Coal power and new energy collaborative planning method and system for power system

Abstract

The invention discloses a coal power and new energy collaborative planning method and system for an electric power system, and the method comprises the steps: determining the dynamic frequency safety constraint and voltage stability constraint of the electric power system; constructing a wind-light combined output scene model based on the four-season labels and the adversarial network, and generating a wind-light load scene in spring, summer, autumn and winter and an extreme wind-light load scene; and solving the coal power and new energy collaborative planning configuration capacity of the power system in different wind-light-load scenes. The method not only considers the traditional capacity investment and operation cost, but also comprehensively optimizes the objective function and combines the complementarity of wind energy and solar energy, the LOLP (Load Load Probability), the maximization of the total revenue of the system and other multi-dimensional indexes, and multi-dimensional optimization can balance the consumption and investment cost of renewable energy sources, improve the collaborative benefit of new energy and coal power, and improve the economic benefit of the system. And the comprehensive parameters of the system are optimized.

Description

Technical Field

[0001] The present invention relates to the field of power system energy collaborative planning, and in particular to a method and system for collaborative planning of coal-fired power and new energy for power systems. Background Art

[0002] The power transmission planning of the Shagohuang New Energy Base must be based on the joint layout of wind power, photovoltaic power, thermal power and energy storage systems to ensure the stability, economy and reliability of the system. In areas with abundant wind and solar resources but large fluctuations, such as Shagohuang, the joint planning of wind, photovoltaic, thermal and energy storage is the key to improving the quality of power output. The instability of wind and solar resources may lead to the situation that the power generation far exceeds the load demand during peak periods, and there may be power shortages during low periods. Summary of the invention

[0003] Purpose of the invention: The present invention aims to provide a collaborative planning method for achieving a balance between coal-fired power and new energy sources such as wind, solar, thermal and storage for power systems; another purpose of the present invention is to provide a collaborative planning system for coal-fired power and new energy sources for power systems.

[0004] Technical solution: The method for coordinated planning of coal-fired power and new energy for power systems of the present invention comprises the following steps:

[0005] (1) Determine the dynamic frequency security constraints and voltage stability constraints of the power system;

[0006] (2) Based on the four seasons labels and adversarial network, a wind-solar joint output scenario model is constructed to generate spring, summer, autumn and winter wind-solar load scenarios and extreme wind-load scenarios;

[0007] (3) Solve the coordinated planning and configuration capacity of coal-fired power and renewable energy in the power system under different wind and solar load scenarios.

[0008] Furthermore, dynamic frequency safety constraints include frequency change rate constraints and quasi-steady-state frequency constraints;

[0009] The frequency change rate constraint is

[0010]

[0011] in, is the equivalent inertia of the system at time t; and They are the state variables of coal-fired power units, wind power units, photovoltaic units and energy storage at time t respectively; is the equivalent inertia of coal power, wind power, photovoltaic power and energy storage; W i g , P i w , P i pv 、E mThey are the capacity of coal-fired power units, the output of wind farms, the output of photovoltaic units and the energy storage capacity; X Rocof is the frequency change rate constraint; ΔP 0 is the power disturbance corresponding to the most serious accident; f 0 is the initial frequency N g 、N w 、N w 、N BES They are the number of coal-fired power, wind power, photovoltaic power and energy storage respectively;

[0012] The quasi-steady-state frequency constraint is

[0013]

[0014] Where Δf ss is the maximum allowable deviation under quasi-steady state; P t L is the system load; It is the frequency modulation power that can be released by conventional units in quasi-steady state; is the frequency modulation power that can be released by the energy storage in the quasi-steady state; D is the synchronous power coefficient.

[0015] Furthermore, the voltage stability constraint includes the static voltage stability margin γ c for

[0016] γ c =(P cr,i -P L0,i ) / b i

[0017] P G +ΔP G (γ c )-P L -γ c b=f(x)

[0018] Among them, P cr,i is the load size of the current operating point of node i; P L0,i is the load size at the voltage collapse point of node i; P cr,i -P L0,i is the absolute load margin; b i is the load growth direction of node i; P G Inject power into the generator base state; P L is the node load base power; b is the load growth mode; ΔP G (γ c ) is the generator dispatch mode; x is the system static state vector, x = [θ T ,V T ] T, θ is the node voltage phase angle vector, V is the node voltage amplitude vector; f(x) is the node injection power function.

[0019] Furthermore, the wind-solar joint output scenario model introduces GP, and the value function Value(G,D) of LGAN-GP is obtained as

[0020]

[0021] Among them, y represents the real sample data, p y It represents the probability distribution followed by these data samples. The calculation is based on the real sample yp y The average expected value of the distribution, D(y|h) is the output score of the discriminator D for the input y under the given condition h. is the noise distribution zp from a specific z The average expectation of the generated data samples. G(Z|h) is the generator G converting random noise z into generated samples under condition h. , λ is the gradient penalty function, It is the second norm of the gradient of the discriminator D(y) with respect to the input y to control the smoothness of the gradient and prevent the gradient from exploding or disappearing.

[0022] Furthermore, the objective function minF of the coordinated planning and configuration capacity of coal-fired power and new energy in the power system is

[0023] minF=∑(C total ,f(η a ,η b )) min

[0024] f(η a ,η b )=α 1 η a +α 2 η b

[0025] Among them, Ctotal is the total system cost, η a is the wind-solar energy complementarity ratio, η b is the LOLP value of the hybrid power generation system. 1 , α 2 is the weight.

[0026] Furthermore, the minimum total system cost is

[0027] minC total =-(C PA ·W pv +C WA ·W w +C EA ·Wbes +C CA ·W g +C s +C G ·W g )

[0028] Among them, C PA , C WA , C EA , C CA , C G are the unit capacity costs of photovoltaic, wind power, energy storage and coal-fired power generation units respectively; W pv , W w , W bes , W g are the capacities of photovoltaic, wind power, energy storage, and coal-fired power units; C s For power generation income.

[0029] Furthermore, the life cycle cost of photovoltaic power generation per unit capacity C PA for

[0030]

[0031] In the formula, CI v represents the investment cost, CO is the operating cost, CM is the maintenance cost, CF represents the failure cost, CD v represents the processing cost; Cl i represents the cost of the i-th item in the investment cost, m represents the number of items included in the investment cost; r represents the loan rate, R represents the annual interest rate, λ represents the discount rate; CM 1 Represents the repair and maintenance cost of the photovoltaic unit, CM 2 represents the maintenance and renewal cost of photovoltaic facilities, CF 1 represents the equipment failure cost; C d represents the cost of disposing of obsolete equipment, and T represents the residual value of fixed assets;

[0032] Wind power life cycle cost per unit capacity C WA for

[0033]

[0034] Where CI is the initial investment cost, CI 1 Design costs for research planning, CI 2 is the purchase and installation cost of wind power equipment, Cl 3 is the land purchase cost, CI 4 Operation and management costs; CR t represents operating cost, CR 1t represents the daily management cost, CR 2trepresents daily operating costs; CS t represents the cost associated with the maintenance of updated facilities, L represents the cost rate of facility maintenance and update; S is the installation cost of the configuration facilities, CT t Costs incurred due to maintenance losses; X t is the percentage of failure cost in the tth year to the initial investment cost; CD represents the residual value gain, and f represents the residual value rate;

[0035] The life cycle cost of energy storage equipment per unit capacity C EA for

[0036]

[0037] In the formula, C sys represents the installation cost of the energy storage system, C rep represents the cost and time required for energy storage system update, C fom represents the fixed annual maintenance cost of the energy storage system, C vom Represents the annual change in the maintenance cost of the energy storage system, C rec Represents its recoverable residual value; among them, C bat is the energy storage battery cost, C pcs is the cost of energy conversion equipment, C bop is the fixed configuration equipment cost; α represents the annual reduction percentage of installation cost, k represents the number of battery replacements, and n represents the number of years of battery storage life cycle; C f-p represents the operation and maintenance cost per unit power, P is the rated power of the energy storage system; C c is the charging cost, t is the rated discharge time, ψ is the system fixed efficiency, D is the number of operating days; y represents the battery recovery ratio;

[0038] Electricity sales cost C A for

[0039]

[0040] In the formula, θ d The price of electricity sold to the system;

[0041] The life cycle cost of coal-fired power units C G

[0042]

[0043] In the formula, w and k represent the system discount rate and the expected operating life of the equipment, respectively, g is the unit installed capacity investment cost of coal-fired power units, W g The installed capacity of coal-fired power plants.

[0044] Furthermore, minC para =min(α1 η a +α 2 η b )

[0045] In the formula, α 1 , α 2 is the weight, η a is the wind-solar energy complementarity ratio, η b is the LOLP value of the hybrid power generation system;

[0046] And the following constraints are met:

[0047] Power balance constraints:

[0048] P t w +P t pv +P t g +P t bes =P t d

[0049] Where P t w , P bes , P t d They are the total wind power output, photovoltaic output, coal-fired power unit output, energy storage equipment output, and external power at time t;

[0050] Energy storage / coal-fired power installed capacity constraints:

[0051]

[0052] In the formula, The upper limit of installed capacity of energy storage-loaded coal-fired power units;

[0053] Wind power / photovoltaic / energy storage / coal-fired power generator output power constraints:

[0054]

[0055] In the formula, f wt 、f pν 、f bs 、f bs The derating factor of the power generation system taking into account various factors;

[0056] Energy storage power station operation status constraints:

[0057] Soc min ≤Soc t ≤Soc max

[0058] In the formula, Soc max 、Soc min The upper and lower limits of the battery storage state of the energy storage power station;

[0059] Parameter constraints:

[0060]

[0061] Channel transmission power range constraints:

[0062]

[0063] In the formula, and They represent the lower and upper limits of the external power transmission at time t respectively;

[0064] Transmission capacity constraints of the outbound transmission channel:

[0065]

[0066] In the formula, χ represents the minimum utilization hours of the delivery channel throughout the year, which is a known quantity.

[0067] Outbound channel adjustment rate constraints:

[0068]

[0069] In the formula, is a 0-1 variable, indicating that the channel is upregulated and downregulated at time t, and is not 1 at the same time. and They represent the lower and upper limits of the forward adjustment of the external power, and They respectively represent the lower limit and upper limit of the reverse adjustment of the external power, and both are known quantities.

[0070] Furthermore, the wind-solar complementarity rate η a for

[0071]

[0072] In the formula, is the average value of the transmitted power, and N is the time period;

[0073] Hybrid power generation system LOLP value η b for

[0074]

[0075] The coal-fired power and new energy collaborative planning system for power systems of the present invention comprises:

[0076] System constraint module, used to perform dynamic frequency security constraints and voltage stability constraints on the power system;

[0077] The model building module builds a wind-solar joint output scenario model based on four seasons labels and adversarial networks, generating spring, summer, autumn, winter, wind-solar-load scenarios and extreme wind-load scenarios;

[0078] The solution module is used to solve the coordinated planning and configuration capacity of coal-fired power and new energy in the power system under different wind and solar load scenarios.

[0079] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: 1. The present invention not only takes into account the traditional capacity investment and operating costs, but also combines the complementarity of wind energy and solar energy, load loss probability LOLP, and system total revenue maximization and other indicators through comprehensive optimization of the objective function. Multi-dimensional optimization can balance the consumption and investment costs of renewable energy, enhance the synergistic benefits of new energy and coal-fired power, and optimize the comprehensive parameters of the system; 2. The present invention introduces flexible scheduling of energy storage and coal-fired power units, and ensures that the system can maintain stable power supply in uncertain load fluctuations by setting different installed capacities and operating constraints. In addition, multiple constraints such as the power range of the transmission channel, the minimum utilization hours, and the regulation rate are also considered to ensure the flexibility and stability of the system. DETAILED DESCRIPTION

[0080] The present invention will be further described below.

[0081] The method for coordinated planning of coal-fired power and new energy for power systems of the present invention comprises the following steps:

[0082] (1) Determine the dynamic frequency security constraints and voltage stability constraints of the power system.

[0083] Dynamic stable and safe operation domain

[0084] The access of electricity to the power system needs to ensure the stability of the system and meet the following constraints:

[0085] (11) Dynamic frequency safety constraints

[0086] (111) Frequency Change Rate Constraint

[0087] At the moment when the system is disturbed, the degree of frequency reduction is only related to the system inertia level. The system inertia in each period is related to the operating status, as shown below:

[0088]

[0089] Where: and are the state variables of thermal power generation unit, wind power generation unit and energy storage at time t; X RocofThe limit for the frequency change rate constraint; is the equivalent inertia of the system at time t; ΔP 0 is the power disturbance corresponding to the most serious accident; f 0 is the initial frequency; W i g , P i w , P i v 、E m are the capacity of coal-fired power units, the output of wind farms and the energy storage capacity; N g 、N w 、N w 、N BES They are the quantities of coal-fired power, wind power, photovoltaic power and energy storage respectively. It is the equivalent inertia of coal power, wind power, photovoltaic power and energy storage.

[0090] (112) Quasi-steady-state frequency constraint

[0091] After the transient process of the system ends, the quasi-steady-state frequency of the system must also be within a certain range, that is, satisfying:

[0092]

[0093] Where: Δf ss is the maximum allowable deviation under quasi-steady state; P t L is the system load; It is the frequency modulation power that can be released by conventional units in quasi-steady state; is the frequency modulation power that can be released by the energy storage in the quasi-steady state; D is the synchronous power coefficient.

[0094] (12) Voltage stability constraints

[0095] (121) Static voltage stability margin

[0096] The voltage stability margin refers to the load distance between the current operating point and the collapse point under the set load growth and generator output direction, which can be expressed as

[0097] γ c =(P cr,i -P L0,i ) / b i (4)

[0098] Where: b i is the load growth direction of node i; P cr,i is the load size of the current operating point of node i; P L0,i is the load size at the voltage collapse point of node i; γ c is a relative load margin indicator, which is regarded as the voltage stability margin in this paper; (Pcr,i -P L0,i ) is the absolute load margin.

[0099] The voltage stability margin calculation model can be expressed as obtaining the maximum γ value of equation (5), that is,

[0100] P G +ΔP G (γ)-P L -γ c b=f(x) (5)

[0101] Where: P G Inject power into the generator base state; P L is the node load base power; b is the load growth mode; ΔP G (γ) is the generator dispatch mode; x is the system static state vector, x = [θ T ,V T ] T , θ is the node voltage phase angle vector, V is the node voltage amplitude vector; f(x) is the node injection power function.

[0102] (2) A wind-solar joint output scenario model is constructed based on the four seasons labels and adversarial network to generate spring, summer, autumn and winter wind-solar-load scenarios and extreme wind-load scenarios.

[0103] The current method for considering uncertainty is not accurate enough. Based on the labelled generative adversarial networks (LGAN) of spring, summer, autumn and winter, it can generate four-season output scenarios and extreme output scenarios. The wind-solar joint output scenario model adds the time label h of the four seasons to both the generator and the discriminator. Combined with the spring, summer, autumn and winter label information of GAN, the LGAN value function Value(G,D) is:

[0104]

[0105] Where z is noise data, G is the generator, D is the discriminator, y is the real data, and h is the spring, summer, autumn and winter label. LGAN generates a large amount of wind power and photovoltaic scene data. The 8760 hours of historical data throughout the year can be divided into 365 scenes, according to the scene division of March to May as spring scenes, June to August as summer scenes, September to November as autumn scenes, and December to February as winter scenes. There are 92 scenes in spring, 92 scenes in summer, 91 scenes in autumn, and 90 scenes in winter. 30 scenes are selected as training sets for each season, and the remaining scenes are used as test sets. Each season is trained separately based on the training set of the scene to obtain a seasonal model, and then random output scenes of four seasons are generated based on the model and the test set, where each test scene can generate 1869 random output scenes. This scene is similar to the statistical characteristics of the actual data, which improves the accuracy of LGAN model training and the efficiency of scene generation.

[0106] GAN needs to design the size of weight limits. Improper design will lead to gradient vanishing or gradient explosion, which will affect the accuracy of wind and solar power output scenarios. Introducing GP into the value function can effectively improve the training speed and stability of the model.

[0107]

[0108] The value function Value(G,D) of LGAN-GP is:

[0109]

[0110] Where λ is the gradient penalty function.

[0111] (3) Solve the coordinated planning and configuration capacity of coal-fired power and renewable energy in the power system under different wind and solar load scenarios.

[0112] (31) Objective function

[0113] Capacity allocation can promote the consumption of wind and solar energy and reduce the wind curtailment rate. However, this method requires higher investment costs. Therefore, capacity allocation needs to be comprehensively evaluated from multiple perspectives.

[0114] minF=∑(C total ,f(η a ,η b )) min (8)

[0115] In the formula, C total is the total system cost.

[0116] (311) The first goal is to maximize revenue, that is, to minimize total cost.

[0117] minC total=-(C PA ·W pv +C WA ·W w +C EA ·W bes +C CA ·W g +C s +C G ·W g ) (9)

[0118] In the formula, C PA , C WA , C EA , C CA , C G are the unit capacity costs of photovoltaic, wind, energy storage and coal-fired power generation units respectively. pv , W w , W bes , W g They are the capacities of photovoltaic, wind, energy storage, and coal-fired power generation units, among which the capacities of photovoltaic and wind are known, while the capacities of energy storage and coal-fired power generation units are unknown. s For power generation income.

[0119] The full life cycle cost of photovoltaic power generation per unit capacity:

[0120]

[0121] In the formula, CI v represents the investment cost, CO represents the operating cost, CM represents the failure cost, CF represents the maintenance cost, and CD represents the v Represents the processing cost. i represents the cost of the i-th item in the investment cost, m represents the total number of items included in the investment cost. r represents the loan rate, R represents the annual interest rate, and λ represents the discount rate. 1 Represents the repair and maintenance cost of the photovoltaic unit, CM 2 represents the maintenance and renewal cost of photovoltaic facilities, CF 1 is the failure cost. d represents the cost of disposing of obsolete equipment, and T represents the residual value of fixed assets.

[0122] Wind power life cycle cost per unit capacity:

[0123]

[0124] Where CI is the initial investment cost, CI 1 Design costs for research planning, CI 2 is the purchase and installation cost of wind power equipment, Cl 3 is the land acquisition and renovation cost, CI4 Operation and management costs; CR t represents operating cost, CR 1t represents the daily management cost, CR 2t Represents daily operating costs. t represents the cost associated with the maintenance of updated facilities, L represents the cost rate of facility maintenance and update. S is the installation cost of the configuration facilities, CT t Costs incurred due to maintenance losses. t is the percentage of failure cost in the tth year to the initial investment cost; CD represents the residual value gain, and f represents the residual value rate of 5%.

[0125] The full life cycle cost of energy storage equipment per unit capacity:

[0126]

[0127] In the formula, C sys represents the installation cost of the energy storage system, C rep represents the cost and time required for energy storage system update, C fom represents the fixed annual maintenance cost of the energy storage system, C vom Represents the annual change in the maintenance cost of the energy storage system, C rec Represents its recoverable residual value; among them, C bat is the energy storage battery cost, C pcs is the cost of energy conversion equipment, C bop is the fixed configuration equipment cost. α represents the annual reduction percentage of installation cost, k represents the number of battery replacements, and n represents the number of years of battery storage life cycle. f-p Represents the operation and maintenance cost per unit power, and P is the rated power of the energy storage system. c is the charging cost, t is the rated discharge time, ψ is the system fixed efficiency, D is the number of operating days, and y represents the battery recovery ratio.

[0128] Electricity sales cost:

[0129]

[0130] In the formula, θ d The price of electricity sold by the system.

[0131] The life cycle cost of coal-fired power units:

[0132]

[0133] Where w and k represent the system discount rate and the expected operating life of the equipment, respectively.

[0134] (312) The second goal is to achieve the optimal comprehensive parameter, which is expressed as follows:

[0135] minC para =min(α 1 η a +α 2 η b ) (15)

[0136] In the formula, α 1 , α 2 is the weight of each indicator.

[0137] Wind-solar complementarity ratio:

[0138] Wind and solar resources have obvious complementary properties in both time and space scales. Maximizing their complementarity can effectively make the overall output smooth. The calculation method of wind-solar complementarity ratio is as follows:

[0139]

[0140] From the above formula, we can observe that η a The smaller it is, the higher the level of synergy between wind energy and photovoltaics.

[0141] Load Loss Probability LOLP:

[0142] Definition η b is the LOLP value of the hybrid power generation system, and the calculation formula is as follows:

[0143]

[0144] It can be clearly seen from the above formula that as η b With the reduction of load, the power generation system achieves a higher degree of load matching.

[0145] (312) Constraints

[0146] Power balance constraints

[0147] P t w +P t pv +P t g +P t bes =P t d (18)

[0148] Where P t w , P t pv , P t g , P bes , P t dThey are the total wind power output, photovoltaic output, coal-fired power unit output, energy storage equipment output, and transmitted power at time t.

[0149] Energy storage / coal-fired power generation capacity constraints

[0150]

[0151] In the formula, It is the upper limit of installed capacity of energy storage and coal-fired power units.

[0152] Wind power / photovoltaic / energy storage / coal-fired power generation unit output power constraints

[0153]

[0154] In the formula, f wt 、f pν 、f bs 、f bs It is the derating factor of the power generation system taking various factors into account.

[0155] Energy storage power station operation status constraints

[0156] Soc min ≤Soc t ≤Soc max (twenty one)

[0157] In the formula, Soc max 、Soc min It is the upper and lower limits of the battery storage status of the energy storage power station.

[0158] Parameter constraints

[0159]

[0160] Constraints related to outbound delivery channels

[0161] 1) Channel transmission power range constraints:

[0162]

[0163] Where: and They are the lower limit and upper limit of the transmitted power at time t, respectively, and are known quantities.

[0164] 2) Transmission capacity constraints of the transmission channel:

[0165]

[0166] Where: χ represents the minimum utilization hours of the delivery channel throughout the year, which is a known quantity.

[0167] 3) Outbound channel adjustment rate constraints:

[0168]

[0169] Where: is a 0-1 variable, indicating that the channel is upregulated and downregulated at time t, and is not 1 at the same time. and They represent the lower and upper limits of the forward adjustment of the external power, and They respectively represent the lower limit and upper limit of the reverse adjustment of the external power, and both are known quantities.

Claims

1. A method for collaborative planning of coal-fired power and new energy for power systems, characterized in that: The following steps are involved: (1) Determine the dynamic frequency security constraints and voltage stability constraints of the power system; (2) Based on the four seasons labels and adversarial network, a wind-solar joint output scenario model is constructed to generate spring, summer, autumn and winter wind-solar load scenarios and extreme wind-load scenarios; (3) Solve the coordinated planning and configuration capacity of coal-fired power and renewable energy in the power system under different wind and solar load scenarios.

2. The method for collaborative planning of coal-fired power and new energy for power systems according to claim 1 is characterized in that: Dynamic frequency safety constraints include frequency change rate constraints and quasi-steady-state frequency constraints; The frequency change rate constraint is in, is the equivalent inertia of the system at time t; and They are the state variables of coal-fired power units, wind power units, photovoltaic units and energy storage at time t respectively; is the equivalent inertia of coal power, wind power, photovoltaic power and energy storage; W i g , P i w , P i pv 、E m They are the capacity of coal-fired power units, the output of wind farms, the output of photovoltaic units and the energy storage capacity; X Rocof is the frequency change rate constraint; ΔP0 is the power disturbance corresponding to the most serious accident; f0 is the initial frequency N g 、N w 、N w 、N BES They are the number of coal-fired power, wind power, photovoltaic power and energy storage respectively; The quasi-steady-state frequency constraint is Where Δf ss is the maximum allowable deviation under quasi-steady state; P t L is the system load; It is the frequency modulation power that can be released by conventional units in quasi-steady state; is the frequency modulation power that can be released by the energy storage in the quasi-steady state; D is the synchronous power coefficient.

3. The method for coordinated planning of coal-fired power and new energy for power systems according to claim 1 is characterized in that: The voltage stability constraint includes the static voltage stability margin γ c for γ c (P cr,i -P L0,i ) / b i P G +ΔP G (c c )-P L -c c b=f(x) Among them, P cr,i is the load size of the current operating point of node i; P L0,i is the load size at the voltage collapse point of node i; P cr,i -P L0,i is the absolute load margin; b i is the load growth direction of node i; P G Inject power into the generator base state; P L is the node load base power; b is the load growth mode; ΔP G (γ c ) is the generator dispatching mode; x is the system static state vector, x = [θ T ,V T ] T , θ is the node voltage phase angle vector, V is the node voltage amplitude vector; f(x) is the node injection power function.

4. The method for collaborative planning of coal-fired power and new energy for power systems according to claim 1 is characterized in that: The wind-solar joint output scenario model introduces GP, and the value function Value(G,D) of LGAN-GP is obtained as follows: Among them, y represents the real sample data, p y It represents the probability distribution followed by these data samples. Based on sample yp y The average expected value of the distribution, D(y|h) is the output score of the discriminator D for the input y under the given condition h; Based on the noise distribution zp z The average expectation of the generated samples; G(Z|h) is the generator G converting random noise z into generated samples under condition h; λ is the gradient penalty function, It is the two-norm of the gradient of the discriminator D(x) with respect to the input y, which is used to control the smoothness of the gradient and prevent the gradient from exploding or disappearing.

5. The method for coordinated planning of coal-fired power and new energy for power systems according to claim 1 is characterized in that: The objective function minF of the coordinated planning and allocation capacity of coal-fired power and new energy in the power system is: minF=∑(C total ,f(η a ,or b )) min f(h a ,or b )=α1η a +a2h b Among them, Ctotal is the total system cost, η a is the wind-solar energy complementarity ratio, η b is the LOLP value of the hybrid power generation system. α1 and α2 are weights.

6. The method for collaborative planning of coal-fired power and new energy for power systems according to claim 5 is characterized in that: The lowest total system cost is minC total =-(C PA ·W pv +C WA ·W w +C EA ·W bes +C CA ·W g +C s +C G ·W g ) Among them, C PA , C WA , C EA , C CA , C G are the unit capacity costs of photovoltaic, wind power, energy storage and coal-fired power generation units respectively; W pv , W w , W bes , W g are the capacities of photovoltaic, wind power, energy storage, and coal-fired power units; C s For power generation income.

7. The method for coordinated planning of coal-fired power and new energy for power systems according to claim 6 is characterized in that: The life cycle cost of photovoltaic power generation per unit capacity C PA for In the formula, CI v represents the investment cost of photovoltaic equipment, CO is the operating cost, CM is the maintenance cost, CF represents the failure cost, CD v represents the processing cost; Cl i represents the cost of the i-th item in the investment cost, m represents the total number of items included in the investment cost; r represents the loan rate, R represents the annual interest rate, and λ represents the discount rate; CM1 represents the repair and maintenance cost of the photovoltaic unit, CM2 represents the maintenance and renewal cost of the photovoltaic facilities, and CF1 represents the equipment failure cost; C d represents the cost of disposing of obsolete equipment, and T represents the residual value of fixed assets; Wind power life cycle cost per unit capacity C WA for Where CI is the initial investment cost of wind power equipment, CI1 is the research, planning and design cost, CI2 is the purchase and installation cost of wind power equipment, Cl3 is the land purchase cost, and CI4 is the operation and management cost; CR t represents operating cost, CR 1t represents the daily management cost, CR 2t represents daily operating costs; CS t represents the cost associated with the maintenance of updated facilities, L represents the cost rate of facility maintenance and update; S is the installation cost of the configuration facilities, CT t Costs incurred due to maintenance losses; X t is the percentage of failure cost in the tth year to the initial investment cost; CD represents the residual value gain, and f represents the residual value rate; The life cycle cost of energy storage equipment per unit capacity C EA for In the formula, C sys represents the installation cost of the energy storage system, C rep represents the cost and time required for energy storage system update, C fom represents the fixed annual maintenance cost of the energy storage system, C vom Represents the annual change in the maintenance cost of the energy storage system, C rec Represents its recoverable residual value; among them, C bat is the energy storage battery cost, C pcs is the cost of energy conversion equipment, C bop is the fixed configuration equipment cost; α represents the annual reduction percentage of installation cost, k represents the number of battery replacements, and n represents the number of years of battery storage life cycle; C f-p represents the operation and maintenance cost per unit power, P is the rated power of the energy storage system; C c is the charging cost, t is the rated discharge time, ψ is the system fixed efficiency, D is the number of operating days; y represents the battery recycling ratio; Electricity sales cost C A for In the formula, θ d The price of electricity sold to the system; The life cycle cost of coal-fired power units C G In the formula, w and k represent the system discount rate and the expected operating life of the equipment, respectively, g is the unit installed capacity investment cost of coal-fired power units, W g The installed capacity of coal-fired power plants.

8. The method for coordinated planning of coal-fired power and new energy for power systems according to claim 7 is characterized in that: The coordinated planning and configuration capacity of coal-fired power and new energy in the power system meets the following constraints: Power balance constraints: P t w +P t pv +P t g +P t bes =P t d Where P t w , P t g , P bes , P t d They are the total wind power output, photovoltaic output, coal-fired power unit output, energy storage equipment output, and external power at time t; Energy storage / coal-fired power installed capacity constraints: In the formula, The upper limit of installed capacity of energy storage-loaded coal-fired power units; Wind power / photovoltaic / energy storage / coal-fired power generator output power constraints: In the formula, f wt 、f pν 、f bs 、f bs The derating factor of the power generation system taking into account various factors; Energy storage power station operation status constraints: Soc min ≤Soc t ≤Soc max In the formula, Soc max , Soc min The upper and lower limits of the battery storage state of the energy storage power station; Parameter constraints: Channel transmission power range constraints: In the formula, and are the lower and upper limits of the transmitted power at time t respectively. Transmission capacity constraints of the outbound transmission channel: In the formula, χ represents the minimum utilization hours of the delivery channel throughout the year, which is a known quantity. Outbound channel adjustment rate constraints: In the formula, is a 0-1 variable, indicating that the channel is upregulated and downregulated at time t, and is not 1 at the same time. and They represent the lower and upper limits of the forward adjustment of the external power, and They respectively represent the lower limit and upper limit of the reverse adjustment of the external power, and both are known quantities.

9. The method for coordinated planning of coal-fired power and new energy for power systems according to claim 8 is characterized in that: Wind-solar energy complementarity ratio η a for In the formula, is the average value of the transmitted power, N is the time period; Hybrid power generation system LOLP value η b for 10. A coal-fired power and new energy collaborative planning system for power systems, characterized in that: include System constraint module, used to perform dynamic frequency security constraints and voltage stability constraints on the power system; The model building module builds a wind-solar joint output scenario model based on four seasons labels and adversarial networks, generating spring, summer, autumn, winter, wind-solar-load scenarios and extreme wind-load scenarios; The solution module is used to solve the coordinated planning and configuration capacity of coal-fired power and new energy in the power system under different wind and solar load scenarios.

Images