CN103956773A - Standby configuration optimization method adopting wind power system unit - Google Patents

Abstract

The invention provides a standby configuration optimization method adopting a wind power system unit. The standby configuration optimization method provides a method for computing two risk costs including outage cost and wind curtailment cost under certain forward-reverse rotation spare capacity by utilizing an integral of a wind power and load joint probability density function, an optimum allocation method for allocating the total system forward-reverse rotation spare capacity to different coal consumption units, and a decision-making method comprehensively considering the risk costs and power generating cost, the total system forward-reverse rotation spare capacity when the power system performs most economical operation and power generating plans and spare allocation capacities of the units. The standby configuration optimization method is suitable for the power system accessed to a wind power plant, can make a decision on the total system forward-reverse rotation spare capacity giving consideration to the economical efficiency and safety and contribution plans and optimum forward-reverse rotation spare capacities of the units, meanwhile unit climbing constraint and network security constraint are considered for spare capacity allocation of the units, and completely available standby application provided by the units is ensured.

Description

Backup configuration optimization method containing wind power system unit

Technical field

The invention belongs to power system operation and dispatching technique field, relate in particular to a kind of backup configuration optimization method containing wind power system.

Background technology

In recent years, along with the continuous increase of wind-electricity integration capacity, wind power generation has become the vital power supply in electric power system.Wind power generation has strong intermittence and stochastic volatility, large-scale wind-electricity integration has brought acid test can to safe operation and the economic dispatch of electrical network, to the enough spinning reserve capacities of electric power system arrangement, is one of the major measure of large-scale wind power and minimizing operation of power networks risk of dissolving.

Enough spinning reserves can greatly reduce the risk of electric power system, but excessive spinning reserve capacity can increase the cost of system, so need reasonably to arrange spinning reserve capacity.The method that spinning reserve is distributed rationally at present mainly contains: Deterministic Methods, probabilistic approach, the method based on cost effectiveness analysis.

First kind Deterministic Methods, the certain percentage that document one < < Optimal wind-thermal generating unit commitment > > (IEEE Transactions on Energy Conversion, 2009 the 29th 25 phases of volume the 13rd page) exerts oneself wind-powered electricity generation is as additional spinning reserve demand.The spinning reserve collocation method of fixed proportion is simple, but easily causes standby deficiency or standby waste, can not make the economy of electric power system reach optimum.

Equations of The Second Kind probabilistic approach, document two < < A new approach to quantify reserve demand in systems with significant installed wind capacity > > (IEEE Transactions on Power Systems, 2005 the 20th 2 phases of volume the 587th page) be take system and are lost Load Probability and determine the stand-by requirement of system as reliability index.The method can guarantee that system is issued to economic optimum in the reliability level of setting, but the setting of reliability level is with very strong subjectivity and cannot answer whether it reasonable.

The method of the 3rd class based on cost effectiveness analysis; the optimum spinning reserve capacity research > > of document three < < large-scale wind powers accesses thermoelectricity systems (protecting electrical power system and control, 2012 the 40th 13 phases of volume the 110th page) is by cost of electricity-generating and lose load cost summation minimum and determine stand-by requirement.The method does not consider that unit ramp loss and Network Security Constraints are on standby impact.

The research of above document has only been considered standbyly in total amount, to meet the total stand-by requirement of system, but fail to propose a kind of cost-effective method, total stand-by requirement is assigned in concrete every unit.

After a large amount of wind-powered electricity generation access electric power systems, the economy of electric power system and fail safe are more difficult to be taken into account.For this reason, need a kind of generation schedule and standby decision-making technique containing wind-powered electricity generation electric power system, generation schedule and the positive and negative spinning reserve partition capacity of the best spinning reserve of the system that can calculate, each unit, to reach the object of taking into account power system economy and fail safe.

Summary of the invention

For traditional standby decision-making and distribution, do not consider fully to meet unit ramp loss and network constraint or other the deficiencies in the prior art, the invention provides a kind of generation schedule and standby decision-making technique containing wind-powered electricity generation electric power system, the method can be determined generation schedule and the standby partition capacity of every unit, unit ramp loss and Network Security Constraints have been considered in standby distribution simultaneously, guarantee that each unit provides standby completely available.

For achieving the above object, the technical solution used in the present invention is a kind of backup configuration optimization method containing wind power system unit, comprises the following steps:

Steps A, utilize the integration type of the joint probability density function of wind-powered electricity generation and two kinds of stochastic variables of load calculate the loss of outage under current positive and negative spinning reserve capacity and abandon windage loss and lose two kinds of risk costs;

Step B, consider standby availability and benefit of saving coal, when the total positive and negative spinning reserve capacity of the system that provides is determined, the standby distribution method at different coal consumption units;

Step C, risk cost is added in cost of electricity-generating target function, by the total positive and negative spinning reserve capacity of system is carried out to optimizing, find the total positive and negative stand-by requirement of system hour of total cost, generation schedule and the positive and negative spinning reserve partition capacity of each unit.

Further, utilize described in steps A the integration type of the joint probability density function of wind-powered electricity generation and two kinds of stochastic variables of load calculate the loss of outage under current positive and negative spinning reserve capacity and abandon windage loss and lose two kinds of risk costs, specifically in accordance with the following methods:

Steps A 1, by normal distribution model, describe the probability distribution of wind-powered electricity generation and load, its expression formula is as follows:

$\left\{\begin{array}{c}{f}_{P_W}\left(P_W\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_W}{e}^{-\frac{{(P_W-{\stackrel{\&OverBar;}{P}}_W)}^2}{{2\mathrm{\&sigma;}}_W^2}}\\ f_{{\mathrm{\&Delta;P}}_L}\left({\mathrm{\&Delta;P}}_L\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_L}{e}^{-\frac{{{\mathrm{\&Delta;p}}_L}^2}{{2\mathrm{\&sigma;}}_L^2}}\end{array}\right.$

In formula: P _wfor actual wind power; for wind power prediction value; Δ P _lfor load prediction error; σ _w, σ _lbe respectively the standard deviation of wind power and load prediction error.

Steps A 2, by Convolution Formula, obtain the two Joint Distribution Z=P _w-Δ P _lprobability density function, its expression formula is as follows:

$f_Z\left(z\right)={\&Integral;}_0^{P_N}{f}_{P_W}\left(P_W\right)f_{{\mathrm{\&Delta;P}}_L}(z-P_W)d{P}_W$

In formula: P _nfor wind-powered electricity generation rated power.

Steps A 3, utilize the joint probability density function of wind power and load prediction error, obtain in the lower power failure expectation of current positive and negative spinning reserve value (being initialized positive and negative spinning reserve value for the first time) and abandon wind and expect;

Electric power system is used whole positive rotation standby, still can not make up the deficiency of system power, will produce power failure.The expression formula of power failure expectation is as follows:

$L_{\mathrm{loss},t}=\underset{R_t^U+Z-{\stackrel{\&OverBar;}{P}}_W<0}{\&Integral;}\mathrm{\&Delta;}{P}_{\mathrm{loss}}{f}_Z\left(z\right)\mathrm{dz}$

In formula: Δ P _lossfor power shortage, have for system standby in the positive rotation of period t, L _{loss, t}for at positive backed-up value being time power failure expectation;

Whole negative spinning reserves is used in electric power system, and surplus that still can not balance sysmte power, can produce and abandon wind.The expression formula of abandoning wind expectation is as follows:

$W_{\mathrm{loss},t}=\underset{R_t^D-(Z-{\stackrel{\&OverBar;}{P}}_W)<0}{\&Integral;}\mathrm{\&Delta;}{W}_{\mathrm{loss}}{f}_Z\left(z\right)\mathrm{dz}$

In formula: Δ W _lossfor abandoning wind power, have for negative standby at period t of system; W _{loss, t}for at negative backed-up value being time abandon wind expectation;

Steps A 4, according to have a power failure expectation and power failure cost, obtain loss of outage, according to abandoning wind expectation and abandoning eolian and obtain and abandon windage loss and lose, expression formula is as follows simultaneously:

$\left\{\begin{array}{c}{F}_{U,t}=c_L{L}_{\mathrm{loss},t}\\ F_{D,t}=c_W{W}_{\mathrm{loss},t}\end{array}\right.$

In formula: c _lfor the unit cost having a power failure; F _u,tfor loss of outage; c _wfor abandoning the unit cost of wind; F _d,tfor abandoning windage loss, lose.

In step B, the total reserve capacity of system, in the distribution of different units, can adopt existing the whole bag of tricks.Invented a kind of standby distribution method of considering standby availability and benefit of saving coal herein, the reserve capacity that every unit of the method consideration is born must be subject to the restriction of unit ramping rate constraints and Network Security Constraints, by most economical method, to standby, be optimized distribution simultaneously, thereby guarantee that unit provides standby availability and economy.Its concrete steps are as follows:

Step B1, according to following formula, obtain three of conventional unit generating curves:

$\left\{\begin{array}{c}{K}_{G0,t}={\stackrel{\&OverBar;}{P}}_{L,t}-{\stackrel{\&OverBar;}{P}}_{W,t}\\ K_{G1,t}={\stackrel{\&OverBar;}{P}}_{L,t}+R_t^U-{\stackrel{\&OverBar;}{P}}_{W,t}\\ K_{G2,t}={\stackrel{\&OverBar;}{P}}_{L,t}-R_t^D-{\stackrel{\&OverBar;}{P}}_{W,t}\end{array}\right.$

In formula: K _{g0, t}for the prediction of the conventional unit of period t is exerted oneself; K _{g1, t}for the maximum output of the conventional unit of period t when the standby whole use of positive rotation; K _{g2, t}for the minimum load of the conventional unit of period t when negative spinning reserve is all used; for the load prediction value of system at period t; for the wind power predicted value at period t;

Step B2, solve the conventional Unit Commitment state under 24 period maximum generation curves;

Step B3, with 3 target functions of corresponding 3 generating curves, the prediction of obtaining each conventional unit under the prerequisite that meets system related constraint is exerted oneself, maximum output and minimum load, three target functions are as follows:

$\left\{\begin{array}{c}{F}_0=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i0,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\\ F_1=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i1,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\\ F_2=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i2,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\end{array}\right.$

In formula: hop count when T is total; N is fired power generating unit sum; P _{i0, t}, P _{i1, t}, P _{i2, t}being respectively fired power generating unit i exerts oneself at corresponding three generating the meritorious of curve of period t; S _istart-up cost for unit i; I _i,tfor the start and stop state of unit i at period t; C _ifor the cost of electricity-generating of fired power generating unit i, as a wherein _i, b _i, c _ifor cost of electricity-generating coefficient;

Step B4,3 target functions distribute according to 3 generating curves respectively that units are meritorious exerts oneself, the prediction of the obtaining unit P that exerts oneself _{i0, t}, maximum output P _{i1, t}, minimum load P _{i2, t}, then by following formula, obtain the standby and negative spinning reserve of positive rotation that unit is born:

$\left\{\begin{array}{c}{R}_t^{\mathrm{Ui}}=P_{i1,t}-P_{i0,t}\\ R_t^{\mathrm{Di}}=P_{i0,t}-P_{i2,t}\end{array}\right.$

In formula: the positive rotation reserve capacity of bearing at period t for going out unit i; the negative spinning reserve capacity of bearing at period t for unit i.

Preferably, described calculating and distribution method containing the positive and negative spinning reserve of wind-powered electricity generation electric power system, is characterized in that, described in solve the conventional Unit Commitment state under 24 period maximum generation curves, specifically in accordance with the following methods:

Step B201, according to following formula, calculate the coa consumption rate at full capacity of each unit, and sort from small to large;

F(P _i,max)＝f _i(P _i,max)/P _i,max

In formula: P _{i, max}the upper limit of exerting oneself for unit i; f _i(P _{i, max}) cost of electricity-generating of prescribing a time limit on exerting oneself for unit i;

Step B202, take and meet maximum generation curve as target, by coa consumption rate sequence at full capacity, determine from small to large the start unit of per period.

Step B203,24 period unit open states are adjusted and made it meet start/stop machine constraint;

The minimum operation of unit and minimum idle time constraint are as follows:

$I_{i,t}=\left\{\begin{array}{c}\mathrm{1,1}\&le;x_{i,t-1}\&le;{T_i}^{\mathrm{on}}\\ 0,-1\&GreaterEqual;x_{i,t-1}>-{T_i}^{\mathrm{off}}\\ 0\mathrm{or}1,\mathrm{otherwise}\end{array}\right.$

In formula: T _i ^on, T _i ^offthe minimum that is respectively unit i allows available machine time and minimum to allow downtime; x _{i, t-1}for the startup-shutdown time of unit i in the t-1 period, work as x _{i, t-1}within>=1 o'clock, represent the available machine time, work as x _{i, t-1}represent the unused time at≤-1 o'clock.

If unit was started shooting in the t-1 period, it does not meet minimum available machine time constraint in the t period and can not shut down; If unit was shut down in the t-1 period, it does not meet minimum constraint downtime in the t period and can not start shooting.

Step B204, heuristic contrary ranking method are revised unit start-stop state;

The unit having started is sorted from big to small by coa consumption rate at full capacity, if unit i residue unit output after the t period meets minimum start constraint and shuts down still meets maximum generation curve power, cut off it.

Preferably, described calculating and distribution method containing the positive and negative spinning reserve of wind-powered electricity generation electric power system, it is characterized in that, described 3 target functions with corresponding 3 generating curves, the prediction of obtaining each conventional unit under the prerequisite that meets system related constraint is exerted oneself, maximum output and minimum load, specifically in accordance with the following methods:

Step B301, meeting under the prerequisite of unit output restriction and system power Constraints of Equilibrium, obtain the unit output under maximum generation curve;

Unit output restriction under maximum generation curve is as follows:

$I_{i,t}{P}_i^{\min }\&le;I_{i,t}{P}_{i1,t}\&le;I_{i,t}{P}_i^{\max }$

In formula: be respectively the bound of exerting oneself of unit i.

System power Constraints of Equilibrium under maximum generation curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i1,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}-R_t^U$

Step B302, meeting under the prerequisite of the constraint such as unit creep speed, unit output restriction, power-balance, obtaining the unit output under minimum generating curve;

The relative unit maximum output of unit minimum load must meet the restriction of unit ramping rate constraints, and formula is as follows:

$\left\{\begin{array}{c}-{\mathrm{\&Delta;P}}_{\mathrm{down},i}\&le;P_{i1,t}-P_{i2,t-1}\&le;{\mathrm{\&Delta;P}}_{\mathrm{up},i}\\ -{\mathrm{\&Delta;P}}_{\mathrm{down},i}\&le;P_{i2,t}-P_{i1,t-1}\&le;{\mathrm{\&Delta;P}}_{\mathrm{up},i}\\ P_{i1,t}-P_{i2,t}\&le;\min ({\mathrm{\&Delta;P}}_{\mathrm{up},i},{\mathrm{\&Delta;P}}_{\mathrm{down},i})\end{array}\right.$

In formula: Δ P _{up, i}, Δ P _{down, i}be respectively the positive and negative creep speed of unit i.

The unit output restriction of unit minimum load is as follows simultaneously:

$I_{i,t}{P}_i^{\min }\&le;I_{i,t}{P}_{i2,t}\&le;I_{i,t}{P}_i^{\max }$

System power Constraints of Equilibrium under minimum generating curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i2,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}-R_t^D$

Step B303, the plan of restriction unit are exerted oneself between unit maximum output and unit minimum load, meeting under the prerequisite of power-balance constraint under plan generating curve, obtain unit plan and exert oneself simultaneously;

The unit plan restriction of exerting oneself is as follows:

I _i,tP _i2,t≤I _i,tP _i0,t≤I _i,tP _i1,t

Power-balance constraint under plan generating curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i0,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}$

Step B304, judge whether to meet branch road transmission capacity constraint, if met, finish; If do not met the relevant unit output of restriction, go to step B301 and redistribute unit output;

The constraint of branch road transmission capacity is as follows:

$\left\{\begin{array}{c}-f_j^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m0,t}\&le;f_j^{\max }\\ -f_J^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m1,t}\&le;f_j^{\max }\\ -f_j^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m2,t}\&le;f_j^{\max }\end{array}\right.$

In formula: G _j,mfor node m shifts distribution factor to the generating of circuit j; P _{m0, t}, P _{m1, t}, P _{m2, t}when respectively corresponding generating curve is for prediction generating curve, maximum generation curve, minimum generating curve, node m is at the injecting power of period t, maximum delivery power for circuit j; M is node sum.

Further, in described step C, risk cost is added in cost of electricity-generating target function, by the total positive and negative spinning reserve capacity of system is carried out to optimizing, find the total positive and negative stand-by requirement of system hour of total cost, generation schedule and the positive and negative spinning reserve partition capacity of each unit, its concrete steps are as follows:

Step C1, obtain consider risk cost and cost of electricity-generating system synthesis this, as shown in the formula:

$F=F_0+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}(F_{U,t}+F_{D,t}+C_t^U+C_t^D)$

In formula: the total cost that F is system; F ₀for the cost of electricity-generating of system, F _u,tfor the loss of outage of system at period t; F _d,tfor the abandon windage loss of system at period t loses; for the positive stand-by cost of system at period t; for the negative stand-by cost of system at period t;

Positive and negative stand-by cost is according to the following formula:

$\left\{\begin{array}{c}{C}_t^U=c_R^U{R}_t^U\\ C_t^D=c_R^D{R}_t^D\end{array}\right.$

In formula: cost coefficient for the positive reserve capacity of system; cost coefficient for the positive reserve capacity of system;

Step C2, the direction that reduces by F be to the positive and negative standby optimizing of carrying out, and finds and make hour positive and negative standby of F, is best positive and negative backed-up value.

Preferably, described calculating and distribution method containing the positive and negative spinning reserve of wind-powered electricity generation electric power system, is characterized in that, the described direction reducing by F is to the positive and negative standby optimizing of carrying out, according to following steps:

Step C201, obtain current positive and negative system synthesis under standby and be originally designated as F ^(k);

Step C202, when last group of positive rotation reserve capacity as follows:

$R^{U,k}=[R_1^{U,k},R_2^{U,k},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{U,k},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{U,k}]$

From the t=1 period, start to align spinning reserve and carry out optimizing, until 24 periods completed.While carrying out optimizing to the positive rotation of period t is standby, keep the positive rotation backed-up value of other periods constant, order replace R ^u,kin obtain new R ^u,kthe total cost F of lower system ^(k1); Order replace R ^u,kin obtain new R ^u,kthe total cost F of lower system ^(k2).

If F ^(k1)< F ^(k)and F ^(k2)> F ^(k), the positive rotation of period t is standby should change by the direction increasing, order $R_t^{U,k+1}=R_t^{U,k}+\mathrm{\&Delta;}{R}_t^{U,k};$

If F ^(k1)> F ^(k)and F ^(k2)< F ^(k), the positive rotation of period t is standby should change by the direction reducing, order $R_t^{U,k+1}=R_t^{U,k}-\mathrm{\&Delta;}{R}_t^{U,k};$

If F ^(k1)> F ^(k)and F ^(k2)> F ^(k), order remodify calculate new this F of system synthesis ^(k1)and F ^(k2), re-start judgement; If when (δ is a less positive number), still there is F ^(k1)> F ^(k)and F ^(k2)> F ^(k), the current positive rotation backed-up value of period t is exactly optimum value,

By the optimizing to 24 period positive rotation backed-up values, can obtain one group of new positive rotation standby:

$R^{U,k+1}=[R_1^{U,k+1},R_2^{U,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{U,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{U,k+1}]$

In like manner, by 24 period negative rotations being turned to the optimizing of backed-up value, can obtain one group of new negative rotation and turn backed-up value:

$R^{D,k+1}=[R_1^{D,k+1},R_2^{D,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{D,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{D,k+1}]$

Step C203, obtain this F of system synthesis under one group of new positive and negative spinning reserve value ^(k+1);

Step C204, judged whether | F ^(k)-F ^(k+1)| < ε, if it is export required standby be best positive and negative standby; Otherwise go to step C202 and continue iteration.

The present invention utilizes the loss of outage of wind-powered electricity generation and the probability density function estimating system of the two kinds of stochastic variables of loading and abandons windage loss and lose, then take the minimization of total system cost as target, calculate the positive and negative spinning reserve of the best of system, the best that in definite system, each unit output plan and per period are born is simultaneously positive and negative standby.

Major advantage of the present invention has: can be by mutually the pining down of stand-by cost and risk cost, and automatic decision goes out the positive and negative spinning reserve of the best of system, has realized the balance of power system economy and fail safe; Determine the positive and negative spinning reserve of total the best of system, the generation schedule of each unit and positive and negative spinning reserve partition capacity simultaneously; Considered unit ramping rate constraints and Network Security Constraints, it is positive and negative standby effectively available that assurance arranges.

Accompanying drawing explanation

Fig. 1 is the flow chart of the backup configuration optimization method containing wind power system unit;

Fig. 2 is analogue system schematic diagram;

Fig. 3 is that the power failure expectation of system shown in Figure 2 and expected gross cost are with just standby variation diagram;

Fig. 4 is that the system of system shown in Figure 2 is abandoned wind expectation and expected gross cost with negative standby variation diagram;

Fig. 5 is the meritorious tidal current chart of maximum of the circuit 7-8 of system shown in Figure 2;

Fig. 6 is the positive reserve capacity figure that 3 fired power generating unit of No. 7 nodes of system shown in Figure 2 are born.

Embodiment

Below in conjunction with the drawings and specific embodiments, further illustrate the present invention, should understand these examples is only not used in and limits the scope of the invention for the present invention is described, after having read the present invention, those skilled in the art all fall within the application's claims limited range to the modification of the various equivalent form of values of the present invention.

Fig. 1 is the flow chart containing the backup configuration optimization method of wind power system unit.As shown in Figure 1, the invention provides a kind of backup configuration optimization method containing wind power system unit, first it utilize the probability density function estimation of wind-powered electricity generation and load envision the loss of outage under positive and negative spinning reserve capacity and abandon windage loss and lose two kinds of risk costs; Then determine standby distribution and cost of electricity-generating under anticipation reserve capacity; Finally by aligning the optimizing of negative spinning reserve, take risk cost, stand-by cost and cost of electricity-generating minimum is target, determines the positive and negative spinning reserve of the best of system, determines the positive and negative reserve capacity that the plan of unit is exerted oneself and born simultaneously.

With reference to figure 1, should specifically there be following steps containing the backup configuration optimization method of wind power system unit:

Step 1, utilize the integration type of the joint probability density function of wind-powered electricity generation and two kinds of stochastic variables of load calculate the loss of outage under current positive and negative spinning reserve capacity and abandon windage loss and lose two kinds of risk costs:

Step 101, by normal distribution model, describe the probability distribution of wind-powered electricity generation and load, its expression is as follows:

$\left\{\begin{array}{c}{f}_{P_W}\left(P_W\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_W}{e}^{-\frac{{(P_W-{\stackrel{\&OverBar;}{P}}_W)}^2}{{2\mathrm{\&sigma;}}_W^2}}\\ f_{{\mathrm{\&Delta;P}}_L}\left({\mathrm{\&Delta;P}}_L\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_L}{e}^{-\frac{{{\mathrm{\&Delta;p}}_L}^2}{{2\mathrm{\&sigma;}}_L^2}}\end{array}\right.$

In formula: P _wfor actual wind power; for wind power prediction value; Δ P _lfor load prediction error; σ _w, σ _lbe respectively the standard deviation of wind power and load prediction error.

Step 102, by Convolution Formula, obtain the two Joint Distribution Z=P _w-Δ P _lprobability density function, its expression formula is as follows:

$f_Z\left(z\right)={\&Integral;}_0^{P_N}{f}_{P_W}\left(P_W\right)f_{{\mathrm{\&Delta;P}}_L}(z-P_W)d{P}_W$

In formula: P _nfor wind-powered electricity generation rated power.

Step 103, utilize the joint probability density function of wind power and load, obtain the power failure expectation under given positive and negative spinning reserve value and abandon wind expectation;

Have a power failure and expect and abandon wind expectation to obtain by following formula:

$\left\{\begin{array}{c}{L}_{\mathrm{loss},t}=\underset{R_t^U+Z-{\stackrel{\&OverBar;}{P}}_W<0}{\&Integral;}\mathrm{\&Delta;}{P}_{\mathrm{loss}}{f}_Z\left(z\right)\mathrm{dz}\\ W_{\mathrm{loss},t}=\underset{R_t^D-(Z-{\stackrel{\&OverBar;}{P}}_W)<0}{\&Integral;}\mathrm{\&Delta;}{W}_{\mathrm{loss}}{f}_Z\left(z\right)\mathrm{dz}\end{array}\right.$

In formula: Δ P _lossfor power shortage, have for system standby in the positive rotation of period t, L _{loss, t}for at positive backed-up value being time power failure expectation; Δ W _lossfor abandoning wind power, have for negative standby at period t of system; W _{loss, t}for at negative backed-up value being time abandon wind expectation;

Step 104, according to having a power failure, expectation and power failure cost are obtained loss of outage; According to abandoning wind expectation and abandoning eolian and obtain and abandon windage loss and lose; Expression formula is as follows:

$\left\{\begin{array}{c}{F}_{U,t}=c_L{L}_{\mathrm{loss},t}\\ F_{D,t}=c_W{W}_{\mathrm{loss},t}\end{array}\right.$

In formula: c _lfor the unit cost having a power failure; F _u,tfor loss of outage; c _wfor abandoning the unit cost of wind; F _d,tfor abandoning windage loss, lose.Step 2, consider standby availability and benefit of saving coal, when the total positive and negative spinning reserve capacity of the system that provides is determined, the standby distribution method at different coal consumption units:

Step 201, according to following formula, obtain three of conventional unit generating curves:

$\left\{\begin{array}{c}{K}_{G0,t}={\stackrel{\&OverBar;}{P}}_{L,t}-{\stackrel{\&OverBar;}{P}}_{W,t}\\ K_{G1,t}={\stackrel{\&OverBar;}{P}}_{L,t}+R_t^U-{\stackrel{\&OverBar;}{P}}_{W,t}\\ K_{G2,t}={\stackrel{\&OverBar;}{P}}_{L,t}-R_t^D-{\stackrel{\&OverBar;}{P}}_{W,t}\end{array}\right.$

In formula: K _{g0, t}for the prediction of the conventional unit of period t is exerted oneself; K _{g1, t}for the maximum output of the conventional unit of period t when the standby whole use of positive rotation; K _{g2, t}for the minimum load of the conventional unit of period t when negative spinning reserve is all used; for the load prediction value of system at period t; for the wind power predicted value at period t;

Step 202, solve the conventional Unit Commitment state under 24 period maximum generation curves;

First calculate the coa consumption rate at full capacity of each unit, press coa consumption rate at full capacity determine from small to large start unit in per period, until the start power of per period is more than or equal to the start power that maximum generation curve needs, coa consumption rate calculating formula is as follows at full capacity:

F(P _i,max)＝f _i(P _i,max)/P _i,max

In formula: P _{i, max}the upper limit of exerting oneself for unit i; f _i(P _{i, max}) cost of electricity-generating of prescribing a time limit on exerting oneself for unit i;

24 period unit open states are adjusted and made it meet start/stop machine constraint, if unit was started shooting in the t-1 period, it does not meet minimum available machine time constraint in the t period and can not shut down; If unit was shut down in the t-1 period, it does not meet minimum constraint downtime in the t period and can not start shooting.The minimum operation of unit and minimum idle time are constrained to:

$I_{i,t}=\left\{\begin{array}{c}\mathrm{1,1}\&le;x_{i,t-1}\&le;{T_i}^{\mathrm{on}}\\ 0,-1\&GreaterEqual;x_{i,t-1}>-{T_i}^{\mathrm{off}}\\ 0\mathrm{or}1,\mathrm{otherwise}\end{array}\right.$

In formula: T _i ^on, T _i ^offthe minimum that is respectively unit i allows available machine time and minimum to allow downtime; x _{i, t-1}for the startup-shutdown time of unit i in the t-1 period, work as x _{i, t-1}within>=1 o'clock, represent the available machine time, work as x _{i, t-1}represent the unused time at≤-1 o'clock.

Finally by heuristic contrary ranking method, unit start-stop state is revised: the unit having started is sorted from big to small by coa consumption rate at full capacity; if unit i residue unit output after the t period meets minimum start constraint and shuts down still meets maximum generation curve power, cut off it.

Step 203, with 3 target functions of corresponding 3 generating curves, the prediction of obtaining each conventional unit under the prerequisite that the meets system related constraint P that exerts oneself _{i0, t}, maximum output P _{i1, t}with minimum load P _{i2, t}, three target functions are as follows:

$\left\{\begin{array}{c}{F}_0=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i0,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\\ F_1=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i1,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\\ F_2=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i2,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\end{array}\right.$

In formula: hop count when T is total; N is fired power generating unit sum; P _{i0, t}, P _{i1, t}, P _{i2, t}being respectively fired power generating unit i exerts oneself at corresponding three generating the meritorious of curve of period t; S _istart-up cost for unit i; I _i,tfor the start and stop state of unit i at period t; C _ifor the cost of electricity-generating of fired power generating unit i, as a wherein _i, b _i, c _ifor cost of electricity-generating coefficient;

The method for solving that unit maximum output, minimum load, prediction are exerted oneself is as follows:

1), under the prerequisite of sufficient unit output restriction and system power Constraints of Equilibrium, obtain the unit output under maximum generation curve;

Unit output restriction under maximum generation curve is as follows:

$I_{i,t}{P}_i^{\min }\&le;I_{i,t}{P}_{i1,t}\&le;I_{i,t}{P}_i^{\max }$

In formula: be respectively the bound of exerting oneself of unit i.

System power Constraints of Equilibrium under maximum generation curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i1,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}-R_t^U$

2), meeting under the prerequisite of the constraint such as unit creep speed, unit output restriction, power-balance, obtain the unit output under minimum generating curve;

The relative unit maximum output of unit minimum load must meet the restriction of unit ramping rate constraints, and formula is as follows:

$\left\{\begin{array}{c}-{\mathrm{\&Delta;P}}_{\mathrm{down},i}\&le;P_{i1,t}-P_{i2,t-1}\&le;{\mathrm{\&Delta;P}}_{\mathrm{up},i}\\ -{\mathrm{\&Delta;P}}_{\mathrm{down},i}\&le;P_{i2,t}-P_{i1,t-1}\&le;{\mathrm{\&Delta;P}}_{\mathrm{up},i}\\ P_{i1,t}-P_{i2,t}\&le;\min ({\mathrm{\&Delta;P}}_{\mathrm{up},i},{\mathrm{\&Delta;P}}_{\mathrm{down},i})\end{array}\right.$

In formula: Δ P _{up, i}, Δ P _{down, i}be respectively the positive and negative creep speed of unit i.

The unit output restriction of unit minimum load is as follows simultaneously:

$I_{i,t}{P}_i^{\min }\&le;I_{i,t}{P}_{i2,t}\&le;I_{i,t}{P}_i^{\max }$

System power Constraints of Equilibrium under minimum generating curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i2,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}-R_t^D$

3), restriction unit plan exerts oneself between unit maximum output and unit minimum load, meeting under the prerequisite of power-balance constraint under plan generating curve, obtain unit plan and exert oneself simultaneously;

The unit plan restriction of exerting oneself is as follows:

I _i,tP _i2,t≤I _i,tP _i0,t≤I _i,tP _i1,t

Power-balance constraint under plan generating curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i0,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}$

4), judge whether to meet branch road transmission capacity constraint, if met, finish; If do not met the relevant unit output of restriction, go to step 1 and redistribute unit output;

The constraint of branch road transmission capacity is as follows:

$\left\{\begin{array}{c}-f_j^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m0,t}\&le;f_j^{\max }\\ -f_J^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m1,t}\&le;f_j^{\max }\\ -f_j^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m2,t}\&le;f_j^{\max }\end{array}\right.$

In formula: G _j,mfor node m shifts distribution factor to the generating of circuit j; P _{m0, t}, P _{m1, t}, P _{m2, t}when respectively corresponding generating curve is for prediction generating curve, maximum generation curve, minimum generating curve, node m is at the injecting power of period t, maximum delivery power for circuit j; M is node sum.

Step 204, by following formula, obtain the standby and negative spinning reserve of positive rotation that unit is born:

$\left\{\begin{array}{c}{R}_t^{\mathrm{Ui}}=P_{i1,t}-P_{i0,t}\\ R_t^{\mathrm{Di}}=P_{i0,t}-P_{i2,t}\end{array}\right.$

In formula: the positive rotation reserve capacity of bearing at period t for going out unit i; the negative spinning reserve capacity of bearing at period t for unit i.

Step 3, risk cost is added in cost of electricity-generating target function, by the total positive and negative spinning reserve capacity of system is carried out to optimizing, finds the total positive and negative stand-by requirement of system hour of total cost, generation schedule and the positive and negative spinning reserve partition capacity of each unit:

Step 301, obtain consider risk cost and cost of electricity-generating system synthesis this, as shown in the formula:

$F=F_0+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}(F_{U,t}+F_{D,t}+C_t^U+C_t^D)$

In formula: the total cost that F is system; F ₀for the cost of electricity-generating of system, F _u,tfor the loss of outage of system at period t; F _d,tfor the abandon windage loss of system at period t loses; for the positive stand-by cost of system at period t; for the negative stand-by cost of system at period t;

Step 302, the direction that reduces by F be to the positive and negative standby optimizing of carrying out, and finds and make hour positive and negative standby of F, is best positive and negative backed-up value;

To positive and negative standby optimizing in accordance with the following methods:

(1), current positive and negative system synthesis under standby is originally designated as F ^(k);

(2), one group of positive rotation reserve capacity is as follows:

$R^{U,k}=[R_1^{U,k},R_2^{U,k},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{U,k},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{U,k}]$

From the t=1 period, start to align spinning reserve and carry out optimizing, until 24 periods completed.While carrying out optimizing to the positive rotation of period t is standby, keep the positive rotation backed-up value of other periods constant, order replace R ^u,kin obtain new R ^u,kthe total cost F of lower system ^(k1); Order replace R ^u,kin obtain new R ^u,kthe total cost F of lower system ^(k2).

If F ^(k1)< F ^(k)and F ^(k2)> F ^(k), the positive rotation of period t is standby should change by the direction increasing, order $R_t^{U,k+1}=R_t^{U,k}+\mathrm{\&Delta;}{R}_t^{U,k};$

If F ^(k1)> F ^(k)and F ^(k2)< F ^(k), the positive rotation of period t is standby should change by the direction reducing, order $R_t^{U,k+1}=R_t^{U,k}-\mathrm{\&Delta;}{R}_t^{U,k};$

If F ^(k1)> F ^(k)and F ^(k2)> F ^(k), order remodify calculate new this F of system synthesis ^(k1)and F ^(k2), re-start judgement; If when (δ is a less positive number), still there is F ^(k1)> F ^(k)and F ^(k2)> F ^(k), the current positive rotation backed-up value of period t is exactly optimum value,

By the optimizing to 24 period positive rotation backed-up values, can obtain one group of new positive rotation standby:

$R^{U,k+1}={[R}_1^{U,k+1},R_2^{U,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{U,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{U,k+1}]$

In like manner, by 24 period negative rotations being turned to the optimizing of backed-up value, can obtain one group of new negative rotation and turn backed-up value:

$R^{D,k+1}=[R_1^{D,k+1},R_2^{D,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{D,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{D,k+1}]$

(3), this F of system synthesis under one group of new positive and negative spinning reserve value ^(k+1);

(4), whether have | F ^(k)-F ^(k+1)| < ε, if it is export required standby be best positive and negative standby; Otherwise go to step (2) and continue iteration.

Fig. 2 is analogue system schematic diagram.As shown in Figure 2, as example of the present invention, analogue system can be IEEE24.In order to test the validity of institute of the present invention extracting method, by 1 wind farm group of No. 9 node accesses in IEEE24 node system, this wind energy turbine set has the wind-driven generator that 200 rated power are 2MW.Utilize the present invention to carry out simulation calculation, analysis of simulation result is as follows:

Shown in Fig. 3, along with the increase of positive backed-up value, the power failure of system expectation constantly reduces, and the total cost of system first reduces rear increase.This is because when positive backed-up value is less than best positive backed-up value, and along with just standby increase, the minimizing of losing load rejection penalty is greater than the increase of just standby expense, and system total cost reduces; Otherwise the total cost of system increases.

Shown in Fig. 4, along with the increase of negative backed-up value, the wind expectation of abandoning of system constantly reduces, and the total cost of system also presents and first reduces the trend increasing afterwards.With positive and negative standby increase, system synthesis originally presents the trend of first falling rear increasing, and corresponding positive and negative spinning reserve is the desired optimal solution of this model during the minimization of total system cost.This paper method can be broken away from the impact that people is decision-making by pining down between cost and risk, the best positive and negative reserve capacity (in Table 1) of automatic acquisition.

Table 2 has provided the result of standby distribution of periods 15, in standby assigning process, considered standby benefit of saving coal herein, meeting under the prerequisite of unit creep speed, the unit that coal consumption cost is higher is preferentially born just standby, the unit that coal consumption unit is slightly low is born negative standby, and the unit that coal consumption cost is minimum keeps completely sending out state and do not bear standby.The meaning of this kind of standby distribution is: just standby according to unit, coal consumption arranges from low to high, when needs increase conventional unit output, preferentially increases the unit output that coal consumption cost is low; Negative standby according to unit, coal consumption arranges from high to low, when needs reduce conventional unit output, preferentially reduces the unit output that coal consumption cost is high, so arranges positive and negative standbyly, can reduce to greatest extent the total consumption of coal cost that uses standby rear system.

As shown in Figure 5, do not consider Network Security Constraints, when system is generated electricity by maximum generation curve, will there is the meritorious trend of branch road and cross the border in the part period of circuit 7-8.Consider Network Security Constraints, the reserve capacity that each unit is born will be reallocated, to meet the security constraint of system.

As shown in Figure 6, in order to meet the Network Security Constraints under maximum generation curve, the just standby minimizing to some extent that 3 units on No. 7 nodes are born after 11 periods.Consider that Network Security Constraints system synthesis originally can increase, but guaranteed standby availability, its essence is and increase the fail safe that coal consumption amount exchanges system for.

The present invention has not only considered the total Optimal Reserve Capacity of system but also considered standby optimum allocation between different coal consumption units, consider unit ramping rate constraints and power system security constraints simultaneously, can provide foundation for dispatcher arranges effectively available positive and negative spinning reserve and unit output plan.

The positive and negative spinning reserve demand of the best of table 1 system

Table 2 periods 15 unit output and standby distribution

Claims (7)

1. containing a backup configuration optimization method for wind power system unit, it comprises the following steps:

Steps A, utilize the integration type of the joint probability density function of wind-powered electricity generation and two kinds of stochastic variables of load calculate the loss of outage under current positive and negative spinning reserve capacity and abandon windage loss and lose two kinds of risk costs;

Step B, consider standby availability and benefit of saving coal, when the total positive and negative spinning reserve capacity of the system that provides is determined, the standby distribution method at different coal consumption units;

Step C, risk cost is added in cost of electricity-generating target function, by the total positive and negative spinning reserve capacity of system is carried out to optimizing, find the total positive and negative stand-by requirement of system hour of total cost, generation schedule and the positive and negative spinning reserve partition capacity of each unit.

2. the backup configuration optimization method containing wind power system unit as claimed in claim 1, it is characterized in that: described in steps A, utilize the integration type of the joint probability density function of wind-powered electricity generation and two kinds of stochastic variables of load calculate the loss of outage under current positive and negative spinning reserve capacity and abandon windage loss and lose two kinds of risk costs, specifically in accordance with the following methods:

Steps A 1, by normal distribution model, describe the probability distribution of wind-powered electricity generation and load, its expression formula is as follows:

$\left\{\begin{array}{c}{f}_{P_W}\left(P_W\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_W}{e}^{-\frac{{(P_W-{\stackrel{\&OverBar;}{P}}_W)}^2}{{2\mathrm{\&sigma;}}_W^2}}\\ f_{{\mathrm{\&Delta;P}}_L}\left({\mathrm{\&Delta;P}}_L\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_L}{e}^{-\frac{{{\mathrm{\&Delta;p}}_L}^2}{{2\mathrm{\&sigma;}}_L^2}}\end{array}\right.$

In formula: P _wfor actual wind power; for wind power prediction value; Δ P _lfor load prediction error; σ _w, σ _lbe respectively the standard deviation of wind power and load prediction error;

Steps A 2, by Convolution Formula, obtain the two Joint Distribution Z=P _w-Δ P _lprobability density function, its expression formula is as follows:

$f_Z\left(z\right)={\&Integral;}_0^{P_N}{f}_{P_W}\left(P_W\right){f_{\mathrm{\&Delta;P}}}_L(z-P_W)d{P}_W$

In formula: P _nfor wind-powered electricity generation rated power;

Steps A 3, utilize the joint probability density function of wind power and load prediction error, obtain the power failure expectation under current positive and negative spinning reserve value and abandon wind expectation;

Electric power system is used whole positive rotation standby, still can not make up the deficiency of system power, will produce power failure, and wherein, the expression formula of the expectation that has a power failure is as follows:

$L_{\mathrm{loss},t}=\underset{R_t^U+Z-{\stackrel{\&OverBar;}{P}}_W<0}{\&Integral;}\mathrm{\&Delta;}{P}_{\mathrm{loss}}{f}_Z\left(z\right)\mathrm{dz}$

In formula: Δ P _lossfor power shortage, have for system standby in the positive rotation of period t, L _{loss, t}for at positive backed-up value being time power failure expectation;

Whole negative spinning reserves is used in electric power system, and surplus that still can not balance sysmte power, can produce and abandon wind, and wherein, the expression formula of abandoning wind expectation is as follows:

$W_{\mathrm{loss},t}=\underset{R_t^D-(Z-{\stackrel{\&OverBar;}{P}}_W)<0}{\&Integral;}\mathrm{\&Delta;}{W}_{\mathrm{loss}}{f}_Z\left(z\right)\mathrm{dz}$

In formula: Δ W _lossfor abandoning wind power, have for negative standby at period t of system; W _{loss, t}for at negative backed-up value being time abandon wind expectation;

Steps A 4, according to have a power failure expectation and power failure cost, obtain loss of outage, according to abandoning wind expectation and abandoning eolian and obtain and abandon windage loss and lose, expression formula is as follows simultaneously:

$\left\{\begin{array}{c}{F}_{U,t}=c_L{L}_{\mathrm{loss},t}\\ F_{D,t}=c_W{W}_{\mathrm{loss},t}\end{array}\right.$

In formula: c _lfor the unit cost having a power failure; F _u,tfor loss of outage; c _wfor abandoning the unit cost of wind; F _d,tfor abandoning windage loss, lose.

3. the backup configuration optimization method containing wind power system unit as claimed in claim 1, it is characterized in that, consider standby availability and benefit of saving coal described in step B, when the total positive and negative spinning reserve capacity of the system that provides is determined, the standby distribution method at different coal consumption units, specifically in accordance with the following methods:

Step B1, according to following formula, obtain three generation load curves under current positive and negative spinning reserve value:

$\left\{\begin{array}{c}{K}_{G0,t}={\stackrel{\&OverBar;}{P}}_{L,t}-{\stackrel{\&OverBar;}{P}}_{W,t}\\ K_{G1,t}={\stackrel{\&OverBar;}{P}}_{L,t}+R_t^U-{\stackrel{\&OverBar;}{P}}_{W,t}\\ K_{G2,t}={\stackrel{\&OverBar;}{P}}_{L,t}-R_t^D-{\stackrel{\&OverBar;}{P}}_{W,t}\end{array}\right.$

In formula: K _{g0, t}for the prediction of the conventional unit of period t is exerted oneself; K _{g1, t}for the maximum output of the conventional unit of period t when the standby whole use of positive rotation; K _{g2, t}for the minimum load of the conventional unit of period t when negative spinning reserve is all used; for the load prediction value of system at period t; for the wind power predicted value at period t;

Step B2, solve the conventional Unit Commitment state under 24 period maximum generation curves;

Step B3, with 3 target functions of corresponding 3 generating curves, the prediction of obtaining each conventional unit under the prerequisite that meets system related constraint is exerted oneself, maximum output and minimum load, three target functions are as follows:

$\left\{\begin{array}{c}{F}_0=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i0,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\\ F_1=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i1,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\\ F_2=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}[C_i\left(P_{i2,t}\right)I_{i,t}+S_i{I}_{i,t}(1-I_{i,t-1})]\end{array}\right.$

In formula: hop count when T is total; N is fired power generating unit sum; P _{i0, t}, P _{i1, t}, P _{i2, t}being respectively fired power generating unit i exerts oneself at corresponding three generating the meritorious of curve of period t; S _istart-up cost for unit i; I _i,tfor the start and stop state of unit i at period t; C _ifor the cost of electricity-generating of fired power generating unit i, as a wherein _i, b _i, c _ifor cost of electricity-generating coefficient;

Step B4,3 target functions distribute according to 3 generating curves respectively that units are meritorious exerts oneself, the prediction of the obtaining unit P that exerts oneself _{i0, t}, maximum output P _{i1, t}, minimum load P _{i2, t}, then by following formula, obtain the standby and negative spinning reserve of positive rotation that each unit is born:

$\left\{\begin{array}{c}{R}_t^{\mathrm{Ui}}=P_{i1,t}-P_{i0,t}\\ R_t^{\mathrm{Di}}=P_{i0,t}-P_{i2,t}\end{array}\right.$

In formula: the positive rotation reserve capacity of bearing at period t for going out unit i; the negative spinning reserve capacity of bearing at period t for unit i.

4. the backup configuration optimization method containing wind power system unit as claimed in claim 3, is characterized in that, described in solve the conventional Unit Commitment state under 24 period maximum generation curves, specifically in accordance with the following methods:

Step B201, according to following formula, calculate the coa consumption rate at full capacity of each unit, and sort from small to large;

F(P _i,max)＝f _i(P _i,max)/P _i,max

In formula: P _{i, max}the upper limit of exerting oneself for unit i; f _i(P _{i, max}) cost of electricity-generating of prescribing a time limit on exerting oneself for unit i;

Step B202, take and meet maximum generation curve as target, by coa consumption rate sequence at full capacity, determine from small to large the start unit of per period;

Step B203,24 period unit open states are adjusted and made it meet start/stop machine constraint;

The minimum operation of unit and minimum idle time constraint are as follows:

$I_{i,t}=\left\{\begin{array}{c}\mathrm{1,1}\&le;x_{i,t-1}\&le;{T_i}^{\mathrm{on}}\\ 0,-1\&GreaterEqual;x_{i,t-1}>-{T_i}^{\mathrm{off}}\\ 0\mathrm{or}1,\mathrm{otherwise}\end{array}\right.$

In formula: T _i ^on, T _i ^offthe minimum that is respectively unit i allows available machine time and minimum to allow downtime; x _{i, t-1}for the startup-shutdown time of unit i in the t-1 period, work as x _{i, t-1}within>=1 o'clock, represent the available machine time, work as x _{i, t-1}represent the unused time at≤-1 o'clock;

If unit was started shooting in the t-1 period, it does not meet minimum available machine time constraint in the t period and can not shut down; If unit was shut down in the t-1 period, it does not meet minimum constraint downtime in the t period and can not start shooting;

Step B204, heuristic contrary ranking method are revised unit start-stop state;

The unit having started is sorted from big to small by coa consumption rate at full capacity, if unit i residue unit output after the t period meets minimum start constraint and shuts down still meets maximum generation curve power, cut off it.

5. the backup configuration optimization method containing wind power system unit as claimed in claim 3, it is characterized in that, described 3 target functions with corresponding 3 generating curves, the prediction of obtaining each conventional unit under the prerequisite that meets system related constraint is exerted oneself, maximum output and minimum load, specifically in accordance with the following methods:

Step B301, meeting under the prerequisite of unit output restriction and system power Constraints of Equilibrium, obtain the unit output under maximum generation curve;

Unit output restriction under maximum generation curve is as follows:

$I_{i,t}{P}_i^{\min }\&le;I_{i,t}{P}_{i1,t}\&le;I_{i,t}{P}_i^{\max }$

In formula: be respectively the bound of exerting oneself of unit i;

System power Constraints of Equilibrium under maximum generation curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i1,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}+R_t^U$

Step B302, meeting under the prerequisite of the constraint such as unit creep speed, unit output restriction, power-balance, obtaining the unit output under minimum generating curve;

The relative unit maximum output of unit minimum load must meet the restriction of unit ramping rate constraints, and formula is as follows:

$\left\{\begin{array}{c}-{\mathrm{\&Delta;P}}_{\mathrm{down},i}\&le;P_{i1,t}-P_{i2,t-1}\&le;{\mathrm{\&Delta;P}}_{\mathrm{up},i}\\ -{\mathrm{\&Delta;P}}_{\mathrm{down},i}\&le;P_{i2,t}-P_{i1,t-1}\&le;{\mathrm{\&Delta;P}}_{\mathrm{up},i}\\ P_{i1,t}-P_{i2,t}\&le;\min ({\mathrm{\&Delta;P}}_{\mathrm{up},i},{\mathrm{\&Delta;P}}_{\mathrm{down},i})\end{array}\right.$

In formula: Δ P _{up, i}, Δ P _{down, i}be respectively the positive and negative creep speed of unit i;

The unit output restriction of unit minimum load is as follows simultaneously:

$I_{i,t}{P}_i^{\min }\&le;I_{i,t}{P}_{i2,t}\&le;I_{i,t}{P}_i^{\max }$

System power Constraints of Equilibrium under minimum generating curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i2,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}-R_t^D$

Step B303, the plan of restriction unit are exerted oneself between unit maximum output and unit minimum load, meeting under the prerequisite of power-balance constraint under plan generating curve, obtain unit plan and exert oneself simultaneously;

The unit plan restriction of exerting oneself is as follows:

I _i,tP _i2,t≤I _i,tP _i0,t≤I _i,tP _i1,t

Power-balance constraint under plan generating curve is as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_{i0,t}+{\stackrel{\&OverBar;}{P}}_{W,t}={\stackrel{\&OverBar;}{P}}_{L,t}$

Step B304, judge whether to meet branch road transmission capacity constraint, if met, finish; If do not met the relevant unit output of restriction, go to step B301 and redistribute unit output;

The constraint of branch road transmission capacity is as follows:

$\left\{\begin{array}{c}-f_j^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m0,t}\&le;f_j^{\max }\\ -f_J^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m1,t}\&le;f_j^{\max }\\ -f_j^{\max }\&le;\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}{G}_{j,m}{P}_{m2,t}\&le;f_j^{\max }\end{array}\right.$

In formula: G _j,mfor node m shifts distribution factor to the generating of circuit j; P _{m0, t}, P _{m1, t}, P _{m2, t}when respectively corresponding generating curve is for prediction generating curve, maximum generation curve, minimum generating curve, node m is at the injecting power of period t, maximum delivery power for circuit j; M is node sum.

6. the backup configuration optimization method containing wind power system unit as claimed in claim 1, it is characterized in that, described in step C, risk cost is added in cost of electricity-generating target function, by the total positive and negative spinning reserve capacity of system is carried out to optimizing, find the total positive and negative stand-by requirement of system hour of total cost, generation schedule and the positive and negative spinning reserve partition capacity of each unit, specifically in accordance with the following methods:

Step C1, obtain consider risk cost and cost of electricity-generating system synthesis this, as shown in the formula:

$F=F_0+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}(F_{U,t}+F_{D,t}+C_t^U+C_t^D)$

In formula: the total cost that F is system; F ₀for the cost of electricity-generating of system, F _u,tfor the loss of outage of system at period t; F _d,tfor the abandon windage loss of system at period t loses; for the positive stand-by cost of system at period t; for the negative stand-by cost of system at period t;

Positive and negative stand-by cost is according to the following formula:

$\left\{\begin{array}{c}{C}_t^U=c_R^U{R}_t^U\\ C_t^D=c_R^D{R}_t^D\end{array}\right.$

In formula: cost coefficient for the positive reserve capacity of system; cost coefficient for the positive reserve capacity of system;

Step C2, the direction that reduces by F be to the positive and negative standby optimizing of carrying out, and finds and make hour positive and negative standby of F, is best positive and negative backed-up value.

7. the backup configuration optimization method containing wind power system unit as claimed in claim 6, is characterized in that, the described direction reducing by F is to the positive and negative standby optimizing of carrying out, according to following steps:

Step C201, obtain current positive and negative system synthesis under standby and be originally designated as F ^(k);

Step C202, when last group of positive rotation reserve capacity as follows:

$R^{U,k}=[R_1^{U,k},R_2^{U,k},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{U,k},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{U,k}]$

From the t=1 period, start to align spinning reserve and carry out optimizing, until 24 periods completed; While carrying out optimizing to the positive rotation of period t is standby, keep the positive rotation backed-up value of other periods constant, order replace R ^u,kin obtain new R ^u,kthe total cost F of lower system ^(k1); Order replace R ^u,kin obtain new R ^u,kthe total cost F of lower system ^(k2);

If F ^(k1)< F ^(k)and F ^(k2)> F ^(k), the positive rotation of period t is standby should change by the direction increasing, order $R_t^{U,k+1}=R_t^{U,k}+\mathrm{\&Delta;}{R}_t^{U,k};$

If F ^(k1)> F ^(k)and F ^(k2)< F ^(k), the positive rotation of period t is standby should change by the direction reducing, order $R_t^{U,k+1}=R_t^{U,k}-\mathrm{\&Delta;}{R}_t^{U,k};$

If F ^(k1)> F ^(k)and F ^(k2)> F ^(k), order remodify calculate new this F of system synthesis ^(k1)and F ^(k2), re-start judgement; If when (δ is a less positive number), still there is F ^(k1)> F ^(k)and F ^(k2)> F ^(k), the current positive rotation backed-up value of period t is exactly optimum value,

By the optimizing to 24 period positive rotation backed-up values, can obtain one group of new positive rotation standby:

$R^{U,k+1}=[R_1^{U,k+1},R_2^{U,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{U,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{U,k+1}]$

By 24 period negative rotations being turned to the optimizing of backed-up value, can obtain one group of new negative rotation and turn backed-up value:

$R^{D,k+1}=[R_1^{D,k+1},R_2^{D,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_t^{D,k+1},\&CenterDot;\&CenterDot;\&CenterDot;,R_{24}^{D,k+1}]$

Step C203, obtain this F of system synthesis under one group of new positive and negative spinning reserve value ^(k+1);

Step C204, judged whether | F ^(k)-F ^(k+1)| < ε, if it is export required standby be best positive and negative standby; Otherwise go to step C202 and continue iteration.