CN102104251A - Microgrid real-time energy optimizing and scheduling method in parallel running mode - Google Patents

Abstract

The invention discloses a microgrid real-time energy optimizing and scheduling method in a parallel running mode. The method comprises the following steps of: firstly, dividing 24 hours in a day into a peak period of time, a general period of time and a trough period of time according to the load condition of a large power grid; and then, monitoring the working state of an accumulator in a microgrid in real time in the real-time process of the microgrid and adopting different energy optimizing strategies according to different periods of time and different working states of the accumulator so as to confirm the active power output of various schedulable micro power supplies, the charging and discharging power of the accumulator and scheduling instructions of the active power interactive with the power grid and the passive power of a passive adjustable power supply in the microgrid. The microgrid suitable for the invention comprises a reproducible energy generator, the schedulable micro power supplies and the accumulator. The invention not only can improve the economy and the reliability of the microgrid, but also can help the great power grid to cut peaks and fill troughs and is favorable to prolonging the service life of the accumulator.

Description

The energy-optimised dispatching method of little real-time power network under a kind of pattern that is incorporated into the power networks

Technical field

The present invention relates to a kind of little Electric Power Network Planning, operation and management, relate in particular to the energy-optimised dispatching method of little real-time power network under a kind of pattern that is incorporated into the power networks.

Background technology

The set that little electrical network is made up of various distributed power sources, energy-storage units, load and control protection system is by the cooperation between relevant control device, for load provides higher reliability and high-quality electric energy.Little electrical network replenishes because of factors such as its environmental friendliness, construction cost are low have become the useful of big electrical network, is more and more paid attention to and studies.Guaranteeing on the basis of little power network safety operation that the operation of the economic optimization of little electrical network is an important content in the energy management research.

At present, doing some researchs aspect the microgrid energy optimized dispatching both at home and abroad, in document [1]～[5] mainly from little operation of power networks cost minimum and environmental benefit set up energy-optimised model aspect best, by each active power that can dispatch little power supply of Optimization Model solving result scheduling, optimization is not but done in its reactive power output, and all do not consider little electrical network self networks structure, make that optimizing the result is short of reality, in addition, just there is not the cycle charging discharge of foundation for the use of the storage battery in little electrical network yet, accumulator cell charging and discharging is not optimized, thereby does not make full use of the economy that storage battery improves little operation of power networks.Pertinent literature:

[1] fourth is bright, Zhang Yingyuan, and the U.S. qin of cogongrass, poplar is Liu Xiaoping. the decline stable state modeling of network system of centralized control is optimized [J] with operation. Automation of Electric Systems, 2009,33 (24): 78-82.

[2] Chen Dawei, Zhu Guiping. little network load optimized distribution [J]. Automation of Electric Systems, 2010,34 (20): 45-48.

[3]Mohamed?FA，Koivo?HN.MicroGrid?Online?Management?andBalancing?Using?Multiobjective?Optimization.2007?IEEE?LAUSANNEPOWERTECH，VOLS?1-5，PP：639-644。

[4] ox inscription, Huang Wei, Guo Jiahuan, Su Ling. the Study on economical operation when little electrical network is incorporated into the power networks [J]. electric power network technique, 2010,34 (11): 39-42.

[5]Mohamed?F.and?H?Koivo.Online?Management?of?MicroGrid?with?BatteryStorage?Using?Multiobjective?Optimization，”the?first?Interna-tional?Conferenceon?Power?Engineering，Energy?and?Electrical?Drives(POWERENG07)，12-14?April2007，Setubal，Portugal.231-236。

Summary of the invention

Overcharge or over-discharge state in order to guarantee that little electrical network internal battery in use is not in, give full play to simultaneously the effect of little electrical network internal battery again, to improve little operation of power networks economy, and help big electrical network to carry out " peak load shifting ", be different from existing microgrid energy optimized dispatching method, the purpose of this invention is to provide the energy-optimised dispatching method of little real-time power network under a kind of pattern that is incorporated into the power networks.

The technical solution used in the present invention is to comprise the steps:

1) according to the load condition of big electrical network with being divided into peak period, section, three kinds of periods of paddy period at ordinary times in one day 24 hours, if big electrical network adopts the peak and valley time electricity price, then just according to the determined peak of the tou power price period, at ordinary times section and the paddy dividing time-steps;

2) in the real time execution process of little electrical network, determine the present located period constantly in each scheduling, monitor the state-of-charge SOC of the storage battery in little electrical network, measure the active power and the reactive power of each node load, measure the active power of output and the reactive power of renewable energy power generation;

3) if current be in the paddy period or at ordinary times the section, then further judge whether the state-of-charge SOC of current storage battery satisfies SOC＜SOC ^Max, be in the peak period as current, then further judge whether the state-of-charge of current storage battery satisfies SOC＞SOC ^Min

4) if current paddy period or the section at ordinary times of being in and do not satisfy SOC＜SOC ^Max, determine that then storage battery can discharge, and be optimized 1, satisfy SOC＜SOC ^Max, determine that then storage battery is not only chargeable but also can discharge, and be optimized 2; If currently be in the peak period, and satisfy SOC＞SOC ^Min, determine that then storage battery can discharge, and be optimized 3, do not satisfy SOC＞SOC ^Min, then determine with firm power charge in batteries, and be optimized 4;

5) obtain the active power and the reactive power output order of each little power supply by the optimization result in the step 4), then it is sent to each little power supply, each little power supply is according to instruction corresponding active power of output and reactive power.

Described step 2) scheduling in is meant that constantly the energy-optimised management of little real-time power network is in little real-time power network running, is a time interval little electrical network to be carried out the primary energy optimized dispatching with per 5 minutes or 15 minutes.

Storage battery state-of-charge SOC in the described step 3) is the parameter that reflection storage battery dump energy accounts for the ratio of its total capacity, is defined as:

$\mathrm{SOC}=\frac{C_{\mathrm{net}}}{C}=1-\frac{\∫\mathrm{Idt}}{C}$

C _Net-storage battery dump energy Ah;

C-storage battery total capacity, Ah;

I-battery discharging electric current, A;

Corresponding SOC ^Max, SOC ^MinFor the bound value of storage battery state-of-charge, get SOC ^Max=0.9, SOC ^Min=0.5.

Optimization 1 in the described step 4) is meant finds the solution the energy-optimised model of being set up 1; energy-optimised model 1 is minimum with little operation of power networks cost to be target; with in little electrical network each can dispatch the startup-shutdown state of little power supply, meritorious output that each can dispatch little power supply, with dealing electrical power, the discharge power of storage battery and the reactive power of each idle output scalable power supply between the big electrical network serve as the optimization variable; energy-optimised model 1 is the mixed integer nonlinear programming problem, its target function:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+{\mathrm{\λ}}_{\mathrm{bat}}+U_P^t{C}_P-U_S^t{I}_S$

In the formula: λ _Bat=(σ P _Bat) Δ t

The t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

Δ t-optimizes the time interval;

σ-designed coefficient;

P _BatThe discharge power of-storage battery;

Its constraints:

(a) trend constraints

$P_i-\underset{j=1}{\overset{j=n}{\mathrm{\Σ}}}[e_i(G_{\mathrm{ij}}{e}_j-B_{\mathrm{ij}}{f}_j)+f_i(G_{\mathrm{ij}}{f}_j+B_{\mathrm{ij}}{e}_j)]=0$

$Q_i-\underset{j=1}{\overset{j=n}{\mathrm{\Σ}}}[f_i(G_{\mathrm{ij}}{e}_j-B_{\mathrm{ij}}{f}_j)-e_i(G_{\mathrm{ij}}{f}_j+B_{\mathrm{ij}}{e}_j)]=0$

In the formula: P _i, Q _i-each node injects active power and reactive power;

e _i, f _iThe real part and the imaginary part of each node voltage of-usefulness complex representation;

C _Ij, B _IjThe real part and the imaginary part of-i node and j node admittance element;

Total node number in the little electrical network of n-.

(b) can dispatch little power supply capacity constraint

$P_i^{\min }\≤P_i^t\≤P_i^{\max }$

In the formula:

-Di i platform can be dispatched the active power output of little power supply in the t period;

-Di i platform can be dispatched the maximum active power output limit value and the minimum active power output limit value of little power supply;

(c) the discharge active power of storage battery constraint

$0\≤P_{\mathrm{bat}}\≤P_{\mathrm{dh}\_\max }^t$

In the formula: P _BatThe discharge power of-storage battery;

But-at the maximum discharge power limit value of t storage battery constantly the time;

(d) heap(ed) capacity constraint that can be mutual between little electrical network and big electrical network, the supply and demand agreement that this may be between them to be reached or the physical transfer capacity limit value of interconnection

$0\≤P_{\mathrm{Pgrid}}^t\≤P_{\mathrm{Pgrid}}^{\max }$

$0\≤P_{\mathrm{Sgrid}}^t\≤P_{\mathrm{Sgrid}}^{\max }$

In the formula:

-little electrical network is purchased the active power of electricity from big electrical network;

-little electrical network is to the active power of big electrical network sale of electricity;

-little electrical network is purchased the maximum active power limit value of electricity from big electrical network;

-little electrical network is to the maximum active power limit value of big electrical network sale of electricity;

(e) can dispatch the shortest continuous operating time and the constraint of the shortest continuous idle time of little power supply

$(T_{i,\mathrm{on}}^{t-1}-\mathrm{MRT})\·(U_i^{t-1}-U_i^t)\≥0$

$(T_{i,\mathrm{off}}^{t-1}-\mathrm{MST})\·(U_i^t-U_i^{t-1})\≥0$

In the formula:

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-t-1 constantly the i platform can dispatch the continuous operating time of little power supply;

-in the t-1 i platform continuous idle time that can dispatch little power supply constantly;

MRT-i platform can be dispatched the minimum continuous operating time of little power supply;

MST-i platform can be dispatched the minimum idle time continuously of little power supply;

(f) can dispatch little power supply active power output rate of change limit

$P_i^t-P_i^{t-1}\≤\mathrm{\Δ}{P}_U$

$P_i^{t-1}-P_i^t\≤\mathrm{\Δ}{P}_D$

In the formula:

-Di i platform can be dispatched the active power output of little power supply in the t period;

Δ P _U-can dispatch little this scheduling of power supply moment power with respect to last one restriction of power increase constantly;

Δ P _D-can dispatch little this scheduling of power supply moment power with respect to last one restriction of power reduction constantly;

(g) buy and sell electric mutual exclusion constraint from big electrical network

$U_P^t+U_S^t\≤1$

(h) the idle output constraint of reactive power output scalable power supply

$0\≤Q_i^t\≤Q_i^{\max }$

In the formula:

-Di i platform reactive power output scalable power supply is in t reactive power output constantly;

The maximum reactive power output limit value of-Di i platform reactive power output scalable power supply.

Optimization 2 in the described step 4) is meant finds the solution the energy-optimised model of being set up 2; energy-optimised model 2 is minimum with little operation of power networks cost to be target; so that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output that each can dispatch little power supply; and the dealing electrical power between the big electrical network; the reactive power of the discharge power of storage battery and each idle output scalable power supply is for optimizing variable; energy-optimised model 2 is the mixed integer nonlinear programming problem; its target function is identical with energy-optimised model 1; its constraints and energy-optimised model 1 are basic identical; just " (c) the discharge active power of the storage battery constraint " in the constraints changes, and " (c) the discharge active power of storage battery constraint " in the constraints of energy-optimised model 2 is:

$-P_{\mathrm{ch}\_\max }^t\≤P_{\mathrm{bat}}\≤P_{\mathrm{dh}\_\max }^t$

In the formula: P _BatThe discharge power of-storage battery; P _Bat＜0, represent to charge in batteries P _Bat＞0, expression is to battery discharging;

The maximum chargeable power of-storage battery;

But-storage battery maximum discharge power.

Optimization 3 in the described step 4) is meant finds the solution the energy-optimised model of being set up 3; energy-optimised model 3 is minimum with little operation of power networks cost to be target; so that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output that each can dispatch little power supply; and the dealing electrical power between the big electrical network; the reactive power of the discharge power of storage battery and each idle output scalable power supply is for optimizing variable; energy-optimised model 3 is the mixed integer nonlinear programming problem; its constraints is identical with energy-optimised model 1; its target function is different slightly with energy-optimised model 1, and difference is the λ in the target function of energy-optimised model 3 _BatWith the target function of energy-optimised model 1 in λ _BatItem designs differently, and the target function of energy-optimised model 3 is:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+{\mathrm{\λ}}_{\mathrm{bat}}+U_P^t{C}_P-U_S^t{I}_S$

In the formula: λ _Bat=δ P _DhΔ t

δ＝a ₁+a ₂·dSOC+a ₃·P _dh+a ₄·dSOC·P _dh+a ₅·dSOC ²

dSOC＝SOC-SOC ^min

The t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

Δ t-optimizes the time interval;

P _BatThe discharge power of-storage battery;

The state-of-charge of SOC-storage battery;

SOC ^MinThe state-of-charge lower limit of-storage battery;

a ₁, a ₂, a ₃, a ₄, a ₅-designed coefficient.

Optimization 4 in the described step 4) is meant finds the solution the energy-optimised model of being set up 4; energy-optimised model 4 is minimum with little operation of power networks cost to be target; with in little electrical network each can dispatch the startup-shutdown state of little power supply, meritorious output that each can dispatch little power supply, with dealing electrical power, the discharge power of storage battery and the reactive power of each idle output scalable power supply between the big electrical network serve as the optimization variable; energy-optimised model 4 is the mixed integer nonlinear programming problem; with respect to energy-optimised model 1, there is not λ in the target function of energy-optimised model 4 _Bat, the target function of energy-optimised model 4 is:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+U_P^t{C}_P-U_S^t{I}_S$

In the formula: the t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

The constraints of the constraints of energy-optimised model 4 and energy-optimised model 1 is basic identical, and being does not have " (c) the discharge active power of storage battery constraint " this item constraint in the constraints of energy-optimised model 4.

In the described step 4) with firm power to charge in batteries, and be optimized 4 and be meant that charging and discharging of accumulator power does not participate in optimizing in 4 optimizing, and with firm power to charge in batteries, this moment, storage battery was equivalent to a constant load, and will contribute to the firm power of charge in batteries in the load at little electrical network network node place at storage battery place and go, and be the constant charge power of storage battery

$P_{\mathrm{bat}}^t=I_{\mathrm{bat}\_\max \_\mathrm{ch}}\·U_{\mathrm{batt}}^t$

In the formula: I _{Bat_max_ch}The maximum charging current that allows of-storage battery, specification is given by producing;

-accumulator voltage.

Optimization result in the described step 5) is meant by solution procedure 4) in optimization 1, optimize 2, optimize 3 and optimize the result that 4 related energy-optimised model 1, energy-optimised model 2, energy-optimised model 3 and energy-optimised models 4 obtain, comprise in little electrical network each can dispatch the startup-shutdown state of little power supply, meritorious output valve that each can dispatch little power supply, with dealing electrical power value, the discharge power value of storage battery and the reactive power value of each idle output scalable power supply between the electrical network greatly.

The present invention compares with background technology, and the beneficial effect that has is:

1) traditional microgrid energy optimized dispatching is not optimized charging and discharging of accumulator, and the present invention is by counting in the target function of Optimization Model at the different period cost functions that discharge and recharge different to battery designs, the paddy that makes at the big electrical network little electrical network of section at ordinary times comes charge in batteries to buy electricity than low electricity price from big electrical network, and allow battery discharging and sell big electrical network with higher electricity price at little electrical network of peak period, both improved the on-road efficiency of little electrical network thus, and helped again big electrical network is carried out " peak load shifting ".

2) traditional microgrid energy optimized dispatching has only been optimized the active power output that respectively can dispatch little power supply, and do not consider via net loss, the present invention has considered little electrical network self networks structure, considered via net loss in feasible the optimization, and with the idle output scalable power supply (power supply that is connected with AC network by converters, the output of its reactive power also is adjustable) reactive power also as optimizing variable, make the active power output and the reactive power output of each power supply output reach total optimization, it is actual to optimize the result, makes little operation of power networks more economically.

3) the battery discharging cost function in the target function of peak energy-optimised model during the period is designed to the function of storage battery state-of-charge and discharge power, when the storage battery state-of-charge more little (but greater than storage battery state-of-charge lower limit SOC ^Min) time this cost function value big more, when discharge power is big more, this cost function value is big more, thereby the guiding storage battery hour reduces discharge power at state-of-charge, make storage battery keep certain charge capacity at any time, provide urgent power to support in the time of transferring lonely network operation to for little electrical network, improved little operation of power networks reliability, can make the state-of-charge of storage battery maintain SOC all the time again simultaneously ^MinAnd SOC ^MaxBetween, prevented overcharge of a battery or overdischarge, help prolonging the useful life of storage battery.

Description of drawings

Fig. 1 is the energy-optimised scheduling flow figure of little real-time power network.

Fig. 2 is the little electrical network exemplary plot of embodiment.

Fig. 3 adopts the present invention that little electrical network is carried out each power supply that real-time energy-optimised scheduling obtains at intraday active power distribution diagram.

Fig. 4 adopts the present invention that little electrical network is carried out each power supply that real-time energy-optimised scheduling obtains at intraday reactive power distribution diagram.

Fig. 5 adopts the present invention that little electrical network is carried out storage battery that real-time energy-optimised scheduling obtains at intraday state-of-charge SOC variation diagram.

Embodiment

The invention will be further described below in conjunction with drawings and Examples.

As shown in Figure 1, the present invention includes following steps:

1) according to the load condition of big electrical network with being divided into peak period, section, three kinds of periods of paddy period at ordinary times in one day 24 hours, if big electrical network adopts the peak and valley time electricity price, then just according to the determined peak of the tou power price period, at ordinary times section and the paddy dividing time-steps;

2) in the real time execution process of little electrical network, determine the present located period constantly in each scheduling, monitor the state-of-charge SOC of the storage battery in little electrical network, measure the active power and the reactive power of each node load, measure the active power of output and the reactive power of renewable energy power generation (photovoltaic generation and wind power generation);

3) if current be in the paddy period or at ordinary times the section, then further judge whether the state-of-charge SOC of current storage battery satisfies SOC＜SOC ^Max, be in the peak period as current, then further judge whether the state-of-charge of current storage battery satisfies SOC＞SOC ^Min

4) if current paddy period or the section at ordinary times of being in and do not satisfy SOC＜SOC ^Max, determine that then storage battery can discharge, and be optimized 1, satisfy SOC＜SOC ^Max, determine that then storage battery is not only chargeable but also can discharge, and be optimized 2; If currently be in the peak period, and satisfy SOC＞SOC ^Min, determine that then storage battery can discharge, and be optimized 3, do not satisfy SOC＞SOC ^Min, then determine with firm power charge in batteries, and be optimized 4;

5) obtain the active power and the reactive power output order of each little power supply by the optimization result in the step 4), then it is sent to each little power supply, each little power supply is according to instruction corresponding active power of output and reactive power.

Storage battery state-of-charge SOC in the described step 3) is the parameter that reflection storage battery dump energy accounts for the ratio of its total capacity, is defined as:

$\mathrm{SOC}=\frac{C_{\mathrm{net}}}{C}=1-\frac{\∫\mathrm{Idt}}{C}$

C _Net-storage battery dump energy Ah;

C-storage battery total capacity, Ah;

I-battery discharging electric current, A;

Corresponding SOC ^Max, SOC ^MinFor the bound value of storage battery state-of-charge, get SOC ^Max=0.9, SOC ^Min=0.5.Overcharge of a battery or overdischarge all can be shortened its useful life, by judging storage battery state-of-charge and SOC ^Max, SOC ^MinBetween relation determine next link, be in order to allow storage battery in charge and discharge process, remain its state-of-charge SOC at SOC ^MinWith SOC ^MaxBetween, to prevent overcharge of a battery or overdischarge.Charge in batteries to its state-of-charge reaches at 0.9 o'clock and will produce the gassing reflection, continues charging and will lose storage battery useful life, can make SOC for this reason ^Max=0.9; DOD is the battery technology parameter that producer provides, and is meant the depth of discharge of storage battery, but promptly storage battery maximum discharge capacity accounts for the ratio of its total capacity, is generally 0.6～0.7, and SOC can be arranged ^Min=1-DOD, but the little electrical network that is incorporated into the power networks will prevent to transfer to the situation of the autonomous operation of lonely net when big electric network fault, the little power supply that needs storage battery and other to participate in voltage-frequency control in this case supports for lonely net provides urgent power, therefore requirement storage battery when being incorporated into the power networks should be kept certain charge capacity at any time, so that storage battery can in time discharge when transferring lonely net to, remedy power shortage because of causing with big electrical breakdown, for this reason, with SOC ^MinSet greatlyyer slightly, get SOC than 1-DOD ^Min=0.5.

Optimization 1 in the described step 4) is meant finds the solution the energy-optimised model of being set up 1; energy-optimised model 1 is minimum with little operation of power networks cost to be target; so that each can dispatch little power supply (as miniature gas turbine in little electrical network; fuel cell; diesel engine generator etc.) startup-shutdown state; the meritorious output that each can dispatch little power supply; and the dealing electrical power between the big electrical network; the discharge power of storage battery and each idle output scalable power supply (power supply that is connected with AC network by converters; its reactive power that outputs to AC network also can be regulated; as big electrical network; fuel cell; miniature gas turbine; storage battery; photovoltaic cell; wind-driven generator etc.) reactive power is for optimizing variable; energy-optimised model 1 is the mixed integer nonlinear programming problem, its target function:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+{\mathrm{\λ}}_{\mathrm{bat}}+U_P^t{C}_P-U_S^t{I}_S$

In the formula: λ _Bat=(σ P _Bat) Δ t

The t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

Δ t-optimizes the time interval;

σ-designed coefficient;

P _BatThe discharge power of-storage battery;

Its constraints:

(a) trend constraints

$P_i-\underset{j=1}{\overset{j=n}{\mathrm{\Σ}}}[e_i(G_{\mathrm{ij}}{e}_j-B_{\mathrm{ij}}{f}_j)+f_i(G_{\mathrm{ij}}{f}_j+B_{\mathrm{ij}}{e}_j)]=0$

$Q_i-\underset{j=1}{\overset{j=n}{\mathrm{\Σ}}}[f_i(G_{\mathrm{ij}}{e}_j-B_{\mathrm{ij}}{f}_j)-e_i(G_{\mathrm{ij}}{f}_j+B_{\mathrm{ij}}{e}_j)]=0$

In the formula: P _i, Q _i-each node injects active power and reactive power;

e _i, f _iThe real part and the imaginary part of each node voltage of-usefulness complex representation;

G _Ij, B _IjThe real part and the imaginary part of-i node and j node admittance element;

Total node number in the little electrical network of n-.

(b) can dispatch little power supply capacity constraint

$P_i^{\min }\≤P_i^t\≤P_i^{\max }$

In the formula:

-Di i platform can be dispatched the active power output of little power supply in the t period;

-Di i platform can be dispatched the maximum active power output limit value and the minimum active power output limit value of little power supply;

(c) the discharge active power of storage battery constraint

$0\≤P_{\mathrm{bat}}\≤P_{\mathrm{dh}\_\max }^t$

In the formula: $P_{\mathrm{dh}\_\max }^t=\min (\frac{(\mathrm{SO}{C}_t-\mathrm{SO}{C}^{\min })\·C_{\mathrm{batt}}^{\′}}{\mathrm{\Δt}},I_{\mathrm{bat}\_\max \_\mathrm{dh}})\·U_{\mathrm{batt}}^t$

-allow the discharge power value in the maximum of t storage battery during the moment;

SOC _tThe state-of-charge of the storage battery that-scheduling monitors constantly;

SOC ^MinThe storage battery state-of-charge lower limit of-setting;

C ' _Batt-battery capacity;

I _{Bat_max_dh}The maximum discharging current that allows of-storage battery, specification is given by producing;

-accumulator voltage;

Δ t-scheduling time interval;

P _BatThe discharge power of-storage battery;

(d) heap(ed) capacity constraint that can be mutual between little electrical network and big electrical network, the supply and demand agreement that this may be between them to be reached or the physical transfer capacity limit value of interconnection

$0\≤P_{\mathrm{Pgrid}}^t\≤P_{\mathrm{Pgrid}}^{\max }$

$0\≤P_{\mathrm{Sgrid}}^t\≤P_{\mathrm{Sgrid}}^{\max }$

In the formula:

-little electrical network is purchased the active power of electricity from big electrical network;

-little electrical network is to the active power of big electrical network sale of electricity;

-little electrical network is purchased the maximum active power limit value of electricity from big electrical network;

-little electrical network is to the maximum active power limit value of big electrical network sale of electricity;

(e) can dispatch the shortest continuous operating time and the constraint of the shortest continuous idle time of little power supply

$(T_{i,\mathrm{on}}^{t-1}-\mathrm{MRT})\·(U_i^{t-1}-U_i^t)\≥0$

$(T_{i,\mathrm{off}}^{t-1}-\mathrm{MST})\·(U_i^t-U_i^{t-1})\≥0$

In the formula: -t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-t-1 constantly the i platform can dispatch the continuous operating time of little power supply;

-in the t-1 i platform continuous idle time that can dispatch little power supply constantly;

MRT-i platform can be dispatched the minimum continuous operating time of little power supply;

MST-i platform can be dispatched the minimum idle time continuously of little power supply;

(f) can dispatch little power supply active power output rate of change limit

$P_i^t-P_i^{t-1}\≤\mathrm{\Δ}{P}_U$

$P_i^{t-1}-P_i^t\≤\mathrm{\Δ}{P}_D$

In the formula:

-Di i platform can be dispatched the active power output of little power supply in the t period;

Δ P _U-can dispatch little this scheduling of power supply moment power with respect to last one restriction of power increase constantly;

Δ P _D-can dispatch little this scheduling of power supply moment power with respect to last one restriction of power reduction constantly;

(g) buy and sell electric mutual exclusion constraint from big electrical network

$U_P^t+U_S^t\≤1$

(h) the idle output constraint of reactive power output scalable power supply

$0\≤Q_i^t\≤Q_i^{\max }$

In the formula:

-Di i platform reactive power output scalable power supply is in t reactive power output constantly;

The maximum reactive power output limit value of-Di i platform reactive power output scalable power supply.

Optimization 2 in the described step 4) is meant finds the solution the energy-optimised model of being set up 2; energy-optimised model 2 is minimum with little operation of power networks cost to be target; so that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output that each can dispatch little power supply; and the dealing electrical power between the big electrical network; the reactive power of the discharge power of storage battery and each idle output scalable power supply is for optimizing variable; energy-optimised model 2 is the mixed integer nonlinear programming problem; its target function is identical with energy-optimised model 1; its constraints and energy-optimised model 1 are basic identical; just " (c) the discharge active power of the storage battery constraint " in the constraints changes, and " (c) the discharge active power of storage battery constraint " in the constraints of energy-optimised model 2 is:

$-P_{\mathrm{ch}\_\max }^t\≤P_{\mathrm{bat}}\≤P_{\mathrm{dh}\_\max }^t$

In the formula: $P_{\mathrm{ch}\_\max }^t=\min (\frac{({\mathrm{SOC}}^{\max }-{\mathrm{SOC}}_t)\·C_{\mathrm{batt}}^{\′}}{\mathrm{\Δt}},I_{\mathrm{bat}\_\max \_\mathrm{ch}})\·U_{\mathrm{batt}}^t$

$P_{\mathrm{dh}\_\max }^t=\max (0,\min (\frac{({\mathrm{SOC}}_t-{\mathrm{SOC}}^{\min })\·C_{\mathrm{batt}}^{\′}}{\mathrm{\Δt}},I_{\mathrm{bat}\_\max \_\mathrm{dh}}))\·U_{\mathrm{batt}}^t$

P _BatThe discharge power of-storage battery; P _Bat＜0, represent to charge in batteries P _Bat＞0, expression is to battery discharging;

The maximum of-storage battery allows charge power;

The maximum of-storage battery allows discharge power;

SOC _tThe state-of-charge of the storage battery that-scheduling monitors constantly;

SOC ^MaxThe storage battery state-of-charge upper limit of-setting;

SOC ^MinThe storage battery state-of-charge lower limit of-setting;

C ' _Batt-battery capacity;

I _{Bat_max_dh}The maximum discharging current that allows of-storage battery, specification is given by producing;

I _{Bat_max_ch}The maximum charging current that allows of-storage battery, specification is given by producing;

-accumulator voltage;

Δ t-scheduling time interval.

Optimization 3 in the described step 4) is meant finds the solution the energy-optimised model of being set up 3; energy-optimised model 3 is minimum with little operation of power networks cost to be target; so that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output that each can dispatch little power supply; and the dealing electrical power between the big electrical network; the reactive power of the discharge power of storage battery and each idle output scalable power supply is for optimizing variable; energy-optimised model 3 is the mixed integer nonlinear programming problem; its constraints is identical with energy-optimised model 1; its target function is different slightly with energy-optimised model 1, and difference is the λ in the target function of energy-optimised model 3 _BatWith the target function of energy-optimised model 1 in λ _BatItem designs differently, and the target function of energy-optimised model 3 is:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+{\mathrm{\λ}}_{\mathrm{bat}}+U_P^t{C}_P-U_S^t{I}_S$

In the formula: λ _Bat=δ P _DhΔ t

δ＝a ₁+a ₂·dSOC+a ₃·P _dh+a ₄·dSOC·P _dh+a ₅·dSOC ²

dSOC＝SOC-SOC ^min

The t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

Δ t-optimizes the time interval;

P _BatThe discharge power of-storage battery;

The state-of-charge of SOC-storage battery;

SOC ^MinThe state-of-charge lower limit of-storage battery;

a ₁, a ₂, a ₃, a ₄, a ₅-designed coefficient,

Optimize 3 and be in the peak period, mainly be in discharge condition at this period storage battery, consider that the little electrical network that is incorporated into the power networks will prevent to transfer to the situation of the autonomous operation of lonely net when big electric network fault, the little power supply that needs storage battery and other to participate in voltage-frequency control in this case provides urgent support for lonely net, therefore requirement storage battery when being incorporated into the power networks should be kept certain charge capacity at any time, so wishing storage battery can discharge when carrying capacity is sufficient more, and when carrying capacity discharge less, accumulator cell charging and discharging cost function in the designed target function that counts energy-optimised model 3 has following characteristic: when state-of-charge is more little, its functional value is big more; When power output is big more, its functional value is big more, again because optimization aim is to make the target function value minimum, therefore adopt this function representation battery discharging cost function will guide storage battery at state-of-charge than hour reducing discharge power, make storage battery can keep certain charge capacity, provide urgent power to support in the time of transferring lonely network operation to for little electrical network, improved little operation of power networks reliability.

Optimization 4 in the described step 4) is meant finds the solution the energy-optimised model of being set up 4; energy-optimised model 4 is minimum with little operation of power networks cost to be target; with in little electrical network each can dispatch the startup-shutdown state of little power supply, meritorious output that each can dispatch little power supply, with dealing electrical power, the discharge power of storage battery and the reactive power of each idle output scalable power supply between the big electrical network serve as the optimization variable; energy-optimised model 4 is the mixed integer nonlinear programming problem; with respect to energy-optimised model 1, there is not λ in the target function of energy-optimised model 4 _Bat, the target function of energy-optimised model 4 is:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+U_P^t{C}_P-U_S^t{I}_S$

In the formula: the t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

The constraints of the constraints of energy-optimised model 4 and energy-optimised model 1 is basic identical, and being does not have " (c) the discharge active power of storage battery constraint " this item constraint in the constraints of energy-optimised model 4.

In the described step 4) with firm power to charge in batteries, and be optimized 4 and be meant that charging and discharging of accumulator power does not participate in optimizing in 4 optimizing, and with firm power to charge in batteries, this moment, storage battery was equivalent to a constant load, and will contribute to the firm power of charge in batteries in the load at little electrical network network node place at storage battery place and go, and be the constant charge power of storage battery

$P_{\mathrm{bat}}^t=I_{\mathrm{bat}\_\max \_\mathrm{ch}}\·U_{\mathrm{batt}}^t$

In the formula: I _{Bat_max_ch}The maximum charging current that allows of-storage battery, specification is given by producing;

-accumulator voltage.

Optimization result in the described step 5) is meant by solution procedure 4) in optimization 1, optimize 2, optimize 3 and optimize the result that 4 related energy-optimised model 1, energy-optimised model 2, energy-optimised model 3 and energy-optimised models 4 obtain, comprise in little electrical network each can dispatch the startup-shutdown state of little power supply, meritorious output valve that each can dispatch little power supply, with dealing electrical power value, the discharge power value of storage battery and the reactive power value of each idle output scalable power supply between the electrical network greatly.

Below in conjunction with accompanying drawing, embodiments of the invention are elaborated, the flow chart of this invention is as shown in Figure 1.

Embodiment:

Consider little electrical network as shown in Figure 2, little electrical network includes diesel engine generator, miniature gas turbine, fuel cell, wind-driven generator, photovoltaic cell and storage battery, and little electrical network is connected with electrical network greatly by point of common coupling (PCC) and is incorporated into the power networks.Suppose with per 15 minutes be a scheduling slot, the current time is 13: 30, adopts the present invention that little electrical network is carried out real-time energy-optimised scheduling, each step division is as follows:

1) will be divided into peak period, section and paddy period at ordinary times in one day 24 hours: the paddy period is 00:00-07:00, and section is 08:00-10:00,16:00-18:00 and 22:00-23:00 at ordinary times, and the peak period is 11:00-15:00 and 19:00-21:00.

2) in the real time execution process of little electrical network, determine the present located period constantly in each scheduling, the state-of-charge SOC of monitoring storage battery, measure the active power and the reactive power of each node load, measure the active power of output and the reactive power of renewable energy power generation (photovoltaic generation and wind power generation); Current time is 13: 30, according to the period of being divided as can be known, currently be in the peak period, suppose the state-of-charge SOC=0.67 of the storage battery that records.

3) currently be in the peak period, judge further whether the state-of-charge of current storage battery satisfies SOC＞SOC ^Min, because SOC ^Min=0.5, so satisfy SOC＞SOC ^Min

4) currently be in the peak period, and satisfy SOC＞SOC ^MinThen be optimized 3; its energy-optimised model 3 of setting up is minimum with little operation of power networks cost to be target; so that each can dispatch little power supply (miniature gas turbine in little electrical network; fuel cell and diesel engine generator) the startup-shutdown state; each can dispatch little power supply (miniature gas turbine; fuel cell; diesel engine generator) meritorious output; and the dealing electrical power between the big electrical network; the discharge power of storage battery and each idle output scalable power supply (power supply that is connected with AC network by converters; its reactive power that outputs to AC network also can be regulated; present embodiment comprises big electrical network; fuel cell; miniature gas turbine; storage battery and wind-driven generator) reactive power for optimizing variable; energy-optimised model 3 is the mixed integer nonlinear programming problem, and its energy-optimised model of setting up 3 is:

Target function:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+{\mathrm{\λ}}_{\mathrm{bat}}+U_P^t{C}_P-U_S^t{I}_S$

In the formula: $C_{\mathrm{OM}}\left(P_i^t\right)=K_{\mathrm{OM}}{P}_i^t\·\mathrm{\Δt}$

$C_P\left(P_{\mathrm{Pgrid}}^t\right)=c_p^t{P}_{\mathrm{Pgrid}}^t\·\mathrm{\Δt}$

$I_S\left(P_{\mathrm{Sgrid}}^t\right)=c_s^t{P}_{\mathrm{Sgrid}}^t\·\mathrm{\Δt}$

The t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

Δ t-optimizes the time interval;

K _OMi-can dispatch the proportionality coefficient of little power supply operation expense;

-little electrical network is purchased the active power of electricity from big electrical network;

-little electrical network is to the active power of big electrical network sale of electricity;

-little electrical network is purchased electric electricity price, unit/kilowatt hour from big electrical network;

-little electrical network is to the electricity price of big electrical network sale of electricity, unit/kilowatt hour;

For diesel engine generator (DG, Diesel Generator), its energy consumption cost is the function of its active power output, can be expressed as follows with quadratic polynomial:

$C_{f\_\mathrm{DG}}\left(P_i^t\right)=(a{\left(P_i^t\right)}^2+b{P}_i^t+c)\mathrm{\Δt}$

In the formula: a, b, c-quadratic polynomial coefficient correlation.

For fuel cell (FC, Fuel Cell) and micro combustion steam turbine (MT, Microturbine), its active power output is proportional with the fuel input variable, corresponding energy consumption cost can be expressed as:

$C_{f\_\mathrm{FC}}\left(P_i^t\right)=\left(c_{\mathrm{ng}}\frac{P_i^t}{{\mathrm{\η}}_i^t}\right)\mathrm{\Δt}$

In the formula:

-meritorious power output;

c _NgFor-the fluid fuel price;

-fuel utilization ratio is the ratio of the calorific value of output electric energy and input fuel;

The battery discharging cost function is designed to:

dSOC＝SOC-SOC ^min

δ＝a ₁+a ₂·dSOC+a ₃·P _dh+a ₄·dSOC·P _dh+a ₅·dSOC ²

λ _bat＝δ·P _dh

In the formula: P _BatThe discharge power of-storage battery;

The state-of-charge of SOC-storage battery;

SOC ^MinThe state-of-charge lower limit of-storage battery;

a ₁, a ₂, a ₃, a ₄, a ₅-designed coefficient;

Constraints:

(a) trend constraints

$P_i-\underset{j=1}{\overset{j=n}{\mathrm{\Σ}}}[e_i(G_{\mathrm{ij}}{e}_j-B_{\mathrm{ij}}{f}_j)+f_i(G_{\mathrm{ij}}{f}_j+B_{\mathrm{ij}}{e}_j)]=0$

$Q_i-\underset{j=1}{\overset{j=n}{\mathrm{\Σ}}}[f_i(G_{\mathrm{ij}}{e}_j-B_{\mathrm{ij}}{f}_j)-e_i(G_{\mathrm{ij}}{f}_j+B_{\mathrm{ij}}{e}_j)]=0$

In the formula: P _i, Q _i-each node injects active power and reactive power;

e _i, f _iThe real part and the imaginary part of each node voltage of-usefulness complex representation;

G _Ij, B _IjThe real part and the imaginary part of-i node and j node admittance element;

Total node number in the little electrical network of n-.

(b) can dispatch little power supply capacity constraint

$P_i^{\min }\≤P_i^t\≤P_i^{\max }$

In the formula:

-Di i platform can be dispatched the active power output of little power supply in the t period;

-Di i platform can be dispatched the maximum active power output limit value and the minimum active power output limit value of little power supply;

(c) the discharge active power of storage battery constraint

$0\≤P_{\mathrm{bat}}\≤P_{\mathrm{dh}\_\max }^t$

In the formula: $P_{\mathrm{dh}\_\max }^t=\min (\frac{(\mathrm{SO}{C}_t-\mathrm{SO}{C}^{\min })\·C_{\mathrm{batt}}^{\′}}{\mathrm{\Δt}},I_{\mathrm{bat}\_\max \_\mathrm{dh}})\·U_{\mathrm{batt}}^t$

-allow the discharge power value in the maximum of t storage battery during the moment;

SOG _tThe state-of-charge of the storage battery that-scheduling monitors constantly;

SOC ^MinThe storage battery state-of-charge lower limit of-setting;

C ' _Batt-battery capacity;

I _{Bat_max_dh}The maximum discharging current that allows of-storage battery, specification is given by producing;

-accumulator voltage;

Δ t-scheduling time interval;

P _BatThe discharge power of-storage battery;

(d) heap(ed) capacity constraint that can be mutual between little electrical network and big electrical network, the supply and demand agreement that this may be between them to be reached or the physical transfer capacity limit value of interconnection

$0\≤P_{\mathrm{Pgrid}}^t\≤P_{\mathrm{Pgrid}}^{\max }$

$0\≤P_{\mathrm{Sgrid}}^t\≤P_{\mathrm{Sgrid}}^{\max }$

In the formula:

-little electrical network is purchased the active power of electricity from big electrical network;

-little electrical network is to the active power of big electrical network sale of electricity;

-little electrical network is purchased the maximum active power limit value of electricity from big electrical network;

-little electrical network is to the maximum active power limit value of big electrical network sale of electricity;

(e) can dispatch the shortest continuous operating time and the constraint of the shortest continuous idle time of little power supply

$(T_{i,\mathrm{on}}^{t-1}-\mathrm{MRT})\·(U_i^{t-1}-U_i^t)\≥0$

$(T_{i,\mathrm{off}}^{t-1}-\mathrm{MST})\·(U_i^t-U_i^{t-1})\≥0$

In the formula:

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-t-1 constantly the i platform can dispatch the continuous operating time of little power supply;

-in the t-1 i platform continuous idle time that can dispatch little power supply constantly;

MRT-i platform can be dispatched the minimum continuous operating time of little power supply;

MST-i platform can be dispatched the minimum idle time continuously of little power supply;

(f) can dispatch little power supply active power output rate of change limit

$P_i^t-P_i^{t-1}\≤\mathrm{\Δ}{P}_U$

$P_i^{t-1}-P_i^t\≤\mathrm{\Δ}{P}_D$

In the formula: -Di i platform can be dispatched the active power output of little power supply in the t period;

Δ P _U-can dispatch little this scheduling of power supply moment power with respect to last one restriction of power increase constantly;

Δ P _D-can dispatch little this scheduling of power supply moment power with respect to last one restriction of power reduction constantly;

(g) buy and sell electric mutual exclusion constraint from big electrical network

$U_P^t+U_S^t\≤1$

(h) the idle output constraint of reactive power output scalable power supply

$0\≤Q_i^t\≤Q_i^{\max }$

In the formula:

-Di i platform reactive power output scalable power supply is in t reactive power output constantly;

The maximum reactive power output limit value of-Di i platform reactive power output scalable power supply;

5) by finding the solution active power and the reactive power output order that above-mentioned energy-optimised model 3 obtains each little power supply, then it is sent to each little power supply, each little power supply is according to instruction corresponding active power of output and reactive power.

Fig. 3 for adopt the present invention carry out each power supply that the energy-optimised scheduling of little real-time power network obtains in one day each constantly active power (P_Grid for and the mutual active power of big electrical network, its value is purchased electricity greater than 0 expression from big electrical network, represents to big electrical network sale of electricity less than 0; P_MT is the active power of miniature gas turbine generating output; P_FC is the active power of fuel cell power generation output; P_DE is the active power of diesel engine generator generating output; P_Bat is a charging and discharging of accumulator power, its value is greater than 0 expression battery discharging, less than 0 expression to charge in batteries), Fig. 4 for adopt the present invention carry out each idle scalable power supply that the energy-optimised scheduling of little real-time power network obtains in one day each reactive power output constantly (Q_Grid is the reactive power of big electrical network to little electrical network output; Q_MT is the reactive power of miniature gas turbine generating output; Q_FC is the reactive power of fuel cell power generation output; Q_Bat is the reactive power of storage battery output; Q_MT is the reactive power of wind turbine power generation output), Fig. 5 is that the storage battery state-of-charge is in intraday situation of change.By Fig. 3 and Fig. 5 as can be known the discharge power P_Bat of storage battery at 0～4h (paddy period), 16～18h (section at ordinary times) and 22～23h (section at ordinary times) are less than 0 (charge in batteries), and at 11～16h (peak period) and 19～22h (peak period) greater than 0 (battery discharging), and with the mutual power P _ Grid of big electrical network in 0～10h (paddy, section at ordinary times) and 16～18h (at ordinary times section) greater than 0 (purchasing) from big electrical network, and at 11～15h (peak period) and 19～21h (peak period) less than 0 (to big electrical network sale of electricity), little like this electrical network is bought electricity to charge in batteries with low electricity price from big electrical network in Pinggu period, and sell electrical network in the battery discharging of peak period and with higher price, both improved the economy of little operation of power networks, and helped again big electrical network is carried out " peak load shifting ".The storage battery state-of-charge maintains 0.5 (SOC all the time as shown in Figure 5 ^Min) and 0.9 (SOC ^Max) between, battery discharging simultaneously is more little near 0.5 o'clock its discharge power more to its state-of-charge, illustrate that the microgrid energy management of adopting the present invention to carry out allows storage battery operate in the safe range all the time, overcharge of a battery or overdischarge have been prevented, and storage battery has kept certain charge capacity (because know that from Fig. 5 SOC is all the time more than 0.6 at any time, bigger than 0.5), can provide urgent power to support when little electrical network transfers lonely network operation under big electrical network breaks down situation.

Claims (9)

1. the energy-optimised dispatching method of little real-time power network under the pattern that is incorporated into the power networks is characterized in that comprising the steps:

1) according to the load condition of big electrical network with being divided into peak period, section, three kinds of periods of paddy period at ordinary times in one day 24 hours, if big electrical network adopts the peak and valley time electricity price, then just according to the determined peak of the tou power price period, at ordinary times section and the paddy dividing time-steps;

2) in the real time execution process of little electrical network, determine the present located period constantly in each scheduling, monitor the state-of-charge SOC of the storage battery in little electrical network, measure the active power and the reactive power of each node load, measure the active power of output and the reactive power of renewable energy power generation;

3) if current be in the paddy period or at ordinary times the section, then further judge whether the state-of-charge SOC of current storage battery satisfies SOC＜SOC ^Max, be in the peak period as current, then further judge whether the state-of-charge of current storage battery satisfies SOC＞SOC ^Min

4) if current paddy period or the section at ordinary times of being in and do not satisfy SOC＜SOC ^Max, determine that then storage battery can discharge, and be optimized 1, satisfy SOC＜SOC ^Max, determine that then storage battery is not only chargeable but also can discharge, and be optimized 2; If currently be in the peak period, and satisfy SOC＞SOC ^Min, determine that then storage battery can discharge, and be optimized 3, do not satisfy SOC＞SOC ^Min, then determine with firm power charge in batteries, and be optimized 4;

5) obtain the active power and the reactive power output order of each little power supply by the optimization result in the step 4), then it is sent to each little power supply, each little power supply is according to instruction corresponding active power of output and reactive power.

2. the little real-time power network energy optimizing method under a kind of pattern that is incorporated into the power networks according to claim 1, it is characterized in that: scheduling described step 2) is meant that constantly the energy-optimised management of little real-time power network is in little real-time power network running, is a time interval little electrical network to be carried out the primary energy optimized dispatching with per 5 minutes or 15 minutes.

3. the little real-time power network energy optimizing method under a kind of pattern that is incorporated into the power networks according to claim 1 is characterized in that: the storage battery state-of-charge SOC in the described step 3) is the parameter that reflection storage battery dump energy accounts for the ratio of its total capacity, is defined as:

$\mathrm{SOC}=\frac{C_{\mathrm{net}}}{C}=1-\frac{\∫\mathrm{Idt}}{C}$

C _Net-storage battery dump energy Ah;

C-storage battery total capacity, Ah;

I-battery discharging electric current, A;

Corresponding SOC ^Max, SOC ^MinFor the bound value of storage battery state-of-charge, get SOC ^Max=0.9, SOC ^Min=0.5.

4. the energy-optimised dispatching method of little real-time power network under a kind of pattern that is incorporated into the power networks according to claim 1; it is characterized in that: the optimization 1 in the described step 4) is meant finds the solution the energy-optimised model of being set up 1; energy-optimised model 1 is minimum with little operation of power networks cost to be target; so that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output that each can dispatch little power supply; and the dealing electrical power between the big electrical network; the reactive power of the discharge power of storage battery and each idle output scalable power supply is for optimizing variable; energy-optimised model 1 is the mixed integer nonlinear programming problem, its target function:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+{\mathrm{\λ}}_{\mathrm{bat}}+U_P^t{C}_P-U_S^t{I}_S$

In the formula: λ _Bat=(σ P _Bat) Δ t

The t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

Δ t-optimizes the time interval;

σ-designed coefficient;

P _BatThe discharge power of-storage battery;

Its constraints:

(a) trend constraints

$P_i-\underset{j=1}{\overset{j=n}{\mathrm{\Σ}}}[e_i(G_{\mathrm{ij}}{e}_j-B_{\mathrm{ij}}{f}_j)+f_i(G_{\mathrm{ij}}{f}_j+B_{\mathrm{ij}}{e}_j)]=0$

$Q_i-\underset{j=1}{\overset{j=n}{\mathrm{\Σ}}}[f_i(G_{\mathrm{ij}}{e}_j-B_{\mathrm{ij}}{f}_j)-e_i(G_{\mathrm{ij}}{f}_j+B_{\mathrm{ij}}{e}_j)]=0$

In the formula: P _i, Q _i-each node injects active power and reactive power;

e _i, f _iThe real part and the imaginary part of each node voltage of-usefulness complex representation;

G _Ij, B _IjThe real part and the imaginary part of-i node and j node admittance element;

Total node number in the little electrical network of n-.

(b) can dispatch little power supply capacity constraint

$P_i^{\min }\≤P_i^t\≤P_i^{\max }$

In the formula:

-Di i platform can be dispatched the active power output of little power supply in the t period;

-Di i platform can be dispatched the maximum active power output limit value and the minimum active power output limit value of little power supply;

(c) the discharge active power of storage battery constraint

$0\≤P_{\mathrm{bat}}\≤P_{\mathrm{dh}\_\max }^t$

In the formula: P _BatThe discharge power of-storage battery;

But-at the maximum discharge power limit value of t storage battery constantly the time;

(d) heap(ed) capacity constraint that can be mutual between little electrical network and big electrical network, the supply and demand agreement that this may be between them to be reached or the physical transfer capacity limit value of interconnection

$0\≤P_{\mathrm{Pgrid}}^t\≤P_{\mathrm{Pgrid}}^{\max }$

$0\≤P_{\mathrm{Sgrid}}^t\≤P_{\mathrm{Sgrid}}^{\max }$

In the formula:

-little electrical network is purchased the active power of electricity from big electrical network;

-little electrical network is to the active power of big electrical network sale of electricity;

-little electrical network is purchased the maximum active power limit value of electricity from big electrical network;

-little electrical network is to the maximum active power limit value of big electrical network sale of electricity;

(e) can dispatch the shortest continuous operating time and the constraint of the shortest continuous idle time of little power supply

$(T_{i,\mathrm{on}}^{t-1}-\mathrm{MRT})\·(U_i^{t-1}-U_i^t)\≥0$

$(T_{i,\mathrm{off}}^{t-1}-\mathrm{MST})\·(U_i^t-U_i^{t-1})\≥0$

In the formula:

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-t-1 constantly the i platform can dispatch the continuous operating time of little power supply;

-in the t-1 i platform continuous idle time that can dispatch little power supply constantly;

MRT-i platform can be dispatched the minimum continuous operating time of little power supply;

MST-i platform can be dispatched the minimum idle time continuously of little power supply;

(f) can dispatch little power supply active power output rate of change limit

$P_i^t-P_i^{t-1}\≤\mathrm{\Δ}{P}_U$

$P_i^{t-1}-P_i^t\≤\mathrm{\Δ}{P}_D$

In the formula:

-Di i platform can be dispatched the active power output of little power supply in the t period;

Δ P _U-can dispatch little this scheduling of power supply moment power with respect to last one restriction of power increase constantly;

Δ P _D-can dispatch little this scheduling of power supply moment power with respect to last one restriction of power reduction constantly;

(g) buy and sell electric mutual exclusion constraint from big electrical network

$U_P^t+U_S^t\≤1$

(h) the idle output constraint of reactive power output scalable power supply

$0\≤Q_i^t\≤Q_i^{\max }$

In the formula: -Di i platform reactive power output scalable power supply is in t reactive power output constantly;

The maximum reactive power output limit value of-Di i platform reactive power output scalable power supply.

5. according to the energy-optimised dispatching method of little real-time power network under claim 1 or the 4 described a kind of patterns that are incorporated into the power networks; it is characterized in that: the optimization 2 in the described step 4) is meant finds the solution the energy-optimised model of being set up 2; energy-optimised model 2 is minimum with little operation of power networks cost to be target; so that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output that each can dispatch little power supply; and the dealing electrical power between the big electrical network; the reactive power of the discharge power of storage battery and each idle output scalable power supply is for optimizing variable; energy-optimised model 2 is the mixed integer nonlinear programming problem; its target function is identical with energy-optimised model 1; its constraints and energy-optimised model 1 are basic identical; just " (c) the discharge active power of the storage battery constraint " in the constraints changes, and " (c) the discharge active power of storage battery constraint " in the constraints of energy-optimised model 2 is:

$-P_{\mathrm{ch}\_\max }^t\≤P_{\mathrm{bat}}\≤P_{\mathrm{dh}\_\max }^t$

In the formula: P _BatThe discharge power of-storage battery; P _Bat＜0, represent to charge in batteries P _Bat＞0, expression is to battery discharging;

The maximum chargeable power of-storage battery;

But-storage battery maximum discharge power.

6. according to the energy-optimised dispatching method of little real-time power network under claim 1 or the 4 described a kind of patterns that are incorporated into the power networks; it is characterized in that: the optimization 3 in the described step 4) is meant finds the solution the energy-optimised model of being set up 3; energy-optimised model 3 is minimum with little operation of power networks cost to be target; so that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output that each can dispatch little power supply; and the dealing electrical power between the big electrical network; the reactive power of the discharge power of storage battery and each idle output scalable power supply is for optimizing variable; energy-optimised model 3 is the mixed integer nonlinear programming problem; its constraints is identical with energy-optimised model 1; its target function is different slightly with energy-optimised model 1, and difference is the λ in the target function of energy-optimised model 3 _BatWith the target function of energy-optimised model 1 in λ _BatItem designs differently, and the target function of energy-optimised model 3 is:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+{\mathrm{\λ}}_{\mathrm{bat}}+U_P^t{C}_P-U_S^t{I}_S$

In the formula: λ _Bat=δ P _DhΔ t

δ＝a ₁+a ₂·dSOC+a ₃·P _dh+a ₄·dSOC·P _dh+a ₅·dSOC ²

dSOC＝SOC-SOC ^min

The t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

Δ t-optimizes the time interval;

P _BatThe discharge power of-storage battery;

The state-of-charge of SOC-storage battery;

SOC ^MinThe state-of-charge lower limit of-storage battery;

a ₁, a ₂, a ₃, a ₄, a ₅-designed coefficient.

7. according to the energy-optimised dispatching method of little real-time power network under claim 1 or the 4 described a kind of patterns that are incorporated into the power networks; it is characterized in that: the optimization 4 in the described step 4) is meant finds the solution the energy-optimised model of being set up 4; energy-optimised model 4 is minimum with little operation of power networks cost to be target; so that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output that each can dispatch little power supply; and the reactive power of the dealing electrical power between the big electrical network and each idle output scalable power supply is for optimizing variable; energy-optimised model 4 is the mixed integer nonlinear programming problem; with respect to energy-optimised model 1, there is not λ in the target function of energy-optimised model 4 _Bat, the target function of energy-optimised model 4 is:

$\min \underset{i=1}{\overset{N}{\mathrm{\Σ}}}(U_i^t{C}_f+U_i^t{C}_{\mathrm{OM}}+U_i^t(1-U_i^{t-1})C_{\mathrm{si}}^t)+U_P^t{C}_P-U_S^t{I}_S$

In the formula: the t-system operation period;

Can dispatch little power supply number in the i-system;

Can dispatch total number of little power supply in the N-system;

-t constantly the i platform can dispatch the state of little power supply, 0 expression is in shut down condition, 1 expression is in running status;

-whether purchase electricity at t little electrical network of the moment from big electrical network, 0 expression is denied that 1 expression is;

-whether little electrical network is to big electrical network sale of electricity constantly at t, and 0 expression is denied that 1 expression is;

C _f-can dispatch the energy consumption cost of little power supply;

C _OM-can dispatch the operation expense of little power supply;

-can dispatch the start-up cost of little power supply;

λ _Bat-designed accumulator cell charging and discharging cost function;

C _P-little electrical network is purchased the expenditure of electricity from big electrical network;

I _S-little electrical network is to the income of big electrical network sale of electricity;

The constraints of the constraints of energy-optimised model 4 and energy-optimised model 1 is basic identical, and being does not have " (c) the discharge active power of storage battery constraint " this item constraint in the constraints of energy-optimised model 4.

8. the little real-time power network energy optimizing method under a kind of pattern that is incorporated into the power networks according to claim 1, it is characterized in that: in the described step 4) with firm power to charge in batteries, and be optimized 4 and be meant that charging and discharging of accumulator power does not participate in optimizing in 4 optimizing, and with firm power to charge in batteries, this moment, storage battery was equivalent to a constant load, and will contribute to the firm power of charge in batteries in the load at little electrical network network node place at storage battery place and go, and be the constant charge power of storage battery

$P_{\mathrm{bat}}^t=I_{\mathrm{bat}\_\max \_\mathrm{ch}}\·U_{\mathrm{batt}}^t$

In the formula: I _{Bat_max_ch}The maximum charging current that allows of-storage battery, specification is given by producing;

-accumulator voltage.

9. according to the little real-time power network energy optimizing method under claim 1 or 4 or 5 or the 6 or 7 described a kind of patterns that are incorporated into the power networks; it is characterized in that: the optimization result in the described step 5) is meant by solution procedure 4) in optimization 1; optimize 2; optimize 3 and optimize 4 related energy-optimised models 1; energy-optimised model 2; energy-optimised model 3 and energy-optimised model 4 and the result that obtains comprises that each can dispatch the startup-shutdown state of little power supply in little electrical network; the meritorious output valve that each can dispatch little power supply; and the dealing electrical power value between the big electrical network; the reactive power value of the discharge power value of storage battery and each idle output scalable power supply.

Images