A rapid screening system for power spot market power based on generator revenue-bid sensitivity matrix

By constructing a generator revenue-bid sensitivity matrix, the problem of insufficient consideration of the impact of nodal load and power flow output in existing technologies is solved, enabling rapid identification and supervision of strategic bidding behavior in the electricity market, and improving market efficiency and transparency.

CN117541276BActive Publication Date: 2024-11-05GUANGDONG ELECTRIC POWER TRADING CENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311219215.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-20
Publication Date
2024-11-05
Estimated Expiration
2043-09-20

AI Technical Summary

Technical Problem

Existing methods fail to adequately consider the impact of nodal load and power flow output when identifying strategic bidding behavior in the electricity market, and cannot provide a basis for regulatory decision-making in advance, resulting in reduced market efficiency and transparency.

Method used

A rapid screening system for market power in the electricity spot market based on the generator revenue-bid sensitivity matrix is ​​established. Through the generation clearing module, the revenue and strategic bidding relationship analysis module, and the sensitivity construction module, the generator revenue-bid sensitivity matrix is ​​constructed to calculate the market power in the electricity spot market and quickly identify generator combinations with greater market power in cooperative strategic bidding.

Benefits of technology

This system can quickly and accurately identify power generator combinations with strong market power in cooperative bidding, maintain fair competition and stable operation of the electricity market, reduce the amount of computation and time required, and provide a basis for regulatory decision-making.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117541276B_ABST
    Figure CN117541276B_ABST
Patent Text Reader

Abstract

The present invention discloses a system for quickly screening market power in the electric power spot market based on a generator revenue-bid sensitivity matrix, comprising a power generation clearing module, a generator revenue and strategic bid relationship analysis module, a generator revenue-bid sensitivity construction module, and an electric power spot market market power screening module. The present invention can be widely used in the supervision of electric power spot market clearing results, can reveal the functional relationship between market members' revenue and bid display, quickly identify generator combinations with higher cooperative strategic bidding market power, and thus maintain fair competition, stability and efficiency in the electric power market.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of electric power market, in particular to a power spot market power rapid screening system based on a power generator revenue-quotation sensitivity matrix. Background Art

[0002] In the past three decades, many countries have established electricity markets to provide a platform for competitive electricity trading for different market participants. In a competitive environment, due to transmission constraints and the relatively concentrated characteristics of the market, market participants may perform strategic bidding behaviors to obtain higher returns. However, the abuse of strategic bidding behaviors has led to abnormal power dispatch and pricing, affecting the efficiency and transparency of the market. In fact, this problem has become an important issue in power markets with highly concentrated power generation resources. Therefore, the prevention, monitoring and mitigation of strategic bidding behaviors have become the key to the safe and stable operation of the power market.

[0003] The above strategic pricing behaviors are common in other markets and are not a special phenomenon in the electricity market. However, in the electricity market, under the widely adopted location marginal pricing (LMP) scheme, the bidding behavior of the unit directly affects the price signals at all other nodes. If a power generation group has power generation resources at multiple nodes, it can bid strategically as a whole in whole or in part to obtain excess revenue. In addition, the implementation of these strategic bidding behaviors will inevitably have an impact on market characteristics including congestion and price levels. Similar strategic bidding behaviors are widely present around the world.

[0004] It can be seen that the identification of strategic bidding behavior is a very important topic. Usually, it is classified as the identification of market power in the power market. There are three main methods to identify market power. The first is simulation verification. Game theory is often used to simulate the market clearing process to study the strategic bidding behavior of market participants. The second is the market power index. At present, the Residual Supply Index (RSI) is the most important market monitoring tool. In addition to RSI, common market power indexes include: Herfindahl-Hirschman Index (HHI), Pivotal Supplier Index (PSI), Must Run Ratio (MRR), etc. Finally, there is the LMP decomposition method, which decomposes LMP into several components. Market power is judged by the weight coefficients of these components.

[0005] In summary, existing methods have conducted extensive research on how to identify market power in the electricity market. However, there are still some deficiencies in the identification of strategic bidding behavior: 1) Current methods for identifying strategic bidding behavior usually focus on the bidding behavior of individual market participants and their relative income, while ignoring the overall node electricity price changes caused by changes in single node quotations and the impact of this change on the benefits of other node market members; 2) Current methods either start from the ex ante market share of power generators without considering the characteristics of the electricity market, or start from the ex post clearing results to punish market participants who implement strategic bidding behavior, thus failing to provide regulatory authorities with relevant decision-making basis in advance. 3) Current methods only consider the quotations and benefits of power generators, and do not take into account the power generation conditions of the power generators themselves (such as node loads and branch currents). Summary of the invention

[0006] The purpose of the present invention is to provide a power spot market power rapid screening system based on the power generator revenue-bid sensitivity matrix, including a power generation clearing module, a power generator revenue and strategic bid relationship analysis module, a power generator revenue-bid sensitivity construction module, and a power spot market power screening module;

[0007] The power generation clearing module stores a real-time market clearing model based on DC optimal power flow;

[0008] The power generator revenue and strategic bid relationship analysis module calls a real-time market clearing model based on DC optimal power flow, and establishes a functional analysis relationship between the revenue of each power generator and the strategic bid of a single power generator;

[0009] The power producer revenue-bid sensitivity construction module retrieves the functional analytical relationship between each power producer revenue and a single power producer strategic bid, and constructs a power producer revenue-bid sensitivity matrix;

[0010] The power spot market power screening module calculates the power spot market power according to the power generator revenue-bid sensitivity matrix.

[0011] Furthermore, the objective function of the real-time market clearing model based on DC optimal power flow is as follows:

[0012]

[0013] Where: represents the collection of strategic bidding power producers; represents the set of non-strategic bidding power producers, p i represents the decision variable of the ith power generator; and are the output decision variable vectors of strategic bidding power plants and non-strategic bidding power plants respectively; and They represent the unit power generation cost coefficients of the i-th strategic bidding power plant and the non-strategic bidding power plant respectively; and They are the unit generation cost coefficient matrices of strategic bidding power plants and non-strategic bidding power plants respectively.

[0014] Furthermore, the constraints of the real-time market clearing model based on DC optimal power flow include system supply and demand balance constraints, branch power flow upper limit constraints, branch power flow lower limit constraints, and power generator private constraints.

[0015] Furthermore, the system supply and demand balance constraints are as follows:

[0016]

[0017] Where: D represents the system node load vector, E represents a column vector whose elements are all 1, G represents the number of generators, N represents the number of nodes, σ S The dual multiplier representing the supply-demand balance constraint of the system; and are the output decision variable vectors of strategic bidding power plants and non-strategic bidding power plants respectively;

[0018] The branch power flow upper limit constraint is as follows:

[0019]

[0020] Where: F MAX represents the upper limit vector of branch power flow, M PTDF represents the power transfer distribution matrix, M NG represents the node connection matrix of the generator, represents the dual multiplier of the upper limit constraint of the branch power flow; L is the number of system branches;

[0021] The branch power flow lower limit constraint is as follows:

[0022]

[0023] Where: F MIN represents the upper limit vector of branch power flow, σ The dual multiplier representing the lower limit constraint of the branch power flow;

[0024] The generator private constraints are as follows:

[0025]

[0026] Where: P MIN and P MAXIndicates the upper and lower limits of the power generation company’s output, T PMIN and T PMAX They represent the dual multipliers of the generator’s private constraints respectively.

[0027] Furthermore, the steps of establishing a functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer include:

[0028] 1) Derive the KKT condition of the real-time market clearing model based on DC optimal power flow, namely:

[0029]

[0030]

[0031] 2) Based on the KKT condition, the optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as the unit power generation cost coefficient The linear function of is:

[0032]

[0033] Where: A0, A + and A - Indicates coefficients, B0, B + and B - represents a constant; λ represents the electricity price of each node;

[0034] 3) Calculate the output P of each power generator, that is:

[0035]

[0036] 4) Based on the output P of each power producer, a functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer is established.

[0037] Furthermore, the optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as the unit power generation cost coefficient The steps of the linear function include:

[0038] 2.1) For any constraint reaching limit in the real-time market clearing model, let the strategic quotation unit be k, rewrite formula (6) to obtain:

[0039]

[0040] Where: E represents the unit matrix, and Respectively represent the coefficient matrix or dual multiplier matrix corresponding to the upper and lower bound constraints; is the strategic bid of strategic power producer k, and is the planning parameter; krepresents a column vector in which all row elements are 0 except the element corresponding to generator k, which is 1; constant vector α -k The kth row element is 0 and the remaining rows of elements represent the bids of non-strategic power producers; The matrix representing the upper bound constraint; The matrix representing the lower bound constraint; Represent the coefficient matrices for reaching the upper and lower bound constraints respectively; denote the dual multipliers that achieve the upper and lower bound constraints respectively; They represent the dual multipliers corresponding to the upper and lower bounds of the private constraints of power generators;

[0041] 2.2) Optimize formula (16) and get:

[0042]

[0043] 2.3) Set the coefficient of the dual multiplier of the unreachable constraint to 0, and obtain the dual multiplier of the power balance constraint and the power flow constraint and the strategic bid of the strategic generator The functional relationship between them is:

[0044]

[0045] Where: A0, A + and A - Respectively represent the corresponding coefficients, B0, B + and B - They represent the corresponding constants respectively;

[0046] 2.4) Based on formula (19)-formula (21), the optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as the unit power generation cost coefficient A linear function of .

[0047] Furthermore, the functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer is as follows:

[0048]

[0049] In the formula, C i is the income of each power generator; i is the output of the i-th generator; G i is the element corresponding to the ith generator in the generator output constant matrix obtained by formula (16). Constant B = B0, B + , B - .

[0050] Furthermore, the power generator revenue-bid sensitivity matrix is ​​as follows:

[0051]

[0052] Among them, the sensitivity of the revenue of the i-th generator to the change of the bid of the k-th generator is As shown below:

[0053]

[0054] In the formula, coefficient A = A0, A + , A - .

[0055] Furthermore, the market power of the electricity spot market is characterized by an index of the strategic bidding intention of any two power generators to cooperate;

[0056] The market power of the electricity spot market is positively correlated with the strategic bidding intention index of any two power generators.

[0057] Furthermore, the strategic bidding intention index of any two power generators is as follows:

[0058]

[0059] Where: i, j represent the serial numbers of the two proposed cooperative power producers, F i,j Represents the cooperative bidding market power index between power generators i and j. represents the sensitivity of the revenue of the ith generator to the change of the quotation of the ith generator; represents the sensitivity of the revenue of the i-th generator to the change of the bid of the j-th generator; represents the sensitivity of the revenue of the jth generator to the change of the bid of the ith generator; It represents the sensitivity of the revenue of the j-th generator to the change of the bid of the j-th generator.

[0060] The technical effect of the present invention is undoubted, and the beneficial effects of the present invention are as follows:

[0061] 1) The present invention fully considers the influence of node load and tidal current output, establishes a method for characterizing the analytical relationship between the revenue of each power generator and the strategic bidding function of a single power generator, and can reveal the correlation between the change of a single power generator's bid and the change of the revenue of all power generators in the system.

[0062] 2) The rapid screening framework for power spot market power based on the power producer revenue-bid sensitivity matrix proposed in the present invention can significantly reduce the computational workload and time of the screening process, and quickly and accurately screen out a combination of power producers with greater cooperative bidding market power.

[0063] The present invention can be widely used in the supervision of clearing results of the electricity spot market, can reveal the functional relationship between the benefits of market members and the quotations displayed, quickly identify combinations of power producers with higher cooperative strategic bidding market power, and thus maintain fair competition, stability and efficiency in the electricity market. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 The strategic bidding situation of the group when there is no network congestion;

[0065] Figure 2 The strategic bidding situation of the units whose real cost is less than the marginal unit cost when there is no network congestion;

[0066] Figure 3 This is the system workflow diagram. DETAILED DESCRIPTION

[0067] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.

[0068] Embodiment 1:

[0069] A power spot market power rapid screening system based on a generator revenue-bid sensitivity matrix, comprising a power generation clearing module, a generator revenue and strategic bid relationship analysis module, a generator revenue-bid sensitivity construction module, and a power spot market power screening module;

[0070] The power generation clearing module stores a real-time market clearing model based on DC optimal power flow;

[0071] The power generator revenue and strategic bid relationship analysis module calls a real-time market clearing model based on DC optimal power flow, and establishes a functional analysis relationship between the revenue of each power generator and the strategic bid of a single power generator;

[0072] The power producer revenue-bid sensitivity construction module retrieves the functional analytical relationship between each power producer revenue and a single power producer strategic bid, and constructs a power producer revenue-bid sensitivity matrix;

[0073] The power spot market power screening module calculates the power spot market power according to the power generator revenue-bid sensitivity matrix.

[0074] Embodiment 2:

[0075] A power spot market power rapid screening system based on a power generator revenue-bid sensitivity matrix, the technical content is the same as that of Example 1, and further, the objective function of the real-time market clearing model based on DC optimal power flow is as follows:

[0076]

[0077] Where: represents the collection of strategic bidding power producers; represents the set of non-strategic bidding power producers, p i represents the decision variable of the ith power generator; and are the output decision variable vectors of strategic bidding power plants and non-strategic bidding power plants respectively; and They represent the unit power generation cost coefficients of the i-th strategic bidding power plant and the non-strategic bidding power plant respectively; and They are the unit generation cost coefficient matrices of strategic bidding power plants and non-strategic bidding power plants respectively.

[0078] Embodiment 3:

[0079] A system for rapid screening of power spot market power based on a generator revenue-bid sensitivity matrix, the technical content of which is the same as any one of Examples 1-2, and further, the constraints of the real-time market clearing model based on DC optimal power flow include system supply and demand balance constraints, branch power flow upper limit constraints, branch power flow lower limit constraints, and generator private constraints.

[0080] Embodiment 4:

[0081] A power spot market power rapid screening system based on a power generator revenue-bid sensitivity matrix, the technical content of which is the same as any one of Embodiments 1-3, and further, the system supply and demand balance constraints are as follows:

[0082]

[0083] Where: D represents the system node load vector, E represents a column vector whose elements are all 1, G represents the number of generators, N represents the number of nodes, σ S The dual multiplier representing the supply-demand balance constraint of the system; and are the output decision variable vectors of strategic bidding power plants and non-strategic bidding power plants respectively;

[0084] The branch power flow upper limit constraint is as follows:

[0085]

[0086] Where: F MAX represents the upper limit vector of branch power flow, M PTDF represents the power transfer distribution matrix, M NG represents the node connection matrix of the generator, represents the dual multiplier of the upper limit constraint of the branch power flow; L is the number of system branches;

[0087] The branch power flow lower limit constraint is as follows:

[0088]

[0089] Where: F MIN represents the upper limit vector of branch power flow, σ The dual multiplier representing the lower limit constraint of the branch power flow;

[0090] The generator private constraints are as follows:

[0091]

[0092] Where: P MIN and P MAX Indicates the upper and lower limits of the power generation company’s output, T PMIN and T PMAX They represent the dual multipliers of the generator’s private constraints respectively.

[0093] Embodiment 5:

[0094] A system for rapid screening of power spot market power based on a power producer revenue-bid sensitivity matrix, the technical content of which is the same as any one of embodiments 1-4, and further, the step of establishing a functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer comprises:

[0095] 1) Derive the KKT condition of the real-time market clearing model based on DC optimal power flow, namely:

[0096]

[0097]

[0098] 2) Based on the KKT condition, the optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as the unit power generation cost coefficient The linear function of is:

[0099]

[0100] Where: A0, A + and A - Indicates coefficients, B0, B + and B - represents a constant; λ represents the electricity price of each node;

[0101] 3) Calculate the output P of each power generator, that is:

[0102]

[0103] 4) Based on the output P of each power producer, a functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer is established.

[0104] Embodiment 6:

[0105] A power spot market power rapid screening system based on power generator revenue-bid sensitivity matrix, the technical content is the same as any one of embodiments 1-5, further, the optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as a unit power generation cost coefficient The steps of the linear function include:

[0106] 2.1) For any constraint reaching limit in the real-time market clearing model, let the strategic quotation unit be k, rewrite formula (6) to obtain:

[0107]

[0108] Where: E represents the unit matrix, and Respectively represent the coefficient matrix or dual multiplier matrix corresponding to the upper and lower bound constraints; is the strategic bid of strategic power producer k, and is the planning parameter; k represents a column vector in which all row elements are 0 except the element corresponding to generator k, which is 1; constant vector α -k The kth row element is 0 and the remaining rows of elements represent the bids of non-strategic power producers; The matrix representing the upper bound constraint; The matrix representing the lower bound constraint; Represent the coefficient matrices for reaching the upper and lower bound constraints respectively; denote the dual multipliers that achieve the upper and lower bound constraints respectively; They represent the dual multipliers corresponding to the upper and lower bounds of the private constraints of power generators;

[0109] 2.2) Optimize formula (16) and get:

[0110]

[0111] 2.3) Set the coefficient of the dual multiplier of the unreachable constraint to 0, and obtain the dual multiplier of the power balance constraint and the power flow constraint and the strategic bid of the strategic generator The functional relationship between them is:

[0112]

[0113] Where: A0, A + and A - Respectively represent the corresponding coefficients, B0, B + and B - They represent the corresponding constants respectively;

[0114] 2.4) Based on formula (19)-formula (21), the optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as the unit power generation cost coefficient A linear function of .

[0115] Embodiment 7:

[0116] A power spot market power rapid screening system based on a power generator revenue-bid sensitivity matrix, the technical content of which is the same as any one of Embodiments 1-6, and further, the functional analytical relationship between the revenue of each power generator and the strategic bid of a single power generator is as follows:

[0117]

[0118] In the formula, C i is the income of each power generator; i is the output of the i-th generator; G i is the element corresponding to the ith generator in the generator output constant matrix obtained by formula (16), that is, P i The constant B = B0, B + , B - .

[0119] Embodiment 8:

[0120] A power spot market power rapid screening system based on a power generator revenue-bid sensitivity matrix, the technical content of which is the same as any one of embodiments 1-7, and further, the power generator revenue-bid sensitivity matrix is ​​as follows:

[0121]

[0122] Among them, the sensitivity of the revenue of the i-th generator to the change of the bid of the k-th generator is As shown below:

[0123]

[0124] In the formula, coefficient A = A0, A + , A - .

[0125] Embodiment 9:

[0126] A system for quickly screening the power of a spot electricity market based on a power generator revenue-bid sensitivity matrix, the technical content of which is the same as any one of embodiments 1-8, and further, the power of the spot electricity market is characterized by an index of strategic bidding intentions of any two power generators for cooperation;

[0127] The market power of the electricity spot market is positively correlated with the strategic bidding intention index of any two power generators.

[0128] Embodiment 10:

[0129] A power spot market power rapid screening system based on a power generator revenue-bid sensitivity matrix, the technical content of which is the same as any one of Embodiments 1-9, and further, the strategic bidding intention index of any two power generators is as follows:

[0130]

[0131] Where: i, j represent the serial numbers of the two proposed cooperative power producers, F i,j Represents the cooperative bidding market power index between power generators i and j. represents the sensitivity of the revenue of the ith generator to the change of the quotation of the ith generator; represents the sensitivity of the revenue of the i-th generator to the change of the bid of the j-th generator; represents the sensitivity of the revenue of the jth generator to the change of the bid of the ith generator; It represents the sensitivity of the revenue of the j-th generator to the change of the bid of the j-th generator.

[0132] Embodiment 11:

[0133] A rapid screening system for power spot market power based on generator revenue-bid sensitivity matrix is ​​as follows:

[0134] The system first determines the marginal power generation cost of each power producer based on historical data, and uses multi-parameter planning theory to derive the functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer within the marginal cost domain, so as to reflect the impact of the strategic unit bid on the revenue of each unit in the system; then, based on sensitivity analysis, the revenue and bid functions of each unit are derived to construct a power producer revenue-bid sensitivity matrix; finally, the sum of the elements of the power producer revenue-bid sensitivity matrix between all two power producer combinations within the power generation group is traversed and calculated as the cooperative strategic bid market power index of the combination, and the combination is sorted according to the index result, so as to realize the rapid screening of the strategic bidding market power exercise of the power generation group.

[0135] The workflow of this system is as follows:

[0136] 1. Representation of the analytical relationship between the revenue of each power producer and the strategic bidding function of a single power producer

[0137] First, the marginal power generation cost of each power producer is determined based on historical data, and the multi-parameter planning theory is used to derive the functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer within the marginal cost domain, so as to reflect the impact of the strategic unit bid on the revenue of each unit in the system. The specific steps are as follows:

[0138] 1) Establish a real-time market clearing model based on DC optimal power flow:

[0139] Objective function: Minimize the power generation cost of the power generator

[0140]

[0141] Where: represents the collection of strategic bidding power producers, represents the set of non-strategic bidding power producers, p i represents the decision variable of the ith power producer, and are the output decision variable vectors of strategic bidding power plants and non-strategic bidding power plants, and They represent the unit power generation cost coefficients of the i-th strategic bidding power plant and the non-strategic bidding power plant, respectively. and They are the unit generation cost coefficient matrices of strategic bidding power plants and non-strategic bidding power plants respectively.

[0142] System supply and demand balance constraints:

[0143]

[0144] Where: D represents the system node load vector, E represents a column vector whose elements are all 1, G represents the number of generators, N represents the number of nodes, σ S Represents the dual multiplier of the supply and demand balance constraint of the system.

[0145] Branch power flow upper limit constraint:

[0146]

[0147] Where: F MAX represents the upper limit vector of branch power flow, M PTDF represents the power transfer distribution matrix, M NG represents the node connection matrix of the generator, Represents the dual multiplier of the branch power flow upper limit constraint.

[0148] Branch power flow lower limit constraint:

[0149]

[0150] Where: F MIN represents the upper limit vector of branch power flow, σ Dual multiplier representing the lower limit constraint of branch power flow.

[0151] Power generator private constraints:

[0152]

[0153] Where: P MIN and P MAX Indicates the upper and lower limits of the power generation company’s output, T PMIN and T PMAX denote the dual multipliers of the corresponding constraints respectively.

[0154] If we do not consider the impact of network constraints on market clearing (assuming that the total system capacity is sufficient and there is no problem of insufficient capacity if a unit is removed), then the electricity prices of all nodes in the entire network are equal, which is equal to the marginal cost of the marginal unit. The market clearing unit selection is to sort all units from low to high according to marginal cost, such as Figure 1 As shown in the figure, the units with costs less than the marginal unit cost are fully powered, and the units with costs greater than the marginal unit cost are not powered. The strategic bidding behavior of the units can be divided into three categories (to avoid negative returns for the units, the reported costs of the units are always greater than or equal to their actual costs):

[0155] a) Units whose actual cost is less than the marginal unit cost: Figure 2 As shown in the figure, if the strategic bid is less than the marginal unit cost, the unit will still be cleared at the marginal unit cost, the power generation will remain unchanged, the strategic bid unit revenue will remain unchanged, and the revenue cannot be raised through strategic bidding; if the strategic bid is greater than or equal to the marginal cost, the unit will be replaced by other units and cannot be selected by the market clearing, and the unit will have no revenue. In summary, for units whose real cost is less than the marginal cost, when there is no network congestion, the optimal bid of the unit in theory is its real bid.

[0156] b) Units whose true cost is equal to the marginal unit cost: This type of unit is a marginal unit. The strategic quotation of a marginal unit tends to be the unit that is closest to and greater than its true cost. In general, the cost difference between adjacent units in the market is not large, so there is not much room for strategic quotations for this type of unit.

[0157] c) Units whose real cost is greater than the marginal unit cost: Units whose real cost is greater than the marginal cost are not selected for market clearing, and the bid will not affect the clearing result anyway.

[0158] When considering network constraints, the orderly line capacity affects the electricity prices of all nodes in the entire network, and the node marginal electricity price mechanism causes the units to falsely report high prices, and the revenue of power generators will change accordingly. Using multi-parameter planning theory, the functional analytical relationship between the revenue of each power generator and the strategic bid of a single power generator in the marginal cost domain is derived, as shown in the next step.

[0159] 2) Using multi-parameter planning theory, derive the functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer

[0160] The KKT conditions for the above optimization problems (1)-(5) are derived as shown in equations (6)-(14):

[0161]

[0162]

[0163] Assume that in the linear optimization problem (1)-(5), there is no degeneration of the original problem or the dual problem, and there is a unique optimal solution. Formula (6) can express the optimal solution as planning parameters The specific process is as follows:

[0164] For a certain constraint reaching limit, let the strategic bidding unit be k, and write equation (6) as:

[0165]

[0166] Where: E represents the unit matrix, and denote the coefficient matrix or dual multiplier matrix corresponding to the upper and lower bound constraints, respectively. is the strategic bid of strategic power producer k, and is the planning parameter. k It represents a column vector whose row elements are all 0 except the element corresponding to generator k, and the constant vector α -k The elements in the kth row are 0, and the elements in the remaining rows represent the bids of non-strategic power producers.

[0167] From the optimality theory, we know that the coefficient matrix on the left side of equation (15) is a full rank matrix. Further deduction shows that:

[0168]

[0169] Further rearrange the coefficients and set the coefficients of the dual multipliers of the unbounded constraints to 0, and we can get the dual multipliers of the power balance constraint and the power flow constraint: Strategic Quotations with Strategic Power Generators The functional relationship between them is:

[0170]

[0171] Where: A0, A + and A - Respectively represent the corresponding coefficients, B0, B + and B - They represent the corresponding constants, which are all calculated by formula (16).

[0172] The electricity price of each node and the strategic bid of the strategic power generator can be calculated by equations (17)-(19): The functional relationship between them is:

[0173]

[0174] From formula (20), we can find that the electricity price λ of each node and the strategic bid of the strategic power generator are There is a one-time relationship between them.

[0175] Similarly, from equations (8)-(11), when a certain constraint reaches a limit, the output of each generator P is a piecewise constant:

[0176]

[0177] The revenue C of each power generator can be calculated by equations (20) and (21): i Strategic bidding with a single generator Functional analysis relationship between them:

[0178]

[0179] From formula (22), we can find that the revenue of each power generator is C i Strategic bidding with a single generator There is a one-time relationship between them.

[0180] 2. A framework for rapid screening of power spot market power based on generator revenue-bid sensitivity matrix:

[0181] 1) Constructing the power producer revenue-bid sensitivity matrix

[0182] After obtaining the functional relationship between the revenue of each power generator and the strategic bid of a single strategic power generator, the sensitivity analysis is used to determine the impact of changes in the bidding strategies of different power generators on the revenue of all power generators in the system. The calculation formula is shown in formula (23):

[0183]

[0184] Where: It represents the sensitivity of the revenue of the i-th generator to the change of the quotation of the k-th generator.

[0185] Therefore, the constructed power producer revenue-bid sensitivity matrix is ​​shown in formula (24):

[0186]

[0187] 2) Calculate the market power index of the strategic bidding intention of the two e-commerce cooperation

[0188] When a power generation group in the system owns power generators at multiple nodes in the system, there will be mutual influence between the power generators, which will further increase the bidding benefits of both. If the traditional multi-parameter planning method is continued to be used to calculate the sensitivity relationship between the power generator's income and the quotation, the calculation scale will increase exponentially and cannot bear the calculation burden. For this reason, the present invention constructs a strategic bidding market power index for cooperation between two power generators based on the power generator income-quotation sensitivity matrix, which is used to quickly judge the possibility of cooperation bidding between two power generators. The calculation method is shown in formula (25):

[0189]

[0190] Where: i, j represent the serial numbers of the two proposed cooperative power producers, F i,j Represents the cooperative bidding market power index between power generators i and j.

[0191] Through formula (25), the original complex combination calculation problem can be converted into a simple algebraic calculation, which greatly reduces the calculation tasks and time when screening possible cooperative bidding power plant combinations.

[0192] 3) Rapid screening of power spot market power

[0193] After calculating the market power index of cooperative strategic bidding between two power generators, the power generator combinations are ranked according to the size of each index group. According to actual needs, the power generator combinations with higher rankings are subject to key supervision, thereby completing the possibility screening of cooperative strategic bidding between the two power generators.

[0194] Embodiment 12:

[0195] A verification experiment of a rapid screening system for power spot market power based on the generator revenue-bid sensitivity matrix is ​​as follows:

[0196] 1. System workflow:

[0197] (1) Establishing a real-time market clearing model based on DC optimal power flow

[0198] Based on the optimal DC power flow, a real-time market clearing model is established with the objective of minimizing the power generation cost of the generator (Formula (1)), which includes the system supply and demand balance constraints (Formula (2)), system power flow constraints (Formulas (3)-(4)) and the generator private constraints (Formula (5)).

[0199] (2) Collect and organize marginal cost information of power generators

[0200] Collect and organize the marginal cost information of each power producer. In a perfectly competitive market, the strategic bid of a marginal unit tends to be the unit that is closest to and greater than its true cost. In a general market, the cost difference between adjacent units is not large, so there is not much room for strategic bids for such units. It is easier to implement the possibility of strategic bidding of power producers based on their marginal costs.

[0201] (3) Derive the functional relationship between the revenue of each power producer and the bid price of the strategic power producer

[0202] (3.1) Based on the implementation of the market clearing model (Equations (1)-(5)), the KKT conditions are derived as shown in Equations (6)-(14);

[0203] (3.2) Select a power generator as the initial strategic power generator;

[0204] (3.3) Taking the collected marginal cost information of power generators as the domain, the functional relationship between the node electricity price and the bid of the strategic power generator is derived according to the calculation method shown in equations (15)-(20);

[0205] (3.4) Then, according to formula (21), the functional relationship between the output of each power producer and the bid of the strategic power producer is calculated;

[0206] (3.5) According to formula (22), calculate the functional relationship between the power generator's revenue and the strategic power generator's bid;

[0207] (3.6) Repeat steps (3.2)-(3.5) to deduce the functional relationship between the impact of changes in bids from different strategic power producers on the overall power producer revenue.

[0208] (4) Constructing the power generator revenue-bid sensitivity matrix

[0209] According to the functional relationship between the revenue of each power producer and the bid of the strategic power producer obtained in step (3), the sensitivity of the revenue change of all power producers in the system when the bid of a single power producer changes is calculated using the method shown in formula (23), and the result is filled into the corresponding position of the power producer revenue-bid sensitivity matrix shown in formula (24), thereby completing the construction of the power producer revenue-bid sensitivity matrix.

[0210] (5) Calculate the market power index of the strategic bidding between two e-commerce companies

[0211] According to the method shown in formula (25), the power generator revenue-bid sensitivity matrix obtained in step (4) is processed to calculate the market power index of the strategic bidding cooperation between the two power generators.

[0212] (6) Rapid screening of power spot market power

[0213] Based on the strategic bidding market power index of the two power generators obtained in step (5), all power generator combinations are ranked and the units with a high probability of exercising market power are determined according to actual needs.

[0214] The specific simulation results are as follows:

[0215] 1) Introduction to simulation data and comparison methods

[0216] The present invention takes the IEEE-30 node and IEEE-118 node standard calculation examples as examples, sets each power generator to quote according to the periodic marginal cost, and the quotation mode is a linear quotation model. The quotation information of each power generator is shown in Table 1 and Table 2. The simulation process is carried out in the above two systems.

[0217] Table 1 IEEE-30 node system unit parameters

[0218]

[0219] Table 2 IEEE-118 node system unit parameters

[0220]

[0221] In the simulation, the ranking results of the market power of the strategic bidding of the two power generators based on marginal cost proposed in this paper and the ranking results of the market clearing model containing the strategic bidding behavior of the actual units are compared to verify the accuracy of the method proposed in this invention:

[0222] M1: The power spot market power rapid screening system based on the power generator revenue-bid sensitivity matrix proposed in the present invention;

[0223] M2: The actual market clearing model with strategic bidding behavior of units.

[0224] The computer hardware environment used for verifying the effect of the present invention is Intel(R) Core(TM) i5-10400U CPU@2.90GHz 16GB RAM.

[0225] 2) Comparison of simulation results

[0226] The ranking results of the market power of the two power generators’ cooperative strategic bidding obtained by M1 and M2 in the IEEE-30 node system are shown in Table 3.

[0227] Table 3 Ranking results of the market power of the two power generators in the strategic bidding system obtained by M1 and M2 in the IEEE-30 node system

[0228]

[0229] It can be seen from Table 3 that the method M1 proposed in the present invention can obtain the same result as the actual clearing result in the small calculation example.

[0230] The ranking results of the market power of the two power generators’ cooperative strategic bidding obtained by M1 and M2 in the IEEE-118 node system are shown in Tables 4 and 5.

[0231] Table 4 Ranking results of the market power of the strategic bidding of two power generators in the IEEE-118 node system obtained by M1 and M2 (ranking 1-99)

[0232]

[0233]

[0234] Table 5 Ranking results of the market power of the strategic bidding of two power generators in the IEEE-118 node system obtained by M1 and M2 (ranking 100-198)

[0235]

[0236]

[0237] Since the IEEE-118 system contains 54 units, the two combinations have The first 198 rankings are shown in Tables 4 and 5. The average deviation of the ranking is 11.26, and the similarity of the first 198 rankings is 90.73%. The ranking of the latter is not shown in detail because the power generation company combination has a small market power and a limited impact on the revenue. However, since this project is to select the power generation company combination with strong market power in the strategic bidding of the two power generation companies, more attention is paid to the top ranking combination, so the ranking accuracy of the first 10% can already meet the actual use needs.

[0238] In summary, the method for rapid screening of market power of cooperative strategic bidding between two power generators based on marginal cost proposed in the present invention can accurately screen out units with high possibility of cooperative strategic bidding in the system, and provide a basis for key supervision by regulatory authorities.

[0239] 3) Comparison of simulation calculations

[0240] To further illustrate the advantage of the proposed method in computing speed, Table 6 shows the computational complexity comparison between the proposed method and the market simulation method.

[0241] Table 6 Comparison of calculation amount between M1 and M2 methods

[0242]

[0243] From the comparison of the contents in Table 6, it can be found that the proposed method does not significantly improve the amount of calculation in small examples, but in large examples, by converting complex clearing simulation calculations into algebraic calculations of equal order, it can achieve a significant reduction in the amount of calculation and improve the calculation efficiency while ensuring a certain accuracy rate.

[0244] Based on the relationship between the revenue of power generators and their bids, this paper proposes a rapid screening system for the market power of strategic bidding of two power generators based on marginal cost. It can quickly identify the combination of power generators with high probability of strategic bidding on the basis of ensuring a high accuracy rate, and help maintain the efficient, stable and fair operation of the power market. The simulation results of the example show that the proposed rapid method can improve the recognition speed while ensuring the accuracy of recognition.

Claims

1. A rapid screening system for power spot market power based on power generator revenue-bid sensitivity matrix, characterized in that: It includes power generation clearing module, power generator revenue and strategic bidding relationship analysis module, power generator revenue-bid sensitivity construction module, and power spot market market power screening module; The power generation clearing module stores a real-time market clearing model based on DC optimal power flow; The power generator revenue and strategic bid relationship analysis module calls a real-time market clearing model based on DC optimal power flow, and establishes a functional analysis relationship between the revenue of each power generator and the strategic bid of a single power generator; The power producer revenue-bid sensitivity construction module retrieves the functional analytical relationship between each power producer revenue and a single power producer strategic bid, and constructs a power producer revenue-bid sensitivity matrix; The power spot market power screening module calculates the power spot market power according to the power generator revenue-bid sensitivity matrix; The constraints of the real-time market clearing model based on DC optimal power flow include system supply and demand balance constraints, branch power flow upper limit constraints, branch power flow lower limit constraints, and power generator private constraints; The system supply and demand balance constraints are as follows: Where: D represents the system node load vector, E represents a column vector whose elements are all 1, G represents the number of generators, N represents the number of nodes, σ S The dual multiplier representing the supply-demand balance constraint of the system; and are the output decision variable vectors of strategic bidding power plants and non-strategic bidding power plants respectively; The branch power flow upper limit constraint is as follows: Where: F MAX represents the upper limit vector of branch power flow, M PTDF represents the power transfer distribution matrix, M NG represents the node connection matrix of the generator, represents the dual multiplier of the upper limit constraint of the branch power flow; L is the number of system branches; The branch power flow lower limit constraint is as follows: Where: F MIN represents the upper limit vector of branch power flow, σ The dual multiplier representing the lower limit constraint of the branch power flow; The generator private constraints are as follows: Where: P MIN and P MAX Indicates the upper and lower limits of the power generation company’s output, T PMIN and T PMAX They represent the dual multipliers of the generator’s private constraints respectively; The steps of establishing the functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer include: 1) Derive the KKT condition of the real-time market clearing model based on DC optimal power flow, namely: 2) Based on the KKT condition, the optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as the unit power generation cost coefficient The linear function of is: Where: A0, A + and A - Indicates coefficients, B0, B + and B - represents a constant; λ represents the electricity price of each node; 3) Calculate the output P of each power generator, that is: In the formula, Represent the coefficient matrices for reaching the upper and lower bound constraints respectively; 4) Based on the output P of each power producer, a functional analytical relationship between the revenue of each power producer and the strategic bid of a single power producer is established.

2. A power spot market power rapid screening system based on power generator revenue-bid sensitivity matrix according to claim 1, characterized in that: The objective function of the real-time market clearing model based on DC optimal power flow is as follows: Where: Represents the collection of strategic bidding power producers; represents the set of non-strategic bidding power producers, p i represents the decision variable of the ith power generator; and are the output decision variable vectors of strategic bidding power plants and non-strategic bidding power plants respectively; and They represent the unit power generation cost coefficients of the i-th strategic bidding power plant and the non-strategic bidding power plant respectively; and They are the unit generation cost coefficient matrices of strategic bidding power plants and non-strategic bidding power plants respectively.

3. A power spot market power rapid screening system based on power generator revenue-bid sensitivity matrix according to claim 1, characterized in that: The optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as the unit power generation cost coefficient The steps of the linear function include: 2.1) For any constraint reaching limit in the real-time market clearing model, let the strategic quotation unit be k, rewrite formula (6) to obtain: Where: E represents the unit matrix, is the strategic bid of strategic power producer k, and is the planning parameter; k represents a column vector in which all row elements are 0 except the element corresponding to generator k, which is 1; constant vector α -k The kth row element is 0 and the remaining rows of elements represent the bids of non-strategic power producers; The matrix representing the upper bound constraint; The matrix representing the lower bound constraint; Represent the coefficient matrices for reaching the upper and lower bound constraints respectively; denote the dual multipliers that achieve the upper and lower bound constraints respectively; They represent the dual multipliers corresponding to the upper and lower bounds of the private constraints of power generators; 2.2) Optimize formula (16) and get: 2.3) Set the coefficient of the dual multiplier of the unreachable constraint to 0, and obtain the dual multiplier of the power balance constraint and the power flow constraint and the strategic bid of the strategic generator The functional relationship between them is: Where: A0, A + and A - Respectively represent the corresponding coefficients, B0, B + and B - They represent the corresponding constants respectively; 2.4) Based on formula (19)-formula (21), the optimal solution of the real-time market clearing model based on DC optimal power flow is expressed as the unit power generation cost coefficient A linear function of .

4. A power spot market power rapid screening system based on power generator revenue-bid sensitivity matrix according to claim 3, characterized in that: The functional analytical relationship between the revenue of each power generator and the strategic bid of a single power generator is as follows: In the formula, C i is the income of each power generator; i is the output of the i-th generator; G i is the element corresponding to the ith generator in the generator output constant matrix obtained by formula (16); constant B = B0, B + , B - ; coefficient A = A0, A + , A - .

5. The power spot market power rapid screening system based on the power generator revenue-bid sensitivity matrix according to claim 1 is characterized in that: The power generator revenue-quotation sensitivity matrix is ​​as follows: Among them, the sensitivity of the revenue of the i-th generator to the change of the bid of the k-th generator is i-k As shown below: In the formula, coefficient A = A0, A + , A - ; G i is the element corresponding to the i-th generator in the generator output constant matrix obtained by formula (16); is the strategic bid of strategic power producer k; C i Profits for all generators.

6. A power spot market power rapid screening system based on power generator revenue-bid sensitivity matrix according to claim 1, characterized in that: The market power of the electricity spot market is characterized by the strategic bidding intention index of any two power generators; The market power of the electricity spot market is positively correlated with the strategic bidding intention index of any two power generators.

7. A power spot market power rapid screening system based on power generator revenue-bid sensitivity matrix according to claim 5, characterized in that: The strategic bidding intention index of any two power generators is as follows: F ij =▽ i-i +▽ i-j +▽ j-i +▽ j-j (25) Where: i, j represent the serial numbers of the two proposed cooperative power producers, F i,j represents the cooperative bidding market power index between power generators i and j; ▽ i-i represents the sensitivity of the revenue of the ith generator to the change in the quotation of the ith generator; ▽ i-j represents the sensitivity of the revenue of the i-th generator to the change in the bid price of the j-th generator; j-i represents the sensitivity of the revenue of the jth generator to the change of the quotation of the ith generator; j-j It represents the sensitivity of the revenue of the j-th generator to the change of the bid of the j-th generator.