Inter-provincial power transaction clearing problem optimization calculation method considering unexecutable power expectation penalty

By establishing a trading model that takes into account the low carbon value of green electricity in the medium and long-term inter-provincial power market and introducing unexecutable power expectation penalty items, the problems of green electricity execution deviation and low computing efficiency are solved, and efficient market resource allocation and Improved computing speed.

CN119066835APending Publication Date: 2024-12-03STATE GRID SICHUAN ELECTRIC POWER CORP ELECTRIC POWER RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410967959.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-18
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The existing technology is difficult to effectively deal with the random characteristics of green electricity in the medium- and long-term inter-provincial power market, resulting in the execution deviation of green electricity winning bids, increasing the burden of market clearance calculation and the complexity of resource allocation, and the existing methods are on a large scale. The calculation efficiency in the market is low, making it difficult to ensure the feasibility and accuracy of the solution results.

Method used

Establish a medium- and long-term inter-provincial power trading model that takes into account the low-carbon value of green electricity, build a punishment expression that considers linearized unexecutable power expectations, and accelerate the solution through fixed acceleration of trading variables, and build an efficient clearance model to improve market resource allocation efficiency.

Benefits of technology

By embedding carbon emission costs and green electricity unenforceable power penalty items, the enforceability of green electricity winning power is improved, subsequent solution adjustments are reduced, market resource allocation efficiency is improved, and calculation speed and solution efficiency are significantly improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066835A_ABST
    Figure CN119066835A_ABST
Patent Text Reader

Abstract

The invention discloses an inter-provincial power transaction clearing problem optimization calculation method considering non-executable power expectation penalty, and the method comprises the following steps: 1), taking the social welfare maximization as an optimization target, and building an inter-provincial power medium and long term transaction model considering the green power low-carbon value; 2) constructing a green electricity non-executable power penalty expression considering output distribution characteristics, and performing linearization processing to obtain a penalty considering linearization non-executable power expectation; 3) correcting the target function of the inter-provincial power medium and long term transaction model based on the penalty considering the linearized non-executable power expectation, and constructing an inter-provincial power medium and long term transaction efficient clearing model; and 4) solving the inter-provincial power medium and long term transaction efficient clearing model to obtain an inter-provincial power medium and long term transaction clearing scheme. The inter-provincial power transaction clearing method for improving the green power configuration scheme executable can improve the executable performance of the clearing scheme, the clearing calculation efficiency and the solving efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of power systems and automation thereof, and in particular to an optimization calculation method for inter-provincial power transaction clearing problem taking into account unexecutable power expectation penalties. Background Art

[0002] The inter-provincial medium- and long-term electricity market is one of the basic links in building a multi-level unified electricity market system and giving full play to the role of market resource allocation, which promotes the medium- and long-term power balance of the power system. However, the large-scale participation of green electricity, mainly wind and solar energy, in the inter-provincial medium- and long-term electricity market will bring severe challenges to the feasibility of market clearing solutions and the efficiency of market clearing calculations.

[0003] On the one hand, the current transaction power in my country's inter-provincial medium- and long-term electricity market needs to be physically executed. However, as the main body of electricity sales, green electricity not only has the excellent characteristics of clean and low carbon, but also has a strong random characteristic in its output. Affected by this characteristic, the actual power that green electricity can generate may not reach its declared power in the inter-provincial medium- and long-term market, resulting in the inability of green electricity to execute the winning bid output stipulated in the medium- and long-term transaction. In order to make up for the execution deviation of the green electricity winning bid power, it is necessary to make multiple plan adjustments in the day-ahead, intraday market transactions or power dispatching, which destroys the optimality of the original inter-provincial medium- and long-term electricity market resource allocation plan. On the other hand, with the participation of a large number of green electricity entities, the scale of the inter-provincial medium- and long-term electricity trading market clearing problem will further increase, resulting in an increasingly heavy burden on model solving. In order to reserve enough time for subsequent power grid security verification and transaction plan adjustment, the solution of the clearing problem needs to be completed in a short time, which puts higher requirements on the computational efficiency of the inter-provincial medium- and long-term electricity market clearing algorithm. Therefore, it is necessary to take into account the impact of the randomness of green electricity in medium- and long-term inter-provincial electricity transactions, so as to ensure the feasibility of green electricity winning bids, reduce subsequent plan adjustments, improve the efficiency of market resource allocation, and study efficient clearing methods to improve the feasibility of clearing plans.

[0004] For the market clearing problem taking into account the uncertainty of green electricity, most existing studies use uncertainty optimization scheduling methods such as chance constraints, fuzzy optimization, stochastic optimization, and robust optimization. However, such methods have a heavy computational burden and are difficult to apply to the medium- and long-term inter-provincial electricity market with a large problem scale. Therefore, in terms of the medium- and long-term inter-provincial electricity market clearing, existing studies still mainly use deterministic methods. For the efficient solution of large-scale market clearing problems, existing studies mainly focus on model scale reduction to reduce the model solution burden and improve computational efficiency. Among them, model scale reduction can be divided into constraint scale reduction and variable scale reduction. Constraint scale reduction mainly uses the physical characteristics of the model, mathematical information, solution information provided by auxiliary optimization problems, or data-driven methods to identify and eliminate ineffective constraints in the clearing model. Variable scale reduction includes variable aggregation and variable fixation. Among them, variable aggregation aims to aggregate similar variables in combination with the physical meaning of variables in the clearing problem; variable fixation mainly fixes the values ​​of some variables in advance based on historical data information, model load level, relaxation solution information, etc., thereby reducing the model feasible domain and improving the solution speed. Although the above methods can improve the efficiency of solving, they usually lose some information and it is difficult to ensure the feasibility and accuracy of the solution. At the same time, existing research mainly focuses on reducing the scale of intra-provincial transaction models, which is difficult to directly apply to inter-provincial power transaction clearing. Summary of the invention

[0005] The purpose of the present invention is to provide an optimization calculation method for the inter-provincial power transaction clearing problem considering the unexecutable power expectation penalty, comprising the following steps:

[0006] 1) Taking social welfare maximization as the optimization goal, establish a medium- and long-term inter-provincial electricity trading model that takes into account the low-carbon value of green electricity;

[0007] 2) Construct an expression for the green power non-executable power penalty that takes into account the output distribution characteristics, and perform linearization to obtain the penalty that takes into account the linearized non-executable power expectation;

[0008] 3) Based on the penalty of linearized unexecutable power expectations, the objective function of the inter-provincial power medium- and long-term trading model is modified to construct an efficient clearing model for inter-provincial power medium- and long-term trading;

[0009] 4) Solve the efficient clearing model for medium- and long-term inter-provincial electricity transactions and obtain the clearing solution for medium- and long-term inter-provincial electricity transactions.

[0010] Furthermore, the objective function maxU of the medium- and long-term inter-provincial electricity trading model taking into account the low-carbon value of green electricity is as follows:

[0011]

[0012] Among them: O represents the set of all seller entities; D represents the set of all buyer entities; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d (t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; represents the cost of carbon emissions;

[0013] Among them, the carbon emission cost C CO2 As shown below:

[0014]

[0015] Among them: F For thermal power units, ο represents the carbon emission factor of the corresponding unit, δ CO2 is the cost per unit of carbon emissions.

[0016] Furthermore, the constraints of the inter-provincial medium- and long-term electricity trading model that takes into account the low-carbon value of green electricity include AC and DC transmission channel power flow calculation constraints, AC and DC transmission channel ATC constraints, seller's winning bid power constraints, buyer's winning bid power constraints, and line change rate constraints.

[0017] Furthermore, the constraints for AC and DC transmission channel power flow calculation are as follows:

[0018]

[0019] Where: T l (t) represents the power flow through the transmission channel l at time t; D r,l Indicates the flow direction of the transaction path r through the transmission channel l, 1 represents the positive direction and -1 represents the reverse direction; μ r,l represents the power flow distribution factor of the transaction path r through the transmission channel l without considering the network loss. For the DC transmission channel, the distribution factor is 1; η r,l represents the network loss conversion coefficient of transaction path r through transmission channel l; R l represents the set of all transaction paths r associated with the transmission channel l; r Represents R l The set of all seller entities related to the path; D r Represents R l The set of all buyer entities related to the path;

[0020] The ATC constraints of AC and DC transmission channels are as follows:

[0021] T l (t)≤ATC l (t) (4)

[0022] Among them: ATC l (t) represents the maximum remaining available transmission capacity of the AC / DC transmission channel l at time t;

[0023] The constraints on the seller's winning bid electricity are as follows:

[0024]

[0025] Where: Q o (t) represents the electricity supply declared by seller o at time t;

[0026] The power constraints for the purchaser's winning bid are as follows:

[0027]

[0028] Where: Q d (t) represents the electricity purchase demand declared by the purchaser d at time t; E r represents the set of all transmission channels on the transaction path r; δ e (t) represents the line loss rate of each transmission channel;

[0029] The line change rate constraint is as follows:

[0030] |T l (t)-T l (t-1)|≤XT l,max (7)

[0031] Among them: XT l,max Indicates the upper limit of the change value of the power flow in adjacent time periods of transmission channel l; T l (t-1) represents the power flow through the transmission channel l at time t-1.

[0032] Furthermore, the non-executable power penalty for green electricity taking into account the output distribution characteristics is as follows:

[0033]

[0034] Where: σ(t) is the green electricity main body O r The penalty coefficient, Green electricity main body r The winning bid for electricity, Unenforceable electricity expectations for green electricity; In order to take into account the output distribution characteristics of green electricity, electricity penalties cannot be implemented.

[0035] Green Electricity Main Bodyr The winning bidder for power As shown below:

[0036]

[0037] in: It is the winning bid electricity of the green power unit located in the mth section; m is the section number of the winning bid electricity of the green power.

[0038] Further, consider the penalty of linearizing the unexecutable power expectation as follows:

[0039]

[0040] Where: m is the segment number of the green power winning bid; is the length of the rectangle of the green generator set located in the mth segment, is the winning bid power of the green power unit located in the mth segment; To consider the penalty of unenforceable power expectation for linearization; is the equivalent penalty coefficient of green electricity in the mth segment; σ(t) is the green electricity entity O r The penalty coefficient.

[0041] Furthermore, the objective function maxG of the efficient clearing model of inter-provincial power medium- and long-term trading is 1 As shown below:

[0042]

[0043] in: To consider the penalty of linearized unexecutable power expectation; O represents the set of all seller entities; D represents the set of all buyers; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d (t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; Represents the cost of carbon emissions.

[0044] Furthermore, the constraints of the efficient clearing model for inter-provincial medium- and long-term electricity transactions are the same as the constraints of the inter-provincial medium- and long-term electricity transactions model that takes into account the low-carbon value of green electricity, including AC and DC transmission channel power flow calculation constraints, AC and DC transmission channel ATC constraints, seller's winning bid power constraints, buyer's winning bid power constraints, and line change rate constraints.

[0045] Furthermore, the steps to solve the efficient clearing model of inter-provincial power medium- and long-term transactions include:

[0046] s1) Decouple each time period, remove the constraints of time period coupling to obtain the decoupled model of the entire time period, and extract each single time period model separately, expressed as m s (t):

[0047]

[0048] s2) Solve each single-period model in parallel to extract transaction feature information within each period;

[0049] s3) Based on the transaction characteristic information, the transaction probability of the purchase and sale of electricity is determined, and the purchase and sale of electricity transactions with a transaction probability less than a preset probability threshold are fixed to zero, and a clearing model m2 with a reduced feasible domain size is constructed;

[0050] s4) Solve the clearing model m2 and obtain the clearing plan for medium- and long-term inter-provincial electricity transactions.

[0051] Furthermore, the transaction characteristic information includes the transaction pairs that were cleared within the time period and the path information they passed through.

[0052] The technical effect of the present invention is undoubted, and the beneficial effects of the present invention are as follows:

[0053] 1) The present invention constructs a medium- and long-term inter-provincial electricity transaction clearing model that takes into account the low-carbon value and random characteristics of green electricity.

[0054] The present invention embeds the carbon emission cost and the penalty for unenforceable green electricity into the objective function, making clean and low-carbon green electricity more advantageous in market competition. At the same time, the part that exceeds the reliable power generation capacity is penalized according to its expectations to improve the enforceability of the green electricity winning bid. Among them, the penalty coefficient of unenforceable green electricity is dynamically set according to the distribution characteristics of green electricity output, so that the penalty term can more accurately reflect the impact of the uncertainty distribution of green electricity on the enforceability of its winning bid.

[0055] 2) The present invention proposes an acceleration strategy for large-scale inter-provincial medium- and long-term electricity trading problems based on fixed transaction variables. In multi-period inter-provincial medium- and long-term electricity trading, the clearing situation of trading pairs is mostly similar to that of the single-period model, and is less affected by the coupling between periods. Based on this, the various periods in the inter-provincial medium- and long-term electricity trading are first decoupled, and the calculation results that can reflect the clearing situation of trading pairs in each period are quickly obtained by solving the single-period model in parallel. Based on this, the relevant transaction variables of the purchase and sale pairs that are unlikely to be traded are fixed to zero and the corresponding transaction paths are eliminated, thereby constructing a small-scale trading model that does not affect the feasibility of the original problem (without simplifying the operating constraints), which greatly improves the solution efficiency while ensuring accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a schematic diagram of the reliable generation capacity of green electricity;

[0057] Figure 2 It is a non-executable power diagram for green electricity;

[0058] Figure 3 It is a linear approximate expression of the expected non-executable power of green electricity;

[0059] Figure 4 It is a process of clearing method for inter-provincial power transaction based on fixed transaction variables;

[0060] Figure 5 The winning bid of green power entity 1 under each method;

[0061] Figure 6 The winning bids of green power entity 2 under various methods;

[0062] Figure 7 It is the relative error convergence curve of M3 and M5. DETAILED DESCRIPTION

[0063] 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.

[0064] Embodiment 1:

[0065] See also Figures 1 to 7 , an optimization calculation method for the inter-provincial power transaction clearing problem considering the expected penalty of unexecutable power, comprising the following steps:

[0066] 1) Taking social welfare maximization as the optimization goal, establish a medium- and long-term inter-provincial electricity trading model that takes into account the low-carbon value of green electricity;

[0067] 2) Construct an expression for the green power non-executable power penalty that takes into account the output distribution characteristics, and perform linearization to obtain the penalty that takes into account the linearized non-executable power expectation;

[0068] 3) Based on the penalty of linearized unexecutable power expectations, the objective function of the inter-provincial power medium- and long-term trading model is modified to construct an efficient clearing model for inter-provincial power medium- and long-term trading;

[0069] 4) Solve the efficient clearing model for medium- and long-term inter-provincial electricity transactions and obtain the clearing solution for medium- and long-term inter-provincial electricity transactions.

[0070] The objective function maxU of the inter-provincial medium- and long-term electricity trading model taking into account the low-carbon value of green electricity is as follows:

[0071]

[0072] Among them: O represents the set of all seller entities; D represents the set of all buyer entities; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d (t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; represents the cost of carbon emissions;

[0073] Among them, carbon emission costs As shown below:

[0074]

[0075] Among them: F For thermal power units, ο represents the carbon emission factor of the corresponding unit, δ CO2 is the cost per unit of carbon emissions.

[0076] The constraints of the inter-provincial medium- and long-term electricity trading model that takes into account the low-carbon value of green electricity include AC and DC transmission channel power flow calculation constraints, AC and DC transmission channel ATC constraints, seller's winning bid power constraints, buyer's winning bid power constraints, and line change rate constraints.

[0077] The constraints for AC and DC transmission channel power flow calculation are as follows:

[0078]

[0079] Where: T l (t) represents the power flow through the transmission channel l at time t; D r,l Indicates the flow direction of the transaction path r through the transmission channel l, 1 represents the positive direction and -1 represents the reverse direction; μ r,l represents the power flow distribution factor of the transaction path r through the transmission channel l without considering the network loss. For the DC transmission channel, the distribution factor is 1; η r,l represents the network loss conversion coefficient of transaction path r through transmission channel l; R l represents the set of all transaction paths r associated with the transmission channel l; r Represents R l The set of all seller entities related to the path; Dr Represents R l The set of all buyer entities related to the path;

[0080] The ATC constraints of AC and DC transmission channels are as follows:

[0081] T l (t)≤ATC l (t) (4)

[0082] Among them: ATC l (t) represents the maximum remaining available transmission capacity of the AC / DC transmission channel l at time t;

[0083] The constraints on the seller's winning bid electricity are as follows:

[0084]

[0085] Where: Q o (t) represents the electricity supply declared by seller o at time t;

[0086] The power constraints for the purchaser's winning bid are as follows:

[0087]

[0088] Where: Q d (t) represents the electricity purchase demand declared by the purchaser d at time t; E r represents the set of all transmission channels on the transaction path r; δ e (t) represents the line loss rate of each transmission channel;

[0089] The line change rate constraint is as follows:

[0090] |T l (t)-T l (t-1)|≤XT l,max (7)

[0091] Among them: XT l,max Indicates the upper limit of the change value of the power flow in adjacent time periods of transmission channel l; T l (t-1) represents the power flow through the transmission channel l at time t-1.

[0092] The non-executable power penalties for green electricity taking into account the output distribution characteristics are as follows:

[0093]

[0094] Where: σ(t) is the green electricity main body O r The penalty coefficient, Green electricity main body r The winning bid for electricity, Unenforceable electricity expectations for green electricity; In order to take into account the output distribution characteristics of green electricity, electricity penalties cannot be implemented.

[0095] Green Electricity Main Body r The winning bidder for power As shown below:

[0096]

[0097] in: It is the winning bid electricity of the green power unit located in the mth section; m is the section number of the winning bid electricity of the green power.

[0098] Consider the penalty for linearizing the unexecutable power expectation as follows:

[0099]

[0100] Where: m is the segment number of the green power winning bid; is the length of the rectangle of the green generator set located in the mth segment, is the winning bid power of the green power unit located in the mth segment; To consider the penalty of unenforceable power expectation for linearization; is the equivalent penalty coefficient of green electricity in the mth segment; σ(t) is the green electricity entity O r The penalty coefficient.

[0101] Objective function maxG of the efficient clearing model for medium- and long-term inter-provincial electricity transactions 1 As shown below:

[0102]

[0103] in: To consider the penalty of linearized unexecutable power expectation; O represents the set of all seller entities; D represents the set of all buyers; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d (t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; Represents the cost of carbon emissions.

[0104] The constraints of the efficient clearing model for inter-provincial medium- and long-term electricity transactions are the same as those of the inter-provincial medium- and long-term electricity transactions model that takes into account the low-carbon value of green electricity, including AC and DC transmission channel power flow calculation constraints, AC and DC transmission channel ATC constraints, seller's winning bid power constraints, buyer's winning bid power constraints, and line change rate constraints.

[0105] The steps to solve the efficient clearing model of inter-provincial power medium- and long-term transactions include:

[0106] s1) Decouple each time period, remove the constraints of time period coupling to obtain the decoupled model of the entire time period, and extract each single time period model separately, expressed as m s (t):

[0107]

[0108] s2) Solve each single-period model in parallel to extract transaction feature information within each period;

[0109] s3) Based on the transaction characteristic information, the transaction probability of the purchase and sale of electricity is determined, and the purchase and sale of electricity transactions with a transaction probability less than a preset probability threshold are fixed to zero, and a clearing model m2 with a reduced feasible domain size is constructed;

[0110] s4) Solve the clearing model m2 and obtain the clearing plan for medium- and long-term inter-provincial electricity transactions.

[0111] Transaction feature information includes the transaction pairs that were cleared within the time period and the paths they passed through.

[0112] Embodiment 2:

[0113] An optimal calculation method for the inter-provincial power transaction clearing problem considering the expected penalty for unexecutable power includes the following steps:

[0114] 1) Taking social welfare maximization as the optimization goal, establish a medium- and long-term inter-provincial electricity trading model that takes into account the low-carbon value of green electricity;

[0115] 2) Construct an expression for the green power non-executable power penalty that takes into account the output distribution characteristics, and perform linearization to obtain the penalty that takes into account the linearized non-executable power expectation;

[0116] 3) Based on the penalty of linearized unexecutable power expectations, the objective function of the inter-provincial power medium- and long-term trading model is modified to construct an efficient clearing model for inter-provincial power medium- and long-term trading;

[0117] 4) Solve the efficient clearing model for medium- and long-term inter-provincial electricity transactions and obtain the clearing solution for medium- and long-term inter-provincial electricity transactions.

[0118] Embodiment 3:

[0119] An optimal calculation method for the inter-provincial power trading clearing problem considering the expected penalty for unexecutable electricity is proposed. The technical content is the same as that of rating 2. Furthermore, the objective function of the inter-provincial power medium- and long-term trading model taking into account the low-carbon value of green electricity is as follows:

[0120]

[0121] Among them: O represents the set of all seller entities; D represents the set of all buyer entities; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d (t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; represents the cost of carbon emissions;

[0122] Among them, carbon emission costs As shown below:

[0123]

[0124] Among them: F For thermal power units, ο represents the carbon emission factor of the corresponding unit, δ CO2 is the cost per unit of carbon emissions.

[0125] Embodiment 4:

[0126] An optimal calculation method for the inter-provincial power transaction clearing problem taking into account the expected penalty for unenforceable electricity. The technical content is the same as any one of ratings 2-3. Furthermore, the constraints of the inter-provincial power medium- and long-term transaction model taking into account the low-carbon value of green electricity include AC and DC transmission channel flow calculation constraints, AC and DC transmission channel ATC constraints, seller's winning power constraints, buyer's winning power constraints, and line change rate constraints.

[0127] Embodiment 5:

[0128] An optimal calculation method for the inter-provincial power trading clearing problem considering the penalty for unenforceable power expectation is provided. The technical content is the same as any one of the ratings 2-4. Furthermore, the AC and DC transmission channel power flow calculation constraints are as follows:

[0129]

[0130] Where: T l (t) represents the power flow through the transmission channel l at time t; D r,lIndicates the flow direction of the transaction path r through the transmission channel l, 1 represents the positive direction and -1 represents the reverse direction; μ r,l represents the power flow distribution factor of the transaction path r through the transmission channel l without considering the network loss. For the DC transmission channel, the distribution factor is 1; η r,l represents the network loss conversion coefficient of transaction path r through transmission channel l; R l represents the set of all transaction paths r associated with the transmission channel l; r Represents R l The set of all seller entities related to the path; D r Represents R l The set of all buyer entities related to the path;

[0131] The ATC constraints of AC and DC transmission channels are as follows:

[0132] T l (t)≤ATC l (t) (4)

[0133] Among them: ATC l (t) represents the maximum remaining available transmission capacity of the AC / DC transmission channel l at time t;

[0134] The constraints on the seller's winning bid electricity are as follows:

[0135]

[0136] Where: Q o (t) represents the electricity supply declared by seller o at time t;

[0137] The power constraints for the purchaser's winning bid are as follows:

[0138]

[0139] Where: Q d (t) represents the electricity purchase demand declared by the purchaser d at time t; E r represents the set of all transmission channels on the transaction path r; δ e (t) represents the line loss rate of each transmission channel;

[0140] The line change rate constraint is as follows:

[0141] |T l (t)-T l (t-1)|≤XT l,max (7)

[0142] Among them: XT l,max Indicates the upper limit of the change value of the power flow in adjacent time periods of transmission channel l; T l(t-1) represents the power flow through the transmission channel l at time t-1.

[0143] Embodiment 6:

[0144] An optimal calculation method for the inter-provincial power trading clearing problem considering the expected penalty for unenforceable power is provided. The technical content is the same as any one of the ratings 2-5. Furthermore, the unenforceable power penalty for green power taking into account the output distribution characteristics is as follows:

[0145]

[0146] Where: σ(t) is the green electricity main body O r The penalty coefficient, Green electricity main body r The winning bid for electricity, Unenforceable electricity expectations for green electricity; In order to take into account the output distribution characteristics of green electricity, electricity penalties cannot be implemented.

[0147] Green Electricity Main Body r The winning bidder for power As shown below:

[0148]

[0149] in: It is the winning bid electricity of the green power unit located in the mth section; m is the section number of the winning bid electricity of the green power.

[0150] Embodiment 7:

[0151] An optimal calculation method for the inter-provincial power trading clearing problem considering the penalty for unenforceable power expectations is provided. The technical content is the same as any one of the ratings 2-6. Furthermore, the penalty for linearized unenforceable power expectations is as follows:

[0152]

[0153] Where: m is the segment number of the green power winning bid; is the length of the rectangle of the green generator set located in the mth segment, is the winning bid power of the green power unit located in the mth segment; To consider the penalty of unenforceable power expectation for linearization; is the equivalent penalty coefficient of green electricity in the mth segment; σ(t) is the green electricity entity O r The penalty coefficient.

[0154] Embodiment 8:

[0155] An optimal calculation method for the inter-provincial power transaction clearing problem considering the expected penalty for unexecutable power. The technical content is the same as any one of the ratings 2-7. Furthermore, the objective function of the inter-provincial power medium- and long-term transaction efficient clearing model is as follows:

[0156]

[0157] in: To consider the penalty of linearized unexecutable power expectation; O represents the set of all seller entities; D represents the set of all buyers; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d (t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; Represents the cost of carbon emissions.

[0158] Embodiment 9:

[0159] An optimal calculation method for the inter-provincial power transaction clearing problem taking into account the expected penalty for unenforceable electricity. The technical content is the same as any one of ratings 2-8. Furthermore, the constraints of the efficient clearing model for inter-provincial power medium- and long-term transactions are the same as the constraints of the inter-provincial power medium- and long-term transactions taking into account the low-carbon value of green electricity, including AC and DC transmission channel flow calculation constraints, AC and DC transmission channel ATC constraints, seller's winning power constraints, buyer's winning power constraints, and line change rate constraints.

[0160] The steps to solve the efficient clearing model of inter-provincial power medium- and long-term transactions include:

[0161] s1) Decouple each time period, remove the constraints of time period coupling to obtain the decoupled model of the entire time period, and extract each single time period model separately, expressed as m s (t):

[0162]

[0163] s2) Solve each single-period model in parallel to extract transaction feature information within each period;

[0164] s3) Based on the transaction characteristic information, the transaction probability of the purchase and sale of electricity is determined, and the purchase and sale of electricity transactions with a transaction probability less than a preset probability threshold are fixed to zero, and a clearing model m2 with a reduced feasible domain size is constructed;

[0165] s4) Solve the clearing model m2 and obtain the clearing plan for medium- and long-term inter-provincial electricity transactions.

[0166] Embodiment 10:

[0167] A method for optimizing the calculation of the inter-provincial power transaction clearing problem taking into account the expected penalty for unenforceable electricity. The technical content is the same as any one of the ratings 2-9. Furthermore, the transaction characteristic information includes the transaction pairs that have been cleared within the time period and the path information they have passed through.

[0168] Embodiment 11:

[0169] An optimization calculation method for the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation is proposed. First, a medium- and long-term inter-provincial power trading model considering the low-carbon value of green electricity is established, with the maximization of social welfare as the optimization goal, and the remaining available transmission capacity of the transmission channel and the transmission power change limit of the adjacent time periods of the transmission channel (time period coupling constraint) as the operation constraints; then, the objective function of the clearing model of the unexecutable power expectation of green electricity is modified, and the unexecutable power penalty of green electricity considering the output distribution characteristics is introduced to improve the executability of the green electricity clearing scheme, and an acceleration method based on pre-fixation of transaction variables is proposed, which can effectively improve the calculation speed at the cost of a small loss of accuracy; finally, the effectiveness of the proposed method in resource allocation effect and the effectiveness of the clearing acceleration strategy are verified through example analysis. The simulation results on three different scale examples show that the inter-provincial power trading clearing method proposed by the present invention to improve the executability of the green electricity configuration scheme has significant improvements in carbon emission reduction and reducing green electricity output deviation, and the proposed acceleration strategy significantly improves the solution efficiency.

[0170] The specific steps are as follows:

[0171] 1. Inter-provincial medium- and long-term electricity trading model taking into account the low-carbon value of green electricity

[0172] The inter-provincial medium- and long-term electricity trading model that takes into account the low-carbon value of green electricity generally takes the maximization of social welfare as the optimization goal, taking into account the carbon emission costs, and taking the remaining available transfer capacity (Available Transfer Capability, ATC) of the transmission channel and the transmission power change limit in adjacent time periods of the transmission channel (time period coupling constraints) as operating constraints. The specific introduction is as follows.

[0173] (1) Objective function: maximizing social welfare

[0174]

[0175] Among them: O represents the set of all seller entities; D represents the set of all buyer entities; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d(t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; Represents the cost of carbon emissions.

[0176] Carbon emissions costs The calculation method is as follows:

[0177]

[0178] Among them: F For thermal power units, ο represents the carbon emission factor of the corresponding unit 0.832, δ CO2 The cost per unit of carbon emissions (45.61 yuan / ton minus the annual average carbon price).

[0179] (2) Operation constraints:

[0180] 1) Constraints on AC and DC transmission channel power flow calculation:

[0181]

[0182] Where: T l (t) represents the power flow through the transmission channel l at time t; D r,l Indicates the flow direction of the transaction path r through the transmission channel l, 1 represents the positive direction and -1 represents the reverse direction; μ r,l represents the power flow distribution factor of the transaction path r through the transmission channel l without considering the network loss. For the DC transmission channel, the distribution factor is 1; η r , l represents the network loss conversion coefficient of transaction path r through transmission channel l; R l represents the set of all transaction paths r associated with the transmission channel l; r Represents R l The set of all seller entities related to the path; D r Represents R l The collection of all buyer entities related to the path.

[0183] 2) ATC constraints for AC and DC transmission channels:

[0184] T l (t)≤ATC l (t) (4)

[0185] Among them: ATC l (t) represents the maximum remaining available transmission capacity of the AC / DC transmission channel l at time t.

[0186] 3) Constraints on the power supply winning bidder:

[0187]

[0188] Where: Q o (t) represents the electricity supply reported by the seller o at time t.

[0189] 4) Purchaser's power bidding constraints:

[0190]

[0191] Where: Q d (t) represents the electricity purchase demand declared by the purchaser d at time t; E r Represents the set of all transmission channels passing through the transaction path r.

[0192] 5) Line change rate constraint:

[0193] |T l (t)-T l (t-1)|≤XT l,max (7)

[0194] Among them: XT l,max It indicates the upper limit of the change value of the power flow in adjacent time periods of transmission channel l.

[0195] The clearing model proposed in the present invention introduces carbon emission costs into the objective function, which can make green electricity have an advantage in market transactions, thereby reducing the carbon emissions of the clearing scheme.

[0196] 2. Correction method of objective function of clearing model considering the non-executable power expectation of green electricity

[0197] The random characteristics of green electricity will affect the feasibility of the medium- and long-term inter-provincial electricity trading clearing plan. Specifically, when the actual power that green electricity can generate is lower than its declared power based on the output forecast, it may be impossible for green electricity to actually implement the medium- and long-term power trading clearing plan, and subsequent plan adjustments need to be introduced, which will destroy the optimality of the original clearing plan. However, if the green electricity winning bid is limited to a conservative range in order to ensure the feasibility of the clearing plan, the actual power that green electricity can generate will be much higher than its winning bid, resulting in insufficient utilization of green electricity generation capacity and wind and solar power abandonment. In this regard, the present invention introduces a non-executable power penalty for green electricity that takes into account the output distribution characteristics. While improving the feasibility of the green electricity clearing plan, it avoids the clearing plan from being too conservative. The specific explanation is as follows:

[0198] 1) Reliable green power generation capacity and unfeasible power expectations

[0199] The reliable power generation capacity of green electricity refers to the output level that green electricity can achieve under a given confidence level. Taking wind power as an example, under a given confidence level α, its reliable power generation capacity is Pα ,like Figure 1 shown.

[0200] The curve in the figure is the probability density function of green power output (which can be obtained through quantile regression, kernel density estimation, etc.), and the reliable power generation capacity of green power P α satisfy:

[0201]

[0202] Among them, O r For green power unit collection, Contribute to the prediction of green electricity.

[0203] When the green power winning bid exceeds its reliable power generation capacity, the excess may be unenforceable. Figure 2 The black curve represents the predicted cumulative distribution function of green power output, and the shaded part is the expectation of green power's unexecutable power. The larger the green power's winning bid, the larger its unexecutable power expectation. The present invention embeds the green power's unexecutable power expectation into the inter-provincial power medium- and long-term transaction clearing model with an acceptable complexity, so as to more accurately describe the correlation between the green power's winning bid and its executability, and improve the effectiveness of the clearing scheme.

[0204] 2) Objective function correction method considering the reliable generation capacity of green electricity

[0205] In order to improve the feasibility of the inter-provincial medium- and long-term electricity trading clearing scheme, the impact of green electricity uncertainty needs to be considered in the clearing model. To this end, the present invention embeds a penalty term related to the unexecutable power expectation of green electricity in the objective function of the clearing model. The modified objective function is shown in formula (9):

[0206]

[0207] Where W is the green electricity non-executable power penalty term. Ideally, this penalty term should be the product of the green electricity non-executable power expectation and the penalty coefficient, as shown in Formula 10:

[0208]

[0209] Among them, σ(t) is the main green electricity O r The penalty coefficient, Green electricity main body r The winning bid for electricity, This can accurately reflect the relationship between the green power winning bid power and its unexecutable power, thereby effectively improving the executability of the green power winning bid power in the clearing plan. The accurate calculation method of is relatively complex and difficult to be directly embedded in the clearing model. Therefore, the present invention uses the area of ​​multiple rectangles to approximate the shadow area corresponding to the unexecutable power expectation, such as Figure 3 shown.

[0210] Depend on Figure 3 It can be seen that according to the cumulative distribution function of green power output, the reliable power generation capacity of green power is taken as the starting point and the maximum output is taken as the end point, which is divided into n segments. The area of ​​each segment uses the area of ​​the rectangle instead of the area of ​​the curved trapezoid, thereby realizing the linear approximation of the unexecutable power expectation. The penalty term after considering the linearized unexecutable power expectation can be expressed as follows:

[0211]

[0212] Where m is the segment number of the green power winning bid (from 0 to n, a total of n+1 segments, where m=0 indicates that the green power output is below the green power reliable output). is the length of the rectangle of the green generator set located in the mth segment, is the winning bid power of the green power unit in the mth segment (i.e., the width of the rectangle). For a green power entity, the sum of its winning bid power in all segments should be equal to the total winning bid power of the entity, that is:

[0213]

[0214] is the equivalent penalty coefficient of green electricity in the mth segment (i.e. Figure 3 The equivalent penalty coefficient increases with the increase of the number of the winning power segment. Therefore, in the optimization process of minimizing the penalty term of the objective function, the segmented winning power of green electricity will start from the segment with a smaller number, ensuring the accuracy of the expected calculation result that the winning power cannot execute.

[0215] Embedding the relevant penalty terms of unexecutable electricity expectations into the objective function can minimize the unexecutable electricity of green electricity in the optimization process, thereby reducing the situation of "actual output of green electricity is less than the bid electricity" at the market clearing level, improving the enforceability of market clearing results, and improving the efficiency of system operation.

[0216] The model after modifying the objective function can be simplified as m 1 ,Right now:

[0217]

[0218] 3. Inter-provincial power transaction clearing method based on fixed transaction variables

[0219] The inter-provincial medium- and long-term electricity trading method proposed in this invention, which takes into account the low-carbon value and random characteristics of green electricity, can improve the resource allocation effect of the market clearing scheme under large-scale access of green electricity, but it needs to introduce new variables and constraints, which will increase the computational burden of the clearing problem. In order to meet the actual market clearing calculation efficiency requirements, this invention proposes an acceleration method based on pre-fixing of transaction variables, which can effectively improve the calculation speed at the cost of a small loss of precision.

[0220] The inter-provincial medium- and long-term electricity trading model constructed by the present invention is a multi-period clearing model with period coupling. However, the clearing conditions of trading pairs in the multi-period model are mostly similar to those in the single-period model. Trading pairs with higher social welfare or lower carbon emission costs will be easier to clear and will be less affected by the coupling between periods. Therefore, the trading conditions of each single-period clearing model when the period coupling constraint is not considered can be used as a reference to exclude trading pairs that have not won the bid under all single-period models, so as to reduce the scale of the original multi-period clearing model. Based on the above ideas, the present invention proposes an inter-provincial electricity trading clearing method based on fixed transaction variables. The specific framework is as follows: Figure 4 shown.

[0221] Considering that a large part of the complexity of the multi-period problem is caused by the coupling constraints between periods, this section will first decouple each period (i.e., remove the constraints of period coupling), obtain the decoupled model of the entire period, and extract each single period model separately, denoted as m s (t):

[0222]

[0223] Secondly, the single-period models are solved in parallel to extract the transaction feature information in each period, including the transaction pairs that have been cleared in the period and the path information they have passed through; then, guided by this information, in the original multi-period coupling problem, the purchase and sale pairs of electricity transactions that are unlikely to be completed are fixed to zero, and the optimization range of the transaction path is narrowed, thereby constructing a transaction model with a smaller feasible domain size. This transaction model is represented by m 2 Specifically, the pseudo code for fixed transaction variables is as follows:

[0224]

[0225]

[0226] After the transaction variables are fixed, a more simplified model (i.e., m 2 ), compared with the original model, due to m 2 The purchase and sale pairs that are unlikely to be traded are excluded, and the optimization range of the transaction paths of feasible purchase and sale pairs is narrowed, and the model scale is greatly reduced compared with the original model. 2All constraints of the original model are retained, and only part of the optimization space of the model is reduced by fixing the transaction variables, ensuring the feasibility of its calculation results for the original clearing problem. At the same time, the transaction variable fixation is based on the solution results of each single-period clearing model, ensuring the accuracy of variable fixation. Therefore, although the optimality of the original clearing problem cannot be fully guaranteed, the model clearing result will be close to the optimal solution of the original model.

[0227] Embodiment 12:

[0228] The verification of the inter-provincial power trading clearing method for improving the executability of the green power configuration scheme described in any one of Examples 1-11 is as follows:

[0229] 1. Example settings

[0230] This embodiment uses actual operation data of inter-provincial medium- and long-term electricity transactions to carry out case analysis to verify the effectiveness of the proposed method.

[0231] This example will conduct a comparative analysis of the clearing method on three different examples (S1-S3), S1 is smaller in scale, and S2 and S3 are larger. S1 and S2 will be used to illustrate the effectiveness of the proposed method in terms of resource allocation; S2 and S3 will be used to illustrate the effectiveness of the proposed clearing acceleration strategy. The scale of each example (S1-S3) is detailed in Table 1:

[0232] Table 1 Problem scale of inter-provincial power trading in each example

[0233]

[0234]

[0235] The comparison method in the example simulation is as follows:

[0236] M0: The traditional method of clearing medium- and long-term inter-provincial electricity transactions, which does not take into account the low-carbon value of green electricity and non-executable electricity.

[0237] M1: A method for clearing medium- and long-term inter-provincial electricity transactions that only considers the low-carbon value of green electricity;

[0238] M2: A method for clearing medium- and long-term inter-provincial electricity transactions that takes into account the low-carbon value of green electricity and unexecutable electricity, but imposes a penalty on all parts that exceed reliable power generation capacity, where the α of reliable power generation capacity is 5%;

[0239] M3: A method for clearing medium- and long-term inter-provincial electricity transactions that considers the low-carbon value of green electricity and unexecutable electricity, with penalties imposed on unexecutable electricity expectations, where the α of reliable power generation capacity is 5%;

[0240] M4: Same as M3, but uses a lossless acceleration strategy based on the adjoint model;

[0241] M5 (proposed method): same as M3, but adopts an acceleration strategy based on fixed transaction variables;

[0242] 2. Construction of a medium- and long-term inter-provincial electricity trading model taking into account the low-carbon value and random characteristics of green electricity

[0243] The inter-provincial medium- and long-term electricity trading model taking into account the low-carbon value of green electricity takes the maximization of social welfare as the optimization goal, as shown in Formula 1, and takes the remaining available transmission capacity of the transmission channel and the transmission power change limit of the adjacent time periods of the transmission channel as the operation constraints, as shown in Formulas 3 to 7. At the same time, considering the situation that green electricity cannot actually implement the medium- and long-term electricity trading clearing plan, in order to ensure the executability of the clearing plan, the present invention introduces the green electricity non-executable power penalty taking into account the output distribution characteristics, and embeds the green electricity non-executable power expectation into the inter-provincial power medium- and long-term trading clearing model with an acceptable complexity, and obtains the revised objective function as shown in Formula 9, which improves the executability of the green electricity clearing plan while avoiding the clearing plan being too conservative.

[0244] 3. Inter-provincial power transaction clearing method based on fixed transaction variables

[0245] In order to meet the actual market clearing calculation efficiency requirements, the present invention takes the transaction situation of each single-period clearing model without considering the period coupling constraint as a reference, excludes the transaction pairs that have not won the bid under all single-period models, so as to reduce the scale of the original multi-period clearing model, and decouples each period to obtain a full-period decoupled model, such as formula (14), and then solves each single-period model in parallel to extract the transaction feature information in each period. With this information as a guide, in the original multi-period coupling problem, the purchase and sale pair power transactions that are unlikely to be traded are fixed to zero, and the optimization range of the transaction path is narrowed, so as to construct a transaction model with a smaller feasible domain scale. The model clearing result will be close to the optimal solution of the original model, at the cost of a small loss of accuracy, effectively improving the calculation speed.

[0246] 4. Simulation analysis

[0247] According to the inter-provincial power medium- and long-term trading model taking into account the low-carbon value and random characteristics of green electricity and the inter-provincial power trading clearing method based on fixed trading variables established in the present invention, the effectiveness of resource allocation and the effectiveness of the acceleration algorithm can be verified through comparative simulation analysis of examples at different scales.

[0248] 1) Verification of resource allocation effectiveness

[0249] In order to verify the effectiveness of the proposed inter-provincial medium- and long-term electricity trading method that takes into account the low-carbon value and random characteristics of green electricity in reducing emissions and improving the feasibility of the clearing scheme, the present invention will take S1 and S2 as examples to compare the total non-executable power of green electricity, the amount of abandoned wind and solar power, the total amount of green electricity exceeding the reliable power generation capacity, the carbon emission cost and the social welfare of the transaction in M1, M2 and M3, as shown in Table 2. Among them, it is assumed that the output distribution of each green electricity entity obeys a normal distribution with a mean of the declared amount, and the actual output is obtained by random sampling of the output distribution. The unit carbon emission factor is 0.8426kgCO2 / kWh, and the unit carbon emission cost is 68.15 yuan / ton.

[0250] In the table, g 1 The total non-executable power of green electricity refers to the part of green electricity winning power that exceeds the actual output of green electricity, in MW; g 2 The amount of wind and solar power abandoned refers to the part of the actual output of green electricity that exceeds the bid power of green electricity, in MW; g 3 The total amount of green electricity exceeding the reliable power generation capacity, in MW; g 4 is the carbon emission cost, in yuan; g 5 Transactional social welfare refers to the social welfare generated only by inter-provincial transactions, and the unit is Yuan.

[0251] Table 2 Resource allocation effect of M1-M3 clearing plan

[0252]

[0253]

[0254] As can be seen from Table 2, M0 only considers the maximization of social welfare, so its carbon emission cost is higher than that of the subsequent methods; M1 maximizes social welfare under the premise of only considering the low-carbon value of green electricity, so it can ensure the optimality of social welfare and low-carbon nature of transactions, but in the resulting clearing scheme, the unexecutable power of green electricity is high, and it is difficult to guarantee the actual resource allocation effect. Compared with M1, M2 effectively reduces the total amount of green electricity exceeding the reliable power generation capacity (in the two examples, the green electricity bid power is almost the same as the reliable power generation capacity) and the unexecutable power of green electricity (in the two examples, it is reduced by 99.35% on average compared with M1). However, since M2 punishes all parts that exceed the reliable power generation capacity, the wind and solar power abandonment in the M2 results also increases (an average increase of 502.77% in the two examples). In contrast, M3 punishes the unexecutable power expectations of green electricity and realizes the reasonable allocation of the winning power. Therefore, compared with M2, M3 breaks through the conservative limit of reliable power generation capacity and effectively reduces the amount of wind and solar power abandonment (an average reduction of 310.62% relative to M2 in the two examples), while also controlling the unexecutable power of green electricity (an average reduction of 99.30% relative to M1 in the two examples). In addition, M2 and M3 will increase some carbon emission costs and reduce some transaction social welfare by reducing the winning power of green electricity (relative to M1) during clearing, but effectively improve the enforceability of the clearing plan.

[0255] In summary, M3 can effectively control social welfare / carbon emission costs and wind and solar curtailment while effectively reducing the non-executable electricity of green electricity, and promote the rational allocation of resources. Below, S1 will be taken as an example to more intuitively illustrate the impact of introducing the penalty term of non-executable electricity of green electricity on the resource allocation effect. Two green electricity entities with different distributions will be taken as examples. Their output cumulative distribution functions in a certain period of time and the winning bids under different methods are detailed in Figure 5 and Figure 6 .

[0256] Depend on Figure 5 and Figure 6 It can be seen that among the two green power entities, M1 introduced the low-carbon value of green electricity compared with M0, resulting in an increase in the winning power of the green power entity; M2 penalized all parts that exceeded the reliable power generation capacity on the basis of M1. Under this method, the winning power of green electricity was almost consistent with the reliable output, but it generated more wind and solar power abandonment; M3 penalized the unexecutable power expectations of green electricity on the basis of M1. When the cumulative distribution function of the green power entity is relatively flat, it allows the winning power to exceed part of the reliable output (such as Figure 5 Green power subject 1), and when the cumulative distribution function of the green power subject is steeper, the winning power will also be limited to the vicinity of the reliable output (such as Figure 5Green electricity is the main body1), thereby reducing the amount of wind and solar power curtailment overall.

[0257] In addition, the test found that introducing carbon emission reduction costs and green electricity non-executable electricity penalty items into the model will increase the complexity of the model and the solution time. Taking S2 as an example, the solution time of M1 is 116.75s, while the solution time of M2 and M3 is 134.54s and 141.02s, respectively, an increase of 15.23% and 20.7%.

[0258] 2) Accelerate algorithm effectiveness verification

[0259] In order to verify the effectiveness of the proposed inter-provincial power transaction clearing method based on fixed transaction variables, the present invention will take S2 ​​and S3 as examples to compare the solution performance and solution results of M3, M4 and M5, as shown in Table 4.

[0260] Table 4 Comparison of solution efficiency and resource allocation effect of M1, M3 and M4

[0261]

[0262] From Table 4, M4 uses the solution of the adjoint model to guide the hot start of the original model, and can obtain the optimal solution without loss, with an average acceleration ratio of 1.53. The M5 method can significantly reduce the time consumed for solution (the average acceleration ratio relative to M3 in the two examples is 2.20). The difference between M5 and M3 in the solution results is smaller. The total non-executable power of green electricity and the amount of wind and solar power abandoned are consistent. The average relative error of transaction social welfare is 0.27%, and the average relative error of carbon emission cost is 0.56%. It can be verified that the proposed acceleration method has little effect on the resource allocation effect. In addition, compared with M4, which can obtain the optimal solution without loss, M5 only simplifies the model, so the steps are simpler than M4, the time consumed is less, and the solution efficiency is higher. In the actual medium- and long-term clearing of inter-provincial electricity, considering the subsequent safety verification and other steps, M5 is more practical in engineering than M4.

[0263] Furthermore, in order to demonstrate the effect of the proposed acceleration method in detail, Table 5 takes S2 as an example to show the specific parameters of M5 in the solution process (including fixed variable ratio, number of iterations, etc.), where m 0 represents the original model constructed by M3, m s (t i ) represents the single-period model obtained after decoupling between periods, m 2 Represents the full-period model after fixing the transaction variables.

[0264] Table 5 Detailed parameters in the solution process of M3 and M5

[0265]

[0266] As can be seen from Table 5, the proposed method first solves all decoupled single-period models in parallel. The solution time of a single single-period model only takes 0.03s, and the total time consumed by this process is 17.81s. Then, guided by the solution results, most of the transaction variables in the original model are fixed, which greatly reduces the space of the feasible domain. For example, 72.6% of the variables are fixed in Table 5. Finally, it only takes 54.50s to solve the model after fixing the variables, and a high-quality feasible solution close to the resource allocation effect of M3 can be obtained (see Table 4). In summary, in S2, the total speedup ratio of M5 reaches 1.95, which proves the effectiveness of the proposed algorithm in accelerating the solution.

[0267] Figure 6 The convergence curve of the relative error of the objective function of the clearing model in the solution process under the two methods M3 and M5 in Example S3 is shown. Figure 6 In the paper, both M3 and M5 are solved using the simplex method. Since the model constructed by M5 is smaller in scale, the relative error of M5 converges faster than that of M3, and a feasible solution close to the optimal one can be found in a shorter time.

[0268] Comprehensive analysis of the simulation results shows that:

[0269] 1. By introducing a penalty for the part of the winning bid that exceeds the reliable power generation capacity in the objective function, the total amount of green electricity that exceeds the reliable power generation capacity is effectively reduced. By penalizing the unexecutable power expectations of green electricity, the reasonable allocation of the winning bid power is achieved, and the amount of wind and solar power abandonment is effectively reduced. At the same time, the unexecutable power of green electricity is controlled, promoting the rational allocation of resources.

[0270] 2. The method proposed in the present invention fixes most of the transaction variables in the original model, greatly reduces the space of the feasible domain, reduces a lot of model solution time, and the relative error converges faster.

[0271] From the experimental results, it can be seen that the inter-provincial power transaction clearing method proposed in the present invention to improve the executability of the green electricity configuration plan can improve the executability of the clearing plan and the clearing calculation efficiency, and significantly improve the solution efficiency, which can provide theoretical and methodological support for medium- and long-term inter-provincial power transactions with large-scale green electricity access.

[0272] In summary, the present invention proposes an optimal calculation method for the inter-provincial power transaction clearing problem considering the penalty for unexecutable power expectations. The method first establishes an inter-provincial power medium- and long-term transaction model that takes into account the low-carbon value of green electricity; then, the objective function of the clearing model of green electricity unexecutable power expectations is corrected, and an acceleration method based on pre-fixed transaction variables is proposed to effectively improve the calculation speed; finally, the effectiveness of the proposed method in terms of resource allocation effect and the effectiveness of the clearing acceleration strategy are verified through example analysis. As a result, the rapid and accurate clearing of inter-provincial power medium- and long-term transactions under large-scale green power access is achieved, ensuring the resource allocation efficiency of the inter-provincial power medium- and long-term market. The simulation analysis of the example shows that the inter-provincial power transaction clearing method proposed in the present invention to improve the executability of the green electricity configuration scheme can improve the executability of the clearing scheme, the clearing calculation efficiency and the solution efficiency.

Claims

1. An optimal calculation method for the inter-provincial power trading clearing problem considering the penalty for unenforceable power expectation, characterized in that: The following steps are involved: 1) Taking social welfare maximization as the optimization goal, establish a medium- and long-term inter-provincial electricity trading model that takes into account the low-carbon value of green electricity; 2) Construct an expression for the green power non-executable power penalty that takes into account the output distribution characteristics, and perform linearization to obtain the penalty that takes into account the linearized non-executable power expectation; 3) Based on the penalty of linearized unexecutable electricity expectations, the objective function of the inter-provincial medium- and long-term electricity trading model is modified, and an efficient clearing model for inter-provincial medium- and long-term electricity trading is constructed. 4) Solve the efficient clearing model for medium- and long-term inter-provincial electricity transactions and obtain the clearing solution for medium- and long-term inter-provincial electricity transactions.

2. According to claim 1, a method for optimizing the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation is characterized in that: The objective function maxU of the inter-provincial medium- and long-term electricity trading model taking into account the low-carbon value of green electricity is as follows: Among them: O represents the set of all seller entities; D represents the set of all buyer entities; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d (t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; represents the cost of carbon emissions; Among them, carbon emission costs As shown below: Among them: F For thermal power units, ο represents the carbon emission factor of the corresponding unit, δ CO2 is the cost per unit of carbon emissions.

3. The method for optimizing the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation according to claim 1 is characterized in that: The constraints of the inter-provincial medium- and long-term electricity trading model that takes into account the low-carbon value of green electricity include AC and DC transmission channel power flow calculation constraints, AC and DC transmission channel ATC constraints, seller's winning bid power constraints, buyer's winning bid power constraints, and line change rate constraints.

4. The method for optimizing the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation according to claim 3 is characterized in that: The constraints for AC and DC transmission channel power flow calculation are as follows: Where: T l (t) represents the power flow through the transmission channel l at time t; D r,l Indicates the flow direction of the transaction path r through the transmission channel l, 1 represents the positive direction and -1 represents the reverse direction; μ r,l represents the power flow distribution factor of the transaction path r through the transmission channel l without considering the network loss. For the DC transmission channel, the distribution factor is 1; η r,l R represents the network loss conversion coefficient of transaction path r through transmission channel l; l represents the set of all transaction paths r associated with the transmission channel l; r Represents R l The set of all seller entities related to the path; D r Represents R l The set of all buyer entities related to the path; The ATC constraints of AC and DC transmission channels are as follows: T l (t)≤ATC l (t) (4) Among them: ATC l (t) represents the maximum remaining available transmission capacity of the AC / DC transmission channel l at time t; The constraints on the seller's winning bid electricity are as follows: Where: Q o (t) represents the electricity supply declared by seller o at time t; The power constraints for the purchaser's winning bid are as follows: Where: Q d (t) represents the electricity purchase demand declared by the purchaser d at time t; E r represents the set of all transmission channels on the transaction path r; δ e (t) represents the line loss rate of each transmission channel; The line change rate constraint is as follows: |T l (t)-T l (t-1)|≤XT l,max (7) Among them: XT l,max Indicates the upper limit of the change value of the power flow in adjacent time periods of transmission channel l; T l (t-1) represents the power flow through the transmission channel l at time t-1.

5. The method for optimizing the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation according to claim 1 is characterized in that: The non-executable power penalties for green electricity taking into account the output distribution characteristics are as follows: Where: σ(t) is the green electricity main body O r The penalty coefficient, Green electricity main body r The winning bid for electricity, Unenforceable electricity expectations for green electricity; In order to take into account the output distribution characteristics of green electricity, electricity penalties cannot be imposed. Green Electricity Main Body r The winning bidder for power As shown below: in: It is the winning bid electricity of the green power unit located in the mth section; m is the section number of the winning bid electricity of the green power.

6. The method for optimizing the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation according to claim 1 is characterized in that: Consider the penalty for linearizing the unexecutable power expectation as follows: Where: m is the segment number of the green power winning bid; is the length of the rectangle of the green generator set located in the mth segment, is the winning bid power of the green power unit located in the mth segment; To consider the penalty of unenforceable power expectation for linearization; is the equivalent penalty coefficient of green electricity in the mth segment; σ(t) is the green electricity entity O r The penalty coefficient.

7. The method for optimizing the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation according to claim 1 is characterized in that: The objective function maxG1 of the efficient clearing model for medium- and long-term inter-provincial electricity transactions is as follows: in: To consider the penalty of linearized unexecutable power expectation; O represents the set of all seller entities; D represents the set of all buyers; R represents the set of all feasible paths, T represents the set of all time periods; X od,r (t) represents the winning bid power of each buyer-seller pair (o, d) at time t on the feasible transaction path r; B d (t), B o (t) respectively represent the declared prices of the buyer d and the seller o at time t; C d,r (t) represents the transmission fee and network loss discount that the buyer d needs to bear when purchasing electricity from the transaction path r at time t; Represents the cost of carbon emissions.

8. The method for optimizing the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation according to claim 1 is characterized in that: The constraints of the efficient clearing model for inter-provincial medium- and long-term electricity transactions are the same as those of the inter-provincial medium- and long-term electricity transactions model that takes into account the low-carbon value of green electricity, including AC and DC transmission channel power flow calculation constraints, AC and DC transmission channel ATC constraints, seller's winning bid power constraints, buyer's winning bid power constraints, and line change rate constraints.

9. The method for optimizing the inter-provincial power trading clearing problem considering the penalty for unexecutable power expectation according to claim 1 is characterized in that: The steps to solve the efficient clearing model of inter-provincial power medium- and long-term transactions include: 1) Decouple each time period, remove the constraints of time period coupling to obtain the decoupled model of the entire time period, and extract each single time period model separately, expressed as m s (t): 2) Solve each single-period model in parallel to extract transaction feature information within each period; 3) Based on the transaction characteristic information, the transaction probability of the purchase and sale power transactions is determined, and the purchase and sale power transactions with transaction probabilities less than the preset probability threshold are fixed to zero, and a clearing model m2 with a reduced feasible domain size is constructed; 4) Solve the clearing model m2 and obtain the clearing plan for medium- and long-term inter-provincial electricity transactions.

10. The method for optimizing the inter-provincial power trading clearing problem considering the non-executable power expectation penalty according to claim 9 is characterized in that: Transaction feature information includes the transaction pairs that were cleared within the time period and the paths they passed through.