CN103986193B - A kind of method that maximum wind grid connection capacity obtains - Google Patents

Abstract

The invention discloses a kind of method that maximum wind grid connection capacity obtains, comprise S1 foundation and can consider the maximum wind grid connection capacity Optimized model that peak regulation is abundance; In model, target function is for target with system grid connection wind-powered electricity generation maximum capacity, constraints comprises the node active power balance constraint of system, the branch road DC power flow active power constraint of system, tributary capacity constraint, conventional power generation usage unit units limits, every day abandon the constraint of wind total amount, cutting load probability and year cutting load Expectation constraint, abandon wind probability and abandon wind Expectation constraint year; Wind-powered electricity generation receives the nonlinear equation in Optimized model to be converted to linear equation by S2, and obtains linear maximum wind grid connection capacity Optimized model; S3 obtains the heap(ed) capacity of wind-electricity integration according to the grid supplemental characteristic obtained, history wind power output data and described linear maximum wind grid connection capacity Optimized model.The present invention introduces abundance Index Constraints and is optimized wind capacity integrated into grid, and the maximum wind grid connection capacity obtained meets the requirement that peak regulation is abundance and generating is abundance; Be linear equation by the non-linear equation in institute's Modling model, significantly reduce computing time.

Description

A kind of method that maximum wind grid connection capacity obtains

Technical field

The invention belongs to the new energy field of electric power system, more specifically, relate to a kind of method that maximum wind grid connection capacity obtains.

Background technology

Current calculating maximum wind grid connection capacity is studied widely.Mainly be divided into four classes: engineering method, dynamic simulation method, restraining factors analytic approach, optimization.Engineering method adopts in mainly calculating in early days, and because its calculating can only the maximum grid connection capacity of approximate estimation wind-powered electricity generation, institute is more unilateral and rough in this way.The various working that dynamic simulation method generally may occur for electric power system carries out Digital Simulation, and the wind capacity integrated into grid limit is determined in the impact such as the voltage fluctuation that these class methods cause electric power system according to wind-electricity integration, the quality of power supply, and amount of calculation is larger.Restraining factors analytic approach considers certain class factor of the maximum grid connection capacity of restriction wind-powered electricity generation, be such as restraining factors with voltage stabilization, the maximum grid connection capacity of wind-powered electricity generation is calculated by P-V (meritorious-voltage) curve, because this method is difficult to consider multiple restraining factors, so have very large limitation in application.Optimization is a kind of method extensively adopted in recent years, and the method can consider multiple restraining factors affecting wind-electricity integration.This method is receptive for electrical network wind-powered electricity generation calculation of capacity as being in the maximized problem of installed capacity of wind-driven power solving Problem with Some Constrained Conditions restriction, thus is solving of optimal problem problem arises.But current optimization method, does not consider in optimizing process that the peak regulation of system is abundance and generating is abundance, make to calculate can be grid-connected wind-powered electricity generation capacity not accurate enough.

At present when planning wind capacity integrated into grid, if the abundance index of systems organization will be considered, be to ask for the maximum access capacity of grid connected wind power mostly in conjunction with Monte Carlo simulation approach.The flow process of the method is as follows: sample to system mode, then simulate by hour management and running, calculate the wind power output time series data under different installed capacity of wind-driven power respectively, and then add up every abundance index, as year generation deficiency probability, peak regulation shortfall probability etc.Afterwards using wherein one or more index as the foundation passing judgment on grid connected wind power access capacity, select the maximum grid connection capacity of the wind-powered electricity generation under given index.This method first supposes wind capacity integrated into grid, then calculate abundance index, if can not meet the abundance requirement of power source planning, then calculates after will adjusting wind capacity integrated into grid again.If the step-length of each adjustment wind capacity integrated into grid is very little, although can obtain comparatively accurate wind power output timing curve, amount of calculation is large especially.Step-length is excessive, and the wind power output timing curve obtained is a kind of "ball-park" estimate, and final like this wind-powered electricity generation of trying to achieve maximum grid connection capacity accuracy is not high, and amount of calculation is very large.

Summary of the invention

For the defect of prior art, the object of the present invention is to provide a kind of method obtaining maximum wind grid connection capacity, be intended to solve when planning wind capacity integrated into grid, cannot take into account in optimizing process peak shaving abundance and generating abundance Index Constraints, thus make result of calculation inaccurate, the problem that computational efficiency is low.

The invention provides a kind of method that maximum wind grid connection capacity obtains, comprise the steps:

S1: set up the maximum wind grid connection capacity Optimized model can considering abundance constraint;

Target function in described maximum wind grid connection capacity Optimized model is with system grid connection wind-powered electricity generation maximum capacity for target, and described bound for objective function comprises system cloud gray model constraint and abundance Index Constraints;

Described system cloud gray model constraint comprises the node active power balance constraint of system, the branch road DC power flow active power constraint of system, tributary capacity constraint, the actual units limits of conventional power generation usage unit, every day abandon wind total amount and retrain;

The inequality constraints of described abundance index comprise cutting load probability and year cutting load Expectation constraint, abandon wind probability and abandon wind Expectation constraint year;

S2: the nonlinear equation in described maximum wind grid connection capacity Optimized model is converted to linear equation, and obtain linear maximum wind grid connection capacity Optimized model;

S3: obtain maximum wind grid connection capacity according to the grid supplemental characteristic obtained, history wind power output data and described linear maximum wind grid connection capacity Optimized model.

Wherein, the node active power balance equation of system described in step S1 is constrained to s is node injecting power and branch road effective power flow incidence matrices, be the active power vector that d days t flow through branch road, element is designated as p _{d, t, ij}, p _{d, t, ij}be the active power that d days t flow through branch road i-j; g _{d, t}for normal power supplies to be gained merit force vector in the reality of d days t, element is designated as g _{d, t, i}, g _{d, t, i}for the normal power supplies of access node i is exerted oneself the reality of d days t is meritorious; u _{d, t}for grid connected wind power to be gained merit force vector in the reality of d days t, element is designated as u _{d, t, i}, u _{d, t, i}for the wind-powered electricity generation of access node i is exerted oneself the reality of d days t is meritorious; be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; r _{d, t}be the cutting load vector of d days t, element is r _{d, t, i}, r _{d, t, i}for node i is in the cutting load amount of d days t; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0; l _{d, t}be the load vector of d days t, element is designated as l _{d, t, i}, l _{d, t, i}it is the load of d days t node i; w _{d, t}for grid connected wind power abandons wind direction amount d days t, element is designated as w _{d, t, i}, w _{d, t, i}for the wind-powered electricity generation of access node i abandons air quantity d days t.

Wherein, described cutting load probability is described year cutting load expectation is described wind probability of abandoning is abandoning wind expectation described year is d is the number of days in timing statistics, and T is the statistics moment number of every day, and d is d days, t is t, be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0, with the system that is respectively in timing statistics D inscribe Load Probability and the threshold value of abandoning wind probability, with the system that is respectively in institute timing statistics D year cutting load expect and abandon in year wind expect threshold values.

Wherein said maximum wind grid connection capacity Optimized model comprises target function MaxP _wn; P _wnfor grid-connected wind-powered electricity generation capacity.

Wherein described linear maximum wind grid connection capacity Optimized model comprises in step s 2: nonlinear node active power balance is converted to linear node active power balance constraint $S\×P_{d,t}^L+g_{d,t}+u_{d,t}+r_{d,t}^y=l_{d,t}+w_{d,t}^y;$ Wherein $r_{d,t,i}^y=y_{d,t}^r{r}_{d,t},w_{d,t}^y=y_{d,t}^w{w}_{d,t}.r_{d,t,i}^y=y_{d,t}^r{r}_{d,t}$ Be equivalent to following three linear inequalities: $0\≤r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}};r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}}{y}_{d,t}^r;r_{d,t,i}-\stackrel{\‾}{r_{d,t,i}}(1-y_{d,t}^r)\≤r_{d,t,i}^y\≤r_{d,t,i};w_{d,t}^y=y_{d,t}^w{w}_{d,t}$ Be equivalent to following three linear inequalities: be the cutting load vector of d days t, element is designated as for continuous variable, represent the cutting load amount of node i d days t; it is the cutting load amount upper limit of d days t node i; be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; be d days t abandon wind direction amount, element is designated as for continuous variable, represent that grid connected wind power abandons air quantity d days t; be the wind-powered electricity generation of d days t access node i abandon the air quantity upper limit; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0; S is node injecting power and branch road effective power flow incidence matrices, be the active power vector that d days t flow through branch road, element is designated as p _{d, t, ij}, p _{d, t, ij}be the active power that d days t flow through branch road i-j; g _{d, t}for normal power supplies to be gained merit force vector in the reality of d days t, element is designated as g _{d, t, ij}, g _{d, t, ij}for the normal power supplies of access node i is exerted oneself the reality of d days t is meritorious; u _{d, t}for grid connected wind power to be gained merit force vector in the reality of d days t, element is designated as u _{d, t, i}, u _{d, t, i}for the wind-powered electricity generation of access node i is exerted oneself the reality of d days t is meritorious; l _{d, t}for the load vector d days t, element is designated as l _{d, t, i}, l _{d, t, i}it is the load of d days t node i;

The branch road DC power flow power constraint of system: p _{d, t, ij}-β _ij(θ _{d, t, i}-θ _{d, t, j})=0; p _{d, t, ij}be the active power that d days t flow through branch road i-j; β _ijfor the susceptance of branch road i-j; θ _{d, t, i}and θ _{d, t, j}be respectively node i and the node j voltage phase angle d days t;

The tributary capacity constraint of system: p _{d, t, ij}be the active power that d days t flow through branch road i-j; for the active power upper limit that every bar branch road i-j can bear;

Conventional power generation usage unit units limits: for conventional power generation usage unit maximum output, g _i for conventional power generation usage unit minimum load;

Node i abandons air quantity constraint 0≤w _{d, t, i}≤ u _{d, t, i}; w _{d, t, i}for the wind-powered electricity generation of access node i abandons air quantity d days t; w _{d, t, i}for the wind-powered electricity generation of access node i is exerted oneself the reality of d days t is meritorious;

System retrains the wind total amount of abandoning of d days w _{d, t, i}be the abandon air quantity of wind-powered electricity generation in t of d days access node i, I is the sum of system node, a and b is respectively the upper and lower bound abandoning wind total amount every day;

Cutting load and the probability constraints abandoning wind $\frac{1}{\mathrm{DT}}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}{y}_{d,t}^r\≤K_p^r$ With $\frac{1}{\mathrm{DT}}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}{y}_{d,t}^w\≤K_p^w;$ be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0; with the system that is respectively is in timing statistics D inscribe Load Probability and the threshold value of abandoning wind probability;

Cutting load and the desired value constraint of abandoning wind: $\frac{1}{D}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}\underset{i=1}{\overset{I}{\mathrm{\Σ}}}{r}_{d,t,i}\≤K_m^r$ With $\frac{1}{D}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}\underset{i=1}{\overset{I}{\mathrm{\Σ}}}{w}_{d,t,i}\≤K_m^w;$ R _{d, t, i}for node i is in the cutting load amount of d days t; w _{d, t, i}for the wind-powered electricity generation of access node i abandons air quantity d days t; with the system that is respectively in institute timing statistics D year cutting load expect and abandon in year wind expect threshold values.

Advantage of the present invention is that adopting the method optimized to overcome traditional analog method calculates the defect inaccurate, efficiency is not high, in computational process, consider cutting load probability simultaneously, cutting load is expected, abandoned wind probability and abandon the abundance indexs such as wind expectation, make the maximum wind grid connection capacity calculated can meet the requirement that the peak regulation of power source planning is abundance and generating is abundance, thus the admissible maximum wind grid connection capacity of evaluating system more accurately.In addition, the non-linear equation in maximum wind grid connection capacity Optimized model is linear equation by the present invention, greatly reduces model complexity and solves difficulty, can reduce computing time significantly like this.In sum, the present invention can instruct the power system planning of wind-electricity integration, realizes the rational and efficient use of wind energy resources, has practical significance.

Accompanying drawing explanation

Fig. 1 is that the power supply containing wind power system some day is exerted oneself and workload demand curve chart;

Fig. 2 is the active power balance schematic diagram of node;

Fig. 3 is the Garver6 node system topology schematic diagram improved;

Fig. 4 is a kind of realization flow figure obtaining the maximum grid connection capacity method of wind-powered electricity generation that the invention process case provides.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

The invention provides a kind of acquisition methods of maximum wind grid connection capacity, comprise the following steps:

(1) data are obtained;

Fig. 1 describes the power supply including wind-electricity integration system for a day for 24 hours and exerts oneself and workload demand curve, three curves are had: solid line is load curve in figure, dotted line is the maximum possible power curve of system, this curve is that the fired power generating unit maximum output of system and grid connected wind power are exerted oneself sum, imaginary point line is the minimum load curve of system, and this curve is that the fired power generating unit minimum load of system and grid connected wind power are exerted oneself sum.In figure, variable capacity refers to the maximum output of the fired power generating unit of system and the difference of minimum load.

Shown in figure, peak regulation deficiency is divided into two time periods: the maximum output of system is less than the time period of system load demand, and system minimum load is greater than the time period of system load demand.The former belongs to the generation deficiency time period, can be operated the power-balance of the system of guarantee in this time period by cutting load, therefore, adopts cutting load index to assess systems generate electricity deficiency; The latter belongs to the generating superfluous time period, and can make the power-balance of the system that ensures by abandoning character and conduct in this time period, employing is abandoned wind index and assessed systems generate electricity surplus.

It can also be seen that from figure, when grid connected wind power capacity increases, grid connected wind power is exerted oneself increase, dotted line then in figure and imaginary point line will on move, obvious cutting load probability and cutting load amount can reduce along with dotted line moves, and abandon wind probability and abandon air quantity and will increase along with imaginary point line moves, so cutting load index and the value of abandoning wind index can limit the access capacity of grid connected wind power.In like manner, when cutting load index and abandon wind index determine time, when namely in figure, the area of peak regulation deficiency is determined, fired power generating unit maximum output and minimum load can affect again the access capacity of grid connected wind power.

From analyzing above, the maximum access capacity of grid connected wind power be calculated, needing to collect following data:

First gathering system network parametric data.Grid parameter comprises node i, node total number I, node injecting power and branch road effective power flow incidence matrices s, line reactance the circuit effective power flow upper limit each moment accesses the load l of each node _{d, t, i}.

Then the power data of gathering system.Access the conventional generator maximum output restriction of each node limit with minimum load g _i , each each moment of node the cutting load amount upper limit each moment of grid connected wind power abandon the air quantity upper limit abandon the upper limit a of wind total amount every day, abandon the lower limit b of wind total amount every day.

After ensureing wind power integration system, electrical network can continue to safely and steadily run, and power system planning department can determine the span of abundance index, calculates the threshold value of four abundance indexs according to the span of these indexs---and abandon wind and expect threshold value, cutting load expect threshold value, abandon wind probability threshold value, cutting load probability threshold value.

Finally because the present invention simulates actual wind power output according to the wind power output of history, therefore need the wind power output data of collecting history with the wind capacity integrated into grid of history

(2) foundation can consider the maximum wind grid connection capacity Optimized model that peak regulation is abundance;

In order to ensure effective utilization of wind energy, optimization aim is decided to be the grid connected wind power maximum capacity of system.And bound for objective function comprises the inequality constraints running constraint and abundance index, wherein run that constraint comprises the node active power balance constraint of system, the branch road DC power flow active power constraint of system, tributary capacity constraint, conventional power generation usage unit units limits, every day abandon wind total amount and retrain, the inequality constraints of abundance index comprise cutting load probability and year cutting load Expectation constraint, abandon wind probability and abandon wind Expectation constraint year.

(2.1) target function

MaxP _wn(1); Target function makes system access wind-powered electricity generation capacity maximum.In formula (1), P _wnfor the access capacity of grid connected wind power, unit: MW.

(2.2) constraints

Set up the node active power balance constraint of system, the constraint of branch road DC power flow active power, tributary capacity constraint, the conventional power generation usage unit units limits of system successively according to the grid supplemental characteristic collected and abandon wind total amount every day below and retrain and the practical significance describing each constraint in detail.

A () node active power balance retrains:

In system, the power-balance situation of each node is as Fig. 2, and the active power that this chart free flow crosses node i is made up of four parts:

Part I is the actual g that exerts oneself of the normal power supplies of access node i _i;

Part II is wind-driven generator or the actual u that exerts oneself of wind energy turbine set of access node i _i-w _i, wherein u _ifor wind-driven generator or the output of wind electric field of access node i, w _iair quantity is abandoned for this wind-driven generator or wind energy turbine set;

Part III is the active power l flowing out node i _i-r _i, wherein l _ifor the load of access node i, r _ifor cutting load amount;

Part IV is the active power S crossing node i from line flows _i× P ^l, wherein S _ifor injecting power and the branch road active power Associate array of node i, its element is s _ijif, s _iwith s _jbetween exist 1 by s _ipoint to s _jdirected edge time, then s _ij=1, otherwise, s _ij=-1, if s _iwith s _jbetween there is no connection, then s _ij=s _ji=0.Associate array s _imiddle element representation is:

P ^lfor branch road effective power flow array, element is designated as p _ij, s _ij× p _ijfor timing, represent that active power flows into node i from node j, s _ij× p _ijfor time negative, represent that active power flows out to node j from node i.Know that the power flowing into node i should equal to flow out the power of node i by figure, the power balance equation therefore setting up node i is as follows: S _i× P ^l+ g _i+ u _i-w _i=l _i-r _i(2)

In the running of electric power system reality, the active power flowing through each node is constantly change in time, but the active power flowing into each node equals the active power flowing out this node all the time, therefore can obtain the node power equilibrium equation in each moment by equation (2) is:

$S\×P_{d,t}^L+g_{d,t}+u_{d,t}+y_{d,t}^r=l_{d,t}+y_{d,t}^w{w}_{d,t}---\left(3\right)$

In equation (3), d represents sky, and t represents the moment, and s is node injecting power and branch road effective power flow incidence matrices, be the active power vector that d days t flow through branch road, element is designated as p _{d, t, ij}, p _{d, t, ij}be the active power that d days t flow through branch road i-j; g _{d, t}for normal power supplies to be gained merit force vector in the reality of d days t, element is designated as g _{d, t, i}, g _{d, t, i}for the normal power supplies of access node i is exerted oneself the reality of d days t is meritorious; u _{d, t}for grid connected wind power to be gained merit force vector in the reality of d days t, element is designated as u _{d, t, i}, u _{d, t, i}for the wind-powered electricity generation of access node i is exerted oneself the reality of d days t is meritorious; be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, and the wind-powered electricity generation of access node i does not abandon wind d days t, value is 0; r _{d, t, i}for node i is in the cutting load amount of d days t; l _{d, t}at the load vector of d days t, element is designated as l _{d, t, i}, l _{d, t, i}it is the load of d days t node i; w _{d, t}for grid connected wind power abandons wind direction amount d days t, element is designated as w _{d, t, i}, w _{d, t, i}for the wind-powered electricity generation of access node i abandons air quantity d days t.Wherein s, l _{d, t}for known quantity, remaining variables is amount to be asked.

The branch road DC power flow active power constraint of (b) system:

In electric power system tide calculates, the effective power flow of branch road ij can be expressed as

in formula (4): P _ij, G _ij, B _ijbe respectively the active power of branch road ij, conductance and susceptance; U _i, U _jbe respectively the voltage magnitude of node i, node j; θ _ijfor the phase angle difference of node i and node j.Now suppose: the normal each node voltage of electric power system run, usually near rated voltage, can think U approx _i=U _j=1; Line resistance is far smaller than line reactance, therefore ignores line resistance; There is not heavy load circuit in system, namely circuit both end voltage phase angle difference is very little, has sin θ _ij≈ θ _i-θ _j.The DC power flow power-balance set up thus between node i and node j retrains as follows:

P _ij-β _ij(θ _i-θ _j)=0 (5); P in formula (5) _ijfor the effective power flow of branch road i-j, β _ijfor the susceptance of branch road i-j, θ _iand θ _jfor the voltage phase angle of node i and node j.

The tributary capacity constraint of (c) system: because the effective power flow passed through in circuit can not exceed the effective power flow upper limit that circuit can bear, therefore set up inequality as follows:

In formula (6) for the effective power flow upper limit that every bar branch road i-j can bear;

(d) conventional power generation usage unit units limits.Exerting oneself of conventional power generation usage unit should be reasonably interval at one, therefore sets up inequality as follows:

In formula (7) for conventional power generation usage unit maximum output, g _i for the minimum load of conventional power generation usage unit.

E () abandons air quantity constraint.Abandoning the air quantity upper limit is the actual u that exerts oneself of wind-powered electricity generation _{d, t}, the air quantity of abandoning in every day and per moment in rational scope, therefore should be set up node i and abandons air quantity constraint: 0≤w _{d, t, i}≤ u _{d, t, i}(8); w _{d, t, i}be the abandon air quantity of wind-powered electricity generation in t of d days access node i, system retrains the wind total amount of abandoning of d days in formula (9), I is the sum of system node, a and b is respectively the upper and lower bound abandoning wind total amount every day.

(2.3) inequality constraints of abundance index

According to abandoning that wind is expected year, year cutting load expect, abandon wind probability, the threshold value of cutting load probability sets up the abundance constraint of probability and expects abundance constraint.

The abundance constraint of probability.Cutting load probability in one section of timing statistics and abandon wind probability and should control in certain limit to meet the abundance requirement of system.Therefore inequality constraints is set up such as formula (10) and formula (11); $\frac{1}{\mathrm{DT}}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}{y}_{d,t}^r\≤K_p^r,\left(10\right);$ $\frac{1}{\mathrm{DT}}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}{y}_{d,t}^w\≤K_p^w,\left(11\right);$ Introduce the unified optimization that these two abundance constraints can realize wind-powered electricity generation capacity and abundance index, make the maximum access capacity of wind-electricity integration obtained can meet cutting load and abandon the probability level of wind.

Expect abundance constraint.Cutting load every day in one section of timing statistics is expected and abandons wind every day to expect control in certain limit to meet the abundance requirement of system.Therefore inequality constraints is set up such as formula (12) and formula (13); $\frac{1}{D}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}\underset{i=1}{\overset{I}{\mathrm{\Σ}}}{r}_{d,t,i}\≤K_m^r,\left(12\right);$ $\frac{1}{D}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}\underset{i=1}{\overset{I}{\mathrm{\Σ}}}{w}_{d,t,i}\≤K_m^w,\left(13\right);$ Introduce the unified optimization that these two abundance constraints can realize wind-powered electricity generation capacity and abundance index, make the maximum access capacity of wind-electricity integration obtained can meet cutting load and abandon the expectation index of wind. for the threshold value of timing statistics D inscribe Load Probability; for abandoning the threshold value of wind probability in timing statistics D; for the threshold value of year cutting load expectation in timing statistics D; for abandoning the threshold value that wind is expected in timing statistics D year.

The present invention, according to the wind power output situation of the wind power output digital simulation reality of the history collected, first calculates the wind power output perunit value of history, then is multiplied by wind-powered electricity generation capacity to be planned, as the wind power output of reality.Namely equation is set up for the wind power output data of history, wherein d, t represent the data that d days t are got, and i represents the node of wind power integration, for the installed capacity of wind-driven power of history, wherein wn represents blower fan rated capacity.

(3) nonlinear equation in wind-electricity integration model is converted to Linear inequalities

In equation (3) with be the product of a binary variable and a continuous variable, thus equation (3) is MIXED INTEGER nonlinear equation, and the linearization step of this equation is as follows:

Introduce new linear variable with them are made to meet equation by linear new variables with substitute into equation (3) and produce new equation (17).

all variablees of equation (17) are all linear, and therefore this new equation is linear equation.

Afterwards nonlinear equation (15) and (16) are converted to six Linear inequalities as follows.

$0\≤r_{d,t,i}^y\≤\stackrel{\‾}{y_{d,t,i}},\left(18\right);r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}}{y}_{d,t}^r,\left(19\right);r_{d,t,i}-\stackrel{\‾}{r_{d,t,i}}(1-y_{d,t}^r)\≤r_{d,t,i}^y\≤r_{d,t,i},\left(20\right);$ $0\≤w_{d,t,i}^y\≤\stackrel{\‾}{w_{d,t,i}},\left(21\right);w_{d,t,i}^y\≤\stackrel{\‾}{w_{d,t,i}}{y}_{d,t}^w,\left(22\right);w_{d,t,i}-\stackrel{\‾}{w_{d,t,i}}(1-y_{d,t}^w)\≤w_{d,t,i}^y\≤w_{d,t,i},\left(23\right);$ Wherein with be respectively r _{d, t, i}and w _{d, t, i}the upper limit, also can be greater than r with one _{d, t, i}and w _{d, t, i}several M of the upper limit replace.

When dispatcher determines to carry out cutting load operation, namely time, inequality group (18)-(20) become $\left[\begin{array}{c}0{\≤r}_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}}\\ r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}}\\ r_{d,t,i}\≤r_{d,t,i}^y\≤r_{d,t,i}\end{array}\right],\left(24\right);$ In formula, the first two equation shows the cutting load variable of each moment t node i all the time its cutting load higher limit is less than 3rd formula shows and time equation become obvious inequality group (24) is of equal value with equation (15).

When dispatcher does not carry out cutting load operation, namely time, inequality group (18)-(20) become $\left[\begin{array}{c}0\≤r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}}\\ r_{d,t,i}^y\≤0\\ r_{d,t,i}-\stackrel{\‾}{r_{d,t,i}}\≤r_{d,t,i}^y\≤r_{d,t,i}\end{array}\right],\left(25\right);$ Show in formula $r_{d,t,i}^y=0,$ And $y_{d,t}^r=1$ Time equation $\left(15\right),r_{d,t,i}^y=y_{d,t}^r{r}_{d,t,i}$ Become obvious inequality group (25) is of equal value with equation (15).

No matter binary variable in sum value 0 or 1, inequality group (18)-(20) are all of equal value with equation (15).In like manner, no matter binary variable value 0 or 1, inequality group (21)-(23) are all of equal value with equation (16).And all variable in inequality group (18)-(23) is all linear, thus inequality group (18)-(23) are all linear, now all equations of whole model are all linear, therefore can obtain the globally optimal solution of model fast.

It is as follows that maximum wind grid connection capacity after conversion optimizes linear model:

Target function: MaxP _wn;

Linear node active power balance constraint:

Wherein be translated into three linear inequalities of equal value:

$0\≤r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}};r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}}{y}_{d,t}^r;r_{d,t,i}-\stackrel{\‾}{r_{d,t,i}}(1-y_{d,t}^r)\≤r_{d,t,i}^y\≤r_{d,t,i};$

be translated into three linear inequalities of equal value:

$0\≤w_{d,t,i}^y\≤\stackrel{\‾}{w_{d,t,i}};w_{d,t,i}^y\≤\stackrel{\‾}{w_{d,t,i}}{y}_{d,t}^w;w_{d,t,i}-\stackrel{\‾}{w_{d,t,i}}(1-y_{d,t}^w)\≤w_{d,t,i}^y\≤w_{d,t,i};$

The branch road DC power flow power constraint of system: p _{d, t, ij}-β _ij(θ _{d, t, i}-θ _{d, t, j})=0;

The tributary capacity constraint of system:

Conventional power generation usage unit units limits:

Node i abandons air quantity constraint 0≤w _{d, t, i}≤ u _{d, t, i};

System retrains the wind total amount of abandoning of d days

Cutting load and the probability constraints abandoning wind $\frac{1}{\mathrm{DT}}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}{y}_{d,t}^r\≤K_p^r$ With $\frac{1}{\mathrm{DT}}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}{y}_{d,t}^w\≤K_p^w;$

Cutting load and the desired value constraint of abandoning wind: $\frac{1}{D}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}\underset{i=1}{\overset{I}{\mathrm{\Σ}}}{r}_{d,t,i}\≤K_m^r$ With $\frac{1}{D}\underset{d=1}{\overset{D}{\mathrm{\Σ}}}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}\underset{i=1}{\overset{I}{\mathrm{\Σ}}}{w}_{d,t,i}\≤K_m^w.$

In model, known parameters is as follows: s is node injecting power and branch road effective power flow incidence matrices; l _{d, t}for the load vector d days t, element is designated as l _{d, t, i}, l _{d, t, i}it is the load (unit: MW) of d days t node i; it is the cutting load amount upper limit of d days t node i; be the wind-powered electricity generation of d days t access node i abandon the air quantity upper limit; β _ijfor the susceptance of branch road i-j; for the effective power flow upper limit (unit: MW) that every bar branch road i-j can bear; for conventional power generation usage unit maximum output (unit: MW), g _i for conventional power generation usage unit minimum load (unit: MW); A and b is respectively the upper and lower bound (unit: MW) abandoning wind total amount every day. for system is in the threshold value of timing statistics D inscribe Load Probability; for system abandons the threshold value of wind probability in timing statistics D; for the threshold value of system year cutting load expectation in timing statistics D; for system abandons the threshold value of wind expectation year in timing statistics D.

The parameter that will solve in model is as follows: be the active power vector that d days t flow through branch road, element is designated as p _{d, t, ij}, p _{d, t, ij}be the active power (unit: MW) that d days t flow through branch road i-j; be the cutting load vector of d days t, element is designated as for continuous variable, represent the cutting load amount (unit: MW) of node i d days t; be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; be d days t abandon wind direction amount, element is designated as for continuous variable, represent that grid connected wind power abandons air quantity d days t; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0; β _ijfor the susceptance of branch road i-j; θ _{d, t, i}and θ _{d, t, j}be respectively node i and the node j voltage phase angle (unit: °) d days t; w _{d, t, i}for the wind-powered electricity generation of access node i abandons air quantity (unit: MW) d days t; for determining the binary variable whether carrying out cutting load operation, value 0 or 1; for determining the binary variable whether carrying out abandoning character and conduct work, value 0 or 1; g _{d, t}for normal power supplies to be gained merit force vector in the reality of d days t, element is designated as g _{d, t, i}, g _{d, t, i}for the normal power supplies of access node i exerts oneself (unit: MW) the reality of d days t is meritorious; u _{d, t}for grid connected wind power to be gained merit force vector in the reality of d days t, element is designated as u _{d, t, i}, u _{d, t, i}for the wind-powered electricity generation of access node i exerts oneself (unit: MW) the reality of d days t is meritorious; P _wnfor grid connected wind power capacity (unit: MW).

The present invention has 2 innovations, innovation one: in the maximum grid connection capacity Optimized model of traditional wind-powered electricity generation, add four abundance Index Constraints, thus establish the maximum wind grid connection capacity Optimized model considering that peak shaving is abundance and generating is abundance, this model can use Solution of Optimization can meet the maximum wind grid connection capacity of the abundance and abundance requirement of generating of the peak regulation of system, and do not need to use the method soundd out to iterate, while raising result of calculation accuracy, greatly reducing computing time.Innovation two: nonlinear node power equilibrium equation is converted into linear equation of equal value with it and Linear inequalities, nonlinear maximum wind grid connection capacity optimization problem is made to become linear planning problem, further improve the efficiency of calculating, more adequately can be met the maximum wind grid connection capacity of the abundance demand of power source planning according to the optimal solution obtained.Can to be made the planning of wind capacity integrated into grid by above 2 known the present invention of innovation and instructing accurately, there is very strong actual use value.

(4) data are substituted into model, calculate

For existing methodical deficiency, the present invention establishes the maximum access capacity of the wind-powered electricity generation based on an abundance constraint model, and this model not only comprises cutting load constraint, further comprises and directly can reflect that wind power integration and the interactional wind of abandoning of system retrain.Be the linear equation with full scale equation equivalence by non-linear equation in model afterwards, thus the optimization method of linear programming can be used to solve target function, therefore computational methods of the present invention have the advantage that precision is high, computational speed is fast.

Below in conjunction with accompanying drawing, the invention process case is described in further detail.

Implementation step 1: obtain measured data and parameter.The implementation case is described in detail to the acquisition methods of the maximum access capacity of wind-electricity integration on the Garver6 node system improved.As shown in Figure 3, system contains 3 generators, 8 circuits, 1 Fans to the structure of system.Collect history wind power output data, system parameters according to implementation step 1, concrete measured data is as shown in table 1-table 4.

Table 1 network parameter

The each node motor of table 2 is exerted oneself and workload demand

Table 3 constrained parameters

Abandon air quantity constrained parameters: constant a gets 0, b and gets 20%*P _wn, namely grid connection capacity 20%.

Table 4 node injecting power and branch road effective power flow incidence matrices S

Node i	1-2	1-4	1-5	2-3	2-4	2-6	3-5	4-6
									1	-1	-1	1	0	0	0	0	0
2	1	0	0	1	-1	-1	0	0
									3	0	0	0	-1	0	0	1	0 10 -->
4	0	1	0	0	1	0	0	-1
									5	0	0	-1	0	0	0	-1	0
6	0	0	0	0	0	1	0	1

Table 1 gives analogue system network parameter, wherein can reach each node to make wind energy turbine set or wind-driven generator exert oneself, by the effective power flow upper limit of every bar branch road be set near 800MW.Each node generator output and workload demand as shown in table 2, node 1,3,6 access motor, node 6 does not connect load.

This simulation example timing statistics is 1 year, i.e. D=365, and in 1 year, every 15 minutes of every day gathered data, within one day, gathers 96 data, i.e. T=96, node total number I=6, the cutting load amount upper limit get 500MW; Abandon the air quantity upper limit get 500MW; The wind power output data of history get the value of a year, history wind capacity integrated into grid for 12200MW.

Implementation step 2: set up and can consider that the maximum wind of abundance constraint receives model.

According to the data Modling model collected.

Implementation step 3: received by maximum wind nonlinear equation in model to be converted to Linear inequalities.

Equation in model (4) is converted to Linear inequalities, sets up new maximum wind and receive model.

Implementation step 4: substitute into data, computation model.

The implementation case uses GAMS (GeneralAlgebraicModelingSystem) software transfer gurobi solver to solve, the each stage following 1. using gurobi to solve MILP solves the stage in advance: simplified model, cancellation redundant constaint, decision problem whether unbounded, or infeasible.

2. solve the stage: utilize heuritic approach to obtain integer feasible solution, obtained a lower bound of former problem by root method of relaxation, finally use branch-bound algorithm, find the optimal solution of former problem.

3. aggregation stages: complete when solving, what export MILP optimization engine solves information.

Wholely solve flow process as shown in Figure 4, obtaining the maximum access capacity of wind-electricity integration after calculating is 374.85MW.

Those skilled in the art will readily understand; the foregoing is only preferred embodiment of the present invention; not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.

Claims (4)

1. a method for maximum wind grid connection capacity acquisition, is characterized in that, comprise the steps:

S1: set up the maximum wind grid connection capacity Optimized model can considering abundance constraint;

Target function in described maximum wind grid connection capacity Optimized model is with system grid connection wind-powered electricity generation maximum capacity for target, and described bound for objective function comprises system cloud gray model constraint and abundance Index Constraints;

Described system cloud gray model constraint comprises the node active power balance constraint of system, the branch road DC power flow active power constraint of system, tributary capacity constraint, the actual units limits of conventional power generation usage unit, every day abandon wind total amount and retrain;

Described abundance Index Constraints comprise cutting load probability and year cutting load Expectation constraint, abandon wind probability and abandon wind Expectation constraint year;

S2: the nonlinear equation in described maximum wind grid connection capacity Optimized model is converted to linear equation, and obtain linear maximum wind grid connection capacity Optimized model;

S3: obtain maximum wind grid connection capacity according to the grid supplemental characteristic obtained, history wind power output data and described linear maximum wind grid connection capacity Optimized model;

The node active power balance equation of system described in step S1 is constrained to $S\×P_{d,t}^L+g_{d,t}+u_{d,t}+y_{d,t}^r{r}_{d,t}=l_{d,t}+y_{d,t}^w{w}_{d,t};$

S is node injecting power and branch road effective power flow incidence matrices, be the active power vector that d days t flow through branch road, element is designated as p _{d, t, ij}, p _{d, t, ij}be the active power that d days t flow through branch road i-j; g _d,tfor normal power supplies to be gained merit force vector in the reality of d days t, element is designated as g _{d, t, i}, g _{d, t, i}for the normal power supplies of access node i is exerted oneself the reality of d days t is meritorious; u _d,tfor grid connected wind power to be gained merit force vector in the reality of d days t, element is designated as u _{d, t, i}, u _{d, t, i}for the wind-powered electricity generation of access node i is exerted oneself the reality of d days t is meritorious; be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; r _d,tbe the cutting load vector of d days t, element is r _{d, t, i}, r _{d, t, i}for node i is in the cutting load amount of d days t; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0; l _d,tbe the load vector of d days t, element is designated as l _{d, t, i}, l _{d, t, i}it is the load of d days t node i; w _d,tfor grid connected wind power abandons wind direction amount d days t, element is designated as w _{d, t, i}, w _{d, t, i}for the wind-powered electricity generation of access node i abandons air quantity d days t.

2. the method for claim 1, is characterized in that, described cutting load probability is described year cutting load expectation is described wind probability of abandoning is $\frac{1}{DT}\underset{d=1}{\overset{D}{\Σ}}\underset{t=1}{\overset{T}{\Σ}}{y}_{d,t}^w\≤K_p^w;$ Abandoning wind expectation described year is $\frac{1}{D}\underset{d=1}{\overset{D}{\Σ}}\underset{t=1}{\overset{T}{\Σ}}\underset{i=1}{\overset{I}{\Σ}}{w}_{d,t,i}\≤K_m^w;$

D is the number of days in timing statistics, and T is the statistics moment number of every day, and d is d days, t is t, be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0, with the system that is respectively in timing statistics D inscribe Load Probability and the threshold value of abandoning wind probability, with the system that is respectively in institute timing statistics D year cutting load expect and abandon in year wind expect threshold values.

3. the method for claim 1, is characterized in that, described maximum wind grid connection capacity Optimized model comprises target function MaxP _wn; P _wnfor grid-connected wind-powered electricity generation capacity.

4. the method for claim 1, is characterized in that, described linear maximum wind grid connection capacity Optimized model comprises in step s 2:

Nonlinear node active power balance is converted to linear node active power balance constraint $S\×P_{d,t}^L+g_{d,t}+u_{d,t}+r_{d,t}^y=l_{d,t}+w_{d,t}^y;$ Wherein be equivalent to following three linear inequalities: $0\≤r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}};r_{d,t,i}^y\≤\stackrel{\‾}{r_{d,t,i}}{y}_{d,t}^r;r_{d,t,i}-\stackrel{\‾}{r_{d,t,i}}(1-y_{d,t}^r)\≤r_{d,t,i}^y\≤r_{d,t,i};w_{d,t}^y=y_{d,t}^w{w}_{d,t}$ Be equivalent to following three linear inequalities: be the cutting load vector of d days t, element is designated as for continuous variable, represent the cutting load amount of node i d days t; it is the cutting load amount upper limit of d days t node i; be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; be d days t abandon wind direction amount, element is designated as for continuous variable, represent that grid connected wind power abandons air quantity d days t; be the wind-powered electricity generation of d days t access node i abandon the air quantity upper limit; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0; S is node injecting power and branch road effective power flow incidence matrices, be the active power vector that d days t flow through branch road, element is designated as p _{d, t, ij}, p _{d, t, ij}be the active power that d days t flow through branch road i-j; g _d,tfor normal power supplies to be gained merit force vector in the reality of d days t, element is designated as g _{d, t, i}, g _{d, t, i}for the normal power supplies of access node i is exerted oneself the reality of d days t is meritorious; u _d,tfor grid connected wind power to be gained merit force vector in the reality of d days t, element is designated as u _{d, t, i}, u _{d, t, i}for the wind-powered electricity generation of access node i is exerted oneself the reality of d days t is meritorious; l _d,tfor the load vector d days t, element is designated as l _{d, t, i}, l _{d, t, i}it is the load of d days t node i;

The branch road DC power flow power constraint of system: p _{d, t, ij}-β _ij(θ _{d, t, i}-θ _{d, t, j})=0; p _{d, t, ij}be the active power that d days t flow through branch road i-j; β _ijfor the susceptance of branch road i-j; θ _{d, t, i}and θ _{d, t, j}be respectively node i and the node j voltage phase angle d days t;

The tributary capacity constraint of system: p _{d, t, ij}be the active power that d days t flow through branch road i-j; for the active power upper limit that every bar branch road i-j can bear;

Conventional power generation usage unit units limits: for conventional power generation usage unit maximum output, g _i for conventional power generation usage unit minimum load;

Node i abandons air quantity constraint 0≤w _{d, t, i}≤ u _{d, t, i}; w _{d, t, i}for the wind-powered electricity generation of access node i abandons air quantity d days t; u _{d, t, i}for the wind-powered electricity generation of access node i is exerted oneself the reality of d days t is meritorious;

System retrains the wind total amount of abandoning of d days w _{d, t, i}be the abandon air quantity of wind-powered electricity generation in t of d days access node i, I is the sum of system node, a and b is respectively the upper and lower bound abandoning wind total amount every day;

Cutting load and the probability constraints abandoning wind with be that the binary system of d days t cutting loads operation starts variable, arbitrary node when d days there is cutting load phenomenon in t, value is 1, otherwise, value is 0; be that d days t are abandoned the binary system that character and conduct does and started variable, grid connected wind power d days t exist abandon wind phenomenon time, value is 1, otherwise, value is 0; with the system that is respectively is in timing statistics D inscribe Load Probability and the threshold value of abandoning wind probability;

Cutting load and the desired value constraint of abandoning wind: with r _{d, t, i}for node i is in the cutting load amount of d days t; w _{d, t, i}for the wind-powered electricity generation of access node i abandons air quantity d days t; with the system that is respectively in institute timing statistics D year cutting load expect and abandon in year wind expect threshold values.