CN104319780B - A kind of transmission and distribution network Global optimization method - Google Patents

Abstract

The present invention relates to a kind of transmission and distribution network Global optimization method, step is：Selected power transmission network and active distribution network orchestration boundary point；The power transmission network reactive Voltage Optimum model containing Constraints is built, and is solved using interior point method；It is three-phase voltage of the three-phase voltage as active distribution network root node using power transmission network optimization photovoltaic conversion；Three-phase reactive Voltage Optimum model of the active distribution network containing Constraints is built, and is solved using interior point method；As the corresponding load power of power transmission network after active distribution network three-phase optimization power is sued for peace；Power transmission network and active distribution network reactive Voltage Optimum alternating iteration, when the adjacent iteration twice change of transmission and distribution network boundary node power meets convergence precision, the optimization convergence of transmission and distribution network Global obtains Global optimisation strategy.The present invention solves transmission and distribution network reactive Voltage Optimum result in the inconsistent of transmission and distribution network boundary node, eliminates the voltage and power mismatch in boundary node, optimizes the idle resource of two-level network.

Description

A kind of transmission and distribution network Global optimization method

Technical field

The present invention relates to a kind of power system reactive power voltage optimization method, in particular to a kind of transmission and distribution network overall situation is idle Voltage optimization method.

Background technology

Reactive power optimization of power system is that on the premise of power network safety operation is ensured, with generator reactive, have load to adjust Pressure transformer tap (OLTC), can switching reactive compensator to exert oneself etc. be control device, realize reactive layered partition balancing, Grid voltage quality is improved, a kind of optimization method of network loss is reduced.

The reactive Voltage Optimum of power transmission network and active distribution network is each independently carried out mostly at present, is carried out at boundary node Equivalence treatment, can cause the power and voltage mismatch at boundary node." hair transmission ＆ distribution overall situation Load flow calculation " (Sun Hongbin, Zhang Baiming, Phase year moral《Hair transmission ＆ distribution overall situation Load flow calculation》1998,22 (12)：39-42 pages) power transmission network and power distribution network are summed up in network structure With the difference of parameter, trend size, model etc.." a kind of area power grid new distributed optimal reactive power " (Li Zhongxu, Liu Yutian 《Power System and its Automation journal》, 2005,17 (2)：80-83 pages) sum up power transmission network and power distribution network on idle work optimization Difference, due to transmission and distribution network network structure and the difference of idle work optimization feature, global nothing should not be built using unified method Work(voltage optimization model, but the document does not disclose the access situation of triphase flow and distributed power source in power distribution network aspect.

As distributed power source accesses power distribution network, active distribution network considers the nothing of the distributed power source for having reactive power compensation planning Work(power is combined with traditional voltage-regulation means, realizes the reactive Voltage Optimum of active distribution network, and optimum results include defeated Power distribution network boundary node data, can cause the change of transmission and distribution network boundary node voltage and power, without coordinate power transmission network and In the case of active distribution network, the result of power transmission network and active distribution network reactive Voltage Optimum may inconsistent or even trend phase Instead.In addition, the access of distributed power source can increase the three-phase imbalance of power distribution network, so need to be calculated using triphase flow dividing Analysis.

In addition, containing on-load transformer tap changer in the reactive Voltage Optimum of power transmission network and active distribution network, can throwing Cut reactive power compensator the control device such as to exert oneself, be discrete variable in Mathematical Modeling, power transmission network and active distribution network it is idle Voltage optimization is mixed-integer nonlinear programming model, and this is accomplished by considering the method that can process discrete variable solving.

Therefore power system reactive power voltage optimization needs the transmission and distribution network Global optimization from from the point of view of the overall situation, To improve computational accuracy, but world model's Unified Solution can not be built, so to be based on the thought of layering and zoning, be transmitted electricity The composition decomposition of net and active distribution network reactive Voltage Optimum, the method that use can process discrete variable solves power transmission network respectively With active distribution network reactive Voltage Optimum model, interaction is coordinated to the reactive Voltage Optimum of power transmission network and active distribution network and is come Reach the target of global optimization；And three-phase imbalance and distributed power source from the aspect of active distribution network reactive Voltage Optimum Reactive-power control, set up three-phase Optimized model；Power transmission network optimum results and active distribution are realized at transmission and distribution network boundary node The data conversion of net three-phase optimum results with interact.

The content of the invention

In view of the shortcomings of the prior art, it is an object of the invention to provide a kind of transmission and distribution network Global optimization side Method, the method is intended to coordinate power transmission network and active distribution network reactive Voltage Optimum, solves the border caused during the two independent calculating The voltage and power mismatch of node, solve transmission and distribution network reactive Voltage Optimum result and differ in the possibility of transmission and distribution network boundary node Cause, while optimizing the idle resource of two-level network, from the angle of global united analysis, use can process discrete variable Complementary Interior-point method solves power transmission network and active distribution network reactive Voltage Optimum model respectively, to power transmission network and active distribution network Reactive Voltage Optimum coordinates interaction to reach the target of global optimization, and realizes power transmission network at transmission and distribution network boundary node The data conversion of optimum results and active distribution network three-phase optimum results with interact.

The purpose of the present invention is realized using following technical proposals：

The present invention provides a kind of transmission and distribution network Global optimization method, and it is theed improvement is that, methods described bag Include following step：

(1) power transmission network and active distribution network orchestration boundary point are selected；

(2) the power transmission network reactive Voltage Optimum model containing Constraints condition is built, using Non-Linear Programming interior point method pair It is solved；

(3) it is three-phase voltage as active distribution using the optimization photovoltaic conversion of each transmission and distribution network boundary node in power transmission network Net the three-phase voltage of corresponding root node；

(4) the three-phase reactive Voltage Optimum mould containing Constraints condition of the active distribution network containing distributed power source is built Type, is solved using Non-Linear Programming interior point method to it；

(5) as power transmission network phase after the three-phase optimization power of each transmission and distribution network boundary node in active distribution network is sued for peace Answer the load power of node；

(6) judge whether adjacent change twice meets convergence essence in an iterative process for the power of transmission and distribution network boundary node Degree, (7) are gone to if meeting, and (2) are gone to if being unsatisfactory for and continues iteration；

(7) convergence according to power transmission network and active distribution network reactive Voltage Optimum at transmission and distribution network boundary node, obtains Global optimisation strategy.

Further, in the step (1), power transmission network and active distribution network orchestration boundary node are selected, to active distribution Net three-phase is modeled, and builds power transmission network and the single-phase coordination interactive frame with three reactive Voltage Optimums for mixing of active distribution network, And single-phase and three reactive Voltage Optimums for mixing coordination interactive modes.

Further, in the step (2), with loss minimization as target, node voltage etc. for constraint set up power transmission network without Work(voltage optimization model, use can process the method based on complementary theory and Non-Linear Programming interior point method of discrete variable to it Solved；Power transmission network reactive Voltage Optimum model is as follows：

Min f (x)=P_l

V_i ^min≤V_i≤V_i ^max1)；

Q_gi ^min≤Q_gi≤Q_gi ^max

Q_ci ^min≤Q_ci≤Q_ci ^max

T_ki ^min≤T_ki≤T_ki ^max

Wherein：P_lIt is the network loss of transmission system；N is transmission system nodes；P_iAnd Q_iThe injection of respectively node i is active Power and reactive power；P_giAnd P_liThe respectively generator active power of node i and load active power；Q_giAnd Q_liRespectively The generator reactive power and reactive load power of node i；G_ijAnd B_ijConductance and electricity respectively between node i and node j Receive；θ_ijIt is phase difference of voltage between node i and node j；V_i、V_i ^minAnd V_i ^maxRespectively the voltage magnitude of node i, voltage are minimum Value and maximum；Q_gi、Q_gi ^minAnd Q_gi ^maxGenerator respectively at node i it is idle exert oneself, idle minimum value and the maximum of exerting oneself Value；Q_ci、Q_ci ^minAnd Q_ci ^maxReactive-load compensation equipment respectively at node i it is idle exert oneself, idle minimum value and the maximum of exerting oneself Value；T_ki、T_ki ^minAnd T_ki ^maxULTC no-load voltage ratio, no-load voltage ratio minimum value and maximum respectively at node i；

The Mathematical Programs With Nonlinear Complementarity Constraints condition for building discrete variable in power transmission network reactive Voltage Optimum is as follows：

Wherein：f_iIt is that i-th ULTC gear or reactive power compensator switching group number are processed by continuous variable The first optimal value for obtaining；f_i(n+1)And f_inRespectively f_iThe actual-gear or switching group number of left and right；f_1xAnd f_2xIt is intermediate variable； μ is smoothing parameter；

Formula 1), 2), 3) constitute containing Constraints condition power transmission network reactive Voltage Optimum model, using in Non-Linear Programming Point method is solved.

Further, in the step (3), during power transmission network and active distribution network reactive Voltage Optimum alternating iteration, It is three-phase voltage as the corresponding root section of active distribution network using the optimization photovoltaic conversion of each transmission and distribution network boundary node in power transmission network The three-phase voltage of point；

It is in the steps below three-phase voltage by the optimization photovoltaic conversion of power transmission network boundary node：By active distribution network root node Voltage is set as three-phase equilibrium, by the single-phase optimization photovoltaic conversion of power transmission network orchestration boundary node for three-phase phase voltage is tried to achieve.

Further, in the step (4), set up containing distribution for constraint with node voltage etc. with loss minimization as target The active distribution network three-phase reactive Voltage Optimum model of power supply, using treatment discrete variable based on complementary theory and non-linear rule Interior point method is drawn to solve；Three-phase reactive Voltage Optimum model containing distributed power source is as follows：

V_i,p ^min≤V_i,p≤V_i,p ^max

Q_idg,p ^min≤Q_idg,p≤Q_idg,p ^max4)；

Q_icap,p ^min≤Q_icap,p≤Q_icap,p ^max

T_ik,p ^min≤T_ik,p≤T_ik,p ^max

Wherein：f_pX () is the network loss of active distribution network system p phases, p ∈ (a, b, c)；F (x) is active distribution network system Network loss；X=[x₁ ^T,x₂ ^T]^T, wherein x₁For distributed power source it is idle exert oneself, reactive power compensator it is idle exert oneself and have load adjust The no-load voltage ratio column vector of pressure transformer, x₂It is the real part and imaginary part column vector of node phase voltage；P_i ^pAnd Q_i ^pRespectively p phases of node i Injection active power and reactive power；P_gi ^pAnd P_li ^pThe respectively generator active power and load wattful power of the p phases of node i Rate, Q_gi ^pAnd Q_li ^pThe respectively generator reactive power and reactive load power of the p phases of node i；V_ri ^pAnd V_mi ^pRespectively node i P phase voltages real part and imaginary part；G_ij ^ptAnd B_ij ^ptIt is the node admittance element between node i and node j, t ∈ (a, b, c)； V_i,p、V_i,p ^minAnd V_i,p ^maxRespectively the p phase voltages of node i, p phase voltages minimum value and maximum；Q_idg,p、Q_idg,p ^minWith Q_idg,p ^maxRespectively i-th p phase of distributed power source it is idle exert oneself, p phases are idle exerts oneself minimum value and maximum；Q_icap,p、 Q_icap,p ^minAnd Q_icap,p ^maxThe p phases of respectively i-th reactive-load compensation equipment are idle to exert oneself, p phases are idle exerts oneself minimum value and maximum Value；T_ik,p、T_ik,p ^minAnd T_ik,p ^maxRespectively i-th the p phases no-load voltage ratio of ULTC, p phase no-load voltage ratio minimum values and maximum Value；

The Mathematical Programs With Nonlinear Complementarity Constraints condition for building discrete variable in active distribution network three-phase reactive Voltage Optimum is as follows：

Wherein：f_i ^pIt is that i-th ULTC p phases gear or reactive power compensator p phase switching groups number are become as continuous The first optimal value that amount treatment is obtained；f_i(n+1) ^pAnd f_in ^pRespectively f_i ^pThe actual-gear or switching group number of left and right；f_1xAnd f_2xFor Intermediate variable；μ is smoothing parameter；

Formula 4), 5), 6) constitute containing Constraints condition active distribution network three-phase reactive Voltage Optimum model, using non-thread Property planning interior point method solved.

Further, the transmission of electricity comprising transmission and distribution network orchestration boundary node is obtained by the step (2) and step (4) Net optimum results and active distribution network three-phase optimum results.

Further, in the step (5), during power transmission network and active distribution network reactive Voltage Optimum alternating iteration, Using bearing as power transmission network respective nodes after the three-phase optimization power summation of each transmission and distribution network boundary node in active distribution network Lotus power.

Further, in the step (6), judge the power of transmission and distribution network boundary node in an iterative process it is adjacent twice Change whether meet convergence precision, if meet if go to (7), gone to if being unsatisfactory for (2) continue iteration；Convergence precision It is the convergence parameter of setting power.

Further, in the step (7), when power adjacent change twice in iteration of transmission and distribution network boundary node When meeting convergence precision, power transmission network and active distribution network reactive Voltage Optimum are final in transmission and distribution network boundary node everywhere convergent To Global optimisation strategy.

Compared with the prior art, the beneficial effect that reaches of the present invention is：

(1) angle from global united analysis of the invention, active distribution network uses triphase flow, it is considered to active distribution Net three-phase imbalance and distributed power source participate in reactive Voltage Optimum, and three-phase reactive Voltage Optimum is carried out in active distribution network aspect In the case of, coordinated based on transmission and distribution network boundary node and complementary Interior-point method is solved, realize transmission and distribution network Global Optimization, the voltage and power mismatch of the boundary node that solution power transmission network and active distribution network are caused when independently calculating, solves transmission ＆ distribution Electric network reactive-load voltage optimization may be inconsistent in the optimum results of transmission and distribution network boundary node, improves power system reactive power voltage The computational accuracy of optimization.

(2) present invention participates in the situation of reactive Voltage Optimum from the angle of global united analysis in distributed power source Under, the idle resource of optimization power transmission network and active distribution network two-stage, and the thought based on layering and zoning, by power transmission network and active The coordination interaction of electric distribution network reactive-voltage optimization, reaches the target of global optimization, realizes that the optimization of global idle resource is coordinated, drop Low via net loss, improves the quality of voltage of power transmission network and active distribution network.

Brief description of the drawings

Fig. 1 is the transmission and distribution network Global optimization method flow chart that the present invention is provided.

Specific embodiment

Specific embodiment of the invention is described in further detail below in conjunction with the accompanying drawings.

Transmission and distribution network Global optimization method mainly includes considering the modeling of active distribution network three-phase, structure power transmission network Single-phase with the active distribution network coordination interactive frame with three reactive Voltage Optimums for mixing, it is proposed that single-phase to be mixed with three The coordination interactive mode of reactive Voltage Optimum；Three-phase imbalance and distribution from the aspect of active distribution network reactive Voltage Optimum Power supply participates in reactive Voltage Optimum, with loss minimization as target, node voltage etc. be that the three-phase containing distributed power source is set up in constraint Reactive Voltage Optimum model, equally with loss minimization as target, node voltage etc. set up power transmission network reactive Voltage Optimum for constraint Model, power transmission network and active distribution network reactive Voltage Optimum be respectively adopted can process discrete variable based on complementary theory and interior The Algorithm for Solving of point method, obtains power transmission network optimum results and active distribution network three-phase comprising transmission and distribution network boundary node excellent Change result；Power transmission network and active distribution network reactive Voltage Optimum alternating iteration, by each transmission and distribution network boundary node in power transmission network Optimization photovoltaic conversion be three-phase voltage as the corresponding root node of active distribution network three-phase voltage；By in active distribution network each As the load power of power transmission network respective nodes after the three-phase optimization power summation of transmission and distribution network boundary node；In power transmission and distribution selvage Under the coordination of boundary's node, power transmission network and active distribution network reactive Voltage Optimum are final in transmission and distribution network boundary node everywhere convergent To Global optimisation strategy.Transmission and distribution network Global optimization method flow chart is as shown in figure 1, specifically include down State step：

(1) power transmission network and active distribution network orchestration boundary point are selected：Consider the modeling of active distribution network three-phase, build power transmission network Single-phase with the active distribution network coordination interactive frame with three reactive Voltage Optimums for mixing, it is proposed that single-phase to be mixed with three The coordination interactive mode of reactive Voltage Optimum.

(2) power transmission network reactive Voltage Optimum model is built, it is solved using Non-Linear Programming interior point method：

With loss minimization as target, node voltage etc. set up power transmission network reactive Voltage Optimum model for constraint, using can The method based on complementary theory and Non-Linear Programming interior point method for processing discrete variable is solved to it, power transmission network reactive voltage Optimized model is as follows：

Min f (x)=P_l

V_i ^min≤V_i≤V_i ^max1)；

Q_gi ^min≤Q_gi≤Q_gi ^max

Q_ci ^min≤Q_ci≤Q_ci ^max

T_ki ^min≤T_ki≤T_ki ^max

Wherein：P_lIt is the network loss of transmission system；N is transmission system nodes；P_iAnd Q_iThe injection of respectively node i is active Power and reactive power；P_giAnd P_liThe respectively generator active power of node i and load active power；Q_giAnd Q_liRespectively The generator reactive power and reactive load power of node i；G_ijAnd B_ijConductance and electricity respectively between node i and node j Receive；θ_ijIt is phase difference of voltage between node i and node j；V_i、V_i ^minAnd V_i ^maxRespectively the voltage magnitude of node i, voltage are minimum Value and maximum；Q_gi、Q_gi ^minAnd Q_gi ^maxGenerator respectively at node i it is idle exert oneself, idle minimum value and the maximum of exerting oneself Value；Q_ci、Q_ci ^minAnd Q_ci ^maxReactive-load compensation equipment respectively at node i it is idle exert oneself, idle minimum value and the maximum of exerting oneself Value；T_ki、T_ki ^minAnd T_ki ^maxULTC no-load voltage ratio, no-load voltage ratio minimum value and maximum respectively at node i.

In order to process the discrete variable in Mathematical Modeling, the accurate solving model of discrete variable is built based on complementary theory, Using document《Meter and discrete variable are based on the idle work optimization of the full Smoothing Newton Method of Constraints》(Lin Jikeng, Shi Weizhao, Wu Nai Tiger, waits to count and idle work optimization Proceedings of the CSEE of the discrete variable based on the full Smoothing Newton Method of Constraints, and 2012,32 (1)：The method of the construction discrete variable Mathematical Programs With Nonlinear Complementarity Constraints condition in 93-100), in structure power transmission network reactive Voltage Optimum The Mathematical Programs With Nonlinear Complementarity Constraints condition of discrete variable is as follows：

Wherein：f_iIt is that i-th ULTC gear or reactive power compensator switching group number are processed by continuous variable The first optimal value for obtaining；f_i(n+1)And f_inRespectively f_iThe actual-gear or switching group number of left and right；f_1xAnd f_2xIt is intermediate variable； μ is smoothing parameter.

Formula 1), 2), 3) constitute containing Constraints condition power transmission network reactive Voltage Optimum model, using in Non-Linear Programming Point method is solved.

(3) it is three-phase voltage as active distribution using the optimization photovoltaic conversion of each transmission and distribution network boundary node in power transmission network Net the three-phase voltage of corresponding root node.The optimization photovoltaic conversion of power transmission network boundary node is that the method for three-phase voltage is：Will be actively Power distribution network root node voltage is set as three-phase equilibrium, is three-phase phase voltage by the single-phase optimization photovoltaic conversion of power transmission network boundary node Try to achieve.

(4) three-phase imbalance and distributed power source participate in reactive voltage from the aspect of active distribution network reactive Voltage Optimum Optimization, with loss minimization as target, node voltage etc. be that the three-phase reactive Voltage Optimum model containing distributed power source is set up in constraint, It is solved using the method based on complementary theory and Non-Linear Programming interior point method that can process discrete variable；Containing distribution The three-phase reactive Voltage Optimum model of formula power supply is as follows：

V_i,p ^min≤V_i,p≤V_i,p ^max

Q_idg,p ^min≤Q_idg,p≤Q_idg,p ^max4)；

Q_icap,p ^min≤Q_icap,p≤Q_icap,p ^max

T_ik,p ^min≤T_ik,p≤T_ik,p ^max

Wherein：f_pX () is the network loss of active distribution network system p phases, p ∈ (a, b, c)；F (x) is active distribution network system Network loss；X=[x₁ ^T,x₂ ^T]^T, x₁For distributed power source it is idle exert oneself, reactive power compensator it is idle exert oneself and on-load voltage regulation become The no-load voltage ratio column vector of depressor, x₂It is the real part and imaginary part column vector of node phase voltage；P_i ^pAnd Q_i ^pThe respectively note of the p phases of node i Enter active power and reactive power；P_gi ^pAnd P_li ^pThe respectively generator active power and load active power of the p phases of node i, Q_gi ^pAnd Q_li ^pThe respectively generator reactive power and reactive load power of the p phases of node i；Formula 4) in power-balance equation about Shu Fangcheng bibliography《Three-phase power flow calculations using the current injection method》(Paulo A.N.Garcia,Jose Luiz R.Pereira,Sandoval Carneiro,Jr., et al.Three-phase power flow calculations using the current injection method.IEEE TRANSACTIONS ON POWER SYSTEMS,2000,15(2):508-514)。V_ri ^pAnd V_mi ^pRespectively The real part and imaginary part of the p phase voltages of node i；G_ij ^ptAnd B_ij ^ptIt is the node admittance element between node i and node j, t ∈ (a, b, c)；V_i,p、V_i,p ^minAnd V_i,p ^maxRespectively the p phase voltages of node i, p phase voltages minimum value and maximum；Q_idg,p、Q_idg,p ^minWith Q_idg,p ^maxRespectively i-th p phase of distributed power source it is idle exert oneself, p phases are idle exerts oneself minimum value and maximum；Q_icap,p、 Q_icap,p ^minAnd Q_icap,p ^maxThe p phases of respectively i-th reactive-load compensation equipment are idle to exert oneself, p phases are idle exerts oneself minimum value and maximum Value；T_ik,p、T_ik,p ^minAnd T_ik,p ^maxRespectively i-th the p phases no-load voltage ratio of ULTC, p phase no-load voltage ratio minimum values and maximum Value；

In order to process the discrete variable in Mathematical Modeling, the accurate solving model of discrete variable is built based on complementary theory, Using document《Meter and discrete variable are based on the idle work optimization of the full Smoothing Newton Method of Constraints》(Lin Jikeng, Shi Weizhao, Wu Nai Tiger, waits to count and idle work optimization Proceedings of the CSEE of the discrete variable based on the full Smoothing Newton Method of Constraints, and 2012,32 (1)：The method of the construction discrete variable Mathematical Programs With Nonlinear Complementarity Constraints condition in 93-100), builds the idle electricity of active distribution network three-phase The Mathematical Programs With Nonlinear Complementarity Constraints condition of discrete variable is as follows in pressure optimization：

Wherein：f_i ^pIt is that i-th ULTC p phases gear or reactive power compensator p phase switching groups number are become as continuous The first optimal value that amount treatment is obtained；f_i(n+1) ^pAnd f_in ^pRespectively f_i ^pThe actual-gear or switching group number of left and right；f_1xAnd f_2xFor Intermediate variable；μ is smoothing parameter；

Formula 4), 5), 6) constitute containing Constraints condition active distribution network three-phase reactive Voltage Optimum model, using non-thread Property planning interior point method solved.

Power transmission network comprising transmission and distribution network orchestration boundary node is obtained by the step (2) and step (4) and optimizes knot Fruit and active distribution network three-phase optimum results.

(5) as power transmission network phase after the three-phase optimization power of each transmission and distribution network boundary node in active distribution network is sued for peace Answer the load power of node.

(6) power transmission network and active distribution network reactive Voltage Optimum alternating iteration, under the coordination of transmission and distribution network boundary node, Judge whether the adjacent change twice in iteration of the power of transmission and distribution network boundary node meets convergence precision, turn if meeting To (7), (2) are gone to if being unsatisfactory for and continues iteration；Convergence precision needs to be set according to different computational accuracies.

(7) power transmission network and active distribution network reactive Voltage Optimum finally give complete in transmission and distribution network boundary node everywhere convergent Office's reactive Voltage Optimum strategy.

Angle from global united analysis of the invention, active distribution network aspect considers the three-phase imbalance of power distribution network Reactive Voltage Optimum is participated in distributed power source, using triphase flow, active distribution network three-phase reactive Voltage Optimum is being carried out In the case of, the complementary Interior-point method coordinated based on transmission and distribution network boundary node and can process discrete variable is solved, and is proposed a kind of The idle resource of transmission and distribution network Global optimization method, optimization power transmission network and active distribution network, by power transmission network and master The coordination interaction of dynamic electric distribution network reactive-voltage optimization, solves the above problems, while realize that the optimization of global idle resource is coordinated, drop Low via net loss, improves the quality of voltage of power transmission network and active distribution network.

Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention rather than its limitations, to the greatest extent Pipe has been described in detail with reference to above-described embodiment to the present invention, and those of ordinary skill in the art still can be to this hair Bright specific embodiment is modified or equivalent, these without departing from spirit and scope of the invention any modification or Equivalent, is applying within pending claims of the invention.

Claims (7)

1. a kind of transmission and distribution network Global optimization method, it is characterised in that methods described comprises the steps：

(1) power transmission network and active distribution network orchestration boundary node are selected；

(2) the power transmission network reactive Voltage Optimum model containing Constraints condition is built, it is entered using Non-Linear Programming interior point method Row is solved；

(3) it is three-phase voltage as active distribution network phase using the optimization photovoltaic conversion of each transmission and distribution network boundary node in power transmission network Answer the three-phase voltage of root node；

(4) the three-phase reactive Voltage Optimum model containing Constraints condition of the active distribution network containing distributed power source is built, is adopted It is solved with Non-Linear Programming interior point method；

(5) will accordingly be saved as power transmission network after the three-phase optimization power summation of each transmission and distribution network boundary node in active distribution network The load power of point；

(6) judge whether adjacent change twice meets convergence precision in an iterative process for the power of transmission and distribution network boundary node；

(7) convergence according to power transmission network and active distribution network reactive Voltage Optimum at transmission and distribution network boundary node, obtains the overall situation Reactive Voltage Optimum strategy；

In the step (2), with loss minimization as target, node voltage set up power transmission network reactive Voltage Optimum model for constraint, It is solved using the method based on complementary theory and Non-Linear Programming interior point method that can process discrete variable；Power transmission network Reactive Voltage Optimum model is as follows：

$\begin{array}{c}\min f\left(x\right)=P_l\\ s.t.P_i=P_{gi}-P_{li}=V_i\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{V}_j\left(G_{ij}{\mathrm{cos\θ}}_{ij}+B_{ij}{\mathrm{sin\θ}}_{ij}\right)\\ Q_i=Q_{gi}-Q_{li}=V_i\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{V}_j\left(G_{ij}{\mathrm{sin\θ}}_{ij}-B_{ij}{\mathrm{cos\θ}}_{ij}\right)\\ {V_i}^{\min }\≤V_i\≤{V_i}^{\max }\\ {Q_{gi}}^{\min }\≤Q_{gi}\≤{Q_{gi}}^{\max }\\ {Q_{ci}}^{\min }\≤Q_{ci}\≤{Q_{ci}}^{\max }\\ {T_{ki}}^{\min }\≤T_{ki}\≤{T_{ki}}^{\max }\end{array}---1);$

Wherein：P_lIt is the network loss of transmission system；N is transmission system nodes；P_iAnd Q_iThe respectively injection active power of node i And reactive power；P_giAnd P_liThe respectively generator active power of node i and load active power；Q_giAnd Q_liRespectively node i Generator reactive power and reactive load power；G_ijAnd B_ijConductance and susceptance respectively between node i and node j；θ_ijFor Phase difference of voltage between node i and node j；V_i、V_i ^minAnd V_i ^maxThe respectively voltage magnitude of node i, voltage minimum and maximum Value；Q_gi、Q_gi ^minAnd Q_gi ^maxGenerator respectively at node i it is idle exert oneself, idle minimum value and the maximum of exerting oneself；Q_ci、 Q_ci ^minAnd Q_ci ^maxReactive-load compensation equipment respectively at node i it is idle exert oneself, idle minimum value and the maximum of exerting oneself；T_ki、 T_ki ^minAnd T_ki ^maxULTC no-load voltage ratio, no-load voltage ratio minimum value and maximum respectively at node i；

The Mathematical Programs With Nonlinear Complementarity Constraints condition for building discrete variable in power transmission network reactive Voltage Optimum is as follows：

$\left\{\begin{array}{c}{f}_{1x}=f_{i(n+1)}-f_i\≥0\\ f_{2x}=f_i-f_{in}\≥0\\ f_{1x}{f}_{2x}=0\end{array}\right.---2);$

$\mathrm{\ψ}(\mathrm{\μ},f_{1x},f_{2x})=f_{1x}+f_{2x}-\sqrt{{(f_{1x}-f_{2x})}^2+4{\mathrm{\μ}}^2}---3);$

Wherein：f_iIt is that i-th ULTC gear or reactive power compensator switching group number process what is obtained by continuous variable First optimal value；f_i(n+1)And f_inRespectively f_iThe actual-gear or switching group number of left and right；f_1xAnd f_2xIt is intermediate variable；μ is light Sliding parameter；

Formula 1), 2), 3) constitute containing Constraints condition power transmission network reactive Voltage Optimum model, using Non-Linear Programming interior point method Solved；

In the step (4), with loss minimization as target, node voltage be constraint set up containing distributed power source active distribution network Three-phase reactive Voltage Optimum model, is solved using treatment discrete variable based on complementary theory and Non-Linear Programming interior point method；Contain The three-phase reactive Voltage Optimum model of distributed power source is as follows：

$\begin{array}{c}\min f\left(x\right)=\underset{p\∈\left\{a,b,c\right\}}{\mathrm{\Σ}}{f}_p\left(x\right)\\ s.t.{P_i}^p={P_{gi}}^p-{P_{li}}^p={V_{ri}}^p(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{t\∈\left\{a,b,c\right\}}{\mathrm{\Σ}}{\left({G_{ij}}^{pt}{V_{rj}}^t-{B_{ij}}^{pt}{V_{mj}}^t\right)}^p+{V_{mi}}^p{\left({G_{ij}}^{pt}{V_{mj}}^t-{B_{ij}}^{pt}{V_{rj}}^t\right)}^p\\ {Q_i}^p={Q_{gi}}^p-{Q_{li}}^p={V_{mi}}^p(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{t\∈\left\{a,b,c\right\}}{\mathrm{\Σ}}{\left({G_{ij}}^{pt}{V_{rj}}^t-{B_{ij}}^{pt}{V_{mj}}^t\right)}^p-{V_{ri}}^p{\left({G_{ij}}^{pt}{V_{mj}}^t-{B_{ij}}^{pt}{V_{rj}}^t\right)}^p\\ {V_{i,p}}^{\min }\≤V_{i,p}\≤{V_{i,p}}^{\max }\\ {Q_{idg,p}}^{\min }\≤Q_{idg,p}\≤{Q_{idg,p}}^{\max }\\ {Q_{icap,p}}^{\min }\≤Q_{icap,p}\≤{Q_{icap,p}}^{\max }\\ {T_{ik,p}}^{\min }\≤T_{ik,p}\≤{T_{ik,p}}^{\max }\end{array}---4);$

Wherein：f_pX () is the network loss of active distribution network system p phases, p ∈ (a, b, c)；F (x) is the network loss of active distribution network system； X=[x₁ ^T,x₂ ^T]^T, wherein x₁For distributed power source it is idle exert oneself, the idle of reactive power compensator is exerted oneself and on-load voltage regulation transformation The no-load voltage ratio column vector of device, x₂It is the real part and imaginary part column vector of node phase voltage；P_i ^pAnd Q_i ^pThe respectively injection of the p phases of node i Active power and reactive power；P_gi ^pAnd P_li ^pThe respectively generator active power and load active power of the p phases of node i, Q_gi ^p And Q_li ^pThe respectively generator reactive power and reactive load power of the p phases of node i；V_ri ^pAnd V_mi ^pRespectively p phases of node i The real part and imaginary part of voltage；G_ij ^ptAnd B_ij ^ptIt is the node admittance element between node i and node j, t ∈ (a, b, c)；V_i,p、 V_i,p ^minAnd V_i,p ^maxRespectively the p phase voltages of node i, p phase voltages minimum value and maximum；Q_idg,p、Q_idg,p ^minAnd Q_idg,p ^maxPoint Not Wei i-th p phase of distributed power source it is idle exert oneself, p phases are idle exerts oneself minimum value and maximum；Q_icap,p、Q_icap,p ^minWith Q_icap,p ^maxThe p phases of respectively i-th reactive-load compensation equipment are idle to exert oneself, p phases are idle exerts oneself minimum value and maximum；T_ik,p、 T_ik,p ^minAnd T_ik,p ^maxRespectively i-th the p phases no-load voltage ratio of ULTC, p phase no-load voltage ratio minimum values and maximum；

The Mathematical Programs With Nonlinear Complementarity Constraints condition for building discrete variable in active distribution network three-phase reactive Voltage Optimum is as follows：

$\left\{\begin{array}{c}{f}_{1x}={f_{i\left(n+1\right)}}^p-{f_i}^p\≥0\\ f_{2x}={f_i}^p-{f_{in}}^p\≥0\\ f_{1x}{f}_{2x}=0\end{array}\right.---5);$

$\mathrm{\ψ}(\mathrm{\μ},f_{1x},f_{2x})={f_{1x}}^p+{f_{2x}}^p-\sqrt{{({f_{1x}}^p-{f_{2x}}^p)}^2+4{\mathrm{\μ}}^2}---6);$

Wherein：f_i ^pIt is that i-th ULTC p phases gear or reactive power compensator p phase switching groups number are pressed at continuous variable The first optimal value that reason is obtained；f_i(n+1) ^pAnd f_in ^pRespectively f_i ^pThe actual-gear or switching group number of left and right；f_1xAnd f_2xIt is centre Variable；μ is smoothing parameter；

Formula 4), 5), 6) constitute containing Constraints condition active distribution network three-phase reactive Voltage Optimum model, using non-linear rule Interior point method is drawn to be solved.

2. Global optimization method as claimed in claim 1, it is characterised in that in the step (1), selectes transmission of electricity Net and active distribution network orchestration boundary node, to active distribution network three-phase model, build power transmission network and active distribution network it is single-phase with The coordination interactive frame of three reactive Voltage Optimums for mixing, and the single-phase coordination with three reactive Voltage Optimums for mixing is handed over Mutual pattern.

3. Global optimization method as claimed in claim 1, it is characterised in that in the step (3), power transmission network and During active distribution network reactive Voltage Optimum alternating iteration, by the optimization voltage of each transmission and distribution network boundary node in power transmission network It is converted into three-phase voltage of the three-phase voltage as the corresponding root node of active distribution network；

It is in the steps below three-phase voltage by the optimization photovoltaic conversion of power transmission network boundary node：By active distribution network root node voltage It is set as three-phase equilibrium, by the single-phase optimization photovoltaic conversion of power transmission network orchestration boundary node for three-phase phase voltage is tried to achieve.

4. Global optimization method as claimed in claim 1, it is characterised in that by the step (2) and step (4) Obtain the power transmission network optimum results comprising transmission and distribution network orchestration boundary node and active distribution network three-phase optimum results.

5. Global optimization method as claimed in claim 1, it is characterised in that in the step (5), power transmission network and During active distribution network reactive Voltage Optimum alternating iteration, by the three-phase of each transmission and distribution network boundary node in active distribution network As the load power of power transmission network respective nodes after optimization power summation.

6. Global optimization method as claimed in claim 1, it is characterised in that in the step (6), judge transmission ＆ distribution Whether adjacent change twice meets convergence precision to the power of power network boundary node in an iterative process, and step is gone to if meeting Suddenly (7), step (2) is gone to if being unsatisfactory for and continues iteration；Convergence precision is the convergence parameter of setting power.

7. Global optimization method as claimed in claim 1, it is characterised in that in the step (7), work as power transmission and distribution When the adjacent change twice in iteration of the power of net boundary node meets convergence precision, power transmission network and the idle electricity of active distribution network Pressure optimization finally gives Global optimisation strategy in transmission and distribution network boundary node everywhere convergent.