CN106780119A - Power distribution network regenerative resource based on many active managements is dissolved method - Google Patents

Abstract

Dissolved method the present invention relates to a kind of power distribution network regenerative resource based on many active managements, various active management measures in weight analysis active distribution network ADN.Temporal characteristicses based on regenerative resource and load, analysis regenerative resource is exerted oneself influence of the various active management measures such as excision, OLTC tap_changings, network reconfiguration to regenerative resource digestion capability.Consideration load, the model of dissolving of regenerative resource temporal characteristicses are established, rapid solving is carried out to model using prim al- dual interior point m ethod, the service condition of power distribution network can be improved, improve digestion capability of the power distribution network to regenerative resource.

Description

Power distribution network regenerative resource based on many active managements is dissolved method

Technical field

The present invention relates to a kind of administration of power networks technology, more particularly to a kind of power distribution network renewable energy based on many active managements Source is dissolved method.

Background technology

With the continuous propulsion of the fast-developing and Process of Urbanization Construction of social economy, electricity needs sustainable growth；Traditional Centralized extensive generating can not meet the requirement that electric energy clean manufacturing and high efficiency of energy are utilized, while returning environmental protection With heavy pressure.Under such background, distributed energy particularly intermittence regenerative resource will be environment-friendly by its Advantage be rapidly developed.Distributed power generation (Distributed Generation, DG) is often referred to be arranged near user Small-sized independent electricity generation system, mainly includes：Miniature gas turbine, wind-driven generator, photovoltaic generator, energy storage etc..DG conducts One of important form using renewable energy power generation, is rapidly developed under policies of various countries promotion, will progressively turn into complete World's power field important component.

With the fast development of DG, DG is accessed and is also brought a series of problem to power distribution network, and such as access point voltage raises, is System bi-directional current etc..Due to current passive management pattern, system is not controlled DG and network itself accordingly, cause be System cannot make full use of DG in the positive role of the aspects such as improvement system losses and quality of voltage.For this passive management mould Formula, many scholars propose the concept of active distribution network (Active Distribution Network, ADN) and active management, And active management is applied in the dissolving of intermittence regenerative resource.Active management is exactly that more careful measurement and assessment are matched somebody with somebody After the system operation data of power network, real-time control is carried out to DG and Distribution Network Equipment and takes certain measure to coordinate. Power distribution network under active management pattern can take control DG's to send power, regulation transformer tapping and reactive-load compensation equipment etc. Various active management measures so that the running status that the distribution system containing DG is optimal.So as to improve power distribution network to DG's Digestion capability, improves the power supply reliability of distribution system, improves the quality of power supply of power distribution network.

The content of the invention

The present invention be directed to the problem that power distribution network regenerative resource is dissolved, it is proposed that a kind of distribution based on many active managements Net regenerative resource is dissolved method, various active management measures in weight analysis active distribution network ADN.Based on regenerative resource and The temporal characteristicses of load, analysis regenerative resource various active managements such as excision, OLTC tap_changings, network reconfiguration of exerting oneself are arranged Apply the influence to regenerative resource digestion capability.

The technical scheme is that：A kind of power distribution network regenerative resource based on many active managements is dissolved method, specifically Comprise the following steps：

1) the power distribution network regenerative resource based on active management is set up to dissolve model, the minimization of object function wind-powered electricity generation and photovoltaic Resection, that is, that wind-powered electricity generation photovoltaic power generation quantity is reached in having the branch road of wind-powered electricity generation and photovoltaic is maximum；

2) setting up constraints includes power-balance constraint and network operation security constraint；

3) data input, including network parameter, distributed electrical source data, wind-powered electricity generation, photovoltaic, the temporal characteristicses data of load, Determine to analyze scene using scene analysis method；Historical data by wind-powered electricity generation, photovoltaic, load in 1 year constitutes A_m×nMatrix, m Represent the number of days of historical data；N represents the every day of wind-powered electricity generation, photovoltaic, load data by the hour；Using k- means clustering algorithms, M n dimension sample cluster the scene number k for being needed；Selection scene, and carry out the tune of respective fields scape optimal power flow problems It is whole；

4) time=0 is put, the intermittent energy source EIAJ in the scene is calculated using prim al- dual interior point m ethod PDIPM, i.e., Minimum cuts machine, while obtaining the optimal value of control variables, time=time+1；

5) return to step 4), until time=setting values, by the generating of the intermittent energy source within accumulation calculating 1 year Amount；

6) later scene is selected, the adjustment of corresponding scene optimal power flow problems, return to step 4 is carried out), until all scenes All over row；

7) the power distribution network regenerative resource allocation optimum under each scene is obtained.

The step 2) in constraints include power-balance constraint and network operation security constraint, it is specific as follows：

1)

2)

3)

4)

5)

6)

7)

8)

9)g_k∈G

Wherein, P_GiAnd Q_GiThe respectively generated power and reactive power of node i；P_Li, Q_LiThe respectively load of node i Active and reactive power；Q_CiIt is the reactive compensation capacity of node i；U_i, U_jThe respectively voltage magnitude of node i and j, j ∈ i are represented All nodes being joined directly together with node i；G_ij、B_ij、δ_ijIt is respectively conductance between node i and j, susceptance and phase difference of voltage； Formula 1) it is node power Constraints of Equilibrium；

Formula 2) it is node voltage constraint, U_i、U_i,maxAnd U_i,minNode i voltage magnitude and the voltage for being allowed are represented respectively Amplitude upper and lower bound；

Formula 3) it is Branch Power Flow constraint, S_i、S_i,maxRepresent respectively in branch road i actual powers and the transimission power for being allowed Limit；

Formula 4) for distributed power source exert oneself bound constraint, wherein P_DG,i,It is respectively i-th DG, DG includes wind-powered electricity generation WT, photovoltaic PV, the active power output of miniature gas turbine MT, and bound；Q_DG,i,Respectively It is that i-th the idle of DG is exerted oneself, and bound；

Formula 5) it is reactive-load compensation constraint, wherein Q_CiI-th switching amount of node reactive-load compensation equipment is represented,WithRepresent the upper and lower bound of reactive-load compensation equipment switching amount；

Formula 6) it is the constraint of OLTC tap taps adjustable range, OLTC_k、WithIt is respectively kth Platform Algorithm of Tap Changing under Load Transformer, tap adjustable range upper and lower bound；

Formula 7) it is load abatement or Demand Side Response constraint, P_Lcur,i、Respectively i-th node load is cut down Amount and load cut down the upper limit；

Formula 8) it is wind-powered electricity generation, the functional relation that photovoltaic is idle to exert oneself with active power output, this formula is adopted only for intermittent DG Use constant power factor control, formula 4) in idle scope for intermittence DG is active, idle uneoupled control；

Formula 9) it is network topology constraint, wherein g during network reconfiguration_kIt is the new network structure for being formed of reconstruct, G is institute Having can meet radial and without isolated node the network topology structure collection that load is powered.

The beneficial effects of the present invention are：Power distribution network regenerative resource of the present invention based on many active managements is dissolved method, Consideration load, the model of dissolving of regenerative resource temporal characteristicses are established, model is quickly asked using prim al- dual interior point m ethod Solution, can improve the service condition of power distribution network, improve digestion capability of the power distribution network to regenerative resource.

Brief description of the drawings

Fig. 1 is the example network structures of improved IEEE of the invention 33；

Fig. 2 is the excision spirogram of intermittent energy source under different scenes of the present invention；

Fig. 3 is the spirogram of dissolving of intermittent energy source under different scenes of the present invention；

Fig. 4 is the 22nd week Tuesday OLTC tap joint position figure of the invention；

Fig. 5 is the result figure of dissolving of the present invention the 22nd week Tuesday scene S2 and S4.

Specific embodiment

Power distribution network regenerative resource dissolve method of the present invention based on many active managements is realized by following approach：

1) the power distribution network regenerative resource based on active management is set up to dissolve model, the minimization of object function wind-powered electricity generation and photovoltaic Resection.Reach within the whole research period, wind-powered electricity generation photovoltaic power generation quantity is maximum.It is specific as follows：

Wherein：It is the i-th typhoon group of motors exerting oneself in t,It is jth platform photovoltaic unit in t Exert oneself, nw, np are respectively the installation number of blower fan and photovoltaic.

2) setting up constraints includes power-balance constraint and network operation security constraint etc., specific as follows：

g_k∈G (10)

Wherein, P_GiAnd Q_GiThe respectively generated power and reactive power of node i；P_Li, Q_LiThe respectively load of node i Active and reactive power；Q_CiIt is the reactive compensation capacity of node i；U_i, U_jThe respectively voltage magnitude of node i and j, j ∈ i are represented All nodes being joined directly together with node i；G_ij、B_ij、δ_ijIt is respectively conductance between node i and j, susceptance and phase difference of voltage； Formula (2) is node power Constraints of Equilibrium.

Formula (3) is constrained for node voltage, U_i、U_i,maxAnd U_i,minNode i voltage magnitude and the electricity for being allowed are represented respectively Pressure amplitude value upper and lower bound；Formula (4) is constrained for Branch Power Flow, S_i、S_i,maxBranch road i actual powers are represented respectively and are allowed The transimission power upper limit.

Formula (5) for distributed power source exert oneself bound constraint, wherein P_DG,i,It is respectively i-th DG, including wind-powered electricity generation (Wind Turbine, WT), photovoltaic (Photovoltaic, PV), miniature gas turbine (Micro- Turbine, MT) active power output, and bound；Q_DG,i,It is respectively i-th DG (Wind turbines, photovoltaic machine Group, schedulable miniature gas turbine) it is idle exert oneself, and bound.

Formula (6) is constrained for reactive-load compensation, wherein Q_CiI-th switching amount of node reactive-load compensation equipment is represented, WithRepresent the upper and lower bound of reactive-load compensation equipment switching amount.

Formula (7) is constrained for OLTC taps adjustable range, OLTC_k、WithIt is respectively that kth platform has Voltage adjustment of on-load transformer tap positions, tap adjustable range upper and lower bound.

Formula (8) is cut down or Demand Side Response constraint for load, P_Lcur,i、Respectively i-th node load is cut Decrement and load cut down the upper limit.

Formula (9) is wind-powered electricity generation, the functional relation that photovoltaic is idle to exert oneself with active power output, and this formula is adopted only for intermittent DG Constant power factor control is used, idle scope is for intermittence DG is active, idle uneoupled control in formula (5).

Formula (10) is network topology constraint, wherein g during network reconfiguration_kIt is the new network structure for being formed of reconstruct, G is It is all to meet radial and without isolated node the network topology structure collection that load is powered.

3) model is solved using prim al- dual interior point m ethod：

Active distribution network problem is actually a problem for active trend management, i.e. optimal power flow problems.The present invention is adopted The optimal power flow problems are asked with prim al- dual interior point m ethod (Primal-Dual Interior Point Method, PDIPM) Solution.Because interior point method is in Powerfactory, modularization in the power system calculation software such as Matpower is not described, only herein Provide following control variables, object function and Lagrangian.

In formula：X is control variables column vector；θ, v are node voltage phase angle and amplitude vector；P_g, Q_gFor generated power with Reactive power vector, including transformer station S/S (Substation), WT, PV and MT；F (X) is object function, f (X) ∈ R¹；It is complete 1 row vector；P_wt, P_pvIt is wind-powered electricity generation, the active vector of photovoltaic；L (X) is Lagrangian, L (X) ∈ R¹； G (X) is equality constraints functions, G (X) ∈ R^l, l is equality constraint number；H₁(X), H₂(X) the inequality upper limit, lower limit are respectively about Beam function, H (X) ∈ R^m, m is inequality constraints number；S₁, S₂It is slack variable, λ, τ₁, τ₂It is dual variable, μ joins for obstacle Number.

Assuming that it is separate between each moment active management measure, different active management measures are analyzed using scene analysis method The situation of dissolving of lower regenerative resource, problem solving flow is as follows：

1) data input, including network parameter, distributed electrical source data, wind-powered electricity generation, photovoltaic, the temporal characteristicses data of load, But determine to analyze scene using scene analysis method.Historical data by wind-powered electricity generation, photovoltaic, load in 1 year constitutes A_m×nMatrix, m Represent the number of days of historical data；N represents the every day of wind-powered electricity generation, photovoltaic, load data by the hour.Using k- means clustering algorithms, M n dimension sample cluster the scene number k for being needed.Selection scene, and carry out the tune of respective fields scape optimal power flow problems It is whole.

2) put time=0, using prim al- dual interior point m ethod PDIPM calculate the scene intermittent energy source EIAJ (i.e. Minimum cuts machine), while obtaining the optimal value of control variables；

If needing to consider network reconfiguration in active management, the maximum mesh as network reconfiguration that intermittent energy source is exerted oneself is needed Scalar functions, so as to obtain the optimal value (including network optimum structure) of control variables.

Time=time+1.

3) return 2), until time=8760 (setting value).By the hair of the intermittent energy source within accumulation calculating 1 year Electricity；

4) later scene is selected, the adjustment of corresponding scene optimal power flow problems is carried out, the 2) step is returned to, until all scenes All over row.

Using the node examples of IEEE 33, the example is a distribution system for 12.66kV single supplies, containing 33 nodes and 5 Bar interconnection, total load is 3715kW, 2300kvar.

The present invention power distribution network is had made some improvements, network structure is shown in Fig. 1, be included in node 2,7,20,21,24, 31 introduce distributed power source, the specific meshed network DG installation situations of IEEE 33 as shown in table 1, it is assumed that blower fan is using double-fed asynchronous Generator, photovoltaic power generation grid-connecting side is equipped with active compensation equipment, and wind-powered electricity generation, photovoltaic are idle can be separately adjustable.Load has been provided with node 0 Adjustable transformer ± 2 × 2.5%, contains five grades of taps.Contain important load at node 24,31, miniature gas turbine is cut Except amount need to be less than 50%.Node voltage allowed band is 0.95~1.05p.u, and branch road 1,2 longtime running rated capacities are 5MVA, Branch road 3,4,5 is 4MVA, and other branch roads are 2MVA.

Table 1

In order to analyze the influence that active management is dissolved to regenerative resource, the present invention uses scene analysis, and concrete scene sets Put as follows：

Scene S1：Regenerative resource is active, it is idle carry out uneoupled control, it is uncontrollable that OLTC taps fix 1.05, MT, Without network reconfiguration；

Scene S2：Regenerative resource is active, it is idle carry out uneoupled control, it is controllable that OLTC taps fix 1.05, MT, nothing Network reconfiguration；

Scene S3：Regenerative resource is active, it is idle carry out uneoupled control, OLTC taps are controllable, and MT is controllable, without network Reconstruct；

Scene S4：Regenerative resource is active, it is idle carry out uneoupled control, it is controllable that OLTC taps fix 1.05, MT, bag Containing network reconfiguration.

Table 2 is given in the case of different scenes, different regenerative resource access capacities, the regenerative resource year amount of dissolving, year Resection and system year network loss results contrast (MWh).IRE is Preliminary batch performance source access capacity, and 1.2 × IRE is intermittence The energy more initially increases by 20%.The use of active management as seen from Figure 2 can substantially reduce the resection of regenerative resource. With the increase of regenerative resource access capacity, scene S1, S2 resection is substantially increased；And scene S3 resection increases in capacity When less, hardly increase resection, when capacity increases above 50%, resection will be substantially increased.This explanation is a large amount of actively The use of control measures can increase access capacity of the system to intermittent energy source, but be not unconfined, be managed more than active , it is necessary to cut-out regenerative resource ensures system safe and stable operation after the limit of power of reason.

Table 2

Active management measure as seen from Figure 3 can increase dissolve amount of the system to regenerative resource.For initial shape The regenerative resource access capacity of state, the year amount of dissolving scene S2 of regenerative resource increased 1.15%, scene S3 compared with scene S1 11.26% is increased compared with scene S1.When regenerative resource access capacity increases by 50%, the year amount of dissolving scene S2 increases compared with scene S1 1.5%, scene S3 increased 25.36% compared with scene S1.

Fig. 4 gives the change of OLTC tap joint positions in the 22nd week 24 hours Tuesday running in scene S3.Big portion Timesharing carves voltage branch point control 1.025, and voltage sets too high (1.05) and can reduce the access amount of regenerative resource, main to use In the load peak period；Voltage sets too low (1) can cause some node voltages more lower limit, be used primarily in the load valley period.

It is large-scale nonlinear combinatorial optimization problem, real-time optimization because power distribution network reconfiguration needs to optimize substantial amounts of switch Scheduling difficulty is larger, so being only analyzed to the 22nd week time period Tuesday in scene S4, is only once weighed in the time period Structure active management.Table 3 and Fig. 5 give the result of dissolving of scene S2 and S4.Under initial regenerative resource access capacity, scene S4 Comparing scene S2 can increase by 7.58% amount of dissolving.When regenerative resource access capacity increases, it is considered to which the amount of dissolving of DR is fast Speed rises；When regenerative resource access capacity increases by 50%, scene S4 compares scene S2 and can increase by 20.09% and dissolves Amount.

Table 3

The present invention analyzes various active management measures in ADN, it is proposed that based on load, regenerative resource temporal characteristicses Regenerative resource is dissolved model, and usage scenario analytic approach is studied regenerative resource digestion capability in ADN.By changing The node Example Verifications of IEEE 33 for entering, have drawn as drawn a conclusion：

The incorrect conduction of regenerative resource can cause access point voltage to raise, and influence the safe and stable operation of power distribution network, actively Management can improve the service condition of power distribution network, improve power distribution network and regenerative resource is dissolved；

The use of active management can be greatly reduced the resection of regenerative resource, but active management raising system pair can The renewable sources of energy dissolve be not it is unconfined, more than after the limit of power of active management, it is necessary to cut-out regenerative resource Guarantee system safe and stable operation；

The application of active management measure is more in ADN, can more improve the access capacity and the amount of dissolving of regenerative resource.OLTC Tap_changing and network reconfiguration can play a positive role for improving power distribution network to dissolving for regenerative resource, but in reality Need to consider the cost of active management in the active management of border, that is, need to consider the adjustment frequency and network reconfiguration of OLTC taps Frequency.

Claims (2)

1. a kind of power distribution network regenerative resource based on many active managements is dissolved method, it is characterised in that specifically include following step Suddenly：

1) set up the power distribution network regenerative resource based on active management to dissolve model, the minimization of object function wind-powered electricity generation and photovoltaic are cut Except amount, that is, wind-powered electricity generation photovoltaic power generation quantity maximum is reached in having the branch road of wind-powered electricity generation and photovoltaic；

2) setting up constraints includes power-balance constraint and network operation security constraint；

3) data input, including network parameter, distributed electrical source data, wind-powered electricity generation, photovoltaic, the temporal characteristicses data of load are used Scene analysis method come determine analyze scene；Historical data by wind-powered electricity generation, photovoltaic, load in 1 year constitutes A_m×nMatrix, m is represented The number of days of historical data；N represents the every day of wind-powered electricity generation, photovoltaic, load data by the hour；Using k- means clustering algorithms, by m N dimension samples cluster the scene number k for being needed；Selection scene, and carry out the adjustment of respective fields scape optimal power flow problems；

4) put time=0, the intermittent energy source EIAJ in the scene is calculated using prim al- dual interior point m ethod PDIPM, i.e., it is minimum Machine is cut, while obtaining the optimal value of control variables, time=time+1；

5) return to step 4), until time=setting values, by the generated energy of the intermittent energy source within accumulation calculating 1 year；

6) later scene is selected, the adjustment of corresponding scene optimal power flow problems, return to step 4 is carried out), until all scenes time Row；

7) the power distribution network regenerative resource allocation optimum under each scene is obtained.

2. the power distribution network regenerative resource based on many active managements is dissolved method according to claim 1, it is characterised in that institute State step 2) in constraints include power-balance constraint and network operation security constraint, it is specific as follows：

$1)---\begin{array}{c}{P}_{Gi}-P_{Li}=U_i\underset{j\∈i}{\mathrm{\Σ}}{U}_j(G_{ij}{\mathrm{cos\δ}}_{ij}+B_{ij}{\mathrm{sin\δ}}_{ij})\\ Q_{Gi}+Q_{Ci}-Q_{Li}=U_i\underset{j\∈i}{\mathrm{\Σ}}{U}_j(G_{ij}{\mathrm{sin\δ}}_{ij}-B_{ij}{\mathrm{cos\δ}}_{ij})\end{array}$

$2)---U_i^{min}\≤U_i\≤U_i^{max}$

$3)---S_i\≤S_i^{\max }$

$4)---\begin{array}{c}{P}_{DG,i}^{min}\≤P_{DG,i}\≤P_{DG,i}^{\max }\\ Q_{DG,i}^{min}\≤Q_{DG,i}\≤Q_{DG,i}^{\max }\end{array}$

$5)---Q_{C,i}^{\min }\≤Q_{C,i}\≤Q_{C,i}^{\max }$

$6)---{\mathrm{OLTC}}_k^{min}\≤{\mathrm{OLTC}}_k\≤{\mathrm{OLTC}}_k^{\max }$

$7)---P_{Lcur,i}\≤P_{Lcur,i}^{\max }$

$8)---\begin{array}{c}{Q}_{WT,i}=g\left(P_{WT,i}\right)\\ Q_{PV,i}=h\left(P_{PV,i}\right)\end{array}$

9)g_k∈G

Wherein, P_GiAnd Q_GiThe respectively generated power and reactive power of node i；P_Li, Q_LiThe load of respectively node i is active And reactive power；Q_CiIt is the reactive compensation capacity of node i；U_i, U_jThe respectively voltage magnitude of node i and j, j ∈ i represent all The node being joined directly together with node i；G_ij、B_ij、δ_ijIt is respectively conductance between node i and j, susceptance and phase difference of voltage；Formula 1) it is node power Constraints of Equilibrium；

Formula 2) it is node voltage constraint, U_i、U_i,maxAnd U_i,minNode i voltage magnitude and the voltage magnitude for being allowed are represented respectively Upper and lower bound；

Formula 3) it is Branch Power Flow constraint, S_i、S_i,maxBranch road i actual powers and the transimission power upper limit for being allowed are represented respectively；

Formula 4) for distributed power source exert oneself bound constraint, wherein P_DG,i,It is respectively i-th DG, DG bags Include wind-powered electricity generation WT, photovoltaic PV, the active power output of miniature gas turbine MT, and bound；Q_DG,i,It is respectively i-th The idle of platform DG is exerted oneself, and bound；

Formula 5) it is reactive-load compensation constraint, wherein Q_CiI-th switching amount of node reactive-load compensation equipment is represented,WithRepresent the upper and lower bound of reactive-load compensation equipment switching amount；

Formula 6) it is the constraint of OLTC tap taps adjustable range, OLTC_k、WithIt is respectively that kth platform has Voltage adjustment of on-load transformer tap positions, tap adjustable range upper and lower bound；

Formula 7) it is load abatement or Demand Side Response constraint, P_Lcur,i、Respectively i-th node load reduction and Load cuts down the upper limit；

Formula 8) it is wind-powered electricity generation, the functional relation that photovoltaic is idle to exert oneself with active power output, this formula is only for intermittent DG using permanent Power factor controlling, formula 4) in idle scope for intermittence DG is active, idle uneoupled control；

Formula 9) it is network topology constraint, wherein g during network reconfiguration_kIt is the new network structure for being formed of reconstruct, G is all energy Meet radial and without isolated node the network topology structure collection that load is powered.