CN105140958A - Method for planning power distribution network comprising photovoltaic power supply - Google Patents

Abstract

The invention relates to the technical field of power distribution networks, in particular to a method for planning a power distribution network comprising photovoltaic power supply. The method comprises the following steps: (1) establishing a target function of social cost; (2) determining a constraint condition; and (3) determining the planning method with a particle swarm algorithm containing chaos optimization, outputting the optimal planning scheme and determining the optimal access position and capacity of the photovoltaic power supply. According to the method, the optimal access position and capacity of the photovoltaic power supply are calculated, so that the network loss of the power distribution network is reduced; the operating cost of the power distribution network is reduced; the voltage of a system is improved; the load capacity of the system is enhanced; safe and effective operation of the power distribution network can be ensured; and the economic benefits can also be further improved.

Description

Containing the distribution network planning method of photo-voltaic power supply

Technical field

The present invention relates to distribution network technology field, be specifically related to a kind of distribution network planning method containing photo-voltaic power supply.

Background technology

Along with the reform of electricity market, technology constantly develops, and photovoltaic generating system grows up on a large scale.This novel forms of electricity generation supports the operation of existing power distribution network to a great extent, small-sized photovoltaic plant can utilize the idle vacant lots such as roof to be arranged near user, large-scale photovoltaic plant needs independently place, can export electric energy independently, and safety flexibly; Advantage by means of its clean environment firendly receives the support energetically of national policy.

Along with making constant progress of technology, following electrical network needs to consider to receive more photo-voltaic power supply.Traditional distribution net work structure will change; existing power distribution network is generally single supply Radial network; the access of photo-voltaic power supply becomes the network of many power supplys; all can produce certain impact to the trend of power distribution network, voltage and relaying protection etc., on-position and the capacity of its influence degree and photo-voltaic power supply are closely bound up.Irrational on-position and capacity will affect the safe and stable operation of power distribution network, affect economic benefit, bring huge loss to country.

At present, the distribution network planning for distributed power source is more, mostly from economy point, under the prerequisite that pursuit social cost is minimum, establishes many different models.But photo-voltaic power supply and other forms of distributed power source distinguish to some extent, its cost model and oneself the feature of exerting oneself, can not use same model to lump together.Therefore, for the feature of photovoltaic system generating, corresponding plan model and suitable processing method to be proposed in planning.

Summary of the invention

In order to the deficiency in solving the problems of the technologies described above, the object of the invention is to: a kind of distribution network planning method containing photo-voltaic power supply is provided, can pass through social cost target function, the best on-position of acquisition photo-voltaic power supply in power distribution network and capacity.

The technical scheme that the present invention adopts for its technical problem of solution is:

The described distribution network planning method containing photo-voltaic power supply, mainly comprise the year investment operating cost of photo-voltaic power supply, network year wear and tear expense, the power purchase expense of superior electrical network and the photo-voltaic power supply social cost of saving because of policy subsidy, comprise the following steps:

(1) target function of social cost is set up:

${\mathrm{minC}}_{total}=C_p+C_{loss}+C_{en}-C_P^U$

In formula, C _totalfor social cost; C _pfor the year investment operating cost of photo-voltaic power supply; C _lossfor network year wear and tear expense; C _enfor the power purchase expense of superior electrical network; for the social cost that photo-voltaic power supply is saved because of policy subsidy;

A, set up photo-voltaic power supply year investment operating cost C _pfunction:

$C_P=\underset{i=1}{\overset{n_p}{\Σ}}({\mathrm{\αC}}_{pi}+W_{pi})$

$\mathrm{\α}=\frac{\mathrm{\γ}{(1+\mathrm{\γ})}^t}{{(1+\mathrm{\γ})}^t-1}$

In formula, n _pfor accessing the photo-voltaic power supply number of power distribution network; α is the year investment average cost coefficient of photo-voltaic power supply; C _piit is the investment cost of i-th photo-voltaic power supply; W _piit is the operation and maintenance cost of i-th photo-voltaic power supply; γ is Annual Percentage Rate; T is the planning time limit;

B, set up network year wear and tear expense C _lossfunction:

$C_{loss}=\underset{i=1}{\overset{n_l}{\Σ}}{C}_0{\mathrm{\τ}}_{1\max }{\mathrm{\ΔP}}_{li}$

In formula, n _lfor distribution network line number; C ₀for unit electricity price (unit/kWh); for annual peak load loss hourage (h) of branch road; Δ P _liit is the active loss of i-th branch road;

C, set up the power purchase expense C of superior electrical network _enfunction:

C _en＝T _max×(P _Σ-P _ΣP)×C ₀

In formula, T _maxfor peak load annual utilization hours; P _Σfor the total burden with power of power distribution network; P _{Σ P}exert oneself for photo-voltaic power supply is always meritorious;

D, set up photo-voltaic power supply because of policy subsidy save social cost function:

$C_P^U=C_1\underset{i=1}{\overset{n_p}{\Σ}}{P}_{\mathrm{\Σ}P}{\mathrm{\τ}}_{2max}$

In formula, C ₁for public subsidies cost (unit/kWh); for year maximum illumination utilizes hourage;

(2) determine constraints, constraints mainly comprises:

E, node voltage retrain:

U _imin≤U _i≤U _imax

In formula, U _ifor the magnitude of voltage of i node; U _iminfor node voltage lower limit; U _imaxfor node voltage higher limit;

F, tributary capacity retrain:

P _j≤P _jmax

In formula, P _jfor flowing through the active power of circuit j; P _jmaxfor the higher limit of jth bar line transmission active power;

G, power distribution network access photo-voltaic power supply capacity-constrained:

S _Pi≤S _imax

S _ΣP≤S _L

In formula, S _pifor the capacity of the photo-voltaic power supply of i node access; S _imaxthe maximum size value of the photo-voltaic power supply of access is allowed for i node; S _{Σ P}for accessing the total capacity of photo-voltaic power supply in power distribution network; S _lfor 10% of distribution network load total amount;

(3) utilize the particle cluster algorithm containing chaos optimization to solve planing method, and export optimum programming scheme, determine best on-position and the capacity of photo-voltaic power supply.

Preferably, the described solution procedure of particle cluster algorithm to planing method containing chaos optimization mainly comprises:

Step 1: input initial data, mainly comprises node and the circuit relevant parameter of power distribution network, particle number of groups, Studying factors c ₁and c ₂, maximum iteration time, inertia weight ω, population maximum update speed v _max, control variables μ, random factor α;

Step 2: in search volume, random generation M particle, initial population is [P ₁, P ₂, P ₃..., P _i..., P _m], use Load flow calculation to judge, P _ifor meeting arbitrary feasible solution of planing method;

Step 3: the fitness calculating each particle, the individual extreme value P of record current particle _i, and therefrom select global extremum P _g;

Step 4: utilize following formula to carry out evolution to particle and upgrade, and calculate the individual extreme value P of more new particle _iwith global extremum P _g;

$v_{i,d}^{k+1}={\mathrm{\ωv}}_{i,d}^k+c_1\·rand\left(\right)\·(p_{i,d}^k-x_{i,d}^k)+c_{2}rand\left(\right)\·(p_{g,d}^k-x_{i,d}^k)$

$x_{i,d}^{k+1}=x_{i,d}^k+v_{i,d}^{k+1}$

In formula, ω is inertia weight; c ₁and c ₂for Studying factors;

Step 5: judge that whether optimal particle is identical with adjacent three generations, if identical, carry out steps d chaos optimization; If different, jump to step e;

Step 6: the optimal location of population is mapped to Logistic function, produce Chaos Variable sequence, then former solution space is returned in inverse mapping, calculate the adaptive value of each feasible solution in above-mentioned feasible solution sequence, and retain adaptive value optimum time corresponding feasible solution, to replace in population the particle that is random;

Step 7: judge whether to reach maximum iteration time, if do not reach, forward step a to, proceed, if reach, forward step f to;

Step 8: export optimum programming scheme.

Compared with prior art, the present invention has following beneficial effect:

When power distribution network increases photo-voltaic power supply, need to determine its rational on-position and capacity; The present invention takes into full account trend, the constraint such as voltage and electric current of power distribution network, adopt the principle planning photo-voltaic power supply of economy optimum, and utilize modified particle swarm optiziation to carry out solving calculating, avoid the middle early problem of conventional particle group algorithm, and then calculate best on-position and the capacity of photo-voltaic power supply, reduce the via net loss of power distribution network, decrease the operating cost of power distribution network, improve the voltage of system, enhance the load capacity of system; Can not only ensure that power distribution network runs safely and effectively, its economic benefit can also be made to be further improved.

Accompanying drawing explanation

Fig. 1 FB(flow block) of the present invention.

Embodiment

Below the embodiment of the present invention is described further:

Embodiment 1

Distribution network planning method containing photo-voltaic power supply of the present invention, mainly comprise the year investment operating cost of photo-voltaic power supply, network year wear and tear expense, the power purchase expense of superior electrical network and the photo-voltaic power supply social cost of saving because of policy subsidy, comprise the following steps:

(1) target function of social cost is set up:

${\mathrm{minC}}_{total}=C_p+C_{loss}+C_{en}-C_P^U$

In formula, C _totalfor social cost; C _pfor the year investment operating cost of photo-voltaic power supply; C _lossfor network year wear and tear expense; C _enfor the power purchase expense of superior electrical network; for the social cost that photo-voltaic power supply is saved because of policy subsidy;

A, set up photo-voltaic power supply year investment operating cost C _pfunction:

$C_P=\underset{i=1}{\overset{n_p}{\Σ}}({\mathrm{\αC}}_{pi}+W_{pi})$

$\mathrm{\α}=\frac{\mathrm{\γ}{(1+\mathrm{\γ})}^t}{{(1+\mathrm{\γ})}^t-1}$

In formula, n _pfor accessing the photo-voltaic power supply number of power distribution network; α is the year investment average cost coefficient of photo-voltaic power supply; C _piit is the investment cost of i-th photo-voltaic power supply; W _piit is the operation and maintenance cost of i-th photo-voltaic power supply; γ is Annual Percentage Rate; T is the planning time limit;

B, set up network year wear and tear expense C _lossfunction:

$C_{loss}=\underset{i=1}{\overset{n_l}{\Σ}}{C}_0{\mathrm{\τ}}_{1\max }{\mathrm{\ΔP}}_{li}$

In formula, n _lfor distribution network line number; C ₀for unit electricity price (unit/kWh); for annual peak load loss hourage (h) of branch road; Δ P _liit is the active loss of i-th branch road;

C, set up the power purchase expense C of superior electrical network _enfunction:

C _en＝T _max×(P _Σ-P _ΣP)×C ₀

In formula, T _maxfor peak load annual utilization hours; P _Σfor the total burden with power of power distribution network; P _{Σ P}exert oneself for photo-voltaic power supply is always meritorious;

D, set up photo-voltaic power supply because of policy subsidy save social cost function:

$C_P^U=C_1\underset{i=1}{\overset{n_p}{\Σ}}{P}_{\mathrm{\Σ}P}{\mathrm{\τ}}_{2max}$

In formula, C ₁for public subsidies cost (unit/kWh); for year maximum illumination utilizes hourage;

(2) determine constraints, constraints mainly comprises:

E, node voltage retrain:

U _imin≤U _i≤U _imax

In formula, U _ifor the magnitude of voltage of i node; U _iminfor node voltage lower limit; U _imaxfor node voltage higher limit;

F, tributary capacity retrain:

P _j≤P _jmax

In formula, P _jfor flowing through the active power of circuit j; P _jmaxfor the higher limit of jth bar line transmission active power;

G, because photo-voltaic power supply is by the impact of the condition such as illumination, temperature, it is exerted oneself and has unsteadiness, considers the stability of system, needs to be limited the capacity of photo-voltaic power supply; Regulation photo-voltaic power supply capacity can not exceed the size of place node load, and the maximum access capacity of photo-voltaic power supply is no more than 10% of electrical network peak load total amount, power distribution network access photo-voltaic power supply capacity-constrained:

S _Pi≤S _imax

S _ΣP≤S _L

In formula, S _pifor the capacity of the photo-voltaic power supply of i node access; S _imaxthe maximum size value of the photo-voltaic power supply of access is allowed for i node; S _{Σ P}for accessing the total capacity of photo-voltaic power supply in power distribution network; S _lfor 10% of distribution network load total amount;

(3) utilize the particle cluster algorithm containing chaos optimization to solve planing method, and export optimum programming scheme, determine best on-position and the capacity of photo-voltaic power supply.

Preferably, the described solution procedure of particle cluster algorithm to planing method containing chaos optimization mainly comprises:

Step 1: input initial data, mainly comprises node and the circuit relevant parameter of power distribution network, particle number of groups, Studying factors c ₁and c ₂, maximum iteration time, inertia weight ω, population maximum update speed v _max, control variables μ, random factor α;

Step 2: in search volume, random generation M particle, initial population is [P ₁, P ₂, P ₃..., P _i..., P _m], use Load flow calculation to judge, P _ifor meeting arbitrary feasible solution of planing method;

Step 3: the fitness calculating each particle, the individual extreme value P of record current particle _i, and therefrom select global extremum P _g;

Step 4: utilize following formula to carry out evolution to particle and upgrade, and calculate the individual extreme value P of more new particle _iwith global extremum P _g;

$v_{i,d}^{k+1}={\mathrm{\ωv}}_{i,d}^k+c_1\·rand\left(\right)\·(p_{i,d}^k-x_{i,d}^k)+c_{2}rand\left(\right)\·(p_{g,d}^k-x_{i,d}^k)$

$x_{i,d}^{k+1}=x_{i,d}^k+v_{i,d}^{k+1}$

In formula, ω is inertia weight; c ₁and c ₂for Studying factors;

Step 5: judge that whether optimal particle is identical with adjacent three generations, if identical, carry out steps d chaos optimization; If different, jump to step e;

Step 6: the optimal location of population is mapped to Logistic function, produce Chaos Variable sequence, then former solution space is returned in inverse mapping, calculate the adaptive value of each feasible solution in above-mentioned feasible solution sequence, and feasible solution corresponding when retaining adaptive value optimum, to replace in population the particle that is random, concrete steps are as follows:

One, by the optimal location p of population _g=(p _{g, 1}, p _{g, 2}, p _{g, 3}..., p _g,n) carry out chaos optimization, be mapped in the domain of definition [0,1] of Logistic equation, produced chaos sequence variable mapping equation adopts following formula,

$y_1^k=(\frac{p_{g,1}}{R},\frac{p_{g,2}}{R},\frac{p_{g,3}}{R},...,\frac{p_{g,d}}{R})$

In formula, R is the maximum permission installed capacity of photo-voltaic power supply;

Two, carry out iteration with Logistic equation and produce Chaos Variable sequence, typical Logistic equation is because the nodes of actual installation photo-voltaic power supply is much smaller than the sum of power distribution network node, namely likelihood ratio higher, utilize above formula to calculate, the result of output is still 0, and the reduced capability of variation, therefore adopts the Logistic equation of random variation, be shown below,

$\left\{\begin{array}{cc}{y}_{n+1}^k=\mathrm{\&mu;}m(1-m)& m<\mathrm{\&alpha;}\\ y_{n+1}^k=0& m\&GreaterEqual;\mathrm{\&alpha;}\end{array}\right.$

In formula, μ is control variables, generally gets μ=4; M=rand () is the random number on [0,1]; α is random factor, gets the constant being less than 1, generally gets 0.15.

Three, chaos sequence is returned former solution space by following formula inverse mapping, produce a Chaos Variable feasible solution sequence;

$p_g^k=(y_1^k\&times;R,y_2^k\&times;R,y_3^k\&times;R,...,y_m^k\&times;R)$

Four, the adaptive value of each feasible solution in above-mentioned feasible solution sequence is calculated, and feasible solution corresponding when retaining adaptive value optimum, be designated as stochastic choice particle from current particle group, uses position replaces the position of the particle selected.

Step 7: judge whether to reach maximum iteration time, if do not reach, forward step a to, proceed, if reach, forward step f to;

Step 8: export optimum programming scheme.

When power distribution network increases photo-voltaic power supply, need to determine its rational on-position and capacity; The present invention takes into full account trend, the constraint such as voltage and electric current of power distribution network, adopt the principle planning photo-voltaic power supply of economy optimum, and utilize modified particle swarm optiziation to carry out solving calculating, avoid the middle early problem of conventional particle group algorithm, and then calculate best on-position and the capacity of photo-voltaic power supply, reduce the via net loss of power distribution network, decrease the operating cost of power distribution network, improve the voltage of system, enhance the load capacity of system; Can not only ensure that power distribution network runs safely and effectively, its economic benefit can also be made to be further improved.

Claims (2)

1. the distribution network planning method containing photo-voltaic power supply, it is characterized in that, planing method mainly comprise photo-voltaic power supply year investment operating cost, network year wear and tear expense, the power purchase expense of superior electrical network and the photo-voltaic power supply social cost of saving because of policy subsidy, comprise the following steps:

(1) target function of social cost is set up:

${\mathrm{minC}}_{total}=C_p+C_{loss}+C_{en}-C_P^U$

In formula, C _totalfor social cost; C _pfor the year investment operating cost of photo-voltaic power supply; C _lossfor network year wear and tear expense; C _enfor the power purchase expense of superior electrical network; for the social cost that photo-voltaic power supply is saved because of policy subsidy;

A, set up photo-voltaic power supply year investment operating cost C _pfunction:

$C_P=\underset{i=1}{\overset{n_p}{\&Sigma;}}({\mathrm{\&alpha;C}}_{pi}+W_{pi})$

$\mathrm{\&alpha;}=\frac{\mathrm{\&gamma;}{(1+\mathrm{\&gamma;})}^t}{{(1+\mathrm{\&gamma;})}^t-1}$

In formula, n _pfor accessing the photo-voltaic power supply number of power distribution network; α is the year investment average cost coefficient of photo-voltaic power supply; C _piit is the investment cost of i-th photo-voltaic power supply; W _piit is the operation and maintenance cost of i-th photo-voltaic power supply; γ is Annual Percentage Rate; T is the planning time limit;

B, set up network year wear and tear expense C _lossfunction:

$C_{loss}=\underset{i=1}{\overset{n_l}{\&Sigma;}}{C}_0{\mathrm{\&tau;}}_{1\max }{\mathrm{\&Delta;P}}_{li}$

In formula, n _lfor distribution network line number; C ₀for unit electricity price (unit/kWh); τ _1xamfor annual peak load loss hourage (h) of branch road; Δ P _liit is the active loss of i-th branch road;

C, set up the power purchase expense C of superior electrical network _enfunction:

C _en＝T _max×(P _Σ-P _ΣP)×C ₀

In formula, T _maxfor peak load annual utilization hours; P _Σfor the total burden with power of power distribution network; P _{Σ P}exert oneself for photo-voltaic power supply is always meritorious;

D, set up photo-voltaic power supply because of policy subsidy save social cost function:

$C_P^U=C_1\underset{i=1}{\overset{n_p}{\&Sigma;}}{P}_{\mathrm{\&Sigma;}P}{\mathrm{\&tau;}}_{2max}$

In formula, C ₁for public subsidies cost (unit/kWh); τ _2xamfor year maximum illumination utilizes hourage;

(2) determine constraints, constraints mainly comprises:

E, node voltage retrain:

U _imin≤U _i≤U _imax

In formula, U _ifor the magnitude of voltage of i node; U _iminfor node voltage lower limit; U _imaxfor node voltage higher limit;

F, tributary capacity retrain:

P _j≤P _jmax

In formula, P _jfor flowing through the active power of circuit j; P _jmaxfor the higher limit of jth bar line transmission active power;

G, power distribution network access photo-voltaic power supply capacity-constrained:

S _Pi≤S _imax

S _ΣP≤S _L

In formula, S _pifor the capacity of the photo-voltaic power supply of i node access; S _imaxthe maximum size value of the photo-voltaic power supply of access is allowed for i node; S _{Σ P}for accessing the total capacity of photo-voltaic power supply in power distribution network; S _lfor 10% of distribution network load total amount;

(3) utilize the particle cluster algorithm containing chaos optimization to solve planing method, and export optimum programming scheme, determine best on-position and the capacity of photo-voltaic power supply.

2. the distribution network planning method containing photo-voltaic power supply according to claim 1, is characterized in that, the described solution procedure of particle cluster algorithm to planing method containing chaos optimization mainly comprises:

Step 1: input initial data, mainly comprises node and the circuit relevant parameter of power distribution network, particle number of groups, Studying factors c ₁and c ₂, maximum iteration time, inertia weight ω, population maximum update speed v _max, control variables μ, random factor α;

Step 2: in search volume, random generation M particle, initial population is [P ₁, P ₂, P ₃..., P _i..., P _m], use Load flow calculation to judge, P _ifor meeting arbitrary feasible solution of planing method;

Step 3: the fitness calculating each particle, the individual extreme value P of record current particle _i, and therefrom select global extremum P _g;

Step 4: utilize following formula to carry out evolution to particle and upgrade, and calculate the individual extreme value P of more new particle _iwith global extremum P _g;

$v_{i,d}^{k+1}={\mathrm{\&omega;v}}_{i,d}^k+c_1\&CenterDot;rand\left(\right)\&CenterDot;(p_{i,d}^k-x_{i,d}^k)+c_{2}rand\left(\right)\&CenterDot;(p_{g,d}^k-x_{i,d}^k)$

$x_{i,d}^{k+1}=x_{i,d}^k+v_{i,d}^{k+1}$

In formula, ω is inertia weight; c ₁and c ₂for Studying factors;

Step 5: judge that whether optimal particle is identical with adjacent three generations, if identical, carry out steps d chaos optimization; If different, jump to step e;

Step 6: the optimal location of population is mapped to Logistic function, produce Chaos Variable sequence, then former solution space is returned in inverse mapping, calculate the adaptive value of each feasible solution in above-mentioned feasible solution sequence, and retain adaptive value optimum time corresponding feasible solution, to replace in population the particle that is random;

Step 7: judge whether to reach maximum iteration time, if do not reach, forward step a to, proceed, if reach, forward step f to;

Step 8: export optimum programming scheme.