CN117526426A - Two-stage distributed optimization method and system for AC/DC power distribution network - Google Patents

Two-stage distributed optimization method and system for AC/DC power distribution network Download PDF

Info

Publication number
CN117526426A
CN117526426A CN202410006126.3A CN202410006126A CN117526426A CN 117526426 A CN117526426 A CN 117526426A CN 202410006126 A CN202410006126 A CN 202410006126A CN 117526426 A CN117526426 A CN 117526426A
Authority
CN
China
Prior art keywords
distribution network
power distribution
power
distributed
operation optimization
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202410006126.3A
Other languages
Chinese (zh)
Other versions
CN117526426B (en
Inventor
乔立
王俊琪
贺继峰
徐秋实
张焱哲
王博
李子寿
卢子敬
莫石
吴启亮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Hubei Electric Power Co Ltd
Economic and Technological Research Institute of State Grid Hubei Electric Power Co Ltd
Original Assignee
State Grid Hubei Electric Power Co Ltd
Economic and Technological Research Institute of State Grid Hubei Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by State Grid Hubei Electric Power Co Ltd, Economic and Technological Research Institute of State Grid Hubei Electric Power Co Ltd filed Critical State Grid Hubei Electric Power Co Ltd
Priority to CN202410006126.3A priority Critical patent/CN117526426B/en
Publication of CN117526426A publication Critical patent/CN117526426A/en
Application granted granted Critical
Publication of CN117526426B publication Critical patent/CN117526426B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/381—Dispersed generators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
    • H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/28—Arrangements for balancing of the load in a network by storage of energy
    • H02J3/32—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
    • H02J3/48—Controlling the sharing of the in-phase component
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
    • H02J3/50—Controlling the sharing of the out-of-phase component
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J5/00—Circuit arrangements for transfer of electric power between ac networks and dc networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/04—Power grid distribution networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/06—Power analysis or power optimisation
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/40—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation wherein a plurality of decentralised, dispersed or local energy generation technologies are operated simultaneously

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

A two-stage distributed optimization method and system for an AC/DC power distribution network comprises the steps of firstly constructing a multi-period operation optimization model of the centralized AC/DC power distribution network, then converting the centralized operation optimization model into a distributed operation optimization model of each sub-network, solving to obtain an optimal solution of boundary coupling variables, constructing an independent operation optimization model of each sub-network based on the optimal solution of the boundary coupling variables, performing linear relaxation on the independent operation optimization model, solving to obtain an optimal solution of a charge/discharge state of an energy storage device, then calculating to obtain an acceleration induction factor of the AC/DC power distribution network based on the optimal solution of the charge/discharge state of the energy storage device, adding the acceleration induction factor into each sub-network independent acceleration model to form each sub-network independent acceleration model, and finally solving the acceleration model to obtain an operation optimization result of the AC/DC power distribution network. The invention effectively improves the convergence and solving efficiency of the model.

Description

Two-stage distributed optimization method and system for AC/DC power distribution network
Technical Field
The invention belongs to the field of power distribution network optimization operation, and particularly relates to a two-stage distributed optimization method and system for an alternating current/direct current power distribution network.
Background
The development of renewable energy technology provides new challenges and higher requirements for flexible access and effective regulation of a power system, and along with continuous access of high-permeability distributed power generation, a traditional alternating-current power distribution network is gradually changed into an alternating-current and direct-current hybrid power distribution network. The power electronic transformer (Power electronic transformer, PET) is a device combining a power electronic device and a high-frequency transformer, and can realize the functions of power grid interconnection, new energy device grid connection, electric energy routing and the like. The open networking form is provided with the characteristics of regional coordination, alternating current-direct current series-parallel connection, multidirectional tide and the like, and can realize the interconnection and complementation of renewable energy sources in a larger range; the plug-and-play AC/DC interface integrates the distributed power supply at a plurality of voltage levels, thereby remarkably reducing the conversion links, realizing flexible regulation and control of tide and being beneficial to the efficient utilization of renewable energy sources. The PET-containing alternating current-direct current hybrid power distribution network provides a new means for renewable energy consumption and becomes a future development trend.
In order to fully exert the flexible regulation and control capability of PET, researchers provide optimal power flow, optimal scheduling and reactive power optimization models of the power distribution network containing PET so as to realize flexible regulation and control of network power flow, operation cost reduction and reactive voltage support. However, researches on the coordination and optimization of multiple regional subnetworks of the PET-containing alternating current-direct current hybrid power distribution network are fresh. Generally, a PET-based alternating current-direct current hybrid power distribution network can realize flexible networking in a larger range, but with the continuous expansion of the scale of an interconnection network, the traditional centralized optimization operation method is difficult to succeed, and the specific reasons are as follows: 1) The problem of mass information, the large scale of distributed power generation and the network thereof, causes the rapid increase of the collected information quantity; 2) The complex problem of the model, the network scale is enlarged to increase the dimension of the decision variable, the optimization model becomes extremely complex, and the solution is difficult to be effectively carried out; 3) The safety problem is that the centralized operation control system can cause the operation breakdown of the whole network as long as a single point of failure occurs.
To achieve distributed operation of a multi-zone autonomous network, scholars have proposed a distributed optimization algorithm comprising: auxiliary problem principle methods, alternate direction multiplier methods, proximity-centric methods, and target cascade analysis methods (Analytical target cascading, ATC), etc., which have been widely used in the field of power system distributed decision making. The ATC belongs to a method for coordinating a hierarchical structure, wherein each element in the hierarchical structure is allowed to independently make a decision, and a parent element coordinates and optimizes the decision of a child element to obtain an overall optimal solution of the problem. Compared with other optimization methods, the ATC method has the advantages of parallel optimization, unrestricted progression, strict convergence proof and the like, and therefore, the ATC method can be applied to distributed operation optimization of the power distribution network. However, for a distributed optimization model of a power distribution network containing energy storage, the charging and discharging modes of the energy storage relate to integer variables, and when the optimization operation period is more, the solving efficiency and convergence of the distributed optimization model can be greatly affected, so that an accurate and efficient solving method is needed.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provides a two-stage distributed optimization method and a system for an AC/DC power distribution network, which can improve the convergence and the calculation efficiency of a model.
In order to achieve the above object, the technical scheme of the present invention is as follows:
in a first aspect, the present invention provides a two-stage distributed optimization method for an ac/dc power distribution network, including:
s1, constructing a multi-period operation optimization model of a centralized AC/DC power distribution network, wherein the multi-period operation optimization model of the centralized AC/DC power distribution network aims at the minimum operation cost, and constraint conditions comprise AC/DC power distribution network tide constraint, power electronic transformer operation constraint, energy storage device operation constraint and distributed power supply operation constraint;
s2, converting the multi-period operation optimization model of the centralized AC/DC power distribution network into a distributed operation optimization model of each subnet, and solving the distributed operation optimization model of each subnet to obtain an optimal solution of the boundary coupling variable;
s3, constructing an independent operation optimization model of each sub-network based on the optimal solution of the boundary coupling variable, and obtaining an operation optimization result of the AC/DC power distribution network by solving the independent operation optimization model of each sub-network.
In the step S1, the objective function of the multi-period operation optimization model of the centralized ac/dc power distribution network includes:
；
in the above-mentioned method, the step of,、/>the number of distributed power supplies in the AC power distribution network and the DC power distribution network is +.>、/>The power generation cost of the distributed power supply i in the AC and DC power distribution networks is +. >、/>Active power of a distributed power supply i in an alternating current power distribution network and a direct current power distribution network in t time periods respectively,/->The method is characterized in that the method is used for purchasing electricity cost of an alternating current/direct current distribution network>Active power injected into the main network at the t period;
in the step S3, the independent operation optimization model of each subnet includes:
independent operation optimization model of communication sub-network:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Continuous and discrete variables in the middle of local variables of the alternating current power distribution network respectively>Is a real number;
DC subnet independent operation optimization model:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable are respectively in the middle of the local variable of the direct current power distribution network.
The step S3 further includes:
s31, performing linear relaxation on the independent operation optimization model of each sub-network, and solving to obtain an optimization solution of the charge and discharge states of the energy storage device in the AC/DC power distribution network;
s32, optimizing solution calculation based on the charge and discharge states of the energy storage device to obtain acceleration induction factors of the AC/DC power distribution network, and adding the acceleration induction factors into the independent operation optimization models of all the sub-networks to form independent operation optimization acceleration models of all the sub-networks;
S33, solving an independent operation optimization acceleration model of each sub-network to obtain an operation optimization result of the AC/DC power distribution network.
In S32, the acceleration induction factor of the AC/DC power distribution networkCalculated based on the following formula:
；
in the above-mentioned method, the step of,for scaling parameters, ++>An optimized solution for the charge and discharge states of the energy storage device;
each subnet independent operation optimization acceleration model comprises:
and (3) an independent operation optimization acceleration model of the alternating current sub-network:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < >>For the active power of the distributed power source i in the alternating current distribution network of the t period +.>Acceleration induction factor for AC distribution network>、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Continuous, discrete variable in local variable of AC distribution network>Is a real number;
DC subnetwork independent operation optimization acceleration model:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>For the number of distributed power sources in a dc distribution network,generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/ >Acceleration induction factor for a DC power distribution network, +.>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable are respectively in the middle of the local variable of the direct current power distribution network.
In the step S2, the distributed operation optimization model of each subnet includes:
operation optimization model of power electronic transformer subnetwork:
；
in the above-mentioned method, the step of,for the objective function of the power electronic transformer subnetwork, < >>The method is characterized in that the method is used for purchasing electricity cost of an alternating current/direct current distribution network>Active power injected into t-period main network, < >>、/>Coefficient vector and weight vector of Lagrange penalty function of alternating current power distribution network respectively>、/>Coefficient vectors and weight vectors of the Lagrange penalty function of the direct current power distribution network are respectively adopted,、/>target variables which are boundary coupling variables of the AC/DC power distribution network respectively, < >>、/>Response variables, which are boundary coupling variables of the ac and dc power distribution network, respectively,/->、/>Equality and inequality constraints of the sub-network of the power electronic transformer, respectively, < >>Is a local variable of a power electronic transformer subnet;
operation optimization model of ac subnetwork:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < > >For the active power of the distributed power source i in the alternating current distribution network of the t period +.>、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Respectively continuous and discrete variables in the local variables of the alternating current power distribution network;
operation optimization model of direct current subnetwork:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>Is a distributed power supply in a direct current distribution networkIs used in the number of (a) and (b),generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable are respectively in the middle of the local variable of the direct current power distribution network.
The alternating current and direct current power distribution network power flow constraint comprises the following steps:
；；
；
；；
in the above-mentioned method, the step of,、/>branch sets of an AC power distribution network and a DC power distribution network respectively, < ->、/>Active power flows of branches ji and ik in t-period alternating current distribution network respectively>、/>Reactive power flows of branches ji and ik in t-period alternating current distribution network respectively>Reactive power output of distributed power supply i in t-period alternating current power distribution network>、/>Active power and reactive power of an alternating-current power distribution network node i are respectively injected into a power electronic transformer PET at t time intervals,/- >、/>The power of the discharging and charging of the alternating current side energy storage device e in the t period is +.>、/>Active and reactive loads of the nodes i of the alternating-current power distribution network in t time periods respectively, < >>、/>For the voltage amplitude of the nodes i, j of the ac distribution network in period t,/>、/>Resistance, reactance, < >/of the ac distribution network branch ij, respectively>、/>For the active power flow of the branches ji, ik of the direct current distribution network in the period t,/>Active power of a direct-current power distribution network node i is injected into a power electronic transformer in a t period, and the power electronic transformer is +.>、/>The power of the discharging and charging of the direct current side energy storage device e in the t period is respectively +.>Active load of node i of direct current power distribution network in t period,/, for>、/>Voltage amplitude of nodes i and j of direct-current power distribution network in t time period, < >>The resistor is the resistance of the branch ij of the direct-current power distribution network;
the power electronic transformer operating constraints include:
；
；
；
；
in the above-mentioned method, the step of,、/>the number of the AC and DC ports of the power electronic transformer is +.>Maximum reactive power allowed to be output for ac port i of the power electronic transformer, +.>Capacity of ac port i for power electronic transformer, +.>Maximum active power allowed to be output for a direct current port i of the power electronic transformer;
the energy storage device operational constraints include:
；
；
；
；
；
in the above-mentioned method, the step of,、/>binary variables of e-discharging and charging states of the energy storage device in t time period respectively, < > >、The power of the energy storage device e is respectively equal to the power of the energy storage device e in the t period>，，/>、/>Maximum power allowed for the energy storage device e, respectively, < >>For t time period the charge capacity of the energy storage device e, < >>、/>The minimum and maximum charge capacity of the energy storage device e are respectively +.>、/>The charge efficiency of the energy storage device is +.>Is the capacity of the energy storage device e;
the distributed power operation constraint includes:
；
；
；
in the above-mentioned method, the step of,、/>respectively outputting the minimum and maximum active power of the alternating current distributed power supply, +.>、/>Minimum and maximum values of reactive power output by the alternating current distributed power supply respectively, < >>、And outputting the minimum and maximum active power values of the direct current distributed power supply respectively.
In a second aspect, the invention provides a two-stage distributed optimization system of an AC/DC power distribution network, which comprises a centralized operation optimization model construction module, a distributed conversion and solving module and a subnet independent model construction and solving module;
the centralized operation optimization model construction module is used for constructing a multi-period operation optimization model of the centralized AC/DC power distribution network with minimum operation cost as a target, and constraint conditions of the multi-period operation optimization model of the centralized AC/DC power distribution network comprise AC/DC power distribution network tide constraint, power electronic transformer operation constraint, energy storage device operation constraint and distributed power supply operation constraint;
The distributed conversion and solving module is used for converting the multi-period operation optimization model of the centralized AC/DC power distribution network into a distributed operation optimization model of each subnet and solving the distributed operation optimization model of each subnet to obtain an optimal solution of the boundary coupling variable;
the subnet independent model construction and solving module is used for constructing an operation optimization model of each subnet independent based on the optimal solution of the boundary coupling variable, and the operation optimization result of the AC/DC power distribution network is obtained by solving the operation optimization model of each subnet independent.
The centralized operation optimization model construction module is used for constructing the following objective functions:
；
in the above-mentioned method, the step of,、/>the number of distributed power supplies in the AC power distribution network and the DC power distribution network is +.>、/>The power generation cost of the distributed power supply i in the AC and DC power distribution networks is +.>、/>Active power of a distributed power supply i in an alternating current power distribution network and a direct current power distribution network in t time periods respectively,/->The method is characterized in that the method is used for purchasing electricity cost of an alternating current/direct current distribution network>Active power injected into the main network at the t period;
the subnet independent model construction and solving module is used for constructing the following operation optimization model of each subnet independently:
independent operation optimization model of communication sub-network:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network- >、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Continuous and discrete variables in the middle of local variables of the alternating current power distribution network respectively>Is a real number;
DC subnet independent operation optimization model:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable are respectively in the middle of the local variable of the direct current power distribution network.
The subnet independent model constructing and solving module comprises a subnet independent model constructing unit, an energy storage parameter optimizing unit, an acceleration induction factor calculating unit, a subnet independent acceleration model constructing unit and an acceleration model solving unit;
the subnet independent model building unit is used for building an operation optimization model of each subnet independent based on the optimal solution of the boundary coupling variable;
the energy storage parameter optimization unit is used for performing linear relaxation on the independent operation optimization model of each sub-network and solving to obtain an optimization solution of the charge and discharge states of the energy storage device in the AC/DC power distribution network;
the acceleration induction factor calculation unit is used for obtaining the acceleration induction factor of the AC/DC power distribution network based on the optimization solution calculation of the charge and discharge states of the energy storage device;
The subnet independent acceleration model construction unit is used for adding an acceleration induction factor into each subnet independent operation optimization model to form each subnet independent operation optimization acceleration model;
and the acceleration model solving unit is used for solving the independent operation optimization acceleration model of each subnet to obtain the operation optimization result of the AC/DC power distribution network.
The acceleration induction factor calculation unit calculates the acceleration induction factor of the AC/DC power distribution network based on the following formula：
；
In the above-mentioned method, the step of,for scaling parameters, ++>An optimized solution for the charge and discharge states of the energy storage device;
the subnet independent acceleration model construction unit is used for constructing the operation optimization acceleration model of each subnet independent as follows:
and (3) an independent operation optimization acceleration model of the alternating current sub-network:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < >>For the active power of the distributed power source i in the alternating current distribution network of the t period +.>Acceleration induction factor for AC distribution network>、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Continuous, discrete variable in local variable of AC distribution network >Is a real number;
DC subnetwork independent operation optimization acceleration model:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>For the number of distributed power sources in a dc distribution network,generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/>Acceleration induction factor for a DC power distribution network, +.>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable are respectively in the middle of the local variable of the direct current power distribution network.
The distributed conversion and solving module is used for converting the multi-period operation optimization model of the centralized AC/DC power distribution network into a distributed operation optimization model of each sub-network as follows:
operation optimization model of power electronic transformer subnetwork:
；
in the above-mentioned method, the step of,for the objective function of the power electronic transformer subnetwork, < >>The method is characterized in that the method is used for purchasing electricity cost of an alternating current/direct current distribution network>Active power injected into t-period main network, < >>、/>Coefficient vector and weight vector of Lagrange penalty function of alternating current power distribution network respectively>、/>Coefficient vectors and weight vectors of the Lagrange penalty function of the direct current power distribution network are respectively adopted,、/>target variables which are boundary coupling variables of the AC/DC power distribution network respectively, < > >、/>Response variables, which are boundary coupling variables of the ac and dc power distribution network, respectively,/->、/>Equality and inequality constraints of the sub-network of the power electronic transformer, respectively, < >>Is a local variable of a power electronic transformer subnet;
operation optimization model of ac subnetwork:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < >>For the active power of the distributed power source i in the alternating current distribution network of the t period +.>、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Respectively continuous and discrete variables in the local variables of the alternating current power distribution network;
operation optimization model of direct current subnetwork:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>For the number of distributed power sources in a dc distribution network,generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable are respectively in the middle of the local variable of the direct current power distribution network.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention discloses a two-stage distributed optimization method of an AC/DC power distribution network, which comprises the steps of firstly constructing a multi-time-period operation optimization model of the centralized AC/DC power distribution network, then converting the multi-time-period operation optimization model of the centralized AC/DC power distribution network into a distributed operation optimization model of each subnet, solving the distributed operation optimization model of each subnet to obtain an optimal solution of boundary coupling variables, constructing an independent operation optimization model of each subnet based on the optimal solution of the boundary coupling variables, and obtaining an operation optimization result of the AC/DC power distribution network by solving the independent operation optimization model of each subnet.
2. According to the two-stage distributed optimization method for the AC/DC power distribution network, acceleration induction factors are introduced into the constructed independent operation optimization model of each sub-network, and the solving efficiency of the mixed integer linear programming model of the power distribution network containing energy storage is improved through induction acceleration.
3. The multi-period AC/DC power distribution network operation optimization model constructed by the two-stage distributed optimization method of the AC/DC power distribution network can fully play the role of peak clipping and valley filling of energy storage.
Drawings
Fig. 1 is a block diagram of an ac/dc hybrid distribution network including a power electronic transformer employed in example 1.
Fig. 2 is a schematic diagram of a target cascade assay.
Fig. 3 is a flow chart of a two-stage distributed solution in embodiments 1, 2.
Fig. 4 is a block diagram of the system described in embodiment 3.
Fig. 5 is a block diagram of the system according to embodiment 4.
Detailed Description
The present invention will be described in further detail with reference to the following detailed description and the accompanying drawings.
The invention provides a two-stage distributed optimization method of an alternating current/direct current power distribution network, which aims at the problem of multi-period optimized operation of the power distribution network, fully plays the role of peak clipping and valley filling of energy storage, builds an optimized operation model of the power distribution network with the minimum operation cost as a target, and introduces a target cascading analysis method to convert a centralized optimization model into a distributed model; furthermore, by means of technical means such as relaxation and acceleration, a two-stage acceleration distributed optimization operation method is provided, and the convergence and the calculation efficiency of the model are improved.
Example 1:
the embodiment of the two-stage distributed optimization method for the AC/DC power distribution network aims at the AC/DC hybrid power distribution network comprising the power electronic transformer shown in fig. 1, and sequentially comprises the following steps:
1. The method comprises the following steps of constructing an initial centralized AC/DC power distribution network multi-period operation optimization model with minimum operation cost, wherein the initial centralized AC/DC power distribution network multi-period operation optimization model comprises AC/DC power distribution network tide constraint, power electronic transformer operation constraint, energy storage device operation constraint, safe operation constraint and distributed power supply operation constraint:
objective function
(1)
In the above-mentioned method, the step of,、/>the number of distributed power supplies in the AC power distribution network and the DC power distribution network is +.>、/>The power generation cost of the distributed power supply i in the AC and DC power distribution networks is +.>、/>Active power of a distributed power supply i in an alternating current power distribution network and a direct current power distribution network in t time periods respectively,/->The method is characterized in that the method is used for purchasing electricity cost of an alternating current/direct current distribution network>And (5) injecting active power into the main network at the t period.
Constraint conditions
The network loss is not considered in this embodiment, and thus, the ac/dc power distribution network power flow constraint can be described as:
(2)
(3)
(4)
(5)
(6)
In the above-mentioned method, the step of,、/>branch sets of an AC power distribution network and a DC power distribution network respectively, < ->、/>Active power flows of branches ji and ik in t-period alternating current distribution network respectively>、/>Reactive power flows of branches ji and ik in t-period alternating current distribution network respectively>Reactive power output of distributed power supply i in t-period alternating current power distribution network>、/>Active power and reactive power of an alternating-current power distribution network node i are respectively injected into a power electronic transformer PET at t time intervals,/- >、/>The power of the discharging and charging of the alternating current side energy storage device e in the t period is +.>、/>Active and reactive loads of the nodes i of the alternating-current power distribution network in t time periods respectively, < >>、/>For the voltage amplitude of the nodes i, j of the ac distribution network in period t,/>、/>Resistance, reactance, < >/of the ac distribution network branch ij, respectively>、/>For the active power flow of the branches ji, ik of the direct current distribution network in the period t,/>Active power of a direct-current power distribution network node i is injected into a power electronic transformer in a t period, and the power electronic transformer is +.>、/>The power of the discharging and charging of the direct current side energy storage device e in the t period is respectively +.>Active load of node i of direct current power distribution network in t period,/, for>、/>Voltage amplitude of nodes i and j of direct-current power distribution network in t time period, < >>Is the resistance of the dc distribution network branch ij.
Power electronic transformer operating constraints:
the circuit topology and the control mode of the PET are not consistent, but the external characteristics and the functional attributes contained in the PET are almost the same, as shown in figure 1, the PET in operation needs to satisfy the active balance equation at all times, namely:
(7)
(8)
(9) JavaScript>
(10)
In the above-mentioned method, the step of,、/>the number of the AC and DC ports of the power electronic transformer is +.>Maximum reactive power allowed to be output for ac port i of the power electronic transformer, +. >Capacity of ac port i for power electronic transformer, +.>The maximum active power that is allowed to be output for the dc port i of the power electronic transformer.
Energy storage device operation constraints:
(11)
(12)
(13)
(14)
(15)
In the above-mentioned method, the step of,、/>binary variables of e-discharging and charging states of the energy storage device in t time period respectively, < >>、The power of the energy storage device e is respectively equal to the power of the energy storage device e in the t period>，，/>、/>Maximum power allowed for the energy storage device e, respectively, < >>For t time period the charge capacity of the energy storage device e, < >>、/>The minimum and maximum charge capacity of the energy storage device e are respectively +.>、/>The charge efficiency of the energy storage device is +.>Is the capacity of the energy storage device e.
Safe operation constraint:
(16)
(17)
In the above-mentioned method, the step of,、/>node voltages allowed by the alternating current distribution network are minimum and maximum respectively, < +.>、The node voltages allowed by the direct current distribution network are minimum and maximum respectively.
Distributed power supply operation constraints:
(18)
(19)
(20)/(1)>
In the above-mentioned method, the step of,、/>respectively outputting the minimum and maximum active power of the alternating current distributed power supply, +.>、/>Minimum and maximum values of reactive power output by the alternating current distributed power supply respectively, < >>、And outputting the minimum and maximum active power values of the direct current distributed power supply respectively.
2. The model is essentially a mixed integer nonlinear programming MINLP problem, and for this purpose, variable substitution is adopted、/>And the linearization method converts the MINLP problem into a mixed integer linear programming MILP problem, and then the multi-period operation optimization model of the centralized AC/DC power distribution network is obtained. In this way, formulas 4, 6, 9 are converted into the following linear form:
(21)
(22)
(23)
In the above-mentioned method, the step of,、/>、/>is a constant coefficient.
Notably, because centralized control has the defects of communication bottleneck, commercial confidentiality and the like, and is difficult to widely apply in multiple autonomous areas, the distributed coordination optimization of the AC/DC power distribution network is realized through a decomposition coordination algorithm.
3. The PET-containing AC/DC power distribution network can be divided into an AC sub-network, a DC sub-network and a PET network, wherein the PET network is used for coupling the AC sub-network and the DC sub-network. Considering this hierarchical structure feature, a two-layer decomposition network is constructed based on a target cascade analysis method, as shown in fig. 2.
The objective function of the centralized optimization model is composed of a summation of a plurality of subelements, i.eWherein->As an objective function of a power electronic transformer +.>Is an objective function of an AC/DC power distribution network. Similarly, the constraint can be decomposed into +. >Andwherein->、/>Respectively inequality and equality constraints of the power electronic transformer,、/>the method is characterized in that the method is respectively an inequality constraint condition and an equality constraint condition of the AC/DC power distribution network, and the constraint conditions are linear constraint by adopting the method of linearization of 21-23.Thus, the centralized optimization problem can be broken down into an operational optimization model of the PET subnetwork, an operational optimization model of the ac subnetwork, and an operational optimization model of the dc subnetwork.
The basic idea of the target cascade analysis method is to continuously split the optimization target according to the system-to-subsystem and then to the component and the like, simultaneously continuously feed back the responses of all stages from bottom to top, and then independently solve each subsystem, and overlap and optimize until convergence is met. The optimization variables of the model are generally divided into local variables、/>、/>And boundary coupling variable->Decomposing the boundary coupling variable into the target variable +.>And response variable +.>When the optimization solution is carried into the subsystem, the variables should satisfy the consistency constraint equation:
(24)
The boundary coupling variable is constrained by adopting an augmentation Lagrangian penalty function, so that distributed optimization is realized. On the basis of the centralized optimization model, the following operation optimization model of the PET subnetwork can be decomposed:
(25)
In the above-mentioned method, the step of,、/>coefficient vector and weight vector of Lagrange penalty function of alternating current power distribution network respectively>、Coefficient vector and weight vector of Lagrange penalty function of direct current distribution network respectively, +.>、/>Target variables which are boundary coupling variables of the AC/DC power distribution network respectively, < >>、/>Response variables, which are boundary coupling variables of the ac and dc power distribution network, respectively,/->Is a local variable of the PET sub-network, which is a continuous variable, is an optimized variable of the model,/-for the model>Is a real number.
Similarly, for an ac subnet, its operational optimization model can be expressed as:
(26)
In the above-mentioned method, the step of,for local variables of an ac power distribution network>、/>And the continuous variable and the discrete variable are respectively in the local variables of the alternating current power distribution network.
For a dc subnet, its operational optimization model can be expressed as:
(27)
In the above-mentioned method, the step of,is a local variable of the direct current sub-network, +.>、/>The continuous variable and the discrete variable in the local variable of the direct current sub-network are respectively. In the optimization of the respective subnetwork +.>、/>To optimize the variables +.>、/>Then it is constant.
On the basis of adopting local optimal power flow, each sub-network performs information interaction with PET, continuously feeds back and transmits, and iteratively solves, so that distributed optimization of different networks is realized.
4. Construction of a first-stage model-relaxed distributed operational optimization model
The operation optimization model of the alternating current sub-network and the operation optimization model of the direct current sub-network are combinedBinary variable of (2)、Relaxation to a continuous variable between 0 and 1, i.e. +.>，/>Thus, the following model was used:
relaxed ac subnetwork operation optimization model:
(28)
Relaxed dc subnetwork operation optimization model:
(29)
Through the relaxation treatment, the AC/DC power distribution network sub-problem is converted into a quadratic programming problem, and the feasibility of model solving can be greatly improved because the quadratic programming problem does not contain integer variables.
5. Solving the subnet optimization model of the solution 25, 28, 29 by adopting a target cascade analysis method, so as to perform first-stage solution, adopting a nested solution method, namely, searching an optimal target and a response value by giving a penalty parameter through an inner loop, updating the penalty parameter according to an optimization result of the inner loop by an outer loop, and finally obtaining an optimization solution of a boundary coupling variable (active, reactive, voltage and the like between adjacent subnets), wherein the specific flow is as shown in fig. 3, and comprises:
5.1 setting an inner and outer circulation index、/>And give the target variable sumResponding to an initial value of a variable;
5.2, solving an alternating current and direct current sub-network optimization model of formulas 28 and 29 in parallel, and solving a PET sub-network operation optimization model of formula 25 to obtain an optimization solution of a target variable and a response variable;
5.3, judging whether the internal circulation is converged based on the following formula, if so, entering the step 5.4; otherwise, letAnd then returning to 5.2 for carrying out the next iteration solution:
(30)
In the above-mentioned method, the step of,is->The sum of the objective functions of all sub-network optimization models during the sub-inner loop iteration, < + >>For the convergence tolerance set, 0.001 is selected in this embodiment;
5.4, judging whether the outer loop converges or not based on the following formula, and if yes, outputting an optimized solution of the boundary coupling variable; otherwise, let、/>And updating penalty parameters later, and returning to 5.2 for carrying out next iteration solution:
(31)
(32)
In the above-mentioned method, the step of,is->Difference between target variable and response variable of secondary outer loop, +.>、/>For the convergence tolerance set, 0.001 is selected in this embodiment;
the penalty parameters are updated based on the following formula:
(33)
(34)
In the above-mentioned method, the step of,、/>for a set constant, ++>、/>Respectively +.>Coefficient vector and weight vector of sub-outer loop penalty function, +.>Is Hadamard product.
6. Based on the optimized solution of the boundary coupling variable, a second stage model, namely an independent operation optimization model of each subnet, is constructed, and the method comprises the following steps:
independent operation optimization model of communication sub-network:
(35)
DC subnet independent operation optimization model:
(36)
7. And solving the independent operation optimization model of each sub-network by adopting a Cplex solver, namely performing second-stage solving to obtain an operation optimization result of the AC/DC power distribution network, wherein the operation optimization result comprises the output of an energy storage device and a distributed power supply.
Example 2:
the overall procedure is the same as in example 1, except that,
since the independent operation optimization model of each subnet shown in equations 30 and 31 contains a large number of binary variables, which essentially belongs to the MILP problem, the feasibility and the computational efficiency of model solution are reduced, for this purpose, the embodiment introduces acceleration induction factors into the independent operation optimization model of each subnet constructed in the second stage, and improves the solution efficiency of the model by inducing acceleration, and the specific construction process is shown in fig. 3, and includes:
6.1, in order to obtain the acceleration induction factor, the linear relaxation problem of MILP needs to be established, namely, the linear relaxation is carried out on the independent operation optimization model of each sub-network, so as to obtain the relaxed independent operation optimization model of each sub-network, which comprises the following steps:
relaxed ac subnetwork independent run optimization model:
(37)
Relaxed dc subnetwork independent run optimization model:
(38)
6.2, solving the independent operation optimization model of each loose subnetwork by adopting a Cplex solver to obtain an optimization solution of the charge and discharge states of the energy storage device of the AC/DC power distribution network 。
6.3, calculating an acceleration induction factor of the AC/DC power distribution network based on the following formula：
(39)
In the above-mentioned method, the step of,for scaling parameters, ++>And the method is an optimized solution of the charge and discharge states of the energy storage device.
6.4, willAdding the operation acceleration model to the operation optimization model independent of each subnet to form an operation optimization acceleration model independent of each subnet, comprising the following steps:
and (3) an independent operation optimization acceleration model of the alternating current sub-network:
(40)
In the above-mentioned method, the step of,acceleration induction factors for the alternating current distribution network;
DC subnetwork independent operation optimization acceleration model:
(41)
In the above-mentioned method, the step of,is an acceleration induction factor of a direct current power distribution network.
And solving the independent operation optimization acceleration model of each sub-network through a Cplex solver to finish the second-stage solving and obtain the operation optimization result of the AC/DC power distribution network.
Example 3:
as shown in FIG. 4, the two-stage distributed optimization system for the AC/DC power distribution network comprises a centralized operation optimization model construction module, a distributed conversion and solving module and a subnet independent model construction and solving module.
The centralized operation optimization model construction module is used for:
constructing an initial centralized AC/DC power distribution network multi-period operation optimization model with the minimum operation cost as a target, wherein constraint conditions of the centralized AC/DC power distribution network multi-period operation optimization model comprise AC/DC power distribution network tide constraint, power electronic transformer operation constraint, energy storage device operation constraint and distributed power supply operation constraint;
And carrying out linearization treatment on nonlinear variables in the initial centralized AC/DC power distribution network multi-period operation optimization model, so as to construct and obtain the centralized AC/DC power distribution network multi-period operation optimization model.
The distributed conversion and solving module is used for:
converting a multi-period operation optimization model of the centralized AC/DC power distribution network into a distributed operation optimization model of each subnet by adopting a target cascading analysis method, wherein the distributed operation optimization model of each subnet comprises:
operation optimization model of PET subnetwork:
in the above-mentioned method, the step of,for the objective function of the power electronic transformer subnetwork, < >>The method is characterized in that the method is used for purchasing electricity cost of an alternating current/direct current distribution network>Active power injected into t-period main network, < >>、/>Coefficient vector and weight vector of Lagrange penalty function of alternating current power distribution network respectively>、/>Coefficient vectors and weight vectors of the Lagrange penalty function of the direct current power distribution network are respectively adopted,、/>target variables which are boundary coupling variables of the AC/DC power distribution network respectively, < >>、/>Response variables, which are boundary coupling variables of the ac and dc power distribution network, respectively,/->、/>Equality and inequality constraints of the sub-network of the power electronic transformer, respectively, < >>Is a local variable of the PET subnet;
operation optimization model of ac subnetwork:
In the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < >>For the active power of the distributed power source i in the alternating current distribution network of the t period +.>、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>And the continuous variable and the discrete variable are respectively in the local variables of the alternating current power distribution network.
Operation optimization model of direct current subnetwork:
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>For the number of distributed power sources in a dc distribution network,generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of the direct current sub-network, +.>、/>The continuous variable and the discrete variable are respectively in the middle of the local variable of the direct current sub-network;
relaxing binary variables in the distributed operation optimization model of each subnet into continuous variables between 0 and 1, and constructing a loose subnet operation optimization model as follows:
relaxed ac subnetwork operation optimization model:
relaxed dc subnetwork operation optimization model:
And (3) iteratively solving the operation optimization model of the PET subnetwork, the relaxed alternating-current subnetwork operation optimization model and the relaxed direct-current subnetwork operation optimization model by adopting a target cascade analysis method to obtain an optimal solution of the boundary coupling variable, wherein the specific flow is shown in the step 5 of the embodiment 1.
The subnet independent model construction and solving module is used for constructing the following subnet independent operation optimization models based on the optimal solution of the boundary coupling variable, and solving the subnet independent operation optimization models through the Cplex solver to obtain the operation optimization result of the AC/DC power distribution network:
independent operation optimization model of communication sub-network:
DC subnet independent operation optimization model:
。
example 4:
the difference from example 3 is that:
as shown in fig. 5, the subnet independent model constructing and solving module comprises a subnet independent model constructing unit, an energy storage parameter optimizing unit, an acceleration induction factor calculating unit, a subnet independent acceleration model constructing unit and an acceleration model solving unit.
The subnet independent model constructing unit is configured to construct an operation optimization model of each subnet independent as in embodiment 3 based on the optimal solution of the boundary coupling variable.
The energy storage parameter optimization unit is used for performing linear relaxation on the independent operation optimization model of each sub-network to obtain:
Relaxed ac subnetwork independent run optimization model:
relaxed dc subnetwork independent run optimization model:
and solving the relaxed operation optimization model by adopting a Cplex solver to obtain an optimization solution of the charge and discharge states of the energy storage device of the AC/DC power distribution network.
The acceleration induction factor calculation unit is used for calculating and obtaining the acceleration induction factor of the AC/DC power distribution network based on the following formula：
In the above-mentioned method, the step of,for scaling parameters, ++>And the method is an optimized solution of the charge and discharge states of the energy storage device.
The subnet independent acceleration model building unit is used for building acceleration induction factorsAnd adding the operation optimization model into the operation optimization model with independent subnetworks to form an operation optimization acceleration model with independent subnetworks as follows:
and (3) an independent operation optimization acceleration model of the alternating current sub-network:
in the above-mentioned method, the step of,acceleration induction factors for the alternating current distribution network;
DC subnetwork independent operation optimization acceleration model:
in the above-mentioned method, the step of,is an acceleration induction factor of a direct current power distribution network.
And the acceleration model solving unit is used for solving the independent operation optimization acceleration model of each subnet through the Cplex solver to obtain an operation optimization result of the AC/DC power distribution network.
To verify the effectiveness of the method of the invention, the following comparative tests were carried out:
(1) The following two scenarios are set:
scene 1: directly solving the distributed optimization model of the AC/DC power distribution network by adopting a target cascade analysis method;
scene 2: the method described in embodiment 1 is used for solving the distributed optimization model of the AC/DC power distribution network.
The convergence of solving the optimization problem for different time periods in each scene is compared, and the result is shown in table 1:
table 1 comparison of convergence results for two scenarios
As the optimization period increases, the problem size will increase, as will the number of integer variables, which may result in the problem being more difficult to solve. As can be seen from table 1, for the problem of small-scale optimization in 24 time periods, both scene 1 and scene 2 can be converged, and compared with scene 1 of the conventional method, the method provided by the invention has fewer inner and outer loop iteration times in scene 2, which indicates that the convergence is faster. For a large-scale optimization model with 96 time periods, the dimension of an integer variable is increased sharply, so that a scene 1 cannot be converged, and when the distributed problem is solved in the first stage, the problem is a convex planning problem through loosening of the integer variable, so that the convergence performance is improved greatly, and after 14 outer-layer loops and 78 inner-layer loops, the algorithm achieves convergence.
(2) The model solving time of example 1 (no acceleration inducing factor was introduced into the model) and example 2 (acceleration inducing factor was introduced into the model) were examined, and the results are shown in table 2:
table 2 model solving time contrast for examples 1, 2
It can be seen that the solving time of the AC/DC sub-problem is respectively 45.3s and 32.4s when the induced acceleration is not adopted, and the solving time after the induced acceleration is respectively reduced to 30.1s and 18.3s, and the solving efficiency is respectively improved by 33.6% and 43.5%, which shows that the method for the induced acceleration provided by the invention can greatly improve the solving efficiency of the mixed integer linear programming problem.

Claims (10)

1. A two-stage distributed optimization method for an AC/DC power distribution network is characterized by comprising the following steps of:
the method comprises the following steps:
s1, constructing a multi-period operation optimization model of a centralized AC/DC power distribution network, wherein the multi-period operation optimization model of the centralized AC/DC power distribution network aims at the minimum operation cost, and constraint conditions comprise AC/DC power distribution network tide constraint, power electronic transformer operation constraint, energy storage device operation constraint and distributed power supply operation constraint;
s2, converting the multi-period operation optimization model of the centralized AC/DC power distribution network into a distributed operation optimization model of each subnet, and solving the distributed operation optimization model of each subnet to obtain an optimal solution of the boundary coupling variable;
S3, constructing an independent operation optimization model of each sub-network based on the optimal solution of the boundary coupling variable, and obtaining an operation optimization result of the AC/DC power distribution network by solving the independent operation optimization model of each sub-network.
2. The two-stage distributed optimization method for an ac/dc power distribution network according to claim 1, wherein:
in the step S1, the objective function of the multi-period operation optimization model of the centralized ac/dc power distribution network includes:
；
in the above-mentioned method, the step of,、/>the number of distributed power supplies in the AC power distribution network and the DC power distribution network is +.>、/>The power generation cost of the distributed power supply i in the AC and DC power distribution networks is +.>、/>Active power of a distributed power supply i in an alternating current power distribution network and a direct current power distribution network in t time periods respectively,/->The method is characterized in that the method is used for purchasing electricity cost of an alternating current/direct current distribution network>Active power injected into the main network at the t period;
in the step S3, the independent operation optimization model of each subnet includes:
independent operation optimization model of communication sub-network:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Continuous, discrete variable in local variable of AC distribution network>Is a real number;
DC subnet independent operation optimization model:
；
In the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable in the local variable of the direct current power distribution network are respectively.
3. A two-stage distributed optimization method for an ac/dc power distribution network according to claim 1 or 2, characterized in that:
the step S3 further includes:
s31, performing linear relaxation on the independent operation optimization model of each sub-network, and solving to obtain an optimization solution of the charge and discharge states of the energy storage device in the AC/DC power distribution network;
s32, optimizing solution calculation based on the charge and discharge states of the energy storage device to obtain acceleration induction factors of the AC/DC power distribution network, and adding the acceleration induction factors into the independent operation optimization models of all the sub-networks to form independent operation optimization acceleration models of all the sub-networks;
s33, solving an independent operation optimization acceleration model of each sub-network to obtain an operation optimization result of the AC/DC power distribution network.
4. A two-stage distributed optimization method for an ac/dc power distribution network according to claim 3, wherein:
in S32, the acceleration induction factor of the AC/DC power distribution networkCalculated based on the following formula:
；
in the above-mentioned method, the step of, For scaling parameters, ++>An optimized solution for the charge and discharge states of the energy storage device;
each subnet independent operation optimization acceleration model comprises:
and (3) an independent operation optimization acceleration model of the alternating current sub-network:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < >>For the active power of the distributed power source i in the alternating current distribution network of the t period +.>Acceleration induction factor for AC distribution network>、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Continuous, discrete variable in local variable of AC distribution network>Is a real number;
DC subnetwork independent operation optimization acceleration model:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>For the number of distributed power sources in the direct current distribution network, < >>Generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/>Acceleration induction factor for a DC power distribution network, +.>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable in the local variable of the direct current power distribution network are respectively.
5. A two-stage distributed optimization method for an ac/dc power distribution network according to claim 1 or 2, characterized in that:
in the step S2, the distributed operation optimization model of each subnet includes:
operation optimization model of power electronic transformer subnetwork:
；
in the above-mentioned method, the step of,for the objective function of the power electronic transformer subnetwork, < >>For the purchase cost of the ac/dc distribution network,active power injected into t-period main network, < >>、/>Coefficient vector and weight vector of Lagrange penalty function of alternating current power distribution network respectively>、/>Coefficient vector and weight vector of Lagrange penalty function of direct current distribution network respectively, +.>、Target variables which are boundary coupling variables of the AC/DC power distribution network respectively, < >>、/>Response variables, which are boundary coupling variables of the ac and dc power distribution network, respectively,/->、/>Equality and inequality constraints of the sub-network of the power electronic transformer, respectively, < >>Is a local variable of a power electronic transformer subnet;
operation optimization model of ac subnetwork:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < >>For the active power of the distributed power source i in the alternating current distribution network of the t period +. >、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Respectively continuous and discrete variables in the local variables of the alternating current power distribution network;
operation optimization model of direct current subnetwork:
；
in the above-mentioned method, the step of,for the purpose of direct current distribution networkMark function->For the number of distributed power sources in the direct current distribution network, < >>Generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable in the local variable of the direct current power distribution network are respectively.
6. A two-stage distributed optimization system of an AC/DC power distribution network is characterized in that:
the system comprises a centralized operation optimization model construction module, a distributed conversion and solving module and a subnet independent model construction and solving module;
the centralized operation optimization model construction module is used for constructing a multi-period operation optimization model of the centralized AC/DC power distribution network with minimum operation cost as a target, and constraint conditions of the multi-period operation optimization model of the centralized AC/DC power distribution network comprise AC/DC power distribution network tide constraint, power electronic transformer operation constraint, energy storage device operation constraint and distributed power supply operation constraint;
The distributed conversion and solving module is used for converting the multi-period operation optimization model of the centralized AC/DC power distribution network into a distributed operation optimization model of each subnet and solving the distributed operation optimization model of each subnet to obtain an optimal solution of the boundary coupling variable;
the subnet independent model construction and solving module is used for constructing an operation optimization model of each subnet independent based on the optimal solution of the boundary coupling variable, and the operation optimization result of the AC/DC power distribution network is obtained by solving the operation optimization model of each subnet independent.
7. The ac/dc distribution network two-stage distributed optimization system according to claim 6, wherein:
the centralized operation optimization model construction module is used for constructing the following objective functions:
；
in the above-mentioned method, the step of,、/>the number of distributed power supplies in the AC power distribution network and the DC power distribution network is +.>、/>The power generation cost of the distributed power supply i in the AC and DC power distribution networks is +.>、/>Active power of a distributed power supply i in an alternating current power distribution network and a direct current power distribution network in t time periods respectively,/->The method is characterized in that the method is used for purchasing electricity cost of an alternating current/direct current distribution network>Active power injected into the main network at the t period;
the subnet independent model construction and solving module is used for constructing the following operation optimization model of each subnet independently:
Independent operation optimization model of communication sub-network:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Continuous, discrete variable in local variable of AC distribution network>Is a real number;
DC subnet independent operation optimization model:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable in the local variable of the direct current power distribution network are respectively.
8. An ac/dc distribution network two-stage distributed optimization system according to claim 6 or 7, wherein:
the subnet independent model constructing and solving module comprises a subnet independent model constructing unit, an energy storage parameter optimizing unit, an acceleration induction factor calculating unit, a subnet independent acceleration model constructing unit and an acceleration model solving unit;
the subnet independent model building unit is used for building an operation optimization model of each subnet independent based on the optimal solution of the boundary coupling variable;
the energy storage parameter optimization unit is used for performing linear relaxation on the independent operation optimization model of each sub-network and solving to obtain an optimization solution of the charge and discharge states of the energy storage device in the AC/DC power distribution network;
The acceleration induction factor calculation unit is used for obtaining the acceleration induction factor of the AC/DC power distribution network based on the optimization solution calculation of the charge and discharge states of the energy storage device;
the subnet independent acceleration model construction unit is used for adding an acceleration induction factor into each subnet independent operation optimization model to form each subnet independent operation optimization acceleration model;
and the acceleration model solving unit is used for solving the independent operation optimization acceleration model of each subnet to obtain the operation optimization result of the AC/DC power distribution network.
9. The ac/dc distribution network two-stage distributed optimization system according to claim 8, wherein:
the acceleration induction factor calculation unit calculates the acceleration induction factor of the AC/DC power distribution network based on the following formula：
；
In the above-mentioned method, the step of,for scaling parameters, ++>An optimized solution for the charge and discharge states of the energy storage device;
the subnet independent acceleration model construction unit is used for constructing the operation optimization acceleration model of each subnet independent as follows:
and (3) an independent operation optimization acceleration model of the alternating current sub-network:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < > >For the active power of the distributed power source i in the alternating current distribution network of the t period +.>Acceleration induction factor for AC distribution network>、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Continuous, discrete variable in local variable of AC distribution network>Is a real number;
DC subnetwork independent operation optimization acceleration model:
；
in the above-mentioned method, the step of,as an objective function of a direct current power distribution network, +.>For the number of distributed power sources in the direct current distribution network, < >>Generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/>Acceleration induction factor for a DC power distribution network, +.>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable in the local variable of the direct current power distribution network are respectively.
10. An ac/dc distribution network two-stage distributed optimization system according to claim 6 or 7, wherein:
the distributed conversion and solving module is used for converting the multi-period operation optimization model of the centralized AC/DC power distribution network into a distributed operation optimization model of each sub-network as follows:
operation optimization model of power electronic transformer subnetwork:
；
In the above-mentioned method, the step of,for the objective function of the power electronic transformer subnetwork, < >>For the purchase cost of the ac/dc distribution network,active power injected into t-period main network, < >>、/>Coefficient vector and weight vector of Lagrange penalty function of alternating current power distribution network respectively>、/>Coefficient vector and weight vector of Lagrange penalty function of direct current distribution network respectively, +.>、Target variables which are boundary coupling variables of the AC/DC power distribution network respectively, < >>、/>Response variables, which are boundary coupling variables of the ac and dc power distribution network, respectively,/->、/>Equality and inequality constraints of the sub-network of the power electronic transformer, respectively, < >>Is a local variable of a power electronic transformer subnet;
operation optimization model of ac subnetwork:
；
in the above-mentioned method, the step of,for the objective function of an ac distribution network->For the number of distributed power sources in an ac distribution network,generating cost for distributed power source i in alternating current distribution network, < >>For period tActive power of distributed power source i in AC distribution network,/->、/>Equality, inequality constraints of ac distribution network, respectively, +.>For local variables of an ac power distribution network>、/>Respectively continuous and discrete variables in the local variables of the alternating current power distribution network;
operation optimization model of direct current subnetwork:
；
in the above-mentioned method, the step of, As an objective function of a direct current power distribution network, +.>For the number of distributed power sources in the direct current distribution network, < >>Generating cost for distributed power supply i in direct current distribution network, < >>For the active power of the distributed power source i in the t-period direct current power distribution network,/>、/>Equality, inequality constraints of the direct current distribution network, respectively,/->Is a local variable of a DC power distribution network, +.>、/>The continuous variable and the discrete variable in the local variable of the direct current power distribution network are respectively.
CN202410006126.3A 2024-01-03 2024-01-03 Two-stage distributed optimization method and system for AC/DC power distribution network Active CN117526426B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410006126.3A CN117526426B (en) 2024-01-03 2024-01-03 Two-stage distributed optimization method and system for AC/DC power distribution network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410006126.3A CN117526426B (en) 2024-01-03 2024-01-03 Two-stage distributed optimization method and system for AC/DC power distribution network

Publications (2)

Publication Number Publication Date
CN117526426A true CN117526426A (en) 2024-02-06
CN117526426B CN117526426B (en) 2024-03-22

Family

ID=89753406

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410006126.3A Active CN117526426B (en) 2024-01-03 2024-01-03 Two-stage distributed optimization method and system for AC/DC power distribution network

Country Status (1)

Country Link
CN (1) CN117526426B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120007532A1 (en) * 2010-07-08 2012-01-12 Tesla Motors, Inc. Rotor Temperature Estimation and Motor Control Torque Limiting for Vector-Controlled AC Induction Motors
CN105958485A (en) * 2016-06-14 2016-09-21 中国电力科学研究院 Power flow calculation method for flexible interconnecting alternating current-direct current hybrid power distribution network
CN109950907A (en) * 2019-02-22 2019-06-28 中国电力科学研究院有限公司 The dispatching method and system of alternating current-direct current mixing power distribution network containing electric power electric transformer
DE102018129810A1 (en) * 2018-11-26 2020-05-28 Technische Universität Darmstadt Method and device for controlling a number of energy-feeding and / or energy-consuming units
CN112347615A (en) * 2020-10-20 2021-02-09 天津大学 Power distribution network hybrid optimization scheduling method considering light storage and fast charging integrated station
CN116260201A (en) * 2022-10-13 2023-06-13 中国电力科学研究院有限公司 Fault recovery method, device, equipment and storage medium for AC/DC hybrid power distribution network
US20230231411A1 (en) * 2022-09-06 2023-07-20 Hunan University Method for flexible coordinated operation of urban distribution network and watershed network
WO2023201916A1 (en) * 2022-04-18 2023-10-26 国网智能电网研究院有限公司 Distributed flexible resource aggregation control apparatus and control method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120007532A1 (en) * 2010-07-08 2012-01-12 Tesla Motors, Inc. Rotor Temperature Estimation and Motor Control Torque Limiting for Vector-Controlled AC Induction Motors
CN105958485A (en) * 2016-06-14 2016-09-21 中国电力科学研究院 Power flow calculation method for flexible interconnecting alternating current-direct current hybrid power distribution network
DE102018129810A1 (en) * 2018-11-26 2020-05-28 Technische Universität Darmstadt Method and device for controlling a number of energy-feeding and / or energy-consuming units
CN109950907A (en) * 2019-02-22 2019-06-28 中国电力科学研究院有限公司 The dispatching method and system of alternating current-direct current mixing power distribution network containing electric power electric transformer
CN112347615A (en) * 2020-10-20 2021-02-09 天津大学 Power distribution network hybrid optimization scheduling method considering light storage and fast charging integrated station
WO2023201916A1 (en) * 2022-04-18 2023-10-26 国网智能电网研究院有限公司 Distributed flexible resource aggregation control apparatus and control method
US20230231411A1 (en) * 2022-09-06 2023-07-20 Hunan University Method for flexible coordinated operation of urban distribution network and watershed network
CN116260201A (en) * 2022-10-13 2023-06-13 中国电力科学研究院有限公司 Fault recovery method, device, equipment and storage medium for AC/DC hybrid power distribution network

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
AIHONG TANG: "Multi-objective_coordination_control_of_distributed_power_flow_controller", 《CSEE JOURNAL OF POWER AND ENERGY SYSTEMS》, vol. 5, no. 3, 30 September 2019 (2019-09-30), pages 348 - 354 *
JIE LIU: "Multiobjective_Optimal_Dispatch_of_Mobile_Energy_Storage_Vehicles_in_Active_Distribution_Networks", 《IEEE SYSTEMS JOURNAL》, vol. 17, no. 1, 31 March 2023 (2023-03-31), pages 804 - 815, XP011934942, DOI: 10.1109/JSYST.2022.3220825 *
王博: "考虑风电不确定性的风蓄火联合优化经济调度研究", 《电力工程技术》, vol. 41, no. 1, 31 January 2022 (2022-01-31), pages 93 - 100 *
阳育德: "大规模电网分层分区无功优化", 《电网技术》, vol. 39, no. 6, 30 June 2015 (2015-06-30), pages 1617 - 1622 *

Also Published As

Publication number Publication date
CN117526426B (en) 2024-03-22

Similar Documents

Publication Publication Date Title
Morstyn et al. Model predictive control for distributed microgrid battery energy storage systems
Abhishek et al. Review of hierarchical control strategies for DC microgrid
Hu et al. Distributed adaptive droop control for optimal power dispatch in DC microgrid
CN109950907B (en) Dispatching method and system for alternating current-direct current hybrid power distribution network containing power electronic transformer
Montoya et al. Numerical methods for power flow analysis in DC networks: State of the art, methods and challenges
Morstyn et al. Scalable energy management for low voltage microgrids using multi-agent storage system aggregation
CN108304972B (en) Active power distribution network frame planning method based on supply and demand interaction and DG (distributed generation) operation characteristics
Cucuzzella et al. Distributed passivity-based control of DC microgrids
CN112467748A (en) Double-time-scale distributed voltage control method and system for three-phase unbalanced active power distribution network
Dan et al. A scenario-based branch-and-bound approach for MES scheduling in urban buildings
Yang et al. Deep learning-based distributed optimal control for wide area energy Internet
Yang et al. Multiagent-based coordination consensus algorithm for state-of-charge balance of energy storage unit
Zhang et al. Hierarchical flexible operation approach on a VSC-MTDC interconnected hybrid grid with a high share of renewable power
Liu et al. A fully distributed economic dispatch method in DC microgrid based on consensus algorithm
Raza et al. Robust nonlinear control of regenerative fuel cell, supercapacitor, battery and wind based direct current microgrid
CN114282779A (en) Intelligent power grid dispatching method based on economic dispatching target and power parameter
CN117526426B (en) Two-stage distributed optimization method and system for AC/DC power distribution network
CN115496427B (en) Flexible interconnection investment planning decision method for multi-microgrid system
Wang et al. Distributed finite-time economic dispatch for islanded microgrids
Lin et al. Distributed optimized consensus-based frequency synchronization in isolated ac microgrids
He et al. A novel control method based on consensus algorithm for microgrids
Malhotra et al. CLPSO based droop optimization technique for DC Microgrid
Zhao et al. Distributed energy management for hybrid AC/DC microgrid parks
Pan et al. Optimal Design and Configuration Strategy for the Physical Layer of Energy Router Based on the Complex Network Theory
Zafeiratou et al. Handling power losses in a DC microgrid through constrained optimization

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant