CN105140971A - AC-DC micro-grid distributed scheduling method based on reweighed acceleration Lagrangian - Google Patents

AC-DC micro-grid distributed scheduling method based on reweighed acceleration Lagrangian Download PDF

Info

Publication number
CN105140971A
CN105140971A CN201510595842.0A CN201510595842A CN105140971A CN 105140971 A CN105140971 A CN 105140971A CN 201510595842 A CN201510595842 A CN 201510595842A CN 105140971 A CN105140971 A CN 105140971A
Authority
CN
China
Prior art keywords
microgrid
node
direct current
represent
branch road
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201510595842.0A
Other languages
Chinese (zh)
Other versions
CN105140971B (en
Inventor
李静
戴文战
赵忠伟
沈忱
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.)
Zhejiang Gongshang University
Original Assignee
Zhejiang Gongshang University
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 Zhejiang Gongshang University filed Critical Zhejiang Gongshang University
Priority to CN201510595842.0A priority Critical patent/CN105140971B/en
Publication of CN105140971A publication Critical patent/CN105140971A/en
Application granted granted Critical
Publication of CN105140971B publication Critical patent/CN105140971B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention discloses an AC-DC micro-grid distributed scheduling method based on reweighed acceleration Lagrangian. According to the method, an AC-DC mixed micro-grid linear cone optimal power flow model including converter power flow restraint is built, an AC-DC mixed micro-grid is divided into an AC micro-grid subsystem and a DC micro-grid subsystem by employing a virtual auxiliary variable method, a distributed coordination optimization method based on local reweighing Lagrangian is proposed, and distributed coordination optimization of resource scheduling of the whole grid is realized. According to the method, parallel and synchronous iterative solution of various subsystems can be realized, global coordination is not needed, only a little information interaction between the adjacent subsystems is needed, the calculating efficiency is high, and the convergence characteristic is good.

Description

A kind of alternating current-direct current micro-capacitance sensor distributed scheduling method accelerating Lagrangian based on heavy weighting
Technical field
The invention belongs to microgrid energy Optimum Scheduling Technology field, particularly relate to a kind of alternating current-direct current mixing micro-capacitance sensor distributed self-discipline economic dispatch method containing blower fan, solar-energy photo-voltaic cell, conventional diesel engine and energy storage device.
Background technology
21 century Mo, regenerative resource will play a leading role in world's energy resource structure.The development of regenerative resource causes the development of micro-capacitance sensor simultaneously, and micro-capacitance sensor can realize economical, efficiently to the variation of user and the high reliability requirement of powering, as a kind of beneficial complement of electrical network.Because photovoltaic cell initially produces direct current, and office and domestic premises electricity consumption mostly are DC load, photovoltaic and energy storage composition direct-current grid are powered to DC load, eliminate many transform parts, improve economy and the reliability of electrical network.Therefore, start to come into one's own in conjunction with the alternating current-direct current mixing micro-capacitance sensor exchanging microgrid and direct-current micro-grid advantage separately.The Optimum Scheduling Technology of research alternating current-direct current mixing micro-capacitance sensor, while guaranteeing that system meets high power supply reliability demand, can realize system cloud gray model total cost minimum.
Alternating current-direct current mixing micro-capacitance sensor optimal power flow problems, except considering the trend Constraints of Equilibrium of ac bus trend Constraints of Equilibrium and DC bus, also comprises the trend constraint of AC-DC transform part.These bring a large amount of nonlinear restrictions to the energy scheduling problem of alternating current-direct current mixing micro-capacitance sensor.Therefore, AC portion and direct current component decoupling zero in micro-capacitance sensor will be mixed, and respectively the nonlinear restriction of various piece is carried out reasonable convex lax, and adopt the method for distributed coordination iteration, significant to the research of alternating current-direct current mixing micro-capacitance sensor scheduling distributed optimization.
Summary of the invention
The present invention is intended to overcome the deficiencies in the prior art, a kind of distributed optimization technology towards alternating current-direct current mixing microgrid energy economic dispatch is provided, this technology based on local heavily weighting Lagrangian distributed optimization method solve alternating current-direct current composite cone optimal power flow problems, guarantee that system realizes total operating cost on reliable and stable operation basis minimum.
For achieving the above object, the technical scheme that the present invention takes specifically comprises following step:
1) the trend decoupling zero of branch road and direct current branch is exchanged in alternating current-direct current mixing micro-capacitance sensor;
Step 1) described in alternating current-direct current mixing micro-capacitance sensor primarily of interchange (AC) microgrid, direct current (DC) microgrid and ACDC converter composition, as shown in Figure 1 example alternating current-direct current mixing micro-capacitance sensor in, 1 #the node b of DC microgrid is connected with the node a1 exchanging microgrid by ACDC converter, and 2 #the node c of DC microgrid is connected with the node a2 exchanging microgrid by ACDC converter.
The stable state physical model of described ACDC converter branch road (b, a1) as shown in Figure 2, wherein V a1represent ac bus voltage, V abrepresent inverter output voltage, V brepresent DC bus-bar voltage, I brepresent DC bus current, R+jX represents converter equivalent reactance, P sand Q srepresent the meritorious and reactive power transmitted between AC system and converter respectively, P dand Q drepresent the meritorious and reactive power that converter exports respectively.Suppose that the operation of ACDC converter itself is complete equipilibrium symmetry and makes R=0, namely do not consider the consume of inverter inside, then the power flow equation of ACDC converter is:
P s=P d(1)
Q s = Q d - ( P d 2 + Q d 2 ) X / V a b 2 - - - ( 2 )
In formula, P d=V bi brepresent the active power of converter DC bus transmission; V ab=kMV brepresent DC bus-bar voltage and inverter output voltage relation and for direct voltage utilance; M is the index of modulation.
Converter adopts determines direct voltage and the control mode exchanging reactive power, is AC microgrid and DC microgrid part, as shown in Figure 3 at converter branch road (b, a1) place by mixing micro-capacitance sensor decomposition.After decoupling zero, the node a1 place connecting converter in AC microgrid with the addition of virtual auxiliary active load and virtual auxiliary reactive power generation machine, when alternating current-direct current mixing micro-capacitance sensor does the whole network tidal current analysis, active power meets formula (1), the idle q of output of virtual auxiliary reactive power generation machine bconstant and meet formula (2), connect the node b place voltage constant of converter in DC microgrid.
According to the trend decoupling method of converter branch road, the alternating current-direct current mixing micro-capacitance sensor shown in Fig. 1, decoupling zero is three parts, and AC microgrid is subsystem a, 1 #dC microgrid is subsystem b, 2 #dC microgrid is subsystem c, as shown in Figure 4, AC microgrid be connected communication line (solid line as in Fig. 4 between subsystem) after the decoupling zero of DC microgrid, represent that AC portion and direct current component need exchange message when subsystems microgrid inside is optimized scheduling.
2) alternating current-direct current composite cone optimal load flow model is set up;
Step 2) described in alternating current-direct current composite cone optimal load flow with electric network active loss and operating cost minimum for target function, that is:
min i m i z e f 0 = Σ t ∈ T Σ ( i , j ) ∈ Ψ a × Ψ d ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ a × Ψ d C j ( p j ( t ) ) - - - ( 3 )
In formula, T represents the running optimizatin cycle; Ψ aand Ψ drepresent the branch road collection of AC microgrid and DC microgrid in alternating current-direct current mixing micro-capacitance sensor respectively; (i, j) represents the branch road pointing to node j in power distribution network from node i (nearer compared to node j distance root node); C j() represents the power supply operating cost function of node j; p jt () represents the active power that node j injects in t; γ represents weight factor; R ijrepresent the resistance of branch road (i, j); w ij(t) and l ijt () is respectively two branch road supplementary variables of definition, meet following equation:
w ij(t):=0.5*(|V i(t)| 2+|I ij(t)| 2),l ij(t):=0.5*(|V i(t)| 2-|I ij(t)| 2)(4)
In formula, I ijt () represents the electric current flow through on t branch road (i, j); V it () represents the voltage at t branch road (i, j) parent node i; Symbol || represent the operator solving variation amplitude.Therefore (w ij(t)-l ij(t)) R ij=| I ij(t) | 2r ij, be the active loss of branch road (i, j).
Step 2) described in alternating current-direct current composite cone optimal load flow model comprises that AC microgrid linearly bores trend constraint, DC microgrid linearly bores trend constraint, the constraint of electric power netting safe running voltage level restraint, feeder current capacity-constrained, unit output, energy storage device energy constraint and ACDC converter Branch Power Flow retrain, specific as follows:
2.1) the AC microgrid described in is linearly bored trend constraint and is comprised two parts: exchange Branch Power Flow linear equality constraints and the inequality constraints of trend second order cone.Described interchange Branch Power Flow linear equality constraints is expressed as:
P i j ( t ) - Σ k : ( j , k ) ∈ Ψ a P j k ( t ) - R i j ( w i j ( t ) - l i j ( t ) ) = p j ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 5 ) Q i j ( t ) - Σ k : ( j , k ) ∈ Ψ a Q j k ( t ) - X i j ( w i j ( t ) - l i j ( t ) ) = q j ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 6 )
w i j ( t ) + l i j ( t ) - v j ( t ) - 2 ( R i j P i j ( t ) + X i j Q i j ( t ) ) + ( R i j 2 + X i j 2 ) ( w i j ( t ) - l i j ( t ) ) = 0 , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 7 )
w i j ( t ) + l i j ( t ) = v i ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 8 )
In formula, Ψ arepresent the branch road collection of AC microgrid in alternating current-direct current mixing micro-capacitance sensor; X ijrepresent the reactance of branch road (i, j); v iand v jrepresent represent respectively the voltage magnitude of branch road (i, j) node i and j square, i.e. v i=| V i(t) | 2and v j=| V j(t) | 2; P ij(t) and Q ijt () is illustrated respectively in t flows through this branch road active power and reactive power at branch road (i, j) parent node i end; p j(t) and q jt () represents the net load active power and reactive power injected at t node j, wherein distributed power source generated output can regard negative bearing power as, namely
p j ( t ) = p j D ( t ) - p j G ( t ) - p j E ( t ) q j ( t ) = q j D ( t ) - q j C ( t ) - q j E ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 9 )
In formula, p jD(t) and q jDt () is illustrated respectively in the meritorious and reactive power that t node j place AC load consumes; p jGt () represents the active power exported at t node j place distributed power source; p jE(t) and q jEt () represents the meritorious and reactive power exported at t node j place diesel engine; q jCt () represents the reactive power exported at t node j place reactive power compensator.
Described Branch Power Flow second order cone inequality, as follows:
( P i j ( t ) ) 2 + ( Q i j ( t ) ) 2 + ( l i j ( t ) ) 2 ≤ w i j ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 10 )
2.2) the DC microgrid described in is linearly bored trend constraint and is comprised two parts: direct current branch trend linear equality constraints and the inequality constraints of trend second order cone.Described direct current branch trend linear equality constraints is expressed as:
P i j ( t ) - Σ k : ( j , k ) ∈ Ψ d P j k ( t ) - R i j ( w i j ( t ) - l i j ( t ) ) = p j ( t ) , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 11 )
w i j ( t ) + l i j ( t ) - v j ( t ) - 2 R i j P i j ( t ) + R i j 2 ( w i j ( t ) - l i j ( t ) ) = 0 , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 12 )
w i j ( t ) + l i j ( t ) = v i ( t ) , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 13 )
In formula, Ψ drepresent the branch road collection of DC microgrid in alternating current-direct current mixing micro-capacitance sensor; v iand v jrepresent represent respectively the direct voltage of branch road (i, j) node i and j square, i.e. v i=| V i(t) | 2and v j=| V j(t) | 2; P ijt () is illustrated respectively in t flows through this branch road active power at branch road (i, j) parent node i end; p jt () represents the net load active power injected at t node j, wherein distributed power source generated output can regard negative bearing power as, namely
p j ( t ) = p j D ( t ) - p j G ( t ) - p j B ( t ) , ∀ ( i , j ) ∈ Ψ b , t ∈ T - - - ( 14 )
In formula, p jDt () represents the active power consumed in t node j place DC load; p jGt active power that () injects at t node j place distributed power source; p jBt () represents the active power provided as controllable burden at t node j place distributed energy storage.
Described Branch Power Flow second order cone inequality, as follows:
( P i j ( t ) ) 2 + ( l i j ( t ) ) 2 ≤ w i j ( t ) , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 15 )
2.3) the electric power netting safe running voltage level restraint described in, is expressed as:
| V a min | 2 ≤ v j ( t ) ≤ | V a m a x | 2 , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 16 )
| V d m i n | 2 ≤ v j ( t ) ≤ | V d m a x | 2 , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 17 )
In formula, V aminand V amaxbe respectively the bound of AC microgrid interior joint j voltage; V dminand V dmaxbe respectively the bound of DC microgrid interior joint j voltage.
2.4) the feeder current capacity-constrained described in, is expressed as:
w i j ( t ) - l i j ( t ) ≤ | I a m a x | 2 , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 18 )
w i j ( t ) - l i j ( t ) ≤ | I d m a x | 2 , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 19 )
In formula, I amaxand I dmaxrepresent that in AC microgrid and DC microgrid, branch road allows the upper current limit flow through respectively.
2.5) the energy storage device energy constraint described in, is expressed as:
E j min ≤ E j ( 0 ) + Σ τ = 1 t p j B ( τ ) · Δ t ≤ E j max - p j B m ≤ p j B ( t ) ≤ p j B m , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 20 )
In formula, E j(0) state-of-charge when expression energy storage device optimization cycle starts, E jminand E jmaxrepresent the bound of a jth distributed energy storage operationally its state-of-charge respectively; p jBmrepresent the maximum charge-discharge electric power that energy storage device allows; Δ t represents the time interval in t to t+1 moment.
2.6) the unit output constraint described in, is expressed as:
p j m i n ≤ p j E ( t ) ≤ p j m a x , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 21 )
- Δp j d ≤ p j E ( t + 1 ) - p j E ( t ) ≤ Δp j u , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 22 )
q j m i n ≤ q j C ( t ) ≤ q j m a x , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 23 )
In formula, p jminand p jmaxrepresent the minimum and maximum active power that node j place conventional diesel engine exports respectively, if node j place does not have diesel engine, make p jmin=p jmax=0; Δ p juwith Δ p jdrepresent the creep speed up and down of node j place diesel engine respectively; q jminand q jmaxrepresent that reactive power compensator minimum and maximum output in node j place is idle, if in like manner node j place does not have diesel engine, makes q respectively jmin=q jmax=0.
2.7) described in ACDC converter Branch Power Flow constraint, according to step 1) described in trend decoupling zero, be expressed as:
q i ( t ) = q d i ( t ) , ∀ ( i , j ) ∈ Ψ p , t ∈ T - - - ( 24 )
p i ( t ) = P j k ( t ) , ∀ ( i , j ) ∈ Ψ p , ( j , k ) ∈ Ψ d , t ∈ T - - - ( 25 )
In formula, Ψ prepresent the branch road collection connecting converter in alternating current-direct current mixing micro-capacitance sensor; p i(t) and q it () represents that the virtual injection of the AC microgrid i node of t connection converter branch road (i, j) is gained merit and reactive power respectively, P jkt () represents active power on t DC microgrid branch road (j, k) and j node connects converter; q di(t) represent t converter branch road (i, j) equivalence reactive power generation machine power output, according to step 1) described in converter control mode its value known constant.
Step 2) the alternating current-direct current mixing micro-capacitance sensor set up linearly bores optimal load flow model, when not only considering single, direction of energy constraint in discontinuity surface, unit output constraint, distributed energy storage charge-discharge electric power and state-of-charge run constraint, also contemplate energy storage charge state across time discontinuity surface between the continuity service requirement that changes.
Step 2) the alternating current-direct current mixing micro-capacitance sensor the set up optimized variable of linearly boring optimal load flow is y:=(y ij(t), (i, j) ∈ Ψ, t ∈ T), wherein Ψ=Ψ a× Ψ drepresent in alternating current-direct current mixing micro-capacitance sensor and exchange branch road and direct current branch collection; Branch road variable y ij(t) :=(P ij(t), Q ij(t), l ij(t), w ij(t), v j(t), p j(t), q j(t)), the feasible zone that alternating current-direct current mixing micro-capacitance sensor linearly bores optimal load flow is that D:={y|y meets constraint (5) ~ (25) } and feasible zone D is convex set.As shown in formula (10) and (15), optimized variable (P ij(t), Q ij(t), l ij(t), w ij(t)) constitute the second order cone space of four peacekeeping three-dimensionals respectively.For alternating current-direct current mixing micro-capacitance sensor, the optimal solution of this problem equation (P can be met ij(t)) 2+ (Q ij(t)) 2+ (l ij(t)) 2=(w ij(t)) 2(P ij(t)) 2+ (l ij(t)) 2=(w ij(t)) 2, it is the accurate convex lax of former non-linear trend that alternating current-direct current mixing micro-capacitance sensor linearly bores Branch Power Flow.
3) the distributed optimization method based on local heavily weighting augmentation Lagrangian is adopted, alternating current-direct current mixing micro-capacitance sensor is resolved into AC microgrid and DC microgrid subsystem, solve alternating current-direct current mixing micro-capacitance sensor and linearly bore optimal load flow, implement the distributed coordination optimization of the whole network energy scheduling.
If containing the individual DC microgrid of n (n>2) in alternating current-direct current mixing micro-capacitance sensor, definition E drepresent the DC bus set of node of converter branch road, E arepresent the ac bus set of node of converter branch road.Applying step 1) described in AC and DC trend decoupling method, this mixing micro-capacitance sensor is decoupled into n+1 subsystem, sets up a communication line between AC and the DC microgrid subsystem after decoupling zero.
Be defined as follows shown variable:
y a:=(P ij(t),Q ij(t),l ij(t),w ij(t),v j(t),p j(t),q j(t),j≠E a,(i,j)∈Ψ a,t∈T)
y d:=(P ij(t),l ij(t),w ij(t),v j(t),p j(t),i≠E d,(i,j)∈Ψ d,t∈T)
z a:=(p i(t),i=E a,t∈T)
z d:=(P ij(t),i=E d,(i,j)∈Ψ d,t∈T)
Then step 2) described in alternating current-direct current mixing micro-capacitance sensor linearly to bore optimal load flow model description be the decomposable form of alternating current-direct current, as follows:
minimizef 0a(y a)+f 0b(y d)(26a)
s.t.(y a,z a)∈D a;(y d,z d)∈D d(26b)
z a=z d.(26c)
D in formula a:={ (y a, z a) | (y a, z a) meet constraint (5) ~ (10), (16), (18), (21) ~ (23) and D d:={ (y d, z d) | meet constraint (11) ~ (15), (17), (19), (20) } represent the subset of former problem feasible zone D respectively and meet D=D a× D d; f 0a(y a) and f 0b(y d) represent the target function of intercommunion subsystem a and direct current subsystem d respectively, that is:
f 0 a ( y a ) = Σ t ∈ T Σ ( i , j ) ∈ Ψ a ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ a C j ( p j ( t ) ) - - - ( 27 a )
f 0 d ( y d ) = Σ t ∈ T Σ ( i , j ) ∈ Ψ d ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ d C j ( p j ( t ) ) - - - ( 27 b )
Visible, f 0=f 0a+ f 0b.
Step 3) described in local heavily weighting augmented lagrangian function, as follows:
L ρ ( y a , z a , y d , z d , λ ) = f 0 a ( y a ) + f 0 d ( y d ) + λ T ( z a - z d ) + ρ σ 2 | | z a - z d | | 2 2 - - - ( 28 )
Equality constraint (26c) dual variable in λ problem of representation (26) in formula, ρ >0 represents punishment parameter, and 1> σ >0 represents weighted factor.
Step 3) described in the distributed optimization method based on local heavily weighting augmentation Lagrangian, comprise following several step:
3.1) initialization: k ← 1 and virtual auxiliary variable initial value z 1, dual variable initial value λ 1.
3.2) variable of AC microgrid subsystem and DC microgrid subsystem is upgraded respectively as follows:
( y a k + 1 , z ^ a k ) = arg min ( y a , z a ) ∈ D a L ρ ( y a , z a , y d k , z d k , λ k ) - - - ( 29 a )
( y d k + 1 , z ^ d k ) = arg min ( y d , z d ) ∈ D d L ρ ( y a k , z a k , y d , z d , λ k ) - - - ( 29 b )
z a k + 1 = z a k + σ ( z ^ a k - z a k ) - - - ( 29 c )
z d k + 1 = z d k + σ ( z ^ d k - z d k ) - - - ( 29 d )
3.3) dual variable λ is upgraded k+1, as follows:
λ k + 1 = λ k + ρ σ ( z a k + 1 - z d k + 1 ) - - - ( 29 e )
3.4) residual error iteration is calculated, as follows:
Δτ 1 = m a x ( | | z ^ a k - z a k | | ∞ , | | z ^ d k - z d k | | ∞ ) , Δτ 2 = | | z a k - z d k | | ∞ - - - ( 29 f )
3.5) Δ τ is judged 1with Δ τ 2whether be less than ε (value of ε gets 0.00001) here, if Δ τ 1with Δ τ 2be not less than ε then make t ← t+1 and jump to step 3.2) continue repeated execution of steps 3.2) ~ step 3.5), until Δ τ 1with Δ τ 2all be less than ε, obtain the optimal solution of problem, thus realize the distributed coordination optimization of alternating current-direct current mixing micro-capacitance sensor scheduling.
In the method that the present invention proposes, the renewal optimization of subsystems and the renewal of dual variable can parallel work-flows, namely when each iteration, and the converter Power Flow Information z that AC microgrid subsystem obtains according to last iteration a kand z d k, solution formula (29a) and (29c) upgrade self optimal solution in like manner, DC microgrid subsystem is also according to the converter Power Flow Information z that last iteration obtains a kand z d k, solution formula (29b) and (29d) upgrade self optimal solution the renewal optimization of two subsystems can walk abreast and carry out, and only needs to transmit the renewal optimization that converter Power Flow Information can carry out next time after having optimized.As residual delta τ 1with Δ τ 2be less than ε, can obtain making the alternating current-direct current mixing microgrid energy scheduling of target function optimum to separate.
The invention has the beneficial effects as follows: the inventive method is set up the alternating current-direct current mixing micro-capacitance sensor comprising the constraint of converter trend and linearly bored optimal load flow model, utilize virtual auxiliary variable method that alternating current-direct current mixing micro-capacitance sensor is decomposed and exchange microgrid and direct-current micro-grid subsystem, propose the distributing coordination optimizing method based on the heavy weighting Lagrangian of local, implement the distributing coordination optimization of the whole network scheduling of resource.The method makes each subsystem can parallel synchronous iterative, and only need a small amount of information interaction without the need to global coordination between contiguous subsystem, the inventive method computational efficiency is higher and convergence property is good.
Accompanying drawing explanation
Fig. 1 is alternating current-direct current mixing micro-capacitance sensor.
Fig. 2 is the stable state physical model of ACDC converter.
Fig. 3 is the decoupling zero of converter Branch Power Flow.
Fig. 4 decomposes based on the alternating current-direct current subsystem of converter trend decoupling zero.
Fig. 5 is 9 node alternating current-direct current mixing micro-capacitance sensor.
Specific implementation method
In order to more clear and intuitive expression thinking of the present invention, alternating current-direct current mixing microgrid energy scheduling distributed coordination optimizing process is described in detail, for alternating current-direct current mixing micro-capacitance sensor as shown in Figure 5, this network comprises diesel engine, asynchronous blower fan, reactive power compensator, photovoltaic cell and storage battery etc., 0 node is the host node of micro-grid connection, the branch road collection Ψ of AC microgrid part a:={ (0,1), (1,2), (2,3), (3,4), (1,5) }, DC microgrid part Ψ b:={ (6,7), (7,8) }, branch road (1,6) is ACDC converter branch road.Maximum, minimum limit that diesel engine is exerted oneself are respectively: 6kW, 3kW; Upper and lower creep speed is respectively 10kW/min, 5kW/min; Maximum, the minimum limit of idle power output are respectively: 19.7kVar, 0.986kVar; Δ t=15min; Storage battery total capacity is 60kWh.
The enforcement of the energy scheduling distributed coordination optimization of alternating current-direct current mixing micro-grid system, comprises following steps:
1) utilize based on virtual auxiliary variable principle, the decoupling zero of alternating current-direct current trend is carried out to the branch road (1,6) connecting ACDC converter.
2) set up alternating current-direct current mixing micro-capacitance sensor and linearly bore optimal load flow model: with system losses and cost of electricity-generating for target function, constraints comprises: the trend constraint after the linear cone trend constraint constraint of 0 ~ 5 node AC microgrid, the linear cone trend constraint constraint of 6 ~ 7 node DC microgrids, the constraint of electric power netting safe running voltage level restraint, feeder current capacity-constrained, unit output, energy storage device energy constraint and ACDC converter branch road (1,6) decoupling zero.
3) the distributed optimization method based on local heavily weighting augmentation Lagrangian is adopted, alternating current-direct current mixing micro-capacitance sensor is resolved into AC microgrid and DC microgrid subsystem, solve alternating current-direct current mixing micro-capacitance sensor and linearly bore optimal load flow, implement the distributed coordination optimization of the whole network energy scheduling.
As step 1) as described in the decoupling zero of alternating current-direct current trend, suppose that the operation of ACDC converter itself is that complete equipilibrium is symmetrical and make R=0, namely do not consider the consume of inverter inside, then the power flow equation of ACDC converter branch road (1,6) is:
P s1=P d6(1)
Q s 1 = Q d 6 - ( P d 6 2 + Q d 6 2 ) X / V a b 2 - - - ( 2 )
In formula, P s1and Q s1represent the meritorious and reactive power transmitted between AC system and converter respectively; P d6and Q d6represent the meritorious and reactive power that converter exports respectively; P d6=V b6i b6, V b6and I b6represent direct voltage and the electric current at DC bus node 6 place respectively; V ab=kMV b6represent the relation of DC bus-bar voltage and inverter output voltage, for direct voltage utilance; M is the index of modulation.
Converter adopts determines direct voltage and the control mode exchanging reactive power, the trend decoupling zero of microgrid and direct-current micro-grid will be exchanged at converter place, after decoupling zero, node 1 place connecting converter in AC microgrid with the addition of virtual auxiliary active load and virtual auxiliary reactive power generation machine, when alternating current-direct current mixing micro-capacitance sensor does the whole network tidal current analysis, active power meets formula (1), the idle formula of output (2) of virtual auxiliary reactive power generation machine, connects the node 6 place voltage constant of converter in DC microgrid.
As step 2) as described in set up alternating current-direct current composite cone optimal load flow model; Step 2) described in alternating current-direct current composite cone optimal load flow with electric network active loss and operating cost minimum for target function, that is:
min i m i z e f 0 = Σ t ∈ T Σ ( i , j ) ∈ Ψ a × Ψ d ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ a × Ψ d C j ( p j ( t ) ) - - - ( 3 )
In formula, T represents the running optimizatin cycle; Ψ aand Ψ drepresent the branch road collection of AC microgrid and DC microgrid in alternating current-direct current mixing micro-capacitance sensor respectively; Ψ represents the branch road collection of power distribution network; (i, j) represents the branch road pointing to node j in power distribution network from node i (nearer compared to node j distance root node); C j() represents the generator cost function of node j; p jt () represents the active power that the generator of node j exports in t; γ represents weight factor; R ijrepresent the resistance of branch road (i, j); w ij(t) and l ijt () is respectively two branch road supplementary variables of definition, meet following equation:
w ij(t):=0.5*(|V i(t)| 2+|I ij(t)| 2),l ij(t):=0.5*(|V i(t)| 2-|I ij(t)| 2)(4)
In formula, I ijt () represents the electric current flow through on t branch road (i, j); V it () represents the voltage at t branch road (i, j) parent node i; Symbol || represent the operator solving variation amplitude.Therefore (w ij(t)-l ij(t)) R ij=| I ij(t) | 2r ij, be the active loss of branch road (i, j).
Establishment step 2) described in the constraints of alternating current-direct current composite cone optimal load flow: comprise that AC microgrid linearly bores trend constraint, DC microgrid linearly bores trend constraint, the constraint of electric power netting safe running voltage level restraint, feeder current capacity-constrained, unit output, trend after energy storage device energy constraint and the decoupling zero of ACDC converter retrain, specific as follows:
Set up 2.1) described in AC microgrid linearly bore trend constraint and comprise two parts: exchange Branch Power Flow linear equality constraints and the inequality constraints of trend second order cone.Described interchange Branch Power Flow linear equality constraints is expressed as:
P i j ( t ) - Σ k : ( j , k ) ∈ Ψ a P j k ( t ) - R i j ( w i j ( t ) - l i j ( t ) ) = p j ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 5 ) Q i j ( t ) - Σ k : ( j , k ) ∈ Ψ a Q j k ( t ) - X i j ( w i j ( t ) - l i j ( t ) ) = q j ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 6 )
w i j ( t ) + l i j ( t ) - v j ( t ) - 2 ( R i j P i j ( t ) + X i j Q i j ( t ) ) + ( R i j 2 + X i j 2 ) ( w i j ( t ) - l i j ( t ) ) = 0 , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 7 )
w i j ( t ) + l i j ( t ) = v i ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 8 )
In formula, X ijrepresent the reactance of branch road (i, j); v iand v jrepresent represent respectively the voltage magnitude of branch road (i, j) node i and j square, i.e. v i=| V i(t) | 2and v j=| V j(t) | 2; P ij(t) and Q ijt () is illustrated respectively in t flows through this branch road active power and reactive power at branch road (i, j) parent node i end; p j(t) and q jt () represents the net load active power and reactive power injected at t node j, wherein distributed power source generated output can regard negative bearing power as, namely
p j ( t ) = p j D ( t ) - p j G ( t ) - p j E ( t ) q j ( t ) = q j D ( t ) - q j C ( t ) - q j E ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 9 )
In formula, p jD(t) and q jDt () is illustrated respectively in the meritorious and reactive power that t node j place AC load consumes; p jGt () represents the active power exported at t node j place distributed power source; p jE(t) and q jEt () represents the meritorious and reactive power exported at t node j place diesel engine; q jCt () represents the reactive power exported at t node j place reactive power compensator.
Described Branch Power Flow second order cone inequality, as follows:
( P i j ( t ) ) 2 + ( Q i j ( t ) ) 2 + ( l i j ( t ) ) 2 ≤ w i j ( t ) , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 10 )
Set up 2.2) described in DC microgrid linearly bore trend constraint and comprise two parts: exchange Branch Power Flow linear equality constraints and the inequality constraints of trend second order cone.Described interchange Branch Power Flow linear equality constraints is expressed as:
P i j ( t ) - Σ k : ( j , k ) ∈ Ψ d P j k ( t ) - R i j ( w i j ( t ) - l i j ( t ) ) = p j ( t ) , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 11 )
w i j ( t ) + l i j ( t ) - v j ( t ) - 2 R i j P i j ( t ) + R i j 2 ( w i j ( t ) - l i j ( t ) ) = 0 , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 12 )
w i j ( t ) + l i j ( t ) = v i ( t ) , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 13 )
In formula, v iand v jrepresent represent respectively the direct voltage of branch road (i, j) node i and j square, i.e. v i=| V i(t) | 2and v j=| V j(t) | 2; P ijt () is illustrated respectively in t flows through this branch road active power at branch road (i, j) parent node i end; p jt () represents the net load active power injected at t node j, wherein distributed power source generated output can regard negative bearing power as, namely
p j ( t ) = p j D ( t ) - p j G ( t ) - p j B ( t ) , ∀ ( i , j ) ∈ Ψ b , t ∈ T - - - ( 14 )
In formula, p jDt () represents the active power consumed in t node j place DC load; p jGt active power that () injects at t node j place distributed power source; p jBt () represents the active power provided as controllable burden at t node j place distributed energy storage.
Described Branch Power Flow second order cone inequality, as follows:
( P i j ( t ) ) 2 + ( l i j ( t ) ) 2 ≤ w i j ( t ) , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 15 )
Set up 2.3) described in electric power netting safe running voltage level restraint, be expressed as:
| V a min | 2 ≤ v j ( t ) ≤ | V a m a x | 2 , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 16 )
| V d m i n | 2 ≤ v j ( t ) ≤ | V d m a x | 2 , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 17 )
In formula, V aminand V amaxbe respectively the bound of AC microgrid interior joint j voltage; V dminand V dmaxbe respectively the bound of DC microgrid interior joint j voltage.
Set up 2.4) described in feeder current capacity-constrained, be expressed as:
w i j ( t ) - l i j ( t ) ≤ | I a m a x | 2 , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 18 )
w i j ( t ) - l i j ( t ) ≤ | I d m a x | 2 , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 19 )
In formula, I amaxand I dmaxrepresent that in AC microgrid and DC microgrid, branch road allows the upper current limit flow through respectively.
Set up 2.5) described in energy storage device energy constraint, be expressed as:
E j min ≤ E j ( 0 ) + Σ τ = 1 t p j B ( τ ) · Δ t ≤ E j max - p j B m ≤ p j B ( t ) ≤ p j B m , ∀ ( i , j ) ∈ Ψ d , t ∈ T - - - ( 20 )
In formula, E j(0) state-of-charge when expression energy storage device optimization cycle starts, E jminand E jmaxrepresent the bound of a jth distributed energy storage operationally its state-of-charge respectively; p jBmrepresent the maximum charge-discharge electric power that energy storage device allows; Δ t represents the time interval in t to t+1 moment.
Set up 2.6) described in unit output constraint, be expressed as:
p j m i n ≤ p j E ( t ) ≤ p j m a x , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 21 )
- Δp j d ≤ p j E ( t + 1 ) - p j E ( t ) ≤ Δp j u , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 22 )
q j m i n ≤ q j C ( t ) ≤ q j m a x , ∀ ( i , j ) ∈ Ψ a , t ∈ T - - - ( 23 )
In formula, p jminand p jmaxrepresent the minimum and maximum active power that node j place conventional diesel engine exports respectively, if node j place does not have diesel engine, make p jmin=p jmax=0; Δ p juwith Δ p jdrepresent the creep speed up and down of node j place diesel engine respectively; q jminand q jmaxrepresent that reactive power compensator minimum and maximum output in node j place is idle, if in like manner node j place does not have diesel engine, makes q respectively jmin=q jmax=0.
Set up 2.7) described in ACDC converter trend constraint, according to step 1) described in trend decoupling zero, be expressed as:
q 1 ( t ) = q d ( t ) , ∀ t ∈ T - - - ( 24 )
p 1 ( t ) = P 67 ( t ) , ∀ t ∈ T - - - ( 25 )
In formula, q dt () represents the reactive power generation machine power output of t converter branch road (1,6) equivalence; p 1(t) and q 1t () represents the meritorious and reactive power that t AC microgrid 1 node injects respectively; P 67t () represents the active power on t DC microgrid branch road (6,7).
Step 2) the alternating current-direct current mixing micro-capacitance sensor set up linearly bores optimal load flow model, when not only considering single, direction of energy constraint in discontinuity surface, unit output constraint, distributed energy storage charge-discharge electric power and state-of-charge run constraint, also contemplate energy storage charge state across time discontinuity surface between the continuity service requirement that changes.
Step 2) the alternating current-direct current mixing micro-capacitance sensor the set up optimized variable of linearly boring optimal load flow is y:=(y ij(t), (i, j) ∈ Ψ, t ∈ T), wherein Ψ=Ψ a× Ψ drepresent in alternating current-direct current mixing micro-capacitance sensor and exchange branch road and direct current branch collection; Branch road variable y ij(t) :=(P ij(t), Q ij(t), l ij(t), w ij(t), v j(t), p j(t), q j(t)), the feasible zone that alternating current-direct current mixing micro-capacitance sensor linearly bores optimal load flow is that D:={y|y meets constraint (5) ~ (25) } and feasible zone D is convex set.As shown in formula (10) and (15), optimized variable (P ij(t), Q ij(t), l ij(t), w ij(t)) constitute the second order cone space of four peacekeeping three-dimensionals respectively.For alternating current-direct current mixing micro-capacitance sensor, the optimal solution of this problem equation (P can be met ij(t)) 2+ (Q ij(t)) 2+ (l ij(t)) 2=(w ij(t)) 2(P ij(t)) 2+ (l ij(t)) 2=(w ij(t)) 2, it is the accurate convex lax of former non-linear trend that alternating current-direct current mixing micro-capacitance sensor linearly bores Branch Power Flow.
Step 3) described in by the distributed optimization method based on local heavily weighting augmentation Lagrangian, alternating current-direct current mixing micro-capacitance sensor is resolved into AC microgrid and DC microgrid subsystem, solve alternating current-direct current mixing micro-capacitance sensor and linearly bore optimal load flow, implement the distributed coordination optimization of the whole network energy scheduling.
Be defined as follows shown variable:
y a:=(P ij(t),Q ij(t),l ij(t),w ij(t),v j(t),p j(t),q j(t),j≠1,(i,j)∈Ψ a,t∈T)
y d:=(P ij(t),l ij(t),w ij(t),v j(t),p j(t),i≠6,(i,j)∈Ψ d,t∈T)
z a:=(p 1(t),t∈T)
z d:=(P 67(t),t∈T)
Then step 2) described in alternating current-direct current mixing micro-capacitance sensor linearly to bore optimal load flow model description be the decomposable form of ac and dc systems, as follows:
minimizef 0a(y a)+f 0b(y d)(26a)
s.t.(y a,z a)∈D a;(y d,z d)∈D d(26b)
z a=z d.(26c)
D in formula a:={ (y a, z a) | (y a, z a) meet constraint (5) ~ (10), (16), (18), (21) ~ (23) and D d:={ (y d, z d) | meet constraint (11) ~ (15), (17), (19), (20) } represent the subset of former problem feasible zone D respectively and meet D=D a× D d; f 0a(y a) and f 0b(y d) represent the target function of intercommunion subsystem a and direct current subsystem d respectively, that is:
f 0 a ( y a ) = Σ t ∈ T Σ ( i , j ) ∈ Ψ a ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ a C j ( p j ( t ) ) - - - ( 27 a )
f 0 d ( y d ) = Σ t ∈ T Σ ( i , j ) ∈ Ψ d ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ d C j ( p j ( t ) ) - - - ( 27 b )
Visible, f 0=f 0a+ f 0b.
Establishment step 3) described in local heavily weighting augmented lagrangian function, as follows:
L ρ ( y a , z a , y d , z d , λ ) = f 0 a ( y a ) + f 0 d ( y d ) + λ T ( z a - z d ) + ρ σ 2 | | z a - z d | | 2 2 - - - ( 28 )
Equality constraint (26c) dual variable in λ problem of representation (26) in formula, ρ >0 represents punishment parameter, and 1> σ >0 represents weighted factor.
Step 3) described in the distributed optimization method based on local heavily weighting augmentation Lagrangian, comprise following several step:
3.1) initialization: k ← 1 and virtual auxiliary variable initial value z 1, dual variable initial value λ 1.
3.2) variable of AC microgrid subsystem and DC microgrid subsystem is upgraded respectively as follows:
( y a k + 1 , z ^ a k ) = arg min ( y a , z a ) ∈ D a L ρ ( y a , z a , y d k , z d k , λ k ) - - - ( 29 a )
( y d k + 1 , z ^ d k ) = arg min ( y d , z d ) ∈ D d L ρ ( y a k , z a k , y d , z d , λ k ) - - - ( 29 b )
z a k + 1 = z a k + σ ( z ^ a k - z a k ) - - - ( 29 c )
z d k + 1 = z d k + σ ( z ^ d k - z d k ) - - - ( 29 d )
3.3) dual variable λ is upgraded k+1, as follows:
λ k + 1 = λ k + ρ σ ( z a k + 1 - z d k + 1 ) - - - ( 29 e )
3.4) residual error iteration is calculated, as follows:
Δτ 1 = m a x ( | | z ^ a k - z a k | | ∞ , | | z ^ d k - z d k | | ∞ ) , Δτ 2 = | | z a k - z d k | | ∞ - - - ( 29 f )
3.5) Δ τ is judged 1with Δ τ 2whether be less than ε (value of ε gets 0.00001) here, if Δ τ 1with Δ τ 2be not less than ε then make t ← t+1 and jump to step 3.2) continue repeated execution of steps 3.2) ~ step 3.5), until Δ τ 1with Δ τ 2all be less than ε, obtain the optimal solution of problem, thus realize the distributed coordination optimization of alternating current-direct current mixing micro-capacitance sensor scheduling.
In the method that the present invention proposes, the renewal optimization of subsystems and the renewal of dual variable can parallel work-flows, namely when each iteration, and the converter Power Flow Information z that AC microgrid subsystem obtains according to last iteration a kand z d k, solution formula (28a) and (29c) upgrade self optimal solution in like manner, DC microgrid subsystem is also according to the converter Power Flow Information z that last iteration obtains a kand z d k, solution formula (29b) and (29d) upgrade self optimal solution the renewal optimization of two subsystems can walk abreast and carry out, and only needs to transmit the renewal optimization that converter Power Flow Information can carry out next time after having optimized.As residual delta τ 1with Δ τ 2be less than ε, can obtain making the alternating current-direct current mixing microgrid energy scheduling of target function optimum to separate.

Claims (5)

1. accelerate an alternating current-direct current micro-capacitance sensor distributed scheduling method of Lagrangian based on heavy weighting, it is characterized in that, the method comprises the following steps:
1) the trend decoupling zero of branch road and direct current branch is exchanged in alternating current-direct current mixing micro-capacitance sensor;
2) alternating current-direct current composite cone optimal load flow model is set up;
3) the distributed optimization method based on local heavily weighting augmentation Lagrangian is adopted, alternating current-direct current mixing micro-capacitance sensor is resolved into AC microgrid and DC microgrid subsystem, solve alternating current-direct current mixing micro-capacitance sensor and linearly bore optimal load flow, implement the distributed coordination optimization of the whole network energy scheduling.
2. a kind of alternating current-direct current micro-capacitance sensor distributed scheduling method accelerating Lagrangian based on heavy weighting according to claim 1, it is characterized in that, step 1) described in alternating current-direct current mixing micro-capacitance sensor primarily of interchange (AC) microgrid, direct current (DC) microgrid and ACDC converter composition; In alternating current-direct current mixing micro-capacitance sensor, in AC microgrid in certain ac bus node a1 and DC microgrid certain DC bus node b between connected by converter, suppose that the operation of ACDC converter itself is complete equipilibrium symmetry and makes R=0, namely the consume of inverter inside is not considered, then this converter branch road (b, a1) stable state physical model trend, as follows:
P s=P d(1)
Q s = Q d - ( P d 2 + Q d 2 ) X / V a b 2 - - - ( 2 )
In formula, V abrepresent inverter output voltage, R+jX represents converter equivalent reactance, P sand Q srepresent the meritorious and reactive power transmitted between AC system and converter respectively, P dand Q drepresent the meritorious and reactive power that converter exports respectively; P simultaneously d=V bi b, V brepresent DC bus-bar voltage, I brepresent DC bus current; V ab=kMV brepresent DC bus-bar voltage and inverter output voltage relation and for direct voltage utilance; M is the index of modulation;
Converter adopts determines direct voltage and the control mode exchanging reactive power, is AC microgrid and DC microgrid part at converter branch road (b, a1) place by mixing micro-capacitance sensor decomposition; After decoupling zero, the node a1 place connecting converter in AC microgrid with the addition of virtual auxiliary active load and virtual auxiliary reactive power generation machine, when alternating current-direct current mixing micro-capacitance sensor does the whole network tidal current analysis, active power meets formula (1), it is constant and meet formula (2) that the output of virtual auxiliary reactive power generation machine is idle, connects the node b place voltage constant of converter in DC microgrid;
According to the trend decoupling method of converter branch road, alternating current-direct current mixing micro-capacitance sensor, decoupling zero is AC microgrid is subsystem a, DC microgrid is subsystem b, AC microgrid is connected communication line with after the decoupling zero of DC microgrid, represent when subsystems microgrid inside is optimized scheduling, AC portion and direct current component need exchange message.
3. a kind of alternating current-direct current micro-capacitance sensor distributed scheduling method accelerating Lagrangian based on heavy weighting according to claim 1, it is characterized in that, step 2) described in alternating current-direct current composite cone optimal load flow with electric network active loss and operating cost minimum for target function, that is:
min i m i z e f 0 = Σ t ∈ T Σ ( i , j ) ∈ Ψ a × Ψ d ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ a × Ψ d C j ( p j ( t ) ) - - - ( 3 )
In formula, T represents the running optimizatin cycle; Ψ aand Ψ drepresent the branch road collection of AC microgrid and DC microgrid in alternating current-direct current mixing micro-capacitance sensor respectively; (i, j) represents the branch road pointing to node j in power distribution network from node i (nearer compared to node j distance root node); C j() represents the power supply operating cost function of node j; p jt () represents the active power that node j injects in t; γ represents weight factor; R ijrepresent the resistance of branch road (i, j); w ij(t) and l ijt () is respectively two branch road supplementary variables of definition, meet following equation:
w ij(t):=0.5*(|V i(t)| 2+|I ij(t)| 2),l ij(t):=0.5*(|V i(t)| 2-|I ij(t)| 2)(4)
In formula, I ijt () represents the electric current flow through on t branch road (i, j); V it () represents the voltage at t branch road (i, j) parent node i; Symbol || represent the operator solving variation amplitude.Therefore (w ij(t)-l ij(t)) R ij=| I ij(t) | 2r ij, be the active loss of branch road (i, j).
Step 2) described in alternating current-direct current composite cone optimal load flow model comprises that AC microgrid linearly bores trend constraint, DC microgrid linearly bores trend constraint, the constraint of electric power netting safe running voltage level restraint, feeder current capacity-constrained, unit output, energy storage device energy constraint and ACDC converter Branch Power Flow retrain, specific as follows:
2.1) the AC microgrid described in is linearly bored trend constraint and is comprised two parts: exchange Branch Power Flow linear equality constraints and the inequality constraints of trend second order cone.Described interchange Branch Power Flow linear equality constraints is expressed as:
In formula, Ψ arepresent the branch road collection of AC microgrid in alternating current-direct current mixing micro-capacitance sensor; X ijrepresent the reactance of branch road (i, j); v iand v jrepresent represent respectively the voltage magnitude of branch road (i, j) node i and j square, i.e. v i=| V i(t) | 2and v j=| V j(t) | 2; P ij(t) and Q ijt () is illustrated respectively in t flows through this branch road active power and reactive power at branch road (i, j) parent node i end; p j(t) and q jt () represents the net load active power and reactive power injected at t node j, wherein distributed power source generated output can regard negative bearing power as, namely
In formula, p jD(t) and q jDt () is illustrated respectively in the meritorious and reactive power that t node j place AC load consumes; p jGt () represents the active power exported at t node j place distributed power source; p jE(t) and q jEt () represents the meritorious and reactive power exported at t node j place diesel engine; q jCt () represents the reactive power exported at t node j place reactive power compensator.
Described Branch Power Flow second order cone inequality, as follows:
2.2) the DC microgrid described in is linearly bored trend constraint and is comprised two parts: direct current branch trend linear equality constraints and the inequality constraints of trend second order cone.Described direct current branch trend linear equality constraints is expressed as:
In formula, Ψ drepresent the branch road collection of DC microgrid in alternating current-direct current mixing micro-capacitance sensor; v iand v jrepresent represent respectively the direct voltage of branch road (i, j) node i and j square, i.e. v i=| V i(t) | 2and v j=| V j(t) | 2; P ijt () is illustrated respectively in t flows through this branch road active power at branch road (i, j) parent node i end; p jt () represents the net load active power injected at t node j, wherein distributed power source generated output can regard negative bearing power as, namely
In formula, p jDt () represents the active power consumed in t node j place DC load; p jGt active power that () injects at t node j place distributed power source; p jBt () represents the active power provided as controllable burden at t node j place distributed energy storage.
Described Branch Power Flow second order cone inequality, as follows:
2.3) the electric power netting safe running voltage level restraint described in, is expressed as:
In formula, V aminand V amaxbe respectively the bound of AC microgrid interior joint j voltage; V dminand V dmaxbe respectively the bound of DC microgrid interior joint j voltage.
2.4) the feeder current capacity-constrained described in, is expressed as:
In formula, I amaxand I dmaxrepresent that in AC microgrid and DC microgrid, branch road allows the upper current limit flow through respectively.
2.5) the energy storage device energy constraint described in, is expressed as:
In formula, E j(0) state-of-charge when expression energy storage device optimization cycle starts, E jminand E jmaxrepresent the bound of a jth distributed energy storage operationally its state-of-charge respectively; p jBmrepresent the maximum charge-discharge electric power that energy storage device allows; Δ t represents the time interval in t to t+1 moment.
2.6) the unit output constraint described in, is expressed as:
In formula, p jminand p jmaxrepresent the minimum and maximum active power that node j place conventional diesel engine exports respectively, if node j place does not have diesel engine, make p jmin=p jmax=0; Δ p juwith Δ p jdrepresent the creep speed up and down of node j place diesel engine respectively; q jminand q jmaxrepresent that reactive power compensator minimum and maximum output in node j place is idle, if in like manner node j place does not have diesel engine, makes q respectively jmin=q jmax=0.
2.7) described in ACDC converter Branch Power Flow constraint, according to step 1) described in trend decoupling zero, be expressed as:
In formula, Ψ prepresent the branch road collection connecting converter in alternating current-direct current mixing micro-capacitance sensor; p i(t) and q it () represents that the virtual injection of the AC microgrid i node of t connection converter branch road (i, j) is gained merit and reactive power respectively, P jkt () represents active power on t DC microgrid branch road (j, k) and j node connects converter; q di(t) represent t converter branch road (i, j) equivalence reactive power generation machine power output, according to step 1) described in converter control mode its value known constant.
Step 2) the alternating current-direct current mixing micro-capacitance sensor the set up optimized variable of linearly boring optimal load flow is y:=(y ij(t), (i, j) ∈ Ψ, t ∈ T), wherein Ψ=Ψ a× Ψ drepresent in alternating current-direct current mixing micro-capacitance sensor and exchange branch road and direct current branch collection; Branch road variable y ij(t) :=(P ij(t), Q ij(t), l ij(t), w ij(t), v j(t), p j(t), q j(t)), the feasible zone that alternating current-direct current mixing micro-capacitance sensor linearly bores optimal load flow is that D:={y|y meets constraint (5) ~ (25) } and feasible zone D is convex set.As shown in formula (10) and (15), optimized variable (P ij(t), Q ij(t), l ij(t), w ij(t)) constitute the second order cone space of four peacekeeping three-dimensionals respectively.For alternating current-direct current mixing micro-capacitance sensor, the optimal solution of this problem equation (P can be met ij(t)) 2+ (Q ij(t)) 2+ (l ij(t)) 2=(w ij(t)) 2(P ij(t)) 2+ (l ij(t)) 2=(w ij(t)) 2, it is the accurate convex lax of former non-linear trend that alternating current-direct current mixing micro-capacitance sensor linearly bores Branch Power Flow.
4. a kind of alternating current-direct current micro-capacitance sensor distributed scheduling method accelerating Lagrangian based on heavy weighting according to claim 1, it is characterized in that, step 3) in, if containing the individual DC microgrid of n (n>2) in alternating current-direct current mixing micro-capacitance sensor, definition E drepresent the DC bus set of node of converter branch road, E arepresent the ac bus set of node of converter branch road.Applying step 1) described in AC and DC trend decoupling method, this mixing micro-capacitance sensor is decoupled into n+1 subsystem, sets up a communication line between AC and the DC microgrid subsystem after decoupling zero.
Be defined as follows shown variable:
y a:=(P ij(t),Q ij(t),l ij(t),w ij(t),v j(t),p j(t),q j(t),j≠E a,(i,j)∈Ψ a,t∈T)
y d:=(P ij(t),l ij(t),w ij(t),v j(t),p j(t),i≠E d,(i,j)∈Ψ d,t∈T)
z a:=(p i(t),i=E a,t∈T)
z d:=(P ij(t),i=E d,(i,j)∈Ψ d,t∈T)
Then step 2) described in alternating current-direct current mixing micro-capacitance sensor linearly to bore optimal load flow model description be the decomposable form of alternating current-direct current, as follows:
minimizef 0a(y a)+f 0b(y d)(26a)
s.t.(y a,z a)∈D a;(y d,z d)∈D d(26b)
z a=z d.(26c)
D in formula a:={ (y a, z a) | (y a, z a) meet constraint (5) ~ (10), (16), (18), (21) ~ (23) and D d:={ (y d, z d) | meet constraint (11) ~ (15), (17), (19), (20) } represent the subset of former problem feasible zone D respectively and meet D=D a× D d; f 0a(y a) and f 0b(y d) represent the target function of intercommunion subsystem a and direct current subsystem d respectively, that is:
f 0 a ( y a ) = Σ t ∈ T Σ ( i , j ) ∈ Ψ a ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ a C j ( p j ( t ) ) - - - ( 27 a )
f 0 d ( y d ) = Σ t ∈ T Σ ( i , j ) ∈ Ψ d ( w i j ( t ) - l i j ( t ) ) R i j + γ Σ t ∈ T Σ ( i , j ) ∈ Ψ d C j ( p j ( t ) ) - - - ( 27 b )
Visible, f 0=f 0a+ f 0b.
Step 3) described in local heavily weighting augmented lagrangian function, as follows:
L ρ ( y a , z a , y d , z d , λ ) = f 0 a ( y a ) + f 0 d ( y d ) + λ T ( z a - z d ) + ρ σ 2 || z a - z d || 2 2 - - - ( 28 )
Equality constraint (26c) dual variable in λ problem of representation (26) in formula, ρ >0 represents punishment parameter, and 1> σ >0 represents weighted factor.
Step 3) described in the distributed optimization method based on local heavily weighting augmentation Lagrangian, comprise following several step:
3.1) initialization: k ← 1 and virtual auxiliary variable initial value z 1, dual variable initial value λ 1.
3.2) variable of AC microgrid subsystem and DC microgrid subsystem is upgraded respectively as follows:
( y a k + 1 , z ^ a k ) = arg m i n ( y a , z a ) ∈ D a L ρ ( y a , z a , y d k , z d k , λ k ) - - - ( 29 a )
( y d k + 1 , z ^ d k ) = arg m i n ( y d , z d ) ∈ D d L ρ ( y a k , z a k , y d , z d , λ k ) - - - ( 29 b )
z a k + 1 = z a k + σ ( z ^ a k - z a k ) - - - ( 29 c )
z d k + 1 = z d k + σ ( z ^ d k - z d k ) - - - ( 29 d )
3.3) dual variable λ is upgraded k+1, as follows:
λ k + 1 = λ k + ρ σ ( z a k + 1 - z d k + 1 ) - - - ( 29 e )
3.4) residual error iteration is calculated, as follows:
Δτ 1 = m a x ( || z ^ a k - z a k || ∞ , || z ^ d k - z d k || ∞ ) , Δτ 2 = || z a k - z d k || ∞ - - - ( 29 f )
3.5) Δ τ is judged 1with Δ τ 2whether be less than ε (value of ε gets 0.00001) here, if Δ τ 1with Δ τ 2be not less than ε then make t ← t+1 and jump to step 3.2) continue repeated execution of steps 3.2) ~ step 3.5), until Δ τ 1with Δ τ 2all be less than ε, obtain the optimal solution of problem, thus realize the distributed coordination optimization of alternating current-direct current mixing micro-capacitance sensor scheduling.
5. a kind of alternating current-direct current micro-capacitance sensor distributed scheduling method accelerating Lagrangian based on heavy weighting according to claim 4, it is characterized in that, the renewal optimization of subsystems and the renewal of dual variable can parallel work-flows, namely when each iteration, the converter Power Flow Information z that AC microgrid subsystem obtains according to last iteration a kand z d k, solution formula (29a) and (29c) upgrade self optimal solution in like manner, DC microgrid subsystem is also according to the converter Power Flow Information z that last iteration obtains a kand z d k, solution formula (29b) and (29d) upgrade self optimal solution the renewal optimization of two subsystems can walk abreast and carry out, and only needs to transmit the renewal optimization that converter Power Flow Information can carry out next time after having optimized.As residual delta τ 1with Δ τ 2be less than ε, can obtain making the alternating current-direct current mixing microgrid energy scheduling of target function optimum to separate.
CN201510595842.0A 2015-09-17 2015-09-17 A kind of alternating current-direct current micro-capacitance sensor distributed scheduling method based on weighting acceleration Lagrangian again Active CN105140971B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510595842.0A CN105140971B (en) 2015-09-17 2015-09-17 A kind of alternating current-direct current micro-capacitance sensor distributed scheduling method based on weighting acceleration Lagrangian again

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510595842.0A CN105140971B (en) 2015-09-17 2015-09-17 A kind of alternating current-direct current micro-capacitance sensor distributed scheduling method based on weighting acceleration Lagrangian again

Publications (2)

Publication Number Publication Date
CN105140971A true CN105140971A (en) 2015-12-09
CN105140971B CN105140971B (en) 2017-10-20

Family

ID=54726228

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510595842.0A Active CN105140971B (en) 2015-09-17 2015-09-17 A kind of alternating current-direct current micro-capacitance sensor distributed scheduling method based on weighting acceleration Lagrangian again

Country Status (1)

Country Link
CN (1) CN105140971B (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108471109A (en) * 2018-04-24 2018-08-31 清华大学 The univesral distribution formula control method and system of the more micro-grid systems of direct current
CN109980684A (en) * 2019-04-02 2019-07-05 云南电网有限责任公司电力科学研究院 A kind of distributed optimization dispatching method based on flexible interconnection micro-capacitance sensor
CN110690719A (en) * 2019-09-18 2020-01-14 国网重庆市电力公司电力科学研究院 Micro-grid battery energy storage configuration method and readable storage medium
CN111525626A (en) * 2020-04-13 2020-08-11 清华大学 Cloud edge cooperative computing-based micro-grid real-time scheduling method and system
CN112271744A (en) * 2020-11-06 2021-01-26 中国电力科学研究院有限公司 Planning configuration method, system and device for regional power grid multipoint layout energy storage system
CN112531686A (en) * 2020-11-24 2021-03-19 绍兴市上虞区舜兴电力有限公司 Micro-grid multilayer control system for optimal power dispatching and voltage control
CN113285485A (en) * 2021-07-23 2021-08-20 南京邮电大学 Power distribution network source network charge storage multi-terminal cooperative voltage regulation method under long, short and multi-time scales

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005261123A (en) * 2004-03-12 2005-09-22 Tokyo Gas Co Ltd Control method and control system of power system
CN104809327A (en) * 2014-09-02 2015-07-29 长沙理工大学 New energy contained electric power dispatching moment uncertainty distribution robustness optimization method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005261123A (en) * 2004-03-12 2005-09-22 Tokyo Gas Co Ltd Control method and control system of power system
CN104809327A (en) * 2014-09-02 2015-07-29 长沙理工大学 New energy contained electric power dispatching moment uncertainty distribution robustness optimization method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
何智文: "交直流混合电力系统无功优化模型及线性规划算法", 《中国优秀硕士学位论文全文数据库》 *
李静: "含间歇性能源的分布式电网优化配置理论与方法研究", 《中国博士学位论文全文数据库》 *

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108471109A (en) * 2018-04-24 2018-08-31 清华大学 The univesral distribution formula control method and system of the more micro-grid systems of direct current
CN108471109B (en) * 2018-04-24 2020-02-21 清华大学 Unified distributed control method and system for direct-current multi-microgrid system
CN109980684A (en) * 2019-04-02 2019-07-05 云南电网有限责任公司电力科学研究院 A kind of distributed optimization dispatching method based on flexible interconnection micro-capacitance sensor
CN110690719A (en) * 2019-09-18 2020-01-14 国网重庆市电力公司电力科学研究院 Micro-grid battery energy storage configuration method and readable storage medium
CN111525626A (en) * 2020-04-13 2020-08-11 清华大学 Cloud edge cooperative computing-based micro-grid real-time scheduling method and system
CN112271744A (en) * 2020-11-06 2021-01-26 中国电力科学研究院有限公司 Planning configuration method, system and device for regional power grid multipoint layout energy storage system
CN112271744B (en) * 2020-11-06 2022-12-20 中国电力科学研究院有限公司 Planning configuration method, system and device for regional power grid multipoint layout energy storage system
CN112531686A (en) * 2020-11-24 2021-03-19 绍兴市上虞区舜兴电力有限公司 Micro-grid multilayer control system for optimal power dispatching and voltage control
CN113285485A (en) * 2021-07-23 2021-08-20 南京邮电大学 Power distribution network source network charge storage multi-terminal cooperative voltage regulation method under long, short and multi-time scales
CN113285485B (en) * 2021-07-23 2021-10-08 南京邮电大学 Power distribution network source network charge storage multi-terminal cooperative voltage regulation method under long, short and multi-time scales
WO2023000624A1 (en) * 2021-07-23 2023-01-26 南京邮电大学 Power distribution network source-grid-load-storage multi-end cooperative voltage regulation method under long, short and multiple time scales

Also Published As

Publication number Publication date
CN105140971B (en) 2017-10-20

Similar Documents

Publication Publication Date Title
CN105140971A (en) AC-DC micro-grid distributed scheduling method based on reweighed acceleration Lagrangian
CN102522746B (en) VSC-HVDC AC/DC optimal power flow method based on primal-dual interior point algorithm
CN109004691B (en) Day-ahead optimal scheduling method for alternating current-direct current hybrid system containing power electronic transformer
CN106541845B (en) A kind of wireless electric automobile charging system actual and control method
CN205509635U (en) Distributing type new forms of energy fill electric pile and hydrogenation station
CN105576803A (en) Distributed new energy charging pile and hydrogen refueling station
CN102545207B (en) Voltage source commutation-high voltage direct current (VSC-HVDC) alternating-direct current optimal power flow method based on predictor-corrector inner point method
CN103236718A (en) Source-network-load automatic control system and method for intelligent microgrid
CN105244869A (en) Dynamic random scheduling control method for power distribution network containing micro-grid
CN104537258A (en) Cone optimization modeling method for allowing distributed stored energy to participate in running adjustment of active power distribution network
CN104979849A (en) Grid-connected operation control method for user side microgrid
CN105186500A (en) Power distribution network energy dispersion coordination and optimization method based on reweighted acceleration Lagrangian
CN105490312A (en) Multi-source reactive power optimization control method for power system
Li et al. Power quality control in wind/fuel cell/battery/hydrogen electrolyzer hybrid micro-grid power system
Li et al. Study on wind-EV complementation on the transmission grid side considering carbon emission
Yang et al. A dual mode distributed economic control for a fuel cell–photovoltaic-battery hybrid power generation system based on marginal cost
CN201750160U (en) Solar energy power supply grid-connected integrated device
Vigneysh et al. Operation and control of wind/fuel cell based hybrid microgrid in grid connected mode
CN113822480A (en) Multi-layer collaborative optimization method and system for rural comprehensive energy system
Sasikumar et al. Design and development of solar charging system for electric vehicles: an initiative to achieve green campus
CN202513586U (en) Intelligent micro-grid using renewable energy for generating power
Aljarhizi et al. Control Management System of a Lithium-ion Battery Charger Based MPPT algorithm and Voltage Control
Ren et al. Modeling and optimization of hybrid wind/PV pumped-storage power system
Tingji et al. A multiport energy hub for the connection of multi power flow: A simulating case
Nanda et al. Performance of battery bank on hybrid microgrid

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
CB03 Change of inventor or designer information
CB03 Change of inventor or designer information

Inventor after: Li Jing

Inventor before: Li Jing

Inventor before: Dai Wenzhan

Inventor before: Zhao Zhongwei

Inventor before: Shen Chen

GR01 Patent grant
GR01 Patent grant