A Distributionally Robust Chance-Constrained Scheduling Method and System for a Multi-Energy Complementary Power Generation System

Abstract

The present invention relates to the technical field of power dispatching, and particularly relates to a distributed robust chance-constrained dispatching method and system for a multi-energy complementary power generation system. The method includes: establishing an objective function with the maximum new energy power generation as the goal, comprehensively considering the outputs of thermal power units, pumped storage units, as well as wind power and photovoltaic units; establishing the operating constraint conditions of the multi-energy complementary power generation system; introducing a distributed robust chance constraint, linearizing the non-linear constraints to characterize the uncertainty of new energy output, and constructing a distributed robust chance-constrained peak shaving dispatching model for the multi-energy complementary power generation system based on the objective function and the constraint conditions; using the mixed integer linear programming method and combining with a solver to solve the dispatching model, obtaining the optimal output allocation for each time period, and obtaining the dispatching method. Through the present invention, the influence of the randomness and volatility of new energy output on the peak shaving dispatching of the power system is effectively solved, and the new energy consumption capacity and operating stability of the multi-energy complementary power generation system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power dispatching, and specifically relates to a distributed robust chance-constrained dispatching method and system for a multi-energy complementary power generation system. Background Art

[0002] With the proposal of the "dual carbon" goal, building a clean and low-carbon power system is a necessary condition for achieving sustainable development. The multi-energy complementary power generation system combines new energy with local conditions, has huge energy potential, and is crucial for the clean and low-carbon transformation of the power system. After large-scale grid connection of new energy mainly based on wind power and photovoltaic power, the randomness and volatility of its own output greatly increase the probability of new energy curtailment in the system. Through the combined and coordinated dispatching of thermal power and pumped storage with new energy, the grid connection and consumption of new energy can be effectively promoted.

[0003] The output of new energy is often difficult to accurately predict and has obvious uncertainty, which in turn affects the peak shaving ability of the system. Stochastic optimization and robust optimization are two commonly used methods to deal with the uncertainty of new energy output. However, stochastic optimization depends on the assumed probability distribution, and in fact, it is difficult to accurately obtain the probability distribution of new energy output. Robust optimization completely ignores the probability information and directly uses the deterministic uncertainty set of new energy output, which will lead to the over-conservative decision-making of the system's peak shaving dispatching. Summary of the Invention

[0004] The present invention provides a distributed robust chance-constrained dispatching method and system for a multi-energy complementary power generation system, thus effectively solving the problems pointed out in the background art.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A distributed robust chance-constrained dispatching method for a multi-energy complementary power generation system, comprising:

[0007] Establish an objective function with the maximum new energy power generation as the goal, and comprehensively consider the output of thermal power units, pumped storage units, and wind power and photovoltaic units;

[0008] Establish the operating constraint conditions of the multi-energy complementary power generation system, including the constraint conditions of the operating characteristics of thermal power units and pumped storage units, the constraint conditions of system peak shaving, and the constraint conditions considering the stability of DC external transmission;

[0009] Introduce a distributed robust chance constraint, linearize the non-linear constraint, characterize the uncertainty of new energy output, and construct a distributed robust chance-constrained peak shaving dispatching model for the multi-energy complementary power generation system with the objective function and constraint conditions;

[0010] Using the mixed-integer linear programming method, combined with a solver to solve the scheduling model, the optimal output allocation of thermal power units, pumped-storage units and new energy units in each time period is obtained, and a distributionally robust chance-constrained peak shaving scheduling method for the multi-energy complementary power generation system is obtained.

[0011] Furthermore, the establishment of the objective function for peak shaving scheduling of the multi-energy complementary power generation system includes:

[0012]

[0013] Among them, is the new energy power generation; is the number of time periods in the scheduling cycle; and are the wind power generation and photovoltaic power generation in time period respectively.

[0014] Furthermore, the establishment of the constraint conditions for the operating characteristics of thermal power units and pumped-storage units includes:

[0015] Lower and upper power limits of thermal power units:

[0016]

[0017] Among them, and are the lower and upper power generation limits of the th thermal power unit respectively; is the power generation of the thermal power unit in time period; is time period of the thermal power unit operating status;

[0018] Power ramp constraint of thermal power units:

[0019]

[0020] Among them, and are the maximum upper and lower ramp rates of the thermal power unit respectively; and are time period of the thermal power unit start and stop operation status respectively;

[0021] Start-stop constraint of thermal power units:

[0022]

[0023]

[0024] Among them, and are respectively the start-up and shutdown state variables of the th thermal power unit in the time period; and are respectively the minimum start-up and shutdown times of the th thermal power unit; represents the maximum number of start-ups and shutdowns of the th thermal power unit within a day;

[0025] Operating state constraints of thermal power units:

[0026]

[0027] Among them, represents the operating state of the thermal power unit in the time period; and respectively represent the start-up state and shutdown state of the thermal power unit in the time period;

[0028] Storage capacity constraints of pumped storage power stations:

[0029]

[0030] Among them, is the water storage volume of the upper reservoir in the time period; and are respectively the minimum and maximum storage capacities of the upper reservoir; and are respectively the flow rates of the th pumped storage unit under power generation and pumping conditions in the time period; is the water level control target at the end of the scheduling period of the upper reservoir; is the number of pumped storage units;

[0031] Start-up and shutdown constraints of pumped storage units:

[0032]

[0033]

[0034]

[0035]

[0036] Among them, and are respectively the 0-1 state variables of the operating states of the th pumped-storage unit under power generation and pumping conditions; and are respectively the 0-1 action variables of the start-up and shutdown actions of the th pumped-storage unit under power generation conditions; and are respectively the 0-1 action variables of the start-up and shutdown actions of the th pumped-storage unit under pumping conditions; and are respectively the maximum number of start-stop times within a day of the th pumped-storage unit under power generation and pumping conditions;

[0037] Power constraint of pumped-storage unit:

[0038]

[0039]

[0040] Among them, and are respectively the lower power limit and the upper power limit of the th pumped-storage unit under power generation conditions; is the upper power limit of the th pumped-storage unit under pumping conditions;

[0041] Power / water conversion efficiency constraint of pumped-storage power station:

[0042]

[0043]

[0044]

[0045] Among them, is the power generation power of the nth pumped-storage unit at time t, is the pumping power of the nth pump at time t, and are respectively the power generation and pumping efficiencies of the th pumped-storage unit; and are respectively the maximum and minimum water discharge volumes in the power generation state; is the maximum water pumping volume in the pumping state.

[0046] Furthermore, the establishment of the constraint conditions for system peak shaving includes:​​​

[0047]

[0048] Among them, is the DC power transmission power of the system during the time period; and is the wind power and photovoltaic power generation during the time period; is the power generation of thermal power units during the time period; is the power generation of pumped storage units during the time period; is the pumping power of pumped storage units during the time period; is the upper limit of the transmission capacity of the DC channel; is the receiving-end load curve; and are slack variables, representing the positive and negative deviations of the sending-end transmission curve matching the receiving-end load curve; is the matching deviation coefficient.

[0049] Furthermore, the establishment of the constraint conditions considering the DC external transmission stability includes:

[0050] DC constant operation duration constraint:

[0051]

[0052]

[0053]

[0054] Among them, is a 0-1 variable, indicating whether the DC power is adjusted during the time period; represents the minimum DC constant operation duration;

[0055] Sending-end system stability constraint:

[0056]

[0057] Among them: is the operating state of the th thermal power unit during the time period; and are respectively the operating states of the th pumped storage unit in the power generation and pumping conditions during the time period; is the number of thermal power units; is the number of pumped storage units; The minimum number of units that need to be started up for thermal power units and pumped-storage units in each time period.

[0058] Furthermore, a distributionally robust chance constraint is introduced. The Wasserstein distance is used to describe the probability distribution fuzzy set of new energy power prediction, and the nonlinear constraints are linearized to characterize the uncertainty of new energy output, including:

[0059] New energy output chance constraint:

[0060]

[0061]

[0062] Among them, is the probability operator; is the planned dispatching wind power output; is the wind power predicted power; is the risk tolerance of wind power output; is the planned dispatching PV output power; is the PV predicted power; is the risk tolerance of PV output;

[0063] The chance constraint is rewritten in the general form:

[0064]

[0065] Among them, is and the vector composed of; is and the vector composed of; , are respectively and the coefficient vectors of; is the confidence level;

[0066] Furthermore, a fuzzy set of new energy predicted power probability distribution based on the Wasserstein distance is constructed :

[0067]

[0068]

[0069]

[0070] Among them, is the empirical distribution of known samples; is the a sample Dirac measure of is the number of samples; the fuzzy set is a set of a series of distributions close to the empirical distribution ; is a random probability distribution; is the support set of the random variable and can be taken as ; is the set composed of all probability distributions within the support set; is the radius of the Wasserstein ball; is the distance between two probability distributions and ; is and joint probability distribution with respect to the marginals and ; is an arbitrary norm on ;

[0071] Based on the fuzzy set , construct the distributionally robust chance-constrained form:

[0072]

[0073] Adopt conditional value-at-risk approximation and saddle point theorem for linearization:

[0074]

[0075] where is a random variable; , , are dual variables generated during the transformation process; is the infinity norm.

[0076] Furthermore, using the mixed-integer linear programming method, combined with a solver to solve the scheduling model, obtain the optimal output allocation of thermal power units, pumped-storage units and new energy units in each time period, and obtain the distributionally robust chance-constrained peak shaving scheduling method for the multi-energy complementary power generation system, including:

[0077] Linearize the non-linear constraints involved in the distributionally robust chance constraints, and transform the uncertainty constraints of the new energy output from non-linear to linear constraints;

[0078] Use the solver to input the new energy output prediction data and model parameters, optimize the objective function value, and solve the optimal scheduling plan that meets the constraint conditions;

[0079] Output the optimal power generation allocation of thermal power units, pumped-storage units and new energy units in each time period, as well as the matching optimization results of DC transmission power and the receiving-end load curve;

[0080] Adjust the parameters of the distributionally robust chance constraint, analyze the new energy accommodation capacity and dispatching optimization effect under different conditions, and finally form the distributionally robust chance constraint peak shaving dispatching method for the multi-energy complementary power generation system.

[0081] Furthermore, use Yalmip to call the Gurobi solver to solve the scheduling model.

[0082] A distributionally robust chance constraint scheduling system for a multi-energy complementary power generation system, the system includes:

[0083] An objective function establishment module, which establishes an objective function, aiming at maximizing the new energy power generation, and comprehensively considers the output of thermal power units, pumped-storage units, and wind and photovoltaic units;

[0084] A constraint condition construction module, which establishes the operation constraint conditions of the multi-energy complementary power generation system, including the constraint conditions of the operation characteristics of thermal power units and pumped-storage units, the constraint conditions of system peak shaving, and the constraint conditions considering the stability of DC external transmission;

[0085] A scheduling model construction module, which introduces a distributionally robust chance constraint, linearizes the non-linear constraints, and characterizes the uncertainty of new energy output, and constructs a distributionally robust chance constraint peak shaving dispatching model for the multi-energy complementary power generation system with the objective function and constraint conditions;

[0086] A scheduling model solving module, which uses the mixed integer linear programming method, combines with the solver to solve the scheduling model, obtains the optimal output allocation of thermal power units, pumped-storage units and new energy units in each time period, and obtains the distributionally robust chance constraint peak shaving dispatching method for the multi-energy complementary power generation system.

[0087] Furthermore, the scheduling model solving module includes:

[0088] A linear transformation unit, which linearizes the non-linear constraints involved in the distributionally robust chance constraint, and transforms the uncertainty constraint of the new energy output from non-linear to linear;

[0089] A solution solving unit, which uses the solver, inputs the new energy output prediction data and model parameters, optimizes the objective function value, and solves the optimal scheduling solution that satisfies the constraint conditions;

[0090] A solution output unit, which outputs the optimal power generation allocation of thermal power units, pumped-storage units and new energy units in each time period, as well as the matching optimization results of DC transmission power and the receiving-end load curve;

[0091] The solution generation unit adjusts the parameters of the distributionally robust chance constraint, analyzes the new energy consumption capacity and the scheduling optimization effect under different conditions, and finally forms the distributionally robust chance constraint peak shaving scheduling method for the multi-energy complementary power generation system.

[0092] Through the technical solution of the present invention, the following technical effects can be achieved:

[0093] It effectively solves the influence of the randomness and volatility of new energy output on the peak shaving scheduling of the power system, and improves the new energy consumption capacity and operation stability of the multi-energy complementary power generation system. Description of the Drawings

[0094] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0095] Figure 1 It is a schematic flow chart of a distributionally robust chance constraint scheduling method for a multi-energy complementary power generation system;

[0096] Figure 2 It is a diagram of the output results of each part of the multi-energy complementary power generation system in the embodiment of the present invention;

[0097] Figure 3 It is a diagram of the comparison results of the system transmission curve and the receiving-end load curve in the embodiment of the present invention;

[0098] Figure 4 It is a diagram of the new energy output results under different parameter values in the distributionally robust chance constraint in the embodiment of the present invention. Detailed Embodiments

[0099] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0100] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0101] Embodiment 1

[0102] As Figure 1As shown in the figure, the present invention provides a distributionally robust chance-constrained scheduling method for a multi-energy complementary power generation system, and the method includes:

[0103] S1: Establish an objective function with the maximum new energy power generation as the goal, and comprehensively consider the output of thermal power units, pumped storage units, wind power units, and photovoltaic units;

[0104] Specifically, by establishing an objective function, with maximizing the new energy power generation as the core, optimize the output coordination of various types of units (thermal power, pumped storage, wind power, photovoltaic) in the multi-energy complementary power generation system, realize the maximized utilization of new energy power generation, reduce the wind and light curtailment rates, and at the same time take into account the system economy and stability, providing an optimization objective basis for the construction of the subsequent scheduling model.

[0105] S2: Establish the operating constraint conditions of the multi-energy complementary power generation system, including the constraint conditions of the operating characteristics of thermal power units and pumped storage units, the constraint conditions of system peak shaving, and the constraint conditions considering the stability of DC external power transmission;

[0106] Specifically, through the setting of constraint conditions, clarify the operating boundaries and scheduling rules of thermal power units and pumped storage units, and at the same time take into account the peak shaving requirements of the system and the stability of DC external power transmission, so as to ensure the actual operability of system operation while achieving the optimization goal (such as maximizing new energy power generation), and avoid unrealistic results or operation risks caused by over-optimization.

[0107] S3: Introduce distributionally robust chance constraints, linearize the non-linear constraints, characterize the uncertainty of new energy output, and construct a distributionally robust chance-constrained peak shaving scheduling model for the multi-energy complementary power generation system with the objective function and constraint conditions;

[0108] Specifically, the purpose of introducing distributionally robust chance constraints and linearizing non-linear constraints is to effectively characterize the uncertainty of new energy output when constructing a peak shaving scheduling model, and at the same time ensure the solvability and solution efficiency of the model. Through distributionally robust chance constraints, the randomness and volatility of new energy output are incorporated into the optimization model, avoiding the deficiencies of stochastic optimization methods that rely entirely on precise probability distribution assumptions and overcoming the problem of excessive conservatism of traditional robust optimization methods, thereby improving the consumption capacity and scheduling flexibility of new energy while ensuring the operation safety of the system.

[0109] S4: Use the mixed integer linear programming method, combine with a solver to solve the scheduling model, obtain the optimal output allocation of thermal power units, pumped storage units, and new energy units in each time period, and obtain a distributionally robust chance-constrained peak shaving scheduling method for the multi-energy complementary power generation system.

[0110] Specifically, based on the established objective function and constraint conditions, the optimal output allocation schemes of thermal power units, pumped-storage units, and new energy units in each time period are obtained, so as to maximize the economy, reliability, and dispatching efficiency of the system operation. By solving the dispatching model, the theoretically optimal dispatching strategy can be transformed into an actual operable dispatching scheme, and at the same time, the effectiveness of the distributionally robust chance-constrained model is verified. Finally, an optimal peak shaving dispatching method for the multi-energy complementary power generation system is formed to improve the new energy consumption and system stability.

[0111] Through the present invention, the influence of the randomness and volatility of new energy output on the peak shaving dispatching of the power system is effectively solved, and the new energy consumption capacity and operation stability of the multi-energy complementary power generation system are improved.

[0112] The embodiment of the present invention includes 10 thermal power units, 6 pumped-storage units, 1 wind farm, and 1 photovoltaic power station, which will be described below in combination with specific application scenarios.

[0113] As a preference of the above embodiment, an objective function for peak shaving dispatching of the multi-energy complementary power generation system is established, including:

[0114]

[0115] Among them, is the new energy generation power; is the number of time periods in the dispatching cycle; and are the generation powers of wind power and photovoltaic power in the time period, respectively.

[0116] As a preference of the above embodiment, constraint conditions for the operating characteristics of thermal power units and pumped-storage units are established, including:

[0117] Lower and upper limits of thermal power unit power:

[0118]

[0119] Among them, and are the lower limit and upper limit of the generation power of the th thermal power unit, respectively; is the generation power of the thermal power unit in the time period; is the operating state of the thermal power unit in the

[0120] Power ramp rate constraint of thermal power units:

[0121]

[0122] Among them, and are respectively the maximum upward and downward ramping rates of the thermal power unit ; and are respectively the start - stop operation status of the thermal power unit during the

[0123] Start - stop constraints of the thermal power unit:

[0124]

[0125]

[0126] Among them, , are respectively the start - stop status variables of the th thermal power unit during the , are respectively the minimum start - up and shut - down times of the th thermal power unit; represents the maximum number of start - stops of the th thermal power unit within a day;

[0127] Operation status constraints of the thermal power unit:

[0128]

[0129] Among them, represents the operation status of the thermal power unit during the , respectively represent the start - up status and shut - down status of the thermal power unit during the

[0130] Storage capacity constraints of the pumped - storage power station:

[0131]

[0132] Among them, is the water storage volume of the upper reservoir during the and are respectively the minimum and maximum storage capacities of the upper reservoir; and are respectively the flow rates of the th pumped - storage unit under power generation and pumping conditions during the is the water level control target at the end of the scheduling period of the upper reservoir; is the number of pumped-storage units;

[0133] Start-stop constraints of pumped-storage units:

[0134]

[0135]

[0136]

[0137]

[0138] Among them, and are respectively the 0-1 state variables of the th pumped-storage unit in the power generation and pumping operation states during the and are respectively the 0-1 action variables of the th pumped-storage unit starting and stopping during the power generation operation in the and are respectively the 0-1 action variables of the th pumped-storage unit starting and stopping during the pumping operation in the and are respectively the th pumped-storage unit's maximum start-stop times within a day during the power generation and pumping operations;

[0139] Power constraints of pumped-storage units:

[0140]

[0141]

[0142] Among them, and are respectively the th pumped-storage unit's lower power limit and upper power limit during the power generation operation; is the th pumped-storage unit's upper power limit during the pumping operation;

[0143] Power / water volume conversion efficiency constraints of pumped-storage power stations:

[0144]

[0145]

[0146]

[0147] Among them, is the power generation power of the nth pumped-storage unit at time t, is the pumping power of the nth pump at time t, , are respectively the power generation and pumping efficiencies of the th pumped-storage unit; , are respectively the maximum and minimum water discharge volumes in the power generation state; is the maximum pumping volume in the pumping state.

[0148] As Figure 2 shown, the output conditions of 10 thermal power units and 6 pumped-storage units are included in the example of the present invention. When the new energy power generation is relatively high, the output of thermal power units is less and the pumped-storage units are used for pumping to maximize the consumption of excess new energy output; when the new energy output is relatively low, the output of thermal power units increases and the pumped-storage units are used for power generation to jointly provide support for the system power transmission.

[0149] The parameters of the thermal power units are shown in Table 1:

[0150] Table 1 Parameters of Thermal Power Units

[0151] Unit number 1 2 3 4 5 6 7 8 9 10 Upper output limit (MW) 60 60 60 60 60 60 60 60 60 60 Lower output limit (MW) 24 24 24 24 24 24 24 24 24 24

[0152] The parameters of the pumped-storage power station are shown in Table 2:

[0153] Table 2 Parameters of Pumped-Storage Power Station

[0154] <![CDATA[Minimum storage capacity of the upper reservoir (m 3 )]]> 0 <![CDATA[Maximum storage capacity of the upper reservoir (m 3 )]]> 2000 <![CDATA[Final storage capacity (m 3 ).]]> 1000 <![CDATA[Initial reservoir capacity (m 3 ).]]> 1000 Maximum number of starts and stops under generating condition of unit (times) 4 Maximum number of starts and stops under pumping condition of unit (times) 4

[0155] The parameters of the pumped-storage units are shown in Table 3:

[0156] Table 3 Parameters of Pumped-Storage Units

[0157] Unit number 1 2 3 4 5 6 Minimum generating power (MW) 20 20 20 20 20 20 Maximum generating power (MW) 50 50 50 50 50 50 <![CDATA[Minimum water discharge (m 3 )]]> 30 30 30 30 30 30 <![CDATA[Maximum water discharge (m 3 )]]> 50 50 50 50 50 50 Maximum pumping power (MW) 50 50 50 50 50 50 <![CDATA[Maximum pumping volume (m 3 )]]> 60 60 60 60 60 60

[0158] As a preference of the above embodiment, the constraint conditions for system peak shaving are established, including:

[0159]

[0160] Among them, is the DC power transmission power of the system in the and are the wind power and photovoltaic power generation in the is the power generation power of the thermal power unit in the is the power generation power of the pumped-storage unit in the is the pumping power of the pumped-storage unit during the time period; is the upper limit of the DC channel transmission capacity; is the receiving-end load curve; and is the slack variable, representing the positive and negative deviations of the sending-end transmission curve matching the receiving-end load curve; is the matching deviation coefficient.

[0161] As Figure 3 shown, the embodiment of the present invention includes a system transmission curve and a receiving-end load curve. The overall trend of the transmission curve is consistent with the load curve, reflecting the peak shaving ability of the system.

[0162] As a preference of the above embodiment, constraint conditions considering the stability of DC external transmission are established, including:

[0163] DC constant operation duration constraint:

[0164]

[0165]

[0166]

[0167] Among them, is a 0-1 variable indicating whether the DC power is adjusted during the time period; represents the minimum DC constant operation duration;

[0168] Sending-end system stability constraint:

[0169]

[0170] Wherein: is the operating state of the th thermal power unit during the time period; , are respectively the operating states of the th pumped-storage unit in the power generation and pumping conditions during the time period; is the number of thermal power units; is the number of pumped-storage units; is the minimum number of units that need to be started up for thermal power units and pumped-storage units in each time period.

[0171] As a preference for the above embodiments, a distributionally robust chance constraint is introduced. The Wasserstein distance is used to describe the probability distribution fuzzy set of new - energy power prediction, and the non - linear constraints are linearized to characterize the uncertainty of new - energy output, including:

[0172] New - energy output chance constraint:

[0173]

[0174]

[0175] where, is the probability operator; is the planned - dispatch wind - power output power; is the wind - power predicted power; is the risk - tolerance of wind - power output; is the planned - dispatch PV - power output power; is the PV - power predicted power; is the risk - tolerance of PV - power output;

[0176] The chance constraint is rewritten in the general form:

[0177]

[0178] where, is and composed of a vector, is and composed of a vector; , are respectively and coefficient vectors; is the confidence level;

[0179] Furthermore, a fuzzy set of new - energy predicted - power probability distribution based on the Wasserstein distance is constructed :

[0180]

[0181]

[0182]

[0183] where, is the empirical distribution of known samples; is the Dirac measure of the - th sample of the random variable, is the number of samples; the fuzzy set is a set of a series of distributions close to the empirical distribution ; is a random probability distribution; is a random variable whose support set can be taken as ; is the set composed of all probability distributions within the support set; is the radius of the Wasserstein ball; is the distance between two probability distributions and ; is and 's joint probability distribution with respect to the marginals and ; is an arbitrary norm on ;

[0184] Based on the fuzzy set , construct the distributionally robust chance-constrained form:

[0185]

[0186] Adopt the conditional value-at-risk approximation and the saddle point theorem for linearization:

[0187]

[0188] where is a random variable; , , are the dual variables generated during the transformation process; is the infinity norm.

[0189] This model constructs a random variable fuzzy set based on the historical data of new energy predicted output. Taking the new energy predicted power generation as the mean and 5% of the mean as the variance, 100 groups of random samples are generated based on the normal distribution as the sample set of the new energy predicted power generation.

[0190] The new energy predicted power generation is shown in Table 4:

[0191] Table 4 New Energy Predicted Power Generation

[0192] Time Wind energy predicted generating power (MW) Photovoltaic predicted generating power (MW) 1 560 0 2 640 0 3 720 0 4 640 0 5 560 0 6 320 0 7 160 10 8 80 20 9 80 40 10 160 80 11 120 120 12 80 160 13 120 170 14 160 60 15 240 120 16 320 80 17 400 40 18 480 20 19 520 0 20 600 0 21 480 0 22 560 0 23 640 0 24 720 0

[0193] Such as Figure 4As shown in the figure, on the one hand, as the confidence level decreases, the chance constraint becomes more relaxed for meeting the requirements of new energy generation power, thereby reducing the conservativeness of the scheduling decision and correspondingly increasing the new energy generation power. On the other hand, as the radius of the Wasserstein ball decreases, the new energy generation power also increases accordingly. This is because the smaller the radius, the smaller the distribution set of the random variable, and the closer the probability distribution is to the empirical distribution of the sample, thus reducing the conservativeness of the distributionally robust chance constraint model. Therefore, in the actual operation of the multi - energy complementary power generation system, the new energy generation power can be increased by setting appropriate confidence levels and Wasserstein ball radii, improving the flexibility of the peak - shaving scheduling decision.

[0194] Preferably, as in the above - mentioned embodiment, using the mixed - integer linear programming method and combining with a solver to solve the scheduling model, the optimal output power distribution of thermal power units, pumped - storage units and new - energy units in each time period is obtained, and a distributionally robust chance - constrained peak - shaving scheduling method for the multi - energy complementary power generation system is obtained, including:

[0195] Linearize the non - linear constraints involved in the distributionally robust chance constraint, and transform the uncertainty constraint of the new - energy output from non - linear to linear.

[0196] Use the solver to input the new - energy output prediction data and model parameters, optimize the objective function value, and solve the optimal scheduling plan that meets the constraint conditions.

[0197] Output the optimal power generation power distribution of thermal power units, pumped - storage units and new - energy units in each time period, as well as the matching optimization result of the DC transmission power and the receiving - end load curve.

[0198] Adjust the parameters of the distributionally robust chance constraint, analyze the new - energy consumption capacity and scheduling optimization effect under different conditions, and finally form the distributionally robust chance - constrained peak - shaving scheduling method for the multi - energy complementary power generation system.

[0199] Specifically, first, linearize the non-linear constraints involved in the distributionally robust chance constraint, transforming the uncertainty of new energy output from non-linear constraints into linear constraints to ensure that the model can be efficiently solved using the mixed-integer linear programming (MILP) method. The linearization simplifies the uncertainty description through mathematical transformation or dual methods, making the model easier to solve while retaining the accuracy of uncertainty characterization. Then, use a solver (such as Yalmip calling Gurobi), input the new energy output prediction data and related model parameters, establish the objective function and operating constraint conditions. During the solution process, optimize the objective function value to ensure the maximization of new energy generation power while satisfying the operating constraints of thermal power units, pumped-storage units, and new energy units. Obtain the global optimal solution quickly through the efficient algorithm of the solver. Subsequently, output the optimal power generation allocation results of thermal power units, pumped-storage units, and new energy units in each time period, and further calculate the matching optimization results of DC transmission power and the receiving-end load curve to verify the practical feasibility of the dispatching scheme in system operation. Finally, by adjusting the key parameters of the distributionally robust chance constraint (such as the confidence level and Wasserstein distance parameter), analyze the new energy consumption capacity and dispatching optimization effect of the system under different conditions. Through this process, further evaluate the sensitivity and robustness of the model, and finally form the peak shaving dispatching method with distributionally robust chance constraint for the multi-energy complementary power generation system, providing reliable technical support for system optimal operation and efficient utilization of new energy.

[0200] As a preference of the above embodiment, use Yalmip to call the Gurobi solver to solve the scheduling model.

[0201] Specifically, Gurobi is a commercially excellent solver, especially suitable for mixed-integer linear programming (MILP) problems, with the advantages of high solving efficiency and strong result reliability, which very well meets the solving requirements of the scheduling model of the present invention. After the scheduling model is linearized, Gurobi can quickly and efficiently handle large-scale linear constraint problems, ensuring the accuracy of the solution and meeting the real-time requirements of power dispatching problems. Combined with Yalmip, a MATLAB-based modeling tool, it is convenient to construct complex objective functions and constraint conditions, and achieve the efficient solution of the model by calling Gurobi. In addition, the parameter tuning function of Gurobi can optimize the solution process according to the model characteristics, improving the convergence speed and result quality, while Yalmip provides flexibility for the sensitivity analysis and expansion of the model. This combination can significantly improve the model solving efficiency and effect, providing reliable technical support for new energy consumption and system dispatching optimization.

[0202] Embodiment 2

[0203] Based on the same inventive concept as a distributionally robust chance-constrained scheduling method for a multi-energy complementary power generation system in the foregoing embodiments, the present invention also provides a distributionally robust chance-constrained scheduling system for a multi-energy complementary power generation system, which includes:

[0204] An objective function establishment module that establishes an objective function, aiming at maximizing the new energy power generation, and comprehensively considering the outputs of thermal power units, pumped storage units, and wind power and photovoltaic units;

[0205] A constraint condition construction module that establishes the operating constraint conditions of the multi-energy complementary power generation system, including the constraint conditions of the operating characteristics of thermal power units and pumped storage units, the constraint conditions of system peak shaving, and the constraint conditions considering the stability of DC external power transmission;

[0206] A scheduling model construction module that introduces distributionally robust chance constraints, linearizes the non-linear constraints, and characterizes the uncertainty of new energy output, and constructs a distributionally robust chance-constrained peak shaving scheduling model for the multi-energy complementary power generation system with the objective function and constraint conditions;

[0207] A scheduling model solving module that uses the mixed integer linear programming method and combines with a solver to solve the scheduling model, obtains the optimal output allocation of thermal power units, pumped storage units, and new energy units in each time period, and obtains a distributionally robust chance-constrained peak shaving scheduling method for the multi-energy complementary power generation system.

[0208] The above scheduling system in the present invention can effectively implement the distributionally robust chance-constrained scheduling method for the multi-energy complementary power generation system, and the technical effects that can be achieved are as described in the foregoing embodiments, which will not be elaborated here.

[0209] As a preference of the above embodiment, the scheduling model solving module includes:

[0210] A linear transformation unit that linearizes the non-linear constraints involved in the distributionally robust chance constraints, and transforms the uncertainty constraints of new energy output from non-linear to linear constraints;

[0211] A solution solving unit that uses a solver, inputs the new energy output prediction data and model parameters, optimizes the objective function value, and solves the optimal scheduling solution that satisfies the constraint conditions;

[0212] A solution output unit that outputs the optimal power generation power distribution of thermal power units, pumped storage units, and new energy units in each time period, as well as the matching optimization result of the DC transmission power and the receiving-end load curve;

[0213] A solution generation unit that adjusts the parameters of the distributionally robust chance constraints, analyzes the new energy consumption capacity and scheduling optimization effects under different conditions, and finally forms a distributionally robust chance-constrained peak shaving scheduling method for the multi-energy complementary power generation system.

[0214] Similarly, for the above optimization solutions of the system, the corresponding optimization effects of the methods in the first embodiment can also be respectively achieved, and details are not described herein again.

[0215] Although the present application has been described in conjunction with specific features and their embodiments, it is obvious that various modifications and combinations can be made without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely exemplary illustrations of the present application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application.

[0216] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A method for scheduling a multi-energy complementary power generation system with distributed robust opportunity constraints, characterized in that: include: Establish an objective function, taking the maximum power generation of renewable energy as the goal, and comprehensively consider the output of thermal power units, pumped storage units, wind power units, and photovoltaic units; Establish the operating constraints of the multi-energy complementary power generation system, including the operating characteristics of thermal power units and pumped storage units, the peak load constraints of the system, and the constraints considering the stability of DC transmission; The distributed robustness opportunity constraint is introduced to linearize the nonlinear constraint, characterize the uncertainty of the output of new energy, and build a distributed robustness opportunity constraint peak-shaving scheduling model for the multi-energy complementary power generation system based on the objective function and constraint conditions; The distributed robust chance constraint is introduced, the Wasserstein distance is used to describe the probability distribution fuzzy set of new energy power prediction, and the nonlinear constraint is linearized to characterize the uncertainty of new energy output, including: Constraints on renewable energy output opportunities: in, is a probability operator; To plan and dispatch wind power output; Power forecasting for wind power; The risk tolerance of wind power output; To plan and dispatch photovoltaic output power; Power prediction for photovoltaics; The risk tolerance of photovoltaic power output; Rewriting the chance constraint in general form: in, for and The vector composed of for and The vector composed of , They are and The coefficient vector of ; is the confidence level; Further construct the fuzzy set of new energy prediction power probability distribution based on Wasserstein distance : in, is the empirical distribution of known samples; is the random variable Samples The Dirac measure of is the number of samples; fuzzy set To be close to the empirical distribution A collection of distributions of ; is a random probability distribution; is a random variable The support set of can be taken as ; is the set of all probability distributions in the support set; is the radius of the Wasserstein sphere; For two probability distributions and The distance between for and About Margin and The joint probability distribution of For Any norm on ; Based on fuzzy sets, construct the distribution of robust chance constraints: The linearization is performed using the conditional value-at-risk approximation and the saddle point theorem: in, is a random variable; , , is the dual variable generated in the transformation process; is the infinite norm; The mixed integer linear programming method is used in combination with the solver to solve the scheduling model, and the optimal output distribution of thermal power units, pumped storage units and new energy units in each time period is obtained, and the peak-shaving scheduling method with distributed robust opportunity constraints for multi-energy complementary power generation systems is obtained.

2. The method for scheduling a multi-energy complementary power generation system with distributed robust opportunity constraints according to claim 1, characterized in that: Establish the objective function of peak load dispatching of multi-energy complementary power generation system, including: in, Power generation for renewable energy; The number of time periods in the scheduling cycle.

3. The multi-energy complementary power generation system distributed blue opportunistic constraint scheduling method according to claim 1 is characterized in that: Establish constraints on the operating characteristics of thermal power units and pumped storage units, including: Power upper and lower limits of thermal power units: in, and Respectively The lower and upper limits of power generation of each thermal power unit; For thermal power units exist The power generated during the period; for Thermal power units The operating status of Power ramp constraints of thermal power units: in, and Thermal power units Maximum up and down climbing rate; and They are Thermal power units Start and stop operation status; Thermal power unit start and stop constraints: in, , Respectively Minimum start-up and shutdown time of each thermal power unit; Indicates Maximum number of starts and stops per day for each thermal power unit; Constraints on the operating status of thermal power units: in, express Thermal power units The operating status of Storage capacity constraints of pumped storage power stations: in, for The amount of water stored in the reservoir during the period; and are the minimum and maximum storage capacities of the upper reservoir, respectively; and They are Time period The flow rate of the pumped storage unit under power generation and pumping conditions; It is the water level control target at the end of the dispatching period of the upper reservoir; is the number of pumped storage units, Pumped storage unit start and stop constraints: in, , They are Time period 0-1 state variables of the operating status of the pumped storage unit under power generation and pumping conditions; , They are Time period 0-1 action variable of the start-up and shutdown actions of the pumped storage unit under power generation conditions; , They are Time period 0-1 action variable of the start-up and shutdown actions of the pumped storage unit under pumping conditions; , Respectively The maximum number of starts and stops of a pumped storage unit in a day under power generation and pumping conditions; Power constraints of pumped storage units: in, , Respectively The lower and upper power limits of each pumped storage unit under power generation conditions; For the The upper power limit of each pumped storage unit under pumping conditions; Power / water conversion efficiency constraints of pumped storage power stations: in, For the Pumped storage units in time The power generation capacity, For the Pumps at time The pumping power, , Respectively The power generation and pumping efficiency of the pumped storage units; , They are the maximum and minimum water discharge under power generation state respectively; It is the maximum pumping volume under pumping state.

4. The method for scheduling a multi-energy complementary power generation system with distributed robust opportunity constraints according to claim 1, characterized in that: Establish the constraints for system peak load regulation, including: in, for System DC transmission power during the period; and for Wind power and photovoltaic power generation power in different time periods; for Power generation capacity of thermal power units during the period; for Power generation capacity of pumped storage units during the time period; for Pumping power of pumped storage units during the period; The upper limit of the DC channel transmission capacity; is the receiving end load curve; and is a slack variable, which indicates the positive and negative deviation of the matching between the transmission curve at the sending end and the load curve at the receiving end; is the matching deviation coefficient.

5. The method for scheduling a multi-energy complementary power generation system with distributed robust opportunity constraints according to claim 1, characterized in that: Establish constraints that take into account the stability of DC transmission, including: DC constant operation time constraints: in, is a 0-1 variable indicating Whether the DC power is adjusted during the time period; Indicates the minimum DC constant operating time, for The system DC transmission power of the period, for The system DC transmission power of the period, The upper limit of the DC channel transmission capacity; Sending end system stability constraints: in: for Time period The operating status of the thermal power units; , They are Time period The operating status of the pumped storage units under power generation and pumping conditions; is the number of thermal power units; is the number of pumped storage units; The minimum number of thermal power units and pumped storage units that need to be started in each period.

6. The method for scheduling multi-energy complementary power generation system with distributed robust opportunity constraints according to claim 1, characterized in that: The mixed integer linear programming method is used in combination with the solver to solve the scheduling model, and the optimal output distribution of thermal power units, pumped storage units and new energy units in each time period is obtained, and the peak-shaving scheduling method with distributed robust opportunity constraints of multi-energy complementary power generation system is obtained, including: Linearizing the nonlinear constraints involved in the distributed robust opportunity constraints, and transforming the uncertainty constraints of the new energy output from nonlinear to linear constraints; Using the solver, input the forecast data of renewable energy output and model parameters, optimize the objective function value, and solve the optimal dispatching plan that meets the constraints; Output the optimal power distribution of thermal power units, pumped storage units and new energy units in each time period, as well as the matching optimization results of DC transmission power and receiving end load curve; The parameters of the distributed peak-loading opportunity constraints are adjusted, the new energy consumption capacity and scheduling optimization effects under different conditions are analyzed, and finally the distributed peak-loading opportunity constraint peak-loading scheduling method of the multi-energy complementary power generation system is formed.

7. The method for scheduling a multi-energy complementary power generation system with distributed robust opportunity constraints according to claim 6, characterized in that: The scheduling model is solved by using Yalmip to call the Gurobi solver.

8. A multi-energy complementary power generation system distributed robust opportunity constraint scheduling system, using the multi-energy complementary power generation system distributed robust opportunity constraint scheduling method as claimed in claim 1, characterized in that: The system comprises: The objective function establishment module establishes the objective function, taking the maximum power generation of new energy as the goal, and comprehensively considering the output of thermal power units, pumped storage units, wind power units, and photovoltaic units; Constraint construction module, which establishes the operating constraints of the multi-energy complementary power generation system, including the operating characteristics of thermal power units and pumped storage units, the system peak load constraints and the constraints considering the stability of DC transmission; The dispatch model construction module introduces the distributed robust opportunity constraint, linearizes the nonlinear constraint, characterizes the uncertainty of the output of new energy, and constructs the distributed robust opportunity constraint peak-shaving dispatch model of the multi-energy complementary power generation system based on the objective function and constraint conditions; The scheduling model solving module uses the mixed integer linear programming method and the solver to solve the scheduling model, obtains the optimal output distribution of thermal power units, pumped storage units and new energy units in each time period, and obtains the peak-shaving scheduling method with distributed robust opportunity constraints for multi-energy complementary power generation systems.

9. The multi-energy complementary power generation system distributed blue opportunistic constraint scheduling system according to claim 8, characterized in that: The scheduling model solving module includes: A linear transformation unit is used to linearize the nonlinear constraints involved in the distributed robust opportunity constraints, and transform the uncertainty constraints of the new energy output from nonlinear to linear constraints; The solution solving unit uses the solver to input the forecast data of new energy output and model parameters, optimize the objective function value, and solve the optimal scheduling solution that meets the constraints; The scheme output unit outputs the optimal power distribution of thermal power units, pumped storage units and new energy units in each time period, as well as the matching optimization results of DC transmission power and receiving end load curve; The scheme generation unit adjusts the parameters of the distributed robust opportunity constraints, analyzes the new energy consumption capacity and scheduling optimization effect under different conditions, and finally forms the distributed robust opportunity constrained peak-shaving scheduling method of the multi-energy complementary power generation system.

Images