A multi-objective game scheduling method and system for a distribution network operator

Abstract

The present invention discloses a multi-objective game scheduling method and system for a distribution network operator. The method includes the following steps: obtaining power consumption data; establishing a distribution network operator model with the goal of minimizing the grid operation cost; establishing a load aggregator cost model for a load aggregator composed of flexible resource users with economy as the goal; taking the established distribution network operator model as the upper-layer model and the established load aggregator cost model as the lower-layer model to construct a master-slave game model, and using an improved grey wolf algorithm to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the distribution network operator's day-ahead scheduling. By modeling flexible resource users and the grid operator, establishing a day-ahead scheduling problem with economy as the goal, and using an improved grey wolf algorithm for solution to obtain the optimal day-ahead time-of-use electricity price, the present invention can improve the power management level on the flexible resource user side and has good practical significance.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply and demand interaction, and particularly to a multi-objective game scheduling method and system for a distribution network operator. Background Art

[0002] Demand response is described as an effective method to induce power consumers to change their power demand from peak hours to off-peak hours within a day, and is an important part of modern power grids. By adjusting shiftable loads, demand response programs can reduce the peak-to-average ratio to maintain the stability of the power grid and avoid the cost of standby generators. Therefore, it is crucial for distribution network operators to establish an effective mechanism to enhance the flexibility of the power system.

[0003] However, the types and characteristics of flexible resources on the user side are becoming increasingly complex and the quantity is increasing. Traditional research tends to focus on specific demand response types for specific flexible resource users; in addition, the current participation of flexible resource users in demand response (DR) is described as a single-agent scheduling optimization problem for distribution network operators, and less consideration is given to objectives such as the electricity consumption benefits of flexible resource users, resulting in a lack of consideration of the game between distribution network operators and flexible resource users. Summary of the Invention

[0004] The purpose of the present invention is to propose a multi-objective game scheduling method and system for a distribution network operator, so as to improve the power management level of flexible resource users.

[0005] To achieve the purpose of the present invention, the technical solutions provided by the present invention are as follows:

[0006] First Aspect

[0007] The present invention provides a multi-objective game scheduling method for a distribution network operator, including the following steps:

[0008] Step S1: Obtain power consumption data;

[0009] Step S2: Based on the obtained power consumption data, establish a distribution network operator model with the goal of minimizing the power grid operation cost;

[0010] Step S3: Based on the obtained power consumption data, establish a load aggregator cost model for a load aggregator composed of flexible resource users with the goal of economy;

[0011] Step S4: Take the distribution network operator model established in Step S2 as the upper-layer model, take the load aggregator cost model established in Step S3 as the lower-layer model, construct a master-slave game model, and use an improved gray wolf algorithm to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the distribution network operator's day-ahead scheduling.

[0012] The second aspect

[0013] The present invention provides a multi - objective game scheduling system for distribution network operators, including an electric energy data acquisition unit and a monitoring host. The electric energy data acquisition unit interacts with the monitoring host, and the monitoring host includes a distribution network operator model establishment unit, a load aggregator cost model establishment unit, and a master - slave game model establishment unit;

[0014] The electric energy data acquisition unit is used to obtain power consumption data;

[0015] The distribution network operator model establishment unit is used to establish a distribution network operator model with the goal of minimizing the grid operation cost based on the obtained power consumption data;

[0016] The load aggregator cost model establishment unit is used to establish a load aggregator cost model for the load aggregator composed of flexible resource users with the goal of economy based on the obtained power consumption data;

[0017] The master - slave game model establishment unit is used to construct a master - slave game model with the established distribution network operator model as the upper - layer model and the established load aggregator cost model as the lower - layer model, and use an improved grey wolf algorithm to solve the master - slave game model to obtain the optimal day - ahead scheduling time - of - use electricity price result for the distribution network operator.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0019] The present invention provides a multi - objective game scheduling method and system for distribution network operators. By modeling flexible resource users and grid operators, establishing a day - ahead scheduling problem with the goal of economy, and using an improved grey wolf algorithm for solution to obtain the optimal day - ahead time - of - use electricity price, the present invention can improve the power management level on the flexible resource user side and has good practical significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the method flow provided by the embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The following further describes the present invention in detail with reference to the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Embodiment 1

[0022] As Figure 1 shown, this embodiment provides a multi - objective game scheduling method for distribution network operators, including the following steps:

[0023] Step S1: Obtain power consumption data;

[0024] It should be noted that the electricity consumption data refers to various electrical energy information obtained after the intelligent electricity consumption monitoring front-end performs real-time operation and processing on the voltage and current signals in the power grid at the installation point.

[0025] Step S2: Based on the obtained electricity consumption data, establish a distribution network operator model with the goal of minimizing the grid operation cost;

[0026] It should be noted that the distribution network operated by the Distribution System Operator (DSO) exchanges electrical energy with the upstream transmission network through a single tie line to balance the source and load at each time period within the region, meet the electricity consumption needs of users, and ensure the safe and stable operation of the distribution network. The DSO conducts electricity trading with the Load Aggregator (LA), Distributed Generation Cluster (DG), and inflexible loads within the region. In the transaction with the DSO, the LA can flexibly adjust its electricity consumption strategy according to the electricity price announced by the DSO to achieve the purpose of reducing its own electricity consumption cost and arbitraging in the electricity market. On the one hand, as the agent of the regional distribution network, the DSO p s and the time-of-use purchase electricity price p b conduct electricity trading with the upstream power grid, ordinary loads, and new energy generation within the region. On the other hand, the DSO formulates the time-of-use selling electricity price for each LA at each moment considering the regional situation p LA and conducts transactions with each LA.

[0027] The DSO encourages flexible resources to participate in DR in the form of price-based demand response by publishing the interactive electricity price with the LA. Within the LA, the LA issues instructions for incentive-based demand response to flexible resource users to enable flexible resources to participate in DR and arbitrage in the energy market. To ensure fair electricity prices and prevent unrestricted arbitrage by the LA, the electricity price satisfies the relational expression:

[0028] ;

[0029] After the DSO publishes the time-of-use electricity price, each LA optimizes its internal operation plan according to the time-of-use electricity price information to minimize the electricity consumption cost and uploads the plan to the DSO. The DSO then obtains the grid operation cost under this electricity price strategy based on the feedback plan. The DSO publishes the time-of-use electricity price again, and the LA adjusts to obtain a new operation plan and feedback it to the DSO. Such dynamic interaction continues until both parties are satisfied and then terminates, and the final time-of-use electricity price and operation plan are obtained.

[0030] Based on the above, in step S2, the objective function C DSO of the distribution network operator model

[0031] ;

[0032] In the formula, λ 1. λ 2 is the weighted cost coefficient, which is non - negative and the sum is 1, f oper is the power grid operation cost function, f Fluc is the load volatility cost function;

[0033] Among them, f oper is as follows:

[0034] ;

[0035] In the formula, C D2LA is the cost of the distribution network operator's transaction with the load aggregator; C DG is the cost of the distribution network operator's transaction with the distributed generation; C Load is the cost of the distribution network operator's transaction with the inflexible load; C D2T is the cost of the distribution network operator's transaction with the transmission network operator; is the time step;

[0036] Specifically, C D2LA , C DG , C Load , C D2T are respectively given by the following formulas:

[0037] ;

[0038] In the formula, A is the number of load aggregators LA; is the time - of - use selling price of LA at time t; q LA,a,t is the a power purchase quantity of LA t at time a , where LA a represents the t th load aggregator; T is the total number of time periods of the game scheduling; t represents the th time period of the scheduling; D is the number of distributed generation clusters DG; DG d represents the d - th distributed generation cluster; represents the scenario π d for which the expectation is calculated, where π d is the scenario of different power generation amounts of DG d under different environmental effects; is the power generation amount of DG d at t time period scenario π d ; is the time-of-use selling price of the superior power grid at time t; represents the expectation calculation for scenario π l ; π l is the scenario of different electricity consumption amounts of the inflexible load; is the electricity consumption amount of the inflexible load t at time period scenario π l ; q Totals,t is t the electricity sales amount from the transmission system operator (TSO) to the distribution system operator (DSO) at time period q Totalb,t is t the electricity purchase amount from the TSO to the DSO at time period. Where q Totals,t and q Totalb,t are given by the following formula:

[0039] ;

[0040] Where f Fluc is as follows:

[0041] ;

[0042] Where σ is the load fluctuation penalty coefficient; q Fluc is the fluctuating power, and its definition is given by the following formula:

[0043] ;

[0044] Step S3: Based on the obtained electricity consumption data, establish a load aggregator cost model for the load aggregator composed of flexible resource users with economy as the goal;

[0045] Specifically, in step S3, the load aggregator cost model C LA includes: the power purchase cost of flexible resource users, the electricity consumption benefit of flexible resource users, and the satisfaction cost of flexible resource users changing their electricity consumption behaviors; specifically as follows:

[0046] ;

[0047] In the formula, represents the cost of the a-th load aggregator; C Buy,a is the electricity purchase cost of flexible resource users, U a is the electricity consumption benefit of flexible resource users, C Dissat,a is the satisfaction cost of flexible resources changing their electricity consumption behavior;

[0048] Among them, C Buy,a is as follows:

[0049] ;

[0050] In the formula, q m,t is the flexible resource user m in the time period t actual electricity consumption; M a represents the set of flexible resource users belonging to the load aggregator; P LA, t is the time-of-use selling electricity price of each load aggregator formulated by the distribution network operator considering the regional situation on a daily basis;

[0051] Among them, U a is as follows:

[0052] ;

[0053] In the formula, is the electricity consumption benefit coefficient of flexible resource users m in the time period t ; is the flexible resource user m in the time period t maximum electricity consumption; T is the total number of time periods for game scheduling;

[0054] Among them, C Dissat,a is as follows:

[0055] ;

[0056] In the formula, α m,t is the demand response satisfaction coefficient of flexible resource users m in the time period t ; h m,t is the flexible resource user mDuring the time period t The planned power consumption before participating in DR.

[0057] Step S4: Take the distribution network operator model established in step S2 as the upper-level model, take the load aggregator cost model established in step S3 as the lower-level model, construct a master-slave game model, and use the improved grey wolf algorithm to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the day-ahead dispatch of the distribution network operator.

[0058] The constraint conditions of the master-slave game model are as follows:

[0059] (1) Each load aggregator needs to satisfy the power balance constraint:

[0060] ;

[0061] (2) When flexible resource users change their electricity consumption behaviors, the reduced load can be classified into curtailable load q Shed,m,t and shiftable load q Shift,m,t :

[0062] ;

[0063] Among them, the curtailable load q Shed,m,t satisfies the maximum curtailment amount constraint:

[0064] ;

[0065] In the formula, q ShedMax,m,t is the maximum curtailment amount of the curtailable load of flexible resource user m during the time period t ;

[0066] Among them, the shiftable load q Shift,m,t satisfies the maximum curtailment amount constraint and the time shift constraint:

[0067] ;

[0068] In the formula, q ShiftUpMax,m,t is the maximum curtailment amount of the shiftable load of flexible resource user m during the time period t ; q ShiftDownMax,m,t is the maximum increment amount of the shiftable load of flexible resource user m during the time period t ;

[0069] Among them, the improved grey wolf algorithm is used to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the distribution network operator's day-ahead scheduling. The specific steps are as follows:

[0070] Step S4.1: Set the number of grey wolf populations E， The maximum number of iterations K， The convergence factor vector g ;

[0071] Step S4.2: Initialize the position vector of each grey wolf, and initialize the random coefficient vectors B and C , and initialize the current iteration number k to 0;

[0072] Step S4.3: Solve the lower-level model and report the solution result to the upper-level model;

[0073] Step S4.4: After the upper-level model receives the solution result, solve the upper-level optimization model;

[0074] Step S4.5: After obtaining the fitness of all grey wolf individuals, select the top 3 wolves with the highest fitness as α wolf, β wolf 、 δ wolves, and the remaining wolves as ω wolves; among them, the α wolf guides the entire wolf pack to make decisions, corresponding to the optimal solution of the algorithm; β wolf and δ wolf assist in management, corresponding to the sub-optimal solution of the algorithm; the ω wolf is responsible for searching according to the information of α wolf, β wolf, δ wolf;

[0075] Step S43.6: Judge whether k is greater than K ? If so, jump to Step S4.7; if not, jump to Step S4.8;

[0076] Step S4.7: Output the position vector of the α wolf and the corresponding upper-level model solution result as the optimal result;

[0077] Step S4.8: Add 1 to the value of k ;

[0078] Step S4.9: Update the convergence factor vector g ;

[0079] Step S4.10: All ωThe wolf updates its position vector and jumps to step S4.3.

[0080] Among them, the distance vector between the gray wolf and the prey is defined D The formula is as follows:

[0081] ;

[0082] In the formula, ∗ represents the multiplication of corresponding elements of the vector; X p is the position vector of the prey; X is the position vector of the gray wolf;

[0083] Among them, the iterative update formula for the position of the gray wolf is defined as follows:

[0084] ;

[0085] Among them, B and C are defined as follows:

[0086] ;

[0087] ;

[0088] Among them, g linearly decreases from 2 to 0 as the number of iterations increases; r 1 and r 2 are random vectors with elements in the range [0,1];

[0089] Define the current ω wolf and α wolf, β wolf, δ wolf's position vectors:

[0090] ;

[0091] Among them, D α 、 D β 、 D δ are the distance vectors between the current gray wolf and α wolf, β wolf, δ wolf respectively; X α 、 X β 、

[0092] X δ are respectively α wolf, β wolf,δ Position vector of the wolf C 1、 C 2、 C 3 is a random coefficient vector.

[0093] Preferably, the convergence factor vector g has the following calculation formula:

[0094] ;

[0095] In the formula, n is a non-linear attenuation exponent.

[0096] Preferably, the current ω wolves use α wolves, β wolves, δ the position vectors of wolves to update their own positions using the first method or the second method:

[0097] The first method:

[0098] ;

[0099] ;

[0100] Among them, B 1、 B 2、 B 3 are calculation coefficient vectors.

[0101] The second method:

[0102] ;

[0103] ;

[0104] In the formula, f 1、 f 2、 f 3 are respectively α wolves, β wolves, δ the fitness of wolves.

[0105] Embodiment 2

[0106] This embodiment provides a multi-objective game scheduling system for a distribution network operator, including an electric energy data acquisition unit and a monitoring host. The electric energy data acquisition unit interacts with the monitoring host, and the monitoring host includes a distribution network operator model establishment unit, a load aggregator cost model establishment unit, and a master-slave game model establishment unit;

[0107] The electric energy data acquisition unit is used to obtain power consumption data;

[0108] The distribution network operator model establishment unit is used to establish a distribution network operator model based on the acquired power consumption data with the goal of minimizing the grid operation cost;

[0109] The load aggregator cost model establishment unit is used to establish a load aggregator cost model for the load aggregator composed of flexible resource users based on the acquired power consumption data with the goal of economy;

[0110] The master-slave game model establishment unit is used to construct a master-slave game model with the established distribution network operator model as the upper layer model and the established load aggregator cost model as the lower layer model, and use the improved grey wolf algorithm to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the distribution network operator's day-ahead scheduling.

[0111] It should be noted that the electric energy data acquisition unit is an intelligent power consumption monitoring front end of the existing technology. It uses the ADE9000 + STM32 chip to form the core architecture, and uses the dedicated electric energy measurement chip ADE9000 to integrate the DSP core to perform real-time operation and processing on the voltage and current signals in the power grid to obtain various power consumption data; the main control unit composed of the STM32 chip completes service processes such as data storage, abnormal alarm, human-machine interface display, and data transmission.

[0112] The monitoring host is responsible for data acquisition, processing, display, and storage, and provides a human-machine interface for the distribution network operator. Flexible resource users can view various power consumption data collected by the intelligent power consumption monitoring front end through the upper computer, and can also query the historical data saved in the database.

[0113] Among them, the objective function of the distribution network operator model C DSO includes: the grid operation cost function and the load volatility cost function; as follows:

[0114] ;

[0115] In the formula, λ 1. λ 2 is the weighted cost coefficient, which satisfies non-negative and the sum is 1, f oper is the grid operation cost function, f Fluc is the load volatility cost function;

[0116] Among them, f oper as follows:

[0117] ;

[0118] In the formula, CLA is the cost of the transaction between the distribution network operator and the load aggregator; C DG is the cost of the transaction between the distribution network operator and the distributed generation; C Load is the cost of the transaction between the distribution network operator and the inelastic load; C D2T is the cost of the transaction between the distribution network operator and the transmission network operator; is the time step;

[0119] wherein, f Fluc is as follows:

[0120] ;

[0121] In the formula, σ is the load fluctuation penalty coefficient; q Fluc is the fluctuating power.

[0122] wherein, the load aggregator cost model C LA includes: the power purchase cost of the flexible resource user, the power consumption benefit of the flexible resource user, and the satisfaction cost of the flexible resource changing the power consumption behavior; as follows:

[0123] ;

[0124] In the formula, C Buy,a is the power purchase cost of the flexible resource user, U a is the power consumption benefit of the flexible resource user, C Dissat,a is the satisfaction cost of the flexible resource changing the power consumption behavior;

[0125] wherein, C Buy,a is as follows:

[0126] ;

[0127] In the formula, q m,t is the flexible resource user m in the time period t of the actual power consumption; M a represents the set of flexible resource users belonging to the load aggregator; P LA, t is the time-of-use selling price of the load aggregator formulated by the distribution network operator considering the regional situation for each moment;

[0128] Among them, U a as the following formula:

[0129] ;

[0130] In the formula, is the electricity consumption benefit coefficient of the flexible resource user m in the time period t ; is the maximum electricity consumption of the flexible resource user m in the time period t ; T is the total number of time periods for game scheduling;

[0131] Among them, C Dissat,a as the following formula:

[0132] ;

[0133] In the formula, α m,t is the demand response satisfaction coefficient of the flexible resource user m in the time period t ; h m,t is the planned electricity consumption of the flexible resource user m in the time period t before participating in DR.

[0134] Among them, the improved grey wolf algorithm is used to solve the master-slave game model to obtain the optimal day-ahead scheduling time-of-use electricity price result of the distribution network operator. The specific steps are as follows:

[0135] Step S4.1: Set the number of grey wolf populations E， the maximum number of iterations K， the convergence factor vector g ;

[0136] Step S4.2: Initialize the position vector of each grey wolf, initialize the random coefficient vectors B and C and initialize the current iteration number k to 0;

[0137] Step S4.3: Solve the lower-layer model and report the solution result to the upper-layer model;

[0138] Step S4.4: After the upper-layer model receives the solution result, solve the upper-layer optimization model;

[0139] Step S4.5: After obtaining the fitness of all grey wolf individuals, select the top 3 wolves with the highest fitness as α wolves, βWolf 、 δ The wolf, and the rest of the wolves act as ω wolves; among them, the α wolf guides the entire wolf pack to make decisions, corresponding to the optimal solution of the algorithm; β the wolf and δ wolves assist in management, corresponding to the sub-optimal solution of the algorithm; the ω wolves are responsible for searching for information according to α wolves, β wolves, δ wolves;

[0140] Step S4.6: Judge whether k is greater than K ? If so, jump to Step S4.7; if not, jump to Step S4.8;

[0141] Step S4.7: Output the position vector of the α wolf and the corresponding upper-layer model solution result as the optimal result;

[0142] Step S4.8: Add 1 to the value of k ;

[0143] Step S4.9: Update the convergence factor vector g ;

[0144] Step S4.10: Update the position vectors of all ω wolves, and jump to Step S4.3.

[0145] Among them, define the distance vector between the gray wolf and the prey D The formula is as follows:

[0146] ;

[0147] In the formula, ∗ represents the multiplication of corresponding elements of the vector; X p is the position vector of the prey; X is the position vector of the gray wolf;

[0148] Define the iterative update formula of the gray wolf position as follows:

[0149] ;

[0150] Among them, B and C are defined as follows:

[0151] ;

[0152] ;

[0153] Among them,g decrease linearly from 2 to 0 as the number of iterations increases; r 1 and r 2 are random vectors for the elements in the range [0, 1];

[0154] Define the current ω wolf and α wolves, β wolves, δ position vectors of wolves:

[0155] ;

[0156] Among them, D α , D β , D δ are the distance vectors between the current grey wolf and α wolves, β wolves, δ wolves respectively; X α , X β , X δ are respectively α wolves, β wolves, δ position vectors of wolves; C 1, C 2, C 3 are random coefficient vectors.

[0157] The optional implementation manners of the embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above implementation manners. Within the technical concept scope of the embodiments of the present invention, various simple modifications can be made to the technical solutions of the embodiments of the present invention, and these simple modifications all fall within the protection scope of the embodiments of the present invention.

Claims

1. A multi-objective game scheduling method for a distribution network operator, characterized in that It includes the following steps: Step S1: Obtain power consumption data; Step S2: Based on the obtained power consumption data, establish a distribution network operator model with the goal of minimizing the grid operation cost; Step S3: Based on the obtained power consumption data, establish a load aggregator cost model for the load aggregator composed of flexible resource users with the goal of economy; Step S4: Take the distribution network operator model established in Step S2 as the upper-level model, take the load aggregator cost model established in Step S3 as the lower-level model, construct a master-slave game model, and use an improved grey wolf algorithm to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the distribution network operator's day-ahead dispatch; In step S2, the objective function C of the distribution network operator model DSO includes: a grid operation cost function and a load volatility cost function, as follows: where λ1 and λ2 are weighted cost coefficients, satisfying non - negativity and the sum being 1, and f oper is the power grid operation cost function, and f Fluc is the load volatility cost function; where f oper is as follows: f oper = (C D2LA + C DG + C Load + C D2T )Δt where C D2LA is the cost of the transaction between the distribution network operator and the load aggregator; C DG is the cost of the transaction between the distribution network operator and the distributed generation; C Load is the cost of the transaction between the distribution network operator and the inelastic load; C D2T is the cost of the transaction between the distribution network operator and the transmission network operator; Δt is the time step; where f Fluc is as follows: f Fluc = σq Fluc Among them, σ is the load fluctuation penalty coefficient; q Fluc is the fluctuating power; In step S3, the load aggregator cost model C LA includes: the electricity purchase cost of flexible resource users, the electricity consumption benefit of flexible resource users, and the satisfaction cost of flexible resources changing their electricity consumption behaviors; specifically, the following formula: where C LA,a represents the cost of the a-th load aggregator; C Buy,a is the electricity purchase cost of flexible resource users, U a is the electricity consumption benefit of flexible resource users, and C Dissat,a is the satisfaction cost for flexible resources to change their electricity consumption behavior; Among them, C Buy,a As shown in the following formula: where q m,t is the actual power consumption of flexible resource user m in time period t; M a represents the set of flexible resource users belonging to the load aggregator; P LA,t is the time-of-use selling price of the load aggregator at each moment formulated by the distribution network operator considering the regional situation in advance; Among them, U a As shown in the following formula: where ω m，t is the electricity consumption benefit coefficient of flexible resource user m in time period t; β m，t is the maximum electricity consumption of flexible resource user m in time period t; T is the total number of time periods for game scheduling; Among them, C Dissat,a As shown in the following formula: where α m,t is the demand response satisfaction coefficient of flexible resource user m in time period t; h m,t is the planned power consumption of flexible resource user m before participating in DR in time period t.

2. The multi-objective game scheduling method for a distribution network operator according to claim 1, wherein In Step S4: The process of using the improved grey wolf algorithm to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the distribution network operator's day-ahead dispatch is as follows: Step S4.1: Set the number of grey wolf populations E, the maximum number of iterations K, and the convergence factor vector g; Step S4.2: Initialize the position vectors of each grey wolf, initialize the random coefficient vectors B and C, and initialize the current iteration number k to 0; Step S4.3: Solve the lower-level model and report the solution result to the upper-level model; Step S4.4: After the upper-level model receives the solution result, solve the upper-level optimization model; Step S4.5: After obtaining the fitness of all grey wolf individuals, select the top 3 wolves with the highest fitness as the α wolf, β wolf, and δ wolf, and the remaining wolves as the ω wolf; among them, the α wolf guides the entire wolf pack to make decisions, corresponding to the optimal solution of the algorithm; the β wolf and δ wolf assist in management, corresponding to the sub-optimal solutions of the algorithm; the ω wolf is responsible for searching according to the information of the α wolf, β wolf, and δ wolf; Step S4.6: Determine whether k is greater than K? If so, jump to Step S4.7; if not, jump to Step S4.8; Step S4.7: Output the position vector of the α wolf and the corresponding upper-level model solution result as the optimal result; Step S4.8: Increment the value of k by 1; Step S4.9: Update the convergence factor vector g; Step S4.10: Update the position vectors of all ω wolves and jump to Step S4.3; Among them, the formula for defining the distance vector D between the grey wolf and the prey is as follows: where * represents element-wise multiplication of vectors; X p is the position vector of the prey; X is the position vector of the grey wolf; Among them, the formula for defining the iterative update of the grey wolf position is as follows: Among them, the definitions of B and C are as follows: B = 2g * r1 - g C＝2r2 Among them, g linearly decreases from 2 to 0 as the number of iterations increases; r1 and r2 are random vectors with elements in the range [0, 1]; Define the position vectors of the current ω wolf, α wolf, β wolf, and δ wolf: Among them, D α , D β , D δ are the distance vectors between the current gray wolf and the α-wolf, β-wolf, and δ-wolf respectively; X α , X β , X δ are the position vectors of the α-wolf, β-wolf, and δ-wolf respectively; C1, C2, and C3 are random coefficient vectors.

3. The multi-objective game scheduling method for a distribution network operator according to claim 2, characterized in that The calculation formula for the convergence factor vector g is as follows: In the formula, n is the non-linear attenuation exponent.

4. A multi-objective game scheduling method for a distribution network operator according to claim 3, characterized in that The current ω wolf updates its own position using the position vectors of the α wolf, β wolf, and δ wolf in the first way or the second way: The first way: Among them, B1, B2, and B3 are calculation coefficient vectors; The second way: In the formula, f1, f2, and f3 are the fitnesses of the α wolf, β wolf, and δ wolf respectively.

5. A multi-objective game scheduling system for a distribution network operator, characterized in that, It includes an electric energy data acquisition unit and a monitoring host. The electric energy data acquisition unit interacts with the monitoring host. The monitoring host includes a distribution network operator model establishment unit, a load aggregator cost model establishment unit, and a master-slave game model establishment unit; The electric energy data acquisition unit is used to obtain power consumption data; The distribution network operator model establishment unit is used to establish a distribution network operator model with the goal of minimizing the grid operation cost based on the obtained power consumption data; The load aggregator cost model establishment unit is used to establish a load aggregator cost model for the load aggregator composed of flexible resource users with economy as the goal based on the obtained power consumption data; The master-slave game model establishment unit is used to construct a master-slave game model with the established distribution network operator model as the upper-layer model and the established load aggregator cost model as the lower-layer model, and use an improved grey wolf algorithm to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the distribution network operator's day-ahead dispatch; The objective function C of the distribution network operator model DSO includes: a power grid operation cost function and a load volatility cost function, as shown in the following formula: where λ1 and λ2 are weighted cost coefficients, satisfying non - negativity and the sum being 1, and f oper is the power grid operation cost function, and f Fluc is the load volatility cost function; where f oper is as follows: f oper = (C D2LA + C DG + C Load + C D2T )Δt where C D2LA is the cost of the transaction between the distribution network operator and the load aggregator; C DG is the cost of the transaction between the distribution network operator and the distributed generation; C Load is the cost of the transaction between the distribution network operator and the inelastic load; C D2T is the cost of the transaction between the distribution network operator and the transmission network operator; Δt is the time step; where f Fluc is as follows: f Fluc = σq Fluc where σ is the load fluctuation penalty coefficient; q Fluc is the fluctuating power; The load aggregator cost model C LA includes: the electricity purchase cost of flexible resource users, the electricity consumption benefit of flexible resource users, and the satisfaction cost of flexible resources changing their electricity consumption behaviors; specifically as follows: where C LA,a represents the cost of the a-th load aggregator; C Buy,a is the power purchase cost of the flexible resource user, U a is the power consumption benefit of the flexible resource user, and C Dissat,a is the satisfaction cost for the flexible resource to change its power consumption behavior; Among them, C Buy,a As shown in the following formula: where q m,t is the actual power consumption of flexible resource user m during time period t; M a represents the set of flexible resource users belonging to the load aggregator; P LA,t is the time-of-use selling price of the load aggregator at each moment formulated by the distribution network operator considering the regional situation in advance; Among them, U a As shown in the following formula: where ω m，t is the electricity consumption benefit coefficient of flexible resource user m in time period t; β m，t is the maximum electricity consumption of flexible resource user m in time period t; T is the total number of time periods for game scheduling; Among them, C Dissat,a As shown in the following formula: where α m,t is the demand response satisfaction coefficient of flexible resource user m in time period t; h m,t is the planned power consumption of flexible resource user m before participating in DR in time period t.

6. The multi-objective game scheduling system for a distribution network operator according to claim 5, characterized in that, In step S4: The process of using the improved grey wolf algorithm to solve the master-slave game model to obtain the optimal time-of-use electricity price result for the distribution network operator's day-ahead dispatch is as follows: Step S4.1: Set the number of grey wolf populations E, the maximum number of iterations K, and the convergence factor vector g; Step S4.2: Initialize the position vector of each grey wolf, initialize the random coefficient vectors B and C, and initialize the current iteration number k to 0; Step S4.3: Solve the lower-layer model and report the solution result to the upper-layer model; Step S4.4: After the upper-layer model receives the solution result, solve the upper-layer optimization model; Step S4.5: After obtaining the fitness of all grey wolf individuals, select the top 3 wolves with the highest fitness as the α wolf, β wolf, and δ wolf, and the remaining wolves as the ω wolf; among them, the α wolf guides the entire wolf pack to make decisions, corresponding to the optimal solution of the algorithm; the β wolf and δ wolf assist in management, corresponding to the sub-optimal solutions of the algorithm; the ω wolf is responsible for searching according to the information of the α wolf, β wolf, and δ wolf; Step S4.6: Judge whether k is greater than K? If so, jump to step S4.7; if not, jump to step S4.8; Step S4.7: Output the position vector of the α wolf and the corresponding upper-layer model solution result as the optimal result; Step S4.8: Increment the value of k by 1; Step S4.9: Update the convergence factor vector g; Step S4.10: Update the position vectors of all ω wolves and jump to step S4.3; Among them, the formula for defining the distance vector D between the grey wolf and the prey is as follows: where * represents element-wise multiplication of vectors; X p is the position vector of the prey; X is the position vector of the grey wolf; Among them, the formula for defining the iterative update of the grey wolf position is as follows: Among them, the definitions of B and C are as follows: B = 2g * r1 - g C＝2r2 Among them, g linearly decreases from 2 to 0 as the number of iterations increases; r1 and r2 are random vectors with elements in the range [0, 1]; Define the position vectors of the current ω wolf, α wolf, β wolf, and δ wolf: Among them, D α , D β , D δ are the distance vectors of the current grey wolf from the alpha wolf, beta wolf, and delta wolf respectively; X α , X β , X δ are the position vectors of the alpha wolf, beta wolf, and delta wolf respectively; C1, C2, and C3 are random coefficient vectors.

Images