CN105098839A - Uncertain wind power output-based coordinated optimization method for wind power grid connection - Google Patents

Abstract

The invention discloses an uncertain wind power output-based coordinated optimization method for wind power grid connection. The method comprises the following steps: collecting related parameters of a system; establishing a wind power prediction error calculation model and a wind power prediction error growth model, and obtaining uncertain wind power output according to the wind power prediction error growth model; on the basis of establishing a weight interval model that load can participate into scheduling and introducing a schedulable load and energy storage system, establishing a wind power cost model and establishing a coordinated optimization scheduling model for wind power grid connection of considering the schedulable load and energy storage system, and obtaining power output of various units within the scheduling period according to the optimization model; and calculating the wind curtailment rate of the system according to the power output of various units, and analyzing various costs and the standby condition of the system. According to the internet coordinated optimization model for wind power grid connection based on the uncertain wind power output and the schedulable load weight interval established by the method, the optimization scheduling scheme with obvious social and economic benefits is provided for scheduling personnel when solving the wind power integration problem.

Description

A kind of based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation

Technical field

The present invention relates to power system operation and scheduling field, particularly a kind of based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation.

Background technology

Current, be both at home and abroad the incorporated into the power networks research of economic dispatch of large-scale wind power is made some progress.Due to features such as wind power output randomness are large and fluctuation is strong, bring larger difficulty must to the grid-connected economic dispatch of large-scale wind power, therefore many scholars to wind power output prediction carried out large quantity research, but still be difficult to obtain predict the outcome accurately, wind power prediction error is also by long-term existence.Energy-storage system can effectively solve the problem, but it is envisaged that energy storage high cost, and efficiency is relatively low, increases the stored energy capacitance of system blindly, can reduce the economy of system equally.Part throttle characteristics and load level affect two important factors of power grid wind electricity digestion capability.Build the background of intelligent grid energetically in country under, load starts to play more and more important role, and it is no longer single electricity consumption side, and starts to carry out interaction as a kind of virtual plant and electrical network.Electrical network, by assessment load level, is reached the modes such as related protocol and is made a part of load incorporate generation schedule with user.

Summary of the invention

The object of the present invention is to provide a kind of based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, setting up, the wind-powered electricity generation increased based on error is uncertain exerts oneself on the basis of model, for uncertain the brought problem of exerting oneself of wind-powered electricity generation, establish wind power cost model; The role participated in system call due to each type load is different, and can the present invention participate in the degree of dispatching according to each type load, and the index of comprehensive each type load proposes the concept of schedulable load weight sector, and establishes model; Wind-electricity integration coordination optimization scheduling model is set up, for dispatching of power netwoks personnel provide the Optimized Operation scheme with obvious economic results in society when solving wind power integration problem in conjunction with energy-storage system.

The present invention is solved by the following technical programs:

Based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, comprise the following steps:

Step 1, the setting of the relevant parameters such as wind-powered electricity generation, fired power generating unit, energy-storage system and schedulable load and collection;

Wind-powered electricity generation parameter comprises the related data for determining wind power cost: wind-powered electricity generation predicts the P that exerts oneself _wF.jt, system burden with power P _d.t, the Wind turbines number N run in dispatching cycle _w, wind-powered electricity generation precision of prediction A _w.jt;

Fired power generating unit relevant parameter: the number N of the fired power generating unit run in dispatching cycle _g, fired power generating unit linearisation cost function coefficient a _i, fired power generating unit units limits upper limit lower limit the climbing that conventional power unit is exerted oneself and rate of descent

Energy-storage system relevant parameter: the efficiency for charge-discharge ψ of energy-storage system, energy storage system capacity bound E _min, E _max, t period energy-storage system power output P _e(t), energy-storage system charge-discharge electric power bound P in the unit interval _e.min, P _e.max, energy-storage system cost coefficient k _e(t);

Schedulable load relevant parameter: correction factor a, b, function coefficients k ₁(t), k ₂(t), t period r type load size P _d.rt, system loading number of types l;

Step 2, according to the setting of step 1 wind-powered electricity generation relevant parameter, draws the pre-permeability of wind-powered electricity generation on this basis, according to the pre-permeability of wind-powered electricity generation and wind power output precision of prediction, obtain wind power output predicated error, based on wind-powered electricity generation error model of growth, draw the uncertain output calculation model of wind-powered electricity generation;

Step 3, the uncertain output calculation model of wind-powered electricity generation is obtained according to step 2, setting up on the weight sector model that load can participate in dispatching and the basis of introducing schedulable load weight sector model and energy-storage system model, establish wind power cost model, and establish consideration schedulable load and energy-storage system wind-electricity integration coordination optimization scheduling model, according to this Optimized model, obtain each unit output in dispatching cycle;

Step 4, abandons wind rate by each unit output computing system, analyzes all kinds of cost of system and spare condition.

Further, in described step 2, the building process of wind-powered electricity generation error model of growth is:

Step 2.1, if certain given wind power output prediction data is at t ₀the error of period is e _t0, and under the condition not introducing other period error, when the t period, by e _t0the error caused is have nothing to do if exist with t and be greater than the constant A of 0, making the growth of error is then claimed to be linear; There is the constant B being greater than 1, make the growth of error is then claimed to make exponential;

Step 2.2, the present invention defines, and the wind power output prediction that predicated error linearly increases is stable, and the wind power output prediction that predicated error exponentially increases is unstable; Suppose that wind power output prediction is at t ₀, t ₁, t ₂..., t _nthe error of period is e _t0, e _t1, e _t2..., e _tnif wind power output prediction is stable, then at t ₁, t ₂..., t _n, t _n+1during the period, by e _t0, e _t1, e _t2..., e _tnthe error increment value caused is

$\left\{\begin{array}{c}\left|e_{t1}^{*}\right|A\left|e_{t0}\right|\\ \left|e_{t2}^{*}\right|=2A\left|e_{t0}\right|+A\left|e_{t1}\right|\\ L\\ \left|e_{tn}^{*}\right|=nA\left|e_{t0}\right|+(n-1)A\left|e_{t1}\right|+L+A\left|e_{tn-1}\right|\\ \left|e_{tn+1}^{*}\right|=nA\left|e_{t0}\right|+(n-1)A\left|e_{t1}\right|+L+A\left|e_{tn}\right|\end{array}\right.,$

Equally, if wind power output predicated error is unstable, then at t ₁, t ₂..., t _n, t _n+1during the period, by e _t0, e _t1, e _t2..., e _tnthe error increment value caused is

$\left\{\begin{array}{c}\left|e_{t1}^{\#}\right|=B\left|e_{t0}\right|\\ \left|e_{t2}^{\#}\right|=B^2\left|e_{t0}\right|+B\left|e_{t1}\right|\\ L\\ \left|e_{tn}^{\#}\right|=B^n\left|e_{t0}\right|+B^{(n-1)}\left|e_{t1}\right|+L+B\left|e_{tn-1}\right|\\ \left|e_{tn+1}^{\#}\right|=B^n\left|e_{t0}\right|+B^{(n-1)}\left|e_{t1}\right|+L+B\left|e_{tn}\right|\end{array}\right..$

Further, in described step 2, wind-powered electricity generation uncertain output calculation model is: if wind power output prediction is stable, then the uncertain output calculation model of wind-powered electricity generation increased based on error is:

Further, in described step 3, energy-storage system model is:

The current state (capacity) of energy-storage system meets for E (t): E (t)=E (t-1)+ψ P _e(t) Δ t, E _min≤ E (t)≤E _max,

The charge-discharge electric power P of day part _et () meets: P _e.min≤ P _e(t)≤P _e.max,

Then t period system call energy-storage system cost model is: C _e(t)=k _e(t) P _e(t).

Further, in described step 3, the process of establishing of schedulable load weight sector model is:

Step S3.1, can the degree that participate in dispatching in each period due to each type load in system different, so these loads are changed into different interval rank; Therefore in system call, just can obtain the rank of each type load in conjunction with actual conditions, and the span of getting properties level is interval as the initial weight of each type load, as r type load in the initial weight interval of t period is: $k_{r.t}^{\±}=\[k_{r.t}^{-},k_{r.t}^{+}\],$

Step S3.2, utilizes fuzzy mathematics relevant knowledge that above formula is carried out Fuzzy Processing, order: $\begin{array}{c}{h}_{r,t}^{-}=(k_{r.t}^{-}-a)/(b-a)\\ h_{r.t}^{+}=(k_{r.t}^{+}-a)/(b-a)\end{array},$ Then each type load at the power interval numbers of t period is: the power interval numbers model that each type load can participate in dispatching is: $\begin{array}{c}{s}_{r.t}^{\±}=\[s_{r.t}^{-},s_{r.t}^{+}\]=1-h_{r.t}^{\±}\\ =\[1-h_{r.t}^{+},1-h_{r.t}^{-}\]\end{array};$ Then the present invention define each type load the t period participate in dispatch weight be: $s_{r.t}=\frac{s_{r.t}^{-}+s_{r.t}^{+}}{2}=1-\frac{h_{r.t}^{+}+h_{r.t}^{-}}{2},$

Step S3.3, finally draws each type load cost calculation model: E _d.rt=k ₁(t) s _r.tp _d.rt+ k ₂(t) (s _r.tp _d.rt) ², then t period system call schedulable load weight sector model is:

Further, in described step 3, wind-powered electricity generation cost model is:

The present invention builds wind power cost model and comprises two parts: one is energy storage cost cost, and two is schedulable load costs; Then:

$F_{W.ji}=k_E\left(t\right)P_E\left(t\right)+k_1\left(t\right)\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}+k_2\left(t\right){\left(\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}\right)}^2$

$\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}={\mathrm{\ζ}}_t{P}_{WU.t}.$

P _E(t)＝(1-ζ _t)P _WU.t

Further, the economic dispatch target function in described step 3 and constraints are:

Target function:

$\min F=\min \underset{t=1}{\overset{T}{\Σ}}(\underset{i=1}{\overset{N_G}{\Σ}}{F}_{G.it}+\underset{j=1}{\overset{N_W}{\Σ}}{F}_{W.jt}),$

Constraints:

(1) power-balance constraint

$\underset{i=1}{\overset{N_G}{\Σ}}{P}_{G.it}+\underset{j=1}{\overset{N_W}{\Σ}}{P}_{W.jt}=P_E\left(t\right)+\underset{r=1}{\overset{l}{\Σ}}{P}_{D.rt},$

(2) unit power output constraint

(3) ramping rate constraints

$\{\begin{array}{c}{P}_{G.it}-P_{G.i(t-1)}\≤{\mathrm{\ξ}}_i^{up}t\\ P_{G.i(t-1)}-P_{G.it}\≤{\mathrm{\ξ}}_i^{down}t\end{array},$

(4) energy-storage system constraint

$\left\{\begin{array}{c}E\left(t\right)=E(t-1)+{\mathrm{\ψP}}_E\left(t\right)\mathrm{\Δ}t\\ E_{\min }\≤E\left(t\right)\≤E_{\max }\\ P_{E.\min }\≤P_E\left(t\right)\≤P_{E.max}\end{array}\right..$

Therefore there is following technique effect herein:

The wind-powered electricity generation that calculates of the uncertain model of exerting oneself of wind-powered electricity generation increased based on error that the present invention sets up is uncertain, and exert oneself can than more comprehensive response error information, for follow-up scheduling provides data more accurately; Schedulable weight sector model can formulate corresponding standard for different stage load at the schedulability of Different periods, avoid the wasting of resources under unified standard, schedulable load is enable fully effectively to participate in the uncertain adjustment of exerting oneself of wind-powered electricity generation, thus the use amount reduced energy-storage system, reduce the reserve capacity of abandoning wind rate and system.The participation of energy-storage system, can make up schedulable load under extreme case and cannot participate in the shortcoming that wind-powered electricity generation regulates, in the process of schedulable load being carried out to weight sector division, bear certain standby effect; The present invention for dispatching of power netwoks personnel solve wind power integration problem time the Optimized Operation scheme with obvious economic results in society is provided.

Accompanying drawing explanation

Accompanying drawing 1 is wind power output change and the uncertain schematic diagram of exerting oneself of wind-powered electricity generation;

Accompanying drawing 2 is energy-storage system discharge and recharge schematic diagrames;

Accompanying drawing 3 is system reserve demand schematic diagrames in dispatching cycle.

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention program is described in further detail.

Fig. 1 illustrates the uncertain component part of exerting oneself of wind-powered electricity generation, because error is sustainable growth to a certain extent, wind-powered electricity generation is uncertain exerts oneself to be increased by wind-powered electricity generation predicated error and wind-powered electricity generation wind-powered electricity generation predicated error and forms;

Accompanying drawing 2 uses energy-storage system discharge and recharge schematic diagram, because energy-storage system is subject to the restriction of capacity and charging and discharging capabilities thereof, depend merely on energy-storage system to carry out uncertain the exerting oneself of regulating wind power and implement more difficult in some cases, therefore, the present invention introduces schedulable load and carrys out collaborative energy-storage system wind-powered electricity generation is uncertain exerts oneself to process;

The present invention is based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, specifically implement as follows:

The setting of step 1, relevant parameter and collection

The setting of step 1.1, wind-powered electricity generation relevant parameter and collection

Wind-powered electricity generation parameter mainly comprises: wind-powered electricity generation predicts the P that exerts oneself _wF.jt, system burden with power P _d.t,the Wind turbines number N run in dispatching cycle _w, wind-powered electricity generation precision of prediction A _w.jt;

The setting of step 1.2, fired power generating unit relevant parameter and collection

Fired power generating unit relevant parameter: the number N of the fired power generating unit run in dispatching cycle _g, fired power generating unit linearisation cost function coefficient a _i, fired power generating unit units limits upper limit lower limit the climbing that conventional power unit is exerted oneself and rate of descent

The setting of step 1.3, energy-storage system relevant parameter and collection

Energy-storage system relevant parameter: the efficiency for charge-discharge ψ of energy-storage system, energy storage system capacity bound E _min, E _max, energy-storage system charge-discharge electric power bound P in the unit interval _e.min, P _e.max, energy-storage system cost coefficient k _e(t);

The setting of step 1.4, schedulable load relevant parameter and collection

Schedulable load relevant parameter: correction factor a, b, function coefficients k ₁(t), k ₂(t), t period r type load size P _d.rt, system loading number of types l.

Step 2, setting according to step 1 wind-powered electricity generation relevant parameter, draw the pre-permeability of wind-powered electricity generation on this basis, according to the pre-permeability of wind-powered electricity generation and wind power output precision of prediction, obtain wind power output predicated error, based on error model of growth, show that wind-powered electricity generation is uncertain and exert oneself.

Wind-powered electricity generation error increases computation model:

If certain given wind power output prediction data is at t ₀the error of period is e _t0, and under the condition not introducing other period error, when the t period, by e _t0the error caused is if

(1) existence and t have nothing to do and are greater than the constant A of 0, make

$\begin{array}{cc}\left|e_t^{*}\right|\≈A(t-t_0)\left|e_{t0}\right|& (t=t_0+1,t_0+2,L),\end{array}$

The growth of error is then claimed to be linear;

(2) there is the constant B being greater than 1, make

$\begin{array}{cc}\left|e_t^{\#}\right|\≈B^{(t-t_0)}\left|e_{t0}\right|& (t=t_0+1,t_0+2,L)\end{array},$

The growth of error is then claimed to make exponential;

The present invention defines, and the wind power output prediction that predicated error linearly increases is stable, and the wind power output prediction that predicated error exponentially increases is unstable;

Suppose that wind power output prediction is at t ₀, t ₁, t ₂..., t _nthe error of period is e _t0, e _t1, e _t2..., e _tnif wind power output prediction is stable, then at t ₁, t ₂..., t _n, t _n+1during the period, by e _t0, e _t1, e _t2..., e _tnthe error increment value caused is

$\left\{\begin{array}{c}\left|e_{t1}^{*}\right|A\left|e_{t0}\right|\\ \left|e_{t2}^{*}\right|=2A\left|e_{t0}\right|+A\left|e_{t1}\right|\\ L\\ \left|e_{tn}^{*}\right|=nA\left|e_{t0}\right|+(n-1)A\left|e_{t1}\right|+L+A\left|e_{tn-1}\right|\\ \left|e_{tn+1}^{*}\right|=nA\left|e_{t0}\right|+(n-1)A\left|e_{t1}\right|+L+A\left|e_{tn}\right|\end{array}\right.,$

Equally, if wind power output predicated error is unstable, then at t ₁, t ₂..., t _n, t _n+1during the period, by e _t0, e _t1, e _t2..., e _tnthe error increment value caused is

$\left\{\begin{array}{c}\left|e_{t1}^{\#}\right|=B\left|e_{t0}\right|\\ \left|e_{t2}^{\#}\right|=B^2\left|e_{t0}\right|+B\left|e_{t1}\right|\\ L\\ \left|e_{tn}^{\#}\right|=B^n\left|e_{t0}\right|+B^{(n-1)}\left|e_{t1}\right|+L+B\left|e_{tn-1}\right|\\ \left|e_{tn+1}^{\#}\right|=B^n\left|e_{t0}\right|+B^{(n-1)}\left|e_{t1}\right|+L+B\left|e_{tn}\right|\end{array}\right.;$

The uncertain output calculation model of wind-powered electricity generation is:

Due to wind-powered electricity generation prediction be stable predict with wind-powered electricity generation be unstable two kinds of situations under, the wind-powered electricity generation uncertain output calculation method increased based on error is identical, the situation of the present invention when this discussion wind power output is predicted as stable.Wind power output is predicted as unstable timing cases and can similarly tries to achieve;

If wind power output prediction is stable, then the uncertain output calculation model of wind-powered electricity generation increased based on error is shown below:

$P_{WU.t}=e_t+\left|e_t^{*}\right|.$

Step 3, to obtain according to step 2 that wind-powered electricity generation is uncertain exerts oneself, setting up on the weight sector model that load can participate in dispatching and the basis of introducing schedulable load weight sector model and energy-storage system model, establish wind power cost model, and establish consideration schedulable load and energy-storage system wind-electricity integration coordination optimization scheduling model, according to this Optimized model, obtain each unit output in dispatching cycle.

Energy-storage system (ESS) model is:

The current state (capacity) of ESS meets for E (t):

E(t)＝E(t-1)+ψP _E(t)Δt，

E _min≤ E (t)≤E _max, through a large amount of optimum experimental, wherein, P _et () is the power output of t period energy-storage system, E _min=0, E _max=40, ψ=0.95,

The charge-discharge electric power P of day part _et () meets:

P _e.min≤ P _e(t)≤P _e.max, through a large amount of optimum experimental wherein, P _e.min=-25, P _e.max=25,

Then t period system call energy-storage system ESS cost model is:

C _E(t)＝k _E(t)P _E(t)。

Schedulable load weight sector model is:

Load weight represents that all kinds of load could participate in the degree of dispatching in systems in which, the role participated in system call due to each type load is different, so can participate in the degree of scheduling according to each type load, and the index of comprehensive each type load, based on interval number, load is divided into 3 types;

Can the degree that participate in dispatching in each period due to each type load in system different, so these loads are changed into different interval rank.Therefore in system call, just can obtain the rank of each type load in conjunction with actual conditions, and the span of getting properties level is interval as the initial weight of each type load, as r type load in the initial weight interval of t period is:

$k_{r.t}^{\±}=\[k_{r.t}^{-},k_{r.t}^{+}\],$

Utilize fuzzy mathematics relevant knowledge that above formula is carried out Fuzzy Processing, order:

$\begin{array}{c}{h}_{r,t}^{-}=(k_{r.t}^{-}-a)/(b-a)\\ h_{r.t}^{+}=(k_{r.t}^{+}-a)/(b-a)\end{array},$ Through a large amount of optimum experimental wherein, a, b are respectively 0.5,1,

Then each type load at the power interval numbers of t period is:

$k_{r.t}^{\±}=\[k_{r.t}^{-},k_{r.t}^{+}\],$

The power interval numbers model that each type load can participate in dispatching is:

$\begin{array}{c}{s}_{r.t}^{\±}=\[s_{r.t}^{-},s_{r.t}^{+}\]=1-h_{r.t}^{\±}\\ =\[1-h_{r.t}^{+},1-h_{r.t}^{-}\]\end{array},$

Then defining each type load herein in the weight that the t period participates in dispatching is:

$s_{r.t}=\frac{s_{r.t}^{-}+s_{r.t}^{+}}{2}=1-\frac{h_{r.t}^{+}+h_{r.t}^{-}}{2},$

Each type load cost calculation model:

E _D.rt＝k ₁(t)s _r.tP _D.rt+k ₂(t)(s _r.tP _D.rt) ²，

Then t period system call schedulable load weight sector model is:

Wind power cost model is:

Wind-powered electricity generation is abandoned wind reason and is mainly comprised the following aspects: 1. Wind turbines increases too fast, grid connected wind power capacity system of system digestion capability far away, then cause electrical network cannot fully to dissolve in time wind-powered electricity generation amount; 2. the ability to send outside of wind-powered electricity generation is not enough, if when wind-powered electricity generation amount fully cannot be dissolved in this locality, if the transmittability of circuit is not enough, then part will be caused to abandon wind; 3. wind-powered electricity generation predicated error; 4. dispatching of power netwoks scarce capacity.

Abandon wind reason in conjunction with above-mentioned wind-powered electricity generation, wind power cost model of building comprises two parts herein: one is energy storage cost cost; Two is schedulable load costs.

$F_{W.ji}=k_E\left(t\right)P_E\left(t\right)+k_1\left(t\right)\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}+k_2\left(t\right){\left(\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}\right)}^2$

$\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}={\mathrm{\ζ}}_t{P}_{WU.t}.$

P _E(t)＝(1-ζ _t)P _WU.t

Target function and constraints are:

Target function:

$\min F=\min \underset{t=1}{\overset{T}{\Σ}}(\underset{i=1}{\overset{N_G}{\Σ}}{F}_{G.it}+\underset{j=1}{\overset{N_W}{\Σ}}{F}_{W.jt}),$

Constraints:

(1) power-balance constraint

$\underset{i=1}{\overset{N_G}{\Σ}}{P}_{G.it}+\underset{j=1}{\overset{N_W}{\Σ}}{P}_{W.jt}=P_E\left(t\right)+\underset{r=1}{\overset{l}{\Σ}}{P}_{D.rt},$

(2) unit power output constraint

(3) ramping rate constraints

$\{\begin{array}{c}{P}_{G.it}-P_{G.i(t-1)}\≤{\mathrm{\ξ}}_i^{up}t\\ P_{G.i(t-1)}-P_{G.it}\≤{\mathrm{\ξ}}_i^{down}t\end{array},$

(4) ESS constraint

$\left\{\begin{array}{c}E\left(t\right)=E(t-1)+{\mathrm{\ψP}}_E\left(t\right)\mathrm{\Δ}t\\ E_{\min }\≤E\left(t\right)\≤E_{\max }\\ P_{E.\min }\≤P_E\left(t\right)\≤P_{E.max}\end{array}\right..$

Step 4, abandon wind rate by each unit output computing system, all kinds of cost of system and spare condition are analyzed.

The present invention, to have 10 machine systems of Large-scale Wind Power field for example, adopts 24h plan, but with 15 minutes for interval (T=96), is optimized emulation by MATLAB simulation software.

Owing to carrying out the division of interval weight to schedulable load, make all kinds of schedulable load can play certain complementation in the uncertain problem of exerting oneself of Different periods process wind-powered electricity generation, add energy-storage system standby effect, schedulable load is made fully to participate in the uncertain adjustment of exerting oneself of wind-powered electricity generation, not only save system cost, and reduction system abandons wind rate.

Accompanying drawing 3 is system reserve demand schematic diagrames in dispatching cycle, can find out in figure, stand-by requirement is all met, there is provided for subsequent use by energy-storage system can not regulate and control the period, there is provided for subsequent use in energy-storage system power storage quota or without during power storage by schedulable load, serve complementary effect.Simultaneously because the schedulability of category-A schedulable load and category-B schedulable load is opposition within the major part period, therefore two class schedulable loads also serve certain complementation within dispatching cycle;

The present invention proposes and a kind ofly to exert oneself with the wind-powered electricity generation of schedulable load weight sector and coordination optimizing method based on wind-powered electricity generation is uncertain.First, the present invention has carried out correlative study to wind-powered electricity generation predicated error, proposes the concept of the pre-permeability of wind-powered electricity generation, in conjunction with wind-powered electricity generation precision of prediction, sets up wind-powered electricity generation predicated error model; Secondly, set up wind-powered electricity generation predicated error model of growth, and establish the uncertain model of exerting oneself of wind-powered electricity generation increased based on error; Consider the schedulable ability of different load, propose the concept of schedulable load weight sector, and establish model; On above Research foundation, set up and exert oneself and the wind-electricity integration Coordination and Optimization Model of schedulable load weight sector based on wind-powered electricity generation is uncertain, for dispatching of power netwoks personnel provide the Optimized Operation scheme with obvious economic results in society when solving wind power integration problem.

Be to be understood that, although this specification is described according to execution mode, but not each execution mode only comprises an independently technical scheme, this narrating mode of specification is only for clarity sake, those skilled in the art should by specification integrally, technical scheme in each execution mode also through appropriately combined, can form other execution modes that it will be appreciated by those skilled in the art that.

A series of detailed description listed is above only illustrating for feasibility execution mode of the present invention; they are also not used to limit the scope of the invention, and allly do not depart from the skill of the present invention equivalent implementations done of spirit or change and all should be included in protection scope of the present invention.

Claims (7)

1., based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, it is characterized in that, comprise the following steps:

Step 1, the setting of the relevant parameters such as wind-powered electricity generation, fired power generating unit, energy-storage system and schedulable load and collection;

Wind-powered electricity generation parameter comprises the related data for determining wind power cost: wind-powered electricity generation predicts the P that exerts oneself _wF.jt, system burden with power P _d.t, the Wind turbines number N run in dispatching cycle _w, wind-powered electricity generation precision of prediction A _w.jt;

Fired power generating unit relevant parameter: the number N of the fired power generating unit run in dispatching cycle _g, fired power generating unit linearisation cost function coefficient a _i, fired power generating unit units limits upper limit lower limit the climbing that conventional power unit is exerted oneself and rate of descent ${\mathrm{\ξ}}_i^{\mathrm{down}};$

Energy-storage system relevant parameter: the efficiency for charge-discharge Ψ of energy-storage system, energy storage system capacity bound E _min, E _max, t period energy-storage system power output P _e(t), energy-storage system charge-discharge electric power bound P in the unit interval _e.min, P _e.max, energy-storage system cost coefficient k _e(t);

Schedulable load relevant parameter: correction factor a, b, function coefficients k ₁(t), k ₂(t), t period r type load size P _d.rt, system loading number of types l;

Step 2, according to the setting of step 1 wind-powered electricity generation relevant parameter, draws the pre-permeability of wind-powered electricity generation on this basis, according to the pre-permeability of wind-powered electricity generation and wind power output precision of prediction, obtain wind power output predicated error, based on wind-powered electricity generation error model of growth, draw the uncertain output calculation model of wind-powered electricity generation;

Step 3, the uncertain output calculation model of wind-powered electricity generation is obtained according to step 2, setting up on the weight sector model that load can participate in dispatching and the basis of introducing schedulable load weight sector model and energy-storage system model, establish wind power cost model, and establish consideration schedulable load and energy-storage system wind-electricity integration coordination optimization scheduling model, according to this Optimized model, obtain each unit output in dispatching cycle;

Step 4, abandons wind rate by each unit output computing system, analyzes all kinds of cost of system and spare condition.

2. according to claims 1 based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, it is characterized in that, in described step 2, the building process of wind-powered electricity generation error model of growth is:

Step 2.1, if certain given wind power output prediction data is at t ₀the error of period is e _t0, and under the condition not introducing other period error, when the t period, by e _t0the error caused is have nothing to do if exist with t and be greater than the constant A of 0, making the growth of error is then claimed to be linear; There is the constant B being greater than 1, make the growth of error is then claimed to make exponential;

Step 2.2, the present invention defines, and the wind power output prediction that predicated error linearly increases is stable, and the wind power output prediction that predicated error exponentially increases is unstable; Suppose that wind power output prediction is at t ₀, t ₁, t ₂..., t _nthe error of period is e _t0, e _t1, e _t2..., e _tnif wind power output prediction is stable, then at t ₁, t ₂..., t _n, t _n+1during the period, by e _t0, e _t1, e _t2..., e _tnthe error increment value caused is

$\left\{\begin{array}{c}\left|e_{t1}^{*}\right|A\left|e_{t0}\right|\\ \left|e_{t2}^{*}\right|=2A\left|e_{t0}\right|+A\left|e_{t1}\right|\\ L\\ \left|e_{tn}^{*}\right|=nA\left|e_{t0}\right|+(n-1)A\left|e_{t1}\right|+L+A\left|e_{tn-1}\right|\\ \left|e_{tn+1}^{*}\right|=nA\left|e_{t0}\right|+(n-1)A\left|e_{t1}\right|+L+A\left|e_{tn}\right|\end{array}\right.,$

Equally, if wind power output predicated error is unstable, then at t ₁, t ₂..., t _n, t _n+1during the period, by e _t0, e _t1, e _t2..., e _tnthe error increment value caused is

$\left\{\begin{array}{c}\left|e_{t1}^{\#}\right|=B\left|e_{t0}\right|\\ \left|e_{t2}^{\#}\right|=B^2\left|e_{t0}\right|+B\left|e_{t1}\right|\\ L\\ \left|e_{tn}^{\#}\right|=B^n\left|e_{t0}\right|+B^{(n-1)}\left|e_{t1}\right|+L+B\left|e_{tn-1}\right|\\ \left|e_{tn+1}^{\#}\right|=B^n\left|e_{t0}\right|+B^{(n-1)}\left|e_{t1}\right|+L+B\left|e_{tn}\right|\end{array}\right..$

3. according to claims 1 based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, it is characterized in that, in described step 2, wind-powered electricity generation uncertain output calculation model is: if wind power output prediction is stable, then the uncertain output calculation model of wind-powered electricity generation increased based on error is:

4. according to claims 1 based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, it is characterized in that, in described step 3, energy-storage system model is:

The current state (capacity) of energy-storage system meets for E (t): E (t)=E (t-1)+Ψ P _e(t) Δ t, E _min≤ E (t)≤E _max,

The charge-discharge electric power P of day part _et () meets: P _e.min≤ P _e(t)≤P _e.max,

Then t period system call energy-storage system cost model is: C _e(t)=k _e(t) P _e(t).

5. according to claims 1 based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, it is characterized in that, in described step 3, the process of establishing of schedulable load weight sector model is:

Step S3.1, can the degree that participate in dispatching in each period due to each type load in system different, so these loads are changed into different interval rank; Therefore in system call, just can obtain the rank of each type load in conjunction with actual conditions, and the span of getting properties level is interval as the initial weight of each type load, as r type load in the initial weight interval of t period is: $k_{r.t}^{\±}=\[k_{r.t}^{-},k_{r.t}^{+}\],$

Step S3.2, utilizes fuzzy mathematics relevant knowledge that above formula is carried out Fuzzy Processing, order: $\begin{array}{c}{h}_{r,t}^{-}=(k_{r.t}^{-}-a)/(b-a)\\ h_{r.t}^{+}=(k_{r.t}^{+}-a)/(b-a)\end{array},$ Then each type load at the power interval numbers of t period is: the power interval numbers model that each type load can participate in dispatching is: $\begin{array}{c}{s}_{r.t}^{\±}=\[s_{r.t}^{-},s_{r.t}^{+}\]=1-h_{r.t}^{\±}\\ =\[1-h_{r.t}^{+},1-h_{r.t}^{-}\]\end{array};$ Then the present invention define each type load the t period participate in dispatch weight be: $s_{r.t}=\frac{s_{r.t}^{-}+s_{r.t}^{+}}{2}=1-\frac{h_{r.t}^{+}+h_{r.t}^{-}}{2},$

Step S3.3, finally draws each type load cost calculation model: E _d.rt=k ₁(t) s _r.tp _d.rt+ k ₂(t) (s _r.tp _d.rt) ², then t period system call schedulable load weight sector model is:

6. according to claims 1 based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, it is characterized in that, in described step 3, wind-powered electricity generation cost model is:

The present invention builds wind power cost model and comprises two parts: one is energy storage cost cost, and two is schedulable load costs; Then:

$F_{W.ji}=k_E\left(t\right)P_E\left(t\right)+k_1\left(t\right)\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}+k_2\left(t\right){\left(\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}\right)}^2$

$\underset{r=1}{\overset{l}{\Σ}}{s}_{r.t}{P}_{D.rt}={\mathrm{\ζ}}_t{P}_{WU.t}.$

P _E(t)＝(1-ζ _t)P _WU.t

7. according to claims 1 based on the uncertain wind-electricity integration coordination optimizing method of exerting oneself of wind-powered electricity generation, it is characterized in that, the economic dispatch target function in described step 3 and constraints are:

Target function:

$\min F=\min \underset{t=1}{\overset{T}{\Σ}}(\underset{i=1}{\overset{N_G}{\Σ}}{F}_{G.it}+\underset{j=1}{\overset{N_W}{\Σ}}{F}_{W.jt}),$

Constraints:

(1) power-balance constraint

$\underset{i=1}{\overset{N_G}{\Σ}}{P}_{G.it}+\underset{j=1}{\overset{N_W}{\Σ}}{P}_{W.jt}=P_E\left(t\right)+\underset{r=1}{\overset{l}{\Σ}}{P}_{D.rt},$

(2) unit power output constraint

(3) ramping rate constraints

$\{\begin{array}{c}{P}_{G.it}-P_{G.i(t-1)}\≤{\mathrm{\ξ}}_i^{up}t\\ P_{G.i(t-1)}-P_{G.it}\≤{\mathrm{\ξ}}_i^{down}t\end{array},$

(4) energy-storage system constraint

$\left\{\begin{array}{c}E\left(t\right)=E(t-1)+{\mathrm{\ψP}}_E\left(t\right)\mathrm{\Δ}t\\ E_{\min }\≤E\left(t\right)\≤E_{\max }\\ P_{E.\min }\≤P_E\left(t\right)\≤P_{E.max}\end{array}\right..$