement mechanisms of the carbon market and CCER market, the linkage and spillover effects among the electricity market, carbon market and CCER market have been strengthened, and the price transmission mechanism between different markets has affected the bidding behavior and profit function of market participants.

[0003] Existing electricity-carbon collaborative trading simulation systems usually only focus on the collaborative trading between the electricity market and the carbon market, or between the electricity market and the CCER market, and do not incorporate the electricity market, carbon market, and CCER market into a unified collaborative trading framework.

[0004] Furthermore, the existing electricity-carbon collaborative trading simulation system rarely involves dynamic bidding games among market participants, and the bidding function adopted cannot reflect the actual bidding adjustment rules. Summary of the invention

[0005] The existing electricity-carbon collaborative trading simulation system has not incorporated the electricity market, carbon market, and CCER market into a unified collaborative trading framework, and rarely involves dynamic bidding games among market participants. The bidding function it adopts cannot reflect the bidding adjustment rules of the frame number.

[0006] In response to the above problems, a simulation system and method for electricity-carbon collaborative trading based on adaptive multi-agents is proposed. By constructing a quotation adjustment model, the quotation of the buyer and seller can be dynamically adjusted according to the completed quotation volume in the bilateral negotiation market or the secondary bulk carbon market, so as to complete all transactions of the planned quotation volume as much as possible. By constructing the first quotation model and the second quotation model, the carbon quota price and CCER price are incorporated into the cost function of the power generation enterprise as variables; the bidding results of the medium- and long-term electricity market are incorporated into the carbon market quotation function as variables through the third quotation model; and the carbon market clearing results are incorporated into the determination function of the average CCER transaction price as variables through the fifth construction unit. These quotation models and construction units reflect the multiple connections between the electricity market and the carbon market and the CCER market at the level of price formation mechanism.

[0007] In the first aspect, a simulation system for electricity-carbon collaborative trading based on adaptive multi-agents includes:

[0008] Get module;

[0009] Adjustment module;

[0010] Update module;

[0011] Simulation module;

[0012] The acquisition module is used to obtain the first quotation of the power generation enterprise in the power bilateral negotiation market, the second quotation in the power centralized bidding market, the third quotation in the primary carbon market, the fourth quotation in the secondary bulk trading carbon market, the fifth quotation in the secondary listed trading carbon market, and the average transaction price of CCER;

[0013] The adjustment module is used to adjust the first quotation and the fourth quotation respectively according to the completed quotation quantities of the buyer and the seller in the power bilateral negotiation market and the secondary bulk transaction carbon market, so as to complete the transaction;

[0014] The updating module is used to construct a strategy updating model according to a reinforcement learning algorithm, and use the strategy updating model to update the second quotation and the third quotation of the centralized power bidding market and the primary carbon market respectively, so as to complete the transaction;

[0015] The simulation module is used to simulate transactions in the electricity bilateral negotiation market, the electricity centralized bidding market, the primary carbon market, the secondary bulk trading carbon market, the secondary listed trading carbon market and the CCER market;

[0016] The first quotation is the quotation of the power generation enterprise in the bilateral electricity negotiation market, the second quotation is the quotation of the power generation enterprise in the centralized electricity bidding market, the third quotation is the quotation of the power generation enterprise in the primary carbon market, the fourth quotation represents the quotation of the power generation enterprise in the secondary bulk trading carbon market, and the fifth quotation is the quotation of the power generation enterprise in the secondary listed trading carbon market;

[0017] The secondary listed carbon market adopts the best price matching algorithm for order matching.

[0018] The fourth quotation in the secondary bulk carbon market and the fifth quotation in the secondary listed carbon market are both within the range of (1-α) to (1+α) of the clearing price in the primary carbon market.

[0019] In conjunction with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the first aspect of the present invention, in a first possible implementation, the acquisition module includes:

[0020] first update unit;

[0021] a first building block;

[0022] The first updating unit is used to update the estimated carbon price based on the clearing price of the primary carbon market in the previous year and the average of the CCER price in the previous year;

[0023] The first updating unit is used to update the clearing price pc of the primary carbon market in the previous year. t-1 and the CCER price P of the previous yearccer，t-1 The average value of the estimated carbon price ecp t , then there is,

[0024] The first construction unit is used to construct a first quotation model according to the estimated carbon price, the carbon emissions per unit of power generation, the carbon quota benchmark value per unit of power generation, and the auction ratio of the primary carbon market:

[0025] P sj，t，min =m j，t +(n j，t -B j，t (1-μ))ecp t (1)

[0026] Among them, P sj，t，min is the lower limit of the power plant's bid in the bilateral electricity negotiation market, and the corresponding upper limit is recorded as P sj，t，max , satisfying P sj，t，max =δ sj，t P sj，t，min , where δ sj，t >1. The bids of power plants in the bilateral electricity negotiation market meet P sj，t，min and P sj，t，max The lower and upper limits of .

[0027] ecp t is the estimated carbon price, m j，t is the unit power generation cost, n j，t is the carbon emission per unit of electricity generation, B j，t is the carbon quota benchmark value per unit of electricity generation, μ is the auction ratio of the primary carbon market;

[0028] The first quotation model is used to calculate and obtain a first quotation in the electricity bilateral negotiation market.

[0029] In conjunction with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the first aspect of the present invention, in a second possible implementation, the acquisition module further includes:

[0030] a second building block;

[0031] The second construction unit is used to construct a second quotation model pebid according to the estimated carbon price, carbon emissions per unit of power generation, carbon quota benchmark value per unit of power generation, auction ratio of the primary carbon market, and expected profit margin. j，t :

[0032] Ecost j，t =m j，t +ecp t ·max{n j，t -B j，t(1-μ), 0} (2),

[0033] pebid j，t =Ecost j，t (1+margin j ) (3),

[0034] Among them, Eco j，t is the expected unit power generation cost of the power generation enterprise, ecp t is the estimated carbon price, m j，t is the unit power generation cost, n j，t is the carbon emission per unit of electricity generation, B j，t is the carbon quota benchmark value per unit of electricity generation, μ is the primary carbon market auction ratio, margin j is the expected profit margin of the power generation enterprise;

[0035] The second quotation model pebid j，t Used to calculate and obtain the second bid of power generation enterprises in the centralized electricity bidding market.

[0036] In combination with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the first aspect of the present invention, in a third possible implementation, the acquisition module further includes:

[0037] The third building block;

[0038] The third construction unit is used to construct a third quotation model pv according to the carbon emissions per unit of power generation, the unit power generation cost and the average electricity price of the power generation enterprise participating in the power market j，t :

[0039]

[0040] Among them, n j，t is the carbon emission per unit of electricity generated, m j，t is the unit power generation cost, pe j，t is the average electricity price of power generation enterprise j participating in the power market;

[0041] The third quotation model pv j，t Used to obtain third-party quotes for calculating the primary carbon market.

[0042] In combination with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the first aspect of the present invention, in a fourth possible implementation, the acquisition module further includes:

[0043] a fourth building block;

[0044] The fourth construction unit is used to construct a fifth quotation model according to the primary carbon market clearing price and the carbon price control ratio:

[0045] (1-α)pc t ≤pc2 t ≤(1+α)pc t (5)

[0046] Calculate the fourth quotation of the secondary bulk carbon market and the fifth quotation of the secondary listed carbon market using the fifth quotation model t :

[0047] Among them, PC t is the clearing carbon price in the primary carbon market, and α is the carbon price control ratio.

[0048] In combination with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the first aspect of the present invention, in a fifth possible implementation, the acquisition module further includes:

[0049] The fifth building block;

[0050] The fifth construction unit constructs a CCER transaction average price model p based on the clearing price of the primary carbon market and the carbon quota demand gap of the power generation industry. ccer，t :

[0051]

[0052] Among them, pc t is the clearing carbon price in the primary carbon market, N ccer，t Indicates the number of CCER certificates that can be traded by power generation enterprises in each period, C gop，t To meet the shortfall in carbon quota demand in the power generation industry.

[0053] In combination with the first possible implementation manner of the first aspect of the present invention, in a sixth possible implementation manner, the adjustment module includes:

[0054] The sixth building block;

[0055] The sixth construction unit is used to construct a quotation adjustment model according to the traded declared quantities, discount coefficients and total declared quantities of the buyer and the seller, so as to adjust the first quotation and the fourth quotation of the buyer and the seller in the power bilateral negotiation market and the secondary bulk transaction carbon market;

[0056] Quote adjustment model:

[0057] P bi，t (r) = (1-α bi，t (r))P bi，t，min +α bi，t (r)P bi，t，max (7)

[0058] P sj，t (r) = α sj，t (r)Psj，t，min +(1-α sj，t (r))P sj，t，max (8)

[0059] Among them, P bi，t (r) represents the buyer’s bid in round r, P sj，t (r) represents the seller’s bid in round r. bi，t，min and P bi，t，max Represent the lower and upper limits of the buyer's offer, P sj，t，min and P sj，t，max Respectively represent the lower and upper limits of the seller's offer. bi，t (r) and α sj，t (r) represent the discount coefficients of the buyer's and seller's quotations, respectively, satisfying:

[0060]

[0061] Where R represents the maximum negotiation round, k bi，t (r) and k sj，t (r) represents the traded volume of buyers and sellers in the rth round, respectively, N bi，t and N sj，t Represents the total declared amount of buyers and sellers, and represent the initial discount factors for buyers and sellers respectively.

[0062] In combination with the second possible implementation manner of the first aspect of the present invention, in a seventh possible implementation manner, the update module includes:

[0063] seventh building block;

[0064] The eighth building block;

[0065] The seventh construction unit is used to construct a preference selection model;

[0066] The eighth construction unit is used to construct a strategy update model according to the preference selection model and the profit of the power generation enterprise in the previous year;

[0067] The strategy update model is used to update the second quotation of the power generation enterprise in the centralized power bidding market and the third quotation in the primary carbon market;

[0068] Propensity Choice Model prop x，j，t :

[0069] prop x，j，t =(1-α)prop x，j，t-1 +φ x，j，t-1 (11)

[0070] Among them, x is the strategy selection parameter, t is the annual parameter, j is the power generation enterprise parameter, and α is the time-proximity parameter. x，j，t-1 satisfy:

[0071]

[0072] Among them, β is the experimental parameter, π j，t-1 is the profit of power generation in the previous year; the strategy update model is as follows:

[0073]

[0074] If x j，t =1:

[0075]

[0076] If x j，t =2:

[0077]

[0078] If x j，t =3:

[0079]

[0080] Among them, t is the annual parameter, pebid j，t+1 The second bid of power generation enterprise j in the centralized power bidding market in the second year, pcbid j，t+1 The third quotation of power generation enterprise J in the primary carbon market in the second year, peb t is the equilibrium price of the previous year, pc t It is the clearing carbon price of the primary carbon market in the previous year.

[0081] In a second aspect, a method for simulating electricity-carbon collaborative trading based on adaptive multi-agents is provided, which adopts the electricity-carbon collaborative trading simulation system described in the first aspect, and includes:

[0082] Step 100: Obtain the first quotation of the power generation enterprise in the power bilateral negotiation market, the second quotation in the power centralized bidding market, the third quotation in the primary carbon market, the fourth quotation in the secondary bulk trading carbon market, the fifth quotation in the secondary listed trading carbon market, and the average transaction price in the CCER market;

[0083] Step 200: In the power bilateral negotiation market and the secondary bulk carbon market, the first quotation and the fourth quotation are adjusted according to the completed quotation quantities of the buyer and the seller, respectively, so as to complete the transaction;

[0084] Step 300: updating the second quotation and the third quotation according to the reinforcement learning strategy to complete the transaction;

[0085] Step 400: Simulate transactions in the electricity bilateral negotiation market, the electricity centralized bidding market, the primary carbon market, the secondary bulk trading carbon market, the secondary listed trading carbon market, and the CCER market;

[0086] Among them, the first quotation is the quotation of the power generation enterprise in the electricity bilateral negotiation market, the second quotation is the quotation of the power generation enterprise in the electricity centralized bidding market, the third quotation is the quotation of the power generation enterprise in the primary carbon market, the fourth quotation represents the quotation of the power generation enterprise in the secondary bulk trading carbon market, and the fifth quotation is the quotation of the power generation enterprise in the secondary listed trading carbon market.

[0087] In conjunction with the adaptive multi-agent-based electricity-carbon collaborative trading simulation method described in the second aspect of the present invention, in a first possible implementation, step 100 includes:

[0088] Step 110: Update the estimated carbon price based on the average of the clearing price of the primary carbon market in the previous year and the CCER price in the previous year.

[0089] Step 120: construct a first quotation model according to the estimated carbon price, the carbon emission per unit of electricity generation, the carbon quota benchmark value per unit of electricity generation, and the primary carbon market auction ratio;

[0090] Step 130: Calculate a first quotation in the power bilateral negotiation market according to the first quotation model;

[0091] In conjunction with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the second aspect of the present invention, in a second possible implementation, step 100 further includes:

[0092] Step 140: construct a second quotation model based on the estimated carbon price, the carbon emissions per unit of electricity generation, the carbon quota benchmark value per unit of electricity generation, and the primary carbon market auction ratio:

[0093] Step 150: Calculate the second bid of the power generation enterprise in the centralized power bidding market according to the second bidding model.

[0094] In conjunction with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the second aspect of the present invention, in a third possible implementation, step 100 further includes:

[0095] Step 160: construct a third quotation model based on the carbon emissions per unit of electricity generation, the unit electricity generation cost, and the average electricity price of the power generation enterprises participating in the electricity market:

[0096] Step 170: Calculate the third quotation of the primary carbon market according to the third quotation model.

[0097] In conjunction with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the second aspect of the present invention, in a fourth possible implementation, step 100 further includes:

[0098] Step 180: Calculate the fourth quotation and the fifth quotation of the secondary bulk trading carbon market and the secondary listed trading carbon market respectively according to the clearing price of the primary carbon market and the carbon price control ratio.

[0099] In conjunction with the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the second aspect of the present invention, in a fifth possible implementation, step 100 further includes:

[0100] Step 190: Calculate the average transaction price of CCER based on the clearing price of the primary carbon market, the carbon emissions of power generation companies, the carbon quotas obtained free of charge by power generation companies, and the carbon quotas obtained by power generation companies through auction in the primary carbon market.

[0101] Implementing the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method described in the present invention, by constructing a quotation adjustment model, it is possible to dynamically adjust the quotations of buyers and sellers based on the completed quotations in the bilateral negotiation market or the secondary bulk carbon market, so as to complete all transactions of the planned quotations as much as possible. By constructing the first quotation model and the second quotation model, the carbon quota price and the CCER price are incorporated as variables into the cost function of the power generation enterprise; the bidding results of the medium- and long-term electricity market are incorporated as variables into the carbon market quotation function through the third quotation model; and the carbon market clearing results are incorporated as variables into the determination function of the CCER average transaction price through the fifth construction unit. These quotation models and construction units reflect the multiple connections between the electricity market and the carbon market and the CCER market at the level of price formation mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0103] Figure 1 It is a schematic diagram of the module composition and connection of the electric-carbon collaborative trading simulation system and method based on adaptive multi-agent in the present invention;

[0104] Figure 2 It is a schematic diagram of the composition and connection of the acquisition module in the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method of the present invention;

[0105] Figure 3It is a schematic diagram of the composition and connection of the adjustment module in the self-adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method of the present invention;

[0106] Figure 4 This is a schematic diagram of the composition and connection of the update module in the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method of the present invention. DETAILED DESCRIPTION

[0107] The following will be combined with the accompanying drawings in the invention to clearly and completely describe the technical solutions in the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0108] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0109] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element.

[0110] It should be understood that the orientation or position relationship indicated by terms such as "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0111] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0112] like Figure 1 , Figure 2 and Figure 3 , Figure 4 . Figure 1 This is a schematic diagram of the module composition and connection of the self-adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method in the present invention. Figure 2 It is a schematic diagram of the composition and connection of the acquisition module in the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method of the present invention; Figure 3 It is a schematic diagram of the composition and connection of the adjustment module in the self-adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method of the present invention; Figure 4 This is a schematic diagram of the composition and connection of the update module in the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method in the present invention. During the simulation, the user can set the following parameters: simulation years, number of power generation companies, number of electricity buyers, initial carbon price, primary carbon market auction ratio, carbon price control ratio, maximum round of electricity bilateral negotiation market, maximum negotiation round of secondary bulk trading carbon market, listing application ratio, CCER carbon offset ratio, excess emission penalty coefficient, emission reduction cost coefficient, and active emission reduction ratio.

[0113] Furthermore, for power generation company j, its unit power generation cost m in the first year is generated: j，1 , carbon emissions per unit of electricity generation n j，1 , Carbon quota benchmark value per unit of electricity generation B j，1 , power generation j，t and margin j ; For electricity purchaser i, generate its electricity demand Q bi，t ; Set the primary carbon market auction ratio μ. At the end of each compliance period, verify that each power generation company’s CO2 emissions are C j，t =n j，t ge j，t , using the baseline method to allocate quotas, the free quota of power generation company j is Free j，t =(1-μ)B j，t ge j，t The government determines the auction quota in the primary carbon market as TC according to the auction ratio. t =∑ j μ·C j，t Since the minimum unit of carbon quota trading is tons, it is necessary to convert Free j，t With TC t Round to the nearest integer.

[0114] To ensure market stability and avoid speculation, it is assumed that the government sets the carbon trading price pc2 t Control the fluctuation within a certain range of the clearing price of the primary carbon market, that is, (1-α)pc t ≤pc2 t ≤(1+α)pc t , where pc tis the clearing carbon price in the primary carbon market, and α is the fluctuation range.

[0115] When α=0, it indicates that the most stringent carbon price control is adopted, requiring the secondary carbon market price to be equal to the primary carbon market clearing price; when α takes a very large value, it can be regarded as no carbon price control; under normal circumstances, α=15% can be taken.

[0116] Generate a scaling factor λ between [0.3, 0.7] sj，t , thereby determining the declared power N of power generation enterprise j in the bilateral negotiated power market sj，t , and N sj，t Satisfy N sj，t =λ sj，t ge j，t Generate a scaling factor λ between [0.3, 0.7] bi，t , thereby determining the electricity demand N reported by the electricity purchaser to participate in the bilateral negotiated electricity market bi，t =λ bi，t Q bi，t . Realize bilateral negotiation and bidding in the electricity market based on dynamic game.

[0117] If one of the following three conditions is met, the entire process of bilateral negotiation and bidding in the power market will be withdrawn: 1. The declared power N of all power generation enterprises in the bilateral negotiation market sj，t All have been sold; 2. The electricity demand reported by all electricity buyers in the bilateral negotiation market N bi，t All satisfied; 3. The maximum number of negotiation rounds R is reached e .

[0118] like Figure 3 , Figure 3 This is a schematic diagram of the composition and connection of the adjustment module in the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method of the present invention.

[0119] After the bidding process is completed, the total transaction volume of the bilateral negotiation market is obtained The weighted average is the average transaction price in the bilateral negotiation market. For each power generation enterprise, the weighted average price of its transaction in the bilateral negotiated electricity market is obtained and trading volume For each electricity buyer, the weighted average price of its transaction in the bilateral negotiated electricity market is obtained: and trading volume

[0120] While conducting bilateral negotiation and bidding in the electricity market, power generation enterprises and electricity buyers conduct transactions in the centralized bidding electricity market. First, calculate the total demand TE of electricity buyers in the centralized bidding market. b，t =∑ i Q bi，t(1-λb i，t The electricity quantity reported by power generation enterprises in the centralized bidding market is the power generation capacity of the power generation enterprise. j，t The remaining amount after deducting the electricity volume reported in the bilateral negotiation market, i.e. qebid j，t =(1-λs j，t )ge j，t , the expected unit power generation cost of the power generation enterprise is Ecoss j，t =m j，t +ecp t ·max{n j，t -B j，t (1-μ), 0}, for the first period (first year), the bid price of the power generation enterprise in the centralized bidding market is pebid j，t =Ecost j，t (1+margin j ); for subsequent periods, the quotes are updated according to the reinforcement learning strategy model.

[0121] The clearing price of the centralized power bidding market adopts the queuing method. All power generation companies are ranked in order from low to high according to their bids (j = 1, 2, ..., n). j＝1 qebid j，t Reach TE b，t When t For power generation company j, if its bid price pebid j，t >peb t , the bid fails. The amount of electricity actually sold by the power generation enterprise in the centralized bidding market is qeb j，t , for pebid j，t <peb t The power generation enterprises, j，t =qebid j，t ; for pebid j，t =peb t The power generation enterprises of these power generation enterprises j，t Need to be allocated proportionally (you can set the allocation function, assuming that for pebid j，t <peb t The total amount of power generation enterprises reported is A, and for pebid j，t =peb t The total amount of power generation enterprises reported is B. For pebid j，t =peb t The power generation enterprise sells qeb of electricity through auction in the centralized bidding market. j，t =qebid j，t *(TE b，t-A) / B). Record the price peb traded by power generation enterprises in the centralized market t and quantity j，t In particular, if the sum of the electricity declared by all power generation enterprises in the centralized power bidding market is less than the sum of the electricity demand of the power purchasers in the centralized power bidding market, that is, ∑ j＝1 qebid j，t <TE b，t When the market clearing price is set to the maximum price declared by the power generation enterprise, that is, peb t =max j {pebid j，t}.

[0122] For each power generation enterprise, the total amount of electricity sold in the medium and long-term electricity market is obtained. j，t , is the sum of the transaction volumes of the bilaterally negotiated power market and the centralized bidding power market, that is

[0123] According to the formula Calculate the upper bound pv of the price declared by each power generation enterprise in the primary carbon market j，t , where pe j，t is the average electricity price of power generation enterprise j participating in the power market, that is, the weighted average electricity price of the bilateral negotiation market and the centralized bidding market. The bid price given by each power generation enterprise in the primary carbon market is not greater than the reserve price, that is, pcbid j，t ≤pv j，t For the first year, the price is between 0 and pv j，t For subsequent years, the quotes are adjusted through the reinforcement learning strategy model.

[0124] Power generation enterprises apply for carbon quotas in the primary carbon market based on their actual power generation. ,in In the primary carbon market, all power generation companies are ranked in descending order according to their bids (j=1, 2, ..., n). When the cumulative carbon quota ∑ j＝1 qcbid j，t Reach the total auction quota TC of the primary market t When the equilibrium price pc is obtained t The auction adopts a unified price auction, and all power generation companies bid at the equilibrium price pc t For power generation company j, if its bid price pcbid j，t <pc t , the bid fails. The carbon quota that power generation enterprises apply for in the primary carbon market is qc j，t , for pcbid j，t >pc t The power generation enterprises, their qcj，t =pcbid j，t ; For pcbid j，t =pc t The power generation enterprises of these power generation enterprises j，t Need to be allocated proportionally (you can set the allocation function, assuming that for pcbid j，t >pc t The total amount of carbon quotas declared by power generation enterprises is A, and for pcbid j，t >pc t The total amount of carbon quotas declared by power generation enterprises is B. j，t =pc t The carbon quota obtained by power generation enterprises from the primary carbon market through auction is qc j，t =qcbid j，t *(TC t -A) / B). In particular, if the total carbon quota declared by all power generation enterprises is less than the total auction quota in the primary carbon market, that is, ∑ j＝1 qcbid j，t <TC t When , the market clearing price is set to the minimum price declared by the power generation enterprises, that is, pc t =min j {pcbid j，t}.

[0125] Generate trading quotas for power generation companies in the secondary bulk carbon market when It means that power generation enterprise j is the seller and sells quotas in the bulk trading market; This means that power generation enterprise j is the buyer and buys quotas in the bulk trading market. Both the buyer and the seller need to round up the amount. For example, if Then the maximum number of tons sold is 2,000, that is, -2,000. Then you need to purchase 1001 tons to meet the carbon emission gap, that is, 1001, rounded up.

[0126] The bids of buyers and sellers in the bulk trading market are subject to price controls that meet the upper and lower limits of the primary carbon market clearing price, i.e. p2bid min =(1-α)pc t , p2bid max =(1+α)pc t For the buyer For sellers For buyers, the volume quoted in the secondary bulk carbon market is For sellers, the volume reported in the secondary bulk carbon market is Repeat the bilateral negotiation and bidding process to continuously update k bi，t (r), k sj，t (r), α bi，t (r), α sj，t (r) until one of the following three conditions is met: 1. All sellers’ declared carbon quotas N in the bulk trading market sj，t are no longer greater than or equal to 100,000 tons; 2. The carbon quota demand of all buyers in the bulk trading market N bi，T No longer greater than or equal to 100,000 tons; 3. The maximum number of negotiation rounds R has been reached block , which is the maximum negotiation round of the bulk trading carbon market set when the program is running.

[0127] like Figure 3 , Figure 3 This is a schematic diagram of the composition and connection of the adjustment module in the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method of the present invention.

[0128] After the entire process of bulk trading is completed, the total trading volume of the secondary bulk trading carbon market is obtained. The average transaction price of the secondary bulk carbon market is obtained by weighted average By recording the transaction volume of each participant in each round, the remaining demand of each participant can be obtained. or supply Calculate the transaction volume of each participant in the entire block transaction process and weighted average transaction price

[0129] For the secondary listed carbon market, the listing declaration ratio is set to split the supply and demand of quotas with an absolute number greater than or equal to 100,000 tons to meet the trading rules of the listed agreement carbon market. Undeclared listing buy and sell orders select the best-priced counterparty from the listing orders. If multiple undeclared buying and selling orders choose the same listing sell order or listing buy order, an undeclared intention order is randomly matched. After the entire listing transaction process is completed, the total transaction volume of the secondary listed carbon market is obtained. The weighted average is the average transaction price of the listed transactions. Get the remaining demand of each participant or supply Calculate the transaction volume of each participant in the entire listing transaction process and weighted average transaction price

[0130] For CCER transactions, N ccer，t Indicates the number of CCER certificates that can be traded by power generation companies in each period. powerrepresents the number of power generation enterprises set, then the number of CCER certificates per period ranges from [λ1*N power ,λ2*N power ], where 0<λ1<λ2. The benchmark price of CCER Equal to the clearing price of the primary carbon market, that is, However, the actual price will be adjusted in real time according to the number of CCER certificates and the carbon quota gap of enterprises. gap，t =∑ j (C j，t -Fee j，t -qc j，t ), which is the carbon emission gap remaining after deducting the carbon quota obtained from the free and actual auctions in the primary carbon market from the total carbon emissions. The average transaction price of CCER can be expressed by the following continuous function:

[0131]

[0132] P ccer，t This function has the following properties:

[0133] 1. When C gap，t <0: When C gap，t >0 o'clock, If and only if C gap，t = 0,

[0134] 2.P ccer，t Follow C gap，t Monotonically increasing;

[0135] 3. When C gap，t >0, P ccer，t Follow C gap，t The rate of increase slows down with the increase of gap，t When P < 0 ccer，t Follow C gap，t The rate of decrease slowed down.

[0136] ρ ccer Indicates the carbon offset ratio of CCER, usually 5%. j，t -Free j，t ≤0, the number of CCERs actually declared by the power generation enterprise is 0; conversely, if the power generation enterprise j meets C j，t -Free j，t >0, the actual number of CCERs declared by the power generation enterprise is: in represents the ceiling function, represents the floor function. If all The sum is less than or equal to N ccer，t , then the CCER certificates actually purchased by each power generation enterprise On the contrary, if all The sum is greater than N ccer，t , then the CCER certificates actually purchased by each power generation enterprise need to be allocated in proportion, that is,

[0137] In order to punish illegal emissions, the government imposes fines on power generation companies for emissions that exceed the quota allowed, i.e. For buyers in the secondary listed carbon market, For sellers in the secondary carbon market, Penalty j，t is the penalty amount imposed by the government on power generation enterprise j, ξ is the penalty coefficient for excess emissions, qc j，t It is the quota successfully bid by power generation enterprise j in the primary carbon market.

[0138] For buyers in the secondary bulk carbon market, For sellers in the secondary bulk carbon market, For buyers in the secondary listed carbon market, For sellers in the secondary listed carbon market, The total profit of power generation enterprise j can be calculated as follows:

[0139]

[0140] like Figure 4 , Figure 4 The schematic diagram of the composition and connection of the update module in the adaptive multi-agent-based electricity-carbon collaborative trading simulation system and method of the present invention is shown in Figure 1. It is defined that the power generation enterprise j adjusts the bid price pebid in the next stage in the centralized power bidding market according to the strategy x∈{1,2,3} j，t+1 And the primary carbon market quotation pebid j，t+1 :

[0141] If x=1:

[0142]

[0143] If x = 2:

[0144]

[0145] If x = 3:

[0146]

[0147] For the bidding strategy selection of power generation enterprises in the second year, define:

[0148]

[0149] r x is a random number on [0,1]. For a randomly generated r x , determine its second-year strategy choice x j，t＝2 , thus combining the clearing price peb1 of the centralized electricity quotation market in the first year and the quotation pebid of power generation enterprise j in the centralized electricity quotation market in the first year j，1 Determine the price pebid of power generation company j in the centralized power quotation market in the second year j，2 Similarly, according to the strategy of the second year, choose x j，t＝2 , combined with the equilibrium carbon price pc1 in the primary carbon market in the first year and the bid pcbid of power generation enterprise j in the primary carbon market in the first year j，1 , determine the bid price of power generation enterprise j in the primary carbon market in the second year pcbid j，2 .

[0150] For the tth year (t=3, 4, ...), define the propensity model prop of the bidding strategy x (x∈{1, 2, 3}) of power generation enterprise j x，j，t ,satisfy:

[0151] prop x，j，t =(1-α)prop x，j，t-1 +φ x，jt-1 (twenty three),

[0152] Where α is the time-close parameter. x，j，t-1 satisfy:

[0153]

[0154] Where β is the experimental parameter. The quotation strategy for the second year can be taken as

[0155] Normalize the selection strategy according to the propensity model:

[0156]

[0157] The strategy update model for power generation enterprises to choose the tth year (t = 3, 4, ...) is:

[0158]

[0159] r x is a random number on [0, 1]. x And the strategy selection probability prob x，j，t , we can get the corresponding strategy selection x j，t, and then get the price pebid of the power generation enterprise in the centralized power bidding market j，t And the price quotes in the primary carbon market pcbid j，t .

[0160] The benchmark carbon quota for power generation enterprises is based on B j，t =0.95*B j，t-1 Update, power generation companies can choose strategies:

[0161] If k = 1, then n j，t =n j，t-1 , m j，t =m j，t-1

[0162] If k = 2, then n j，t =0.95*n j，t-1 ,

[0163] in, Represents the emission reduction cost coefficient. If the power generation enterprise is penalized for excessive emissions in year t-1, Penalty j，t-1 > 0, then k j，t =2.

[0164] At the end of each year, a random number r is randomly generated from a uniform distribution on (0, 1) k , when r k <ρ k (0≤ρ k ≤1), k j，t =2, simulating the behavior of power generation companies to actively reduce emissions, and ρ k It is the proportion of active emission reduction.

[0165] At the end of the operation, the price and volume simulation results of medium- and long-term electricity trading, carbon trading, CCER trading, and the overall carbon emissions of the power generation industry are exported, and the parameters set by the user and the operation log are output.

[0166] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An adaptive multi-agent-based electricity-carbon collaborative trading simulation system, characterized by comprising: Get module; Adjustment module; Update module; Simulation module; The acquisition module is used to obtain the first quotation of the power generation enterprise in the power bilateral negotiation market, the second quotation in the power centralized bidding market, the third quotation in the primary carbon market, the fourth quotation in the secondary bulk trading carbon market, the fifth quotation in the secondary listed trading carbon market, and the average transaction price in the CCER market; The adjustment module is used to adjust the first quotation and the fourth quotation respectively according to the completed quotation quantities of the buyer and the seller in the power bilateral negotiation market and the secondary bulk transaction carbon market, so as to complete the transaction; The updating module is used to construct a strategy updating model according to the reinforcement learning strategy, and use the strategy updating model to update the second quotation and the third quotation so as to complete the transaction; The simulation module is used to simulate transactions in the power bilateral negotiation market, the power centralized bidding market, the primary carbon market, the secondary bulk trading carbon market, the secondary listed trading carbon market and the CCER market; wherein the first quotation is the quotation of the power generation enterprise in the power bilateral negotiation market, the second quotation is the quotation of the power generation enterprise in the power centralized bidding market, the third quotation is the quotation of the power generation enterprise in the primary carbon market, the fourth quotation represents the quotation of the power generation enterprise in the secondary bulk trading carbon market, and the fifth quotation is the quotation of the power generation enterprise in the secondary listed trading carbon market; The secondary listed carbon market adopts the best price matching algorithm for order matching.

2. The self-adaptive multi-agent-based electricity-carbon collaborative trading simulation system according to claim 1 is characterized in that: The acquisition module comprises: first update unit; a first building block; The first updating unit is used to update the estimated carbon price according to the clearing price of the primary carbon market in the previous year and the average transaction price of CCER in the previous year; The first construction unit is used to construct a first quotation model according to the estimated carbon price, the carbon emissions per unit of power generation, the carbon quota benchmark value per unit of power generation, and the auction ratio of the primary carbon market; The first quotation model is used to calculate and obtain a first quotation in the electricity bilateral negotiation market.

3. The self-adaptive multi-agent-based electricity-carbon collaborative trading simulation system according to claim 1 is characterized in that: The acquisition module also includes: a second building block; The second construction unit is used to construct a second quotation model according to the estimated carbon price, the carbon emissions per unit of power generation, the carbon quota benchmark value per unit of power generation, the auction ratio of the primary carbon market, and the expected profit margin; The second bidding model is used to calculate and obtain the second bid of the power generation enterprise in the centralized power bidding market.

4. The adaptive multi-agent-based electricity-carbon collaborative trading simulation system according to claim 1 is characterized in that: The acquisition module also includes: The third building block; The third construction unit is used to construct a third quotation model according to the carbon emissions per unit of power generation, the unit power generation cost, and the average electricity price of the power generation enterprise participating in the power market; The third quotation model is used to calculate and obtain the third quotation of the primary carbon market.

5. The adaptive multi-agent-based electricity-carbon collaborative trading simulation system according to claim 1 is characterized in that: The acquisition module also includes: a fourth building block; The fourth construction unit is used to construct a fourth quotation model according to the primary carbon market clearing price and the carbon price control ratio; The fourth quotation model is used to calculate and obtain the fourth quotation of the secondary bulk trading carbon market and the fifth quotation of the secondary listed trading carbon market.

6. The self-adaptive multi-agent-based electricity-carbon collaborative trading simulation system according to claim 1 is characterized in that: The acquisition module also includes: The fifth building block; The fifth construction unit is used to obtain the carbon quota gap of the entire power generation industry based on the carbon emissions of the power plant and the carbon quota obtained in the primary carbon market auction, and to construct a CCER transaction average price model based on the number of tradable CCERs in circulation and the clearing price of the primary carbon market; The CCER average transaction price model is used to calculate the average transaction price of the CCER market.

7. The self-adaptive multi-agent-based electricity-carbon collaborative trading simulation system according to claim 1 is characterized in that: The adjustment module comprises: The sixth building block; The sixth construction unit is used to construct a quotation adjustment model according to the traded declared quantities, discount coefficients and total declared quantities of buyers and sellers to adjust the first and fourth quotations of buyers and sellers in the electricity bilateral negotiation market and the secondary bulk carbon market.

8. The self-adaptive multi-agent-based electricity-carbon collaborative trading simulation system according to claim 1 is characterized in that: The update module includes: seventh building block; The eighth building block; The seventh construction unit is used to construct a preference selection model; The eighth construction unit is used to construct a strategy update model according to the preference selection model and the profit of the power generation enterprise in the previous year; The strategy update model is used to update the power generation enterprise's third quotation in the centralized power bidding market and the fourth quotation in the primary carbon market.

9. A simulation method for electricity-carbon collaborative trading based on adaptive multi-agent, using the simulation system according to any one of claims 1 to 8, characterized in that: include: Step 100: Obtain the first quotation of the power generation enterprise in the power bilateral negotiation market, the second quotation in the power centralized bidding market, the third quotation in the primary carbon market, the fourth quotation in the secondary bulk trading carbon market, the fifth quotation in the secondary listed trading carbon market, and the average transaction price in the CCER market; Step 200: In the power bilateral negotiation market and the secondary bulk carbon market, the first quotation and the fourth quotation are adjusted according to the completed quotation quantities of the buyer and the seller, respectively, so as to complete the transaction; Step 300: updating the second quotation and the third quotation according to the reinforcement learning algorithm to complete the transaction; Step 400: Simulate transactions in the electricity bilateral negotiation market, the electricity centralized bidding market, the primary carbon market, the secondary bulk trading carbon market, the secondary listed trading carbon market, and the CCER market; Among them, the first quotation is the quotation of the power generation enterprise in the electricity bilateral negotiation market, the second quotation is the quotation of the power generation enterprise in the electricity centralized bidding market, the third quotation is the quotation of the power generation enterprise in the primary carbon market, the fourth quotation represents the quotation of the power generation enterprise in the secondary bulk trading carbon market, and the fifth quotation is the quotation of the power generation enterprise in the secondary listed trading carbon market.

10. The method for simulating electricity-carbon collaborative trading based on adaptive multi-agent according to claim 9 is characterized in that: The step 100 comprises: Step 110: Update the estimated carbon price based on the average of the clearing price of the primary carbon market in the previous year and the CCER price in the previous year. Step 120: construct a first quotation model according to the estimated carbon price, the carbon emission per unit of electricity generation, the carbon quota benchmark value per unit of electricity generation, and the primary carbon market auction ratio; Step 130: Calculate a first quotation in the electricity bilateral negotiation market according to the first quotation model.

Images