Day-ahead electric power spot market clearing method based on quantum approximate optimization

Abstract

The invention relates to a day-ahead electric power spot market clearing method based on quantum approximate optimization, and the method comprises the steps: building a day-ahead electric power spot market clearing model, and abbreviating the day-ahead electric power spot market clearing model; the abbreviated day-ahead electric power spot market clearing model is decoupled into a discrete variable model and a continuous variable model; the discrete variable model is quantized through a quantum approximation optimization algorithm, the unit start-stop state and the bid winning electric quantity of the day-ahead electric power spot market are obtained through continuous iterative optimization, and the node electricity price is obtained through solving partial derivative of a Lagrange multiplier objective function constructed in the solving process; and carrying out security check on the current clearing result of the electric power spot market, and carrying out market settlement after the requirement is met. The real-time performance and convergence of electric power spot market transaction are improved, the workload of workers in the electric power market is reduced, and a supporting effect is achieved for guaranteeing safe, stable, economical and efficient operation of the electric power spot market.

Description

Technical Field

[0001] The present invention relates to the technical field of power market, and in particular to a day-ahead power spot market clearing method based on quantum approximate optimization. Background Art

[0002] The electricity spot market involves multiple variables and constraints, including electricity demand, supply, cost, market rules, etc. These factors constitute a large-scale complex problem. Traditional optimization algorithms often require a lot of calculations and iterations when dealing with such complex problems, which is time-consuming.

[0003] The optimization problem of the electricity spot market requires finding the global optimal solution, but traditional optimization algorithms are difficult to find the global optimal solution and are prone to falling into the local optimal solution. In addition, the rules and market conditions of the electricity spot market may change constantly, so the optimization algorithm needs to have strong robustness to adapt to the changes, but traditional optimization algorithms are highly sensitive to the initial solution and parameter settings, and the algorithm robustness is poor. Summary of the invention

[0004] In order to solve the problem of low solving efficiency of traditional algorithms when facing large-scale problems, the purpose of the present invention is to provide a day-ahead electricity spot market clearing method based on quantum approximate optimization, which has the advantage of fast convergence and ensures the efficiency of spot transactions.

[0005] To achieve the above object, the present invention adopts the following technical solution: a method for clearing the day-ahead electricity spot market based on quantum approximate optimization, the method comprising the following steps in order:

[0006] (1) Taking the day-ahead transaction declaration information and the load forecast curve provided by the power dispatching agency as the market optimization boundary conditions, a day-ahead power spot market clearing model is established, and the day-ahead power spot market clearing model is abbreviated to obtain the objective function and abbreviated constraint conditions of the abbreviated day-ahead power spot market clearing model. The abbreviated constraint conditions include three types: unit start-stop variable constraints, unit bid power constraints, and constraints including start-stop variables and bid power;

[0007] (2) According to the objective function of the abbreviated day-ahead electricity spot market clearing model and the abbreviated constraints, the abbreviated day-ahead electricity spot market clearing model is decoupled into a discrete variable model and a continuous variable model;

[0008] (3) The discrete variable model is quantized through a quantum approximate optimization algorithm, and the start and stop states of the units are optimized. The winning bid power of the units is optimized through a continuous variable model. The start and stop states and winning bid power of the units in the day-ahead power spot market are obtained through continuous iterative optimization. The node electricity price is obtained by taking the partial derivative of the Lagrange multiplier objective function constructed in the solution process.

[0009] (4) Conduct security verification on the day-ahead electricity spot market clearing results and conduct market settlement if they meet the requirements. The day-ahead electricity spot market clearing results include the start and stop status of the units, the amount of electricity won in the bid, and the node electricity price;

[0010] The step (1) specifically refers to: providing the power dispatching agency with the operation technical parameters, wherein the operation technical parameters include the maximum and minimum output of the unit and the ramp rate, and using the operation technical parameters as the parameters for clearing the power spot market; the power dispatching agency performs new energy and load forecasting, and establishes a day-ahead power spot market clearing model with the goal of minimizing the total system cost under the condition of meeting the day-ahead market load demand and reserve demand, and the formula of its objective function is as follows:

[0011]

[0012] Where F represents the total cost of power generation in the system, f(P i,t ) represents the fuel cost of unit i in period t, represents the no-load cost of unit i, represents the startup cost of unit i; T represents the number of scheduling cycle periods; N G Indicates the number of units; u i,t The variable representing the start and stop of unit i in time period t, with a value of 0 or 1, 1 for operation and 0 for shutdown; Indicates whether unit i switches from shutdown state to startup state in time period t. If yes, the value is 1, otherwise, the value is 0;

[0013] The constraints of the day-ahead electricity spot market clearing model include power balance constraints, system reserve constraints, upper and lower limits of unit output constraints, unit ramp-up constraints, unit ramp-down constraints, line power flow constraints, unit start-up times constraints, unit shutdown times constraints, unit minimum shutdown time constraints, and unit maximum shutdown time constraints.

[0014] The power balance constraint is:

[0015]

[0016] Where N j represents the number of load nodes, P i,t represents the winning bid power of unit i in time period t, D j,t represents the power demand of load node j in time period t;

[0017] The system reserve constraints are:

[0018]

[0019] Where P i,max represents the maximum output of unit i in time period t, and r represents the system reserve rate requirement;

[0020] The upper and lower limits of the unit output are:

[0021] u i,t P i,min ≤P i,t ≤u i,t P i,max

[0022] Where P i,min represents the minimum output of unit i in time period t;

[0023] The climbing constraint of the unit is:

[0024]

[0025] In the formula, Indicates the climbing ability of unit i;

[0026] The downhill climbing constraint of the unit is:

[0027]

[0028] In the formula, It represents the down-slope climbing capability of unit i;

[0029] The power flow constraint of the line is:

[0030] P l,min ≤P l,t ≤P l,max

[0031] Where P l,t , P l,max , P l,min They represent the power flow power, upper limit of power flow and lower limit of power flow of the transmission line respectively;

[0032] The startup times of the unit are constrained as follows:

[0033]

[0034] In the formula, Indicates the maximum number of times that unit i switches from shutdown to startup;

[0035] The shutdown times of the unit are constrained as follows:

[0036]

[0037] In the formula, Indicates whether unit i switches from the start state to the stop state in time period t. If yes, the value is 1, otherwise the value is 0; Indicates the maximum number of times that unit i switches from shutdown to startup;

[0038] The minimum downtime constraint of the unit is:

[0039]

[0040] In the formula, is the running time of unit i; is the minimum startup time limit of unit i;

[0041] The maximum downtime constraint of the unit is:

[0042]

[0043] In the formula, is the downtime of unit i; is the minimum downtime limit of unit i;

[0044] The day-ahead electricity spot market clearing model is simplified, and the objective function of the abbreviated day-ahead electricity spot market clearing model is obtained as follows:

[0045]

[0046] In the formula, is the fuel cost associated with the unit's winning bid for electricity, are the no-load costs and startup costs associated with the unit start-stop variables;

[0047] The constraints of the abbreviated day-ahead electricity spot market clearing model include three types: unit start-stop variable constraints, unit bid power constraints, and constraints that include start-stop variables and bid power.

[0048] The unit start and stop variable constraints:

[0049]

[0050] In the formula, V is a constant in the simplified constraints;

[0051] The winning bid power constraints of the unit:

[0052]

[0053] In the formula, J are all constants in the simplified constraints;

[0054] The constraints including the start / stop variables and the bid electricity are:

[0055]

[0056] In the formula, Y are all constants in the simplified constraints.

[0057] The step (2) specifically refers to: adding the unit start-stop variable constraint to the objective function of the abbreviated day-ahead power spot market clearing model, introducing a new continuous variable To replace u in the objective function of the abbreviated day-ahead electricity spot market clearing model i,t , and obtain the new objective function H and constraints:

[0058] The new objective function H is:

[0059] Constraints:

[0060]

[0061]

[0062]

[0063]

[0064] Where m is a positive number ranging from 1000 to 10000; is the fuel cost associated with the unit's winning bid for electricity, are the no-load costs and startup costs associated with the unit start-stop variables; V, J, and Y are all constants;

[0065] definition in, represents an indicator function, when hour, The value is 0, otherwise it is +∞;

[0066] At this time, the continuous variables are separated from the discrete variables, and the new objective function H and its constraints are abbreviated as the new objective function G and related constraints:

[0067] The new objective function G is:

[0068] Constraints:

[0069] Among them, A0 and A1 are constant coefficients, f0(u i,t ) is a discrete variable model, The model is a continuous variable.

[0070] The step (3) specifically refers to: firstly, the three parameters Initialized to u0, P0, λ0 respectively;

[0071] Then, the quantum approximate optimization algorithm is used to quantize and solve the following objective function G2 to optimize the unit start and stop variables for the Kth iteration:

[0072]

[0073] Where K represents the number of iterations; the penalty factor Q is a positive number ranging from 1000 to 10000; represents the multiplier in the k-1th iteration;

[0074] The discrete variable model is quantized to solve the unit start and stop variables of the Kth iteration The specific steps are as follows:

[0075] First, use the Pauli operator σ z Convert the objective function G2 into a quantum Ising model and determine N according to the number of units. G qubits, and then the uniformly distributed quantum superposition state |ψ0> of the quantum gate H is applied to each qubit:

[0076]

[0077] Where z is an n-bit binary code, which is the 2^ NG possible results, 0 represents shutdown, 1 represents startup;

[0078] Then set the initial Hamiltonian matrix H B and the target Hamiltonian matrix H based on the quantum Ising model C , and write the corresponding unitary transformation based on the matrix:

[0079]

[0080]

[0081] Where U B , U C Respectively represent the corresponding H B , H C Unitary transformation of ; β, γ are the parameters with ;

[0082] Further, the corresponding parameter-containing quantum gate matrix is ​​derived based on the unitary transformation, and the corresponding parameter-containing quantum gate is built on the quantum circuit to obtain the final quantum state |ψ(β,γ)>:

[0083]

[0084] Where d represents the depth of the quantum circuit;

[0085] Then H CThe expected minimum is the goal:

[0086] minE(β,γ)=<ψ(β,γ)|H C |ψ(β,γ)>

[0087] Where E(β,γ) represents H C expectations;

[0088] The optimal solution β is obtained by variational quantum circuit optimization * and γ * ; Finally, the optimal solution β * and γ * Substitute the parameter-containing quantum gate matrix to make it a constant matrix, and sample and measure the quantum state |ψ(β,γ)>:

[0089] Finally, the probability distribution result of the set of unit start-stop solutions is obtained, and the one with the highest probability is the unit start-stop variable of the Kth iteration. The optimal solution of

[0090] The continuous variable model uses a classical computer to update the vector containing the unit's winning bid power λ t,K :

[0091]

[0092]

[0093] Until K>Kmax or The iteration stops and the unit start and stop variables u in the day-ahead electricity spot market are obtained. i,t 、Winning bid quantity P i,t ; Kmax is the maximum number of iterations;

[0094] The Lagrangian objective function L for minimizing the system operation cost constructed during the solution process is as follows:

[0095]

[0096] Finally, the Lagrangian objective function L is used to calculate the power demand D j,t , find the partial derivative to get LMP i,t :

[0097]

[0098] Where, LMP i,t The node electricity price is the electricity price of node i in time period t. The node electricity price refers to the minimum additional power generation cost required by the system when the node adds a new unit of active power demand.

[0099] It can be seen from the above technical scheme that the beneficial effects of the present invention are: first, the present invention embeds a quantum approximate optimization algorithm in a large-scale day-ahead electricity spot market clearing model, and utilizes the advantage of fast convergence of quantum computing in solving complex problems to ensure the high efficiency of spot transactions; second, the present invention has rapid adaptability and can adjust the solution strategy in a short time to adapt to the rapidly changing supply and demand situation and prices in the electricity spot market; third, the present invention improves the real-time and convergence of electricity spot market transactions, reduces the workload of electricity market staff, and plays a supporting role in ensuring the safe, stable, economical and efficient operation of the electricity spot market. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] Figure 1 is a flow chart of the method of the present invention;

[0101] Figure 2 A flow chart of the method for solving the day-ahead electricity spot market clearing results;

[0102] Figure 3 The flowchart is a method for solving the start-stop variables of the unit for the Kth iteration by quantizing the discrete variable model. DETAILED DESCRIPTION

[0103] like Figure 1 As shown, a method for clearing the day-ahead electricity spot market based on quantum approximate optimization comprises the following steps in order:

[0104] (1) Taking the day-ahead transaction declaration information and the load forecast curve provided by the power dispatching agency as the market optimization boundary conditions, a day-ahead power spot market clearing model is established, and the day-ahead power spot market clearing model is abbreviated to obtain the objective function and abbreviated constraint conditions of the abbreviated day-ahead power spot market clearing model. The abbreviated constraint conditions include three types: unit start-stop variable constraints, unit bid power constraints, and constraints including start-stop variables and bid power;

[0105] (2) According to the objective function of the abbreviated day-ahead electricity spot market clearing model and the abbreviated constraints, the abbreviated day-ahead electricity spot market clearing model is decoupled into a discrete variable model and a continuous variable model;

[0106] (3) The discrete variable model is quantized through a quantum approximate optimization algorithm, and the start and stop states of the units are optimized. The winning bid power of the units is optimized through a continuous variable model. The start and stop states and winning bid power of the units in the day-ahead power spot market are obtained through continuous iterative optimization. The node electricity price is obtained by taking the partial derivative of the Lagrange multiplier objective function constructed in the solution process.

[0107] (4) Conduct security verification on the day-ahead electricity spot market clearing results and conduct market settlement if they meet the requirements. The day-ahead electricity spot market clearing results include the start and stop status of the units, the amount of electricity won in the bid, and the node electricity price;

[0108] The step (1) specifically refers to: providing the power dispatching agency with the operation technical parameters, wherein the operation technical parameters include the maximum and minimum output of the unit and the ramp rate, and using the operation technical parameters as the parameters for clearing the power spot market; the power dispatching agency performs new energy and load forecasting, and establishes a day-ahead power spot market clearing model with the goal of minimizing the total system cost under the condition of meeting the day-ahead market load demand and reserve demand, and the formula of its objective function is as follows:

[0109]

[0110] Where F represents the total cost of power generation in the system, f(P i,t ) represents the fuel cost of unit i in period t, represents the no-load cost of unit i, represents the startup cost of unit i; T represents the number of scheduling cycle periods; N G Indicates the number of units; u i,t The variable representing the start and stop of unit i in time period t, with a value of 0 or 1, 1 for operation and 0 for shutdown; Indicates whether unit i switches from shutdown state to startup state in time period t. If yes, the value is 1, otherwise, the value is 0;

[0111] The constraints of the day-ahead electricity spot market clearing model include power balance constraints, system reserve constraints, upper and lower limits of unit output constraints, unit ramp-up constraints, unit ramp-down constraints, line power flow constraints, unit start-up times constraints, unit shutdown times constraints, unit minimum shutdown time constraints, and unit maximum shutdown time constraints.

[0112] The power balance constraint is:

[0113]

[0114] Where N j represents the number of load nodes, P i,t represents the winning bid power of unit i in time period t, D j,t represents the power demand of load node j in time period t;

[0115] The system reserve constraints are:

[0116]

[0117] Where P i,max represents the maximum output of unit i in time period t, and r represents the system reserve rate requirement;

[0118] The upper and lower limits of the unit output are:

[0119] u i,t P i,min ≤P i,t ≤u i,t P i,max

[0120] Where P i,min represents the minimum output of unit i in time period t;

[0121] The climbing constraint of the unit is:

[0122]

[0123] In the formula, Indicates the climbing ability of unit i;

[0124] The downhill climbing constraint of the unit is:

[0125]

[0126] In the formula, It represents the down-slope climbing capability of unit i;

[0127] The power flow constraint of the line is:

[0128] P l,min ≤P l,t ≤P l,max

[0129] Where P l,t , P l,max , P l,min They represent the power flow power, upper limit of power flow and lower limit of power flow of the transmission line respectively;

[0130] The startup times of the unit are constrained as follows:

[0131]

[0132] In the formula, Indicates the maximum number of times that unit i switches from shutdown to startup;

[0133] The shutdown times of the unit are constrained as follows:

[0134]

[0135] In the formula, Indicates whether unit i switches from the start state to the stop state in time period t. If yes, the value is 1, otherwise the value is 0; Indicates the maximum number of times that unit i switches from shutdown to startup;

[0136] The minimum downtime constraint of the unit is:

[0137]

[0138] In the formula, is the running time of unit i; is the minimum startup time limit of unit i;

[0139] The maximum downtime constraint of the unit is:

[0140]

[0141] In the formula, is the downtime of unit i; is the minimum downtime limit of unit i;

[0142] The day-ahead electricity spot market clearing model is simplified, and the objective function of the abbreviated day-ahead electricity spot market clearing model is obtained as follows:

[0143]

[0144] In the formula, is the fuel cost associated with the unit's winning bid for electricity, are the no-load costs and startup costs associated with the unit start-stop variables;

[0145] The constraints of the abbreviated day-ahead electricity spot market clearing model include three types: unit start-stop variable constraints, unit bid power constraints, and constraints that include start-stop variables and bid power.

[0146] The unit start and stop variable constraints:

[0147]

[0148] In the formula, V is a constant in the simplified constraints;

[0149] The winning bid power constraints of the unit:

[0150]

[0151] In the formula, J are all constants in the simplified constraints;

[0152] The constraints including the start / stop variables and the bid electricity are:

[0153]

[0154] In the formula, Y are all constants in the simplified constraints.

[0155] The step (2) specifically refers to: adding the unit start-stop variable constraint to the objective function of the abbreviated day-ahead power spot market clearing model, introducing a new continuous variable To replace u in the objective function of the abbreviated day-ahead electricity spot market clearing model i,t , and obtain the new objective function H and constraints:

[0156] The new objective function H is:

[0157] Constraints:

[0158]

[0159]

[0160]

[0161]

[0162] Where m is a positive number ranging from 1000 to 10000; is the fuel cost associated with the unit's winning bid for electricity, are the no-load costs and startup costs associated with the unit start-stop variables; V, J, and Y are all constants;

[0163] definition in, represents an indicator function, when hour, The value is 0, otherwise it is +∞;

[0164] At this time, the continuous variables are separated from the discrete variables, and the new objective function H and its constraints are abbreviated as the new objective function G and related constraints:

[0165] The new objective function G is:

[0166] Constraints:

[0167] Among them, A0 and A1 are constant coefficients, f0(u i,t ) is a discrete variable model, The model is a continuous variable.

[0168] like Figure 2 As shown, the step (3) specifically refers to: firstly, the three parameters Initialized to u0, P0, λ0 respectively;

[0169] Then, the quantum approximate optimization algorithm is used to quantize and solve the following objective function G2 to optimize the unit start and stop variables for the Kth iteration:

[0170]

[0171] Where K represents the number of iterations; the penalty factor Q is a positive number ranging from 1000 to 10000; represents the multiplier in the k-1th iteration;

[0172] like Figure 3 As shown, the discrete variable model is quantized to solve the unit start and stop variables of the Kth iteration The specific steps are as follows:

[0173] First, use the Pauli operator σ z Convert the objective function G2 into a quantum Ising model and determine N according to the number of units. G qubits, and then the uniformly distributed quantum superposition state |ψ0> of the quantum gate H is applied to each qubit:

[0174]

[0175] Where z is an n-bit binary code, which is the 2^ NG possible results, 0 represents shutdown, 1 represents startup;

[0176] Then set the initial Hamiltonian matrix H B and the target Hamiltonian matrix H based on the quantum Ising model C , and write the corresponding unitary transformation based on the matrix:

[0177]

[0178]

[0179] Where U B , U C Respectively represent the corresponding H B , H C Unitary transformation of ; β, γ are the parameters with ;

[0180] Further, the corresponding parameter-containing quantum gate matrix is ​​derived based on the unitary transformation, and the corresponding parameter-containing quantum gate is built on the quantum circuit to obtain the final quantum state |ψ(β,γ)>:

[0181]

[0182] Where d represents the depth of the quantum circuit;

[0183] Then H C The expected minimum is the goal:

[0184] minE(β,γ)=<ψ(β,γ)|H C |ψ(β,γ)>

[0185] Where E(β,γ) represents H C expectations;

[0186] The optimal solution β is obtained by variational quantum circuit optimization * and γ * ; Finally, the optimal solution β * and γ * Substitute the parameter-containing quantum gate matrix to make it a constant matrix, and sample and measure the quantum state |ψ(β,γ)>:

[0187] Finally, the probability distribution result of the set of unit start-stop solutions is obtained, and the one with the highest probability is the unit start-stop variable of the Kth iteration. The optimal solution of

[0188] The continuous variable model uses a classical computer to update the vector containing the unit's winning bid power λ t,K :

[0189]

[0190]

[0191] Until K>Kmax or The iteration stops and the unit start and stop variables u in the day-ahead electricity spot market are obtained. i,t 、Winning bid quantity P i,t ; Kmax is the maximum number of iterations; the classical computer is an ordinary computer different from a quantum computer.

[0192] The Lagrangian objective function L for minimizing the system operation cost constructed during the solution process is as follows:

[0193]

[0194] Finally, the Lagrangian objective function L is used to calculate the power demand D j,t , find the partial derivative to get LMP i,t :

[0195]

[0196] Where, LMP i,t The node electricity price is the electricity price of node i in time period t. The node electricity price refers to the minimum additional power generation cost required by the system when the node adds a new unit of active power demand.

[0197] In summary, the present invention embeds a quantum approximate optimization algorithm in a large-scale day-ahead electricity spot market clearing model, and utilizes the advantage of fast convergence of quantum computing in solving complex problems to ensure the high efficiency of spot transactions; the present invention has rapid adaptability and can adjust the solution strategy in a short time to adapt to the rapidly changing supply and demand conditions and prices in the electricity spot market; the present invention improves the real-time and convergence of electricity spot market transactions, reduces the workload of electricity market staff, and plays a supporting role in ensuring the safe, stable, economical and efficient operation of the electricity spot market.

Claims

1. A method for clearing the day-ahead electricity spot market based on quantum approximate optimization, characterized by: The method comprises the following steps in order: (1) Taking the day-ahead transaction declaration information and the load forecast curve provided by the power dispatching agency as the market optimization boundary conditions, a day-ahead power spot market clearing model is established, and the day-ahead power spot market clearing model is abbreviated to obtain the objective function and abbreviated constraint conditions of the abbreviated day-ahead power spot market clearing model. The abbreviated constraint conditions include three types: unit start-stop variable constraints, unit bid power constraints, and constraints including start-stop variables and bid power; (2) According to the objective function of the abbreviated day-ahead electricity spot market clearing model and the abbreviated constraints, the abbreviated day-ahead electricity spot market clearing model is decoupled into a discrete variable model and a continuous variable model; (3) The discrete variable model is quantized through a quantum approximate optimization algorithm, and the start and stop states of the units are optimized. The winning bid power of the units is optimized through a continuous variable model. The start and stop states and winning bid power of the units in the day-ahead power spot market are obtained through continuous iterative optimization. The node electricity price is obtained by taking the partial derivative of the Lagrange multiplier objective function constructed in the solution process. (4) Conducting safety verification on the day-ahead electricity spot market clearing results and conducting market settlement after meeting the requirements. The day-ahead electricity spot market clearing results include the start / stop status of the unit, the winning bid power, and the node electricity price.

2. The method for clearing the day-ahead electricity spot market based on quantum approximate optimization according to claim 1 is characterized in that: The step (1) specifically refers to: providing the power dispatching agency with the operation technical parameters, wherein the operation technical parameters include the maximum and minimum output of the unit and the ramp rate, and using the operation technical parameters as the parameters for clearing the power spot market; The power dispatching agency conducts new energy and load forecasting, and establishes a day-ahead power spot market clearing model with the goal of minimizing the total system cost under the condition of meeting the day-ahead market load demand and reserve demand. The formula of its objective function is as follows: Where F represents the total cost of power generation in the system, f(P i,t ) represents the fuel cost of unit i in period t, represents the no-load cost of unit i, represents the startup cost of unit i; T represents the number of scheduling cycle periods; N G Indicates the number of units; u i,t The variable representing the start and stop of unit i in time period t, with a value of 0 or 1, 1 for operation and 0 for shutdown; Indicates whether unit i switches from shutdown state to startup state in time period t. If yes, the value is 1, otherwise, the value is 0; The constraints of the day-ahead electricity spot market clearing model include power balance constraints, system reserve constraints, upper and lower limits of unit output constraints, unit ramp-up constraints, unit ramp-down constraints, line power flow constraints, unit start-up times constraints, unit shutdown times constraints, unit minimum shutdown time constraints, and unit maximum shutdown time constraints. The power balance constraint is: Where N j represents the number of load nodes, P i,t represents the winning bid power of unit i in time period t, D j,t represents the power demand of load node j in time period t; The system reserve constraints are: Where P i,max represents the maximum output of unit i in time period t, and r represents the system reserve rate requirement; The upper and lower limits of the unit output are: u i,t Q i,min ≤P i,t ≤u i,t Q i,max Where P i,min represents the minimum output of unit i in time period t; The climbing constraint of the unit is: In the formula, Indicates the climbing ability of unit i; The downhill climbing constraint of the unit is: In the formula, It represents the down-slope climbing capability of unit i; The power flow constraint of the line is: P l,min ≤P l,t ≤P l,max Where P l,t , P l,max , P l,min They represent the power flow power, upper limit of power flow and lower limit of power flow of the transmission line respectively; The startup times of the unit are constrained as follows: In the formula, Indicates the maximum number of times that unit i switches from shutdown to startup; The shutdown times of the unit are constrained as follows: In the formula, Indicates whether unit i switches from the start state to the stop state in time period t. If yes, the value is 1, otherwise the value is 0; Indicates the maximum number of times that unit i switches from shutdown to startup; The minimum downtime constraint of the unit is: In the formula, is the running time of unit i; T i on is the minimum startup time limit of unit i; The maximum downtime constraint of the unit is: In the formula, is the downtime of unit i; T i off is the minimum downtime limit of unit i; The day-ahead electricity spot market clearing model is simplified, and the objective function of the abbreviated day-ahead electricity spot market clearing model is obtained as follows: In the formula, is the fuel cost associated with the unit's winning bid for electricity, are the no-load costs and startup costs associated with the unit start-stop variables; The constraints of the abbreviated day-ahead electricity spot market clearing model include three types: unit start-stop variable constraints, unit bid power constraints, and constraints that include start-stop variables and bid power. The unit start and stop variable constraints: In the formula, V is a constant in the simplified constraints; The winning bid power constraints of the unit: In the formula, J are all constants in the simplified constraints; The constraints including the start / stop variables and the bid electricity are: In the formula, Y are all constants in the simplified constraints.

3. The method for clearing the day-ahead electricity spot market based on quantum approximate optimization according to claim 1 is characterized in that: The step (2) specifically refers to: adding the unit start-stop variable constraint to the objective function of the abbreviated day-ahead power spot market clearing model, introducing a new continuous variable To replace u in the objective function of the abbreviated day-ahead electricity spot market clearing model i,t , and obtain the new objective function H and constraints: The new objective function H is: Constraints: Where m is a positive number ranging from 1000 to 10000; is the fuel cost associated with the unit's winning bid for electricity, are the no-load costs and startup costs associated with the unit start-stop variables; V, J, and Y are all constants; definition in, represents an indicator function, when hour, The value is 0, otherwise it is +∞; At this time, the continuous variables are separated from the discrete variables, and the new objective function H and its constraints are abbreviated as the new objective function G and related constraints: The new objective function G is: Constraints: Among them, A0 and A1 are constant coefficients, f0(u i,t ) is a discrete variable model, The model is a continuous variable.

4. The method for clearing the day-ahead electricity spot market based on quantum approximate optimization according to claim 1 is characterized in that: The step (3) specifically refers to: firstly, the three parameters Initialized to u0, P0, λ0 respectively; Then, the quantum approximate optimization algorithm is used to quantize and solve the following objective function G2 to optimize the unit start and stop variables for the Kth iteration: G2： In the formula, K represents the number of iterations; The penalty factor Q is a positive number ranging from 1000 to 10000; represents the multiplier in the k-1th iteration; The discrete variable model is quantized to solve the unit start and stop variables of the Kth iteration The specific steps are as follows: First, use the Pauli operator σ z Convert the objective function G2 into a quantum Ising model and determine N according to the number of units. G qubits, and then the uniformly distributed quantum superposition state |ψ0> of the quantum gate H is applied to each qubit: Where z is an n-bit binary code, which is the 2^ NG possible results, 0 represents shutdown, 1 represents startup; Then set the initial Hamiltonian matrix H B and the target Hamiltonian matrix H based on the quantum Ising model C , and write the corresponding unitary transformation based on the matrix: Where U B , U C Respectively represent the corresponding H B , H C Unitary transformation of ; β, γ are the parameters with ; Further, the corresponding parameter-containing quantum gate matrix is ​​derived based on the unitary transformation, and the corresponding parameter-containing quantum gate is built on the quantum circuit to obtain the final quantum state |ψ(β,γ)>: Where d represents the depth of the quantum circuit; Then H C The expected minimum is the goal: min E(β,γ)=<ψ(β,γ)|H C |ψ(b,c)> Where E(β,γ) represents H C expectations; The optimal solution β is obtained by variational quantum circuit optimization * and γ * ; Finally, the optimal solution β * and γ * Substitute the parameter-containing quantum gate matrix to transform it into a constant matrix, and sample and measure the quantum state |ψ(β,γ)>; Finally, the probability distribution result of the set of unit start-stop solutions is obtained, and the one with the highest probability is the unit start-stop variable of the Kth iteration. The optimal solution of The continuous variable model uses a classical computer to update the vector containing the unit's winning bid power λ t,K : Until K>Kmax or The iteration stops and the unit start and stop variables u in the day-ahead electricity spot market are obtained. i,t 、Winning bid quantity P i,t ; Kmax is the maximum number of iterations; The Lagrangian objective function L for minimizing the system operation cost constructed during the solution process is as follows: Finally, the Lagrangian objective function L is used to calculate the power demand D j,t , find the partial derivative to get LMP i,t : Where, LMP i,t The node electricity price is the electricity price of node i in time period t. The node electricity price refers to the minimum additional power generation cost required by the system when the node adds a new unit of active power demand.

Images