CN109687440A - One kind is relaxed control distributed power generation investment and distribution plan optimization method under environment - Google Patents
One kind is relaxed control distributed power generation investment and distribution plan optimization method under environment Download PDFInfo
- Publication number
- CN109687440A CN109687440A CN201811603130.9A CN201811603130A CN109687440A CN 109687440 A CN109687440 A CN 109687440A CN 201811603130 A CN201811603130 A CN 201811603130A CN 109687440 A CN109687440 A CN 109687440A
- Authority
- CN
- China
- Prior art keywords
- frog
- population
- power generation
- memeplex
- distributed power
- 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
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
-
- 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]
Abstract
It relaxs control distributed power generation investment and distribution plan optimization method under environment the embodiment of the invention discloses one kind, including model construction and improved multiple target shuffled frog leaping algorithm, the model construction includes objective function, constraint condition and Uncertainty Management, and the improved multiple target shuffled frog leaping algorithm includes shuffled frog leaping algorithm mechanism and step；The present invention, have studied the combined optimization problem of distributed power generation investment and distribution network construction planning, profit distribution coefficient is introduced during constructing objective function to reach two-win purpose, method proposed by the invention can make the significantly more efficient investment and planning for encouraging various distributed generation technologies of DNO, the uncertainty of the following various parameters has been allowed also for, designer can have been helped more accurately to develop planning.
Description
Technical field
The present invention relates to distributed generation technology field more particularly to one kind distributed power generations under environment of relaxing control to throw
Money and distribution plan optimization method.
Background technique
Distributed generation resource refers to the low capacity power generating equipment for being directly arranged at power distribution network or being distributed near load, with
The drawbacks of centralized power generation, long distance power transmission and bulk power grid interconnect constantly shows, conventional power generation increased costs and people's environmental protection
Consciousness constantly enhancing, distributed power generation more and more attention has been paid to, however distributed power generation unit access power distribution network, can change and match
The direction of energy in electric line, planning operation or even safety benefit and economic benefit to power distribution network can all bring a fixing
It rings, under the market environment relaxed control, the investment of distributed power generation and Operation Decision are determined by distributed power generation quotient,
And guarantee that power grid security reliability service is the responsibility of distribution network operation business, in this context, the investment operation of DG can band to DNO
What kind of, which comes, influences, and how to guide DGO and DNO to make rational planning for, science decision, the two interests are taken into account and increase society
The total welfare of meeting, is good problem to study.
Since distributed power generation distributes, current research distributed power generation and distribution closely bound up with distribution network planning rationally
The pertinent literature that network planning is drawn can be divided into single planning and comprehensive coordination two classes of planning: Qian Zheshi according to the type of decision variable
In the case where not changing system feeder line and substation configures, the installation site and capacity of distributed power generation are optimized, and
The latter is then the integrated planning of distributed power generation and power distribution network substation or feeder line etc., is a kind of global optimization planning, according to point
The computation model of cloth generation optimization configuration can be divided into single object optimization and two kinds of multiple-objection optimization, the former can be divided into again
Following three classes: from investment angle, with the minimum optimization aim of power distribution company cost of investment；From loss angle, most with distribution network loss
Small is optimization aim；From environmental benefit angle, the installed capacity that generates electricity in a distributed manner is up to optimization aim, multi-objective restriction,
The Model for Multi-Objective Optimization of distributed generation planning is established from two angles of economy and safety, such as uses distributed power generation
Cost of investment minimum, loss minimization and air extract maximum establish Model for Multi-Objective Optimization, mostly only from one party
Economic interests it is optimal establish Optimized model for objective function, do not account between DNO and DGO existing interest relations and its
The influence that may cause during distributed power generation and distribution network planning lacks to the harmony of interests machine between DNO and DGO
System is studied.
Summary of the invention
It relaxs control distributed power generation investment and distribution plan optimization side under environment the embodiment of the invention provides one kind
Method, including model construction and improved multiple target shuffled frog leaping algorithm, the improved multiple target shuffled frog leaping algorithm include mixed
Conjunction leapfrogs algorithm mechanism and step, and the shuffled frog leaping algorithm is a kind of novel heuristic group based on global collaborative search
Body evolution algorithm carries out heuristic search by heuristic function, to find the solution of combinatorial optimization problem, the mixing leapfrogs
The parameter that algorithm includes specifically includes that F, the quantity of frog group；M, the quantity of population；N, the quantity of frog in population；Smax, most
Allow step-length of beating greatly；Px, globally optimal solution；Pb, locally optimal solution；Pw, the worst solution in part；Q, the number of frog in subgroup
Amount；LS, part member evolution number；SF, global thoughts communication number, the standard leapfrog algorithm the step of it is as follows:
S0, initialization: generating an initial frog population F at random in feasible zone, determines m and n, i.e., in entire population
Quantity F=m × n of frog；
S1, a virtual population is generated: in solution space, generate F frog U (1), U (2) ... ..., U
(F), wherein n is dimension variable, each frog indicates a candidate solution of solution space in optimization problem, and i-th of frog is in n
Coordinate in dimension space is U (i)=(Ui1, Ui2 ..., Uin), and the performance of calculated U (i) is indicated with f (i)；
S2, to frog divided rank: by F frog according to performance quality carry out descending arrangement, generate array X={ U
(i), f (i), i=1,2 ..., F }, i=1 indicates the position (performance) of this frog preferably, records best frog in population
Position Px=U (1)；
S3, frog is grouped: is respectively put into different memeplex, array X is divided into m memeplex:Y1,
Y2 ..., Ym have n frog, i.e. Yk=[U (j) k, f (j) k ∣ U (j) k=U (k+m (j-1)), f (j) k in each memeplex
=f (k+m (j-1)), j=1 ..., n], wherein k=1,2 ..., m；
S4, every group of memeplex execute memetic and evolve: in each population, every frog all can be by other frogs
Influence, evolved by memetic, so that every frog is towards target position Step wise approximation；
S5, frog jump movement between population: it is performed in each memeplex after certain memetic evolves, it will
Each subgroup Y1, Y2 ..., Ym are merged into X, i.e. X={ Yk, k=1,2 ..., m } simultaneously re-starts descending arrangement to X, and updates
Best frog Pg in population；
S6, termination condition: if meeting stopping criterion for iteration, stopping search, and otherwise re-executes step3, general feelings
Under condition, after the circulation for performing certain number is evolved, represents algorithm when the frog preferably solved is no longer changed and stops,
The standard that can also be terminated by defining maximum evolution number as algorithm.
Optionally, the evolutionary step of the S4 includes:
S4-0, im=0 is set, im indicates the counting to memeplex, changes 0 between m, the quantity m with memeplex
Compare, if iN=0, iN indicate evolution number, compared with the maximum evolution times N allowed in each memeplex, each
In memeplex, Pb and Pw respectively indicate performance preferably with the worst frog, and Pg indicates frog best in entire population, sub
Population is evolved every time can only improve the position of the worst frog Pw, can not improve global worst frog；
S4-1, im=im+1；
S4-2, iN=iN+1；
S4-3, the position for adjusting the worst frog, method of adjustment are as follows: mobile distance the Di=rand () * (Pb-Pw) of frog
(14), new position Pw=Pw+Di, (Dmax >=Di >=-Dmax) (15), wherein rand () is the random number between 0 to 1,
Dmax is the maximum distance for allowing frog mobile；
If S4-4, execution aforesaid operations can obtain the frog of a more good position, one can be generated preferably
Solution, then just replacing original frog with the frog of new position, otherwise, just replaces Pb with Pg, repeats the above process；
It is just random to generate a new explanation substitution if S4-5, the above method cannot still generate the frog of more good position
The originally frog Pw of the worst position；
If S4-6, iN < N, 4-2 is executed；
If S4-7, im < m, 4-1 is executed, otherwise frog jumps, and re-executes global search.
Optionally, the model construction includes objective function, constraint condition and Uncertainty Management.
Optionally, the constraint condition includes conventional constraint and exhaust emission constraint.
Optionally, the Uncertainty Management includes Wind turbines power generation Uncertainty Management and the market demand and motor
Uncertainty Management.
As can be seen from the above technical solutions, the embodiment of the present invention has the advantage that
1, the invention proposes one for solving the multiple target dynamic model of distributed power generation and distribution network planning, and
Using improved multiple target mixing leapfrog decision making algorithm solve model built, used in Double Step algorithm, first looking for can
Make the noninferior solution of DNO and DGO profit maximization simultaneously, preferred plan scheme is then chosen from Candidate Set again, finally by institute
Established model is applied in an actual distribution network, demonstrates the flexibility and validity of model built by comparative analysis,
The purpose of the present invention is not to formulate to force trading rules, but propose a transaction new approaches for both sides, while can have
Effect meets technology, economy and environmental constraints, and method proposed by the invention can make the significantly more efficient encouragement of DNO various
The investment and planning of distributed generation technology, have allowed also for the uncertainty of the following various parameters, can help to plan
Person more accurately develops planning.
2, the present invention takes into account the interests of distribution network operation business Yu distributed power generation quotient, have studied distributed power generation invest and
The combined optimization problem of distribution network construction planning introduces profit distribution coefficient during constructing objective function to reach double
Purpose is won, while needing to meet power distribution network safe and stable operation and emission reduction targets constraint, for uncertainty present in model
Factor is determined using two-point estimate method, and the present invention carries out multiple target shuffled frog leaping algorithm using random simplex method excellent
Change for solving model built, finally by taking IEEE33 Node power distribution system as an example, is tested by the comparative analysis between different situations
The validity and feasibility of method proposed by the invention are demonstrate,proved.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
Some embodiments of the present invention, for those of ordinary skill in the art, without any creative labor,
It can also be obtained according to these attached drawings other attached drawings.
Fig. 1 is two-point estimate method implementation flow chart in the present invention；
Fig. 2 is that the random simplex method in the present invention optimizes shuffled frog leaping algorithm flow chart；
Fig. 3 is IEEE33 Node power distribution system wiring diagram in the present invention；
Fig. 4 is the target function value in the present invention under the situation of α=0 when each scheme Pareto optimality；
Fig. 5 is the target function value in the present invention under the situation of α ≠ 0 when each scheme Pareto optimality；
Fig. 6 is 1 different type DG characteristic of table in the present invention；
Fig. 7 is blower technical characteristic in the present invention；
Fig. 8 is workload demand, bidding price adjustment coefficient and Duration Prediction result in the present invention；
Fig. 9 is required other data used in studying in the present invention；
Figure 10 is the plan arrangement of solution 1 in the present invention；
The value and DNO and DGO satisfaction that Figure 11 is α in each scheme in the present invention；
Figure 12 is the plan arrangement of scheme 16 in scene 2 in the present invention.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawing and specific embodiment party
The present invention is described in further detail for formula.Obviously, described embodiments are only a part of the embodiments of the present invention, and
The embodiment being not all of.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work
Under the premise of every other embodiment obtained, shall fall within the protection scope of the present invention.
Embodiment 1
Fig. 1-Fig. 5 is please referred to, the present invention the following technical schemes are provided: one kind is relaxed control throw under environment by distributed power generation
Money and distribution plan optimization method, including model construction and improved multiple target shuffled frog leaping algorithm, the improved multiple target
Shuffled frog leaping algorithm includes shuffled frog leaping algorithm mechanism and step, from the shuffled frog leaping algorithm the step of from the point of view of, which, which has, patrols
Volume clearly frame structure, but SLFA uses formula (14), (15) calculating or with Pg generation during carrying out local optimal searching
For Pb, when occurring, there is no then generate a frog in the case where more excellent frog at random in search space to continue optimizing, but this
When may will appear the new frog position might as well before worst frog the case where, therefore in the present invention with random single
Pure shape method improves SLFA local search ability and is presented to policymaker as far as possible to effectively avoid the generation of the above problem
More representative noninferior solutions avoids falling into locally optimal solution, and wherein RNM method is the improvement to simplex method, can make
The direction of search is more accurate, finds out potential substitution point that may be present near the direction of search, and improved multiple target mixing leapfrogs
Algorithm flow chart as shown in Fig. 2, when solving the combined optimization model of distributed power generation and distribution network planning using IMO-SFLA,
Firstly the need of it is clear that the practical significance that frog refers in the problem in SFLA, from the foregoing, it can be understood that the algorithm passes through determination
Type, investment time, position and the capacity and distribution network line of distributed power generation unit and the construction situation of substation make
It obtains DGO and DNO and reaches the state of two-win, while realizing emission reduction targets, therefore, the frog in SFLA is in problem of the invention
The optimal location and state, frog individual adaptation degree for indicating distributed power generation quotient and the investment of distribution network operation business are asked corresponding to this
The functional value of topic scalar functions, the shuffled frog leaping algorithm are a kind of novel heuristic groups based on global collaborative search
Evolution algorithm carries out heuristic search by heuristic function (any mathematical function), so that the solution of combinatorial optimization problem is found,
The algorithm has concept simple, and adjusting parameter is few, and calculating speed is fast, and global search optimizing ability is strong, it is easy to accomplish the features such as,
The algorithm that leapfrogs is derived from the research to frog foraging behavior, and basic thought is: live a frog population in a piece of wetland,
Discrete in wetland to be dispersed with many stones, frog goes to find the more ground of food by carrying out jump between different stone
Side, every frog individual have the culture of oneself, while can realize the exchange of information between frog by the exchange of culture, from
And the ability of entire population search of food is promoted, the entire frog group of wetland is divided into different sons first by the execution of algorithm
Group, each sub-group are known as a memeplex, suffer from the culture of oneself, execute local searching strategy respectively, in subgroup
Each individual in body has the culture of oneself, and affects other individuals, is also influenced, is passed through by other individuals
Memetic evolves to develop, and after group Swarm Evolution to certain phase, the exchange for carrying out thought between each sub-group again is (complete
Office's information exchange) realize hybrid operation between sub-group, until set condition meets, the shuffled frog leaping algorithm
The parameter for including specifically includes that F, the quantity of frog group；M, the quantity of population；N, the quantity of frog in population；Smax, maximum permit
Perhaps it beats step-length；Px, globally optimal solution；Pb, locally optimal solution；Pw, the worst solution in part；Q, the quantity of frog in subgroup；LS,
Part member evolution number；SF, global thoughts communication number, the standard leapfrog algorithm the step of it is as follows:
S0, initialization: generating an initial frog population F at random in feasible zone, determines m and n, i.e., in entire population
Quantity F=m × n of frog；
S1, a virtual population is generated: in solution spaceGenerate F frog U (1), U (2) ... ..., U
(F), wherein n is dimension variable, each frog indicates a candidate solution of solution space in optimization problem, and i-th of frog is in n
Coordinate in dimension space is U (i)=(Ui1, Ui2 ..., Uin), and the performance of calculated U (i) is indicated with f (i)；
S2, to frog divided rank: by F frog according to performance quality carry out descending arrangement, generate array X={ U
(i), f (i), i=1,2 ..., F }, i=1 indicates the position (performance) of this frog preferably, records best frog in population
Position Px=U (1)；
S3, frog is grouped: is respectively put into different memeplex, array X is divided into m memeplex:Y1,
Y2 ..., Ym have n frog, i.e. Yk=[U (j) k, f (j) k ∣ U (j) k=U (k+m (j-1)), f (j) k in each memeplex
=f (k+m (j-1)), j=1 ..., n], wherein k=1,2 ..., m；
S4, every group of memeplex execute memetic and evolve: in each population, every frog all can be by other frogs
Influence, evolved by memetic, so that every frog is towards target position Step wise approximation；
S5, frog jump movement between population: it is performed in each memeplex after certain memetic evolves, it will
Each subgroup Y1, Y2 ..., Ym are merged into X, i.e. X={ Yk, k=1,2 ..., m } simultaneously re-starts descending arrangement to X, and updates
Best frog Pg in population；
S6, termination condition: if meeting stopping criterion for iteration, stopping search, and otherwise re-executes S3, ordinary circumstance
Under, after the circulation for performing certain number is evolved, represents algorithm when the frog preferably solved is no longer changed and stops,
The standard that can be terminated by defining maximum evolution number as algorithm,
Specifically, the evolutionary step of the S4 includes:
S4-0, im=0 is set, im indicates the counting to memeplex, changes 0 between m, the quantity m with memeplex
Compare, if iN=0, iN indicate evolution number, compared with the maximum evolution times N allowed in each memeplex, each
In memeplex, Pb and Pw respectively indicate performance preferably with the worst frog, and Pg indicates frog best in entire population, sub
Population is evolved every time can only improve the position of the worst frog Pw, can not improve global worst frog；
S4-1, im=im+1；
S4-2, iN=iN+1；
S4-3, the position for adjusting the worst frog, method of adjustment are as follows: mobile distance the Di=rand () * (Pb-Pw) of frog
(14), new position Pw=Pw+Di, (Dmax >=Di >=-Dmax) (15), wherein rand () is the random number between 0 to 1,
Dmax is the maximum distance for allowing frog mobile；
If S4-4, execution aforesaid operations can obtain the frog of a more good position, one can be generated preferably
Solution, then just replacing original frog with the frog of new position, otherwise, just replaces Pb with Pg, repeats the above process；
It is just random to generate a new explanation substitution if S4-5, the above method cannot still generate the frog of more good position
The originally frog Pw of the worst position；
If S4-6, iN < N, 4-2 is executed；
If S4-7, im < m, 4-1 is executed, otherwise frog jumps, and re-executes global search.
Specifically, the model construction includes objective function, constraint condition and Uncertainty Management, the objective function
Including two main bodys that DNO and DGO are in the chain of market interest, i.e., connection is close and there are interests contradictions, if only paying attention to a certain
The interests of side necessarily cause the decline of market efficiency therefore to follow different mesh when carrying out investment decision in view of DNO and DGO
It marks, the present invention wishes by formulating a Revenue Sharing Mechanism interests between DNO and DGO to be coordinated, and guides DGO
The interests of DNO are considered when carrying out investment decision, to realize Pareto optimality, objective function of the invention is as follows:
max min{ξ1,ξ2}
Wherein,
ξ1=(1- α) × DNOp
ξ2=(DGOp-DGOc)+α×DNOp
In formula, ξ 1 and ξ 2 respectively indicate the profit of DNO and DGO；α indicates DNO to the profit distribution ratio of DGO；DNOp table
Show that DNO bring profit is given in the presence due to DG；DGOp and DGOc respectively indicates the income and cost of DGO.
DNO is since the profit that DG exists and obtains by calculating cost difference of DNO when without there is DG by that can be obtained
It arrives, specific method are as follows: calculate the cost of DNO twice, be once the cost calculated in the case of no DG is participated in, i.e.,
DNOnodgc；Once be calculate have DG participate in planning in the case of cost, i.e. DNOdgc, this two-part difference precisely due to
The presence of DG, DNO surplus profit DNOp obtained:
Wherein, the cost DNOc of DNO mainly includes four parts: line loss cost CL, track investment and O&M cost CF, being become
The discharge costs TEC of power station investment and O&M cost CS and institute's power purchase energy, in the present invention, it is assumed that all investments all exist
The beginning of the year carries out, then DNOc can be expressed as:
Wherein t and the t ' expression of years, year；T indicates planning horizon, year；ω indicates load level, KW；N ω indicates planning
The load level set just predicted, KW；P ω indicates rate for incorporation into the power network when load level is ω, member/kWh；Plosst, ω are indicated
In t, active loss when load level is ω, kW；η ω indicates the duration of load level ω, h；D is discount
Rate, %；Nl indicates new route number in power distribution network, item；Cl indicates the specific investment cost cost of distribution line, Wan Yuan；Ctr is indicated
The cost of investment of substation, Wan Yuan；Dl indicates line length, KM；γ lt ' and ψ trt ' are route and the substation in t ' year respectively
The Boolean variable of investment, 0 or 1；TEt indicates the total emission volumn of t, ton.
The scene that the present invention is considered is all that DG is operated normally under environment, and the cost DGOc of DGO includes O&M cost CO
With cost of investment CI, failure cost, expression are put aside are as follows:
Wherein COn indicates the O&M cost of n distributed power generation unit, and CIn indicates n distributed power generation unit
Cost of investment, in this model consider have m class different distributed generator set type；I indicates that bus, Nb indicate distribution
Net median generatrix quantity, item；COdg indicates the unit operating cost of DG, Wan Yuan/KW；Pdgi, t, ω indicate DG in t, demand water
When putting down as ω, the active power exported to bus i, KW；δ dgi, t is indicated for bus i, in boolean's change that the DG of t is invested
Amount, 0 or 1；CIdg indicates the specific investment cost cost of DG, Wan Yuan/KW.
The revenue stream of DGO depends on the role of DGO in the market, can be in power selling income, the price type of electric energy
Bilateral contracts electricity price can also use the market price with market guidance, the electricity price of DGO of the present invention, and income GDOp can be expressed
Are as follows:
Specifically, the constraint condition includes conventional constraint and exhaust emission constraint, in order to guarantee the safety and stability of system operation
Property, operation of power networks conventional constraint condition must be taken into consideration during constructing Revenue Sharing Mechanism, specifically include that power-balance about
Beam, voltage constraint, DG units limits, route and substation capacity constraint.
Power-balance constraint expression formula are as follows:
Wherein Pneti, t, ω and Qneti, t, ω are illustrated respectively in t and export only when desired level is ω to bus i
Active and reactive power, KW, Kvar；PDi, t, ω and QDi, t, ω are illustrated respectively in t, when desired level is ω, bus
Active and reactive requirement on i, KW, Kvar；Qdgi, t, ω indicate that DG in t, when desired level is ω, is exported to bus i
Reactive power, Kvar.
Voltage constraint expression formula:
Wherein Vmin and Vmax respectively indicates minimum, most high operation voltage limitation, KV；Vi, t, ω indicate needing in t
Under the conditions of seeking horizontal ω, the voltage class of bus i, KV.
DG units limits expression formula are as follows:
Wherein Pdglim indicates the maximum run-limiting of unit DG, KW.
Route and substation capacity constraint expression formula are as follows:
Wherein Il, t, ω indicate route l in t, current strength when desired level is ω, A；Il indicates planning horizon
Current strength when beginning in route l, A；Indicate the current strength of newly-increased route when t, A.I and j respectively indicates route l
Transmitting terminal and receiving end, Ztl indicate impedance of the route l in t, Ω.
Wherein Sgr idt, ω indicate that, in t, desired level flows through the apparent energy of substation, VA when being ω；Str table
Show the amount of capacity of substation when planning horizon starts, KVA；It indicates when t since newly-increased or reconstruction becomes
Power station and increased amount of capacity, KVA.
Currently, environmental problem is more taken seriously.Environmental benefit is included in constraint condition by the present invention, system discharge by
Total amount limitation, is shown below:
Et≤Elim
In formula, Et indicates the total emission volumn of t, ton；Elim is discharge limit, ton.
Total discharge amount mainly includes two parts: being generated by the discharge of mains network generation and by renewable generating set
Discharge, therefore, total emission volumn Et can be indicated are as follows:
Wherein eg and edg respectively indicates power grid and the emission factor of DG, and Pgt, ω are indicated in t, when desired level is ω
The electric energy bought by power grid, KWh.
Specifically, the Uncertainty Management includes Wind turbines power generation Uncertainty Management and the market demand and motor
Uncertainty Management, the output power of wind power generating set depend primarily on local wind speed, and historical wind speed data is usually by gas
As part is recorded as unit of hour, under normal circumstances, the variation of wind speed is counted as a stochastic variable, can make
Local wind speed is depended primarily on the output power that Weibull distribution is expressed as wind power generating set, historical wind speed data is logical
Often recorded as unit of hour by meteorological part, under normal circumstances, the variation of wind speed is counted as a stochastic variable,
Weibull distribution expression can be used are as follows:
Wherein vk indicates the wind speed in the area k, m/s；Ak and bk respectively indicate shape index and scaled index, calculating side
Method are as follows:
ak=(σk/μk)-1.086
Wherein μ k and σ k respectively indicates wind speed mean value and standard deviation of the area k in special time period.
According to known wind speed profile function, the output power of wind power generating set can be by according to the technical characteristic of blower
Following formula determines:
Wherein vin, vout and vrated respectively indicate the incision of blower, cuts out and rated wind speed, m/s；Prated is indicated
The rated power of blower, KW；δ wi, t is indicated for bus i, in boolean's change that the distributed wind-power generator unit of t is invested
Amount, 0 or 1.
In general, power load and the electricity price based on market be it is very uncertain, the present invention in order to simplify meter
It calculates, power load and electricity price level is defined, and is uncertain by adjusting coefficient reflection system bring.
In system, long-term active and reactive power demand can be indicated are as follows:
Wherein, PDi, base and QDi, substantially active and reactive power when base is illustrated respectively in First Year on bus i
Demand, KW, Kvar；PDi, t, ω and QDi, t, ω are illustrated respectively in t, when desired level is ω, active on bus i and
Reactive power demand, KW, Kvar；τ ω indicates that workload demand regulation coefficient when desired level is ω, β are load growth rate, %.
Under environment of opening the markets, power purchase price is to be determined by competition, therefore electricity price is simultaneously under different desired levels
Be not it is constant, in order to without loss of generality, it is assumed that the electricity price level in the case where different demands are horizontal are as follows:
pω=p γω
Wherein p is base price, member/KWh；γ ω indicates the bidding price adjustment coefficient when desired level is ω, and assumes
This regulation coefficient is known.
For the uncertainty of long-term load demand and electricity price, handled using formula (11) and (12), however formula
In the value of τ ω and γ ω be still uncertain, present invention assumes that probability density functions of these uncertain values meet logarithm
Normal distribution, at the same the average and standard deviation of τ ω and γ ω probability density function be it is determining, it is specific to determine that method is as follows
It is described:
In the present invention, the average value and variance of probability density function are determined using two-point estimate method, specific method is such as
Under:
Assuming that there is a function, this method is for solving the probability density function in known uncertain variable xi
Under the conditions of, the problem of how solving the probability density function of Y.Implementing procedure is as shown in Figure 1.
Embodiment 2
Fig. 3 and Fig. 6-Figure 12 is please referred to, the present invention surveys proposed method by IEEE-33 node power distribution net system
It calculates, system structure is as shown in figure 3, have 32 branches, 5 interconnection switch branches, the average load of each load point in the system
And power factor (PF) is respectively 55.5KW and 0.9285, which passes through the substation of a 20KV, and St=0tr, s=
20MW, the single-machine capacity Captr=10MW of transformer, ten thousand yuan of monovalent Ctr=50, present invention consideration exists simultaneously non-renewable
With renewable DG situation, Fig. 6 gives the correlation properties of Gas Generator Set, diesel oil, cogeneration of heat and power and wind power generating set, Fig. 7
Give the technical characteristic of wind power generating set, it is contemplated that four kinds of load levels, i.e. low ebb load, waist lotus, Ji He and height
Peak load, Fig. 8 give the workload demand of prediction and the regulation coefficient value of electricity price and respective duration, it is assumed that demand
The standard deviation sigma D ω of the horizontal adjustment coefficient and standard deviation sigma λ ω of electricity price level regulation coefficient is the 2% of respective mean value, and search is calculated
The termination condition of method is to reach maximum number of iterations, other emulation are assumed and the characteristic of distributed unit is not as shown in figure 9, ought
There are when the investment of distributed unit, the overall cost of ownership of DNO is 0.29 hundred million yuan, in order to verify having for established model of the invention
Effect property, it is contemplated that profitless distribute and have two kinds of situations of profit distribution
Scene 1: profit distribution, that is, α=0 is not present: the situation of profitless distribution is analyzed first, in this trifle, it is assumed that
Since the profit that DG exists and generates all is obtained by DNO in system, that is, α=0% is assumed, in this case, before Pareto optimality
Along there is 20 noninferior solutions, the value of objective function is as shown in figure 3, the plan of solution 1 arranges as shown in Figure 10, in the program
In, the net profit of DNO and DGO are positive value, and have used three kinds of DG, i.e. wind power generating set, Gas Generator Set and thermoelectricity connection
Unit is produced, Figure 10 gives installation and the investment time of bus simultaneously, in this scenario, has carried out the reinforcing of transmission line of electricity,
There is no the investment of substation；
Scene 2: there are profit distribution, that is, α ≠ 0: in the present case, it is assumed that DNO is while considering itself benefit, for drum
It encourages DGO investment and a certain proportion of profit distribution, i.e. α ≠ 0 is carried out to DGO, but is uncertain, in this scene, benefit of the DNO to DGO
Profit allocation proportion β is determined that there are 20 noninferior solutions, in all solutions, mesh in Pareto forward position obtained by optimization process
Offer of tender numerical value is positive value, this also means that lucrative on DNO and distributed power generation under either a program, not Tongfang
The difference of case is the difference of profit amount, therefore DNO from DGO has different tendencies during the choosing of the ratio of scheme, sharp
The amplitude of variation for moistening allocation proportion is very big, and minimum 31.2%, it is up to 99.1%, simulation result is as shown in figure 4, Fig. 4 is provided
The satisfaction of the profit of DNO and DGO under each scheme, profit distribution ratio α and each Interest Main Body profit maximization is as schemed
Shown in 11, when choosing final scheme, it is optimal for choosing the scheme for two objective functions when min-satisfaction degree maximum
Scheme, i.e. scheme 16 have used Wind turbines, Gas Generator Set, diesel engine unit and four kinds of cogeneration unit in this scenario
DG, the installation of bus and investment plan are as shown in figure 12, exist simultaneously track strengthening in this scenario and substation reinforces.
The present invention calculates proposed method by IEEE-33 node power distribution net system, system structure as shown in figure 3,
There are 32 branches, 5 interconnection switch branches in the system, the average load and power factor (PF) of each load point are respectively
55.5KW and 0.9285, the network pass through the substation of a 20KV, and St=0tr, s=20MW, and the single machine of transformer holds
Captr=10MW, ten thousand yuan of monovalent Ctr=50 are measured, present invention consideration exists simultaneously non-renewable and renewable DG situation, and Fig. 6 gives
The correlation properties of Gas Generator Set, diesel oil, cogeneration of heat and power and wind power generating set are gone out, Fig. 7 gives wind power generating set
Technical characteristic, it is contemplated that four kinds of load levels, i.e. low ebb load, waist lotus, Ji He and peak load, Fig. 8 give prediction
Workload demand and electricity price regulation coefficient value and respective duration, it is assumed that the standard deviation sigma of desired level regulation coefficient
The standard deviation sigma λ ω of D ω and electricity price level regulation coefficient is the 2% of respective mean value, and the termination condition of searching algorithm is to reach most
Big the number of iterations, other emulation are assumed and the characteristic of distributed unit is as shown in figure 9, when there is no distributed unit investment,
The overall cost of ownership of DNO is 0.29 hundred million yuan.
In summary: key problem in technology point of the invention is:
The first, the harmony of interests Optimized model between DNO and DGO is established, it is intended to solve different type distributed generator
Group (only has power output is probabilistic can be again in the machine set type that the present invention is considered comprising distributed wind-power generator one kind
The raw energy) and power distribution network joint planning problem, belong to Global Optimization Model；
The second, when optimization aim is set the present invention it is innovative the respective profit of DNO and DGO is passed through into distribution of interests
It is organically fused together, while joined emission reduction constraint in constraint condition, and leapfroged certainly using the mixing of improved multiple target
Plan algorithm solves established model；
Third carries out multiple scenarios by taking IEEE-33 node system as an example, demonstrates the validity of model built in text
And practicability.
It relaxs control distributed power generation investment and distribution plan optimization side under environment to one kind provided by the present invention above
Method is described in detail, and for those of ordinary skill in the art, thought according to an embodiment of the present invention is being embodied
There will be changes in mode and application range, in conclusion the content of the present specification should not be construed as to limit of the invention
System.
Claims (5)
- Distributed power generation investment and distribution plan optimization method under environment 1. one kind is relaxed control, including model construction and improved Multiple target shuffled frog leaping algorithm, it is characterised in that: the improved multiple target shuffled frog leaping algorithm includes shuffled frog leaping algorithm machine Reason and step, the shuffled frog leaping algorithm are a kind of novel heuristic Swarm Evolution algorithms based on global collaborative search, are led to It crosses heuristic function and carries out heuristic search, so that the solution of combinatorial optimization problem is found, the parameter that the shuffled frog leaping algorithm includes Specifically include that F, the quantity of frog group；M, the quantity of population；N, the quantity of frog in population；Smax, maximum allowable bounce step-length； Px, globally optimal solution；Pb, locally optimal solution；Pw, the worst solution in part；Q, the quantity of frog in subgroup；LS, part member are evolved secondary Number；SF, global thoughts communication number, the standard leapfrog algorithm the step of it is as follows:S0, initialization: generating an initial frog population F at random in feasible zone, determine m and n, i.e., frog in entire population Quantity F=m × n；S1, a virtual population is generated: in solution spaceF frog U (1) of generation, U (2) ... ..., U (F), Middle n is dimension variable, each frog indicates a candidate solution of solution space in optimization problem, and i-th of frog is in n-dimensional space Coordinate be U (i)=(Ui1, Ui2 ..., Uin), the performance of calculated U (i) is indicated with f (i)；S2, to frog divided rank: by F frog according to performance quality carry out descending arrangement, generate array X={ U (i), f (i), i=1,2 ..., F }, i=1 indicates the position (performance) of this frog preferably, records the position of best frog in population Px=U (1)；S3, frog is grouped: is respectively put into different memeplex, array X is divided into m memeplex:Y1, Y2 ..., There are n frog, i.e. Yk=[U (j) k, f (j) k ∣ U (j) k=U (k+m (j-1)), f (j) k=f (k+m in Ym, each memeplex (j-1)), j=1 ..., n], wherein k=1,2 ..., m；S4, every group of memeplex execute memetic and evolve: in each population, every frog all can be by the shadow of other frogs It rings, is evolved by memetic, so that every frog is towards target position Step wise approximation；S5, frog jump movement between population: it is performed in each memeplex after certain memetic evolves, it will be each Subgroup Y1, Y2 ..., Ym are merged into X, i.e. X={ Yk, k=1,2 ..., m } simultaneously re-starts descending arrangement, and Population Regeneration to X In best frog Pg；S6, termination condition: if meeting stopping criterion for iteration, stopping search, and otherwise re-executes step3, under normal circumstances, After the circulation for performing certain number is evolved, represents algorithm when the frog that preferably solves is no longer changed and stop, it can also be with The standard terminated by defining maximum evolution number as algorithm.
- 2. one kind according to claim 1 is relaxed control under environment distributed power generation investment and distribution plan optimization method, It is characterized by: the evolutionary step of the S4 includes:S4-0, im=0 being set, im indicates the counting to memeplex, changes 0 between m, compared with the quantity m of memeplex, If iN=0, iN indicate evolution number, compared with the maximum evolution times N allowed in each memeplex, in each memeplex In, Pb and Pw respectively indicate performance preferably and the worst frog, and Pg indicates frog best in entire population, sub- population every time into Change the position that can only improve the worst frog Pw, global worst frog can not be improved；S4-1, im=im+1；S4-2, iN=iN+1；S4-3, the position for adjusting the worst frog, method of adjustment are as follows: mobile distance the Di=rand () * (Pb-Pw) (14) of frog, New position Pw=Pw+Di, (Dmax >=Di >=-Dmax) (15), wherein rand () is the random number between 0 to 1, and Dmax is The maximum distance for allowing frog mobile；If S4-4, execution aforesaid operations can obtain the frog of a more good position, a preferably solution can be generated, that Just replace original frog, otherwise, just replace Pb with Pg with the frog of new position, repeats the above process；It is just random to generate a new explanation substitution originally if S4-5, the above method cannot still generate the frog of more good position The frog Pw of the worst position；If S4-6, iN < N, 4-2 is executed；If S4-7, im < m, 4-1 is executed, otherwise frog jumps, and re-executes global search.
- 3. one kind according to claim 1 is relaxed control under environment distributed power generation investment and distribution plan optimization method, It is characterized by: the model construction includes objective function, constraint condition and Uncertainty Management.
- 4. one kind according to claim 3 is relaxed control under environment distributed power generation investment and distribution plan optimization method, It is characterized by: the constraint condition includes conventional constraint and exhaust emission constraint.
- 5. one kind according to claim 3 is relaxed control under environment distributed power generation investment and distribution plan optimization method, It is characterized by: the Uncertainty Management includes that Wind turbines power generation Uncertainty Management and the market demand and motor are uncertain Property processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811603130.9A CN109687440A (en) | 2018-12-26 | 2018-12-26 | One kind is relaxed control distributed power generation investment and distribution plan optimization method under environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811603130.9A CN109687440A (en) | 2018-12-26 | 2018-12-26 | One kind is relaxed control distributed power generation investment and distribution plan optimization method under environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109687440A true CN109687440A (en) | 2019-04-26 |
Family
ID=66189676
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811603130.9A Pending CN109687440A (en) | 2018-12-26 | 2018-12-26 | One kind is relaxed control distributed power generation investment and distribution plan optimization method under environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109687440A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111697704A (en) * | 2020-07-04 | 2020-09-22 | 李英超 | Power supply information processing method of smart power grid |
| WO2022032960A1 (en) * | 2020-08-11 | 2022-02-17 | 国网河北省电力有限公司沧州供电分公司 | Evaluation method based on multi-algorithm fusion for offshore wind power grid connection mode |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809279A (en) * | 2016-03-03 | 2016-07-27 | 河海大学 | Multi-objective quantum Shuffled Frog Leaping Algorithm (SFLA) based water resource optimization and diversion method |
| CN106991442A (en) * | 2017-03-30 | 2017-07-28 | 中国矿业大学 | The self-adaptive kernel k means method and systems of shuffled frog leaping algorithm |
| CN107681669A (en) * | 2017-06-22 | 2018-02-09 | 国家电网公司 | Using the power network distribution idle work optimization method of shuffled frog leaping algorithm |
| CN107994582A (en) * | 2017-10-24 | 2018-05-04 | 中国南方电网有限责任公司电网技术研究中心 | Reconstruction method of power distribution network and system containing distributed generation resource |
-
2018
- 2018-12-26 CN CN201811603130.9A patent/CN109687440A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809279A (en) * | 2016-03-03 | 2016-07-27 | 河海大学 | Multi-objective quantum Shuffled Frog Leaping Algorithm (SFLA) based water resource optimization and diversion method |
| CN106991442A (en) * | 2017-03-30 | 2017-07-28 | 中国矿业大学 | The self-adaptive kernel k means method and systems of shuffled frog leaping algorithm |
| CN107681669A (en) * | 2017-06-22 | 2018-02-09 | 国家电网公司 | Using the power network distribution idle work optimization method of shuffled frog leaping algorithm |
| CN107994582A (en) * | 2017-10-24 | 2018-05-04 | 中国南方电网有限责任公司电网技术研究中心 | Reconstruction method of power distribution network and system containing distributed generation resource |
Non-Patent Citations (4)
| Title |
|---|
| 王园媛: "求解组合优化问题的混合蛙跳算法的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王晓笛: "基于改进蛙跳算法的多目标优化问题研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 陈政等: "放松管制环境下基于改进布谷鸟搜索算法的DGO与DNO效益均衡模型", 《电力建设》 * |
| 马志伟等: "放松管制环境下的配网侧分布式发电协调规划模型", 《水电能源科学》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111697704A (en) * | 2020-07-04 | 2020-09-22 | 李英超 | Power supply information processing method of smart power grid |
| CN111697704B (en) * | 2020-07-04 | 2021-05-11 | 李英超 | Power supply information processing method of smart power grid |
| WO2022032960A1 (en) * | 2020-08-11 | 2022-02-17 | 国网河北省电力有限公司沧州供电分公司 | Evaluation method based on multi-algorithm fusion for offshore wind power grid connection mode |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rouzbahani et al. | A review on virtual power plant for energy management | |
| Sedighizadeh et al. | Stochastic multi-objective economic-environmental energy and reserve scheduling of microgrids considering battery energy storage system | |
| Nagarajan et al. | Combined economic emission dispatch of microgrid with the incorporation of renewable energy sources using improved mayfly optimization algorithm | |
| Liu et al. | A blockchain-enabled secure power trading mechanism for smart grid employing wireless networks | |
| Maghouli et al. | A scenario-based multi-objective model for multi-stage transmission expansion planning | |
| Qiu et al. | A literature survey of optimal power flow problems in the electricity market context | |
| Yi et al. | Coordinated operation strategy for a virtual power plant with multiple DER aggregators | |
| CN109523060A (en) | Ratio optimization method of the high proportion renewable energy under transmission and distribution network collaboration access | |
| Liu et al. | Worst-case conditional value-at-risk based bidding strategy for wind-hydro hybrid systems under probability distribution uncertainties | |
| CN108565900A (en) | A kind of distributed energy optimizing operation method based on game theory | |
| Peng et al. | Optimized economic operation strategy for distributed energy storage with multi-profit mode | |
| Meena et al. | Modified taguchi-based approach for optimal distributed generation mix in distribution networks | |
| CN109687440A (en) | One kind is relaxed control distributed power generation investment and distribution plan optimization method under environment | |
| Cao et al. | An interactive tri-level multi-energy management strategy for heterogeneous multi-microgrids | |
| Peng et al. | Review on bidding strategies for renewable energy power producers participating in electricity spot markets | |
| Yan et al. | Stochastic optimal scheduling strategy of cross-regional carbon emissions trading and green certificate trading market based on Stackelberg game | |
| Sreenivasulu et al. | A coordinated stochastic dispatch model for hybrid energy markets with renewable energy uncertainties using moth flame optimization | |
| Slochanal et al. | Generation expansion planning in the competitive environment | |
| Wang et al. | Research on short‐term and mid‐long term optimal dispatch of multi‐energy complementary power generation system | |
| Mohseni et al. | A game-theoretic approach to model interruptible loads: application to micro-grid planning | |
| Xie et al. | Emission-constrained optimal generation scheduling incorporating equity principles | |
| Abbasi et al. | Return on investment in transmission network expansion planning considering wind generation uncertainties applying non-dominated sorting genetic algorithm | |
| Feng et al. | Economic research on application of electric heating system based on game theory | |
| Drobyazko et al. | Formation of the intelligent energy system based on digital technologies in urban management | |
| CN110350575A (en) | Meter and wind-powered electricity generation receive the transmission system planing method of ability and Demand Side Response |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190426 |