Source network load storage collaborative planning method and system considering multi-agent mixed game

Abstract

The application relates to a source-network-load-storage collaborative planning method and system considering multi-agent mixed games, the method comprising the following steps: establishing a power distribution network collaborative planning model considering master-slave-cooperation mixed games of source-network-load-storage agents, the power distribution network collaborative planning model being a double-layer model, comprising an upper-layer master-slave game sub-model of a power distribution network operator and other agents and a lower-layer cooperation game sub-model between source-load-storage agents; and adopting a solving strategy combining a target cascade analysis method and an adaptive coefficient alternating direction multiplier method to solve the power distribution network collaborative planning model, so as to obtain an optimal planning scheme of line new construction, distributed power and energy storage devices. Compared with the prior art, the application has the advantages of being capable of realizing efficient configuration and coordinated utilization of source-network-load-storage resources, improving planning reliability and the like.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of power distribution networks, and relates to a power distribution network planning method, in particular to a source-grid-load-storage collaborative planning method and system considering multi-agent mixed game. BACKGROUND

[0002] With the deepening of the power market reform, various types of social capital join the construction of the power distribution network, presenting a competitive situation of interest subjects. The rapid development of the four aspects of the new power distribution system, i.e. source, grid, load and storage, increases the number of elements of the power distribution system and the number of operation and maintenance subjects, and the interaction between the source and the load becomes more in-depth, and the operation characteristics become more complex. Each subject has its own interest demands, and also needs to bear the important task of maintaining the efficient operation of the system. The diversified subjects and the competitive environment put forward higher requirements for the power distribution network planning, and the actual needs of considering the specific interest demands of different subjects bring new problems to the fine planning of the power distribution system. Specifically, the construction of the distributed energy station easily leads to the increase of peak load and the overload rate of the line exceeding the safety standard, the access of the distributed photovoltaic leads to the increase of the line power flow peak-valley difference rate, and the energy storage device needs to be accessed to improve the operation level of the system. These problems lead to the increase of the investment cost of the power grid planning. Under this background, how to consider the interest demands of various subjects to reasonably depict the game situation, and then realize the efficient configuration and coordinated utilization of the source-grid-load-storage resources in the planning is a problem to be solved.

[0003] The existing researches rarely optimize the demand of each subject independently, or do not reflect the marketized game relationship of the multi-side subjects of the source-grid-load-storage in the power distribution network planning, and it is difficult to promote the consumption of new energy and the improvement of the power user's electricity behavior. SUMMARY

[0004] The purpose of the application is to overcome the defects of the prior art and provide a source-grid-load-storage collaborative planning method and system considering multi-agent mixed game, which is suitable for 110kV and below power distribution networks and can realize the efficient configuration and coordinated utilization of source-grid-load-storage resources.

[0005] The purpose of the application can be realized by the following technical solutions:

[0006] A source-grid-load-storage collaborative planning method considering multi-agent mixed game, comprising the following steps:

[0007] A power distribution network collaborative planning model considering master-slave-cooperative mixed game of source-grid-load-storage subjects is established, the source-grid-load-storage subjects include a power distribution network operator, a distributed power source operator, an energy storage operator and a power user, the distributed power source operator, the energy storage operator and the power user form a source-load-storage side subject, the power distribution network collaborative planning model is a double-layer model, including an upper-layer master-slave game sub-model of the power distribution network operator and other subjects and a lower-layer cooperative game sub-model between the source-load-storage side subjects.

[0008] The distribution network collaborative planning model is solved by using a solving strategy combining a target cascade analysis method and an adaptive coefficient alternating direction multiplier method, to obtain an optimal planning scheme of line new construction, distributed power supply and energy storage device.

[0009] Further, the upper-layer master-slave game sub-model is a planning and decision model with the distribution network operator as the leading party, and the target is to maximize the annual comprehensive benefit of the distribution network operator, and the constraint conditions include system power flow constraint, system operation safety constraint, newly constructed line constraint and traded power balance constraint.

[0010] Further, the lower-layer cooperative game sub-model includes a comprehensive cost minimization sub-problem and a benefit distribution maximization sub-problem, wherein the comprehensive cost minimization sub-problem takes the sum of the power costs of the distributed power supply operator, the energy storage operator and the power user as the target, and the constraint conditions include traded power constraint and power transaction balance constraint, and the benefit distribution maximization sub-problem takes the maximum increment of the benefits of each subject after cooperation as the target, and the constraint conditions include cooperative benefit constraint and traded power price constraint.

[0011] Further, the objective function of the benefit distribution maximization sub-problem is expressed as:

[0012]

[0013] In the formula, is the benefit after cooperation of the benefit subject n; is the benefit before cooperation of the benefit subject n; C n is the annual comprehensive cost of the subject n; F is the number of subjects; α n is the bargaining power of the benefit subject n, and the bargaining power is obtained based on a power mapping function.

[0014] Further, when the distribution network collaborative planning model is solved, firstly, the double-layer model is decoupled by using the target cascade analysis method, and then the lower-layer cooperative game sub-model is distributedly solved by using the adaptive coefficient alternating direction multiplier method.

[0015] Further, in the target cascade analysis method, the end condition of the external loop iteration is:

[0016]

[0017] In the formula, is the power expected to be traded between the benefit subject alliance and the distribution network operator; is the power expected to be traded by the distribution network operator, and the superscripts k and k o are the number of internal and external loop iterations of the target cascade analysis method, Ns is the number of scenarios, T is the number of time periods, and ε is a convergence threshold. o is a convergence threshold.

[0018] Further, in the target cascade analysis method, the penalty factor is updated at the end of the external loop iteration, and the update formula is:

[0019]

[0020] wherein σ is an update constant, is the penalty factor at the k o , k o +1 external loop iteration.

[0021] Further, the distributed solving includes: first solving a comprehensive cost minimization sub-problem to obtain the transaction power between each subject, and then solving a benefit distribution maximization sub-problem to obtain the expected transaction electricity price of the subject.

[0022] Further, the distributed solving specifically includes:

[0023] An augmented Lagrange function of each sub-problem is established;

[0024] According to the expected transaction electricity quantity / price information provided by other subjects, each subject autonomously updates the decision, and transmits the updated expected transaction electricity quantity / price information to other subjects;

[0025] The Lagrange multiplier is updated, and the penalty factor is adaptively adjusted;

[0026] After the iteration termination condition is met, the final solving result is obtained.

[0027] The application also provides a source-grid-load-storage collaborative planning system considering a multi-subject mixed game, which comprises one or more processors, a memory and one or more programs stored in the memory, and the one or more programs comprise instructions for executing the source-grid-load-storage collaborative planning method considering the multi-subject mixed game.

[0028] Compared with the prior art, the application has the following beneficial effects:

[0029] 1. The application constructs a power distribution network collaborative planning model considering the master-slave-cooperative mixed game of source-grid-load-storage subjects, innovatively combines the master-slave game relationship between the power distribution network operator and the source-grid-load-storage subjects with the cooperative game inside the source-grid-load-storage subjects, effectively stimulates the enthusiasm of market subjects in participating in planning, effectively balances the interests of the source-grid-load-storage subjects on each side, promotes the consumption of new energy, reduces the grid supply load of the system, improves the power consumption behavior of power users, realizes flexible and efficient configuration and coordinated utilization of source-grid-load-storage power resources, and improves the planning level of the power distribution network.

[0030] 2、The application balances the interests of multiple parties through the master-slave cooperation mixed game mechanism, optimizes the overall allocation efficiency of resources, the upper master-slave game sub-model takes maximizing the comprehensive benefits of the power distribution network operator as the target, and the lower cooperation game sub-model is used for encouraging new energy investment and user side demand response, reducing the dependence on the main network, improving the system flexibility and new energy consumption capacity, solving the problem of insufficient consideration of individual interests in the traditional method, and significantly improving the rationality of the planning scheme.

[0031] 3、The application adopts a model solving method based on the target cascade analysis method and the adaptive alternating direction multiplier method, realizes decoupling and distributed iteration solving of the double-layer model, can achieve global convergence, is efficient in solving, and the solving result is reliable. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 It is a multi-agent mixed game relationship schematic diagram of the application;

[0033] Figure 2 It is a whole solving flow schematic diagram of the power distribution network collaborative planning model considering the source network load storage agent master-slave cooperation mixed game of the application;

[0034] Figure 3 It is an IEEE33 node system planning initial topology in the embodiment of the application;

[0035] Figure 4 It is typical daily output data in the embodiment of the application;

[0036] Figure 5 It is a power distribution network operator simulation transaction result graph in scenario 2 in the embodiment of the application;

[0037] Figure 6 It is a demand response power load power result graph in the embodiment of the application;

[0038] Figure 7 It is a power user simulation transaction result graph in scenario 2 in the embodiment of the application;

[0039] Figure 8 It is an algorithm iteration convergence process graph in the embodiment of the application;

[0040] Figure 9 It is a graph of Q1 target function change of each agent in the embodiment of the application. DETAILED DESCRIPTION

[0041] The application will be described in detail below in combination with the drawings and specific embodiments. The embodiments are implemented on the premise of the technical scheme of the application, detailed implementation modes and specific operation processes are given, but the protection scope of the application is not limited to the following embodiments.

[0042] The embodiment provides a source-network-load-storage collaborative planning method considering multi-agent mixed game, which comprises the following steps: establishing a power distribution network collaborative planning model considering master-slave-cooperation mixed game of source-network-load-storage agents, wherein the source-network-load-storage agents comprise a power distribution network operator, a distributed power source operator, an energy storage operator and a power user, the distributed power source operator, the energy storage operator and the power user form source-load-storage side agents, the power distribution network collaborative planning model is a double-layer model, comprising an upper-layer master-slave game sub-model of the power distribution network operator and other agents and a lower-layer cooperation game sub-model between the source-load-storage side agents; and a solving strategy combining a target cascade analysis method and an adaptive coefficient alternating direction multiplier method is used to solve the power distribution network collaborative planning model, so as to obtain an optimal planning scheme of line new construction, distributed power source and energy storage device.

[0043] The method innovatively combines the master-slave game relationship between the power distribution network operator and the source-network-load-storage agents with the cooperation game among the source-network-load-storage agents, and solves based on the target cascade analysis method and the adaptive alternating direction multiplier method, so that the interests of multiple parties can be balanced, the overall configuration efficiency of resources can be optimized, and the rationality and reliability of power distribution network planning can be improved.

[0044] I. Game relationship

[0045] The embodiment relates to the game relationship of four interest agents and two planning levels. The power distribution network operator decides a line new construction scheme with the target of reducing line expansion cost, network loss and operation cost; the distributed power source (DG) operator and the energy storage operator respectively decide the capacity configuration and access position of the DG and the energy storage device with the target of reducing investment cost and increasing power selling income; and the power user decides a power consumption mode with the target of minimizing power consumption cost according to a power price signal and an incentive mechanism. The network frame line influences the site selection and capacity determination of the DG and the energy storage, the access position and capacity of the DG and the energy storage influence the power selling income of the power distribution network operator and the power purchase selection of the power user, and the power consumption scheme of the power user influences the power consumption price and profit income of the power distribution network operator, the DG operator and the energy storage operator. However, the interests of the parties are compatible to a certain extent, and there is a possibility of cooperation and win-win. The DG and the energy storage relieve the power supply pressure of the power distribution network in operation, and the user-side demand response promotes the balance between power supply and demand and guarantees the stable operation of the system. The direct market transaction of the DG, the energy storage and the power user breaks the monopoly of the power distribution network operator, increases the selectivity of power interaction, and can avoid the phenomenon of abandoned wind and light and increase the utilization rate of the energy storage battery to a certain extent.

[0046] Reference Figure 1As shown, the distribution network operator dominates the distribution network construction and management, has high decision-making power, and can be regarded as the game leader. In the upper layer problem, the distribution network operator establishes the network structure of the line and transmits it to the lower layer followed by other subjects. The lower layer subjects exist in a cooperative game relationship and can independently conduct market transactions in addition to purchasing and selling electricity from the distribution network. The DG operator and the energy storage operator determine the access capacity and location according to the existing topology structure and optimize the operation scheme. At the same time, the power users determine the power consumption mode and power purchase selection according to the electricity price signal, and transmit the power generation and consumption plan to the upper layer distribution network operator. The distribution network operator re-plans the network structure and formulates the electricity price according to the latest configuration scheme and the power generation and consumption plan, and the two-layer game coordination iteration is carried out until the equilibrium is reached, realizing the win-win of the four parties.

[0047] II. Collaborative planning model of distribution network

[0048] 1. Upper layer master-slave game sub-model

[0049] In this embodiment, the upper layer master-slave game sub-model is a planning and decision model led by the network side subject, considering the new construction of the network line, taking the annual comprehensive income of the distribution network operator as the target, and the objective function is:

[0050]

[0051] In the formula, C DSO , are the annual comprehensive income, the annual electricity selling income, the annual electricity purchasing cost from the upper grid and the annual investment cost of the distribution network operator respectively; f t T , f t G are the electricity selling price and the upper grid price at t period respectively; are the electricity selling amount at t period and the substation purchasing amount at s scenario respectively; is the set of to-be-constructed lines; l i is the length of the i-th to-be-constructed line; is the unit length investment cost of the i-th to-be-constructed line; is a 0-1 variable indicating whether the selected line is newly constructed; N s is the number of scenarios; D s is the number of days of s scenario; b is the discount rate; y is the service life of the equipment.

[0052] The constraint conditions considered by the upper layer master-slave game sub-model include system power flow constraints, system operation safety constraints, new line construction constraints and transaction power balance constraints.

[0053] 1) System power flow constraints

[0054]

[0055] where: u(j) is the set of head nodes of branches with j as the end node; w(j) is the set of end nodes of branches with j as the head node; P ij,s,t , P jk,s,t and Q ij,s,t , Q jk,s,t are the active and reactive power flowing through feeder (i, j) and (j, k) in scenario s and period t; U i,s,t , U j,s,t are the voltage magnitudes of node i and j in scenario s and period t; I ij,s,t is the current magnitude of line ij in scenario s and period t; P j,s,t and Q j,s,t are the active and reactive power injected by node j; and are the active and reactive power of DG at node j; is the charge-discharge power of energy storage at node j; and are the outage power, power transferred out and power transferred in of load at node j; and are the active and reactive power of load at node j; R ij , X ij , Z ij are the resistance, reactance and impedance of line ij, respectively.

[0056] 2) System operation security constraints

[0057] U min ≤ U i,t ≤ U max (10)

[0058]

[0059] where: U max and U min are the upper and lower voltage limits; I ij,max is the security current of line ij; is a 0-1 variable indicating whether line ij is on or off; U i,t , I ij,t are the voltage magnitude of node i and current magnitude of line ij in period t.

[0060] 3) New line construction constraints

[0061]

[0062] where: is the set of lines to be constructed at new node a; indicates whether line i at new node a is new or not.

[0063] 4) Trading power balance constraint

[0064]

[0065] wherein: are the trading power of DG operator, energy storage operator, power user and interest subject m, respectively; Φ0is the set of other interest subjects except itself.

[0066] 2) Lower-level cooperative game sub-model

[0067] The lower-level cooperative game sub-model is constructed based on the planning decision model of source-load-storage follower subjects, which includes DG operator, energy storage operator and power user. The following is a detailed description:

[0068] DG operator: Considering the interests of DG operators mainly based on photovoltaic and wind power, the installation location and capacity of distributed wind power and distributed photovoltaic are determined to maximize the annual comprehensive income of the operator. The objective function can be expressed as:

[0069]

[0070] wherein: C DG , are the annual comprehensive income, annual electricity sales income, government new energy subsidy cost, annual investment cost and annual operation and maintenance cost of the DG operator, respectively; are the electricity sales price of the DG operator, the new energy subsidy price of the government, and the operation and maintenance fee per unit of DG power generation, respectively; Ω DG is the set of DG candidate installation nodes; c DG is the unit capacity investment cost of DG; is the installed capacity of DG at node i; is a 0-1 variable, indicating whether the ith candidate node is connected to DG.

[0071] The constraint conditions include:

[0072] 1) DG installation capacity constraint

[0073]

[0074] wherein: is the maximum installed capacity of DG at node i.

[0075] 2) DG operation constraint

[0076]

[0077] wherein: are the upper and lower limits of DG active and reactive power output at node i in scenario s and period t, respectively.

[0078] 3) DG transaction power balance constraint

[0079]

[0080] wherein: is the transaction power of DG and benefit subject m at node i in s scenario t period.

[0081] Energy storage operator: The energy storage operator maximizes its annual comprehensive income by arbitrage of low storage and high sale of electricity, determines the installation location and capacity of the energy storage device. The objective function is as follows:

[0082]

[0083] wherein: C ESS , is the annual comprehensive income, annual electricity sale income, new energy subsidy cost, annual investment cost and annual operation and maintenance cost of the energy storage operator, respectively; is the electricity sale price, electricity purchase price, new energy subsidy price and operation and maintenance unit price of the energy storage operator when charging and discharging per unit of electricity, respectively; Ω ESS is the set of energy storage candidate installation nodes; is the investment price per unit capacity of the energy storage; is the installation capacity of the energy storage; is the investment decision variable of the energy storage.

[0084] The constraint conditions include:

[0085] 1) Energy storage installation capacity constraint

[0086]

[0087] wherein: is the maximum installation capacity of ESS.

[0088] 2) Energy storage operation constraint

[0089]

[0090] wherein: represents the charging and discharging state of the energy storage at node i in s scenario t period, which is a 0-1 variable, 1 for discharging and 0 for charging; is the energy storage capacity at node i in s scenario t period; is the charging and discharging efficiency of the energy storage, respectively; is the upper and lower limit of the energy storage capacity, respectively; is the energy storage capacity at the initial and end time of operation, respectively; Δt is the duration of period t.

[0091] 3) Energy storage transaction power balance constraint

[0092]

[0093] wherein: is the transaction power of the energy storage and the benefit subject m at node i in time period t of scenario s.

[0094] Power consumer: considering both interruptible load and transferable load, the power consumer adjusts the power consumption behavior to maximize the reduction of power consumption cost according to the power price signal and the grid incentive mechanism. The objective function is as follows:

[0095] max C DR = C IL + C TL (35)

[0096]

[0097] wherein: C DR , C IL , C TL are the reduced power consumption cost, the subsidy income of the interruptible load, and the reduced power consumption cost of the transferable load, respectively, after the power consumer participates in the demand response; f IL , f TL are the interruptible load compensation price and the load purchase price. is the interruptible load power, the power transferred out, and the power transferred in at node i in time period t of scenario s.

[0098] 1) Demand response load constraint

[0099]

[0100] wherein: and and and are the upper and lower limits of the interruptible load power and the transferred in / out load power at node i in time period t of scenario s, respectively.

[0101] 2) Power consumer transaction power balance constraint

[0102]

[0103] wherein: is the transaction power of the power consumer and the benefit subject m at node i in time period t of scenario s.

[0104] Further, the power distribution network collaborative planning model of the master-slave game stage leader and follower is as follows:

[0105]

[0106] wherein: C UThe comprehensive benefits of the alliance year.

[0107] Followers form an interest alliance and respond to the decision of the distribution network operator through cooperation. In cooperation, the maximum comprehensive benefit is taken as the goal, and the benefit of the follower after cooperation needs to be ensured not to be damaged. In the cooperative game, the Nash negotiation method can effectively solve the game between multiple stakeholders and effectively depict the cooperative interaction between the source, load and storage stakeholders in the market. The standard model is as follows:

[0108]

[0109] In the formula, F n is the benefit of the subject n after cooperation; is the benefit of the subject n before cooperation, that is, the negotiation breaking point; F is the number of subjects.

[0110] The above standard model is a multivariate coupled non-convex nonlinear problem, which can be decomposed and converted into two sub-problems to be solved in turn. The equilibrium solution of the model represents that the benefits of the negotiating subjects reach the Pareto optimality.

[0111] Further, the lower cooperative game sub-model in this embodiment is respectively a comprehensive cost minimization sub-problem and a benefit distribution maximization sub-problem.

[0112] The comprehensive cost minimization sub-problem (Q1) is as follows:

[0113] The comprehensive cost is the sum of the electric energy costs of the DG operator, the energy storage operator and the power user, and the objective function is:

[0114]

[0115] In the formula, C n is the annual comprehensive cost of the subject n.

[0116] The constraint conditions are as follows:

[0117] 1) Trading power constraint

[0118] |P n-m,s,t |≤P n-m,s,max (47)

[0119] In the formula, P n-m,s,t , P n-m,s,max are the trading power between the interest subject n and the interest subject m and the upper limit of the trading power.

[0120] 2) Electric energy trading balance constraint

[0121]

[0122] In the formula, P n,s,t is the total output power of the interest subject n in the s scenario t period.

[0123] The profit distribution maximization sub-problem (Q2) is as follows:

[0124] After the comprehensive cost minimization is achieved, the profit distribution is considered according to the electricity contribution of each subject in the cooperation. The greater the electricity contribution, the greater the bargaining power of the subject in the profit distribution. It is generally believed that providing electricity contributes more than consuming electricity. An electricity mapping function is established based on an exponential function of a natural constant to measure the bargaining power:

[0125]

[0126]

[0127] In the formula: and are the total energy provided and consumed by the profit subject n, respectively; and are the maximum energy provided and consumed by the alliance subject, respectively; a n is the bargaining power of the profit subject n.

[0128] The profit distribution maximization sub-problem aims to maximize the income increment after the subject participates in the cooperation compared to before the cooperation, and is converted into a minimum value problem in the form of a logarithmic function of a strictly monotonically increasing convex function:

[0129]

[0130] In the formula: is the income of the profit subject n after cooperation; is the income of the profit subject n before cooperation; w n-m,s,t is the transaction electricity price between the profit subject n and the profit subject m in the s scenario t period.

[0131] The constraint conditions are as follows:

[0132] 1) Cooperation income constraint

[0133] Each subject participating in cooperation and contributing can obtain income.

[0134]

[0135] 2) Transaction electricity price constraint

[0136] w s,t,min ≤w n-m,s,t ≤w s,t,max (55)

[0137] In the formula: w s,t,max , w s,t,min are the upper and lower limits of the transaction electricity price in the s scenario t period, respectively.

[0138] III. Solution strategy

[0139] In this embodiment, when the power distribution network collaborative planning model is solved, firstly, the analytical target cascading (ATC) method is used to decouple the double-layer model, and then the alternating direction method of multipliers (ADMM) is used to solve the lower-layer cooperative game sub-model in a distributed manner. The lower-layer sub-problem is divided into multiple independent optimization models, each subject can solve autonomously and reach global convergence through alternating iteration, as shown in reference Figure 2 .

[0140] Since the interaction between the power distribution network operator and other interest subject alliance exists in the upper-layer model and Q1, Q2 only determines the transaction price according to the solution of the transaction power between the alliance subjects. In the interaction process, the solution of Q2 does not affect the solution of Q1, nor is it passed to the upper-layer problem. Therefore, the coupling of the double-layer model only exists in the upper-layer problem and Q1, and the following consistency constraint is met:

[0141]

[0142] In the formula: is the power expected to be traded between the interest subject alliance and the power distribution network operator; is the power expected to be traded by the power distribution network operator.

[0143] The constraint is decoupled, and a corresponding penalty term is added to the objective function to relax the consistency constraint. The decoupled target is:

[0144]

[0145] In the formula: k, k o is the number of internal and external loop iterations of the target cascading analysis method.

[0146] The external loop converges when the following condition is met:

[0147]

[0148] To ensure the convergence of the model, the penalty factor can be updated at the end of the external iteration:

[0149]

[0150] In the formula: σ is usually 1<σ<3.

[0151] After decoupling, the power distribution network operator optimizes the network structure in the upper-layer problem, and the interest subject alliance optimizes the configuration scheme and power consumption mode in the lower-layer problem. The power information is exchanged between the upper-layer and lower-layer models to guide the planning decisions of each subject.

[0152] Further, in the two sub-problems of the lower layer model, the power transaction balance and the price balance are coupled, auxiliary variables can be introduced to convert the original constraints into double-coupled constraints. The equality holds indicates that cooperation is reached between the two subjects.

[0153]

[0154] In the formula: P m-n,s,t is the transaction power between the interest subject m and the interest subject n; w m-n,s,t is the transaction price between the interest subject j and the interest subject i.

[0155] After decoupling, the alternating direction multiplier method can be used to solve the two sub-problems in a distributed manner. First, solve Q1 to get the transaction power between each subject, and then solve Q2 to get the expected transaction price of the subject. The specific steps are as follows:

[0156] 1) Establish the augmented Lagrangian function of the sub-problem.

[0157]

[0158] In the formula: is the Lagrange multiplier; is the penalty factor.

[0159] 2) According to the expected transaction power / price information provided by other subjects, each subject updates its decision independently and transmits its expected transaction power / price to other subjects. Repeat the process until each cooperative subject updates its decision.

[0160]

[0161] In the formula: k is the iteration number.

[0162] 3) Update the Lagrange multiplier.

[0163]

[0164] 4) Adaptive penalty coefficient adjustment.

[0165] The value of the penalty coefficient will affect the convergence speed and the feasibility of the solution. In order to improve the convergence speed of the alternating direction multiplier method, the primal feasibility and the dual feasibility are decreased at a consistent speed, and the penalty coefficient can be dynamically adjusted at each iteration. The present application introduces an adaptive adjustment mechanism to dynamically correct the penalty coefficient:

[0166]

[0167] In the formula: are the original residual error and dual residual error of the kth iteration of Q1; δ>1, τ1>1, τ2>1, and δ=10, τ1= τ2=2 are usually taken. Similarly, the dynamic adjustment can be made

[0168] 5) If the condition is met, the kth iteration converges, and the iteration is terminated.

[0169]

[0170] 6) If the iteration number k>kmax, the algorithm does not converge, and the iteration is terminated; if not, let k=k+1, and return to step 2) to continue iteration and solving. max

[0171] If the above method is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0172] To verify the effectiveness of the source-network-load-storage planning method proposed in the present application, an example is set based on an IEEE 33-node system. Due to the increasing load demand, new load nodes 34-37 are added to the distribution network, and the load data is shown in Table 1. After the increase, the maximum load of the system is 4175 kW+j2560 kvar. The purchase price of the distribution network operator from the upper-level power grid is 0.38 yuan / (kW·h). The new line to be built is shown by the Figure 3 red dashed line in the middle, and the length and other data of the alternative line are shown in Table 2. The new line construction cost is 100,000 yuan / km, and the resistance and reactance are 0.25 Ω / km and 0.4 Ω / km. The planning period is 10 years, and the discount rate is 6%.

[0173] The related parameters of the DG and the energy storage are shown in Table 3. It is assumed that the user side is a transferable load and participates in the price-type demand response. The time-of-use price is shown in Table 4. The 25th node of the system is an interruptible load node, the interruptible time is 11:00-22:00, and the interruptible load subsidy is 0.4 yuan / (kW·h). Based on the annual historical data of a certain distribution network, the typical day scenarios are obtained by Gaussian mixture clustering, as shown in Figure 4 .​

[0174] Table 1 new node load data

[0175]

[0176] Table 2 alternative line parameters

[0177]

[0178] Table 3 related parameters

[0179]

[0180]

[0181] Table 4 electricity price parameters

[0182]

[0183] To verify the effectiveness of the power distribution network source network load storage collaborative planning method based on mixed game proposed in the application, the following three methods are used to solve and compare the planning results.

[0184] Method 1: the method of the application, that is, considering the master-slave-cooperative multi-agent mixed game double-layer planning.

[0185] Method 2: power distribution network planning method based on master-slave game. That is, only considering the master-slave game between the power distribution network operator and other subjects, at this time the planning goal is the maximum comprehensive benefit of the power distribution network operator and the source load storage subject alliance.

[0186] Method 3: traditional power distribution network collaborative planning method. That is, the power distribution network source network load storage planning without considering the game relationship between multiple subjects, at this time the planning goal is to maximize the overall benefit of the planning.

[0187] Methods 1, 2 and 3 compare and analyze the necessity of considering multi-agent game in power distribution network planning. Methods 1 and 2 compare and analyze the results of interest subjects participating in planning under different game relationships, and verify the effectiveness of the cooperation of the game followers on the source load storage side.

[0188] The planning results of different methods are shown in Table 5. In method 1, the DG operator has the largest access capacity of photovoltaic and wind turbine, because when considering the interests of the subject itself, if the wind and light can be consumed locally, the DG operator will increase the investment of DG to increase the operating income. In method 3, the power distribution network operator has less power purchase cost from the upper grid, so the demand for DG is less. From the line planning results, it can be seen that the length of each new line in method 3 is the shortest, because for the planning goal of maximizing the overall benefit, under the condition of ensuring system safety and flexibility of resources, the smaller the line impedance, the smaller the loss, and the greater the investment and operating income. In methods 1 and 2, there is a game between the power distribution network operator and other subjects, and the power generation and consumption decisions of each subject will affect the selection scheme of the power distribution network for new lines to some extent. For example, in method 1, node 21 has distributed wind power access, and for the DG operator, connecting node 35 at node 21 or 22 can expand the radial network, which is conducive to the transaction between the DG operator and the power user. At the same time, the line 22-35 is shorter than the line 21-35, so the final line construction scheme is 22-35. From the planning results of energy storage, the access capacity of energy storage in method 1 is the largest. In method 1, the energy storage operator can cooperate with the DG operator and the power user to purchase power at a lower price, and sell it to the nearby load at a higher price to promote the use of energy storage devices and increase income. From the above analysis, under the influence of considering the multi-subject game, the grid-connected capacity of distributed energy after planning increases, which helps to reduce the dependence of the power distribution network on the upper grid, promotes energy utilization, and reduces carbon emissions.

[0189] Table 5 Planning results of different methods

[0190]

[0191] The planning annual comprehensive income / expense of each benefit subject obtained by different methods is shown in Table 6. It can be seen that the income / expense of each operator in method 2 is higher than that in method 3. In method 1, the planning income of the DG operator and the power distribution network operator is smaller, because the benefit subjects have bargaining power, and during planning, each operator can propose a price demand according to the expected electricity interaction to reach a consensus, and the electricity price at this time is often lower than the initial pricing to promote electricity interaction between the two parties. However, for the whole, the annual comprehensive total cost of the source, network, load and storage side subjects planned in method 1 is 511.11 million yuan, which is 9.47 million yuan and 26.67 million yuan less than that in methods 2 and 3 respectively. Therefore, in the planning, considering the multi-subject can maximize the overall benefit by sacrificing the interests of some subjects, increase the enthusiasm of other subjects to participate in the construction of the power distribution network, and promote the rational allocation of resources.

[0192] Table 6 Planning income / expense of benefit subjects

[0193]

[0194] The economic efficiency of each subject in the planning under the mixed game mode is analyzed by comparing the costs of each subject under different methods in combination with the planning results.

[0195] The benefits and costs of the distribution network operator under different planning methods are shown in Table 7. It can be seen that the planning considering the mixed game reduces the power purchase of the distribution network from the upper grid, reduces the power selling benefit, and slightly increases the investment cost. The simulation trading results of the distribution network operator in the typical day scenario 2 are shown in Table 7. It can be seen that the distribution network mainly obtains benefits by selling power to power users, and has less interaction with DG and energy storage. Therefore, the distribution network selling benefit is less than that of other methods. In method 1, the large access of distributed energy to a certain extent alleviates the problem of increased network loss of the distribution network caused by the influence of the line planning of the cooperative subject. Therefore, the cooperation of the source, load and storage subjects will reduce the planning benefit of the distribution network operator, but from the perspective of the whole system, it can reduce the dependence of the distribution network on the upper grid and improve the power supply flexibility. Figure 5

[0196] Table 7 Benefits and costs of the distribution network operator

[0197]

[0198] The benefits and costs of the DG operator are shown in Table 8. In method 1, the access capacity of DG is the largest, so the investment cost is higher. In the operation transaction, the transaction power of method 1 is the largest, and due to the influence of the cooperative game with other alliance subjects, the DG selling price is lower, so the selling benefit is lower than that of method 2.

[0199] Table 8 Benefits and costs of the DG operator

[0200]

[0201] The benefits and costs of the energy storage operator are shown in Table 9. In method 1, the capacity of the energy storage is increased, and the charging and discharging capacity of the energy storage is increased, so as to effectively utilize the transaction with the DG operator and the arbitrage of low storage and high generation to improve the operation benefit. Therefore, through the cooperation with the source and load subjects, the investment capacity of the energy storage device can be increased, the comprehensive benefit of the energy storage operator can be improved, and the storage side subjects can actively participate in the planning.

[0202] Table 9 Benefits and costs of the energy storage operator

[0203]

[0204] ​The power user's various fees are shown in Table 10. In the three methods, the interruptible load actively participates in the response, and the maximum interruptible power is adopted to reduce the power purchase cost and obtain the interruptible subsidy.

[0205] Table 10: Power user's various fees

[0206]

[0207] Taking the typical day scenario 2 as an example, the demand response load power of the power user in method 1 and the simulation transaction results are shown in Figure 6 and Figure 7 respectively. The peak load period is 11:00-13:00 and 18:00-22:00, at which time the power user shifts the load to the valley period and the normal period to reduce the power cost. At the same time, the power user interrupts part of the load according to the power demand from 11:00 to 22:00. As shown in Table 10, the power user in method 1 reduces the power cost the most, which is 3,240,500 yuan less than that in method 2. This is because in the cooperative game, the power user can preferentially trade with other subjects except the distribution network operator and participate in the negotiation to determine the power purchase price. As shown in Figure 7 , in the period with less DG power generation, the power user has greater bargaining power, so it can purchase power at a lower price than the original DG operator's power selling price. In the period from 8:00 to 19:00, the photovoltaic power generation increases, at which time the DG operator has stronger bargaining power, and the photovoltaic power selling price increases. When the DG power generation cannot meet the demand, the power user will purchase part of the power from the energy storage operator at a lower price than the time-of-use price. Based on this, through the cooperation of the alliance of interest subjects, the economic benefit of the load side in the planning can be effectively improved, and the enthusiasm of the power user can be improved.

[0208] The convergence of the algorithm used in the application is shown in Figure 8 , in the external loop of the target cascade analysis algorithm, 8 iterations are performed to achieve convergence. In the last external loop, the sub-problem Q1 is solved by the adaptive coefficient alternating direction multiplier method, and the dual residual converges to within 10-1 after 25 iterations. At this time, under the decision of the line planning and selection and capacity, the power transaction consensus between the interest subjects is reached. In the interest distribution stage, the sub-problem Q2 converges successfully after 10 iterations.

[0209] To verify the effectiveness of dynamically updating the penalty coefficient through the adaptive adjustment mechanism in the solving process of the alternating direction multiplier method, taking the sub-problem Q1 as an example, different initial penalty coefficients are set for comparison. Considering that if the penalty coefficient is not selected properly, the classical alternating direction multiplier method takes too long to solve, according to ρ=10 2The adaptive coefficient alternating direction multiplication method converges after 25 iterations, and the maximum number of iterations is set to 25. Analysis ρ = 10 2 and ρ = 10 4 Under the two values, the two solution methods affect the change of the objective function value of each subject. The results are as follows: Figure 9 As shown in the figure, under the classical alternating direction multiplier method, ρ=10 2 When ρ = 10, the objective function does not reach the optimal value after 25 iterations and converges slowly, indicating that the penalty coefficient is too small at this time; 4 When , the objective function value shows an oscillating characteristic in the early iteration process, indicating that the penalty coefficient is too large. For the adaptive coefficient alternating direction multiplier method, the initial value of ρ is 10 2 When the objective function value of each subject converges faster than the classical alternating direction multiplier method; the initial value of ρ is 10 4 , the oscillation process in the iteration is weakened. This is because, if ρ is small, dynamically increasing ρ can increase the weight of the penalty term of the objective function and accelerate the convergence process; if ρ is large, dynamically reducing ρ can prevent the Lagrange multiplier from growing too fast, thereby weakening the oscillation. Therefore, by adaptively adjusting the penalty coefficient, the initial value of ρ can be set within a large range, effectively avoiding the difficulty in solving the problem caused by improper selection of the penalty coefficient. In addition, in the bi-level programming model established by the present invention, after reaching the convergence condition of Q1, the convergence requirements of the outer loop must also be met. Therefore, dynamically adjusting the penalty coefficient during the alternating iteration process can reduce the number of internal loops in each outer loop process and improve computational efficiency.

[0210] In other embodiments, a source-grid-load-storage collaborative planning system considering multi-agent hybrid games may be provided, comprising one or more processors, a memory, and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the source-grid-load-storage collaborative planning method considering multi-agent hybrid games as described above.

[0211] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A source-grid-load-storage collaborative planning method considering multi-agent hybrid game, characterized by: The following steps are involved: A distribution network collaborative planning model is established that considers a master-slave-cooperative hybrid game among source, grid, load, and storage entities. The source, grid, load, and storage entities include distribution network operators, distributed power generation operators, energy storage operators, and electricity users. The distributed power generation operators, energy storage operators, and electricity users constitute the source, load, and storage side entities. The distribution network collaborative planning model is a two-layer model, including an upper-layer master-slave game sub-model between the distribution network operator and other entities, and a lower-layer cooperative game sub-model between the source, load, and storage side entities. The distribution network collaborative planning model is solved by combining the target cascade analysis method and the adaptive coefficient alternating direction multiplier method to obtain the optimal planning scheme for new line construction, distributed power generation and energy storage devices. The upper-level master-slave game sub-model is a planning and decision-making model dominated by distribution network operators, with the goal of maximizing the annual comprehensive revenue of distribution network operators. The constraints considered include system power flow constraints, system operation safety constraints, new line constraints, and transaction power balance constraints. The lower-level cooperative game sub-model includes a comprehensive cost minimization sub-problem and a benefit distribution maximization sub-problem. The comprehensive cost minimization sub-problem aims to minimize the sum of the electricity costs of distributed power generation operators, energy storage operators, and power users, and the considered constraints include transaction power constraints and electricity transaction balance constraints. The benefit distribution maximization sub-problem aims to maximize the incremental benefits of each subject after participating in the cooperation compared to before the cooperation, and the considered constraints include cooperation benefit constraints and transaction electricity price constraints. When solving the distribution network collaborative planning model, the target cascade analysis method is first used to decouple the two-layer model, and then the adaptive coefficient alternating direction multiplier method is used to perform a distributed solution to the lower-layer cooperative game sub-model; In the target cascade analysis method, the end condition of the outer loop iteration is: Where, The power that the stakeholder alliance expects to trade with the distribution network operator; is the power that the distribution network operator expects to trade, with superscripts k and k o is the number of iterations of the internal and external loops of the target cascade analysis method, N s is the number of scenes, T is the number of time periods, ε o is the convergence threshold; The distributed solution includes: first solving the comprehensive cost minimization sub-problem to obtain the transaction power between each subject, and then solving the benefit distribution maximization sub-problem to obtain the subject's expected transaction electricity price; The distributed solution specifically includes: Establish the augmented Lagrangian function of each subproblem; Based on the expected transaction volume / price information provided by other entities, each entity independently updates its decision and transmits the updated expected transaction volume / price information to other entities; Update the Lagrange multiplier and adaptively adjust the penalty factor; After the iteration termination condition is met, the final solution is obtained.

2. The source-grid-load-storage collaborative planning method considering multi-agent hybrid game according to claim 1 is characterized in that: The objective function of the benefit distribution maximization sub-problem is expressed as: Where, The benefits after cooperation between stakeholders n; C is the benefit of the stakeholder n before cooperation; n is the annual comprehensive cost of entity n; Φ is the number of entities; α n is the bargaining power of the stakeholder n, which is obtained based on the power mapping function.

3. The source-grid-load-storage collaborative planning method considering multi-agent hybrid game according to claim 1 is characterized in that: In the target cascade analysis method, the penalty factor is updated at the end of the outer loop iteration. The update formula is: Where σ is the update constant, kth o 、k o +1 penalty factor for outer loop iteration.

4. A source-grid-load-storage collaborative planning system considering multi-agent hybrid game, characterized by: It includes one or more processors, a memory and one or more programs stored in the memory, and the one or more programs include instructions for executing the source-grid-load-storage collaborative planning method considering multi-agent hybrid game as described in any one of claims 1-3.

Images