Water-light-storage complementary and direct-current external sending two-stage random optimization scheduling method and system

Abstract

The present application relates to the technical field of power system optimization scheduling, and particularly relates to a water-light-storage complementary and DC external sending two-stage random optimization scheduling method and system, which comprises the following steps: establishing a water-light-storage complementary and DC external sending system model containing thermal power units, photovoltaic units, cascade hydropower, pumped storage units, static var compensators and DC external sending; taking the minimization of operation cost as an objective function, and taking node balance constraints, branch power flow constraints, thermal power unit constraints, photovoltaic output constraints, pumped storage power station constraints, cascade hydropower constraints, static var compensator constraints and branch operation constraints as constraint conditions, a two-stage random optimization scheduling model is constructed; considering photovoltaic uncertainty by using a conditional generative adversarial network (CGAN), typical scenarios are generated, and the generated scenarios are reduced by using an Euclidean hierarchical clustering method (EAHCM); the two-stage random optimization scheduling model is solved to obtain an optimization scheduling result, and the goal of economic scheduling is achieved.

Description

Technical Field

[0001] This invention relates to the field of power system optimization scheduling technology, and in particular to a two-stage stochastic optimization scheduling method and system for hydro-solar-storage complementarity and DC transmission. Background Technology

[0002] With the introduction of the "dual carbon" target, clean energy is continuously developing and growing, gradually becoming the main force in energy transformation. At present, the integration of the industrial chain and the mechanism of ecological synergy urgently need breakthroughs. Insufficient cross-industry collaboration will lead to poor connection between green electricity and energy-intensive industries, ultimately resulting in the abandonment of hydropower and solar power.

[0003] In remote areas, new energy and large-scale hydropower bases are often far from electricity load centers. Therefore, high-voltage direct current (HVDC) transmission becomes a crucial means of achieving inter-regional power transmission and promoting the consumption of hydropower and photovoltaic power. However, the peak-shaving capacity of these transmission lines has become a major obstacle to the efficiency of renewable energy utilization. In this situation, pumped-storage hydropower stations with excellent peak-shaving capabilities, when combined with hydro-solar power generation systems, can significantly improve the dynamic adjustment capability of the power grid and optimize the utilization efficiency of renewable energy resources.

[0004] Pumped storage hydroelectric power stations, as a special type of energy storage power source, are of great significance to the stable operation of the power system. Compared with conventional power generation methods, this technology achieves large-scale energy storage, is flexible, and can quickly respond to dynamic load changes, improving the safety, stability, reliability, and power quality of the power system, thereby optimizing the power supply structure and reducing overall power system losses. Therefore, as the scale of the power system expands and the proportion of renewable energy continues to increase, the strategic position of pumped storage hydroelectric power stations in the power grid is becoming increasingly crucial.

[0005] While some progress has been made in optimizing DC power transmission, several issues remain, including insufficient coordination between the power source and grid sides, and inadequate optimization of inter-regional dispatch strategies. The consideration of uncertainties in renewable energy generation is relatively simplistic, failing to achieve a balance between stochasticity and economic efficiency. Furthermore, most existing models neglect reactive power and voltage constraints in the AC grid, which may impact the safety and economic viability of actual operation.

[0006] The information disclosed in this background section is intended only to enhance the understanding of the general background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] This invention provides a two-stage stochastic optimization scheduling method and system for hydro-solar-storage complementarity and DC transmission, thereby effectively solving the problems in the background technology.

[0008] To achieve the above objectives, the technical solution adopted by this invention is: a two-stage stochastic optimization scheduling method for hydro-solar-storage complementarity and DC transmission, comprising the following steps:

[0009] Establish a model of a hydro-solar-storage complementary system and DC transmission system, including thermal power units, photovoltaic units, cascade hydropower, pumped storage units, static var compensators, and DC transmission.

[0010] With the objective function of minimizing the two-stage system operating cost, and with constraints such as node balance constraints, branch power flow constraints, thermal power unit constraints, photovoltaic power output constraints, pumped storage power station constraints, cascade hydropower constraints, static var compensator constraints, and tributary operation constraints, a two-stage stochastic optimization scheduling model for a hydro-solar-storage complementary and DC transmission system considering photovoltaic uncertainty is constructed.

[0011] We use Conditional Generative Adversarial Network (CGAN) to consider photovoltaic uncertainties, generate typical scenarios, and then reduce the generated scenarios using the Euclidean hierarchical clustering method EAHCM.

[0012] Solve the two-stage stochastic optimization scheduling model to obtain the optimized scheduling result.

[0013] Furthermore, the objective function is:

[0014] ;

[0015] In the formula: For the first stage of decision-making, this represents the unit start-up and shutdown plan; for The feasible domain; Let the objective function be the first-stage optimization problem. Let be a random variable, representing the uncertainty of renewable energy generation and load; for The probability distribution; These are the decision variables for the second stage. The objective function for the second-stage optimization problem; for The feasible domain; Indicates the expected value; , , and These represent the time period, the number of thermal power units, photovoltaic power stations, and pumped storage power stations, respectively. Number the time period; , and These are the numbers for thermal power units, photovoltaic power plants, and pumped storage power plants, respectively. express Periodic thermal power units The amount of electricity generated; Indicates thermal power unit The cost of no-load operation; express Periodic thermal power units The start / stop status; The unit price of fuel; and They represent Periodic thermal power units The cost of powering on and off; Indicates the unit light abandonment penalty factor; and They are respectively Periodic photovoltaic power station The predicted output and actual power generation; and These are the start-up and shutdown costs for a single variable speed generator unit, respectively. and They are respectively Pumped storage power station Total number of pumped storage units started and shut down.

[0016] Furthermore, the branch power flow constraint is as follows:

[0017] ;

[0018] ;

[0019] ;

[0020] ;

[0021] ;

[0022] In the formula: and These are the corresponding branches in the admittance matrix. The real and imaginary parts; for Time Node and The phase angle difference between them; and They are respectively Time Node and The voltage amplitude; and For the line Active and reactive power flow limits; and Represents a node Upper and lower limits of voltage amplitude.

[0023] Furthermore, it also includes linearizing the power flow constraints of the aforementioned branches:

[0024] Assumption Very small and the voltage at each node is close to the rated voltage, so that , , , Then we have:

[0025] ;

[0026] ;

[0027] Will As an independent variable, a mathematical transformation of the nonlinear voltage amplitude term is used. , to nonlinear terms Transform into linear and quadratic terms:

[0028] ;

[0029] By substituting the above equation into the first two equations, we obtain the following: and Linearized network model:

[0030] ;

[0031] ;

[0032] at this time:

[0033] ;

[0034] Assuming under the basic conditions, and The value is The loss is decomposed into voltage angle term and voltage amplitude term:

[0035] ;

[0036] According to the first-order Taylor series expansion, the term , The linearization is as follows:

[0037] ;

[0038] in yes A function; because As the independent variable, firstly Transform into The function is then expanded using a first-order Taylor series:

[0039] ;

[0040] Finally add , can be obtained The complete formula, Similarly, we can conclude that:

[0041] ;

[0042] .

[0043] Furthermore, the cascade hydropower constraints are as follows:

[0044] ;

[0045] ;

[0046] ;

[0047] ;

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] ;

[0053] ;

[0054] ;

[0055] ;

[0056] ;

[0057] ;

[0058] ;

[0059] In the formula: and They represent Cascade hydropower stations The meritorious and the ineffective contributions; express Periodic hydroelectric power station The operating status is indicated by "1" for power-on and "0" for power-off. and Cascade hydropower Maximum and minimum active power output; and Cascade hydropower Maximum and minimum reactive power output; Indicates cascade hydropower The power factor; and These represent cascade hydropower. The upward and downward climbing limits; It is a composite constant used to simplify power calculations, reflecting the combined effects of turbine efficiency, water density, and gravitational acceleration; Represented as cascade hydropower The efficiency of the turbine; Density of water, unit: kg / m³; Expresses gravitational acceleration, unit: m / s²; for Periodic hydroelectric power station Hydropower head; for Periodic hydroelectric power station The power generation flow rate; and Hydropower stations The maximum and minimum permissible power generation flow rates; and Hydropower stations The maximum and minimum permissible head for generating electricity; Indicates hydroelectric power station exist Storage capacity for a given period of time; for Periodic hydroelectric power station Inbound traffic; for Periodic hydroelectric power station The discharge flow rate; and These represent the cases considering water retention. Periodic hydroelectric power station direct upstream power station Power generation flow and water discharge flow; Indicates hydroelectric power station The time it takes for water to flow directly downstream; express Periodic hydroelectric power station The natural water inflow; and Indicates hydroelectric power station Maximum and minimum allowed storage capacity; and These represent the initial and final storage capacities for scheduling.

[0060] Furthermore, the Conditional Generative Adversarial Network (CGAN) is used to consider photovoltaic uncertainties and generate typical scenarios, including:

[0061] Set conditions to generate the objective function of the CGAN adversarial network:

[0062] ;

[0063] In the formula, This is a gradient penalty term; Represents real samples and generate samples Random sampling along a straight line;

[0064] Set the loss functions for the conditional generative adversarial network (CGAN) generator and discriminator:

[0065] ;

[0066] ;

[0067] The Wasserstein distance is introduced into the discriminator loss function to measure the distance between generated samples and real samples, and its definition is as follows:

[0068] ;

[0069] In the formula, The joint distribution representing the optimal path, for Distribution and The distance between them; This indicates that all probability distributions can be represented Convert to probability distribution The set of feasible solutions In the middle, take an infimum; Represents real samples With generated samples The distance between them; due to The solution is quite complex, so the Kantorovich-Rubinstein dual form is used to describe the generation of samples. Compared with real samples The distance between them is expressed as:

[0070] ;

[0071] In the formula, The constant of the Lipschitz function; This indicates that the discriminator function satisfies the Lipschitz continuity condition of constant.

[0072] Furthermore, the reduction of the generated scene using the Euclidean hierarchical clustering method EAHCM includes the following steps:

[0073] First, calculate the Euclidean distance and mean value among the cluster members. and center point medoid vector;

[0074]

[0075]

[0076] In the formula, It is a medoid vector. It is the mean vector of the cluster;

[0077] Then, EAHCM selects the option with the smallest distance from other options; the following formula represents the selection part of EAHCM, where... This is the representative vector for the corresponding cluster;

[0078] ;

[0079] After selecting a representative time period within the cluster, EAHCM measures the dissimilarity between the representative vectors of all adjacent clusters; then, the two adjacent clusters with the highest similarity are merged; the following formula is used to calculate the dissimilarity between two adjacent clusters:

[0080] ;

[0081] In the formula, and Clustering and The representative vector; in each iteration, EAHCM merges the two clusters, and this process continues until the optimal number of clusters is reached.

[0082] Furthermore, solving the two-stage stochastic optimization scheduling model includes:

[0083] The two-stage stochastic optimal scheduling model of a photovoltaic-hydro-storage complementary system and a DC transmission system considering photovoltaic uncertainties was solved using mixed-integer linear programming, and the optimal scheduling results were obtained.

[0084] This invention also includes a two-stage stochastic optimization scheduling system for hydro-solar-storage complementarity and DC transmission, using the method described above. The system includes:

[0085] The system modeling unit is used to establish a model of a hydro-solar-storage complementary system that includes thermal power units, photovoltaic units, cascade hydropower, pumped storage units, static var compensators, and DC transmission.

[0086] The scheduling model modeling unit is used to construct a two-stage stochastic optimization scheduling model for a hydro-solar-storage complementary system and a DC transmission system that takes the minimization of the two-stage system operation cost as the objective function and the constraints as node balance constraints, branch power flow constraints, thermal power unit constraints, photovoltaic power output constraints, pumped storage power station constraints, cascade hydropower constraints, static var compensator constraints, and tributary operation constraints.

[0087] The scene generation unit is used to generate typical scenes by taking into account photovoltaic uncertainties using the Conditional Generative Adversarial Network (CGAN), and to reduce the generated scenes using the Euclidean hierarchical clustering method EAHCM.

[0088] The solution unit is used to solve the two-stage stochastic optimization scheduling model and obtain the optimized scheduling result.

[0089] The present invention also includes a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as described above.

[0090] The present invention also includes a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described above.

[0091] The beneficial effects of this invention are as follows: By constructing a two-stage stochastic optimization scheduling model for a hydro-solar-storage complementary system and a DC transmission system that considers photovoltaic uncertainties, the system scheduling cost can be reduced to a certain extent and the utilization efficiency of renewable energy can be improved by coordinating the power generation adjustment work with the configuration of pumped storage power stations in the hydro-solar complementary scheduling. Furthermore, with the adjustment of DC transmission power, the system operating cost continues to decrease, verifying that DC power optimization has a significant effect on improving system economy. It also points out the huge optimization potential in the coordinated operation of DC transmission and generator units, providing a key basis for further improving system scheduling strategies. By using a conditional generative adversarial network (CGAN) to consider photovoltaic uncertainties, typical scenarios are generated, and the generated scenarios are reduced using the Euclidean hierarchical clustering method EAHCM. Solving the two-stage stochastic optimization scheduling model yields optimized scheduling results, reducing curtailed solar power and achieving the goal of economical power system scheduling. Attached Figure Description

[0092] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0093] Figure 1 This is a flowchart of the method in Example 1;

[0094] Figure 2 This is a schematic diagram of the system structure in Example 1;

[0095] Figure 3 This is a flowchart of the optimized scheduling method in Example 2;

[0096] Figure 4 This is a schematic diagram of the nodes of the hydro-solar-storage power system in Example 2;

[0097] Figure 5 The above are the daily forecast data curves for the hydro-solar-storage power system in Example 2.

[0098] Figure 6 This is a schematic diagram of the structure of the computer device of the present invention. Detailed Implementation

[0099] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0100] Example 1:

[0101] like Figure 1 As shown: A two-stage stochastic optimization scheduling method for hydro-solar-storage complementarity and DC transmission includes the following steps:

[0102] Establish a model of a hydro-solar-storage complementary system and DC transmission system, including thermal power units, photovoltaic units, cascade hydropower, pumped storage units, static var compensators, and DC transmission.

[0103] With the objective function of minimizing the two-stage system operating cost, and with constraints such as node balance constraints, branch power flow constraints, thermal power unit constraints, photovoltaic power output constraints, pumped storage power station constraints, cascade hydropower constraints, static var compensator constraints, and tributary operation constraints, a two-stage stochastic optimization scheduling model for a hydro-solar-storage complementary and DC transmission system considering photovoltaic uncertainty is constructed.

[0104] We use Conditional Generative Adversarial Network (CGAN) to consider photovoltaic uncertainties, generate typical scenarios, and then reduce the generated scenarios using the Euclidean hierarchical clustering method EAHCM.

[0105] Solve the two-stage stochastic optimization scheduling model to obtain the optimized scheduling result.

[0106] By constructing a two-stage stochastic optimization scheduling model for a hydro-solar-storage complementary system and a DC transmission system that considers photovoltaic uncertainties, the system scheduling cost can be reduced to a certain extent and the utilization efficiency of renewable energy can be improved by coordinating the power generation adjustment work with the configuration of pumped storage stations in the hydro-solar complementary scheduling. Combined with DC transmission power adjustment, the system operating cost continues to decrease, verifying that DC power optimization has a significant effect on improving system economy. It also points out the huge optimization potential in the coordinated operation of DC transmission and generator units, providing a key basis for further improving system scheduling strategies. Using a conditional generative adversarial network (CGAN) to consider photovoltaic uncertainties, typical scenarios are generated, and the generated scenarios are reduced using the Euclidean hierarchical clustering method EAHCM. Solving the two-stage stochastic optimization scheduling model yields optimized scheduling results, reducing curtailed solar power and achieving the goal of economical power system scheduling.

[0107] In this embodiment, the objective function is:

[0108] ;

[0109] In the formula: For the first stage of decision-making, this represents the unit start-up and shutdown plan; for The feasible domain; Let the objective function be the first-stage optimization problem. Let be a random variable, representing the uncertainty of renewable energy generation and load; for The probability distribution; These are the decision variables for the second stage. The objective function for the second-stage optimization problem; for The feasible domain; Indicates the expected value; , , and These represent the time period, the number of thermal power units, photovoltaic power stations, and pumped storage power stations, respectively. Number the time period; , and These are the numbers for thermal power units, photovoltaic power plants, and pumped storage power plants, respectively. express Periodic thermal power units The amount of electricity generated; Indicates thermal power unit The cost of no-load operation; express Periodic thermal power units The start / stop status; The unit price of fuel; and They represent Periodic thermal power units The cost of powering on and off; Indicates the unit light abandonment penalty factor; and They are respectively Periodic photovoltaic power station The predicted output and actual power generation; and These are the start-up and shutdown costs for a single variable speed generator unit, respectively. and They are respectively Pumped storage power station Total number of pumped storage units started and shut down.

[0110] Branch flow constraints are:

[0111] ;

[0112] ;

[0113] ;

[0114] ;

[0115] ;

[0116] In the formula: and These are the corresponding branches in the admittance matrix. The real and imaginary parts; for Time Node and The phase angle difference between them; and They are respectively Time Node and The voltage amplitude; and For the line Active and reactive power flow limits; and Represents a node Upper and lower limits of voltage amplitude.

[0117] This also includes linearizing the branch power flow constraints:

[0118] Assumption Very small and the voltage at each node is close to the rated voltage, so that , , , Then we have:

[0119] ;

[0120] ;

[0121] Will As an independent variable, a mathematical transformation of the nonlinear voltage amplitude term is used. , to nonlinear terms Transform into linear and quadratic terms:

[0122] ;

[0123] By substituting the above equation into the first two equations, we obtain the following: and Linearized network model:

[0124] ;

[0125] ;

[0126] at this time:

[0127] ;

[0128] Assuming under the basic conditions, and The value is The loss is decomposed into voltage angle term and voltage amplitude term:

[0129] ;

[0130] According to the first-order Taylor series expansion, the term , The linearization is as follows:

[0131] ;

[0132] in yes A function; because As the independent variable, firstly Transform into The function is then expanded using a first-order Taylor series:

[0133] ;

[0134] Finally add , can be obtained The complete formula, Similarly, we can conclude that:

[0135] ;

[0136] .

[0137] The constraints of cascade hydropower are:

[0138] ;

[0139] ;

[0140] ;

[0141] ;

[0142] ;

[0143] ;

[0144] ;

[0145] ;

[0146] ;

[0147] ;

[0148] ;

[0149] ;

[0150] ;

[0151] ;

[0152] ;

[0153] In the formula: and They represent Cascade hydropower stations The meritorious and the ineffective contributions; express Periodic hydroelectric power station The operating status is indicated by "1" for power-on and "0" for power-off. and Cascade hydropower Maximum and minimum active power output; and Cascade hydropower Maximum and minimum reactive power output; Indicates cascade hydropower The power factor; and These represent cascade hydropower. The upward and downward climbing limits; It is a composite constant used to simplify power calculations, reflecting the combined effects of turbine efficiency, water density, and gravitational acceleration; Represented as cascade hydropower The efficiency of the turbine; Density of water, unit: kg / m³; Expresses gravitational acceleration, unit: m / s²; for Periodic hydroelectric power station Hydropower head; for Periodic hydroelectric power station The power generation flow rate; and Hydropower stations The maximum and minimum permissible power generation flow rates; and Hydropower stations The maximum and minimum permissible head for generating electricity; Indicates hydroelectric power station exist Storage capacity for a given period of time; for Periodic hydroelectric power station Inbound traffic; for Periodic hydroelectric power station The discharge flow rate; and These represent the cases considering water retention. Periodic hydroelectric power station direct upstream power station Power generation flow and water discharge flow; Indicates hydroelectric power station The time it takes for water to flow directly downstream; express Periodic hydroelectric power station The natural water inflow; and Indicates hydroelectric power station Maximum and minimum allowed storage capacity; and These represent the initial and final storage capacities for scheduling.

[0154] In this embodiment, the conditional generative adversarial network CGAN is used to consider photovoltaic uncertainties and generate typical scenarios, including:

[0155] Set conditions to generate the objective function of the CGAN adversarial network:

[0156] ;

[0157] In the formula, This is a gradient penalty term; Represents real samples and generate samples Random sampling along a straight line;

[0158] Set the loss functions for the conditional generative adversarial network (CGAN) generator and discriminator:

[0159] ;

[0160] ;

[0161] The Wasserstein distance is introduced into the discriminator loss function to measure the distance between generated samples and real samples, and its definition is as follows:

[0162] ;

[0163] In the formula, The joint distribution representing the optimal path, for Distribution and The distance between them; This indicates that all probability distributions can be represented Convert to probability distribution The set of feasible solutions In the middle, take an infimum; Represents real samples With generated samples The distance between them; due to The solution is quite complex, so the Kantorovich-Rubinstein dual form is used to describe the generation of samples. Compared with real samples The distance between them is expressed as:

[0164] ;

[0165] In the formula, The constant of the Lipschitz function; This indicates that the discriminator function satisfies the Lipschitz continuity condition of constant.

[0166] The generated scene is reduced using the Euclidean hierarchical clustering method EAHCM, including the following steps:

[0167] First, calculate the Euclidean distance and mean value among the cluster members. and center point medoid vector;

[0168]

[0169]

[0170] In the formula, It is a medoid vector. It is the mean vector of the cluster;

[0171] Then, EAHCM selects the option with the smallest distance from other options; the following formula represents the selection part of EAHCM, where... This is the representative vector for the corresponding cluster;

[0172] ;

[0173] After selecting a representative time period within the cluster, EAHCM measures the dissimilarity between the representative vectors of all adjacent clusters; then, the two adjacent clusters with the highest similarity are merged; the following formula is used to calculate the dissimilarity between two adjacent clusters:

[0174] ;

[0175] In the formula, and Clustering and The representative vector; in each iteration, EAHCM merges the two clusters, and this process continues until the optimal number of clusters is reached.

[0176] Solving the two-stage stochastic optimization scheduling model includes:

[0177] The two-stage stochastic optimal scheduling model of a photovoltaic-hydro-storage complementary system and a DC transmission system considering photovoltaic uncertainties was solved using mixed-integer linear programming, and the optimal scheduling results were obtained.

[0178] like Figure 2 As shown, this embodiment also includes a two-stage stochastic optimization scheduling system for hydro-solar-storage complementarity and DC transmission, using the method described above. The system includes:

[0179] The system modeling unit is used to establish a model of a hydro-solar-storage complementary system that includes thermal power units, photovoltaic units, cascade hydropower, pumped storage units, static var compensators, and DC transmission.

[0180] The scheduling model modeling unit is used to construct a two-stage stochastic optimization scheduling model for a hydro-solar-storage complementary system and a DC transmission system that takes the minimization of the two-stage system operation cost as the objective function and the constraints as node balance constraints, branch power flow constraints, thermal power unit constraints, photovoltaic power output constraints, pumped storage power station constraints, cascade hydropower constraints, static var compensator constraints, and tributary operation constraints.

[0181] The scene generation unit is used to generate typical scenes by taking into account photovoltaic uncertainties using the Conditional Generative Adversarial Network (CGAN), and to reduce the generated scenes using the Euclidean hierarchical clustering method EAHCM.

[0182] The solution unit is used to solve the two-stage stochastic optimization scheduling model and obtain the optimized scheduling result.

[0183] Example 2:

[0184] like Figure 3 As shown in this embodiment, a two-stage stochastic optimization scheduling method for a hydro-solar-storage complementary and DC transmission system considering photovoltaic uncertainties is provided. This method establishes a model of the hydro-solar-storage complementary and DC transmission system, including thermal power units, photovoltaic units, cascade hydropower, pumped storage units, static var compensators, and DC transmission. The objective function is to minimize the two-stage system operating cost. Constraints are set on the units (thermal power units, photovoltaic output, pumped storage units, cascade hydropower, static var compensators, etc.), branch power flow constraints, node balance constraints, and tributary operation constraints. An optimal scheduling model for the hydro-solar-storage power system considering DC transmission is established. Then, considering photovoltaic uncertainties, a conditional generative adversarial network (CGAN) method is used to generate typical photovoltaic scenarios. Finally, the two-stage stochastic optimization scheduling result of the hydro-solar-storage complementary and DC transmission system considering photovoltaic uncertainties is obtained, reducing curtailment of solar power and achieving the goal of economical power system scheduling.

[0185] Specifically, in the above embodiment, the expression for the objective function is:

[0186] ;

[0187] In the formula: For the first stage of decision-making, this represents the unit start-up and shutdown plan; for The feasible domain; Let the objective function be the first-stage optimization problem. Let be a random variable, representing the uncertainty of renewable energy generation and load; for The probability distribution; These are the decision variables for the second stage. The objective function for the second stage optimization problem; for The feasible domain; This represents the expected value. , , and These represent the time period, the number of thermal power units, photovoltaic power stations, and pumped storage power stations, respectively. Number the time period; , and These are the numbers for thermal power units, photovoltaic power plants, and pumped storage power plants, respectively. express Periodic thermal power units The amount of electricity generated; Indicates thermal power unit The cost of no-load operation; express Periodic thermal power units The start / stop status; The unit price of fuel; and They represent Periodic thermal power units The cost of powering on and off; Indicates the unit light abandonment penalty factor; and They are respectively Periodic photovoltaic power station The predicted output and actual power generation; and These are the start-up and shutdown costs for a single variable speed generator unit, respectively. and They are respectively Pumped storage power station Total number of pumped storage units started and shut down.

[0188] Furthermore, in the above embodiments, the expression for the node balance constraint is:

[0189] ;

[0190] ;

[0191] In the formula: , , and These are the numbers for the cascade hydropower stations, converter stations, loads, and reactive power compensation devices, respectively. , and Number the nodes; Indicates connection to the busbar The collection of devices on it; and They represent Periodic thermal power units The meritorious and the ineffective contributions; and They represent Cascade hydropower stations The meritorious and the ineffective contributions; and They are respectively Pumped storage power station Total pumping and power generation capacity; and They represent Time-of-use converter station Active and reactive power input on the AC side; and They represent Time-of-use load The active and reactive power demands; express Time-limited static var compensator The amount of reactive power compensation; and They represent Time-of-use transmission lines The trend of meritorious and ineffective actions.

[0192] The expression for branch flow constraints is:

[0193] ;

[0194] ;

[0195] ;

[0196] ;

[0197] ;

[0198] In the formula: and These are the corresponding branches in the admittance matrix. The real and imaginary parts; for Time Node and The phase angle difference between them; and They are respectively Time Node and The voltage amplitude; and For the line Active and reactive power flow limits; and Represents a node Upper and lower limits of voltage amplitude.

[0199] The linearization process of the branch power flow constraints is as follows:

[0200] Assumption Very small and the voltage at each node is close to the rated voltage, so that , , , Then we have:

[0201] ;

[0202] ;

[0203] Will As an independent variable, a mathematical transformation of the nonlinear voltage amplitude term is used. Without sacrificing accuracy, the nonlinear term Transform into linear and quadratic terms:

[0204] ;

[0205] By substituting the above equation into the first two equations, we obtain the following: and Linearized network model:

[0206] ;

[0207] ;

[0208] at this time:

[0209] ;

[0210] In the above two equations, the power flow equations are linear except for the losses. As shown in the equations above, the influence of voltage magnitude on losses is taken into account.

[0211] by For example, assuming the basic conditions, and The value is The loss is decomposed into voltage angle term and voltage amplitude term:

[0212] ;

[0213] According to the first-order Taylor series expansion, the term , It can be linearized as follows:

[0214] ;

[0215] As shown in the above formula, yes A function. Because As the independent variable, firstly Transform into The function is then expanded using a first-order Taylor series:

[0216] ;

[0217] Finally add , can be obtained The complete formula, Similarly, we can conclude that:

[0218] ;

[0219] .

[0220] The expression for the operating constraints of thermal power units is:

[0221] ;

[0222] ;

[0223] ;

[0224] ;

[0225] ;

[0226] ;

[0227] ;

[0228] ;

[0229] ;

[0230] In the formula: Indicates thermal power unit exist The time during which the system remains operational before the specified time period; Indicates thermal power unit exist The time during which the device must remain switched off before the specified time period; and thermal power units Minimum power-on and minimum power-off times; express Periodic thermal power units The start / stop status; and They represent Periodic thermal power units The cost of powering on and off; and thermal power units Unit start-up and shutdown costs; and They represent Periodic thermal power units The meritorious and the ineffective contributions; and thermal power units Maximum and minimum active power output; and thermal power units Maximum and minimum reactive power output; Indicates thermal power unit The power factor; and They represent thermal power units The limits of upward and downward climbing.

[0231] The expression for the photovoltaic output constraint is:

[0232] ;

[0233] In the formula: and They are respectively Periodic photovoltaic power station The predicted output and actual power generation.

[0234] The expression for the constraints of a pumped storage power station is:

[0235] ;

[0236] ;

[0237] ;

[0238] ;

[0239] ;

[0240] ;

[0241] ;

[0242] ;

[0243] ;

[0244] ;

[0245] ;

[0246] The expression for the constraints of the cascade hydropower units is:

[0247] ;

[0248] ;

[0249]

[0250] ;

[0251] ;

[0252] ;

[0253] ;

[0254] ;

[0255] ;

[0256] ;

[0257] ;

[0258] ;

[0259] ;

[0260] ;

[0261] ;

[0262] In the formula: and They represent Cascade hydropower stations The meritorious and the ineffective contributions; express Periodic hydroelectric power station The operating status is indicated by "1" for power-on and "0" for power-off. and Cascade hydropower Maximum and minimum active power output; and Cascade hydropower Maximum and minimum reactive power output; Indicates cascade hydropower The power factor; and These represent cascade hydropower. The upward and downward climbing limits; It is a composite constant used to simplify power calculations, reflecting the combined effects of turbine efficiency, water density, and gravitational acceleration; Represented as cascade hydropower The efficiency of the turbine; This indicates the density of water (unit: kg / m³). This represents the acceleration due to gravity (unit: m / s²). for Periodic hydroelectric power station Hydropower head; for Periodic hydroelectric power station The power generation flow rate; and Hydropower stations The maximum and minimum permissible power generation flow rates; and Hydropower stations The maximum and minimum permissible head for generating electricity; Indicates hydroelectric power station exist Storage capacity for a given period of time; for Periodic hydroelectric power station Inbound traffic; for Periodic hydroelectric power station The discharge flow rate; and These represent the cases considering water retention. Periodic hydroelectric power station direct upstream power station Power generation flow and water discharge flow; Indicates hydroelectric power station The time it takes for water to flow directly downstream; express Periodic hydroelectric power station The natural water inflow; and Indicates hydroelectric power station Maximum and minimum allowed storage capacity; and These represent the initial and final storage capacities for scheduling.

[0263] The expression for the static var compensator constraint is:

[0264] ;

[0265] In the formula: express Time-limited static var compensator The amount of reactive power compensation; and These represent static var compensators (SVCs). The maximum and minimum reactive power compensation.

[0266] The expression for the DC operating constraint is:

[0267] ;

[0268] ;

[0269] ;

[0270] ;

[0271] ;

[0272] ;

[0273] ;

[0274] ;

[0275] ;

[0276] ;

[0277] ;

[0278] In the formula: and They represent Time-of-use converter station Active and reactive power input on the AC side; express Time-of-use converter station Active power output from the DC side; and They represent converter stations Upper and lower limits of active power on the AC side; and They represent converter stations Upper and lower limits of reactive power on the AC side; and They represent converter stations DC-side active power upper and lower limits; and DC transmission lines for adjacent time periods Maximum upward and downward adjustment limits for transmission power; and They represent Time-based delivery routes The upward and downward adjustment states are indicated by "1" for an adjustment in transmission direction and "0" for no change. and These are the transmission lines within and outside the scheduling cycle. The maximum number of downward and upward adjustments.

[0279] The objective function of the Conditional Generative Adversarial Network (CGAN) for scene generation is:

[0280] ;

[0281] In the formula, This is a gradient penalty term; Represents real samples and generate samples Random sampling along a straight line.

[0282] In the traditional CGAN training process, the generator aims to improve the quality of generated samples. The output value in the discriminator, while the discriminator's target is to reduce the generated samples. Output values ​​in the discriminator and improve the real samples The output value in the discriminator, where the loss functions of the generator and discriminator are defined as follows:

[0283] ;

[0284] ;

[0285] Original GANs are prone to training difficulties and pattern collapse during training because the loss function of the discriminator D is equivalent to the JS divergence. When the generated samples differ too much from the real samples, gradient vanishing occurs, leading to training difficulties. To address this issue, this paper introduces Wasserstein distance into the discriminator loss function to measure the distance between generated and real samples, as defined below:

[0286] ;

[0287] In the formula, The joint distribution representing the optimal path, for Distribution and The distance between them; This indicates that all probability distributions can be represented Convert to probability distribution The set of feasible solutions In the middle, take an infimum; Represents real samples With generated samples The distance between them. Due to The solution is quite complex. In CGAN, the Kantorovich-Rubinstein dual form is usually used to describe the generation of samples. Compared with real samples The distance between them is expressed as:

[0288] ;

[0289] In the formula, It is the constant of the Lipschitz function; This indicates that the discriminator function satisfies the Lipschitz continuity condition of constant.

[0290] The Euclidean Hierarchical Clustering Method with Reduced Scenario (EAHCM) includes the following steps:

[0291] First, calculate the Euclidean distance and mean value between cluster members. ) and medoid )vector.

[0292] ;

[0293] ;

[0294] In the formula, It is a medoid vector. It is the mean vector of the cluster.

[0295] Then, EAHCM selects the option with the smallest distance from the other options. This makes the representative vector more similar to other members within the cluster, thus preserving the original temporal order of the data. The following formula represents the selection part of EAHCM, where... This is the representative vector for the corresponding cluster.

[0296] ;

[0297] After selecting a representative time period within the clusters, EAHCM measures the dissimilarity between the representative vectors of all adjacent clusters. Then, the two adjacent clusters with the highest similarity are merged. The following formula calculates the dissimilarity between two adjacent clusters.

[0298] ;

[0299] In the formula, and Clustering and The representative vector. In each iteration, EAHCM merges the two clusters, and this process continues until the optimal number of clusters is reached.

[0300] Finally, the scheduling model of the hydro-solar-storage power system was solved using mixed integer linear programming to obtain the optimized scheduling results.

[0301] In practice, solvers such as Cplex and Gurobi can be used to solve the problem.

[0302] To evaluate the effectiveness of the proposed two-stage stochastic optimization model for a hydro-solar-storage complementary system and DC transmission system considering photovoltaic uncertainties, this study selects a six-node system for case verification. This system comprises three thermal power generating units (total installed capacity 700MW), two cascade hydropower stations (total installed capacity 410MW), one photovoltaic power station (installed capacity 800MW), one pumped storage power station (installed capacity 100MW), four load nodes, and one DC transmission line. A schematic diagram of its structure can be found in [reference needed]. Figure 4 .

[0303] G1, G2, and G3, three thermal power plants, form the backbone of the power system and provide fundamental support for the hydro-solar-storage complementary power system. The cascade hydropower stations consist of upstream and downstream stations, with H1 being the upstream station. Power flow converges at six nodes, then undergoes converter operation before being transmitted via DC. Daily forecast data can be found here. Figure 5 The penalty for wasting light is 500 yuan / (MW•h), and no load loss is allowed.

[0304] Against the backdrop of uncertainties in photovoltaic (PV) power generation, and based on the objective of minimizing system operating costs, this paper explores the effects of configuring variable-speed pumped-storage power stations, optimizing DC power transmission, and the impact of PV uncertainties on the scheduling scheme of a hydro-PV-storage complementary power generation system. Three specific examples are used for in-depth analysis:

[0305] Example 1: Under the condition that the DC power transmission is the preset actual power plan, without considering pumped storage power stations, calculate the two-stage random optimization scheduling.

[0306] Example 2: Based on Example 1, a pumped storage power station is added to calculate the two-stage random optimal scheduling.

[0307] Example 3: Based on Example 2, the condition of a fixed daily total DC transmission volume is added to further optimize the power.

[0308] This system includes the configuration of thermal power units, cascade hydropower stations, pumped storage power stations, and various power lines and loads within the power system. The operating parameters and constraints for each piece of equipment are set as follows:

[0309] The technical and economic parameters of the three thermal power units included in this system are shown in Table 1 and Table 2, respectively:

[0310] Table 1 Technical Parameters

[0311]

[0312] Table 2 Economic Parameters

[0313]

[0314] This system includes two cascade hydropower stations, H1 and H2, whose parameters include hydraulic characteristics, power generation capacity, operational constraints, and reservoir scheduling characteristics, as detailed in Tables 3, 4, 5, and 6:

[0315] Table 3 Hydraulic characteristics

[0316]

[0317] Table 4 Power Generation Capacity

[0318]

[0319] Table 5 Operational Constraints

[0320]

[0321] The system is configured with 5 reversible pumped storage units, and their operating parameters and energy conversion characteristics are shown in Table 6:

[0322] Table 6 Reservoir Dispatch Characteristics

[0323]

[0324] This system comprises 7 transmission lines and 4 load nodes, and its electrical parameters and load requirements are shown in Tables 7 and 8:

[0325] Table 7 Electrical Parameters

[0326]

[0327] Table 8 Load Demand

[0328]

[0329] Modeling and solving were performed for three cases respectively, and the total system cost comparison is shown in Table 9. The operating cost includes the power generation energy consumption cost, no-load cost and start-up and shutdown cost of thermal power units, as well as the start-up and shutdown cost of pumped storage units.

[0330] Table 9 Comparison of Total System Costs

[0331]

[0332] In hydro-solar hybrid dispatching, coordinating power generation adjustments with pumped-storage power stations can reduce system dispatching costs and improve renewable energy utilization efficiency to some extent. When combined with DC power transmission for regulation, system operating costs continue to decrease, validating the significant role of DC power optimization in improving system economics. This also highlights the substantial optimization potential in the coordinated operation of DC transmission and generator units, providing crucial evidence for further refining system dispatching strategies.

[0333] Please see Figure 6 The diagram shows a structural schematic of a computer device provided in an embodiment of this application. An embodiment of this application provides a computer device 400, including a processor 410 and a memory 420. The memory 420 stores a computer program executable by the processor 410. When the computer program is executed by the processor 410, it performs the method described above.

[0334] This application embodiment also provides a storage medium 430, on which a computer program is stored, and the computer program is executed by a processor 410 to perform the above method.

[0335] The storage medium 430 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0336] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. "A plurality of" means two or more, unless otherwise explicitly specified.

[0337] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0338] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0339] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0340] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0341] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0342] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0343] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A water-light storage complementary and DC external sending two-stage random optimization scheduling method, characterized in that, The method comprises the following steps: a water-light-storage complementary and DC transmission system model containing thermal power units, photovoltaic units, cascade hydropower, pumped storage units, static var compensators and DC transmission is established; a two-stage stochastic optimization scheduling model of the water-light-storage complementary and DC transmission system considering photovoltaic uncertainty is constructed, taking the minimization of two-stage system operation cost as an objective function, and taking node balance constraints, branch power flow constraints, thermal power unit constraints, photovoltaic output constraints, pumped storage power station constraints, cascade hydropower constraints, static var compensator constraints and branch operation constraints as constraint conditions; a conditional generative adversarial network (CGAN) is used to consider photovoltaic uncertainty and generate typical scenarios, and the generated scenarios are reduced through an Euclidean hierarchical clustering method (EAHCM); the two-stage stochastic optimization scheduling model is solved to obtain an optimal scheduling result; the objective function is: ； In the formula: is the first-stage decision variable, representing the unit commitment plan; is the feasible region of ; is the objective function of the first-stage optimization problem; is a random variable, representing the uncertainty of renewable energy generation and load; is the probability distribution of ; is the second-stage decision variable; is the objective function of the second-stage optimization problem; is the feasible region of ; represents the expected value; , , and represent the number of time periods, thermal power units, photovoltaic power stations and pumped storage power stations, respectively; is the time period number; , and are the thermal power unit, photovoltaic power station and pumped storage power station numbers, respectively; represents the power generation of the thermal power unit in the time period ; represents the no-load cost of the thermal power unit ; represents the start-stop state of the thermal power unit in the time period ; is the unit price of fuel; and represent the start-up and shutdown costs of the thermal power unit in the time period , respectively; represents the unit penalty factor for abandoned light; and are the day-ahead forecast value and actual power generation of the photovoltaic power station in the time period , respectively; and are the start-up and shutdown costs of a single variable-speed unit; and are the total number of start-up and shutdown units of the pumped storage power station in the time period . 2.The two-stage stochastic optimization scheduling method of water-light-storage complementary and DC external transmission according to claim 1, characterized in that, the branch power flow constraints are: ； ； ； ； ； In the formula: and These are the corresponding branches in the admittance matrix. The real and imaginary parts; for Time Node and The phase angle difference between them; and They are respectively Time Node and The voltage amplitude; and For the line Active and reactive power flow limits; and Represents a node Upper and lower limits of voltage amplitude. 3.The two-stage stochastic optimization scheduling method of water-light-storage complementary and DC external transmission according to claim 2, characterized in that, the branch power flow constraints are linearized: Assume Very small and each node voltage close to the rated voltage, so that , , , Then, there are: ； ； Will As one argument, the mathematical transformation of the non-linear voltage amplitude term , the non-linear term is converted into a linear term and a quadratic term: ； By substituting the above equation into the first two equations, the linearized network model with and is obtained. ； ； at this time: ； Assume that in the basic case condition, and the value of ; the loss is decomposed into voltage angle terms and voltage amplitude terms: ； According to the first order Taylor series expansion, the linearization of the term , is as follows: ； where is a function of as an argument, first transforms into a function, and then employs a first-order Taylor series expansion: ； Finally add The complete formula is By analogy, we have​ ； 。 4.The two-stage stochastic optimization scheduling method of water-light-storage complementary and DC external transmission according to claim 1, characterized in that, the cascade hydropower constraints are: ； ； ； ； ； ； ； ； ； ； ； ； ； ； ； In the formula: and They represent Cascade hydropower stations The meritorious and the ineffective contributions; express Periodic hydroelectric power station The operating status is indicated by "1" for power-on and "0" for power-off. and Cascade hydropower Maximum and minimum active power output; and Cascade hydropower Maximum and minimum reactive power output; Indicates cascade hydropower The power factor; and These represent cascade hydropower. The upward and downward climbing limits; It is a composite constant used to simplify power calculations, reflecting the combined effects of turbine efficiency, water density, and gravitational acceleration; Represented as cascade hydropower The efficiency of the turbine; Density of water, unit: kg / m³; Expresses gravitational acceleration, unit: m / s²; for Periodic hydroelectric power station Hydropower head; for Periodic hydroelectric power station The power generation flow rate; and Hydropower stations The maximum and minimum permissible power generation flow rates; and Hydropower stations The maximum and minimum permissible head for generating electricity; Indicates hydroelectric power station exist Storage capacity for a given period of time; for Periodic hydroelectric power station Inbound traffic; for Periodic hydroelectric power station The discharge flow rate; and These represent the cases considering water retention. Periodic hydroelectric power station direct upstream power station Power generation flow and water discharge flow; represents a hydropower plant time of water flow to the immediate downstream; represents time period hydropower plant natural inflow of the hydropower plant; and represents a hydropower plant maximum and minimum reservoir capacity allowed; and respectively, the beginning and end of dispatch reservoir capacity.

5. The two-stage stochastic optimization scheduling method of water-light-storage complementation and DC external transmission according to claim 1, characterized in that, the conditional generative adversarial network (CGAN) is used to consider photovoltaic uncertainty and generate typical scenarios, comprising the following steps: a conditional generative adversarial network (CGAN) objective function is set; ； wherein is a gradient penalty term; denotes real samples and generated samples are sampled randomly on a straight line; loss functions of the generator and the discriminator of the conditional generative adversarial network (CGAN) are set; ； ； a Wasserstein distance is introduced into the discriminator loss function to measure the distance between generated samples and real samples, and the definition is as follows: ； where the joint distribution representing the optimal path, is the distribution and the distance between them; denotes the infimum over the set of all feasible solutions that transform the probability distribution into the probability distribution ; denotes the distance between the real samples and the generated samples ; since solving this is complex, the Kantorovich-Rubinstein duality is used to describe the distance between the generated samples and the real samples , which is given by:​ ； In the formula, is a Lipschitz function constant; represents that the discriminator function satisfies the Lipschitz continuity condition of the constant.

6. The two-stage stochastic optimization scheduling method of water-light-storage complementation and DC external transmission according to claim 5, characterized in that, after representative time periods are selected in the cluster, the EAHCM measures the dissimilarity between representative vectors of adjacent clusters; then, the two adjacent clusters with the highest similarity are merged; the dissimilarity of the two adjacent clusters is calculated as follows: First compute the Euclidean distance between cluster members and the mean and the center point medoid vector; wherein is the medoid vector, is the mean vector of the cluster; Then, the EAHCM selects the option with the smallest distance to the other options; the selection part of the EAHCM is given by the following equation, where is the representative vector of the corresponding cluster. ； the two-stage stochastic optimization scheduling model is solved, comprising the following steps: ； where and are the representative vectors of the clusters and respectively; at each iteration, EAHCM merges two clusters, and this process continues until the optimal number of clusters is reached. 7.The two-stage stochastic optimization scheduling method of water-light-storage complementary and DC external transmission according to claim 1, characterized in that, the two-stage stochastic optimization scheduling model of the water-light-storage complementary and DC transmission system considering photovoltaic uncertainty is solved by using a mixed integer linear programming method to obtain an optimal scheduling result. The system comprises the method according to any one of claims 1 to 7.

8. A water-light-storage complementary and DC external transmission two-stage stochastic optimization scheduling system, characterized in that, a system model modeling unit is configured to establish a water-light-storage complementary and DC transmission system model containing thermal power units, photovoltaic units, cascade hydropower, pumped storage units, static var compensators and DC transmission; a scheduling model modeling unit is configured to construct a two-stage stochastic optimization scheduling model of the water-light-storage complementary and DC transmission system considering photovoltaic uncertainty, taking the minimization of two-stage system operation cost as an objective function, and taking node balance constraints, branch power flow constraints, thermal power unit constraints, photovoltaic output constraints, pumped storage power station constraints, cascade hydropower constraints, static var compensator constraints and branch operation constraints as constraint conditions; a scenario generation unit is configured to use a conditional generative adversarial network (CGAN) to consider photovoltaic uncertainty and generate typical scenarios, and reduce the generated scenarios through an Euclidean hierarchical clustering method (EAHCM); a solving unit is configured to solve the two-stage stochastic optimization scheduling model to obtain an optimal scheduling result. ​ 9. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1-7.

10. A storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the method of any one of claims 1-7.

Images