CN104537258A - Cone optimization modeling method for allowing distributed stored energy to participate in running adjustment of active power distribution network - Google Patents

Abstract

A cone optimization modeling method for allowing distributed stored energy to participate in running adjustment of an active power distribution network comprises the steps that (1), element parameters of the active power distribution network to be adjusted in an optimization mode are read; (2), according to the parameters provided by the step (1), a time sequence optimization model is established for allowing the distributed stored energy to participate in running adjustment of the active power distribution network, wherein an objective function with the minimum active loss is set, and the running constraint of the power distribution network and the running constraint of an energy storage system are each considered; (3), according to the standard form min {cTx/Ax=b, x<K} for second-order cone optimization, cone model conversion is performed on the time sequence optimization model in the step (2), wherein linearization is performed on the running constrain of the objective function and the running constraint of the active power distribution network, cone conversion is performed on the capacity constraint of stored energy inverters, and a non-linear rotating cone constraint is introduced. The complexity degree of an optimization model function relation is greatly lowered, and meanwhile the requirements for rapid convergence and optimal solving are met.

Description

Distributed energy storage participates in the cone Optimization Modeling method of active power distribution network runing adjustment

Technical field

The present invention relates to a kind of modeling method of active power distribution network traffic control.Particularly relate to a kind of cone Optimization Modeling method that distributed energy storage participates in active power distribution network runing adjustment.

Background technology

The electrical energy production of conventional electric power system, conveying, distribution and consumption have extremely strong simultaneity, due to the storage of large-scale electric energy cannot be realized, namely the time of running of electrical network is in namely to send out uses state, and the electrical network regulating power that generator relies on the inertia of self to provide also is extremely limited, thus result in the appearance of the larger peak-valley difference produced by part throttle characteristics.In the peak of power consumption period, heavier load uncertainty is transported to user from generator, and the equipment such as the circuit in system, transformer are all in higher load factor, make the corresponding increase of network loss; In the low power consumption period, generator is in light condition, reduces the generating efficiency of generator itself.In recent years, along with the continuous increase of electricity needs, petering out of fossil energy and going from bad to worse of environmental problem, be that distributed power generation (distributed generation, the DG) technology of core obtains international very big concern and development with renewable energy utilization.But, the distributed power source being representative with wind-force, photovoltaic generation for electric energy conversion and export time there is obvious randomness and intermittence, to distribution system normally run bring much more probabilistic while, propose new challenge also to conventional electrical distribution network operation mode and dispatching method.Power distribution network not containing DG is " passive ", and the electric energy that access user uses is provided by upper level power transmission network, when power distribution network access DG produces bi-directional current, claims this system for " active distribution system ".Active distribution system possesses the complicated distribution that combination controls various distributed energy (distributed energy resource, DER, as DG, controllable burden, energy storage etc.) ability.

Extensive energy storage technology, as a kind of means realizing stored energy and power bi-directional and regulate, obtains application in distribution system aspect at present, and it is that the operational management of power distribution network provides new thinking of development.To power distribution network itself; the access of Large Copacity, high-efficiency energy-storage device can realize the scale storage of electric energy; bring subversiveness to the constraint of Real-time Balancing between generating and load to break through; and then effectively can alleviate the need for electricity of peak load; simultaneously also can margin capacity in minimizing system, thus improve the economy of system cloud gray model.And during towards application scenarioss such as regenerative resources, the randomness that one side accumulator system can effectively suppress renewable energy system to be exerted oneself and undulatory property, improve the safety and reliability that distribution system is run; On the other hand accumulator system also can improve the receiving ability of distribution system to blower fan, photovoltaic distributed power supply, reduces because primary energy burns the discharge of the pollutant such as carbon dioxide, sulphuric dioxide produced, thus obtains huge environmental benefit.Therefore, how to utilize accumulator system to participate in runing adjustment and the optimization of distribution trend, giving full play to accumulator system and exert oneself for peak load shifting, smooth distribution formula power supply and improve the effect of system running state, is problem demanding prompt solution.

What distributed energy storage participated in active power distribution network runing adjustment problem is in the nature mathematical optimization problem, although be not quite similar for the optimization aim of the existing model of this problem, but the constraint condition considered mainly comprises trend constraint, branch current retrains, the operation constraint of the systems such as voltage level restraint self and the operation constraint of distributed energy storage, and solving of this kind of problem often needs to take a large amount of computing times, this be due to: first, for the optimization of power distribution network, its objective function non-convex function often, and relate to a large amount of equation about control variable and state variable and inequality constrain condition in Optimized model, with there is again complicated funtcional relationship between variations per hour, therefore the runing adjustment of power distribution network belongs to complicated extensive non-convex nonlinear optimal problem, secondly, for the optimization of distributed energy storage, it has obvious temporal characteristics, its running optimizatin is no longer confined to single time discontinuity surface, but to expand to when there are temporal aspect multiple on discontinuity surface, and then cause its decision variable dimension with the swift growth of sequential section number, thus further increase the calculating scale of active power distribution network running optimizatin problem.Above-mentioned two factors result in the complicated and scale that distributed energy storage participates in active power distribution network runing adjustment problem jointly.Therefore, be badly in need of a kind of accurately, the computation model of the above-mentioned optimization problem of rapid solving and algorithm.

Summary of the invention

Technical matters to be solved by this invention is, provides a kind of distributed energy storage that can fast, accurately solve to participate in the cone Optimization Modeling method of active power distribution network runing adjustment.

The technical solution adopted in the present invention is: a kind of distributed energy storage participates in the cone Optimization Modeling method of active power distribution network runing adjustment, comprises the steps:

1) according to the active power distribution network of adjustment to be optimized, read the primary element parameter in active power distribution network, network topology annexation, distributed power source on-position, type and capacity, the initial value of distributed energy storage on-position, inverter rated power, state-of-charge and operation limit value, load and distributed power source operation characteristic prediction curve, system reference voltage and reference power;

2) according to step 1) the active power distribution network parameter that provides sets up the timing optimization model that distributed energy storage participates in active power distribution network runing adjustment problem, comprise: choosing root node is balance node, the active loss of setting active power distribution network is minimum is objective function, and consider the constraint of active power distribution network trend respectively, operation voltage level retrains, branch current retrains, energy storage inverter capacity-constrained, energy storage inverter charge-discharge electric power retrains, energy storage charge state consecutive variations retrains, energy storage charge state runs constraint, the condition of optimization cycle energy storage charge state at whole story equated constraint,

3) according to the canonical form min{c that second order cone is optimized ^tx|Ax=b, x ∈ Κ }, to step 2) described in the distributed energy storage timing optimization model that participates in active power distribution network runing adjustment problem carry out Based On The Conic Model conversion, wherein, c, A, b are constant, and K is the cartesian product of limited non-NULL point convex cone, uses rotating cone represent, described Based On The Conic Model transforms and comprises: carry out linearization to the minimum objective function of active power distribution network active loss, the constraint of active power distribution network trend, operation voltage level constraint, branch current constraint, carry out cone to energy storage inverter capacity-constrained to transform, and introduce the constraint of non-linear rotating cone, thus obtain the minimum objective function of active power distribution network active loss after transforming, transform after the constraint of active power distribution network trend, transform after operation voltage level constraint, transform after branch current constraint and the energy storage inverter capacity-constrained after transforming;

By step 2) in provide be Linear Constraints energy storage inverter charge-discharge electric power constraint, energy storage charge state consecutive variations retrains, energy storage charge state runs constraint and distributed energy storage optimization cycle state-of-charge at whole story equated constraint, the objective function that active power distribution network active loss after described conversion is minimum, active power distribution network trend constraint after conversion, operation voltage level constraint after conversion, branch current constraint after conversion and the energy storage inverter capacity-constrained after transforming, and the common cone Optimized model forming distributed energy storage participation active power distribution network runing adjustment of the non-linear rotating cone constraint introduced,

4) cone is utilized to optimize software for calculation to step 3) the cone Optimized model that obtains is optimized and solves.

Step 2) described in distributed energy storage participate in the timing optimization model of active power distribution network runing adjustment problem specifically:

(1) the minimum objective function of the active power distribution network active loss described in is expressed as:

$\min \underset{t=0}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{P}_i\left(t\right)\mathrm{\&Delta;t}---\left(1\right)$

In formula, T is the running optimizatin cycle, and Δ t is the material calculation in the running optimizatin cycle, and n is system node number; P _it active power sum that () injects for t node i place, the equality constraint of the about intrafascicular effective power flow of the active power distribution network trend provided with following formula represents;

(2) the active power distribution network trend constraint representation described in is:

$\left\{\begin{array}{c}{P}_i\left(t\right)=G_{\mathrm{ii}}{V}_i{\left(t\right)}^2+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}{V}_i\left(t\right)V_j\left(t\right)[G_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\\ =P_i^{\mathrm{DG}}\left(t\right)+P_i^{\mathrm{ESS}}\left(t\right)+P_i^{\mathrm{LD}}\left(t\right),i=2,...,n\\ Q_i\left(t\right)=-B_{\mathrm{ii}}{V}_i{\left(t\right)}^2-\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}{V}_i\left(t\right)V_j\left(t\right)[B_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)-G_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\\ =Q_i^{\mathrm{DG}}\left(t\right)+Q_i^{\mathrm{ESS}}\left(t\right)+Q_i^{\mathrm{LD}}\left(t\right),i=2,...,n\end{array}\right.---\left(2\right)$

In formula, the set of the adjacent node that Ω (i) is node i; V _i(t), V _j(t) and θ _ijt () is respectively t node i, the voltage magnitude of j and phase angle difference; G _ii, B _ii, G _ijand B _ijbe respectively the self-conductance in bus admittance matrix, from susceptance, transconductance and mutual susceptance; Q _it reactive power sum that () injects for t node i place; be respectively active power and the reactive power of the injection of t node i place distributed power source, load and distributed energy storage;

(3) the operation voltage level constraint representation described in is:

V _imin≤V _i(t)≤V _imax,i＝1,…,n (3)

In formula, V _imaxand V _iminbe respectively the bound of node i voltage magnitude;

(4) the branch current constraint representation described in is:

$\begin{array}{c}{I}_{\mathrm{ij}}{\left(t\right)}^2=(G_{\mathrm{ij}}^2+B_{\mathrm{ij}}^2)[V_i{\left(t\right)}^2+V_j{\left(t\right)}^2-2V_i\left(t\right)V_j\left(t\right)\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\&le;{I_{\mathrm{ij}\max }}^2\\ i=1,...,n,j\&Element;\mathrm{\&Omega;}\left(i\right)\end{array}---\left(4\right)$

In formula, I _ijt () is the current amplitude of t branch road ij, I _ijmaxit is the current amplitude upper limit of branch road ij;

(5) the energy storage inverter capacity-constrained described in is expressed as:

$\sqrt{P_k^{\mathrm{ESS}}{\left(t\right)}^2+Q_k^{\mathrm{ESS}}{\left(t\right)}^2}\&le;S_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(5\right)$

In formula, Ω _eSSfor the set of distributed energy storage system; be respectively active power and the reactive power of the energy storage inverter output of t kth; for the rated capacity of a kth energy storage inverter;

(6) the energy storage inverter charge-discharge electric power constraint representation described in is:

$\left\{\begin{array}{c}-P_{k\max }^{\mathrm{ESS}}\&le;P_k^{\mathrm{ESS}}\left(t\right)\&le;P_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}\\ -Q_{k\max }^{\mathrm{ESS}}\&le;Q_k^{\mathrm{ESS}}\left(t\right)\&le;Q_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}\end{array}\right.---\left(6\right)$

In formula, be respectively a kth energy storage inverter active power and the reactive power discharge and recharge upper limit;

(7) the energy storage charge state consecutive variations constraint described in can be expressed as:

$E_k^{\mathrm{ESS}}(t+\mathrm{\&Delta;t})-E_k^{\mathrm{ESS}}\left(t\right)=P_k^{\mathrm{ESS}}\left(t\right)\mathrm{\&Delta;t},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(7\right)$

In formula, for the state-of-charge of a t kth distributed energy storage;

(8) energy storage charge state described in runs constraint representation:

$E_{k\min }^{\mathrm{ESS}}\&le;E_k^{\mathrm{ESS}}\left(t\right)\&le;E_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(8\right)$

In formula, be respectively the operation limit value of a kth distributed energy storage state-of-charge;

(9) distributed energy storage optimization cycle state-of-charge at the whole story equated constraint described in is expressed as:

$E_k^{\mathrm{ESS}}\left(0\right)=E_k^{\mathrm{ESS}}\left(T\right),k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(9\right)$

In formula, be respectively the state-of-charge in a kth distributed energy storage optimization cycle moment at the whole story.

Step 2) set up the timing optimization model that distributed energy storage participates in active power distribution network runing adjustment problem, not only consider charge-discharge electric power and state-of-charge operation constraint from discontinuity surface time single, and consider continuity and the sequential relationship of state-of-charge change between adjacent time section, and the service requirement that optimization cycle state-of-charge at the whole story is equal.

Step 3) described in Based On The Conic Model transform, specifically:

First, the mode of being replaced by variable is to step 2) the minimum objective function of the active power distribution network active loss that provides carries out linearization, namely utilizes $\left\{\begin{array}{c}{X}_i\left(t\right)=V_i{\left(t\right)}^2/\sqrt{2}\\ Y_{\mathrm{ij}}\left(t\right)=V_i\left(t\right)V_j\left(t\right)\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\\ Z_{\mathrm{ij}}\left(t\right)=V_i\left(t\right)V_j\left(t\right)\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\end{array}\right.$ By the V in objective function _i(t), V _j(t), θ _ijt the non-linear form of () sum of products trigonometric function is replaced, obtain the objective function that the active power distribution network active loss after transforming is minimum:

$\min \underset{t=0}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\{\sqrt{2}{G}_{\mathrm{ii}}{X}_i\left(t\right)+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[G_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)\left]\right\}\mathrm{\&Delta;t}---\left(10\right);$

Secondly, to step 2) provide same containing V _i(t), V _j(t), θ _ijthe constraint condition of (t) variable: active power distribution network trend retrains, operation voltage level retrains and branch current constraint converts accordingly, the branch current constraint after the operation voltage level obtained respectively after the active power distribution network trend constraint after transforming, conversion retrains and transforms:

$\left\{\begin{array}{c}{P}_i\left(t\right)=\sqrt{2}{G}_{\mathrm{ii}}{X}_i\left(t\right)+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[G_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)]\\ =P_i^{\mathrm{DG}}\left(t\right)+P_i^{\mathrm{ESS}}\left(t\right)+P_i^{\mathrm{LD}}\left(t\right),i=2,...,n\\ Q_i\left(t\right)=-{\sqrt{2}B}_{\mathrm{ii}}{X}_i\left(t\right)-\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[B_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)-G_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)]\\ =Q_i^{\mathrm{DG}}\left(t\right)+Q_i^{\mathrm{ESS}}\left(t\right)+Q_i^{\mathrm{LD}}\left(t\right),i=2,...,n\end{array}\right.---\left(11\right)$

$\frac{V_{i\min }^2}{\sqrt{2}}\&le;X_i\left(t\right)\&le;\frac{V_{i\max }^2}{\sqrt{2}},i=1,...,n---\left(12\right)$

$\begin{array}{c}{I}_{\mathrm{ij}}{\left(t\right)}^2=(G_{\mathrm{ij}}^2+B_{\mathrm{ij}}^2)[\sqrt{2}{X}_i\left(t\right)+\sqrt{2}{X}_j\left(t\right)-{2Y}_{\mathrm{ij}}\left(t\right)]\&le;{I_{\mathrm{ij}\max }}^2\\ i=1,...,n,j\&Element;\mathrm{\&Omega;}\left(i\right)\end{array}---\left(13\right)$

Then, to step 2) the Nonlinear Constraints energy storage inverter capacity-constrained that provides carries out formal argument, and make it the constraint requirements meeting rotating cone K, obtain the energy storage inverter capacity-constrained after transforming:

$2\frac{S_{i\max }^{\mathrm{ESS}}}{\sqrt{2}}\frac{S_{i\max }^{\mathrm{ESS}}}{\sqrt{2}}\&GreaterEqual;P_i^{\mathrm{ESS}}{\left(t\right)}^2+Q_i^{\mathrm{ESS}}{\left(t\right)}^2,i\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(14\right),$

Step 3) described in the constraint of non-linear rotating cone:

2X _i(t)X _j(t)≥Y _ij(t) ²+Z _ij(t) ²,i＝1,…,n,j∈Ω(i) (15)。

Step 3) by the introducing of the linearization of objective function, the linearization of constraint condition and rotating cone constraint condition, will with V _i(t), θ _ij(t) and mathematical model for decision variable is carried out equivalence and is transformed, and defines with X _i(t), Y _ij(t), Z _ij(t), distributed energy storage for decision variable participates in the cone Optimized model of active power distribution network runing adjustment, makes the nonlinear optimal problem of original function relation complexity be converted into second order cone optimization problem and solves.

Distributed energy storage of the present invention participates in the cone Optimization Modeling method of active power distribution network runing adjustment, greatly simplifies the complexity of Optimized model funtcional relationship, has graceful cone geometry concurrently simultaneously, can ensure optimization problem quick, accurately solve.The present invention is in distributed energy storage Optimization Modeling, take into full account the running boundary of energy-storage units and inverter thereof, the quick control characteristic of inverter is not only considered from discontinuity surface time single, and sequential, discontinuity surface time each is carried out unified Modeling, thus the running optimizatin of distributed energy storage is made to form an organic whole between discontinuity surface, adjacent time section and on whole optimization cycle when single.Cone optimization method of the present invention can carry out Unify legislation to Power Flow Problem and distributed energy storage running optimizatin problem, the estimate simultaneously of complicated nonlinear optimal problem and high dimensional nonlinear system of equations is achieved, avoid loaded down with trivial details iteration and a large amount of tests, computing velocity has and promotes significantly.And, because bore the geometry of the grace had and special processing mode, the optimality of the solution of institute's Solve problems can be ensured, apply it to distributed energy storage and participate in, in the optimization problem of power distribution network runing adjustment, optimum system cloud gray model scheme can being obtained.Visible, the requirement that cone optimization method can meet Fast Convergent simultaneously and accurately solve.

Accompanying drawing explanation

Fig. 1 is IEEE 33 node example and distributed power source, distributed energy storage on-position figure;

Fig. 2 is the cone Optimization Modeling method flow diagram that a kind of distributed energy storage of the present invention participates in active power distribution network runing adjustment;

Fig. 3 is load and distributed power source operation characteristic prediction curve;

Fig. 4 a is distributed energy storage charge-discharge electric power optimum results;

Fig. 4 b is distributed energy storage state-of-charge result of variations;

Fig. 5 a is 16 Nodes distributed energy storage prioritization scheme comparison diagrams under different model;

Fig. 5 b is 32 Nodes distributed energy storage prioritization scheme comparison diagrams under different model.

Embodiment

Below in conjunction with embodiment and accompanying drawing, the cone Optimization Modeling method that distributed energy storage of the present invention participates in active power distribution network runing adjustment is described in detail.

Distributed energy storage of the present invention participates in the cone Optimization Modeling method of active power distribution network runing adjustment, for containing in the distribution system running optimizatin research of distributed power source and distributed energy storage, the cone Optimization Softwares such as MOSEK, CPLEX can be adopted to carry out simulated implementation.The present invention adopts MOSEK software, with IEEE 33 bus test system shown in Fig. 1 for embodiment.

Distributed energy storage of the present invention participates in the cone Optimization Modeling method of active power distribution network runing adjustment, as shown in Figure 2, comprises the steps:

1) according to the active power distribution network of adjustment to be optimized, read the primary element parameter in active power distribution network, network topology annexation, distributed power source on-position, type and capacity, distributed energy storage on-position, inverter rated power, state-of-charge (state of charge, SOC) initial value and operation limit value, load and distributed power source operation characteristic prediction curve, system reference voltage and reference power etc.;

For the present embodiment, first read the resistance value of circuit element in IEEE 33 node system, the active power of load cell, reactive power, network topology annexation; Secondly, the on-position of setting 4 550kVA Wind turbines is node 13,18,31,33, and the on-position of 2 600kVA photovoltaic systems is node 15,30; Again, the on-position of setting 2 distributed energy storage systems is node 16,32, the rated capacity of the two inverter is 500kVA, rated power is 400kW, specified energy storage capacity is respectively 1600kWh and 800kWh, state-of-charge initial value is respectively 50.0% and 12.5%, and state-of-charge runs limit value and is respectively 6.25%/87.5% and 6.25%/95.0%; Then, in units of sky, utilize load forecasting method to simulate the day operation curve of load and wind-powered electricity generation, photovoltaic, as shown in Figure 3; Finally, the reference voltage arranging system is 12.66kV, reference power is 100MVA.

2) according to step 1) the active power distribution network parameter that provides sets up the timing optimization model that distributed energy storage participates in active power distribution network runing adjustment problem, comprise: choosing root node is balance node, the active loss of setting active power distribution network is minimum is objective function, and consider the constraint of active power distribution network trend respectively, operation voltage level retrains, branch current retrains, energy storage inverter capacity-constrained, energy storage inverter charge-discharge electric power retrains, energy storage charge state (SOC) consecutive variations retrains, energy storage charge state runs constraint, the condition of optimization cycle energy storage charge state at whole story equated constraint etc.,

Described distributed energy storage participates in the timing optimization model of active power distribution network runing adjustment problem specifically:

(1) the minimum objective function of the active power distribution network active loss described in can be expressed as:

$\min \underset{t=0}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{P}_i\left(t\right)\mathrm{\&Delta;t}---\left(1\right)$

In formula, T is the running optimizatin cycle, and Δ t is the material calculation in the running optimizatin cycle, and n is system node number; P _it active power sum that () injects for t node i place, the equality constraint of the about intrafascicular effective power flow of active power distribution network trend that available following formula (2) provides represents;

(2) the active power distribution network trend constraint described in can be expressed as:

$\left\{\begin{array}{c}{P}_i\left(t\right)=G_{\mathrm{ii}}{V}_i{\left(t\right)}^2+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}{V}_i\left(t\right)V_j\left(t\right)[G_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\\ =P_i^{\mathrm{DG}}\left(t\right)+P_i^{\mathrm{ESS}}\left(t\right)+P_i^{\mathrm{LD}}\left(t\right),i=2,...,n\\ Q_i\left(t\right)=-B_{\mathrm{ii}}{V}_i{\left(t\right)}^2-\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}{V}_i\left(t\right)V_j\left(t\right)[B_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)-G_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\\ =Q_i^{\mathrm{DG}}\left(t\right)+Q_i^{\mathrm{ESS}}\left(t\right)+Q_i^{\mathrm{LD}}\left(t\right),i=2,...,n\end{array}\right.---\left(2\right)$

In formula, the set of the adjacent node that Ω (i) is node i; V _i(t), V _j(t) and θ _ijt () is respectively t node i, the voltage magnitude of j and phase angle difference; G _ii, B _ii, G _ijand B _ijbe respectively the self-conductance in bus admittance matrix, from susceptance, transconductance and mutual susceptance; Q _it reactive power sum that () injects for t node i place; be respectively active power and the reactive power of the injection of t node i place distributed power source, load and distributed energy storage;

(3) the operation voltage level constraint described in can be expressed as:

V _imin≤V _i(t)≤V _imax,i＝1,…,n (3)

In formula, V _imaxand V _iminbe respectively the bound of node i voltage magnitude;

(4) the branch current constraint described in can be expressed as:

$\begin{array}{c}{I}_{\mathrm{ij}}{\left(t\right)}^2=(G_{\mathrm{ij}}^2+B_{\mathrm{ij}}^2)[V_i{\left(t\right)}^2+V_j{\left(t\right)}^2-2V_i\left(t\right)V_j\left(t\right)\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\&le;{I_{\mathrm{ij}\max }}^2\\ i=1,...,n,j\&Element;\mathrm{\&Omega;}\left(i\right)\end{array}---\left(4\right)$

In formula, I _ijt () is the current amplitude of t branch road ij, I _ijmaxit is the current amplitude upper limit of branch road ij;

(5) the energy storage inverter capacity-constrained described in can be expressed as:

$\sqrt{P_k^{\mathrm{ESS}}{\left(t\right)}^2+Q_k^{\mathrm{ESS}}{\left(t\right)}^2}\&le;S_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(5\right)$

In formula, Ω _eSSfor the set of distributed energy storage system; be respectively active power and the reactive power of the energy storage inverter output of t kth; for the rated capacity of a kth energy storage inverter;

(6) the energy storage inverter charge-discharge electric power constraint described in can be expressed as:

$\left\{\begin{array}{c}-P_{k\max }^{\mathrm{ESS}}\&le;P_k^{\mathrm{ESS}}\left(t\right)\&le;P_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}\\ -Q_{k\max }^{\mathrm{ESS}}\&le;Q_k^{\mathrm{ESS}}\left(t\right)\&le;Q_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}\end{array}\right.---\left(6\right)$ In formula, be respectively a kth energy storage inverter active power and the reactive power discharge and recharge upper limit;

(7) the energy storage charge state consecutive variations constraint described in can be expressed as:

$E_k^{\mathrm{ESS}}(t+\mathrm{\&Delta;t})-E_k^{\mathrm{ESS}}\left(t\right)=P_k^{\mathrm{ESS}}\left(t\right)\mathrm{\&Delta;t},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(7\right)$

In formula, for the state-of-charge of a t kth distributed energy storage;

(8) energy storage charge state described in is run constraint and can be expressed as:

$E_{k\min }^{\mathrm{ESS}}\&le;E_k^{\mathrm{ESS}}\left(t\right)\&le;E_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(8\right)$

In formula, be respectively the operation limit value of a kth distributed energy storage state-of-charge;

(9) distributed energy storage optimization cycle state-of-charge at the whole story equated constraint described in can be expressed as:

$E_k^{\mathrm{ESS}}\left(0\right)=E_k^{\mathrm{ESS}}\left(T\right),k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(9\right)$

In formula, be respectively the state-of-charge in a kth distributed energy storage optimization cycle moment at the whole story.

Set up the timing optimization model that distributed energy storage participates in active power distribution network runing adjustment problem, not only consider charge-discharge electric power and state-of-charge operation constraint from discontinuity surface time single, and consider continuity and the sequential relationship of state-of-charge change between adjacent time section, and the service requirement that optimization cycle state-of-charge at the whole story is equal.

3) according to the canonical form min{c that second order cone is optimized ^tx|Ax=b, x ∈ Κ }, to step 2) described in the distributed energy storage timing optimization model that participates in active power distribution network runing adjustment problem carry out Based On The Conic Model conversion, wherein, c, A, b are constant, and K is the cartesian product of limited non-NULL point convex cone, uses rotating cone represent, described Based On The Conic Model transforms and comprises: carry out linearization to the minimum objective function of active power distribution network active loss, the constraint of active power distribution network trend, operation voltage level constraint, branch current constraint, carry out cone to energy storage inverter capacity-constrained to transform, and introduce the constraint of non-linear rotating cone, thus obtain the minimum objective function of active power distribution network active loss after transforming, transform after the constraint of active power distribution network trend, transform after operation voltage level constraint, transform after branch current constraint and the energy storage inverter capacity-constrained after transforming;

By step 2) in provide be Linear Constraints energy storage inverter charge-discharge electric power constraint, energy storage charge state consecutive variations retrains, energy storage charge state runs constraint and distributed energy storage optimization cycle state-of-charge at whole story equated constraint, the objective function that active power distribution network active loss after described conversion is minimum, active power distribution network trend constraint after conversion, operation voltage level constraint after conversion, branch current constraint after conversion and the energy storage inverter capacity-constrained after transforming, and the common cone Optimized model forming distributed energy storage participation active power distribution network runing adjustment of the non-linear rotating cone constraint introduced,

Step 3) described in Based On The Conic Model transform, specifically:

First, the mode of being replaced by variable is to step 2) the minimum objective function of the active power distribution network active loss that provides of formula (1) carries out linearization, namely utilizes $\left\{\begin{array}{c}{X}_i\left(t\right)=V_i{\left(t\right)}^2/\sqrt{2}\\ Y_{\mathrm{ij}}\left(t\right)=V_i\left(t\right)V_j\left(t\right)\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\\ Z_{\mathrm{ij}}\left(t\right)=V_i\left(t\right)V_j\left(t\right)\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\end{array}\right.$ By the V in objective function _i(t), V _j(t), θ _ijt the non-linear form of () sum of products trigonometric function is replaced, obtain the objective function that the active power distribution network active loss after transforming is minimum:

$\min \underset{t=0}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\{\sqrt{2}{G}_{\mathrm{ii}}{X}_i\left(t\right)+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[G_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)\left]\right\}\mathrm{\&Delta;t}---\left(10\right);$

Secondly, to step 2) provide same containing V _i(t), V _j(t), θ _ijconstraint equation (2) ~ (4) of (t) variable: the constraint of active power distribution network trend, operation voltage level constraint and branch current constraint convert accordingly, obtain the operation voltage level constraint after the active power distribution network trend constraint after transforming, conversion and the constraint of the branch current after transforming respectively, shown in (11) ~ (13):

$\left\{\begin{array}{c}{P}_i\left(t\right)=\sqrt{2}{G}_{\mathrm{ii}}{X}_i\left(t\right)+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[G_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)]\\ =P_i^{\mathrm{DG}}\left(t\right)+P_i^{\mathrm{ESS}}\left(t\right)+P_i^{\mathrm{LD}}\left(t\right),i=2,...,n\\ Q_i\left(t\right)=-{\sqrt{2}B}_{\mathrm{ii}}{X}_i\left(t\right)-\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[B_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)-G_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)]\\ =Q_i^{\mathrm{DG}}\left(t\right)+Q_i^{\mathrm{ESS}}\left(t\right)+Q_i^{\mathrm{LD}}\left(t\right),i=2,...,n\end{array}\right.---\left(11\right)$

$\frac{V_{i\min }^2}{\sqrt{2}}\&le;X_i\left(t\right)\&le;\frac{V_{i\max }^2}{\sqrt{2}},i=1,...,n---\left(12\right)$

$\begin{array}{c}{I}_{\mathrm{ij}}{\left(t\right)}^2=(G_{\mathrm{ij}}^2+B_{\mathrm{ij}}^2)[\sqrt{2}{X}_i\left(t\right)+\sqrt{2}{X}_j\left(t\right)-{2Y}_{\mathrm{ij}}\left(t\right)]\&le;{I_{\mathrm{ij}\max }}^2\\ i=1,...,n,j\&Element;\mathrm{\&Omega;}\left(i\right)\end{array}---\left(13\right)$

Then, to step 2) Nonlinear Constraints energy storage inverter capacity-constrained formula (5) that provides carries out formal argument, make it the constraint requirements meeting rotating cone K, obtain the energy storage inverter capacity-constrained after transforming, shown in (14):

$2\frac{S_{i\max }^{\mathrm{ESS}}}{\sqrt{2}}\frac{S_{i\max }^{\mathrm{ESS}}}{\sqrt{2}}\&GreaterEqual;P_i^{\mathrm{ESS}}{\left(t\right)}^2+Q_i^{\mathrm{ESS}}{\left(t\right)}^2,i\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(14\right)$

And formula (6) ~ (9) are Linear Constraints, meet the canonical form that second order cone is optimized, without the need to transforming;

Step 3) described in the constraint of introducing non-linear rotating cone:

2X _i(t)X _j(t)≥Y _ij(t) ²+Z _ij(t) ²,i＝1,…,n,j∈Ω(i) (15)

Now, the objective function in above-mentioned formula (6) ~ (13) and constraint condition are variable X _i(t), Y _ij(t), Z _ij(t) and linear function form, formula (14), (15) are rotating cone constraint type, can meet second order cone optimize canonical form.

Step 3) by the introducing of the linearization of objective function, the linearization of constraint condition and rotating cone constraint condition, will with V _i(t), θ _ij(t) and mathematical model for decision variable is carried out equivalence and is transformed, and defines with X _i(t), Y _ij(t), Z _ij(t), distributed energy storage for decision variable participates in the cone Optimized model of active power distribution network runing adjustment, makes the nonlinear optimal problem of original function relation complexity be converted into second order cone optimization problem and solves.

4) utilize cone to optimize software for calculation to step 3) the cone Optimized model that obtains is optimized and solves, and utilizes GAMS CONOPT solver to step 2) in basic model be optimized and solve.

5) export the optimum results of cone Optimized model, and the result of calculation of cone Optimized model and basic model is compared checking.

Distributed energy storage participates in the prioritization scheme of active power distribution network runing adjustment as shown in Figure 4; Solve the optimum results of cone Optimized model and basic model to such as shown in Fig. 5 (a), 5 (b).

Observation Fig. 4 is known, and distributed energy storage can play an active part in the adjustment of the active power distribution network electric energy equilibrium of supply and demand.For the distributed energy storage at node 32 place, the access of wind-powered electricity generation and photovoltaic makes system power stream generation larger fluctuation, exert oneself at distributed power source comparatively large/little, need for electricity is less/larger time, electric energy is carried/absorbed to accumulator system to electrical network, and reach complete period maximum charge-discharge electric power in disparities between supply and demand great 2:30,16:30 moment.But, the charging and discharging state of Fig. 4 and size not with the electric energy supplydemand relationship one_to_one corresponding shown in Fig. 2, this is because distributed energy storage is owing to being limited to the constraint of its inverter rated power and SOC operation limit value, in the long-play participating in power distribution network is optimized, it not merely with the electrical energy demands of discontinuity surface time single for rely on, but based on the operating condition of whole optimization cycle, namely the balance of grid generation and load electricity consumption is realized first on the whole, and then get down to its charging and discharging state of local modulation and size, thus play its vital role that global energy is managed to greatest extent.

From Fig. 5 (a), 5 (b), the optimum results of cone Optimized model is consistent with the optimum results of former basic model, demonstrates cone Optimization Modeling method rationality and correctness.In addition, enter to compare to the counting yield solving two models, result display cone Optimized model solve more rapidly and efficiently, as shown in table 1.

Solution efficiency comparative result under the different model of table 1

Distributed energy storage of the present invention participates in the cone Optimization Modeling method of active power distribution network runing adjustment, based on the timing optimization that power distribution network on discontinuity surface time multiple runs, establish to minimize the whole network active loss for objective function, consider system load flow constraint, operation voltage level retrains, branch current retrains, energy storage inverter capacity-constrained, energy storage inverter charge-discharge electric power retrains, energy storage SOC consecutive variations retrains, energy storage SOC runs constraint, the cone Optimized model of the conditions such as optimization cycle energy storage at whole story SOC equated constraint.When model conversation, first, according to the canonical form that second order cone is optimized, (non-NULL point convex cone imports under partial order, the problem of the linear objective function under linear equality, linear inequality constraint condition), objective function is carried out linearization by the mode of being replaced by variable, carries out linearization to the constraint condition of distribution system self simultaneously; Then, the non-linear form for energy storage inverter capacity-constrained converts, and makes it the requirement meeting rotating cone constraint type; Finally, introduce rotating cone constraint according to the funtcional relationship of new variables, form cone Optimized model.Wherein, rotating cone constraint ensure that the consistance of cone Optimized model and master mould.Compared with the non-convex nonlinear model participating in active power distribution network runing adjustment with distributed energy storage in the past, the objective function of cone Optimized model is linear, and its feasible zone is retrained by linear equality, inequality and non-linear rotating cone to form, thus search volume is limited within the scope of closed convex cone, search volume is made to have certain slickness, closure and symmetry, greatly simplifie the complexity of Optimized model funtcional relationship, have graceful cone geometry concurrently simultaneously, can ensure optimization problem quick, accurately solve.

In distributed energy storage Optimization Modeling, take into full account the running boundary of energy-storage units and inverter thereof.Wherein, the exchange that inverter is mainly used in charging and discharging state and power controls, and its running boundary mainly considers the rated capacity constraint of energy storage inverter, and meritorious and reactive power discharge and recharge constraint; Energy-storage units is mainly used in the storage of electric charge, its running boundary mainly considers the cyclophysis that the sequential relationship of SOC change, the operation limit value of SOC and SOC change, and is namely presented as that SOC meets with the consecutive variations of charge-discharge electric power, SOC the constraint condition that limit value requires and optimization cycle SOC at the whole story is equal in sequential respectively.Therefore, the present invention not only considers the quick control characteristic of inverter from discontinuity surface time single for distributed energy storage Optimization Modeling, and sequential, discontinuity surface time each is carried out unified Modeling, thus the running optimizatin of distributed energy storage is made to form an organic whole between discontinuity surface, adjacent time section and on whole optimization cycle when single.

In counting yield, cone optimization method of the present invention can carry out Unify legislation to Power Flow Problem and distributed energy storage running optimizatin problem, the estimate simultaneously of complicated nonlinear optimal problem and high dimensional nonlinear system of equations is achieved, avoid loaded down with trivial details iteration and a large amount of tests, computing velocity has and promotes significantly; On the other hand, because bore the geometry of the grace had and special processing mode, the optimality of the solution of institute's Solve problems can be ensured, apply it to distributed energy storage and participate in, in the optimization problem of power distribution network runing adjustment, optimum system cloud gray model scheme can being obtained.Visible, the requirement that cone optimization method can meet Fast Convergent simultaneously and accurately solve.

Claims (5)

1. distributed energy storage participates in a cone Optimization Modeling method for active power distribution network runing adjustment, it is characterized in that, comprises the steps:

1) according to the active power distribution network of adjustment to be optimized, read the primary element parameter in active power distribution network, network topology annexation, distributed power source on-position, type and capacity, the initial value of distributed energy storage on-position, inverter rated power, state-of-charge and operation limit value, load and distributed power source operation characteristic prediction curve, system reference voltage and reference power;

2) according to step 1) the active power distribution network parameter that provides sets up the timing optimization model that distributed energy storage participates in active power distribution network runing adjustment problem, comprise: choosing root node is balance node, the active loss of setting active power distribution network is minimum is objective function, and consider the constraint of active power distribution network trend respectively, operation voltage level retrains, branch current retrains, energy storage inverter capacity-constrained, energy storage inverter charge-discharge electric power retrains, energy storage charge state consecutive variations retrains, energy storage charge state runs constraint, the condition of optimization cycle energy storage charge state at whole story equated constraint,

3) according to the canonical form min{c that second order cone is optimized ^tx|Ax=b, x ∈ Κ }, to step 2) described in the distributed energy storage timing optimization model that participates in active power distribution network runing adjustment problem carry out Based On The Conic Model conversion, wherein, c, A, b are constant, and K is the cartesian product of limited non-NULL point convex cone, uses rotating cone represent, described Based On The Conic Model transforms and comprises: carry out linearization to the minimum objective function of active power distribution network active loss, the constraint of active power distribution network trend, operation voltage level constraint, branch current constraint, carry out cone to energy storage inverter capacity-constrained to transform, and introduce the constraint of non-linear rotating cone, thus obtain the minimum objective function of active power distribution network active loss after transforming, transform after the constraint of active power distribution network trend, transform after operation voltage level constraint, transform after branch current constraint and the energy storage inverter capacity-constrained after transforming;

By step 2) in provide be Linear Constraints energy storage inverter charge-discharge electric power constraint, energy storage charge state consecutive variations retrains, energy storage charge state runs constraint and distributed energy storage optimization cycle state-of-charge at whole story equated constraint, the objective function that active power distribution network active loss after described conversion is minimum, active power distribution network trend constraint after conversion, operation voltage level constraint after conversion, branch current constraint after conversion and the energy storage inverter capacity-constrained after transforming, and the common cone Optimized model forming distributed energy storage participation active power distribution network runing adjustment of the non-linear rotating cone constraint introduced,

4) cone is utilized to optimize software for calculation to step 3) the cone Optimized model that obtains is optimized and solves.

2. distributed energy storage according to claim 1 participates in the cone Optimization Modeling method of active power distribution network runing adjustment, it is characterized in that, step 2) described in distributed energy storage participate in the timing optimization model of active power distribution network runing adjustment problem specifically:

(1) the minimum objective function of the active power distribution network active loss described in is expressed as:

$\min \underset{t=0}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{P}_i\left(t\right)\mathrm{\&Delta;t}---\left(1\right)$

In formula, T is the running optimizatin cycle, and Δ t is the material calculation in the running optimizatin cycle, and n is system node number; P _it active power sum that () injects for t node i place, the equality constraint of the about intrafascicular effective power flow of the active power distribution network trend provided with following formula represents;

(2) the active power distribution network trend constraint representation described in is:

$\left\{\begin{array}{c}{P}_i\left(t\right)=G_{\mathrm{ii}}{V}_i{\left(t\right)}^2+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}{V}_i\left(t\right)V_j\left(t\right)[G_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\\ =P_i^{\mathrm{DG}}\left(t\right)+P_i^{\mathrm{ESS}}\left(t\right)+P_i^{\mathrm{LD}}\left(t\right),i=2,...,n\\ Q_i\left(t\right)=-B_{\mathrm{ii}}{V}_i{\left(t\right)}^2-\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}{V}_i\left(t\right)V_j\left(t\right)[R_{\mathrm{ij}}\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)-G_{\mathrm{ij}}\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\\ =Q_i^{\mathrm{DG}}\left(t\right)+Q_i^{\mathrm{ESS}}\left(t\right)+Q_i^{\mathrm{LD}}\left(t\right),i=2,...,n\end{array}\right.---\left(2\right)$

In formula, the set of the adjacent node that Ω (i) is node i; V _i(t), V _j(t) and θ _ijt () is respectively t node i, the voltage magnitude of j and phase angle difference; G _ii, B _ii, G _ijand B _ijbe respectively the self-conductance in bus admittance matrix, from susceptance, transconductance and mutual susceptance; Q _it reactive power sum that () injects for t node i place; be respectively active power and the reactive power of the injection of t node i place distributed power source, load and distributed energy storage;

(3) the operation voltage level constraint representation described in is:

V _imin≤V _i(t)≤V _imax,i＝1,…,n (3)

In formula, V _imaxand V _iminbe respectively the bound of node i voltage magnitude;

(4) the branch current constraint representation described in is:

$L_{\mathrm{ij}}{\left(t\right)}^2=(G_{\mathrm{ij}}^2+B_{\mathrm{ij}}^2)[V_i{\left(t\right)}^2+V_j{\left(t\right)}^2-{2V}_i\left(t\right)V_j\left(t\right)\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\left(t\right)]\&le;{L_{\mathrm{ij}\max }}^2$

(4)

i＝1,…,n,j∈Ω(i)

In formula, I _ijt () is the current amplitude of t branch road ij, I _ijmaxit is the current amplitude upper limit of branch road ij

(5) the energy storage inverter capacity-constrained described in is expressed as:

$\sqrt{P_k^{\mathrm{ESS}}{\left(t\right)}^2+Q_k^{\mathrm{ESS}}{\left(t\right)}^2}\&le;S_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(5\right)$

In formula, Ω _eSSfor the set of distributed energy storage system; be respectively active power and the reactive power of the energy storage inverter output of t kth; for the rated capacity of a kth energy storage inverter;

(6) the energy storage inverter charge-discharge electric power constraint representation described in is:

$\left\{\begin{array}{c}-P_{k\max }^{\mathrm{ESS}}\&le;P_k^{\mathrm{ESS}}\left(t\right)\&le;P_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}\\ -Q_{k\max }^{\mathrm{ESS}}\&le;Q_k^{\mathrm{ESS}}\left(t\right)\&le;Q_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}\end{array}\right.---\left(6\right)$

In formula, be respectively a kth energy storage inverter active power and the reactive power discharge and recharge upper limit;

(7) the energy storage charge state consecutive variations constraint described in can be expressed as:

$E_k^{\mathrm{ESS}}(t+\mathrm{\&Delta;t})-E_k^{\mathrm{ESS}}\left(t\right)=P_k^{\mathrm{ESS}}\left(t\right)\mathrm{\&Delta;t},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(7\right)$

In formula, for the state-of-charge of a t kth distributed energy storage;

(8) energy storage charge state described in runs constraint representation:

$E_{k\min }^{\mathrm{ESS}}\&le;E_k^{\mathrm{ESS}}\left(t\right)\&le;E_{k\max }^{\mathrm{ESS}},k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(8\right)$

In formula, be respectively the operation limit value of a kth distributed energy storage state-of-charge;

(9) distributed energy storage optimization cycle state-of-charge at the whole story equated constraint described in is expressed as:

$E_k^{\mathrm{ESS}}\left(0\right)=E_k^{\mathrm{ESS}}\left(T\right),k\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(9\right)$

In formula, be respectively the state-of-charge in a kth distributed energy storage optimization cycle moment at the whole story.

3. distributed energy storage according to claim 1 participates in the cone Optimization Modeling method of active power distribution network runing adjustment, it is characterized in that, step 2) set up the timing optimization model that distributed energy storage participates in active power distribution network runing adjustment problem, not only consider charge-discharge electric power and state-of-charge operation constraint from discontinuity surface time single, and consider continuity and the sequential relationship of state-of-charge change between adjacent time section, and the service requirement that optimization cycle state-of-charge at the whole story is equal.

4. distributed energy storage according to claim 1 participates in the cone Optimization Modeling method of active power distribution network runing adjustment, it is characterized in that, step 3) described in Based On The Conic Model transform, specifically:

First, the mode of being replaced by variable is to step 2) the minimum objective function of the active power distribution network active loss that provides carries out linearization, namely utilizes $\left\{\begin{array}{c}{X}_i\left(t\right)=V_i{\left(t\right)}^2/\sqrt{2},\\ Y_{\mathrm{ij}}\left(t\right)=V_i\left(t\right)V_j\left(t\right)\cos {\mathrm{\&theta;}}_{\mathrm{ij}}\\ Z_{\mathrm{ij}}\left(t\right)=V_i\left(t\right)V_j\left(t\right)\sin {\mathrm{\&theta;}}_{\mathrm{ij}}\end{array}\right.$ By the V in objective function _i(t), V _j(t), θ _ijt the non-linear form of () sum of products trigonometric function is replaced, obtain the objective function that the active power distribution network active loss after transforming is minimum:

$\min \underset{t=0}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\{\sqrt{2}{G}_{\mathrm{ii}}{X}_i\left(t\right)+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[G_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)\left]\right\}\mathrm{\&Delta;t}---\left(10\right);$

Secondly, to step 2) provide same containing V _i(t), V _j(t), θ _ijthe constraint condition of (t) variable: active power distribution network trend retrains, operation voltage level retrains and branch current constraint converts accordingly, the branch current constraint after the operation voltage level obtained respectively after the active power distribution network trend constraint after transforming, conversion retrains and transforms:

$\left\{\begin{array}{c}{P}_i\left(t\right)=\sqrt{2}{G}_{\mathrm{ii}}{X}_i\left(t\right)+\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[G_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)+B_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)]\\ =P_i^{\mathrm{DG}}\left(t\right)+P_i^{\mathrm{ESS}}\left(t\right)+P_i^{\mathrm{LD}}\left(t\right),i=2,...,n\\ Q_i\left(t\right)=-{\sqrt{2}B}_{\mathrm{ii}}{X}_i\left(t\right)-\underset{j\&Element;\mathrm{\&Omega;}\left(i\right)}{\mathrm{\&Sigma;}}[R_{\mathrm{ij}}{Y}_{\mathrm{ij}}\left(t\right)-G_{\mathrm{ij}}{Z}_{\mathrm{ij}}\left(t\right)]\\ =Q_i^{\mathrm{DG}}\left(t\right)+Q_i^{\mathrm{ESS}}\left(t\right)+Q_i^{\mathrm{LD}}\left(t\right),i=2,...,n\end{array}\right.---\left(11\right)$

$\frac{V_{i\min }^2}{\sqrt{2}}\&le;X_i\left(t\right)\&le;\frac{V_{i\max }^2}{\sqrt{2}},i=1,...,n---\left(12\right)$

$L_{\mathrm{ij}}{\left(t\right)}^2=(G_{\mathrm{ij}}^2+B_{\mathrm{ij}}^2)[\sqrt{2}{X}_i\left(t\right)+\sqrt{2}{X}_j\left(t\right)-{2Y}_{\mathrm{ij}}\left(t\right)]\&le;{I_{\mathrm{ij}\max }}^2$

(13)

i＝1,…,n,j∈Ω(i)

Then, to step 2) the Nonlinear Constraints energy storage inverter capacity-constrained that provides carries out formal argument, and make it the constraint requirements meeting rotating cone K, obtain the energy storage inverter capacity-constrained after transforming:

$2\frac{S_{i\max }^{\mathrm{ESS}}}{\sqrt{2}}\frac{S_{i\max }^{\mathrm{ESS}}}{\sqrt{2}}\&GreaterEqual;P_i^{\mathrm{ESS}}{\left(t\right)}^2+Q_i^{\mathrm{ESS}}{\left(t\right)}^2,i\&Element;{\mathrm{\&Omega;}}_{\mathrm{ESS}}---\left(14\right),$

Step 3) described in the constraint of non-linear rotating cone:

2X _i(t)X _j(t)≥Y _ij(t) ²+Z _ij(t) ²,i＝1,…,n,j∈Ω(i) (15)。

5. distributed energy storage according to claim 1 participates in the cone Optimization Modeling method of active power distribution network runing adjustment, it is characterized in that, step 3) by the introducing of the linearization of objective function, the linearization of constraint condition and rotating cone constraint condition, will with V _i(t), θ _ij(t) and mathematical model for decision variable is carried out equivalence and is transformed, and defines with X _i(t), Y _ij(t), Z _ij(t), distributed energy storage for decision variable participates in the cone Optimized model of active power distribution network runing adjustment, makes the nonlinear optimal problem of original function relation complexity be converted into second order cone optimization problem and solves.