CN103872694B - The capacity optimization of regional wind power group energy-accumulating power station and auxiliary peak regulating method thereof - Google Patents

Abstract

The invention discloses a kind of capacity optimization of regional wind power group energy-accumulating power station and auxiliary peak regulating method thereof, comprise: 1) consider that wind farm group power fluctuation stabilizes the BESS complex function discharge and recharge implementation method with peak load regulation network, optimize computation model so that BESS operating cost is minimum for target builds capacity; 2) utilize ultra-short term wind power prediction, synchronously participate in power fluctuation with BESS and stabilize with Income Maximum during peak load regulation network for target builds real-time optimization moving model; 3) cost sharing of BESS system cloud gray model and distribution of income pattern and method are proposed.Adopt technical scheme of the present invention can plan the energy-storage system of the required configuration of regional wind power, optimize and the rational stored energy capacitance of computing economy, there is provided energy-storage system to assist the prioritization scheme participating in peak load regulation network, the reasonable Effec-tive Function level of lifting region wind field configuration energy-storage system simultaneously.

Description

The capacity optimization of regional wind power group energy-accumulating power station and auxiliary peak regulating method thereof

Technical field

The present invention relates to a kind of capacity optimization of regional wind power group energy-accumulating power station and auxiliary peak regulating method thereof.

Background technology

The lasting lifting of intermittent power supply permeability, makes uncertain factor in power grid energy equilibrium system constantly increase, and current electric grid regulation and control face immense pressure.Energy-storage system changes the ability of energy flow time and space idea because possessing, give birth to, and become the important feasible pattern promoting intermittent power supply controllability and utilization ratio as the new link application in conventional energy stream transfer chain ^[1].

For configuring battery energy storage power station in wind energy turbine set dispersion energy storage and regional wind power group, (English of battery energy storage power station is BatteryEnergyStorageStation to current Chinese scholars, be called for short BESS) all expand correlative study, wherein disperse energy storage mainly to concentrate on the optimum calculating of capacity of energy-storage system and the Optimal Control Strategy of energy-storage system, its function is mainly positioned stabilizing of wind power fluctuation.Wherein, calculation of capacity belongs to planning problem, is to utilize typical history data, builds optimization object function according to the leading factor affecting storage energy operation, determines to realize the optimum stored energy capacitance of wind energy turbine set expectation needed for power stage; The latter is to realize the controlling and adjustment to actual storage energy operation system, and its target is generally the economical operation of energy-storage system or the validity of smoothing fluctuations.

With wind energy turbine set disperse energy storage first than, BESS functional localization has stronger extensibility, it possesses stand-by power supply, intermittent power supply is stabilized, peak regulation, frequency modulation, transient state the are gained merit multi-functional such as emergency response and transient voltage first support.Research at present for BESS mainly concentrates on the fields such as BESS performance comparison, the functional analysis of sound state and supervisory control system.

Ding Ming, Xu Nin's week, Bi Rui etc. the 3 class battery energy storage power station systems based on comprehensive modeling can comparative analysis [J]. Automation of Electric Systems, 2011,35(15): 34-39 establishes comprehensive compatible model to the BESS being energy-accumulating medium with lithium battery, sodium-sulphur battery and flow battery, and has carried out comparative analysis to performance, effect and economic index etc.

Ding Ming, Xu Ningzhou, Lin Gende. the research [J] of the static function of battery energy storage power station. electrotechnics journal, 2012,27(10): 242-248 then inquires into the static function of BESS respectively, and proposes discharge and recharge and the scheduling strategy of optimization accordingly.

For domestic built BESS demonstration project, document Lee builds woods, Xie Zhijia, Huidong etc. and national wind-light storage transmission demonstration project is introduced and typical mode of operation analysis [J]. Automation of Electric Systems, 2013,37(1): 59-64; Lu Zhigang, Wang Ke, Liu Yi. Shenzhen Baoqing lithium battery energy storage battery power station key technology and the complete method for designing of system [J]. Automation of Electric Systems, 2013,37(1): 65-69 describes key technology and the operational mode of national wind-light storage transmission demonstration project and Shenzhen Baoqing BESS respectively.

Document Cui Qiang, Wang Xiuli, Liu Zuyong. take into account the research of interlock electricity price and the performance analysis [J] thereof of energy-accumulating power station operation under market environment. Proceedings of the CSEE, 2013,33(13): 62-68 then establishes BESS and Generation Side, the Spot Price dynamic game that responds for electrical measurement and user links model, and analyzes the benefit under its market environment.

Above-mentioned document has positive effect for advancing the correlative study of BESS, but flexible for functional localization, that operation fitness is high BESS system, clearer and more definite correlative study is had not yet to see to its field such as carrying out practically pattern, complex function implementation strategy.

Summary of the invention

The object of the invention is for overcoming above-mentioned the deficiencies in the prior art, a kind of capacity optimization of regional wind power group energy-accumulating power station and auxiliary peak regulating method thereof are provided.

For achieving the above object, the present invention adopts following technical proposals:

The capacity optimization of regional wind power group energy-accumulating power station and an auxiliary peak regulating method thereof, comprising:

1) consider that wind farm group power fluctuation stabilizes the BESS complex function discharge and recharge implementation method with peak load regulation network, optimize computation model so that BESS operating cost is minimum for target builds capacity;

2) utilize ultra-short term wind power prediction, synchronously participate in power fluctuation with BESS and stabilize with Income Maximum during peak load regulation network for target builds real-time optimization moving model;

3) cost sharing of BESS system cloud gray model and distribution of income pattern and method are proposed.

In step 1), wind farm group BESS stabilizes in power fluctuation process, as wind farm group power P (t) >P _reft, time (), BESS system is charged; P (t) <P _reftime (t), BESS system discharge; And as P (t)=P _reft time (), be then in floating charge state, wherein P (t) is wind farm group power, P _reft () is for expecting power output; When charge power, state-of-charge (SOC) or charge power rate of change are greater than BESS charge power rate of change maximum, then there will be and abandon wind; Shown in (1-3),

P(t)-P _ref(t)>P _max-cha（1）

SOC(t)>SOC _max（2）

ΔP _cha(t)>ΔP _max-cha（3）

In formula, P _max-chafor the maximum charge power of BESS; SOC _maxfor the up limit value of SOC, get 1.0; Δ P _chat ()=P (t)-P (t-1) is charge power rate of change; Δ P _max-chafor the charge power rate of change maximum of BESS;

Equally, when discharge power and rate of change thereof are greater than charge power rate of change maximum, or SOC lower than descending limit value time, shown in (4-6), then can occur stabilizing insufficient because stabilizing underpower;

｜P(t)-P _ref(t)｜>P _max-discha（4）

SOC(t)<SOC _min（5）

ΔP _discha(t)>ΔP _max-discha（6）

In formula, P _max-discha>0 is the maximum discharge power of BESS; SOC _minfor the descending limit value of SOC, get 0.1; Δ P _dischat () is discharge power rate of change; Δ P _max-dischafor the discharge power rate of change maximum of BESS.

In step 1), BESS participates in peak regulation can realize the storage of paddy lotus phase wind energy, has both been beneficial to the pressure reducing thermoelectricity degree of depth peak regulation, again and can be peak load and provide power supply to supply, simultaneously for BESS idle capacity provides income operational mode; The discharge and recharge behavior participating in peak regulation by according to BESS during paddy lotus and peak load can charge and discharge energy, invariable power stores or release within a certain period of time;

E _cha=P _cha·Δt _cha（7）

E _discha=P _discha·Δt _discha（8）

Wherein, E _cha, E _dischabe respectively the charging of participation peak regulation, discharge energy; P _cha, P _dischabe respectively charge-discharge electric power during peak regulation; Δ t _cha, Δ t _dischabe respectively the discharge and recharge duration.

In described step 1) with BESS operating cost minimum for target build capacity optimize computation model be:

minC=C _cons·k _T/k+C _aba+C _lack+C _outline；

In formula, minC is the minimum cost run with BESS; K is that BESS service life amounts to a year number; k _tgot interval duration T by cost calculation and amounted to a year number; C _consfor construction cost; C _abafor abandoning eolian; C _lackfor desired output punishment cost; The out-of-limit punishment cost C of SOC _outline.

C _cons=γ·V+ε·V（9）

$C_{\mathrm{aba}}=\mathrm{\&beta;}\&CenterDot;\{\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{aba}-1}\left(t\right)\&CenterDot;(P\left(t\right)-P_{\mathrm{ref}}\left(t\right)-{P\max -\mathrm{cha}}_{)}\mathrm{\&Delta;t}))$

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{aba}-2}\left(t\right)\&CenterDot;[(P\left(t\right)-P_{\mathrm{ref}}\left(t\right))\mathrm{\&Delta;t}-$ （10）

(SOC _max-SOC(t-1))V]

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{aba}-3}\left(t\right)\&CenterDot;(\mathrm{\&Delta;P}\left(t\right)-\mathrm{\&Delta;}{P}_{\max -\mathrm{cha}})\mathrm{\&Delta;t}\}$

$C_{\mathrm{lack}}=\mathrm{\&alpha;}\&CenterDot;\{\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{lack}-1}\left(t\right)\&CenterDot;\left(\right|P\left(t\right)-P_{\mathrm{ref}}\left(t\right)|-P_{\max -\mathrm{discha}})\mathrm{\&Delta;t}))$

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{lack}-2}\left(t\right)\&CenterDot;\left[\right|P\left(t\right)-P_{\mathrm{ref}}\left(t\right)|\mathrm{\&Delta;t}-$ （11）

(SOC(t-1)-SOC _min)V]

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{lack}-3}\left(t\right)\&CenterDot;\left(\right|\mathrm{\&Delta;P}\left(t\right)|-\mathrm{\&Delta;}{P}_{\max -\mathrm{discha}})\mathrm{\&Delta;t}\}$

$C_{\mathrm{outline}}=\mathrm{\&eta;}\&CenterDot;\{\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{outline}-1}\left(t\right)\&CenterDot;(\mathrm{SOC}\left(t\right)-{\mathrm{SOC}}_{\mathrm{Hline}})V$ （12）

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{outline}-2}\left(t\right)|\mathrm{SOC}\left(t\right)-{\mathrm{SOC}}_{\mathrm{Lline}}|\}$

In formula, γ is unit capacity batteries energy storage basic cost; V is the rated capacity of BESS; ε is BESS installation cost; T by the time window length of extraction service data; β be unit capacity abandon eolian; α causes the unit punishment cost of lacked capacity for stabilizing underpower; η be SOC out-of-limit run time unit capacity punishment cost; SOC (t-1) is corresponding previous moment SOC value; Δ t is the sampling interval; SOC _hline, SOC _llinebe respectively SOC and normally run upper lower limit value, work as SOC(t) >SOC _hlinetime, BESS still can charge to SOC _max, but its out-of-limit working capacity will produce out-of-limit punishment cost, SOC _hlinevalue is 0.85, SOC _llinevalue is 0.20; In like manner, SOC _min<SOC (t) <SOC _llinetime will produce descending out-of-limit punishment cost; F _aba(t), F _lack(t), F _outlinet () is respectively and abandons wind, stabilize not enough or out-of-limit running status Boolean quantity, and its expression formula is as shown in (13-15):

Described step 2) in ultra-short term refer to day to be that real time execution optimizes time scale.

Described step 2) in BESS synchronously participate in power fluctuation stabilize with Income Maximum during peak load regulation network for target build day level real-time optimization moving model be:

$Q_d=G-[\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{C}_{\mathrm{sti}}-\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{C\&prime;}_{\mathrm{sti}}]---\left(22\right)$

In formula, Q _dfor day level real time execution benefit; G is peak load regulation network income; C _sti(i=1,2,3,4)) be corresponding section operation cost; C' _sti(i=1,2,3,4) only participate in operating cost when wind power fluctuation is stabilized for BESS; On-road efficiency optimization turns to target with the maximum of Qd, P _p(t), P _v(t) and P _st () is the optimized variable under this target; Work as Q _dshow during >0 that participating in electrical network behavior has income, and Q _dlarger then income is more considerable; And when the risk cost added value that peak load regulation network income and power fluctuation are stabilized quite even is not enough to balance, Q _dto reduce gradually until be zero, Q _dshow that BESS has neither part nor lot in peak load regulation network when being zero.

Be that real time execution optimizes time scale with day, the market behavior determines regional peakload and low ebb duration of load application scope [t under guiding _pi, t _pl], [t _vi, t _vl], and electric discharge and charge power be respectively P _p(t), P _v(t), and its corresponding electricity price r _p, r _vif Δ t is the step-length of going forward one by one of t, i.e. the sampling interval, then peak load regulation network income is:

$\mathrm{vlG}=\underset{t=t_{\mathrm{pi}}}{\overset{t_{\mathrm{pl}}}{\mathrm{\&Sigma;}}}{r}_p\&CenterDot;P_p\left(t\right)\&CenterDot;\mathrm{\&Delta;t}-\underset{t=t_{\mathrm{vi}}}{\overset{t_{\mathrm{vl}}}{\mathrm{\&Sigma;}}}{r}_v\&CenterDot;P_v\left(t\right)\&CenterDot;\mathrm{\&Delta;t}---\left(17\right)$

Consider the discharge and recharge behavior in peak valley lotus period, day part discharge and recharge and moving model as follows:

T _pi≤ t≤t _plinterval:

$\left\{\begin{array}{c}{\mathrm{SOC}}_{\mathrm{st}1}\left(t\right)=\mathrm{SOC}\left(t_{\mathrm{pi}}\right)+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{pl}}}{\overset{t_{\mathrm{pl}}}{\mathrm{\&Sigma;}}}{P}_p\left(t\right)\&CenterDot;\mathrm{\&Delta;t}+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{pl}}}{\overset{t_{\mathrm{pl}}}{\mathrm{\&Sigma;}}}{P}_s\left(s\right)\&CenterDot;\mathrm{\&Delta;t}\\ C_{\mathrm{st}1}=C_{\mathrm{aba}}\left({\mathrm{SOC}}_{\mathrm{st}1}\right)+C_{\mathrm{lack}}\left(\mathrm{SO}{C}_{\mathrm{st}1}\right)+C_{\mathrm{outline}}\left({\mathrm{SOC}}_{\mathrm{st}1}\right)\end{array}\right.---\left(18\right)$

In formula, SOC _st1t () is t _pi≤ t≤t _plinterval SOC variable expression; SOC (t _pi) be the initial SOC numerical value in this interval, correspond to t _pithe numerical value in moment; P _st () stabilizes charge-discharge electric power for this interval power fluctuation; C _st1for this section operation cost calculation formula, wherein can obtain according to formula (10)-(12), for BESS charge-discharge characteristics parameter as P _max-cha, SOC _maxetc. under the prerequisite determined, this interval cost calculation formula is the function of SOC variable.This interval B ESS has peak load electric discharge peak regulation and the double operation function stabilized of power fluctuation.

T _pl<t<t _viinterval:

$\left\{\begin{array}{c}{\mathrm{SOC}}_{\mathrm{st}2}\left(t\right)=\mathrm{SOC}\left(t_{\mathrm{pl}}\right)+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{pl}}}{\overset{t_{\mathrm{vi}}}{\mathrm{\&Sigma;}}}{P}_s\left(t\right)\&CenterDot;\mathrm{\&Delta;t}\\ C_{\mathrm{st}2}=C_{\mathrm{aba}}\left({\mathrm{SOC}}_{\mathrm{st}2}\right)+C_{\mathrm{lack}}\left({\mathrm{SOC}}_{\mathrm{st}2}\right)+C_{\mathrm{outline}}\left({\mathrm{SOC}}_{\mathrm{st}2}\right)\end{array}\right.---\left(19\right)$

In formula, the SOC variable expression that SOCst2 (t) is this time interval; SOC (t _pl) be the initial SOC numerical value in this interval; C _st2for this section operation cost calculation formula.This interval is stabilized for the main operation action of BESS with power fluctuation.

T _vi≤ t≤t _vlinterval:

$\left\{\begin{array}{c}{\mathrm{SOC}}_{\mathrm{ST}3}\left(T\right)=\mathrm{SOC}\left(t_{\mathrm{vi}}\right)-\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{vl}}}{\overset{t_{\mathrm{vl}}}{\mathrm{\&Sigma;}}}{P}_v\left(t\right)\&CenterDot;\mathrm{\&Delta;t}+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{vl}}}{\overset{t_{\mathrm{vl}}}{\mathrm{\&Sigma;}}}{P}_s\left(t\right)\&CenterDot;\mathrm{\&Delta;t}\\ C_{\mathrm{st}3}=C_{\mathrm{aba}}\left({\mathrm{SOC}}_{\mathrm{st}3}\right)+C_{\mathrm{lack}}\left({\mathrm{SOC}}_{\mathrm{st}3}\right)+C_{\mathrm{outline}}\left({\mathrm{SOC}}_{\mathrm{st}3}\right)\end{array}\right.---\left(3\right)$

In formula, SOC _st3t SOC variable expression that () is this time interval; SOC (t _vi) be the initial SOC numerical value in this interval; C _st3for this section operation cost calculation formula.This interval B ESS has the double operation behavior that the paddy lotus phase charges and power fluctuation is stabilized.

T _vl<t<t _piinterval:

$\left\{\begin{array}{c}{\mathrm{SOC}}_{\mathrm{st}4}\left(t\right)=\mathrm{SOC}\left(t_{\mathrm{vl}}\right)+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{vl}}}{\overset{t_{\mathrm{pi}}}{\mathrm{\&Sigma;}}}{P}_s\left(t\right)\&CenterDot;\mathrm{\&Delta;t}\\ C_{\mathrm{st}4}=C_{\mathrm{aba}}\left({\mathrm{SOC}}_{\mathrm{st}4}\right)+C_{\mathrm{lack}}\left({\mathrm{SOC}}_{\mathrm{st}4}\right)+C_{\mathrm{outline}}\left({\mathrm{SOC}}_{\mathrm{st}4}\right)\end{array}\right.---\left(21\right)$

In formula, SOC _st4t SOC variable expression that () is this time interval; SOC (t _vl) be the initial SOC numerical value in this interval; C _st4for this section operation cost calculation formula.This interval is stabilized for the main operation action of BESS with power fluctuation.

The invention has the beneficial effects as follows the energy-storage system of the required configuration of planning regional wind power, optimize and the rational stored energy capacitance of computing economy, there is provided energy-storage system to assist the prioritization scheme participating in peak load regulation network, the reasonable Effec-tive Function level of lifting region wind field configuration energy-storage system simultaneously.

Accompanying drawing explanation

Fig. 1 (a) be in simulation example January typical case day stabilize power stage design sketch;

Fig. 1 (b) is the SOC curve chart of typical case's day January in simulation example;

Fig. 2 (a) be in simulation example July typical case day stabilize power stage design sketch;

Fig. 2 (b) is the SOC curve chart of typical case's day July in simulation example;

Fig. 3 (a) stabilizes design sketch after participating in peak regulation in simulation example;

Fig. 3 (b) is the SOC curve chart after participating in peak regulation in simulation example.

Embodiment

Below in conjunction with drawings and Examples, the present invention is further described.

The capacity optimization of regional wind power group energy-accumulating power station and an auxiliary peak regulating method thereof, comprising:

One, consider that wind farm group power fluctuation stabilizes the BESS complex function discharge and recharge implementation method with peak load regulation network, optimize computation model so that BESS operating cost is minimum for target builds capacity;

Wind farm group BESS stabilizes in power fluctuation process, as wind farm group power P (t) >P _reft, time (), BESS system is charged; P (t) <P _reftime (t), BESS system discharge; And as P (t)=P _reft time (), be then in floating charge state, wherein P (t) is wind farm group power, P _reft () is for expecting power output; When charge power, state-of-charge (SOC) or charge power rate of change are greater than BESS charge power rate of change maximum, then there will be and abandon wind; Shown in (1-3),

P(t)-P _ref(t)>P _max-cha（1）

SOC(t)>SOC _max（2）

ΔP _cha(t)>ΔP _max-cha（3）

In formula, P _max-chafor the maximum charge power of BESS; SOC _maxfor the up limit value of SOC, get 1.0; Δ P _chat ()=P (t)-P (t-1) is charge power rate of change; Δ P _max-chafor the charge power rate of change maximum of BESS;

Equally, when discharge power and rate of change thereof are greater than charge power rate of change maximum, or SOC lower than descending limit value time, shown in (4-6), then can occur stabilizing insufficient because stabilizing underpower;

｜P(t)-P _ref(t)｜>P _max-discha（4）

SOC(t)<SOC _min（5）

ΔP _discha(t)>ΔP _max-discha（6）

In formula, P _max-discha>0 is the maximum discharge power of BESS; SOC _minfor the descending limit value of SOC, get 0.1; Δ P _dischat () is discharge power rate of change; Δ P _max-dischafor the discharge power rate of change maximum of BESS.

Participate in peak load regulation network and run link, its reason is one, and the seasonal variation that wind-powered electricity generation cluster is exerted oneself makes its part-time interval have available idle capacity; Its two, its income comes from market environment paddy at lower night lotus phase rate for incorporation into the power network lower than electricity price of peak load.BESS participates in peak regulation can realize the storage of paddy lotus phase wind energy, has both been beneficial to the pressure reducing thermoelectricity degree of depth peak regulation, again and can be peak load and provide power supply to supply, simultaneously for BESS idle capacity provides income operational mode; The discharge and recharge behavior participating in peak regulation by according to BESS during paddy lotus and peak load can charge and discharge energy, invariable power stores or release within a certain period of time;

E _cha=P _cha·Δt _cha（7）

E _discha=P _discha·Δt _discha（8）

Wherein, E _cha, E _dischabe respectively the charging of participation peak regulation, discharge energy; P _cha, P _dischabe respectively charge-discharge electric power during peak regulation; Δ t _cha, Δ t _dischabe respectively the discharge and recharge duration.

BESS participates in peak load regulation network, because of its taking capacity, must abandon wind or stabilize under powered risk in lift portion charge and discharge interval when wind power fluctuation is stabilized, cause the rising of BESS operating cost.

With BESS operating cost minimum for target build capacity optimize computation model be:

minC=C _cons·k _T/k+C _aba+C _lack+C _outline；（16）

In formula, minC is the minimum cost run with BESS; K is that BESS service life amounts to a year number; k _tgot interval duration T by cost calculation and amounted to a year number; C _consfor construction cost; C _abafor abandoning eolian; C _lackfor desired output punishment cost; The out-of-limit punishment cost C of SOC _outline.

Set up and consider that the capacity planning model of operating cost need be taken into account BESS and build the system expenditure putting into operation, minimum for target function with its overall cost.Operating cost mainly comprises construction cost C _conswith operating cost C _run, wherein C _runto behavior cause and abandon wind, cause fluctuation shown in formula (4-6) and stabilize deficiency or the excessive charge and discharge of SOC and cause producing during out-of-limit influence on system operation BESS life-span, therefore C formula (1-3) Suo Shi _runcomprise and abandon eolian C _aba, desired output punishment cost C _lackpunishment cost C out-of-limit with SOC _outline.In general, C _conswith C _runmutual containing, the rated capacity V of obvious BESS affects C _conswith C _runthe key factor of balance and overall cost.Concrete account form is such as formula shown in (9-12):

C _cons=γ·V+ε·V（9）

$C_{\mathrm{aba}}=\mathrm{\&beta;}\&CenterDot;\{\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{aba}-1}\left(t\right)\&CenterDot;(P\left(t\right)-P_{\mathrm{ref}}\left(t\right)-{P\max -\mathrm{cha}}_{)}\mathrm{\&Delta;t}))$

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{aba}-2}\left(t\right)\&CenterDot;[(P\left(t\right)-P_{\mathrm{ref}}\left(t\right))\mathrm{\&Delta;t}-$ （10）

(SOC _max-SOC(t-1))V]

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{aba}-3}\left(t\right)\&CenterDot;(\mathrm{\&Delta;P}\left(t\right)-\mathrm{\&Delta;}{P}_{\max -\mathrm{cha}})\mathrm{\&Delta;t}\}$

$C_{\mathrm{lack}}=\mathrm{\&alpha;}\&CenterDot;\{\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{lack}-1}\left(t\right)\&CenterDot;\left(\right|P\left(t\right)-P_{\mathrm{ref}}\left(t\right)|-P_{\max -\mathrm{discha}})\mathrm{\&Delta;t}))$

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{lack}-2}\left(t\right)\&CenterDot;\left[\right|P\left(t\right)-P_{\mathrm{ref}}\left(t\right)|\mathrm{\&Delta;t}-$ （11）

(SOC(t-1)-SOC _min)V]

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{lack}-3}\left(t\right)\&CenterDot;\left(\right|\mathrm{\&Delta;P}\left(t\right)|-\mathrm{\&Delta;}{P}_{\max -\mathrm{discha}})\mathrm{\&Delta;t}\}$

$C_{\mathrm{outline}}=\mathrm{\&eta;}\&CenterDot;\{\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{outline}-1}\left(t\right)\&CenterDot;(\mathrm{SOC}\left(t\right)-{\mathrm{SOC}}_{\mathrm{Hline}})V$ （12）

$+\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}{F}_{\mathrm{outline}-2}\left(t\right)|\mathrm{SOC}\left(t\right)-{\mathrm{SOC}}_{\mathrm{Lline}}|\}$

In formula, γ is unit capacity batteries energy storage basic cost; V is the rated capacity of BESS; ε is BESS installation cost; T by the time window length of extraction service data; β be unit capacity abandon eolian; α causes the unit punishment cost of lacked capacity for stabilizing underpower; η be SOC out-of-limit run time unit capacity punishment cost; SOC (t-1) is corresponding previous moment SOC value; Δ t is the sampling interval; SOC _hline, SOC _llinebe respectively SOC and normally run upper lower limit value, work as SOC(t) >SOC _hlinetime, BESS still can charge to SOC _max, but its out-of-limit working capacity will produce out-of-limit punishment cost, SOC _hlinevalue is 0.85, SOC _llinevalue is 0.20; In like manner, SOC _min<SOC (t) <SOC _llinetime will produce descending out-of-limit punishment cost; F _aba(t), F _lack(t), F _outlinet () is respectively and abandons wind, stabilize not enough or out-of-limit running status Boolean quantity, and its expression formula is as shown in (13-15):

Two, utilize ultra-short term wind power prediction, synchronously participate in power fluctuation with BESS and stabilize with Income Maximum during peak load regulation network for target builds real-time optimization moving model;

Ultra-short term refers to day to be that real time execution optimizes time scale.

With BESS synchronously participate in power fluctuation stabilize with Income Maximum during peak load regulation network for target build day level real-time optimization moving model be:

$Q_d=G-[\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{C}_{\mathrm{sti}}-\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{C\&prime;}_{\mathrm{sti}}]---\left(22\right)$

In formula, Q _dfor day level real time execution benefit; G is peak load regulation network income; C _sti(i=1,2,3,4)) be corresponding section operation cost; C' _sti(i=1,2,3,4) only participate in operating cost when wind power fluctuation is stabilized for BESS; On-road efficiency is optimized with Q _dmaximumly turn to target, P _p(t), P _v(t) and P _st () is the optimized variable under this target; Work as Q _dshow during >0 that participating in electrical network behavior has income, and Q _dlarger then income is more considerable; And when the risk cost added value that peak load regulation network income and power fluctuation are stabilized quite even is not enough to balance, Q _dto reduce gradually until be zero, Q _dshow that BESS has neither part nor lot in peak load regulation network when being zero.

Run cost optimization determines the BESS rated capacity V of balanced basic cost and operating cost under being to ensure to stabilize effect prerequisite, but possesses the feasibility introducing peak regulating function in actual motion: one, considers the annual cycles that wind power exports, k _tgeneral value is 1; Its two, theoretical research shows, output of wind electric field has obvious seasonal variation, visits multiple wind field such as Zhucheng, Weihai further on the spot, and field data shows that each wind field all possesses this feature.Can obtain thus, V is as ensureing the annual average size value stabilizing effect, and in wind power output in certain large, season that fluctuating range is large, by being all used for, V ensures that effect is stabilized in fluctuation in season; And in wind power output little or mild season of exerting oneself, V will possess the volume space participating in peak regulation behavior.

Utilize wind power output calendar variation feature, with reference to ultra-short term power prediction data, abundant equilibrium utilizes BESS capacity, makes it possess participation fluctuation simultaneously and stabilizes the feasibility with peak load regulation network.

Peak load regulation network brings benefits on the one hand, on the other hand because of the use to capacity, must improve the risk that BESS operating cost increases.Therefore, the target of real time execution benefit optimization is that balance peak regulation income and operating cost increase risk, makes overall benefit optimum.Specifically, be that real time execution optimizes time scale with day, the market behavior determines regional peakload and low ebb duration of load application scope [t under guiding _pi, t _pl], [t _vi, t _vl], and electric discharge and charge power be respectively P _p(t), P _v(t), and its corresponding electricity price r _p, r _vif Δ t is the step-length of going forward one by one of t, i.e. the sampling interval, then peak load regulation network income is:

$G=\underset{t=t_{\mathrm{pi}}}{\overset{t_{\mathrm{pl}}}{\mathrm{\&Sigma;}}}{r}_p\&CenterDot;P_p\left(t\right)\&CenterDot;\mathrm{\&Delta;t}-\underset{t=t_{\mathrm{vi}}}{\overset{t_{\mathrm{vl}}}{\mathrm{\&Sigma;}}}{r}_v\&CenterDot;P_v\left(t\right)\&CenterDot;\mathrm{\&Delta;t}---\left(17\right)$

Consider the discharge and recharge behavior in peak valley lotus period, day part discharge and recharge and moving model as follows:

T _pi≤ t≤t _plinterval:

$\left\{\begin{array}{c}{\mathrm{SOC}}_{\mathrm{st}1}\left(t\right)=\mathrm{SOC}\left(t_{\mathrm{pi}}\right)+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{pl}}}{\overset{t_{\mathrm{pl}}}{\mathrm{\&Sigma;}}}{P}_p\left(t\right)\&CenterDot;\mathrm{\&Delta;t}+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{pl}}}{\overset{t_{\mathrm{pl}}}{\mathrm{\&Sigma;}}}{P}_s\left(s\right)\&CenterDot;\mathrm{\&Delta;t}\\ C_{\mathrm{st}1}=C_{\mathrm{aba}}\left({\mathrm{SOC}}_{\mathrm{st}1}\right)+C_{\mathrm{lack}}\left(\mathrm{SO}{C}_{\mathrm{st}1}\right)+C_{\mathrm{outline}}\left({\mathrm{SOC}}_{\mathrm{st}1}\right)\end{array}\right.---\left(18\right)$

In formula, SOC _st1t () is t _pi≤ t≤t _plinterval SOC variable expression; SOC (t _pi) be the initial SOC numerical value in this interval, correspond to t _pithe numerical value in moment; P _st () stabilizes charge-discharge electric power for this interval power fluctuation; C _st1for this section operation cost calculation formula, wherein can obtain according to formula (10)-(12), for BESS charge-discharge characteristics parameter as P _max-cha, SOC _maxetc. under the prerequisite determined, this interval cost calculation formula is the function of SOC variable.This interval B ESS has peak load electric discharge peak regulation and the double operation function stabilized of power fluctuation.

T _pl<t<t _viinterval:

$\left\{\begin{array}{c}{\mathrm{SOC}}_{\mathrm{st}2}\left(t\right)=\mathrm{SOC}\left(t_{\mathrm{pl}}\right)+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{pl}}}{\overset{t_{\mathrm{vi}}}{\mathrm{\&Sigma;}}}{P}_s\left(t\right)\&CenterDot;\mathrm{\&Delta;t}\\ C_{\mathrm{st}2}=C_{\mathrm{aba}}\left({\mathrm{SOC}}_{\mathrm{st}2}\right)+C_{\mathrm{lack}}\left({\mathrm{SOC}}_{\mathrm{st}2}\right)+C_{\mathrm{outline}}\left({\mathrm{SOC}}_{\mathrm{st}2}\right)\end{array}\right.---\left(19\right)$

In formula, SOC _st2t SOC variable expression that () is this time interval; SOC (t _pl) be the initial SOC numerical value in this interval; C _st2for this section operation cost calculation formula.This interval is stabilized for the main operation action of BESS with power fluctuation.

T _vi≤ t≤t _vlinterval:

$\left\{\begin{array}{c}{\mathrm{SOC}}_{\mathrm{ST}3}\left(T\right)=\mathrm{SOC}\left(t_{\mathrm{vi}}\right)-\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{vl}}}{\overset{t_{\mathrm{vl}}}{\mathrm{\&Sigma;}}}{P}_v\left(t\right)\&CenterDot;\mathrm{\&Delta;t}+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{vl}}}{\overset{t_{\mathrm{vl}}}{\mathrm{\&Sigma;}}}{P}_s\left(t\right)\&CenterDot;\mathrm{\&Delta;t}\\ C_{\mathrm{st}3}=C_{\mathrm{aba}}\left({\mathrm{SOC}}_{\mathrm{st}3}\right)+C_{\mathrm{lack}}\left({\mathrm{SOC}}_{\mathrm{st}3}\right)+C_{\mathrm{outline}}\left({\mathrm{SOC}}_{\mathrm{st}3}\right)\end{array}\right.---\left(3\right)$

In formula, SOC _st3t SOC variable expression that () is this time interval; SOC (t _vi) be the initial SOC numerical value in this interval; C _st3for this section operation cost calculation formula.This interval B ESS has the double operation behavior that the paddy lotus phase charges and power fluctuation is stabilized.

T _vl<t<t _piinterval:

$\left\{\begin{array}{c}{\mathrm{SOC}}_{\mathrm{st}4}\left(t\right)=\mathrm{SOC}\left(t_{\mathrm{vl}}\right)+\frac{1}{V}\&CenterDot;\underset{t=t_{\mathrm{vl}}}{\overset{t_{\mathrm{pi}}}{\mathrm{\&Sigma;}}}{P}_s\left(t\right)\&CenterDot;\mathrm{\&Delta;t}\\ C_{\mathrm{st}4}=C_{\mathrm{aba}}\left({\mathrm{SOC}}_{\mathrm{st}4}\right)+C_{\mathrm{lack}}\left({\mathrm{SOC}}_{\mathrm{st}4}\right)+C_{\mathrm{outline}}\left({\mathrm{SOC}}_{\mathrm{st}4}\right)\end{array}\right.---\left(21\right)$

In formula, SOC _st4t SOC variable expression that () is this time interval; SOC (t _vl) be the initial SOC numerical value in this interval; C _st4for this section operation cost calculation formula.This interval is stabilized for the main operation action of BESS with power fluctuation.

In real-time peaking operation benefit optimisation strategy implementation procedure, part issue-resolution is as follows:

1) area and calendar variation is considered, [t _pi, t _pl], [t _vi, t _vl] value difference, run actual in conjunction with Shandong Power, choose [19:00,22:00], [2:00,5:00] is the peak valley lotus period; Be simplification problem simultaneously, P _p(t), P _vt () is taken as definite value, i.e. firm power release or storage;

2) stand-by power supply such as diesel engine generator will become the important way making up wind power prediction deviation, therefore referenced wind power prediction think its error can obtain wind energy turbine set power autonomous to a certain degree make up digestion; Can introduce further in real time execution benefit optimizing process with the relative less ultra-short term wind power prediction of time error, assist the adjusting and optimizing in line computation;

3) P _p(t), P _vt () will be exerted oneself at [t to former wind farm group plan _pi, t _pl], [t _vi, t _vl] period has an impact, now according to real time execution optimum results, in pre-scheduled time level, (as 2 hours) report P to control centre in advance _p(t), P _vt () numerical value, for power adjustment in control centre's provides the reaction time.

Three, the cost sharing of BESS system cloud gray model and distribution of income pattern and method are proposed.

Cost sharing problem

These pool schemes blame interconnected principle, based on the potential benefited contrast run according to being benefited and carrying on a shoulder pole.For the C of operating cost _runpart, considers C _runthe uniformity that in production process, wind farm group is benefited and harmony, adopt equal offshoot program; For C _conspart, by each wind energy turbine set installed capacity pro rate, reason is as follows: 1) in wind farm group, each wind field geographical position is close, the conditional likelihood such as weather, landform, wind-resources distribution relative equilibrium, outside the plan such as removing maintenance stops, each wind energy turbine set actual power generation and installed capacity possess linear approximate relationship; 2) though wind speed and direction real-time change, consider that each wind field adjoins mutually, the fluctuation of each wind field has horizontal lag characteristic, and the difference in magnitude opposite sex that longitudinally fluctuates is same and installed capacity has certain relation.

Income Distribution Problem

Run income comprise potential income that power fluctuation stabilizes and participate in directly being benefited of peak load regulation network, potential income has generally been contained in the rate for incorporation into the power network agreement of electricity power enterprise and Utilities Electric Co., and this programme is only for direct benefited assignment problem.According to benefited property principle, bear income at most large.Income G comes from the lifting of paddy lotus phase storing electricity in peak load forward price value, and therefore G is according to [t _vi, t _vl] period each wind energy turbine set generating total amount ratio distribute.

Simulation example:

In visualstudio2010, c++ programming language is adopted to carry out modeling, by calling the optimal solution of outside solver CPLEX Solve problems.Choose region, Shandong wind field, comprise the wind energy turbine set that 3 geographical position are adjoined, its installed capacity is respectively 45MW, 40MW and 65MW, and grid-connected power samples is spaced apart 10min, with unit capacity (MWh) BESS basic cost for fiducial value, each parameter perunit value as shown in Table 1.

Form 1 parameter value table

BESS capacity planning

Choose the grid-connected power data of certain annual each wind field, what be up to that target determines region wind field and each dispersion wind field according to annual wind power utilization stabilizes target value, determines region wind field BESS and dispersion BESS calculation of capacity as shown in Table 2 based on stored energy capacitance planing method in literary composition.

As can be seen from form 2 result of calculation, compare the overall energy storage of 3 pieces of dispersion wind fields and drop into, region wind field BESS capacity requirement significantly reduces by 37.02%, and the space smoothing effect of region wind field effectively reduces configuration capacity needed for energy storage.Further analyzed area wind field BESS stabilizes effect and running status, chooses the typical day breeze power curve in January and July respectively, stabilizes effect and SOC respectively as Fig. 1, shown in 2.

Form 2 calculation of capacity result

Can find out, January, Shandong District wind field major part wind-resources was sufficient, and wind power numerical value is comparatively large, simultaneously with fluctuating by a relatively large margin.As can be seen from Fig. 1 (a), the stored energy capacitance determined herein can realize the output expecting power January substantially, simultaneously SOC fluctuation, and with out-of-limit process.Its reason is to choose year herein for the sample data cycle, and the optimum capacity obtained thus will ensure that the annual overall situation stabilizes effect, and for the different time cross-section of local wind's power fluctuation degree, it stabilizes effect also will have difference.

Further observation Fig. 2 can find out, due to the Shandong District abatement of wind in July, power stage is on the low side and fluctuation is less, and now utilize planning energy storage will realize desired output completely, SOC occurs without out-of-limit situation simultaneously, and remains on about 0.5.Be not difficult thus to draw, based on the calendar variation of wind power output, in the season that wind power output is more weak, BESS will possess stronger participation peak regulation feasibility.

Real-time peak regulation model-based optimization

According to real-time peaking operation benefit optimization method, for typical case's day breeze power output in January, if participate in peak regulation will aggravate the out-of-limit degree of SOC further, and the increase that now peak regulation income is not enough to offset BESS operating cost can be obtained as calculated, now BESS all participates in that power fluctuation stabilizes will be optimized operation state.

And for participating in peak regulation feasibility stronger July, the redundancy of its SOC will provide volume space for peak regulation.Choose and typical case's day participate in peak load regulation network, its income is shown in shown in form 3, stabilizes effect and SOC as shown in Figure 3.

Form 3 peak regulation income calculation result

Can be obtained by form 3, July region wind field BESS participates in peak regulation can obtain considerable income, and each wind field is made a profit with revenue relations separately in various degree by load duty simultaneously; Observe Fig. 3 (a) can find out, because peak load regulation network makes to stabilize the relative Fig. 2 (a) of effect weaken to some extent taking of capacity, but still effective output of the power that meets the expectation; And shown in Fig. 3 (b), SOC ' is not for participating in the SOC change curve of peak regulation, by relatively finding out, SOC is in run at high level after peak charging, add the up out-of-limit risk of SOC, be in low level and run after peak regulation electric discharge, descending out-of-limit risk increases, but under the mutual restriction between peak regulation income and operating cost increase, can optimal feasible solution be searched out; Each month in year risk return profile as shown in Table 4, meet the seasonal variation of wind power output as seen, method of the present invention can the considerable extra operation income of feasible region wind field BESS simultaneously.

Form 4 annual each month income statistics

To sum up can obtain, utilize the annual service data of each wind energy turbine set of region wind field, the optimum programming capacity of region wind field configuration BESS can be obtained; On this volumetric basis, utilize the calendar variation that wind field is exerted oneself, introduce BESS peak regulating function, real data shows that this operational mode brings considerable year's purchase by for BESS.

By reference to the accompanying drawings the specific embodiment of the present invention is described although above-mentioned; but not limiting the scope of the invention; one of ordinary skill in the art should be understood that; on the basis of technical scheme of the present invention, those skilled in the art do not need to pay various amendment or distortion that creative work can make still within protection scope of the present invention.

Claims (5)

1. the capacity optimization of regional wind power group energy-accumulating power station and an auxiliary peak regulating method thereof, is characterized in that, comprising:

1) consider that wind farm group power fluctuation stabilizes the BESS complex function discharge and recharge implementation method with peak load regulation network, optimize computation model so that BESS operating cost is minimum for target builds capacity;

2) utilize ultra-short term wind power prediction, synchronously participate in power fluctuation with BESS and stabilize with Income Maximum during peak load regulation network for target builds real-time optimization moving model;

3) cost sharing of BESS system cloud gray model and distribution of income pattern and method are proposed;

Step 1) in, wind farm group BESS stabilizes in power fluctuation process, and when wind farm group power P (t) >Pref (t), BESS system is charged; During P (t) <Pref (t), BESS system discharge; And as P (t)=Pref (t), be then in floating charge state, wherein P (t) is wind farm group power, and Pref (t) is for expecting power output; When charge power, state-of-charge and SOC or charge power rate of change are greater than BESS charge power rate of change maximum, then there will be and abandon wind; Shown in 1-formula 3,

P(t)-Pref(t)>Pmax-cha(1)

SOC(t)>SOCmax(2)

ΔPcha(t)>ΔPmax-cha(3)

In formula, Pmax-cha is the maximum charge power of BESS; SOCmax is the up limit value of SOC, gets 1.0; Δ Pcha (t)=P (t)-P (t-1) is charge power rate of change; Δ Pmax-cha is the charge power rate of change maximum of BESS;

Equally, when discharge power and rate of change thereof are greater than charge power rate of change maximum, or SOC lower than descending limit value time, shown in 4-formula 6, then can occur stabilizing insufficient because stabilizing underpower;

|P(t)-Pref(t)|>Pmax-discha(4)

SOC(t)<SOCmin(5)

ΔPdischa(t)>ΔPmax-discha(6)

In formula, Pmax-discha>0 is the maximum discharge power of BESS; SOCmin is the descending limit value of SOC, gets 0.1; Δ Pdischa (t) is discharge power rate of change; Δ Pmax-discha is the discharge power rate of change maximum of BESS.

2. the capacity optimization of regional wind power group energy-accumulating power station as claimed in claim 1 and auxiliary peak regulating method thereof, it is characterized in that, step 1) in, BESS participates in the storage that peak regulation can realize paddy lotus phase wind energy, both the pressure reducing thermoelectricity degree of depth peak regulation had been beneficial to, power supply is provided to supply, simultaneously for BESS idle capacity provides income operational mode again and for peak load; The discharge and recharge behavior participating in peak regulation by according to BESS during paddy lotus and peak load can charge and discharge energy, invariable power stores or release within a certain period of time;

E _discha＝P _discha·Δt _discha(8)

Wherein, E _cha, E _dischabe respectively the charging of participation peak regulation, discharge energy; P _cha, P _dischabe respectively charge-discharge electric power during peak regulation, Δ t _cha, Δ t _dischabe respectively the discharge and recharge duration.

3. the capacity optimization of regional wind power group energy-accumulating power station as claimed in claim 1 and auxiliary peak regulating method thereof, is characterized in that, described step 1) in minimumly with BESS operating cost for target builds capacity optimization computation model be:

minC＝C _cons·k _T/k+C _aba+C _lack+C _outline；

In formula, minC is the minimum cost run with BESS; K is that BESS service life amounts to a year number; k _tgot interval duration T by cost calculation and amounted to a year number; C _consfor construction cost; C _abafor abandoning eolian; C _lackfor desired output punishment cost; The out-of-limit punishment cost C of SOC _outline.

4. the capacity optimization of regional wind power group energy-accumulating power station as claimed in claim 1 and auxiliary peak regulating method thereof, is characterized in that, described step 2) in ultra-short term refer to day to be that real time execution optimizes time scale.

5. the capacity optimization of regional wind power group energy-accumulating power station as claimed in claim 1 and auxiliary peak regulating method thereof, it is characterized in that, described step 2) in BESS synchronously participate in power fluctuation stabilize with Income Maximum during peak load regulation network for target build day level real-time optimization moving model be:

$Q_d=G-\&lsqb;\underset{i=1}{\overset{4}{\&Sigma;}}{C}_{sti}-\underset{i=1}{\overset{4}{\&Sigma;}}{C}_{sti}^{\&prime;}\&rsqb;$

In formula, Q _dfor day level real time execution benefit; G is peak load regulation network income; C _stifor corresponding section operation cost; C ' _stifor BESS only participates in operating cost when wind power fluctuation is stabilized, wherein, i=1,2,3,4; On-road efficiency is optimized with Q _dmaximumly turn to target, P _p(t), P _v(t) and P _st () is the optimized variable under this target; Work as Q _dshow during >0 that participating in electrical network behavior has income, and Q _dlarger then income is more considerable; And when the risk cost added value that peak load regulation network income and power fluctuation are stabilized quite even is not enough to balance, Q _dto reduce gradually until be zero, Q _dshow that BESS has neither part nor lot in peak load regulation network when being zero.