CN114744612A - Two-stage day-ahead economic dispatching method for off-grid micro-grid - Google Patents

Two-stage day-ahead economic dispatching method for off-grid micro-grid Download PDF

Info

Publication number
CN114744612A
CN114744612A CN202210329613.4A CN202210329613A CN114744612A CN 114744612 A CN114744612 A CN 114744612A CN 202210329613 A CN202210329613 A CN 202210329613A CN 114744612 A CN114744612 A CN 114744612A
Authority
CN
China
Prior art keywords
load
cost
microgrid
time
electric automobile
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.)
Pending
Application number
CN202210329613.4A
Other languages
Chinese (zh)
Inventor
黄衍松
夏晓荣
胡鹏飞
王飞
王晴
候慧
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.)
Wuhan University of Technology WUT
Jingmen Power Supply Co of State Grid Hubei Electric Power Co Ltd
Original Assignee
Wuhan University of Technology WUT
Jingmen Power Supply Co 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 Wuhan University of Technology WUT, Jingmen Power Supply Co of State Grid Hubei Electric Power Co Ltd filed Critical Wuhan University of Technology WUT
Priority to CN202210329613.4A priority Critical patent/CN114744612A/en
Publication of CN114744612A publication Critical patent/CN114744612A/en
Pending legal-status Critical Current

Links

Images

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
    • H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
    • 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
    • H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
    • 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
    • H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
    • H02J3/14—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load by switching loads on to, or off from, network, e.g. progressively balanced loading
    • H02J3/144—Demand-response operation of the power transmission or distribution network
    • 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
    • H02J3/24—Arrangements for preventing or reducing oscillations of power in 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
    • H02J3/28—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
    • H02J3/322—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means the battery being on-board an electric or hybrid vehicle, e.g. vehicle to grid arrangements [V2G], power aggregation, use of the battery for network load balancing, coordinated or cooperative battery charging
    • 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
    • H02J3/388—Islanding, i.e. disconnection of local power supply from the network
    • 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
    • H02J2300/10—The dispersed energy generation being of fossil origin, e.g. diesel generators
    • 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
    • H02J2300/22—The renewable source being solar energy
    • H02J2300/24—The renewable source being solar energy of photovoltaic 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
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/28—The renewable source being wind energy
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention provides a two-stage day-ahead economic dispatching method for an off-grid microgrid, which utilizes chaotic phase space reconstruction, multi-target particle swarm, data driving, linear programming and other methods, reduces negative influences such as wind abandon and load loss caused by source load uncertainty through flexible resource regulation and control, and can give consideration to system efficiency and reliability on the basis of reducing dispatching cost.

Description

Two-stage day-ahead economic dispatching method for off-grid microgrid
Technical Field
The invention belongs to the technical field of economic dispatching of power systems, and particularly relates to a two-stage day-ahead economic dispatching method for an off-grid micro-grid.
Background
The off-grid micro-grid plays a significant role in solving the problems of on-site access and consumption of renewable energy sources and electricity utilization in remote areas and island areas far away from inland. The generated energy of renewable energy is influenced by environmental conditions, great uncertainty is generated on the operation of the off-grid micro-grid, and in addition, the fluctuation of the load also has uncertainty. Therefore, how to relieve the negative influence caused by load uncertainty has important practical significance on the aspect of stable operation of the off-grid micro-grid.
Disclosure of Invention
Aiming at the defects of the existing research, the invention provides a two-stage day-ahead economic dispatching method for an off-grid micro-grid.
The technical scheme of the invention is a two-stage day-ahead economic dispatching method for an off-grid micro-grid, which specifically comprises the following steps:
step 1: processing historical data of the wind power, the photovoltaic power and the load power by adopting phase space reconstruction, and predicting by adopting an extreme learning machine to obtain predicted values of the wind power, the photovoltaic power and the load power;
step 2: according to the predicted values of the wind power, the photovoltaic power and the load power obtained in the step 1, establishing an off-grid micro-grid multi-target economic dispatching model considering demand response, and solving by adopting a multi-target particle swarm algorithm and a fuzzy membership function to obtain a first-stage dispatching scheme;
and step 3: the first-stage scheduling scheme is applied to a scheduling day and one week before the scheduling day, data of one week before the scheduling day is used as a training set, the scheduling day is used as a test set, renewable energy abandonment and load loss generated by the scheduling day under the first-stage scheduling scheme are predicted by adopting a machine learning algorithm, a high-energy-carrying load model and a standby power supply model are built, renewable energy abandonment is absorbed by adopting a storage battery, and the load loss is absorbed by adopting a frequency modulation power supply, so that the second-stage scheduling scheme is obtained.
Preferably, the phase space reconstruction described in step 1 is specifically as follows:
step 1.1: let chaotic time series be x1,x2,L,xN-1,xNTo xt(t ═ 1,2, L, N- (m-1) τ), transformed as follows:
xt＝(xt,x(t+τ),x(t+2τ),L,x[t+(m-1)τ])T
where τ is the delay time, m is the embedding dimension,
according to a phase space reconstruction method, a chaos time sequence x is divided into1,x2,L,xN-1,xNConversion into a new data space of delay τ and dimension m, i.e.
Figure BDA0003572706940000021
Wherein each column represents a vector or phase point,
the extreme learning machine in the step 1 is characterized in that:
step 1.2: the extreme learning machine is used for solving the algorithm of the single hidden layer neural network, when an activation function is infinite or infinitesimal, the parameters of the single hidden layer feedforward neural network do not need to be adjusted completely, the weight and the bias between an input layer and a hidden layer can be randomly selected before training, and the connection weight beta between the hidden layer and an output layer is
β＝H+D′
Wherein H+D' is the transpose of the ideal output of the network, which is the generalized inverse of the hidden layer output matrix.
Preferably, the demand response described in step 2 is specifically as follows:
step 2.1: considering that the demand response includes electric vehicles, transferable loads and interruptible loads,
(1) an electric automobile is supposed to be limited by the driving habits of users, generally the electric automobile leaves a micro-grid system in the morning and returns to the micro-grid system after the journey is finished in the evening,
the initial load early peak initial time of the microgrid is assumed to be Tstart，mThe late peak starting time is Tstart，nThe return journey time of the ith electric automobile is t0(i) The charging start time is Tstart，char(i) At discharge start time Tstart，dischar(i) The charge and discharge starting time of the electric automobile is determined by comparing the return time of the electric automobile user with the starting time of the load peak of the microgrid at morning and evening, namely:
when t is0(i)＜Tstart，m，Tstart，char(i)＝t0(i)；
When T isstart，m≤t0(i)≤Tstart，nThen T isstart，dischar(i)＝Tstart，n；
When t is0(i)＞Tstart，nThen T isstart，dischar(i)＝t0(i)，
The maximum discharging electric quantity of the electric automobile ensures that the residual electric quantity meets the daily running requirement of a user and cannot exceed the maximum discharging depth of the electric automobile set by the electric automobile, namely, the maximum discharging quantity of the electric automobile measures the minimum value between the maximum discharging quantity and the minimum value:
Figure BDA0003572706940000031
determining charging time length T of ith electric automobile by using charging and discharging initial time and maximum discharging amount of electric automobilechar(i) Duration of discharge Tdischar(i) And orderly charging and discharging load P of electric automobileEV(t)，
Tdischar(i)＝Cdischar(i)/Pd
Tchar(i)＝(Cdischar(i)+s(i)*w)/Pc
Figure BDA0003572706940000032
When T ∈ [ T ]start,char(i),Tstart,char(i)+Tchar(i)-1]A (i, t) ═ 1; otherwise, a (i, t) ═ 0;
when T ∈ [ T ]start,dischar(i),Tstart,dischar(i)+Tdischar(i)-1]B (i, t) ═ 1; otherwise, b (i, t) is 0,
wherein N is the number of electric vehicles, PcAnd PdRespectively is charging and discharging power and Pc＞0,PdW and fr are respectively the power consumption per kilometer and the maximum depth of discharge, s (i) is the driving mileage of the ith electric automobile,
Figure BDA0003572706940000033
and CevThe charging state upper limit and the charging amount of the electric automobile are respectively, a (i, t) and b (i, t) are respectively the charging state and the discharging state of the ith electric automobile in a time period t, and when the electric automobile is charged, the value of a (i, t) is 1; when the electric automobile discharges, the b (i, t) value is 1;
(2) transferable load
Transferable load model: the load transfer-in and transfer-out model is
Figure BDA0003572706940000034
Figure BDA0003572706940000035
In the formula, Pin(t) and Pout(t) load values transferred in and out for time period t, respectively; n is a radical ofTLFor the total number of transferable load types, NT′LThe number of transferable load types with the operation duration longer than one scheduling period; h is a total ofmaxMaximum value of power supply duration for the transferable load units; x is a radical of a fluorine atomk(t) the number of load transfer units of the kth class starting to operate in the period of t; y isk(t) the number of kth type load transferring-out units which start to operate in a period of t; pl.kFor the power of the kth transferable load in the l working period, wherein k is more than or equal to 0 and less than or equal to NTL，
Figure BDA0003572706940000041
In the formula, xTL(t) is the actual load transfer at time t, XTL(t) transferable load capacity at time t,
the subsidy cost of the micro-grid system to the transferable load is as follows:
Figure BDA0003572706940000042
in the formula: cTL,costSubsidizing costs for the microgrid for transferable loads, cTLThe unit price is uniformly subsidized for the micro-grid to the transferable load,
(3) interruptible load
Interruptible load model: the load time sequence expression after the interruptible load control is adopted as
Figure BDA0003572706940000043
In the formula, P' (t) is the load demand after the interruptible load is adopted in the period t; p (t) is the original predicted load demand in the time period; y is a user number capable of providing interruptible load, and the total number of the users is Y; u (y, t) is a variable from 0 to 1, representing the state of whether the yth user is selected during the period t; s (y, t) is the interruption capacity of the tth user t period,
the subsidy cost of the micro-grid system to the interruptible load is as follows:
Figure BDA0003572706940000044
Figure BDA0003572706940000045
in the formula: cIL,costSubsidization cost for the microgrid to interruptible loads, cILUnit price, P, for a unified subsidy of a microgrid to transferable loadsIL(t) using the load demand after the interruptible load for a period of t,
the off-grid microgrid multi-target economic dispatching model in the step 2 is specifically as follows:
step 2.2: the established first-stage microgrid multi-target economic dispatching model mainly considers demand response, and model optimization indexes comprise high-efficiency indexes, namely, the highest local utilization rate (expressed by the minimum mean value of microgrid net loads) f of renewable energy sources1(ii) a Economic indicator, i.e. minimum integrated operating cost f of microgrid system2(ii) a And load reliability index, i.e. minimum rate of loss of load of the microgrid f3，
Figure BDA0003572706940000051
f2＝min[CES,cost+CEV,cost+Cde,cost+CTL,cost+CIL,cost]
Figure BDA0003572706940000052
CES,cost＝CES,om+CES,loss
Cde,cost＝Cde,fuel+Cde,om+Cde,EN+Cde,start
Figure BDA0003572706940000053
(s.t.PEV(t)≤0)
P′load-pre(t)＝-Ppv-pre(t)-Pwind-pre(t)+PEV(t)+
PTL(t)+PIL(t)+Pload-pre(t)
Of formula (II) to (III)'load-pre(t) microgrid net load at time t after one-stage scheduling by using predicted wind-solar load, CES，costFor the comprehensive operating costs of energy storage, CEV，costThe discharge subsidy cost of the micro-grid system to the electric vehicle, Cde，costComprehensive operating costs for the operation of diesel units, CTL，costSubsidizing costs for the microgrid to transferable loads, Pload-pre(t) predicting the uncontrollable load of the microgrid at the moment t, cEVThe unit price is uniformly subsidized for the discharge of the micro-grid system to the electric automobile,
the constraint conditions that each unit in the model needs to satisfy are as follows:
(1) state of charge and output constraints of an energy storage unit
SOCmax≤SOC(t)≤SOCmin
Figure BDA0003572706940000054
In the formula, SOCmax、SOCminAnd
Figure BDA0003572706940000055
respectively representing the state of charge and the maximum and minimum output of the energy storage unit;
(2) output upper and lower limits and climbing rate constraint of diesel engine set
Figure BDA0003572706940000061
Figure BDA0003572706940000062
(3) Power balance equation
Figure BDA0003572706940000063
In the formula (I), the compound is shown in the specification,
Figure BDA0003572706940000064
the maximum and minimum output of the diesel engine set are respectively;
Figure BDA0003572706940000065
respectively are the upper limit and the lower limit of the climbing speed of the diesel engine set,
the multi-target particle swarm algorithm and the fuzzy membership function in the step 2 are as follows:
step 2.3: the multi-target particle swarm optimization algorithm comprises the following steps:
(1) initializing a particle swarm, wherein the size of the swarm is N, the dimension of the particle is D, and randomly initializing the position x of each particleidAnd velocity vid(wherein i-1, 2L, N, D-1, 2L, D);
(2) calculating an adaptive value of the particle;
(3) computing particle individual and global historical best PiAnd Pg；
(4) Adopting the full size/small fitness function values to compare and select a new best fitness function value;
(5) updating the speed and the position of the particle according to the following formula, and if a certain dimension of the particle exceeds a boundary, reinitializing the dimension data;
Figure BDA0003572706940000066
Figure BDA0003572706940000067
where w is the inertial weight, k is the current iteration number, vidIs the velocity of the particles, c1And c2A non-negative constant, called the acceleration factor; r is1And r2Is distributed over [0,1 ]]A random number of intervals;
(6) screening non-inferior solutions in the current particle swarm, adding the non-inferior solutions into the elite set, and removing the inferior solutions in the elite set;
(7) if the end condition is met, ending the circulation if the end condition is met, otherwise returning to the step (2),
the finally obtained elite set is a non-inferior solution set obtained by a multi-target particle swarm algorithm,
by utilizing the multi-objective particle swarm algorithm, a series of non-inferior solutions can be obtained, a set formed by the non-inferior solutions is called a non-inferior solution set, also called Pareto Front,
after a non-inferior solution set is obtained, a fuzzy membership function is adopted to select a proper final solution, and one non-inferior solution x in the Pareto Front is consideredkSatisfaction with the ith sub-objective function
Figure BDA0003572706940000073
Is composed of
Figure BDA0003572706940000071
fi max、fi minRespectively, the ith sub-target maximum-minimum function value, and therefore,
Figure BDA0003572706940000074
the value range is (0,1) for xkIn other words, the overall satisfaction of all sub-goals is
Figure BDA0003572706940000072
M and N are the numbers of non-inferior solutions and sub-targets respectively, mukA larger value of (d) indicates a higher overall satisfaction for this non-inferior solution.
Preferably, the storage battery model in step 3 is specifically as follows:
step 3.1: the storage battery energy storage system is generally measured by a State of Charge (SOC), where the SOC is a ratio of an energy storage remaining capacity divided by a rated capacity, and the remaining capacity of the storage battery at time t is determined by the remaining capacity at the previous time and a charging or discharging capacity of the storage battery during a time period [ t-1, t ]:
SOC(t)＝SOC(t-1)+PES(t)Δtηc/Ees
SOC(t)＝SOC(t-1)+PES(t)Δt/(Eesηd)
wherein SOC (t) and SOC (t-1) are the states of charge of the storage battery at t and t-1, respectively, PES(t) is the output power of the battery pack at time t, ηcAnd ηdCharging and discharging efficiencies for the storage battery, respectively, EesIs the battery rated capacity of the storage battery,
the expenses generated by the energy storage unit in the economic dispatching process comprise the operation and maintenance cost C of the energy storage unitES,omAnd loss cost C caused by charge-discharge conversionES，loss，
CES,om＝|PES(t)|*Kom,ES
CES,loss＝nB*(Ccost,change/nBN)
In the formula, Kom,ESAnd Ccost,changeRespectively representing the running cost coefficient and the replacement cost of the energy storage unit; n isBAnd nBNRespectively the number of charging and discharging conversion in one period of the energy storage unit and the rated charging and discharging number in the life cycle,
the frequency modulation power supply model in the step 3 is concretely as follows:
step 3.2: the goal of the FM power scheduling model is to minimize FM cost, i.e.
Cf＝CfES,cost+CfIL,cost
In the formula, CfES,costAnd CfIL,costThe frequency modulation costs of the frequency modulated battery and the interruptible load respectively,
the constraint that the frequency-modulated power supply needs to satisfy has a power balance equation, i.e.
Pload-cut＝PfES+PfIL
In the formula, PfESAnd PfILBattery output for second stage dispatchAnd interruptible load power.
The invention has the beneficial effects that:
(1) a demand response strategy is considered on the load side, and the effect of stabilizing system net load fluctuation is achieved;
(2) the two-stage island scheduling strategy has higher renewable energy abandonment and load loss prediction precision, so that the economy, the efficiency and the load loss rate of the system in an island operation state are improved.
Drawings
FIG. 1 is a schematic diagram of a two-stage day-ahead economic dispatching method of an off-grid microgrid of the present invention;
FIG. 2 is a diagram of an off-grid microgrid system of the present invention;
FIG. 3 is a diagram of wind power, photovoltaic and load power prediction results of the present invention;
FIG. 4 is a graph of the output of each load during one stage of scheduling according to the present invention;
FIG. 5 is a graph of the output of each operating unit during a phase of scheduling according to the present invention;
FIG. 6 is a diagram showing the result of the wind curtailment in the two-stage battery dispatching of the present invention;
FIG. 7 is a diagram of the two-stage FM power scheduling result of the present invention.
Detailed Description
In order to facilitate understanding and implementation of the present invention for persons of ordinary skill in the art, the present invention is further described in detail with reference to the drawings and the implementation examples, and it is to be understood that the implementation examples described herein are only for illustration and explanation of the present invention and are not to be construed as limiting the present invention.
The following describes an embodiment of the present invention with reference to fig. 1 to 7, and fig. 1 is a diagram of an off-grid microgrid system according to the present invention, and includes the following specific steps:
step 1: historical data of the wind power, the photovoltaic power and the load power are processed by adopting phase space reconstruction, a limit learning machine is adopted to predict to obtain predicted values of the wind power, the photovoltaic power and the load power,
the phase space reconstruction described in step 1 is specifically as follows:
step 1.1: setting chaos timeSequence x1,x2,L,xN-1,xNTo xt(t ═ 1,2, L, N- (m-1) τ), transformed as follows:
xt＝(xt,x(t+τ),x(t+2τ),L,x[t+(m-1)τ])T
where τ is the delay time, m is the embedding dimension,
according to a phase space reconstruction method, a chaos time sequence x is divided into1,x2,L,xN-1,xNConversion into a new data space of delay τ and dimension m, i.e.
Figure BDA0003572706940000091
Wherein each column represents a vector or phase point,
the extreme learning machine in the step 1 is characterized in that:
step 1.2: the extreme learning machine is used for solving the algorithm of the single hidden layer neural network, when an activation function is infinite or differentiable, the parameters of the single hidden layer feedforward neural network do not need to be adjusted completely, the weight and the bias between an input layer and a hidden layer can be randomly selected before training, and the connection weight beta between the hidden layer and an output layer is
β＝H+D′
Wherein H+D' is the transpose of the ideal output of the network, which is the generalized inverse of the hidden layer output matrix.
FIG. 2 is a diagram of wind power, photovoltaic and load power prediction results.
Step 2: and (3) establishing an off-grid micro-grid multi-target economic dispatching model considering demand response according to the predicted values of the wind power, the photovoltaic power and the load power obtained in the step (1), and solving by adopting a multi-target particle swarm algorithm and a fuzzy membership function to obtain a first-stage dispatching scheme.
The demand response described in step 2 is specifically as follows:
step 2.1: consider that demand response includes electric vehicles, transferable loads, and interruptible loads.
(1) Provided is an electric automobile. Assuming that the electric automobile is limited by the driving habit of a user, the electric automobile generally drives away from the micro-grid system in the morning and returns to the micro-grid system after the journey is finished in the evening.
The initial load early peak starting time of the micro-grid is assumed to be Tstart，mThe late peak starting time is Tstart，nThe return journey time of the ith electric automobile is t0(i) The charging start time is Tstart，char(i) At discharge start time Tstart，dischar(i) In that respect The charge and discharge starting time of the electric automobile is determined by comparing the return time of the electric automobile user with the starting time of the load peak of the microgrid load at morning and evening, namely:
when t is0(i)＜Tstart，m，Tstart，char(i)＝t0(i)；
When T isstart，m≤t0(i)≤Tstart，nThen T isstart，dischar(i)＝Tstart，n；
When t is0(i)＞Tstart，nThen T isstart，dischar(i)＝t0(i)。
The maximum discharging electric quantity of the electric automobile ensures that the residual electric quantity meets the daily running requirement of a user and cannot exceed the maximum discharging depth of the electric automobile set by the electric automobile, namely, the maximum discharging quantity of the electric automobile measures the minimum value between the maximum discharging quantity and the minimum value:
Figure BDA0003572706940000101
determining charging time length T of ith electric automobile by using charging and discharging initial time and maximum discharging amount of electric automobilechar(i) Duration of discharge Tdischar(i) And orderly charging and discharging load P of electric automobileEV(t)。
Tdischar(i)＝Cdischar(i)/Pd
Tchar(i)＝(Cdischar(i)+s(i)*w)/Pc
Figure BDA0003572706940000102
When T ∈ [ T ]start,char(i),Tstart,char(i)+Tchar(i)-1]A (i, t) ═ 1; otherwise, a (i, t) ═ 0;
when T ∈ [ T ]start,dischar(i),Tstart,dischar(i)+Tdischar(i)-1]B (i, t) ═ 1; otherwise, b (i, t) ═ 0.
Wherein N is the number of electric vehicles, PcAnd PdRespectively is charging and discharging power and Pc＞0,PdW and fr are respectively the power consumption per kilometer and the maximum depth of discharge, s (i) is the driving mileage of the ith electric automobile,
Figure BDA0003572706940000105
and CevThe charging state upper limit and the charging amount of the electric automobile are respectively, a (i, t) and b (i, t) are respectively the charging state and the discharging state of the ith electric automobile in a time period t, and when the electric automobile is charged, the value of a (i, t) is 1; when the electric automobile is discharged, the b (i, t) value is 1.
(2) Transferable load
Transferable load model: the load is transferred into and out of the model as
Figure BDA0003572706940000103
Figure BDA0003572706940000104
In the formula, Pin(t) and Pout(t) load values transferred in and out for time period t, respectively; n is a radical ofTLFor the total number of transferable load types, NT′LThe number of transferable load types with the operation duration longer than one scheduling period; h ismaxMaximum value of power supply duration for the transferable load units; x is the number ofk(t) the number of load transfer units of the kth class starting to operate in the period of t; y isk(t) is the start of the t periodThe number of units is transferred out from the k-th load in operation; p isl.kFor the power of the kth transferable load in the l working period, wherein k is more than or equal to 0 and less than or equal to NTL。
Figure BDA0003572706940000111
In the formula, xTL(t) is the actual load transfer at time t, XTL(t) is the transferable load capacity at time t.
The subsidy cost of the micro-grid system to the transferable load is as follows:
Figure BDA0003572706940000112
in the formula: cTL,costSubsidizing costs for the microgrid for transferable loads, cTLAnd the unit price is uniformly subsidized for the micro-grid to the transferable loads.
(3) Interruptible load
Interruptible load model: the load time sequence expression after the interruptible load control is adopted as
Figure BDA0003572706940000113
In the formula, P' (t) is the load requirement after the interruptible load is adopted in the period t; p (t) is the original predicted load demand in the time period; y is a user number capable of providing interruptible load, and the total number of the users is Y; u (y, t) is a variable from 0 to 1, representing the state of whether the yth user is selected during the period t; s (y, t) is the interrupt capacity for the tth user t period.
The subsidy cost of the micro-grid system to the interruptible load is as follows:
Figure BDA0003572706940000114
Figure BDA0003572706940000115
in the formula: cIL,costSubsidizing costs for the microgrid for interruptible loads, cILUnit price, P, for a unified subsidy of a microgrid to transferable loadsIL(t) the load demand after the interruptible load is adopted for the period t.
The off-grid microgrid multi-target economic dispatching model in the step 2 is specifically as follows:
step 2.2: the established first-stage microgrid multi-target economic dispatching model mainly considers demand response, and model optimization indexes comprise high-efficiency indexes, namely, the highest local utilization rate (expressed by the minimum mean value of microgrid net loads) f of renewable energy sources1(ii) a Economic indicator, i.e. minimum integrated operating cost f of microgrid system2(ii) a And load reliability index, i.e. minimum rate of loss of load of the microgrid f3。
Figure BDA0003572706940000121
f2＝min[CES,cost+CEV,cost+Cde,cost+CTL,cost+CIL,cost]
Figure BDA0003572706940000122
CES,cost＝CES,om+CES,loss
Cde,cost＝Cde,fuel+Cde,om+Cde,EN+Cde,start
Figure BDA0003572706940000123
(s.t.PEV(t)≤0)
P′load-pre(t)＝-Ppv-pre(t)-Pwind-pre(t)+PEV(t)+
PTL(t)+PIL(t)+Pload-pre(t)
Of formula (II) to (III)'load-pre(t) microgrid net load at time t after one-stage scheduling by using predicted wind-solar load, CES，costFor the comprehensive operating costs of energy storage, CEV，costThe discharge subsidy cost of the micro-grid system to the electric vehicle, Cde，costComprehensive operating costs for the operation of diesel units, CTL，costSubsidizing costs for the microgrid to transferable loads, Pload-pre(t) the load of the microgrid which is predicted to be uncontrollable at the time t, cEVThe unit price is uniformly subsidized for the discharge of the micro-grid system to the electric automobile.
The constraint conditions that each unit in the model needs to satisfy are as follows:
(1) state of charge and output constraints of an energy storage unit
SOCmax≤SOC(t)≤SOCmin
Figure BDA0003572706940000125
In the formula, SOCmax、SOCminAnd
Figure BDA0003572706940000126
the state of charge and the maximum and minimum output of the energy storage unit are respectively.
(2) Output upper and lower limits and climbing rate constraint of diesel engine set
Figure BDA0003572706940000127
Figure BDA0003572706940000128
(3) Power balance equation
Figure BDA0003572706940000124
In the formula (I), the compound is shown in the specification,
Figure BDA0003572706940000134
the maximum and minimum output of the diesel engine set are respectively;
Figure BDA0003572706940000135
respectively the upper limit and the lower limit of the climbing speed of the diesel engine set.
The multi-target particle swarm algorithm and the fuzzy membership function in the step 2 are as follows:
step 2.3: the multi-objective particle swarm optimization algorithm comprises the following steps.
(1) Initializing a particle swarm, wherein the size of the particle swarm is N, the dimension of the particle is D, and the position x of each particle is randomly initializedidAnd velocity vid(wherein i ═ 1,2L, N, D ═ 1,2L, D).
(2) Calculating an adaptation value of a particle
(3) Computing particle individual and global history best PiAnd Pg。
(4) And comparing by adopting the full large/small fitness function values, and selecting a new best.
(5) The particle velocity and position are updated according to the following formula, and if a certain dimension of the particle exceeds the boundary, the dimension data is reinitialized.
Figure BDA0003572706940000131
Figure BDA0003572706940000132
Where w is the inertial weight, k is the current iteration number, vidIs the velocity of the particles, c1And c2A non-negative constant, called the acceleration factor; r is1And r2Is distributed in [0,1 ]]Random number of intervals.
(6) Screening non-inferior solutions in the current particle swarm, adding the non-inferior solutions into the elite set, and removing the inferior solutions in the elite set.
(7) And (4) whether a termination condition is met, if so, ending the circulation, and if not, returning to the step (2).
And finally obtaining the elite set which is the non-inferior solution set obtained by the multi-target particle swarm algorithm.
A series of non-inferior solutions can be obtained by using a multi-objective particle swarm algorithm, and a set formed by the non-inferior solutions is called a non-inferior solution set and is also called Pareto Front.
After a non-inferior solution set is obtained, a fuzzy membership function is adopted to select an appropriate final solution. Consider a non-inferior solution x of Pareto FrontkSatisfaction with the ith sub-objective function
Figure BDA0003572706940000136
Is composed of
Figure BDA0003572706940000133
fi max、fi minRespectively, the ith sub-target maximum-minimum function value, and therefore,
Figure BDA0003572706940000137
the value range is (0, 1).
For xkTo say, the overall satisfaction of all sub-targets is
Figure BDA0003572706940000141
M and N are the numbers of non-inferior solutions and sub-targets respectively, mukA larger value of (d) indicates a higher overall satisfaction for this non-inferior solution.
Fig. 3 and 4 are graphs of the output of each load and the output of each operation unit in a one-stage scheduling scheme, respectively.
And step 3: and applying the first-stage scheduling scheme to the scheduling day and one week before the scheduling day, taking data of one week before the scheduling day as a training set and the scheduling day as a test set, and predicting the abandonment and the load loss of the renewable energy generated by the scheduling day under the first-stage scheduling scheme by adopting a machine learning algorithm. Firstly, a classification algorithm XGboost is used for distinguishing a load losing moment and a wind abandoning moment, and then a regression algorithm extreme learning machine is used for carrying out regression prediction on the load losing quantity at the load losing moment and the wind abandoning quantity at the wind abandoning moment. Because the predicted value and the actual value of the uncontrollable load have larger correlation with the previous week uncontrollable load, the data of the week before the scheduling day is selected as training data. And establishing a high-energy-carrying load model and a standby power supply model, adopting a storage battery to absorb renewable energy for abandonment, and adopting a frequency modulation power supply to absorb lost load to obtain a second-stage scheduling scheme.
The storage battery model in the step 3 is specifically as follows:
step 3.1: the storage battery energy storage system generally measures its storage capacity according to a State of Charge (SOC), which is a ratio of the storage remaining capacity divided by a rated capacity. The residual capacity of the storage battery at the moment t is determined by the residual capacity at the last moment and the charging or discharging capacity of the storage battery in the period [ t-1, t ]:
SOC(t)＝SOC(t-1)+PES(t)Δtηc/Ees
SOC(t)＝SOC(t-1)+PES(t)Δt/(Eesηd)
wherein SOC (t) and SOC (t-1) are the state of charge of the storage battery at the time t and t-1 respectively,
PES(t) is the output power of the battery pack at time t, ηcAnd ηdCharging and discharging efficiencies for the storage battery, respectively, EesIs the battery rated capacity of the storage battery.
The cost generated by the energy storage unit in the economic dispatching process comprises the operation and maintenance cost C of the energy storage unitES,omAnd loss cost C caused by charge-discharge conversionES，loss。
CES,om＝|PES(t)|*Kom,ES
CES,loss＝nB*(Ccost,change/nBN)
In the formula, Kom,ESAnd Ccost,changeRespectively representing the running cost coefficient and the replacement cost of the energy storage unit; n isBAnd nBNThe number of times of charge-discharge conversion in one period of the energy storage unit and the rated charge-discharge number of times in the service life period are respectively.
The frequency modulation power supply model in the step 3 is specifically as follows:
step 3.2: the goal of the FM power scheduling model is to minimize FM cost, i.e.
Cf＝CfES,cost+CfIL,cost
In the formula, CfES,costAnd CfIL,costThe frequency modulation costs of the frequency modulated battery and the interruptible load, respectively.
The constraint that the frequency-modulated power supply needs to satisfy has a power balance equation, i.e.
Pload-cut＝PfES+PfIL
In the formula, PfESAnd PfILRespectively for the battery output and interruptible load power scheduled for the second stage.
Fig. 5 and fig. 6 are the result of the wind curtailment in the battery scheduling and the result of the frequency modulation power scheduling, respectively. According to analysis results, the prediction accuracy rates of the wind abandoning moment and the load losing moment are both 80%, and the good prediction accuracy provides accurate prepositive work for spare capacity scheduling, so that the system economy, the efficiency and the load loss rate are improved.
It should be understood that the above description of the preferred embodiments is given for clarity and not for any purpose of limitation, and that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (4)

1. A two-stage day-ahead economic dispatching method for an off-grid microgrid is characterized by comprising the following steps:
step 1: processing historical data of the wind power, the photovoltaic power and the load power by adopting phase space reconstruction, and predicting by adopting an extreme learning machine to obtain predicted values of the wind power, the photovoltaic power and the load power;
step 2: according to the predicted values of the wind power, the photovoltaic power and the load power obtained in the step 1, establishing an off-grid micro-grid multi-target economic dispatching model considering demand response, and solving by adopting a multi-target particle swarm algorithm and a fuzzy membership function to obtain a first-stage dispatching scheme;
and step 3: the first-stage scheduling scheme is applied to the scheduling day and one week before the scheduling day, data of one week before the scheduling day is used as a training set, the scheduling day is used as a test set, renewable energy abandonment and load loss generated by the scheduling day under the first-stage scheduling scheme are predicted by adopting a machine learning algorithm, a high-energy-carrying load model and a standby power supply model are established, the renewable energy abandonment is absorbed by adopting a storage battery, and the load loss is absorbed by adopting a frequency modulation power supply, so that the second-stage scheduling scheme is obtained.
2. The two-stage day-ahead economic dispatching method for the off-grid microgrid according to claim 1, characterized in that:
the phase space reconstruction described in step 1 is specifically as follows:
step 1.1: let chaotic time series be x1,x2,L,xN-1,xNTo xt(t ═ 1,2, L, N- (m-1) τ), transformed as follows:
xt＝(xt,x(t+τ),x(t+2τ),L,x[t+(m-1)τ])T
where τ is the delay time, m is the embedding dimension,
according to a phase space reconstruction method, a chaos time sequence x is divided into1,x2,L,xN-1,xNConversion into a new data space of delay τ and dimension m, i.e.
Figure FDA0003572706930000011
Wherein each column represents a vector or phase point,
the extreme learning machine in the step 1 is characterized in that:
step 1.2: the extreme learning machine is used for solving the algorithm of the single hidden layer neural network, when an activation function is infinite or infinitesimal, the parameters of the single hidden layer feedforward neural network do not need to be adjusted completely, the weight and the bias between an input layer and a hidden layer can be randomly selected before training, and the connection weight beta between the hidden layer and an output layer is
β＝H+D′
Wherein H+D' is the transpose of the ideal output of the network, which is the generalized inverse of the hidden layer output matrix.
3. The two-stage day-ahead economic dispatching method for the off-grid microgrid according to claim 1, characterized in that:
the demand response described in step 2 is specifically as follows:
step 2.1: considering that the demand response includes electric vehicles, transferable loads and interruptible loads,
(1) an electric automobile is supposed to be limited by the driving habits of users, generally the electric automobile leaves a micro-grid system in the morning and returns to the micro-grid system after the journey is finished in the evening,
the initial load early peak starting time of the micro-grid is assumed to be Tstart，mThe late peak starting time is Tstart，nThe return journey time of the ith electric automobile is t0(i) The charging start time is Tstart，char(i) At discharge start time Tstart，dischar(i) The charge and discharge starting time of the electric automobile is determined by comparing the return time of the electric automobile user with the starting time of the load peak of the microgrid at morning and evening, namely:
when t is0(i)＜Tstart，m，Tstart，char(i)＝t0(i)；
When T isstart，m≤t0(i)≤Tstart，nThen T isstart，dischar(i)＝Tstart，n；
When t is0(i)＞Tstart，nThen T isstart，dischar(i)＝t0(i)。
The maximum discharging electric quantity of the electric automobile ensures that the residual electric quantity meets the daily running requirement of a user and cannot exceed the maximum discharging depth of the electric automobile set by the electric automobile, namely, the maximum discharging quantity of the electric automobile measures the minimum value between the maximum discharging quantity and the minimum value:
Figure FDA0003572706930000021
determining charging time length T of ith electric automobile by using charging and discharging initial time and maximum discharging amount of electric automobilechar(i) Duration of discharge Tdischar(i) And orderly charging and discharging load P of electric automobileEV(t)，
Tdischar(i)＝Cdischar(i)/Pd
Tchar(i)＝(Cdischar(i)+s(i)*w)/Pc
Figure FDA0003572706930000031
When T ∈ [ T ]start,char(i),Tstart,char(i)+Tchar(i)-1]A (i, t) ═ 1; otherwise, a (i, t) ═ 0;
when T ∈ [ T ]start,dischar(i),Tstart,dischar(i)+Tdischar(i)-1]B (i, t) ═ 1; otherwise, b (i, t) is 0,
wherein N is the number of electric vehicles, PcAnd PdRespectively is charging and discharging power and Pc＞0,PdW and fr are respectively the power consumption per kilometer and the maximum depth of discharge, s (i) is the driving mileage of the ith electric automobile,
Figure FDA0003572706930000032
and CevThe upper and lower limits of the electric vehicle state of charge and the battery capacity, a (i, t) and b (i, t) are respectively the ith electric vehicleA charging state and a discharging state in a period t, when the electric automobile is charged, the value of a (i, t) is 1; when the electric automobile discharges, the b (i, t) value is 1;
(2) transferable load
Transferable load model: the load is transferred into and out of the model as
Figure FDA0003572706930000033
Figure FDA0003572706930000034
In the formula, Pin(t) and Pout(t) load values transferred in and out for time period t, respectively; n is a radical ofTLIs total transferable load type, N'TLThe number of transferable load types with the operation duration longer than one scheduling period; h ismaxMaximum value of power supply duration for transferable load units; x is a radical of a fluorine atomk(t) the number of load transfer units of the kth class starting to operate in the period of t; y isk(t) the number of kth type load transfer-out units which start to operate in the period of t; pl.kFor the power of the kth transferable load in the l working period, wherein k is more than or equal to 0 and less than or equal to NTL，
Figure FDA0003572706930000035
In the formula, xTL(t) is the actual load transfer at time t, XTL(t) transferable load capacity for time t,
the subsidy cost of the micro-grid system to the transferable load is as follows:
Figure FDA0003572706930000036
in the formula: cTL,costSubsidizing costs for the microgrid for transferable loads, cTLIs rotatable for micro-gridThe unified subsidy unit price of the load shifting;
(3) interruptible load
Interruptible load model: the load time sequence expression after the interruptible load control is adopted as
Figure FDA0003572706930000041
In the formula, P' (t) is the load demand after the interruptible load is adopted in the period t; p (t) is the original predicted load demand in the time period; y is a user number capable of providing interruptible load, and the total number of the users is Y; u (y, t) is a variable from 0 to 1, representing the state of whether the yth user is selected during the period t; s (y, t) is the interruption capacity of the tth user t period,
the subsidy cost of the micro-grid system to the interruptible load is as follows:
Figure FDA0003572706930000042
Figure FDA0003572706930000043
in the formula: cIL,costSubsidizing costs for the microgrid for interruptible loads, cILUnit price, P, for a unified subsidy of a microgrid to transferable loadsIL(t) using the load demand after the interruptible load for a period of t,
the off-grid microgrid multi-target economic dispatching model in the step 2 is specifically as follows:
step 2.2: the established first-stage microgrid multi-target economic dispatching model mainly considers demand response, model optimization indexes comprise high-efficiency indexes, namely the local utilization rate of renewable energy is highest, and f is expressed by the minimum mean value of microgrid net loads in the model1(ii) a Economic indicator, i.e. minimum integrated operating cost f of microgrid system2(ii) a And load reliability index, i.e. minimum rate of load loss f of microgrid3，
Figure FDA0003572706930000044
f2＝min[CES,cost+CEV,cost+Cde,cost+CTL,cost+CIL,cost]
Figure FDA0003572706930000045
CES,cost＝CES,om+CES,loss
Cde,cost＝Cde,fuel+Cde,om+Cde,EN+Cde,start
Figure FDA0003572706930000051
(s.t.PEV(t)≤0)
P′load-pre(t)＝-Ppv-pre(t)-Pwind-pre(t)+PEV(t)+PTL(t)+PIL(t)+Pload-pre(t)
Of formula (II) to (III)'load-pre(t) microgrid net load at time t after one-stage scheduling by using predicted wind-solar load, CES，costFor the comprehensive operating costs of energy storage, CEV，costThe discharge subsidy cost of the micro-grid system to the electric vehicle, Cde，costComprehensive operating costs for the operation of diesel units, CTL，costSubsidizing costs for the microgrid to transferable loads, Pload-pre(t) the load of the microgrid which is predicted to be uncontrollable at the time t, cEVThe unit price is uniformly subsidized for the discharge of the micro-grid system to the electric automobile,
the constraint conditions that each unit in the model needs to satisfy are as follows:
(1) state of charge and output constraints of an energy storage unit
SOCmax≤SOC(t)≤SOCmin
Figure FDA0003572706930000052
In the formula, SOCmax、SOCminAnd
Figure FDA0003572706930000053
respectively representing the state of charge and the maximum and minimum output of the energy storage unit;
(2) output upper and lower limits and climbing rate constraint of diesel engine set
Figure FDA0003572706930000054
Figure FDA0003572706930000055
(3) Power balance equation
Figure FDA0003572706930000056
In the formula (I), the compound is shown in the specification,
Figure FDA0003572706930000057
the maximum and minimum output of the diesel engine set are respectively;
Figure FDA0003572706930000058
respectively representing the upper limit and the lower limit of the climbing speed of the diesel engine unit;
the multi-target particle swarm algorithm and the fuzzy membership function in the step 2 are as follows:
step 2.3: the multi-target particle swarm optimization algorithm comprises the following steps:
(1) initializing a particle swarm, wherein the size of the swarm is N, the dimension of the particle is D, and randomly initializing the position x of each particleidAnd velocity vid(wherein i-1, 2L, N, D-1, 2L, D);
(2) calculating an adaptive value of the particle;
(3) computing particle individual and global historical best PiAnd Pg；
(4) Adopting the full size/small fitness function values to compare and select a new best fitness function value;
(5) updating the speed and position of the particle according to the following formula, if a certain dimension of the particle exceeds the boundary, reinitializing the dimension data,
Figure FDA0003572706930000061
Figure FDA0003572706930000062
where w is the inertial weight, k is the current iteration number, vidIs the velocity of the particles, c1And c2A non-negative constant, called acceleration factor; r is1And r2Is distributed over [0,1 ]]A random number of intervals;
(6) screening non-inferior solutions in the current particle swarm, adding the non-inferior solutions into the elite set, and removing the inferior solutions in the elite set;
(7) if the end condition is met, ending the circulation if the end condition is met, otherwise returning to the step (2),
the finally obtained elite set is a non-inferior solution set obtained by a multi-target particle swarm algorithm,
by utilizing the multi-target particle swarm algorithm, a series of non-inferior solutions can be obtained, a set formed by the non-inferior solutions is called a non-inferior solution set and is also called a Pareto Front, namely a Pareto Front,
after a non-inferior solution set is obtained, a fuzzy membership function is adopted to select a proper final solution, and one non-inferior solution x in the Pareto Front is consideredkSatisfaction with the ith sub-objective function
Figure FDA0003572706930000063
Is composed of
Figure FDA0003572706930000064
fi max、fi minRespectively, the ith sub-target maximum-minimum function value, and therefore,
Figure FDA0003572706930000065
the value range is (0,1),
for xkTo say, the overall satisfaction of all sub-targets is
Figure FDA0003572706930000066
M and N are the numbers of non-inferior solutions and sub-targets respectively, mukA larger value of (d) indicates a higher overall satisfaction for this non-inferior solution.
4. The two-stage day-ahead economic dispatching method for the off-grid microgrid according to claim 1, characterized in that:
the storage battery model in the step 3 is specifically as follows:
step 3.1: the storage battery energy storage system is generally measured by a State of Charge (SOC), where the SOC is a ratio of an energy storage remaining capacity divided by a rated capacity, and the remaining capacity of the storage battery at time t is determined by the remaining capacity at the previous time and a charging or discharging capacity of the storage battery during a time period [ t-1, t ]:
SOC(t)＝SOC(t-1)+PES(t)Δtηc/Ees
SOC(t)＝SOC(t-1)+PES(t)Δt/(Eesηd)
wherein SOC (t) and SOC (t-1) are the states of charge of the storage battery at t and t-1 respectively, PES(t) is the output power of the battery pack at time t, ηcAnd ηdAre respectively provided withCharging and discharging efficiency for batteries EesIs the battery rated capacity of the storage battery,
the cost generated by the energy storage unit in the economic dispatching process comprises the operation and maintenance cost C of the energy storage unitES,omAnd loss cost C caused by charge-discharge conversionES，loss，
CES,om＝|PES(t)|*Kom,ES
CES,loss＝nB*(Ccost,change/nBN)
In the formula, Kom,ESAnd Ccost,changeRespectively representing the running cost coefficient and the replacement cost of the energy storage unit; n isBAnd nBNRespectively the number of charging and discharging conversion in one period of the energy storage unit and the rated charging and discharging number in the life cycle,
the frequency modulation power supply model in the step 3 is concretely as follows:
step 3.2: the goal of the FM power scheduling model is to minimize FM cost, i.e.
Cf＝CfES,cost+CfIL,cost
In the formula, CfES,costAnd CfIL,costThe frequency modulation costs of the frequency modulated battery and the interruptible load respectively,
the constraint that the frequency-modulated power supply needs to satisfy has a power balance equation, i.e.
Pload-cut＝PfES+PfIL
In the formula, PfESAnd PfILBattery output and interruptible load power scheduled for the second stage, respectively.
CN202210329613.4A 2022-03-30 2022-03-30 Two-stage day-ahead economic dispatching method for off-grid micro-grid Pending CN114744612A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210329613.4A CN114744612A (en) 2022-03-30 2022-03-30 Two-stage day-ahead economic dispatching method for off-grid micro-grid

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210329613.4A CN114744612A (en) 2022-03-30 2022-03-30 Two-stage day-ahead economic dispatching method for off-grid micro-grid

Publications (1)

Publication Number Publication Date
CN114744612A true CN114744612A (en) 2022-07-12

Family

ID=82279234

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210329613.4A Pending CN114744612A (en) 2022-03-30 2022-03-30 Two-stage day-ahead economic dispatching method for off-grid micro-grid

Country Status (1)

Country Link
CN (1) CN114744612A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116451880A (en) * 2023-06-16 2023-07-18 华北电力大学 Distributed energy optimization scheduling method and device based on hybrid learning

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116451880A (en) * 2023-06-16 2023-07-18 华北电力大学 Distributed energy optimization scheduling method and device based on hybrid learning
CN116451880B (en) * 2023-06-16 2023-09-12 华北电力大学 Distributed energy optimization scheduling method and device based on hybrid learning

Similar Documents

Publication Publication Date Title
CN110119886B (en) Active distribution network dynamic planning method
CN109193815B (en) Combined heat and power dispatching method for improving wind power consumption
CN111900727B (en) PSO-based optical storage, charging and replacement integrated charging station collaborative optimization scheduling method and device
Qi et al. Comprehensive optimized hybrid energy storage system for long-life solar-powered wireless sensor network nodes
CN112713618B (en) Active power distribution network source network load storage cooperative optimization operation method based on multi-scene technology
Bao et al. Battery energy storage system load shifting control based on real time load forecast and dynamic programming
CN103795079A (en) Off-grid grid-connected hybrid photovoltaic power generation control system and economical operation optimization method thereof
CN110138006B (en) Multi-microgrid coordinated optimization scheduling method considering new energy electric vehicle
CN111626527B (en) Intelligent power grid deep learning scheduling method considering fast/slow charging/discharging form of schedulable electric vehicle
CN109948823B (en) Self-adaptive robust day-ahead optimization scheduling method for light storage charging tower
CN104200297A (en) Energy optimizing dispatching method of home hybrid power supply system in real-time power price environment
CN107565607A (en) A kind of micro-capacitance sensor Multiple Time Scales energy dispatching method based on Spot Price mechanism
CN111555356B (en) Random variable analysis method for regional comprehensive energy system based on coupled Markov model
CN111162524A (en) Control method and system for electric vehicle charging user to access power distribution network
Chaoui et al. Deep reinforcement learning energy management system for multiple battery based electric vehicles
CN116151486B (en) Multi-time-scale random optimization method and device for photovoltaic charging station with energy storage system
CN111786417A (en) Distributed new energy consumption-oriented active power distribution network multi-target interval optimization scheduling method
CN203933038U (en) From the grid-connected mixing photovoltaic power generation control system of net
CN114448044A (en) Bidirectional quick-charging ordered charging and discharging method and system for power changing station
CN112101632B (en) Building load optimal scheduling method based on distributed estimation algorithm
CN115549159A (en) Large-scale electric vehicle grouping scheduling strategy considering peak shaving cost
CN105896580A (en) Multi-objective optimization control method and device of microgrid
CN114744612A (en) Two-stage day-ahead economic dispatching method for off-grid micro-grid
CN114725930A (en) Self-adaptive power system scheduling method and device
TWI639962B (en) Particle Swarm Optimization Fuzzy Logic Control Charging Method Applied to Smart Grid

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