CN105552965A - Chance constraint planning based optimal configuration method of distributed energy source - Google Patents

Abstract

The invention provides a chance constraint planning based optimal configuration method of a distributed energy source. The method comprises the following steps of building a comprehensive optimal configuration model of the distributed energy source; determining a chance constraint condition of the comprehensive optimal configuration model of the disturbed energy source; determining an energy storage system configuration principle of the comprehensive optimal configuration module of the distributed energy source; and figuring out the comprehensive optimal configuration model of the distributed energy source. With the method proposed by the invention, the acceptable capability of a power distribution net on the distrusted energy source is effectively improved, the energy efficiency target and the low carbon target of a planning scheme can be met, economic requirement also can be met, the problems of site selection and constant volume of the distributed energy sources in places with different resource levels and economic development degrees can be solved, the application of the distributed energy source at a large scale is ensured, and the environmental pollution and energy crisis caused by fossil energy can be effectively relieved.

Description

A kind of distributed energy Optimal Configuration Method based on chance constrained programming

Technical field

The present invention relates to distribution network planning technology and survey field, be specifically related to a kind of distributed energy Optimal Configuration Method based on chance constrained programming.

Background technology

Along with the extensive access of electric automobile, energy storage, the distributed power source distributed energy, power distribution network is become " active " from " passive ", trend is become " two-way " from " unidirectional ", present " polyphyly " feature of more sophisticated, profound influence will be produced to distribution network load characteristic, Supply Security, reliability and asset utilization ratio etc.On the one hand, distributed energy access power distribution network, produces very important impact by the operation of electric power system with planning: the uncertainty that distributed energy distributes in the time and space, adds the difficulty that operation of power networks controls; The nonlinear characteristic of electric automobile, energy storage etc., can produce harmonic pollution, affects the electrical network quality of power supply; Traditional distribution network planning criterion possibly cannot be applicable to the sight that distributed energy accesses on a large scale, and the extensive access of distributed energy proposes new requirement to distribution network planning.On the other hand, the extensive access of distributed energy brings new opportunities and challenges to grid company, is conducive to grid company developing power sales, expands load control resource, improve electrical network kurtosis, promote electrical network utilization ratio.

At this under the new situation, the Optimal Configuration Method of research distributed energy, not only can be active distribution network planning and foundation is provided, and the economy of initiatively distribution system operation can be ensured, improve entire system efficiency, reduce operation and maintenance expenses use, improve rate of return on investment, for conventional electrical distribution net is to Modern power distribution net upgrading lay a good foundation.And at present, not yet have and effectively can improve the receiving ability optimization method of power distribution network to distributed energy.

Summary of the invention

In view of this, a kind of distributed energy Optimal Configuration Method based on chance constrained programming provided by the invention, the method effectively improves the receiving ability of power distribution network to distributed energy; Energetic efficiency objectives and the low-carbon (LC) target of programme can be met; also cost-effectiveness requirement can be met; different resource level, the addressing of the economic development level Area distribution formula energy and constant volume problem can be solved; guarantee the large-scale application of distributed energy, the environmental pollution that effective alleviation fossil energy causes and energy crisis.

The object of the invention is to be achieved through the following technical solutions:

Based on a distributed energy Optimal Configuration Method for chance constrained programming, described method comprises the steps:

Step 1. sets up distributed energy complex optimum allocation models;

Step 2. determines the constraints condition of opportunity of described distributed energy complex optimum allocation models;

Step 3. determines the energy-storage system equipping rules of described distributed energy complex optimum allocation models;

Step 4. solves described distributed energy complex optimum allocation models.

Preferably, described step 1 comprises:

1-1. sets up distributed energy complex optimum allocation models:

$\left\{\begin{array}{c}{\mathrm{minZ}}_{\cos t}=\mathrm{\ω}\·C^T\\ \mathrm{\ω}=\[{\mathrm{\ω}}_1,{\mathrm{\ω}}_2,{\mathrm{\ω}}_3,{\mathrm{\ω}}_4,{\mathrm{\ω}}_5\]\\ C=\[C_{DER},C_{oper},C_{main},C_{loss},C_{carb}\]\end{array}\right.---\left(1\right)$

In formula (1), ω is weight coefficient vector and weight coefficient vector ω is determined by analytic hierarchy process (AHP) or interval based AHP; C is cost vector; C _dERfor construction cost; C _operfor operating cost; C _mainfor maintenance cost; C _lossfor cost of losses; C _carbfor carbon emission amount penalties; Z _costfor distributed energy integrated configuration value;

1-2. decomposes described distributed energy complex optimum allocation models:

$\left\{\begin{array}{c}{C}_{DER}=\underset{i=1}{\overset{n_D}{\Σ}}{f}_{D_i}\·C_{D_i}\·P_{D_i}\·\frac{r_{0i}{(1+r_{0i})}^{m_{Di}}}{{(1+r_{0i})}^{m_{Di}}-1}\\ C_{oper}=\underset{i=1}{\overset{n_D}{\Σ}}{f}_{D_i}\·C_{O_i}\·P_{D_i}\·T_{D_i}\\ C_{main}=\underset{i=1}{\overset{n_D}{\Σ}}{f}_{D_i}\·C_{M_i}\·P_{D_i}\·T_{D_i}\\ C_{loss}=\frac{365}{4}\underset{s=1}{\overset{4}{\Σ}}\underset{h=1}{\overset{24}{\Σ}}\underset{j=1}{\overset{n_B}{\Σ}}{C}_{ele}\·I_{bshj}^2\·R_{bj}\\ C_{carb}=\mathrm{\ψ}\underset{i=1}{\overset{n_D}{\Σ}}{f}_{D_i}\·P_{D_i}\·T_{D_{\max i}}\·{\mathrm{\ζ}}_i\end{array}\right.---\left(2\right)$

In formula (2), n _dfor the system containing distributed power source can be accessed by the nodes to be selected of DER; f _difor flag bit, and f _divalue is 0 or 1, wherein, and f _dithe node i place access DER of access blower fan or photovoltaic is represented when=1; C _difor the construction cost of unit capacity DER; P _direpresent the installed capacity of node i place DER, and P _di>0; r _0iand m _direpresent discount rate and depreciable life respectively; C _oifor the operating cost of node i place unit capacity DER within the unit interval; T _difor the running time in year of node i place DER; S represents 1 year 4 season, and h represents 24 periods of typical case's day various quarters; C _mifor the maintenance cost of node i place unit capacity DER within the unit interval; J is branch road; n _bfor system branch number; C _elefor degree electricity price lattice; I _bshjfor typical case various quarters, day part day flows through the electric current of branch road j; R _bjfor the resistance of branch road j; T _dmaxifor the year maximum output hourage of node i place DER; ξ _ifor degree electrical carbon discharge capacity, described degree electrical carbon discharge capacity is direct carbon emission amount; ψ is carbon emission penalties coefficient, and unit be unit/kg, ψ according to regional economic development level and low-carbon (LC) requirement, value is different.

Preferably, described step 2 comprises:

Determine the constraints condition of opportunity of described distributed energy complex optimum allocation models, described constraints condition of opportunity is under any operating mode, and the not out-of-limit probability of branch power is not less than α 0 and the not out-of-limit probability of node voltage is not less than the constraints of β 0:

$\left\{\begin{array}{c}\mathrm{Pr}ob\left\{\right|p_j|\≤p_{j\max },j=1,2,\mathrm{...},n_B\}\≥{\mathrm{\α}}_0\\ \mathrm{Pr}ob\{u_{\min }\≤u\≤u_{\max }\}\≥{\mathrm{\β}}_0\end{array}\right.---\left(3\right)$

In formula (3), Prob{*} is the probability that event { * } is set up; α ₀, β ₀be confidence level; p _jfor the active power on branch road j; p _jmaxfor the power limit that branch road j allows; U is node voltage vector; u _maxfor the node voltage upper limit; u _minfor node voltage lower limit.

Preferably, the energy-storage system equipping rules of the allocation models of distributed energy complex optimum described in described step 3 comprises:

A. adopt decentralized configuration mode to carry out energy storage configuration to described containing each distributed power source access point in the system of distributed power source, be arranged in each distributed power source access point place by energy-storage system;

B. energy storage configuration is not carried out to the described load bus containing the access miniature gas turbine in the system of distributed power source;

C. to the load bus of the access blower fan in the described system containing distributed power source and photovoltaic, all configuration has the energy storage device of capacity.

Preferably, the capacity of described energy storage device is determined according to the power producing characteristics of node load characteristic and DG, i.e. the capacity of energy storing device C of described energy storage device _eSSfor:

$0<C_{ESS}\&le;\frac{(t_b-t_a)\sqrt{{\mathrm{\&sigma;}}_{DG}^2+{\mathrm{\&sigma;}}_{load}^2}}{\sqrt{n}}\&CenterDot;Y_{\mathrm{\&alpha;}/2}---\left(4\right)$

In formula (4), t _afor the initial time of short-term load forecasting; t _bfor the termination time of short-term load forecasting; N is the sample number of predictive period; Y _α/2for upside α quantile corresponding when confidential interval is 1-α; σ ² _dGfor DG exerts oneself predicated error distribution variance; σ ² _loadfor load prediction error distribution variance.

Preferably, described step 4 comprises:

Described distributed energy complex optimum allocation models after 4-1. initialization decomposition and algorithm parameter;

4-2. initialization distributed energy position and capacity;

4-3. verifies described constraints condition of opportunity;

4-4. intelligent optimization algorithm solves the described distributed energy complex optimum allocation models after initialization.

Preferably, described 4-1 comprises:

D. according to demand history data, the normal distribution that each load bus load initial value and planning level year obey is determined;

E. according to wind speed and illumination historical data, wind speed and illumination parameter is determined;

F. the confidence level parameter alpha of chance constrained programming is set ₀and β ₀; And determine parameter value in described distributed energy complex optimum allocation models.

Preferably, described 4-2 comprises:

According to the wind-resources in area, light resources and construction condition, determine the distributed energy type of load bus priority access to be accessed; For undistinguishable load bus to be accessed, then the on-position of random initializtion distributed energy and capacity, complete the initialization of distributed energy position and capacity.

Preferably, described 4-3 comprises:

With the constraints condition of opportunity of Monte-carlo Simulation Method verification access scheme;

If constraints condition of opportunity is set up, then enter 4-4;

Otherwise return step 2.

Preferably, described 4-4 comprises:

With the described distributed energy complex optimum allocation models after the genetic algorithm in intelligent optimization algorithm and PSO Algorithm initialization, until meet the condition of convergence or reach the iterations upper limit, export solving result.

As can be seen from above-mentioned technical scheme, the invention provides a kind of distributed energy Optimal Configuration Method based on chance constrained programming, set up distributed energy complex optimum allocation models; Determine the constraints condition of opportunity of distributed energy complex optimum allocation models; Determine the energy-storage system equipping rules of described distributed energy complex optimum allocation models; Solve distributed energy complex optimum allocation models.The method that the present invention proposes effectively improves the receiving ability of power distribution network to distributed energy; Energetic efficiency objectives and the low-carbon (LC) target of programme can be met; also cost-effectiveness requirement can be met; different resource level, the addressing of the economic development level Area distribution formula energy and constant volume problem can be solved; guarantee the large-scale application of distributed energy, the environmental pollution that effective alleviation fossil energy causes and energy crisis.

With immediate prior art ratio, technical scheme provided by the invention has following excellent effect:

1, in technical scheme provided by the present invention, by setting up distributed energy complex optimum allocation models; Determine the constraints condition of opportunity of distributed energy complex optimum allocation models; Determine the energy-storage system equipping rules of described distributed energy complex optimum allocation models; Solve distributed energy complex optimum allocation models.Effectively improve the receiving ability of power distribution network to distributed energy.

2, technical scheme provided by the present invention, not only can be active distribution network planning and provides foundation, and can ensure the economy of initiatively distribution system operation, improve entire system efficiency, reduce operation and maintenance expenses use, improve rate of return on investment, for conventional electrical distribution net is to Modern power distribution net upgrading lay a good foundation.

3, technical scheme provided by the present invention; energetic efficiency objectives and the low-carbon (LC) target of programme can be met; also cost-effectiveness requirement can be met; different resource level, the addressing of the economic development level Area distribution formula energy and constant volume problem can be solved; guarantee the large-scale application of distributed energy, the environmental pollution that effective alleviation fossil energy causes and energy crisis.

4, technical scheme provided by the invention, is widely used, and has significant Social benefit and economic benefit.

Accompanying drawing explanation

Fig. 1 is the flow chart of a kind of distributed energy Optimal Configuration Method based on chance constrained programming of the present invention;

Fig. 2 is the schematic flow sheet of step 1 in distributed energy Optimal Configuration Method of the present invention;

Fig. 3 is the schematic flow sheet of step 4 in distributed energy Optimal Configuration Method of the present invention;

Fig. 4 is the energy storage decentralized configuration schematic diagram figure contained in the power distribution network energy-storage system of DG in embody rule example of the present invention;

Fig. 5 is the energy storage centralized configuration schematic diagram figure contained in the power distribution network energy-storage system of DG in embody rule example of the present invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on embodiments of the invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.

As shown in Figure 1, the invention provides a kind of distributed energy Optimal Configuration Method based on chance constrained programming, comprise the steps:

Step 1. sets up distributed energy complex optimum allocation models;

Step 2. determines the constraints condition of opportunity of distributed energy complex optimum allocation models;

Step 3. determines the energy-storage system equipping rules of described distributed energy complex optimum allocation models;

Step 4. solves described distributed energy complex optimum allocation models.

As shown in Figure 2, step 1 comprises:

1-1. sets up distributed energy complex optimum allocation models:

$\left\{\begin{array}{c}{\mathrm{minZ}}_{\cos t}=\mathrm{\&omega;}\&CenterDot;C^T\\ \mathrm{\&omega;}=\&lsqb;{\mathrm{\&omega;}}_1,{\mathrm{\&omega;}}_2,{\mathrm{\&omega;}}_3,{\mathrm{\&omega;}}_4,{\mathrm{\&omega;}}_5\&rsqb;\\ C=\&lsqb;C_{DER},C_{oper},C_{main},C_{loss},C_{carb}\&rsqb;\end{array}\right.---\left(1\right)$

In formula (1), ω is weight coefficient vector and weight coefficient vector ω is determined by analytic hierarchy process (AHP) or interval based AHP; C is cost vector; C _dERfor construction cost; C _operfor operating cost; C _mainfor maintenance cost; C _lossfor cost of losses; C _carbfor carbon emission amount penalties; Z _costfor distributed energy integrated configuration value;

1-2. decomposes distributed energy complex optimum allocation models:

$\left\{\begin{array}{c}{C}_{DER}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{D_i}\&CenterDot;P_{D_i}\&CenterDot;\frac{r_{0i}{(1+r_{0i})}^{m_{Di}}}{{(1+r_{0i})}^{m_{Di}}-1}\\ C_{oper}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{O_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_i}\\ C_{main}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{M_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_i}\\ C_{loss}=\frac{365}{4}\underset{s=1}{\overset{4}{\&Sigma;}}\underset{h=1}{\overset{24}{\&Sigma;}}\underset{j=1}{\overset{n_B}{\&Sigma;}}{C}_{ele}\&CenterDot;I_{bshj}^2\&CenterDot;R_{bj}\\ C_{carb}=\mathrm{\&psi;}\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_{\max i}}\&CenterDot;{\mathrm{\&zeta;}}_i\end{array}\right.---\left(2\right)$

In formula (2), n _dfor the system containing distributed power source can be accessed by the nodes to be selected of DER; f _difor flag bit, and f _divalue is 0 or 1, wherein, and f _dithe node i place access DER of access blower fan or photovoltaic is represented when=1; C _difor the construction cost of unit capacity DER; P _direpresent the installed capacity of node i place DER, and P _di>0; r _0iand m _direpresent discount rate and depreciable life respectively; C _oifor the operating cost of node i place unit capacity DER within the unit interval; T _difor the running time in year of node i place DER; S represents 1 year 4 season, and h represents 24 periods of typical case's day various quarters; C _mifor the maintenance cost of node i place unit capacity DER within the unit interval; J is branch road; n _bfor system branch number; C _elefor degree electricity price lattice; I _bshjfor typical case various quarters, day part day flows through the electric current of branch road j; R _bjfor the resistance of branch road j; T _dmaxifor the year maximum output hourage of node i place DER; ξ _ifor degree electrical carbon discharge capacity, degree electrical carbon discharge capacity is direct carbon emission amount; ψ is carbon emission penalties coefficient, and unit be unit/kg, ψ according to regional economic development level and low-carbon (LC) requirement, value is different.

Wherein, step 2 comprises:

Determine the constraints condition of opportunity of distributed energy complex optimum allocation models, constraints condition of opportunity is under any operating mode, and the not out-of-limit probability of branch power is not less than α 0 and the not out-of-limit probability of node voltage is not less than the constraints of β 0:

$\left\{\begin{array}{c}\mathrm{Pr}ob\left\{\right|p_j|\&le;p_{j\max },j=1,2,\mathrm{...},n_B\}\&GreaterEqual;{\mathrm{\&alpha;}}_0\\ \mathrm{Pr}ob\{u_{\min }\&le;u\&le;u_{\max }\}\&GreaterEqual;{\mathrm{\&beta;}}_0\end{array}\right.---\left(3\right)$

In formula (3), Prob{*} is the probability that event { * } is set up; α ₀, β ₀be confidence level; p _jfor the active power on branch road j; p _jmaxfor the power limit that branch road j allows; U is node voltage vector; u _maxfor the node voltage upper limit; u _minfor node voltage lower limit.

Wherein, the energy-storage system equipping rules of the complex optimum of distributed energy described in step 3 allocation models comprises:

A. adopt decentralized configuration mode to carry out energy storage configuration to containing each distributed power source access point in the system of distributed power source, be arranged in each distributed power source access point place by energy-storage system;

B. energy storage configuration is not carried out to the load bus containing the access miniature gas turbine in the system of distributed power source;

C. to the load bus containing the access blower fan in the system of distributed power source and photovoltaic, all configuration has the energy storage device of capacity.

Wherein, the capacity of energy storage device is determined according to the power producing characteristics of node load characteristic and DG, i.e. the capacity of energy storing device C of energy storage device _eSSfor:

$0<C_{ESS}\&le;\frac{(t_b-t_a)\sqrt{{\mathrm{\&sigma;}}_{DG}^2+{\mathrm{\&sigma;}}_{load}^2}}{\sqrt{n}}\&CenterDot;Y_{\mathrm{\&alpha;}/2}---\left(4\right)$

In formula (4), t _afor the initial time of short-term load forecasting; t _bfor the termination time of short-term load forecasting; N is the sample number of predictive period; Y _α/2for upside α quantile corresponding when confidential interval is 1-α; σ ² _dGfor DG exerts oneself predicated error distribution variance; σ ² _loadfor load prediction error distribution variance.

As shown in Figure 3, step 4 comprises:

Distributed energy complex optimum allocation models after 4-1. initialization decomposition and algorithm parameter;

4-2. initialization distributed energy position and capacity;

4-3. verifies constraints condition of opportunity;

4-4. intelligent optimization algorithm solves the distributed energy complex optimum allocation models after initialization.

Wherein, 4-1 comprises:

D. according to demand history data, the normal distribution that each load bus load initial value and planning level year obey is determined;

E. according to wind speed and illumination historical data, wind speed and illumination parameter is determined;

F. the confidence level parameter alpha of chance constrained programming is set ₀and β ₀; And determine parameter value in distributed energy complex optimum allocation models.

Wherein, 4-2 comprises:

According to the wind-resources in area, light resources and construction condition, determine the distributed energy type of load bus priority access to be accessed; For undistinguishable load bus to be accessed, then the on-position of random initializtion distributed energy and capacity, complete the initialization of distributed energy position and capacity.

Wherein, 4-3 comprises:

With the constraints condition of opportunity of Monte-carlo Simulation Method verification access scheme;

If constraints condition of opportunity is set up, then enter 4-4;

Otherwise return step 2.

Wherein, 4-4 comprises:

With the distributed energy complex optimum allocation models after the genetic algorithm in intelligent optimization algorithm and PSO Algorithm initialization, until meet the condition of convergence or reach the iterations upper limit, export solving result.

The invention provides a kind of embody rule example of the distributed energy Optimal Configuration Method based on chance constrained programming, as follows:

(1) distributed energy complex optimum allocation models is set up

Efficiency, according to physical viewpoint, refers in using energy source, ratio that is that play a role and quantity of energy that is actual consumption.According to above-mentioned definition, to plan that electrical network efficiency is optimum, average annual construction cost, operating cost and maintenance cost in conjunction with low-carbon (LC) target and distributed energy, realize configuring the complex optimum of distributed energy, Mathematical Modeling is as follows.

$\begin{array}{c}{\mathrm{minZ}}_{\cos t}=\mathrm{\&omega;}\&CenterDot;C^T\\ \mathrm{\&omega;}=\&lsqb;{\mathrm{\&omega;}}_1,{\mathrm{\&omega;}}_2,{\mathrm{\&omega;}}_3,{\mathrm{\&omega;}}_4,{\mathrm{\&omega;}}_5\&rsqb;\\ C=\&lsqb;C_{DER},C_{oper},C_{main},C_{loss},C_{carb}\&rsqb;\end{array}---\left(1\right)$

In formula:

ω---weight coefficient vector;

C---cost vector.Comprise construction cost, operating cost, maintenance cost, cost of losses and carbon emission amount penalties.

Formula (1) is complex optimum target function, represents that in planning horizon, comprehensive cost minimizes, and can be analyzed to again:

$\begin{array}{c}{C}_{DER}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{D_i}\&CenterDot;P_{D_i}\&CenterDot;\frac{r_{0i}{(1+r_{0i})}^{m_{Di}}}{{(1+r_{0i})}^{m_{Di}}-1}\\ C_{oper}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{O_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_i}\\ C_{main}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{M_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_i}\\ C_{loss}=\frac{365}{4}\underset{s=1}{\overset{4}{\&Sigma;}}\underset{h=1}{\overset{24}{\&Sigma;}}\underset{j=1}{\overset{n_B}{\&Sigma;}}{C}_{ele}\&CenterDot;I_{bshj}^2\&CenterDot;R_{bj}\\ C_{carb}=\mathrm{\&psi;}\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_{\max i}}\&CenterDot;{\mathrm{\&zeta;}}_i\end{array}---\left(2\right)$

In formula:

N _d---system can be accessed by the nodes to be selected of DER;

F _di---flag bit, its value is 0 or 1, f _di=1 represents node i place access DER;

C _di---the construction cost of unit capacity DER;

C _oi---the operating cost of node i place unit capacity DER within the unit interval;

C _mi---the maintenance cost of node i place unit capacity DER within the unit interval.

As node i place access blower fan or photovoltaic, think its C _oi=0; P _direpresent the installed capacity (P of node i place DER _di>0); r _0iand m _direpresent discount rate and depreciable life respectively; T _difor the running time in year of node i place DER; C _lossrepresent the year cost of losses of the rear system of DER access; S represents 1 year 4 season, and h represents 24 periods of typical case's day various quarters; C _elefor degree electricity price lattice, n _bfor system branch number, I _bshjrepresent that typical case various quarters, day part day flows through the electric current of branch road j, R _bjfor the resistance of branch road j; C _carbrepresent DER carbon emission penalties, T _dmaxifor the year maximum output hourage of node i place DER, ξ _ifor degree electrical carbon discharge capacity (direct carbon emission amount does not comprise indirect carbon emission amount), ψ is carbon emission penalties coefficient, and unit is unit/kg, and according to regional economic development level and low-carbon (LC) requirement, value is different.

Weight coefficient vector ω can be determined by the decision-making technique such as analytic hierarchy process (AHP), interval based AHP.

(2) clear and definite constraints

This embody rule example adopts following constraints condition of opportunity:

$\begin{array}{c}\mathrm{Pr}ob\left\{\right|p_j|\&le;p_{j\max },j=1,2,\mathrm{...},n_B\}\&GreaterEqual;{\mathrm{\&alpha;}}_0\\ \mathrm{Pr}ob\{u_{\min }\&le;u\&le;u_{\max }\}\&GreaterEqual;{\mathrm{\&beta;}}_0\end{array}---\left(3\right)$

In formula:

The probability that Prob{*}---event { * } is set up;

α ₀, β ₀---confidence level;

P _j---the active power on circuit j;

P _jmax---the power limit that circuit j allows;

U---node voltage vector;

U _max---the node voltage upper limit;

U _min---node voltage lower limit.

Formula (3) represents under any operating mode, and the not out-of-limit probability of branch power is not less than α ₀, the not out-of-limit probability of node voltage is not less than β ₀.

(3) energy-storage system collocation method is determined

Energy storage is used in the system containing distributed power source, stabilizing power fluctuation, improve grid stability, improve in the quality of power supply there is irreplaceable support and optimization function, but because its cost of investment is high, special shortcoming battery energy storage also being existed to cycle life multi-cycle finite, so from the viewpoint of the economy of cost of investment and the validity of functional realiey, configure rational capacity during stored energy application and be necessary.At present, energy storage is mainly being divided into decentralized configuration and centralized configuration two kinds containing the configuration mode in the distribution system of distributed power source, respectively as shown in Figures 4 and 5, the energy storage equipping rules herein considered mainly contain following some:

A. adopt decentralized configuration mode, namely energy-storage system is arranged near each distributed power source access point;

B. for the load bus of access miniature gas turbine, energy storage device is not configured;

C. for the load bus of access blower fan, photovoltaic, all configure the energy storage device of certain capacity, its capacity is determined jointly by the power producing characteristics of node load characteristic and DG, and method is as follows:

If certain load bus short-term load forecasting and the DG that connects exert oneself, predicated error meets Δ P respectively _load~ N (μ _load, σ ² _load), Δ P _dG~ N (μ _dG, σ ² _dG), by the feature of Lie Wei-Edward Lindberg central-limit theorem and normal distribution, itself and distribution W=Δ P _load+ Δ P _dGmeet W ~ N (μ _load+ μ _dG, σ ² _load+ σ ² _dG), then the configurable capacity of energy storing device of this node is:

$0<C_{ESS}\&le;\frac{(t_b-t_a)\sqrt{{\mathrm{\&sigma;}}_{DG}^2+{\mathrm{\&sigma;}}_{load}^2}}{\sqrt{n}}\&CenterDot;Y_{\mathrm{\&alpha;}/2}---\left(4\right)$

In formula:

T _a---the initial time of short-term load forecasting;

T _b---the termination time of short-term load forecasting;

The sample number of n---predictive period;

Y _α/2---upside α quantile corresponding when confidential interval is 1-α;

σ ² _dG---DG exerts oneself predicated error distribution variance;

σ ² _load---load prediction error distribution variance.

Confidential interval 1-α generally gets 95%, corresponding quantile Y _α/2be 1.96.During practical application, the typical day can choosing 1 year 4 season carries out short-term load forecasting, and the capacity of energy storing device of each Joint Enterprise gets the mean value of result of calculation.

(4) model solution

A. initialization model and algorithm parameter.

● according to demand history data, determine normal distribution N ~ (μ that each load bus load initial value and planning level year obey _l, σ _l ²).

● according to wind speed, illumination historical data, determine wind speed, illumination relevant parameter, namely determine k, c parameter of Weibull distribution (wind speed) and α, β parameter of Beta distribution (illumination).

● the confidence level parameter alpha of setting chance constrained programming ₀, β ₀; Provide major parameter value in model.

B. distributed energy position and capacity initialization.

According to regional actual conditions, consider wind, light resources and construction condition, determine the distributed energy type of load bus priority access to be accessed; For undistinguishable load bus to be accessed, then the on-position of random initializtion distributed energy and capacity.

C. constraints verification

By the constraints of Monte Carlo simulation verification access scheme, if condition meets, then perform step 4, otherwise return step 2 and re-execute.

D. by intelligent optimization algorithm, model is solved.

By solving model by the intelligent optimization algorithm such as genetic algorithm, particle cluster algorithm, until meet the condition of convergence or reach maximum iteration time, and Output rusults.

The distributed energy Optimal Configuration Method based on chance constrained programming that this embody rule example provides; according to the focal point of investment decision person; model can meet energetic efficiency objectives and the low-carbon (LC) target of programme; also cost-effectiveness requirement can be met; different resource level, the addressing of the economic development level Area distribution formula energy and constant volume problem can be solved preferably; guarantee the large-scale application of distributed energy, the environmental pollution that effective alleviation fossil energy causes and energy crisis.

Above embodiment is only in order to illustrate that technical scheme of the present invention is not intended to limit; although with reference to above-described embodiment to invention has been detailed description; those of ordinary skill in the field still can modify to the specific embodiment of the present invention or equivalent replacement; and these do not depart from any amendment of spirit and scope of the invention or equivalent replacement, it is all being applied within the claims of the present invention awaited the reply.

Claims (10)

1. based on a distributed energy Optimal Configuration Method for chance constrained programming, it is characterized in that, described method comprises the steps:

Step 1. sets up distributed energy complex optimum allocation models;

Step 2. determines the constraints condition of opportunity of described distributed energy complex optimum allocation models;

Step 3. determines the energy-storage system equipping rules of described distributed energy complex optimum allocation models;

Step 4. solves described distributed energy complex optimum allocation models.

2. the method for claim 1, is characterized in that, described step 1 comprises:

1-1. sets up distributed energy complex optimum allocation models:

$\left\{\begin{array}{c}{\mathrm{minZ}}_{\cos t}=\mathrm{\&omega;}\&CenterDot;C^T\\ \mathrm{\&omega;}=\&lsqb;{\mathrm{\&omega;}}_1,{\mathrm{\&omega;}}_2,{\mathrm{\&omega;}}_3,{\mathrm{\&omega;}}_4,{\mathrm{\&omega;}}_5\&rsqb;\\ C=\&lsqb;C_{DER},C_{oper},C_{main},C_{loss},C_{carb}\&rsqb;\end{array}\right.---\left(1\right)$

In formula (1), ω is weight coefficient vector and weight coefficient vector ω is determined by analytic hierarchy process (AHP) or interval based AHP; C is cost vector; C _dERfor construction cost; C _operfor operating cost; C _mainfor maintenance cost; C _lossfor cost of losses; C _carbfor carbon emission amount penalties; Z _costfor distributed energy integrated configuration value;

1-2. decomposes described distributed energy complex optimum allocation models:

$\left\{\begin{array}{c}{C}_{DER}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{D_i}\&CenterDot;P_{D_i}\&CenterDot;\frac{r_{0i}{(1+r_{0i})}^{m_{Di}}}{{(1+r_{0i})}^{m_{Di}}-1}\\ C_{oper}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{O_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_i}\\ C_{main}=\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;C_{M_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_i}\\ C_{loss}=\frac{365}{4}\underset{s=1}{\overset{4}{\&Sigma;}}\underset{h=1}{\overset{24}{\&Sigma;}}\underset{j=1}{\overset{n_B}{\&Sigma;}}{C}_{ele}\&CenterDot;I_{bshj}^2\&CenterDot;R_{bj}\\ C_{carb}=\mathrm{\&psi;}\underset{i=1}{\overset{n_D}{\&Sigma;}}{f}_{D_i}\&CenterDot;P_{D_i}\&CenterDot;T_{D_{\max i}}\&CenterDot;{\mathrm{\&zeta;}}_i\end{array}\right.---\left(2\right)$

In formula (2), n _dfor the system containing distributed power source can be accessed by the nodes to be selected of DER; f _difor flag bit, and f _divalue is 0 or 1, wherein, and f _dithe node i place access DER of access blower fan or photovoltaic is represented when=1; C _difor the construction cost of unit capacity DER; P _direpresent the installed capacity of node i place DER, and P _di>0; r _0iand m _direpresent discount rate and depreciable life respectively; C _oifor the operating cost of node i place unit capacity DER within the unit interval; T _difor the running time in year of node i place DER; S represents 1 year 4 season, and h represents 24 periods of typical case's day various quarters; C _mifor the maintenance cost of node i place unit capacity DER within the unit interval; J is branch road; n _bfor system branch number; C _elefor degree electricity price lattice; I _bshjfor typical case various quarters, day part day flows through the electric current of branch road j; R _bjfor the resistance of branch road j; T _dmaxifor the year maximum output hourage of node i place DER; ξ _ifor degree electrical carbon discharge capacity, described degree electrical carbon discharge capacity is direct carbon emission amount; ψ is carbon emission penalties coefficient, and unit be unit/kg, ψ according to regional economic development level and low-carbon (LC) requirement, value is different.

3. method as claimed in claim 2, it is characterized in that, described step 2 comprises:

Determine the constraints condition of opportunity of described distributed energy complex optimum allocation models, described constraints condition of opportunity is under any operating mode, and the not out-of-limit probability of branch power is not less than α 0 and the not out-of-limit probability of node voltage is not less than the constraints of β 0:

$\left\{\begin{array}{c}\mathrm{Pr}ob\left\{\right|p_j|\&le;p_{j\max },j=1,2,...,n_B\}\&GreaterEqual;{\mathrm{\&alpha;}}_0\\ \mathrm{Pr}ob\{u_{\min }\&le;u\&le;u_{\max }\}\&GreaterEqual;{\mathrm{\&beta;}}_0\end{array}\right.---\left(3\right)$

In formula (3), Prob{*} is the probability that event { * } is set up; α ₀, β ₀be confidence level; p _jfor the active power on branch road j; p _jmaxfor the power limit that branch road j allows; U is node voltage vector; u _maxfor the node voltage upper limit; u _minfor node voltage lower limit.

4. the method for claim 1, is characterized in that, the energy-storage system equipping rules of the allocation models of distributed energy complex optimum described in described step 3 comprises:

A. adopt decentralized configuration mode to carry out energy storage configuration to described containing each distributed power source access point in the system of distributed power source, be arranged in each distributed power source access point place by energy-storage system;

B. energy storage configuration is not carried out to the described load bus containing the access miniature gas turbine in the system of distributed power source;

C. to the load bus of the access blower fan in the described system containing distributed power source and photovoltaic, all configuration has the energy storage device of capacity.

5. method as claimed in claim 4, it is characterized in that, the capacity of described energy storage device is determined according to the power producing characteristics of node load characteristic and DG, i.e. the capacity of energy storing device C of described energy storage device _eSSfor:

$0<C_{ESS}\&le;\frac{(t_b-t_a)\sqrt{{\mathrm{\&sigma;}}_{DG}^2+{\mathrm{\&sigma;}}_{load}^2}}{\sqrt{n}}\&CenterDot;Y_{\mathrm{\&alpha;}/2}---\left(4\right)$

In formula (4), t _afor the initial time of short-term load forecasting; t _bfor the termination time of short-term load forecasting; N is the sample number of predictive period; Y _α/2for upside α quantile corresponding when confidential interval is 1-α; σ ² _dGfor DG exerts oneself predicated error distribution variance; σ ² _loadfor load prediction error distribution variance.

6. the method for claim 1, is characterized in that, described step 4 comprises:

Described distributed energy complex optimum allocation models after 4-1. initialization decomposition and algorithm parameter;

4-2. initialization distributed energy position and capacity;

4-3. verifies described constraints condition of opportunity;

4-4. intelligent optimization algorithm solves the described distributed energy complex optimum allocation models after initialization.

7. method as claimed in claim 6, it is characterized in that, described 4-1 comprises:

D. according to demand history data, the normal distribution that each load bus load initial value and planning level year obey is determined;

E. according to wind speed and illumination historical data, wind speed and illumination parameter is determined;

F. the confidence level parameter alpha of chance constrained programming is set ₀and β ₀; And determine parameter value in described distributed energy complex optimum allocation models.

8. method as claimed in claim 6, it is characterized in that, described 4-2 comprises:

According to the wind-resources in area, light resources and construction condition, determine the distributed energy type of load bus priority access to be accessed; For undistinguishable load bus to be accessed, then the on-position of random initializtion distributed energy and capacity, complete the initialization of distributed energy position and capacity.

9. method as claimed in claim 6, it is characterized in that, described 4-3 comprises:

With the constraints condition of opportunity of Monte-carlo Simulation Method verification access scheme;

If constraints condition of opportunity is set up, then enter 4-4;

Otherwise return step 2.

10. method as claimed in claim 6, it is characterized in that, described 4-4 comprises:

With the described distributed energy complex optimum allocation models after the genetic algorithm in intelligent optimization algorithm and PSO Algorithm initialization, until meet the condition of convergence or reach the iterations upper limit, export solving result.