CN114169916A - Market member quotation strategy making method suitable for novel power system - Google Patents

Market member quotation strategy making method suitable for novel power system Download PDF

Info

Publication number
CN114169916A
CN114169916A CN202111360571.2A CN202111360571A CN114169916A CN 114169916 A CN114169916 A CN 114169916A CN 202111360571 A CN202111360571 A CN 202111360571A CN 114169916 A CN114169916 A CN 114169916A
Authority
CN
China
Prior art keywords
market
power
constraint
load
climbing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111360571.2A
Other languages
Chinese (zh)
Inventor
张粒子
唐成鹏
王进
肖艳炜
邓晖
陆承宇
孙珂
周子青
章枫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Zhejiang Electric Power Co Ltd
North China Electric Power University
Electric Power Research Institute of State Grid Zhejiang Electric Power Co Ltd
Original Assignee
State Grid Zhejiang Electric Power Co Ltd
North China Electric Power University
Electric Power Research Institute of State Grid Zhejiang Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by State Grid Zhejiang Electric Power Co Ltd, North China Electric Power University, Electric Power Research Institute of State Grid Zhejiang Electric Power Co Ltd filed Critical State Grid Zhejiang Electric Power Co Ltd
Priority to CN202111360571.2A priority Critical patent/CN114169916A/en
Publication of CN114169916A publication Critical patent/CN114169916A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0206—Price or cost determination based on market factors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/08—Auctions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Game Theory and Decision Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a market member quotation strategy making method suitable for a novel power system, and belongs to the technical field of power systems and power markets. The method comprises the following steps: step 1: determining market boundary conditions, determining market trading, clearing, pricing and settlement processes and modes, and establishing a market clearing mode and a market pricing mechanism; step 2: constructing a double-layer optimization model of the optimal quotation of market members according to the market clearing mode and the market pricing mechanism in the step 1; and step 3: and (3) solving the double-layer optimization model established in the step (2) by adopting an iterative solution of deep reinforcement learning to obtain the optimal quotation strategy of the market subject. The invention is not only suitable for various market main bodies, but also can be flexibly matched with different market mechanism designs, thereby providing powerful support for the production and operation of the market main bodies in the complex power market environment in the future.

Description

Market member quotation strategy making method suitable for novel power system
Technical Field
The invention relates to the technical field of power systems and power markets, in particular to a market member quotation strategy making method suitable for a novel power system.
Background
The quotation strategy of the market member refers to a whole set of strategies adopted by the market member when making a quotation decision according to market rules when the power market carries out power trading. In an ideal, fully competitive power market, social benefits are maximized, and the optimal quotation strategy for market members is to quote according to the marginal cost of the generator set. However, the electricity market is not a fully competitive market, but rather is closer to the oligopolistic market, as determined by the particularities of the electricity industry. Since electricity cannot be stored in large quantities and the production and consumption of electricity must be balanced in real time, electricity cannot be freely bought and sold without restriction as other commodities. Factors such as limited number of power generation companies, large investment scale, transmission constraints and transmission losses determine that in the power market, only a few power generation companies may provide power in some regions, which forms an oligopolistic market. Market members may not make price quotes at marginal cost as a price quote strategy, but rather take advantage of imperfections in market structure and rules to increase profits, which is referred to as strategic price quotes. At present, the methods used by researchers at home and abroad for constructing the optimal quotation strategy of market members can be roughly classified into the following categories:
(1) method based on cost analysis
The basic idea of the method is that the electricity generation cost of an electricity generation company plus reasonable profit is taken as a declaration price, and the core of the method is to accurately analyze and predict the electricity generation cost. The method is simple and easy to master for a power generation company, and has the defect that the quotation conditions of other competitors are not taken into consideration, and the profit of the power generation company is difficult to maximize. However, the power generation cost analysis is the basis of the price quoted by the power generation company, and the power generation company needs to calculate the power generation cost of the unit, and the price quoted by the power generation company is 'central number'.
(2) Method for predicting market electricity price
The basic principle of the method is that corresponding electricity prices in the power market are predicted at first, and a power generation company only needs to declare a price slightly lower than the predicted price. Current methods of predicting electricity prices include time series prediction methods, gray system methods, regression analysis, artificial neuron network prediction methods, combinatorial prediction methods, wavelet analysis methods, and the like.
(3) Method for estimating quotation behaviors of other members
The method constructs the quotation strategy of the method by estimating the quotation behaviors of other market members, and the method adopts the principle that the quotation behavior of an opponent is estimated by a probability or fuzzy method, an optimal bidding model is established on the basis, and the optimal quotation strategy is obtained by solving the optimization problem. The difficulty of this method is whether the bidding behavior of the competitor can be accurately estimated, which can be usually estimated by using a statistical method or a predictive method, but the accurate estimation of the bidding behavior of the competitor requires the knowledge of sufficient historical data.
(4) Method based on game theory
The game theory is a theory for specially researching how 2 or more than 2 conflicting individuals make respective optimization decisions under the condition of interaction, and can be applied to the research of market member quotation strategies. The principle of the method is that a game model (such as a Cournot model, a Stackelberg model and a supply function model) is constructed according to the power transaction, then balance points of the models are searched, and the balance points correspond to the optimal quotation strategy. The method has two problems, namely, a plurality of simplifying assumptions are made when the models are applied, the obtained balance point possibly has little significance for constructing a quotation strategy, and whether the balance point exists in the electric power market is not determined theoretically.
In addition to the above methods, some researchers have also applied artificial intelligence methods such as neural networks, genetic algorithms, genetic programming and finite state automata, evolutionary computing, Q-learning, etc. to construct adaptive, evolutionary bidding strategies.
In a plurality of quotation strategy methods, power generation cost analysis is the basis of bidding of a power generator, the method is simple and easy to implement, but the quotation condition of a competitor is not considered, and the maximization of the profit of the competitor is difficult to realize; the methods for predicting the market electricity price and estimating the quotation behaviors of other power generation companies need a large amount of historical data as supports, the data is not sufficient in the primary stage of the construction of the electric power spot market, the market structure and the trading rules are still in adjustment, and the market price is difficult to be accurately predicted; the game theory method has obvious advantages in the aspects of solving the problems of double games and perfect information games, but has not ideal effects in the aspects of multi-player games, incomplete and imperfect information processing. In summary, the prior art still has defects, and is difficult to adapt to multi-subject competition in a complex power market environment. Particularly, in the process of building a novel power system taking new energy as a main body, new energy, stored energy, demand side response and other emerging main bodies are rapidly developed, the power market mechanism is continuously adjusted and changed, and the prior art is difficult to adapt to market member quotation strategy formulation in the future power market.
The invention provides a general market member optimal quotation strategy making method by considering various market main bodies such as new energy, thermal power, energy storage, demand side response and the like widely existing in a novel electric power system in the future and market mechanisms such as an electric energy market, a flexible climbing auxiliary service market and the like.
Disclosure of Invention
The invention aims to provide a market member quotation strategy making method suitable for a novel power system, which is characterized by comprising the following steps of:
step 1: determining market boundary conditions, determining market trading, clearing, pricing and settlement processes and modes, and establishing a market clearing model and a market pricing mechanism;
step 2: establishing a market main body quotation decision model according to new energy, thermal power, energy storage equipment and demand side response resources, and then establishing a double-layer optimization model of the optimal quotation of market members by combining the market clearing model and the market pricing mechanism in the step 1;
and step 3: and (3) solving the double-layer optimization model established in the step (2) by adopting an iterative solution of deep reinforcement learning to obtain the optimal quotation strategy of the market subject.
The market clearing model in the step 1 is a combined clearing model of electric energy and flexible climbing auxiliary service; the market pricing mechanism comprises a node marginal price pricing mechanism of the electric energy market and an opportunity cost unified pricing mechanism of the auxiliary service market.
The market subject quotation decision model in the step 2 is as follows:
the new energy unit:
maximizing profit as objective function
Figure BDA0003358866210000031
Wherein the subscript I represents the generator set number, I belongs to Iwind∪Isolar，IwindAnd IsolarRespectively representing the collection of wind power and photovoltaic units;
Figure BDA0003358866210000032
and gi,tRespectively representing the market price and the winning power of the electric energy of the unit at the time t and the time i; Δ T represents the time interval of a unit time period, T represents the total number of time periods;
thermal power generating unit:
objective function
Figure BDA0003358866210000033
Figure BDA0003358866210000034
Figure BDA0003358866210000035
Figure BDA0003358866210000036
Wherein the content of the first and second substances,
Figure BDA0003358866210000037
for the purpose of the electric energy yield,
Figure BDA0003358866210000038
in order to flexibly climb the market for the benefit,
Figure BDA0003358866210000039
the cost of electricity generation; i belongs to Icoal∪Igas，IcoalAnd IgasRespectively representing the set of coal-fired and gas-fired units;
Figure BDA00033588662100000310
and
Figure BDA00033588662100000311
respectively represents the clearing price of the up-grade and down-grade climbing service in the flexible climbing market in the time period t,
Figure BDA00033588662100000312
and
Figure BDA00033588662100000313
respectively representing the middle scalar of the climbing service of the unit up and down in the t period i;
Figure BDA00033588662100000314
and
Figure BDA00033588662100000315
representing a power generation cost parameter of the unit i; u. ofi,tThe variable is a 0-1 variable representing the starting and stopping states of the unit i in a t period, the value is 1 representing the starting state, and the value is 0 representing the stopping state;
Figure BDA0003358866210000041
which represents the cost of the empty load,
Figure BDA0003358866210000042
representing start-stop cost;
pumped storage power stations:
the objective function comprises electric energy yield and flexible climbing market yield
Figure BDA0003358866210000043
The subscript J represents the pumped storage power station number, J represents the pumped storage power station set, and J belongs to J;
Figure BDA0003358866210000044
and
Figure BDA0003358866210000045
respectively representing the power generation and pumping power of the j power station in the t period.
The electric energy and flexible climbing auxiliary service combined clear model comprises an SCUC model and a corresponding SCED model; the SCUC model is as follows:
an objective function:
Figure BDA0003358866210000046
Figure BDA0003358866210000047
Figure BDA0003358866210000048
Figure BDA0003358866210000049
Figure BDA00033588662100000410
Figure BDA00033588662100000411
Figure BDA00033588662100000412
wherein the content of the first and second substances,
Figure BDA0003358866210000051
represents the utility of the conventional load of the b node, mbRepresents the slope of the load demand curve; db,tAnd
Figure BDA0003358866210000052
respectively representing a middle scalar and a maximum load value of a b node in a t period;
Figure BDA0003358866210000053
represents the cost of the generator set i calculated based on the quote,
Figure BDA0003358866210000054
converting variables for starting;
Figure BDA0003358866210000055
representing the cost of pumped-storage power station j, CnewAnd CstoreRespectively representing basic quotation parameters of the new energy and the pumped storage unit; d is an element of DRIndicating participation in demand response DRLoad of (d, C)CAPThe upper limit of the market quote is expressed,
Figure BDA0003358866210000056
a baseline load for load aggregate quotient d over a time period t;
Figure BDA0003358866210000057
and
Figure BDA0003358866210000058
for increasing or decreasing its load during the period t, correspondingly
Figure BDA0003358866210000059
And
Figure BDA00033588662100000510
represents its invocation cost, whered、βdIs a cost parameter;
Figure BDA00033588662100000511
and
Figure BDA00033588662100000512
respectively represents the demand price parameter V of the climbing capacity product up and down in the t periodt upAnd Vt downThe shortage variable of the climbing capacity product is the t time period; k is a radical ofiFor the quote variable of the i unit, kjIs the quoted price variable of the j power station;
and (4) quotation constraint:
0≤ki≤Kmax (14)
0≤kj≤Kmax (15)
wherein, KmaxRepresenting an upper limit of the quoted price;
and (4) system constraint: including power balance constraints, forward and reverse network flow constraints, and conventional load demand constraints; specifically, the following are shown:
Figure BDA00033588662100000513
wherein, Fl maxRepresenting the transmission capacity limit of line l, fl-i、fl-j、fl-b、fl-dRespectively representing generator power transmission distribution factors of a generator set i, a pumped storage j, a node b and a load aggregation quotient d to a line l;
flexible grade climbing auxiliary service demand constraints
Figure BDA00033588662100000514
Wherein the content of the first and second substances,
Figure BDA0003358866210000061
and
Figure BDA0003358866210000062
the bid amount is bid for the climbing capacity of the load aggregation provider;
Figure BDA0003358866210000063
and
Figure BDA0003358866210000064
the target amount of the climbing capacity of the flexible climbing market is determined by the predicted load fluctuation and the uncertainty requirement, and the calculation formula is as follows:
Figure BDA0003358866210000065
in the formula (I), the compound is shown in the specification,
Figure BDA0003358866210000066
the predicted value of the system net load representing the t period,
Figure BDA0003358866210000067
and
Figure BDA0003358866210000068
is the maximum possible deviation of the upper and lower due to the net load prediction error in the t +1 period;
when the 95% confidence level is considered,
Figure BDA0003358866210000069
and
Figure BDA00033588662100000610
corresponding to 97.5% and 2.5% quantiles of the predicted error distribution, respectively; if the net load prediction error is assumed to follow the Gaussian distribution, i.e., xi-N (0, sigma)2) Then the 95% confidence interval is-1.96 σ to 1.96 σ, and the above equation is rewritten as:
Figure BDA00033588662100000611
and (3) operation constraint of the new energy unit:
Figure BDA00033588662100000612
in which I ∈ Iwind∪IsolarThe maximum output gmax i, t is determined by the predicted output at time t;
and (3) operation constraint of the thermal power generating unit: the method comprises the following steps of unit output upper and lower limit constraint, climbing capacity constraint, capacity upper limit constraint in climbing capacity, minimum continuous start-stop time constraint and unit start-stop logic constraint; specifically, the following are shown:
Figure BDA00033588662100000613
wherein I ∈ Icoal∪Igas，
Figure BDA0003358866210000071
And
Figure BDA0003358866210000072
respectively representing the minimum output and the maximum output of the unit i;
Figure BDA0003358866210000073
and
Figure BDA0003358866210000074
for the time T of operation or stoppage of the unit ii onAnd Ti offRespectively limiting the minimum shutdown time and the minimum startup time of the unit i;
incentive-type demand response constraints: the method comprises the following steps of responding to load power constraint, responding to total power consumption balance constraint before and after response, power consumption load constraint, climbing capacity constraint and bid capacity upper limit constraint in climbing capacity; specifically, the following are shown:
Figure BDA0003358866210000075
wherein the content of the first and second substances,
Figure BDA0003358866210000076
and
Figure BDA0003358866210000077
respectively the response capacity and the response load limit value of the load aggregation quotient d;
Figure BDA0003358866210000078
and
Figure BDA0003358866210000079
represents the up-down climbing capability within 1 min;
and (3) operation constraint of the pumped storage power station: the method comprises the following steps of power generation and pumping power constraint, climbing capacity constraint, bid capacity upper limit constraint in climbing capacity, storage capacity constraint and storage capacity balance constraint; specifically, the following are shown:
Figure BDA00033588662100000710
wherein J is equal to J,
Figure BDA0003358866210000081
and
Figure BDA0003358866210000082
maximum power for generating power and pumping water of the power station j respectively; ej,tRepresents the storage capacity, eta, of the power station j in the time period tjThe pumping efficiency is expressed and is usually 0.75;
Figure BDA0003358866210000083
and
Figure BDA0003358866210000084
respectively representing the minimum and maximum storage capacity electric quantity of the power station j; t is t0And t-1Respectively representing the initial period and the final period of a complete operation cycle of the pumped-storage power station.
The SCED model is based on the SCUC modeli,tAnd
Figure BDA0003358866210000085
optimal value u obtained by SCUC optimizationi,t *And
Figure BDA0003358866210000086
instead, the thermal power unit operation constraints do not include minimum continuous on-off time constraints and unit start-up and shut-down logic constraints.
The invention has the beneficial effects that:
the invention is not only suitable for various market main bodies, but also can be flexibly matched with different market mechanism designs, thereby providing powerful support for the production and operation of the market main bodies in the complex power market environment in the future.
Drawings
FIG. 1 is a flow chart of a market member quotation strategy making method for a novel power system according to the present invention;
FIG. 2 is a block diagram of a two-layer optimization model of the present invention;
FIG. 3 is a block diagram of a DDPG solution two-layer model.
Detailed Description
The invention provides a market member quotation strategy making method suitable for a novel power system, and the market member quotation strategy making method is further explained by combining the attached drawings and a specific embodiment.
The electric power market competition is a typical Complex Adaptive System (CAS) problem, not only are the main bodies numerous, but also the behavior actions among the main bodies can influence each other, and the main bodies can adjust the quotation strategies of the main bodies through continuous tests and learning, so that the evolution of the whole system is promoted. Therefore, the market member quotation strategy can not only focus on the cost of the market member, but also needs to predict other competitor strategies, and the optimal quotation strategy is the Nash equilibrium result of one market competition.
The invention provides a market member quotation strategy making method suitable for a novel power system. The process comprises the following steps: firstly, determining market boundary conditions such as market modes, mechanisms, rules and the like; secondly, determining a market competition target, performing behavior modeling on various resources in the market, and establishing a double-layer optimization model of optimal quotation of market members; and finally, solving the model by adopting an iterative solution method based on deep reinforcement learning to obtain the optimal quotation strategy of the market members. FIG. 1 is a flow chart of a market member quotation strategy making method for a novel power system according to the present invention; the method comprises the following steps:
the method comprises the following steps: market boundary conditions are determined. And (3) defining market trading, clearing, pricing and settlement processes and modes, and establishing a market clearing and pricing model. The invention considers possible market modes under the condition of high new energy occupation ratio in the future, namely an electric energy and flexible climbing auxiliary service combined clearing mode, wherein a node marginal price pricing mechanism is adopted in an electric energy market, and an opportunity cost unified pricing mechanism is adopted in an auxiliary service market.
Step two: and (4) modeling the behavior of the market member. The most common market competition goal is to seek profit maximization, and under the assumption, corresponding market quotation decision models are respectively established for main bodies possibly existing in a future novel power system, such as new energy, thermal power, energy storage equipment, demand side response resources and the like. The quotation decision models of various main bodies and the market clearing and pricing models jointly form a double-layer optimization model of the optimal quotation of the market members.
Step three: and (6) solving the model. Because the market clearing model is a mixed integer quadratic programming problem, the model transformation method in the prior art is not suitable, and an iterative solution of deep reinforcement learning can be selected. Through reasonable setting of the artificial neural network, the intelligent learning mode, the deep learning parameters and the like, the double-layer optimization model can be efficiently solved, and the optimal quotation strategy of the market main body is obtained.
The market member quotation strategy making method suitable for the novel power system comprises two parts, namely a double-layer optimization model and a solving algorithm. FIG. 2 is a block diagram of a two-layer optimization model of the present invention; the upper model is a quotation decision model aiming at maximizing profits of various resources, and the profit calculation mode corresponds to a market pricing and settlement mechanism; the lower model is a clearing model corresponding to a market trading and clearing mechanism, and the clearing model is mainly considered in combination with the electric energy and flexible climbing auxiliary service. For the model solving algorithm, in order to avoid complicated model transformation and facilitate flexible combination or setting of various resources or mechanisms, the method is suitable for adopting an iterative solution method based on deep reinforcement learning.
And (3) upper layer: market subject quotation decision model
Firstly, market behavior modeling is respectively carried out on wind power, photovoltaic and thermal power units and energy storage equipment represented by pumped storage. The new energy unit only participates in the electric energy market, and the income maximization of the electric energy market is achieved; the thermal power generating unit and the pumped storage unit simultaneously participate in the electric energy and flexible climbing market, and the total profit of the two markets is maximized. Considering that the power consumption of a single adjustable load is small, the invention considers that the adjustable load provides incentive-based demand response (IDR) in the form of a load aggregator, and quotes at actual cost, and does not consider the strategic behavior. It should be noted that: firstly, the operation characteristics of the unit are considered in the lower-layer market clearing; secondly, the operation and maintenance cost of pumping and storing energy is considered in the operation characteristic constraint of the pumping and sending efficiency; and thirdly, the market main body only submits the electric energy market quotation without quotation for flexible climbing, and the price of the electric energy market quotation is determined by opportunity cost.
1) New energy machine set (I belongs to I)wind∪Isolar)
Figure BDA0003358866210000091
0≤ki≤Kmax (2)
Equation (1) is an objective function for maximizing yield. Wherein the subscript I denotes the generator set number, IwindAnd IsolarRespectively representing the collection of wind power and photovoltaic units;
Figure BDA0003358866210000101
and gi,tRespectively representing the market price and the winning power of the electric energy of the unit at the time t and the time i; Δ T represents the time interval of the unit period, and T represents the total number of periods.
Equation (2) is an offer constraint. Wherein k isiFor the quote variable of the i unit, KmaxIndicating an upper limit for the bid.
2) Thermal power generating unit (I belongs to I)coal∪Igas)
Figure BDA0003358866210000102
Wherein the content of the first and second substances,
Figure BDA0003358866210000103
Figure BDA0003358866210000104
Figure BDA0003358866210000105
the constraint conditions are as in equation (2).
The target function (3) is derived from the electric energy
Figure BDA0003358866210000106
Flexible climbing market revenue
Figure BDA0003358866210000107
And cost of electricity generation
Figure BDA0003358866210000108
The three parts are combined. Wherein, IcoalAnd IgasRespectively representing the set of coal-fired and gas-fired units;
Figure BDA0003358866210000109
and
Figure BDA00033588662100001010
respectively represents the clearing price of the up-grade and down-grade climbing service in the flexible climbing market in the time period t,
Figure BDA00033588662100001011
and
Figure BDA00033588662100001012
respectively representing the middle scalar of the climbing service of the unit up and down in the t period i;
Figure BDA00033588662100001013
and
Figure BDA00033588662100001014
representing a power generation cost parameter of the unit i; u. ofi,tIs a variable 0-1 representing the starting and stopping states of the unit in the t period i, the value of 1 represents the starting state, the value of 0 represents the shutdown stateA state;
Figure BDA00033588662100001015
which represents the cost of the empty load,
Figure BDA00033588662100001016
indicating the start-stop cost.
3) Pumped storage power station (J epsilon J)
Figure BDA00033588662100001017
0≤kj≤Kmax (8)
The formula (7) is an objective function and comprises two parts of electric energy yield and flexible climbing market yield. Wherein, subscript J represents the pumped storage power station number, and J represents the pumped storage power station set;
Figure BDA00033588662100001018
and
Figure BDA00033588662100001019
respectively representing the power generation and pumping power of the j power station in the t period. Equation (8) is the quotation constraint.
The lower layer: electric energy and flexible climbing auxiliary service combined clearing model
The combined clear Security Constrained Unit Commission (SCUC) model aims at maximizing social benefits, and the constraint conditions comprise system constraint, flexible climbing demand constraint, demand response relevant constraint and operation characteristic constraint of various generator sets and energy storage equipment.
1) An objective function:
Figure BDA0003358866210000111
wherein the content of the first and second substances,
Figure BDA0003358866210000112
Figure BDA0003358866210000113
Figure BDA0003358866210000114
Figure BDA0003358866210000115
Figure BDA0003358866210000116
Figure BDA0003358866210000117
the target function formula (9) is composed of the electricity utilization effect of the conventional load, the power generation cost of the generator set and the energy storage equipment, the calling cost of demand response resources, and the shortage cost of flexible climbing service. Wherein the content of the first and second substances,
Figure BDA0003358866210000118
the utility of the conventional load of the node b is shown, the expression is shown as the formula (10), mb(<0) Represents the slope of the load demand curve; db,tAnd
Figure BDA0003358866210000119
respectively representing a middle scalar and a maximum load value of a b node in a t period;
Figure BDA00033588662100001110
representing the cost of the generator set i calculated based on the quote, the expressions of the new energy set and the conventional set are respectively (11) and (12),
Figure BDA00033588662100001111
converting variables for starting;
Figure BDA00033588662100001112
and (3) representing the cost of the pumped storage power station j, wherein the expression is shown as a formula (13). CnewAnd CstoreAnd respectively representing basic quotation parameters of the new energy and the pumped storage unit.
For IDR resources, D ∈ DRIndicating participation in demand response DRLoad of (d, C)CAPThe upper limit of the market quote is expressed,
Figure BDA00033588662100001113
a baseline load for load aggregate quotient d over a time period t;
Figure BDA00033588662100001114
and
Figure BDA00033588662100001115
for increasing or decreasing its load during the period t, correspondingly
Figure BDA00033588662100001116
And
Figure BDA00033588662100001117
representing its calling cost, is determined by equations (14) - (15), where αd、βd(more than or equal to 0) is a cost parameter.
For the market of flexible climbing a slope,
Figure BDA00033588662100001118
and
Figure BDA00033588662100001119
respectively represents the demand price parameters of the climbing capacity products up and down in the time period t,
Figure BDA00033588662100001120
and
Figure BDA00033588662100001121
the shortage variable of the climbing capacity product is the shortage variable of the climbing capacity product.
2) And (4) system constraint:
Figure BDA0003358866210000121
the system constraints are respectively a power balance constraint, a forward network power flow constraint, a reverse network power flow constraint and a conventional load demand constraint from top to bottom. Wherein the content of the first and second substances,
Figure BDA0003358866210000122
representing the transmission capacity limit of line l, fl-i、fl-j、fl-b、fl-dAnd respectively representing the generator power transmission distribution factors of the generator set i, the pumped storage j, the node b and the load aggregation quotient d to the line l.
3) Flexible grade climbing auxiliary service demand constraints
Figure BDA0003358866210000123
In the formula (I), the compound is shown in the specification,
Figure BDA0003358866210000124
and
Figure BDA0003358866210000125
the bid amount is bid for the climbing capacity of the load aggregation provider;
Figure BDA0003358866210000126
and
Figure BDA0003358866210000127
the target amount of the climbing capacity of the flexible climbing market is determined by the predicted load fluctuation and the uncertainty requirement, and the calculation formula is as follows:
Figure BDA0003358866210000128
in the formula (I), the compound is shown in the specification,
Figure BDA0003358866210000129
the predicted value of the net load of the system (namely the difference between the predicted load and the predicted output of the new energy) representing the t period,
Figure BDA00033588662100001210
and
Figure BDA00033588662100001211
is the maximum possible deviation of the upper and lower values of the t +1 period due to the net load prediction error. When the 95% confidence level is considered,
Figure BDA00033588662100001212
and
Figure BDA00033588662100001213
corresponding to 97.5% and 2.5% quantiles, respectively, of the prediction error distribution. If the net load prediction error is assumed to follow the Gaussian distribution, i.e., xi-N (0, sigma)2) Then the 95% confidence interval may be determined to be-1.96 σ to 1.96 σ, with the above formula rewritten as:
Figure BDA00033588662100001214
4) new energy unit operation constraint (I belongs to I)wind∪Isolar)：
Figure BDA00033588662100001215
In the formula, the maximum output
Figure BDA00033588662100001216
Determined by the predicted effort over time t.
5) Thermal power generating unit operation constraint (I belongs to I)coal∪Igas)：
Figure BDA0003358866210000131
The operation constraint of the thermal power generating unit is respectively the upper and lower limit constraint of the output of the unit, the climbing capability constraint, the upper limit constraint of the bid amount in the climbing capability, the minimum continuous start-stop time constraint and the start-stop logic constraint of the unit from top to bottom. Wherein the content of the first and second substances,
Figure BDA0003358866210000132
and
Figure BDA0003358866210000133
respectively representing the minimum output and the maximum output of the unit i;
Figure BDA0003358866210000134
and
Figure BDA0003358866210000135
for the time t the run or down time of the unit i,
Figure BDA0003358866210000136
and
Figure BDA0003358866210000137
respectively, unit i minimum down time and on time limits.
6) Incentive type demand response dependent constraints
Figure BDA0003358866210000138
The IDR related constraints are respectively as follows from top to bottom: responding to load power constraint, responding to total power consumption balance constraint before and after response, power consumption load constraint, climbing capacity constraint and bidding capacity upper limit constraint in climbing capacity. Wherein the content of the first and second substances,
Figure BDA0003358866210000139
and
Figure BDA00033588662100001310
are respectively asThe response capacity and the response load limit value of the load aggregation quotient d;
Figure BDA00033588662100001311
and
Figure BDA00033588662100001312
indicating the up-and-down climbing capability within 1 min.
7) Pumped storage power station operation constraints (J epsilon J)
Figure BDA0003358866210000141
The pumped storage operation constraint is respectively power generation and pumped power constraint, climbing capacity constraint, upper limit of bid capacity in climbing capacity, storage capacity constraint and storage capacity balance constraint from top to bottom. Wherein the content of the first and second substances,
Figure BDA0003358866210000142
and
Figure BDA0003358866210000143
maximum power for generating power and pumping water of the power station j respectively; ej,tRepresents the storage capacity, eta, of the power station j in the time period tjThe pumping efficiency is expressed and is usually 0.75;
Figure BDA0003358866210000144
and
Figure BDA0003358866210000145
respectively representing the minimum and maximum storage capacity electric quantity of the power station j; t is t0And t-1Respectively representing the initial period and the final period of a complete operation cycle of the pumped-storage power station.
The integrated electric energy and flexible climbing auxiliary service market combined clearing model consists of the SCUC model and a corresponding SCED model, and firstly, a unit start-stop plan is determined by the SCUC model; on the basis, the SCED model is optimized to obtain prices of two markets and medium scalars of various resources. The SCED model is similar to the above models except that0/1 variable ui,tAnd
Figure BDA0003358866210000146
optimum value by SCUC optimization
Figure BDA0003358866210000147
And
Figure BDA0003358866210000148
instead, the genset operating characteristic constraints do not include constraints related to the start-stop of the genset. Electric energy market node price through SCED model
Figure BDA0003358866210000149
Can be obtained by calculating the shadow price of the power balance constraint and the network power flow constraint (corresponding to the formula (16)), and the flexible climbing market clearing price
Figure BDA00033588662100001410
And
Figure BDA00033588662100001411
shadow price determination by hill climbing capability assistance service requirement constraint (equation (17)).
Thus, the construction of the double-layer optimization model is completed. In the model framework, the upper-layer market subject optimizes the decision-making price variable ki、kjTransmitted to the lower layer, and solved by the market combined clearing model to obtain the market price of the electric energy
Figure BDA00033588662100001412
Winning capacity g of each resourcei,t、
Figure BDA00033588662100001413
Db,t、
Figure BDA00033588662100001414
Etc., and flexible ramp market price
Figure BDA00033588662100001415
Winning bid amount of each resource
Figure BDA00033588662100001416
And the like. And optimizing the quotation strategy of each main body according to the market clearing and settlement income, and finally realizing balance through continuous interactive iteration to obtain the optimal quotation strategy of the market main body.
Model solution based on deep reinforcement learning
FIG. 3 is a block diagram of a DDPG solution two-layer model. The whole model framework can be iteratively solved based on a depth reinforcement learning method, a depth Deterministic strategy Gradient algorithm (DDPG) is selected in the invention, wherein the market clearing model belongs to the mixed integer quadratic programming problem, and optimized calculation can be carried out by means of a commercial solver Gurobi. Each main body is used as an Agent with the ability of self-learning, and the market clearing process interacting with the main body is used as the environment. The key elements of the method comprise:
1) and a state s: taking the average income of all units, the average price of climbing service in the market and the average price of climbing service in the market as the states
Figure BDA0003358866210000151
2) Action a: the action of the Agent is a quotation decision variable kiOr
Figure BDA0003358866210000152
3) Strategy muθ: expressed by mapping relation between states and actions
Figure BDA0003358866210000153
Adding a certain noise
Figure BDA0003358866210000154
Time of flight
Figure BDA0003358866210000155
Figure BDA0003358866210000156
Where z represents the number of iterations and the noise standard deviation vzComprises the following steps:
Figure BDA0003358866210000157
in the formula,. epsilon.represents the attenuation factor, and is usually a very small positive number, TBIndicating the capacity of the reservoir, TtrainIs the total number of training sessions.
4) Reporting r: in return for the profit that the Agent gets by interacting with the environment (corresponding to equation (1) or (3) or (7)).
5) The training goal of the master Q network (with parameter ω) is to minimize the mean square error between the sample Q value and the target Q value, with an error function of:
Figure BDA0003358866210000158
wherein the content of the first and second substances,
Figure BDA0003358866210000159
ω is updated as follows:
Figure BDA00033588662100001510
in the formula, NbatRepresents the sample set size of the sample, γ ∈ [0,1 ]]In order to be a factor of the discount,
Figure BDA00033588662100001511
to learn the rate.
6) The deterministic policy gradient formula of the main policy network (with the parameter θ) is:
Figure BDA00033588662100001512
θ is updated as follows:
Figure BDA00033588662100001513
in the formula (I), the compound is shown in the specification,
Figure BDA00033588662100001514
is the learning rate of the master policy network.
7) Target network soft update:
Figure BDA0003358866210000161
the whole algorithm is programmed and calculated based on PyTorch and Gurobi frameworks.
The system of the invention consists of 28 generator sets and 20 conventional loads. Wherein, the power supply is generally installed with 960MW, including 6 thermal power generating units and installed with 480 MW; 10 photovoltaic units are installed with 300 MW; 10 wind turbines are installed with 180 MW. On the basis, 1 60MW pumped storage unit and 1 load aggregation quotient with the maximum response capacity of 60MW are additionally added to construct different example scenarios. The effect of adopting the optimal quotation strategy is analyzed for thermal power, and the comparison standard is a marginal cost quotation scene.
Analysis of results for different power supply configurations:
the system formed by thermal power and new energy is used as a basic case, a pumped storage unit or an IDR is added, and the market competition results when each main body adopts an optimal quotation strategy are shown in table 1.
TABLE 1 results of market competition for different power source configurations
Figure BDA0003358866210000162
Compared with marginal cost quotation, the thermal power generating unit improves profits by adopting an optimal quotation strategy. In the system, the power generation cost of new energy, energy storage equipment and demand response resources is lower than that of thermal power, and the market supply and demand are loose, so that the optimal thermal power quotation is generally low, and the profit increasing rate of the optimal quotation strategy is about 4-9%.
Analyzing results of different new energy ratio scenes:
based on case 4, a plurality of scenes are constructed by gradually increasing 20% of the new energy installed capacity, and the market competition results are as follows:
TABLE 2 market competition results under different new energy ratio scenarios
Figure BDA0003358866210000171
With the gradual increase of the new energy occupation ratio, the thermal power profit is gradually reduced, so that the optimal quotation strategy is gradually close to the marginal cost quotation. When the installed occupancy of the new energy is higher than 64%, the optimal quotation strategy of the thermal power is to quote according to marginal cost.
The present invention is not limited to the above embodiments, and any changes or substitutions that can be easily made by those skilled in the art within the technical scope of the present invention are also within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (5)

1. A market member quotation strategy making method adapting to a novel power system is characterized by comprising the following steps:
step 1: determining market boundary conditions, determining market trading, clearing, pricing and settlement processes and modes, and establishing a market clearing model and a market pricing mechanism;
step 2: establishing a market main body quotation decision model according to new energy, thermal power, energy storage equipment and demand side response resources, and then establishing a double-layer optimization model of the optimal quotation of market members by combining the market clearing model and the market pricing mechanism in the step 1;
and step 3: and (3) solving the double-layer optimization model established in the step (2) by adopting an iterative solution of deep reinforcement learning to obtain the optimal quotation strategy of the market subject.
2. The method for making the market member quotation strategy applicable to the novel power system according to claim 1, wherein the market clearing model in the step 1 is a combined clearing model of electric energy and flexible climbing auxiliary services; the market pricing mechanism comprises a node marginal price pricing mechanism of the electric energy market and an opportunity cost unified pricing mechanism of the auxiliary service market.
3. The method for making the market member quotation strategy applicable to the novel power system according to claim 1, wherein the market subject quotation decision model in the step 2 is as follows:
the new energy unit:
maximizing profit as objective function
Figure FDA0003358866200000011
Wherein the subscript I represents the generator set number, I belongs to Iwind∪Isolar，IwindAnd IsolarRespectively representing the collection of wind power and photovoltaic units;
Figure FDA0003358866200000012
and gi,tRespectively representing the market price and the winning power of the electric energy of the unit at the time t and the time i; Δ T represents the time interval of a unit time period, T represents the total number of time periods;
thermal power generating unit:
objective function
Figure FDA0003358866200000013
Figure FDA0003358866200000014
Figure FDA0003358866200000021
Figure FDA0003358866200000022
Wherein the content of the first and second substances,
Figure FDA0003358866200000023
for the purpose of the electric energy yield,
Figure FDA0003358866200000024
in order to flexibly climb the market for the benefit,
Figure FDA0003358866200000025
the cost of electricity generation; i belongs to Icoal∪Igas，IcoalAnd IgasRespectively representing the set of coal-fired and gas-fired units; lambda [ alpha ]t upAnd λt downRespectively represents the clearing price of the up-grade and down-grade climbing service in the flexible climbing market in the time period t,
Figure FDA0003358866200000026
and
Figure FDA0003358866200000027
respectively representing the middle scalar of the climbing service of the unit up and down in the t period i;
Figure FDA0003358866200000028
and
Figure FDA0003358866200000029
representing a power generation cost parameter of the unit i; u. ofi,tThe variable is a 0-1 variable representing the starting and stopping states of the unit i in a t period, the value is 1 representing the starting state, and the value is 0 representing the stopping state;
Figure FDA00033588662000000210
which represents the cost of the empty load,
Figure FDA00033588662000000211
representing start-stop cost;
pumped storage power stations:
the objective function comprises electric energy yield and flexible climbing market yield
Figure FDA00033588662000000212
The subscript J represents the pumped storage power station number, J represents the pumped storage power station set, and J belongs to J;
Figure FDA00033588662000000213
and
Figure FDA00033588662000000214
respectively representing the power generation and pumping power of the j power station in the t period.
4. The market member quotation strategy formulation method applicable to the novel power system according to claim 2, wherein the electric energy and flexible hill climbing assistance service combined clearing model comprises an SCUC model and a corresponding SCED model; the SCUC model is as follows:
an objective function:
Figure FDA00033588662000000215
Figure FDA0003358866200000031
Figure FDA0003358866200000032
Figure FDA0003358866200000033
Figure FDA0003358866200000034
Figure FDA0003358866200000035
Figure FDA0003358866200000036
wherein the content of the first and second substances,
Figure FDA0003358866200000037
represents the utility of the conventional load of the b node, mbRepresents the slope of the load demand curve; db,tAnd
Figure FDA0003358866200000038
respectively representing a middle scalar and a maximum load value of a b node in a t period;
Figure FDA0003358866200000039
represents the cost of the generator set i calculated based on the quote,
Figure FDA00033588662000000310
converting variables for starting;
Figure FDA00033588662000000311
representing the cost of pumped-storage power station j, CnewAnd CstoreRespectively represent new energy and water pumping storageBasic quotation parameters of the energy unit; d is an element of DRIndicating participation in demand response DRLoad of (d, C)CAPThe upper limit of the market quote is expressed,
Figure FDA00033588662000000312
a baseline load for load aggregate quotient d over a time period t;
Figure FDA00033588662000000313
and
Figure FDA00033588662000000314
for increasing or decreasing its load during the period t, correspondingly
Figure FDA00033588662000000315
And
Figure FDA00033588662000000316
represents its invocation cost, whered、βdIs a cost parameter;
Figure FDA00033588662000000317
and
Figure FDA00033588662000000318
respectively represents the demand price parameter V of the climbing capacity product up and down in the t periodt upAnd Vt downThe shortage variable of the climbing capacity product is the t time period; k is a radical ofiFor the quote variable of the i unit, kjIs the quoted price variable of the j power station;
and (4) quotation constraint:
0≤ki≤Kmax (14)
0≤kj≤Kmax (15)
wherein, KmaxRepresenting an upper limit of the quoted price;
and (4) system constraint: including power balance constraints, forward and reverse network flow constraints, and conventional load demand constraints; specifically, the following are shown:
Figure FDA0003358866200000041
wherein, Fl maxRepresenting the transmission capacity limit of line l, fl-i、fl-j、fl-b、fl-dRespectively representing generator power transmission distribution factors of a generator set i, a pumped storage j, a node b and a load aggregation quotient d to a line l;
flexible grade climbing auxiliary service demand constraints
Figure FDA0003358866200000042
Wherein the content of the first and second substances,
Figure FDA0003358866200000043
and
Figure FDA0003358866200000044
the bid amount is bid for the climbing capacity of the load aggregation provider;
Figure FDA0003358866200000045
and
Figure FDA0003358866200000046
the target amount of the climbing capacity of the flexible climbing market is determined by the predicted load fluctuation and the uncertainty requirement, and the calculation formula is as follows:
Figure FDA0003358866200000047
in the formula (I), the compound is shown in the specification,
Figure FDA0003358866200000048
systematic net representation of t time periodThe predicted value of the load is calculated,
Figure FDA0003358866200000049
and
Figure FDA00033588662000000410
is the maximum possible deviation of the upper and lower due to the net load prediction error in the t +1 period;
when the 95% confidence level is considered,
Figure FDA00033588662000000411
and
Figure FDA00033588662000000412
corresponding to 97.5% and 2.5% quantiles of the predicted error distribution, respectively; if the net load prediction error is assumed to follow the Gaussian distribution, i.e., xi-N (0, sigma)2) Then the 95% confidence interval is-1.96 σ to 1.96 σ, and the above equation is rewritten as:
Figure FDA00033588662000000413
and (3) operation constraint of the new energy unit:
Figure FDA0003358866200000051
in which I ∈ Iwind∪IsolarThe maximum output gmax i, t is determined by the predicted output at time t;
and (3) operation constraint of the thermal power generating unit: the method comprises the following steps of unit output upper and lower limit constraint, climbing capacity constraint, capacity upper limit constraint in climbing capacity, minimum continuous start-stop time constraint and unit start-stop logic constraint; specifically, the following are shown:
Figure FDA0003358866200000052
wherein I ∈ Icoal∪Igas，
Figure FDA0003358866200000053
And
Figure FDA0003358866200000054
respectively representing the minimum output and the maximum output of the unit i;
Figure FDA0003358866200000055
and
Figure FDA0003358866200000056
for the time T of operation or stoppage of the unit ii onAnd Ti offRespectively limiting the minimum shutdown time and the minimum startup time of the unit i;
incentive-type demand response constraints: the method comprises the following steps of responding to load power constraint, responding to total power consumption balance constraint before and after response, power consumption load constraint, climbing capacity constraint and bid capacity upper limit constraint in climbing capacity; specifically, the following are shown:
Figure FDA0003358866200000061
wherein the content of the first and second substances,
Figure FDA0003358866200000062
and
Figure FDA0003358866200000063
respectively the response capacity and the response load limit value of the load aggregation quotient d;
Figure FDA0003358866200000064
and
Figure FDA0003358866200000065
represents the up-down climbing capability within 1 min;
and (3) operation constraint of the pumped storage power station: the method comprises the following steps of power generation and pumping power constraint, climbing capacity constraint, bid capacity upper limit constraint in climbing capacity, storage capacity constraint and storage capacity balance constraint; specifically, the following are shown:
Figure FDA0003358866200000066
wherein J is equal to J,
Figure FDA0003358866200000067
and
Figure FDA0003358866200000068
maximum power for generating power and pumping water of the power station j respectively; ej,tRepresents the storage capacity, eta, of the power station j in the time period tjThe pumping efficiency is expressed and is usually 0.75;
Figure FDA0003358866200000069
and
Figure FDA00033588662000000610
respectively representing the minimum and maximum storage capacity electric quantity of the power station j; t is t0And t-1Respectively representing the initial period and the final period of a complete operation cycle of the pumped-storage power station.
5. The method for making market member quotation strategy adapting to novel power system according to claim 4, wherein the SCED model is based on SCUC modeli,tAnd
Figure FDA0003358866200000071
optimal value u obtained by SCUC optimizationi,t *And
Figure FDA0003358866200000072
instead, the thermal power unit operation constraints do not include minimum connectionsContinuous shutdown time constraint and unit start-stop logic constraint.
CN202111360571.2A 2021-11-17 2021-11-17 Market member quotation strategy making method suitable for novel power system Pending CN114169916A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111360571.2A CN114169916A (en) 2021-11-17 2021-11-17 Market member quotation strategy making method suitable for novel power system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111360571.2A CN114169916A (en) 2021-11-17 2021-11-17 Market member quotation strategy making method suitable for novel power system

Publications (1)

Publication Number Publication Date
CN114169916A true CN114169916A (en) 2022-03-11

Family

ID=80479627

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111360571.2A Pending CN114169916A (en) 2021-11-17 2021-11-17 Market member quotation strategy making method suitable for novel power system

Country Status (1)

Country Link
CN (1) CN114169916A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114884101A (en) * 2022-07-04 2022-08-09 华中科技大学 Pumped storage dispatching method based on self-adaptive model control prediction
CN115860163A (en) * 2022-05-11 2023-03-28 国网山东省电力公司东营供电公司 New energy power generation deviation evaluation method and system based on system operation indexes

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115860163A (en) * 2022-05-11 2023-03-28 国网山东省电力公司东营供电公司 New energy power generation deviation evaluation method and system based on system operation indexes
CN115860163B (en) * 2022-05-11 2024-05-07 国网山东省电力公司东营供电公司 New energy power generation deviation evaluation method and system based on system operation index
CN114884101A (en) * 2022-07-04 2022-08-09 华中科技大学 Pumped storage dispatching method based on self-adaptive model control prediction
CN114884101B (en) * 2022-07-04 2022-09-30 华中科技大学 Pumped storage dispatching method based on self-adaptive model control prediction

Similar Documents

Publication Publication Date Title
Li et al. Distributed tri-layer risk-averse stochastic game approach for energy trading among multi-energy microgrids
CN110276698B (en) Distributed renewable energy transaction decision method based on multi-agent double-layer collaborative reinforcement learning
Guo et al. Optimal energy management of multi-microgrids connected to distribution system based on deep reinforcement learning
Chen et al. Research on day-ahead transactions between multi-microgrid based on cooperative game model
Song et al. Economic-environmental equilibrium-based bi-level dispatch strategy towards integrated electricity and natural gas systems
Huang et al. A control strategy based on deep reinforcement learning under the combined wind-solar storage system
CN112217195B (en) Cloud energy storage charging and discharging strategy forming method based on GRU multi-step prediction technology
Tao et al. Deep reinforcement learning based bidding strategy for EVAs in local energy market considering information asymmetry
CN114169916A (en) Market member quotation strategy making method suitable for novel power system
Zhang et al. Bi-level stochastic real-time pricing model in multi-energy generation system: A reinforcement learning approach
CN112084705A (en) Grid-connected coordination planning method and system for comprehensive energy system
Wehinger et al. Modeling electricity wholesale markets with model predictive and profit maximizing agents
Liu et al. Research on bidding strategy of thermal power companies in electricity market based on multi-agent deep deterministic policy gradient
CN116207739B (en) Optimal scheduling method and device for power distribution network, computer equipment and storage medium
CN116862551A (en) New energy consumption price decision method considering user privacy protection
Peng et al. Review on bidding strategies for renewable energy power producers participating in electricity spot markets
Peng et al. Sequential coalition formation for wind-thermal combined bidding
Bartels et al. Influence of hydrogen on grid investments for smart microgrids
Vincent et al. Sustainable microgrid design with peer-to-peer energy trading involving government subsidies and uncertainties
Liu et al. Reinforcement learning-based energy trading and management of regional interconnected microgrids
Zeng et al. Optimal configuration planning of vehicle sharing station-based electro-hydrogen micro-energy systems for transportation decarbonization
Wang et al. A tri-level framework for distribution-level market clearing considering strategic participation of electrical vehicles and interactions with wholesale market
Qiu et al. Local integrated energy system operational optimization considering multi‐type uncertainties: A reinforcement learning approach based on improved TD3 algorithm
CN116862144A (en) Multi-park low-carbon scheduling method and system based on double games
CN116451880B (en) Distributed energy optimization scheduling method and device based on hybrid learning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination